Introduction

Apart from shigellosis, Shigella species are the pioneer cause of the bacillary dysentery in human. Shigella-associated diarrhea in children remains an important global human health problem, especially in developing countries [1]. It is estimated that 13.2% of all diarrhea-related mortality in the world are caused by Shigella species, making them the second most common cause of these deaths [1, 2]. Shigella is a member of the Enterobacterales family, and generally, four Shigella species, including Shigella sonnei, Shigella boydii, Shigella dysenteriae, and Shigella flexneri are responsible for shigellosis or milder forms of diarrheal disease [3]. Among them, S. sonnei is the most reported and sometimes dominant species in developed countries that causes mild, watery, or bloody diarrhea. While in low- and middle-income nations, S. flexneri is the major species that causes epidemic and endemic shigellosis. Likewise, the other two species (S. boydii and S. dysenteriae) are relatively rare [2, 4].

Fluoroquinolones belong to a family of synthetic broad-spectrum antibiotics. In addition to being effective antibiotics against multidrug-resistant bacteria, they are also among the most prescribed antibiotics for treating patients with shigellosis [4, 5]. Thus, quinolone-resistant Shigella species received a high rank among the global priority list of antibiotic-resistant bacteria published by the World Health Organization (WHO) [2, 3]. In recent years, fluoroquinolone-resistant Shigella species have emerged in many parts of the globe, particularly in Asia due to excessive use of these antibiotics [6].

Based on the available evidence, the resistance to fluoroquinolones in bacteria, including Shigella species, occurs mainly by point mutations in chromosomal genes encoding DNA gyrase (gyrA and gyrB) and topoisomerase IV (parC and parE) located in quinolone resistance-determining regions (QRDRs) [3, 5,6,7]. However, the plasmid-mediated quinolone resistance (PMQR) genes are also involved in this process through four main mechanisms, including the aminoglycoside acetyltransferase-encoding enzyme variant aac(6’)-lb-cr, qnr genes, efflux-pump-encoding qepA and oqxAB genes, and a recently detected phosphorylase gene (crpP) [3, 5, 8]. The aac(6’)-lb-cr gene codes a variant of the aminoglycoside acetyltransferase that acetylates aminoglycosides such as amikacin and kanamycin and fluoroquinolones such as ciprofloxacin and norfloxacin [9]. Currently, seven types of qnr genes have been demonstrated to be transferrable by plasmids, namely qnrA, qnrB, qnrC, qnrD, qnrE, qnrS, and qnrVC [10]. A different qnr protein belonging to the pentapeptide repeat protein (PRP) family is encoded by each gene that alters the target of quinolones [11]. A 56-kb conjugative plasmid, pOLA52, encodes oqxAB efflux pump via oqxA and oqxB genes, which confers resistance to several compounds, including fluoroquinolone antibiotics [12]. The quinolone efflux pump A (qepA) gene was first described in an Escherichia coli isolate from a urine specimen in 2007 [13]. This gene encodes a protein of 511 amino acids that significantly reduces the susceptibility to quinolones [12]. ATP-dependent phosphorylation of ciprofloxacin by CrpP enzyme that encoded by crpP gene, leads to a reduction in ciprofloxacin susceptibility. The crpP gene have mostly been identified in Pseudomonas aeruginosa, with lesser frequency reported in E. coli, Acinetobacter baumannii, and Klebsiella pneumoniae [8].

Also, a substantial amount of research has focused on the association between antibiotic resistance and biofilm production in bacteria. However, researchers are still attempting to define the association among creation of biofilms and resistance to antibiotics, as there are inconsistent results among different bacteria species [14]. Moreover, few studies have investigated the association between the formation of biofilms in Shigella isolates and quinolone resistance.

Consequently, the purpose of this study was to examine the frequency of PMQR genes and biofilm development in various species of clinical quinolone-resistant Shigella isolates. Also, the enterobacterial repetitive intergenic consensus polymerase chain reaction (ERIC-PCR) technique was used to examine the clonal relatedness of quinolone-resistant isolates.

Materials and methods

Ethics statement

Ethics approval was obtained from the University of Babylon, Hilla, Iraq (Project No: M220601) compliant with the Declaration of Helsinki. All methods were performed in accordance with the relevant guidelines and regulations of the University of Babylon, Hilla, Iraq. Written consent was obtained from parents or guardians of all participants.

Study design and sample collection

In this cross-sectional study, stool samples were collected from 150 pediatric patients (younger than 10 years) with diarrhea referred to the private hospitals and treatment centers from November 2020 to December 2021. Exclusionary criteria were antibiotic use within 72 h of sample collection and being over 10 years in age, while inclusion criteria included any manifestation of fever, abdominal pain, cramps, vomiting or diarrhea. Sterile plastic containers were used to collect stool samples, which were immediately transferred to the microbiology laboratory for analysis.

Bacterial isolation and identification

A portion of all samples were inoculated on Hektoen Enteric (HE) Agar and Xylose Lysine Desoxycholate (XLD) Agar (Condalab, Madrid, Spain) and incubated at 37 ºC for 24 h. Also, a small percentage of specimens were enriched in 5 ml of Gram-negative (GN) broth (Condalab, Madrid, Spain) for six hours at 37 °C. The enriched specimens were sub-cultured on HE and XLD plates and placed at 37 ºC for 24 h [4]. The isolates that appeared green and red on HE and XLD, respectively, were considered presumptive Shigella isolates and identified using standard biochemical tests, including sulfide indole motility (SIM), citrate and urea hydrolysis, methyl red (MR) and Voges–Proskauer (VP), Kligler Iron Agar (KIA), and Lysine Iron Agar (LIA) [15]. Shigella isolates were confirmed by the VITEK 2 Compact system (bioMérieux, Marcy l’Étoile, France) according to the manufacturer protocol [16]. All isolates were further confirmed by polymerase chain reaction (PCR) of putative integrase as a specific marker for Shigella genus using previously described primers (Bioneer, South Korea) (Table 1) [17]. The QIAamp DNA Minikit (Qiagen GmbH, Hilden, Germany) was used for DNA extraction from Shigella isolates following the manufacturer instructions. PCR reaction (25 μL) was performed with the following program: 95 °C for 5 min, 35 cycles of 95 °C for 40 s, 60 °C for 30 s, 72°C for 30 s, and final extension at 72 °C for 5 min. The PCR products (5 μL) were loaded into 1% agarose gel and placed in 1X tris–acetate-EDTA (TAE) buffer contained ethidium bromide (0.5 μg/mL). The electrophoresis was performed for 45 min and amplicons were analyzed in a Bio-Rad gel documentation system (USA). In all experiments, S. sonnei ATCC® 25,931™ and S. flexneri ATCC® 29,903™ were used as quality control.

Table 1 Primer sequences that were used in this study
Full size table

Antimicrobial susceptibility testing (AST)

The minimal inhibitory concentrations (MICs) of ciprofloxacin (32–0.03125 μg/mL), levofloxacin (32–0.03125 μg/mL), gatifloxacin (64–0.0625 μg/mL), ofloxacin (64–0.0625 μg/mL), imipenem (64–0.0625 μg/mL), meropenem (64–0.0625 μg/mL), ampicillin (512–0.05 μg/mL), amoxicillin/clavulanic acid (512–0.05 μg/mL), azithromycin (512–0.05 μg/mL), cefotaxime (64–0.0625 μg/mL), ceftriaxone (64–0.0625 μg/mL), ceftazidime (256–0.025 μg/mL), tetracycline (256–0.025 μg/mL), chloramphenicol (512–0.05 μg/mL), trimethoprim/sulfamethoxazole (64–0.0625 μg/mL), and cefepime (256–0.025 μg/mL) (Sigma-Aldrich, Darmstadt, Germany) were determined by broth microdilution method according to the Clinical Laboratory Standard Institute (CLSI) 2021 principles [18]. The Mueller–Hinton broth (Merck, Darmstadt, Germany) was used for AST. The breakpoints for resistance or susceptibility to each antibiotic were determined according to the CLSI 2021 criteria [18]. Multidrug-resistant (MDR) isolates were defined as those resistant to three or more antibiotics belong to different category [14]. Escherichia coli ATCC® 25,922™ was used as the control strain.

PMQR genes detection by PCR method

Uniplex PCR method was used to detect 11 PMQR genes in quinolone-resistant isolates using previously described primers (Bioneer, South Korea) as follows: qnrA, qnrB, qnrC, qnrD, qnrE, qnrS, qnrVC, aac(6′)-Ib-cr, qepA, oqxAB, and crpP [19,20,21,22]. The PCR reaction (25 μL) was performed in a Bio-Rad thermocycler (USA) with the following temperatures: 95 °C for 5 min, 35 cycles of 95 °C for 30 s, annealing (Table 1) for 30 s, 72°C for 30 s, and final extension at 72 °C for 5 min. The electrophoresis and visualization of the PCR products were performed similar to the previous stage.

Biofilm formation

A 96-well microtiter plate was used to assess the biofilm formation of quinolone-resistant and quinolone-susceptible isolates based on the previously described method with minor modifications [14]. The isolates were cultivated overnight in tryptic soy broth (Condalab, Madrid, Spain) and then were adjusted to a 0.5 McFarland standard turbidity. A total of 150 μL of these suspensions was dispensed in microtiter plate wells and incubated overnight at 37 °C. An initial three washes with phosphate-buffered saline (PBS) were done, followed by 20 min of fixation with methanol and 15 min of staining with 0.1% (v/v) crystal violet solution. Biofilm formation was measured at 550 nm optical density (OD550) using a microtiter plate reader (Thermo Fisher Scientific, USA). The isolates were categorized into four groups: non-biofilm (OD ≤ ODc), low-biofilm (ODc < OD ≤ 2 ODc), medium-biofilm (2 ODc < OD ≤ 4 ODc), and strong-biofilm (OD > 4 ODc) [14].

Clonal association by ERIC-PCR

The clonal association of quinolone-resistant isolates were investigated by ERIC-PCR. ERIC-PCR was performed in 25 μL volume using following program: 95 °C (15 min), 35 cycles [95 °C (30 s), 40 °C (45 s), and 72 °C (6 min)], and a final extension at 72 °C (10 min). The ERIC-PCR primers are shown in Table 1 [23]. Using a standard 1-kb DNA ladder (Fermentas, USA) as a reference, the ERIC-PCR patterns were evaluated by comparing positions of band lanes in each agarose gel (2%). To analyze the ERIC-PCR patterns, dendrograms were generated based on unweighted pair group method with arithmetic mean (UPGMA) and the Dice similarity coefficient using GelJ version 2.0 software [23, 24]. Shigella isolates with 90.0% or more similarity were considered to be clonally related.

Statistical analysis

Descriptive statistical analysis (frequency and prevalence of antibiotic resistance rates and PMQR genes) were performed using Statistical Package for the Social Sciences (SPSS) version 20.0 (SPSS Inc., Chicago, IL, USA). Fisher's exact test or Chi-square test were used to determine any significant correlations (P-value < 0.05) between variables.

Results

Shigella isolates

During the study period, a total of 95 presumptive Shigella isolates were collected and primarily identified using standard phenotypic and biochemical tests. Using VITEK 2 Compact system, the Shigella isolates were confirmed on species level in the following order: S. sonnei (53, 55.8%), S. flexneri (39, 41.1%), and S. boydii (3, 3.2%). No S. dysenteriae isolate was detected. All identified isolates showed the 159 bp specific band of putative integrase of Shigella genus in PCR assay. Shigella isolates were mostly retrieved from male patients (59, 62.1%) including 31 (52.5%) S. sonnei, 25 (42.4%) S. flexneri, and 3 (5.1%) S. boydii. The remaining were collected from females (36, 37.9%). There was no significant differences between the prevalence of Shigella species and gender of patients (P-value = 0.334). Also, most of the isolates were identified in pediatric patients aged from 4–6 years old (42, 44.2%), followed by 2–4 years old (28, 29.5%), 6–8 years old (13, 13.7%), < 2 years old (9, 9.5%), and 8–10 years old (3, 3.2%).

Antibiotic resistance rates

The AST revealed that 98.9% (n = 94/95) and 97.9% (n = 93/95) of Shigella isolates were susceptible to imipenem and meropenem, respectively (Table 2). S. sonnei, S. boydii, and S. flexneri showed the susceptibility rates of 100.0% (n = 53/53), 100.0% (n = 3/3), and 97.4% (n = 38/39) towards imipenem, respectively. The highest resistance rates of the Shigella isolates were against ampicillin (92.6%, n = 88/95), followed by amoxicillin/clavulanic acid (85.3%, n = 81/95), trimethoprim/sulfamethoxazole (76.8%, n = 73/95), ceftriaxone (56.8%, n = 54/95), and cefotaxime (50.5%, n = 48/95). All three species S. boydii (100.0%, n = 3/3), S. sonnei (90.6%, n = 48/53), and S. flexneri (94.9%, n = 37/39) were highly resistant to ampicillin, with no significant difference (P > 0.05). In total, 42 of 95 (44.2%) Shigella isolates were quinolone-resistant including 26 (61.9%) S. sonnei and 16 (38.1%) S. flexneri. No S. boydii isolate was quinolone-resistant. All 42 quinolone-resistant isolates were simultaneously resistant against two or more quinolones (Table 2). The highest co-resistance rate was observed against levofloxacin–ofloxacin (23.8%, n = 10/42), followed by ciprofloxacin–levofloxacin (21.4%, n = 9/42). The highest resistance rate was against ciprofloxacin (31.6%, n = 30/95) followed by ofloxacin (28.4%, n = 27/95), levofloxacin (26.3%, n = 25/95), and gatifloxacin (15.8%, n = 15/95), respectively. All isolates including 42 quinolone-resistant strains were MDR.

Table 2 Antibiotic resistance rates and quinolone co-resistance of the different Shigella species
Full size table

Prevalence of PMQR genes

In total, the PMQR genes were detected in 92.9% (39/42) of quinolone-resistant Shigella isolates including all quinolone-resistant S. flexneri (100.0%, n = 16/16) and 88.5% (n = 23/26) of quinolone-resistant S. sonnei. Four different PMQR genes were found to be associated with both quinolone-resistant S. sonnei and S. flexneri isolates in this study (Table 3, Fig. 1, Fig. 2, and Fig. 3). The occurrence of these genes in their descending order of positivity was as follows: qnrS (52.4%) > qnrA and aac(6′)-Ib-cr (33.3%) > qnrB (19.0%). The prevalence of different genes in Shigella species was as follows: 61.5% and 37.5% (qnrS), 19.2% and 56.3% (qnrA), 38.5% and 25.0 (aac(6′)-Ib-cr), and 19.2% and 18.8% (qnrB) for S. sonnei and S. flexneri, respectively. The co-incidence of PMQR determinants was detected in 40.5% (17/42) of quinolone-resistant isolates (Table 3). Statistically, there was no significant difference (P > 0.05) in the prevalence of the four PMQR genes in both Shigella species. The qnrC, qnrD, qnrE, qnrVC, qepA, oqxAB, and crpP genes were not detected in any isolate.

Table 3 Prevalence of plasmid-mediated quinolone resistance (PMQR) genes in different species of quinolone-resistant clinical Shigella isolates
Full size table
Fig. 1
figure 1

Gel electrophoresis of qnrS gene (428 bp). Lane 1: DNA ladder (100 bp); Lane 2: Control positive; Lanes 3–8: qnrS-positive Shigella sonnei isolates; Lanes 9, 10, 12, and 14: qnrS-positive Shigella flexneri isolates; Lanes 11 and 13: qnrS-negative Shigella sonnei

Full size image
Fig. 2
figure 2

Gel electrophoresis of qnrB (488 bp) and aac(6’)-lb-cr (260 bp). Lane 1: DNA ladder (100 bp); Lane 2: Positive control of qnrB gene; Lanes 3: qnrS-positive Shigella sonnei isolate; Lane 4: Negative control: DNA/RNA free water; Lane 5: Positive control of aac(6’)-lb-cr gene; Lanes 6–8: aac(6’)-lb-cr-positive Shigella flexneri isolates; Lanes 9–11: aac(6’)-lb-cr-positive Shigella sonnei isolates; Lanes 12 and 13: aac(6’)-lb-cr-negative Shigella flexneri isolates

Full size image
Fig. 3
figure 3

Gel electrophoresis of qnrA gene (630 bp). Lane 1 and 12: DNA ladder (100 bp); Lane 2: Negative control: DNA/RNA free water; Lanes 3: Positive control of qnrA gene; Lanes 4–6: qnrA-positive Shigella sonnei isolates; Lanes 7, 8, 9, and 11: qnrA-positive Shigella flexneri isolates; Line 10: qnrA-negative Shigella flexneri isolates

Full size image

Biofilm formation

In total, 52.8% (28/53) of quinolone-susceptible isolates including 55.6% (15/27) of S. sonnei, 43.5% (10/23) of S. flexneri, and 100.0% (3/3) of S. boydii were low-biofilm producers. None of the quinolone-susceptible isolates were strong-biofilm producers or medium-biofilm producers. However, 64.3% (27/42) of quinolone-resistant isolates showed the biofilm formation phenomenon including 28.6% (12/42) strong-biofilm producers, 23.8% (10/42) medium-biofilm producers, and 11.9% (5/42) low-biofilm producers. Also, 15 isolates (35.7%, 15/42) were non-biofilm producers. In terms of Shigella species, 30.8% (8/26), 23.1% (6/26), 11.5% (3/26), and 34.6% (9/26) of quinolone-resistant S. sonnei were strong-biofilm producers, medium-biofilm producers, low-biofilm producers, and non-biofilm producers, respectively. Meanwhile, 25.0% (4/16), 25.0% (4/16), 12.5% (2/16), and 37.5% (6/16) of quinolone-resistant S. flexneri were strong-biofilm producers, medium-biofilm producers, low-biofilm producers, and non-biofilm producers, respectively. Biofilm formation was not significantly different between quinolone-resistant S. sonnei and quinolone-resistant S. flexneri isolates (P-value > 0.999). Moreover, the biofilm formation was not significantly different between quinolone-resistant and quinolone-susceptible Shigella isolates (P-value = 0.299).

Clonal association

Based on the ERIC-PCR, 16 quinolone-resistant S. flexneri isolates were distributed to 4 clusters (A to D) including 2 to 3 isolates and 6 singletons with 12 various ERIC-types (E1 to E12), indicating a high genetic diversity among the isolates (Fig. 4). The ERIC-PCR patterns included 3 to 8 bands with various sizes ranging from 100 to 1000 bp for each isolate. Also, the 26 quinolone-resistant S. sonnei isolates were distributed to 3 clusters (A to C) including 2 to 3 isolates and 19 singletons with 24 various ERIC-types (E1 to E24), showing a high genetic diversity among the isolates (Fig. 5). The ERIC-PCR patterns included 3 to 9 bands with various sizes that ranged from 100 to 1300 bp for each isolate.

Fig. 4
figure 4

The UPGMA dendrogram with the Dice coefficient of quinolone-resistant Shigella flexneri isolates in the basis of ERIC-PCR patterns

Full size image
Fig. 5
figure 5

The UPGMA dendrogram with the Dice coefficient of quinolone-resistant Shigella sonnei isolates in the basis of ERIC-PCR patterns

Full size image

Discussion

In this study, S. sonnei (58.0%) was the most common cause of the shigellosis followed by S. flexneri (41.1%) and S. boydii (3.2%). In contrast to this study, previous reports suggested S. flexneri as the leading species of shigellosis in developing countries [25, 26]. However, most developing countries, including those in Southeast Asia, are recording a shift in the major etiological agent of shigellosis to S. sonnei [27]. In line with the current study, Teimourpour et al. [28] from Iran, reported S. sonnei (69.9%) as the most predominant species followed by S. flexneri (19.4%) and S. boydii (9.9%). On the other hand, the absence of S. dysenteriae in this study was not surprising, as the isolation of these species is relatively rare when compared to the other three species [4, 6]. Yu et al. [3] from China, also corroborated this result by recording a complete absence of S. dysenteriae in a study involving 118 Shigella isolates. These discrepancies may be due to the differences in the sample size, the race of studies population, and the hygiene situation of studied region.

Shigella infections are distributed differently based on age and gender worldwide. This study revealed the most incidence rate of Shigella isolates in males and in pediatric patients aged from 4–6 years old. In line with the current observations, Abbasi et al. [4] from Iran, Liu et al. [15] from China, and Jain et al. [25] from India reported a higher frequency rate of Shigella isolates in males than that of females. The descriptive epidemiology identified varying rates of shigellosis cases among males and females. Nevertheless, there is a lack of comprehensive explanations on the factors contributing to this disparity and the methods of transmission [29]. One of the reasons for this difference may be women's better personal hygiene habits [26, 29]. Moreover, males are more likely to be exposed to the environments contaminated with Shigella species [15]. Previous studies reported a more prevalence rate of Shigella infections in pediatric patients under 5 years which was consistent with our findings [15, 30]. Due to their low immunity and lack of prior exposure, children under five are more susceptible to shigellosis than older children [15].

Shigella species isolated in this study displayed a wide spectrum of antibiotic resistance. The highest resistance rates were found to ampicillin (92.6%), amoxicillin/clavulanic acid (85.3%), and trimethoprim/sulfamethoxazole (76.8%). S. boydii completely resisted ampicillin, amoxicillin/clavulanic acid, cefotaxime, and trimethoprim/sulfamethoxazole with a rate of 100.0%. Despite having a low prevalence of isolation, S. boydii could pose a serious health concern. Similar results were previously reported by Abbasi et al. [4], Zamanlo et al. [31] and Farahani et al. [32] from neighboring geographical area, Liu et al. [16] from China, and Phiri et al. [26] from Malawi. In another study carried out in Kenya by Leting et al. [33], the Shigella isolates displayed almost similar resistance rate against ampicillin (100.0%) and amoxicillin/clavulanic acid (84.2%). This implied that these group of antibiotics are no more suitable for the treatment of shigellosis. This could have resulted due to the extensive usage of these antibiotics, especially in the treatment of diarrhea leading to high degrees of resistance. According to Behruznia et al. [34], Shigella species have become resistant to first-line drugs (trimethoprim–sulfamethoxazole and ampicillin), so that they are no longer recommended for the treatment of shigellosis. The increase in resistance by Shigella species calls for concerns as the United Kingdom, Northern Ireland, and numerous other nations have all reported cases of extensively drug-resistant (XDR) Shigella sonnei [35]. Besides, our findings revealed that almost 50.0% of Shigella isolates were resistant to third generation cephalosporins that was in line with the previous reports from Iran [4, 31]. However, in contrast to our findings, Madhavan et al. [30] from India and Liu et al. [36] from China reported lower resistance rates for these antibiotics. The high resistance rate to third generation cephalosporins in this study may be due to the emergence of extended-spectrum beta-lactamases (ESBLs)-producing Shigella isolates [37,38,39]. In recent years, there have been reports of the emergence of Shigella strains carrying ESBL genes from some parts of the world including China, Iran, and England [37,38,39].

In this study, the focus was on quinolone-resistant strains. In total, 44.2% of Shigella isolates were simultaneously resistant against two or more quinolones including 61.9% S. sonnei and 38.1% S. flexneri. This resistance rate was much higher than a previous report from neighboring country Iran in which only 4.4% of isolates were quinolone-resistant [28]. Also, in this study, the most resistance rate among quinolones was related to ciprofloxacin (31.6%) that was in line with a previous report from China [6]. Due to the widespread use of fluoroquinolones as first-line antibiotics to treat diarrhea in recent years, quinolone resistance has increased considerably. A lack of local antimicrobial resistance surveillance, easy access to antibiotics, and inappropriate prescription of antibiotics in our region may all be contributing factors to the situation's deterioration. However, the high quinolone-resistant rate for S. sonnei in this study was in good parallel with a previous report that showed 76% (313/411) of S. sonnei isolates collected from several countries were quinolone-resistant [40]. A previous study from India has also reported quinolone-resistant S. sonnei brought on by a combination of qnrB and gyrA mutations [41]. Based on the available evidence, men who have sex with men (MSM) in Taiwan are frequently infected with quinolone-resistant S. sonnei that has the same QRDR mutations [42].

The PMQR determinants have escalated into a serious global issue over the past few years [6]. To the best of our knowledge, this study was the first to investigate 11 different PMQR genes in quinolone-resistant Shigella isolates. The qnrS (52.4%) was the most frequent PMQR gene followed by qnrA and aac(6′)-Ib-cr (33.3%), and qnrB (19.0%). The qnrC, qnrD, qnrE, qnrVC, qepA, oqxAB, and crpP genes were not detected in any isolate. In a previous study from India [41], 18.7% of S. sonnei isolates harbored qnrC gene and no isolate was positive for qepA, qnrS, and qnrA. In another study from Iran, qnrC and qnrD genes were not detected in any Shigella isolates [43]. Also, in a previous research from China [44], qnrA, qnrC, qnrD and qepA genes were not found in S. flexneri isolates. However, Yang et al. [45] from China, reported the qepA gene in 2.05% of Shigella isolates. Moreover, based on the available evidence, there is no report of the presence of qnrE, qnrVC, and crpP genes in Shigella isolates from different countries.

Statistically, there was no significant difference in the prevalence of the four PMQR genes between S. sonnei and S. flexneri species. In a previous study from China, aac (6′)-Ib-cr was the most frequent gene found in quinolone-resistant Shigella isolates [46]. The frequency of selecting chromosomal mutants to decrease ciprofloxacin activity by N-acetylation of its piperazinyl substituent has been shown to be significantly increased by aac (6′)-Ib-cr [46]. In this study, qnrA was detected in 19.2% and 56.3% of S. sonnei and S. flexneri isolates, respectively. However, in a previous studies from China [44, 46], the Shigella isolates were negative for this gene. Based on evidence, China had a majority prevalence of qnrS in S. sonnei and S. flexneri, while India had a majority prevalence of qnrB [47]. Also, similar to this study, qepA gene was not detected in S. sonnei isolates from China [48]. In a previous study from India, the qnrS1 was detected in 47% of S. boydii isolate and no S. sonnei was positive for qnr genes that was in contrast to this study [49].

In this study, 52.8% of quinolone-susceptible and 64.3% of quinolone-resistant isolates were biofilm producers and the biofilm formation rates of resistant and susceptible isolates were not significantly different. In line with this study, no association was found between the biofilm formation and antibiotic resistance in Pseudomonas aeruginosa and Acinetobacter baumannii isolates in previous studies [50, 51]. However, several investigations have shown the positive correlation between biofilm production and antibiotic resistance [52, 53]. Despite extensive research, Shigella species have not been thoroughly characterized in terms of association of their biofilm formation with antimicrobial resistance. Hence, further studies are needed to reveal the molecular mechanisms involved in these processes.

Based on the ERIC-PCR, quinolone-resistant S. flexneri and S. sonnei isolates had a high genetic diversity. This incredibly complex pattern makes it possible that various strains of the same species circulate in our region. These results were in good parallel with the previous reports from Malawi [26], Peru [54], and Latin America [55]. To prove that the Shigella strains in this study were genetically unrelated, additional research utilizing more exact typing techniques, such as whole-genome sequencing (WGS) and multi-locus sequence typing (MLST), would be required [26]. This study had some limitations as the other quinolone resistance mechanisms (mutations in QRDRs) were not investigated.

Conclusion

This study demonstrated the predominance of S. sonnei and S. flexneri in our region. The Shigella isolates showed high resistance to ampicillin, amoxicillin/clavulanic acid, and trimethoprim/sulfamethoxazole and all isolates were MDR. It seems that qnrS, qnrA, and aac(6′)-Ib-cr play significant role in the quinolone resistance among Shigella isolates in our region. There was no statistical association between biofilm formation and quinolone resistance. Also the quinolone-resistant S. flexneri and S. sonnei isolates had a high genetic diversity. Hence, antibiotic therapy needs to be routinely monitored based on the surveillance programs.