Molecular Detection of Fluoroquinolone Resistance among Multidrug-, Extensively Drug-, and Pan-Drug-Resistant Campylobacter Species in Egypt

In recent times, resistant foodborne pathogens, especially of the Campylobacter species, have created several global crises. These crises have been compounded due to the evolution of multidrug-resistant (MDR) bacterial pathogens and the emergence of extensively drug-resistant (XDR) and pan-drug-resistant (PDR) strains. Therefore, this study aimed to investigate the development of resistance and the existence of both XDR and PDR among Campylobacter isolates. Moreover, we explored the use of the polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) technique for the detection of fluoroquinolone (FQ)-resistant Campylobacter isolates. A total of 120 Campylobacter isolates were identified depending on both phenotypic and genotypic methods. Of note, cefoxitin and imipenem were the most effective drugs against the investigated Campylobacter isolates. Interestingly, the majority of our isolates (75%) were MDR. Unfortunately, both XDR and PDR isolates were detected in our study with prevalence rates of 20.8% and 4.2%, respectively. All FQ-resistant isolates with ciprofloxacin minimum inhibitory concentrations ≥4 µg/mL were confirmed by the genetic detection of gyrA chromosomal mutation via substitution of threonine at position 86 to isoleucine (Thr-86-to-Ile) using the PCR-RFLP technique. Herein, PCR-RFLP was a more practical and less expensive method used for the detection of FQ resistant isolates. In conclusion, we introduced a fast genetic method for the identification of FQ-resistant isolates to avoid treatment failure through the proper description of antimicrobials.


Introduction
Campylobacteriosis, caused by Campylobacter species (spp.), is a worldwide foodborne bacterial disease with zoonotic importance. Most of the human campylobacteriosis cases are caused by Campylobacter jejuni (C. jejuni) and the closely related Campylobacter coli Regarding the species level, C. jejuni was the predominant species (45.7%), followed by C. coli (11.4%). The highest isolation rates of C. jejuni isolates were observed in human stool swabs (71.4%), followed by chicken cloacal swabs and breast muscle samples (68.6% each). Meanwhile, the highest isolation rates of C. coli isolates were detected in the chicken liver (22.9%), followed by human stool swabs (14.3%) ( Table 1). Furthermore, there were statistically significant differences (p < 0.001) in the prevalence of Campylobacter spp., C. jejuni and C. coli among different sample types.

Antimicrobial Susceptibility Testing of Campylobacter Isolates 2.2.1. Antimicrobial Susceptibility Profiles of Campylobacter Species from Various Sources
Analysis of the antimicrobial susceptibility of the recovered 120 Campylobacter isolates against the examined 24 antimicrobials showed that all the tested isolates were resistant to amoxycillin, ampicillin, cephalothin and erythromycin. Moreover, high resistance rates were detected against trimethoprim-sulfamethoxazole (98.3%), followed by nalidixic acid (97.5%), clarithromycin (96.7%) and azithromycin and clindamycin (95% each). On the other hand, the lowest resistance rates were observed against amikacin (39.2%), imipenem (40.8%) and cefoxitin (45.8%) ( Table 2).
Regarding the species level, C. jejuni isolates were 100% resistant to nalidixic acid, trimethoprim-sulfamethoxazole and clarithromycin, while C. coli isolates were 100% resistant to clindamycin. Additionally, our results showed higher resistance rates of C. jejuni than of C. coli isolates for the tested antimicrobials except for cefoxitin, cefoperazone, imipenem, doxycycline, azithromycin, tobramycin, amikacin, colistin and clindamycin ( Figure 1). There were statistically significant differences in the resistance prevalence among C. jejuni and C. coli isolates against sulbactam-ampicillin, amoxycillinclavulanic acid, cefoxitin, ciprofloxacin, trimethoprim-sulfamethoxazole and chloramphenicol (p = 0.01, 0.023, 0.011, 0.016, 0.039 and 0.018, respectively). Additionally, there were higher significant differences in the resistance prevalence among C. jejuni and C. coli isolates for cefepime, nalidixic acid, clarithromycin and gentamicin (p = 0.005, 0.007, 0.001 and 0.003, respectively). Meanwhile, there were no statistically significant differences in the resistance profiles among C. jejuni and C. coli isolates for the other tested antimicrobials (p > 0.05) (Figure 1). According to the isolates' sources, higher resistance rates were observed among human Campylobacter isolates than the chicken ones for the investigated antimicrobials except for sulbactam-ampicillin, amoxycillin-clavulanic acid, imipenem, aztreonam, trimethoprimsulfamethoxazole, clarithromycin and amikacin ( Figure 2). There were statistically significant differences in the resistance prevalence among human and chicken Campylobacter isolates against trimethoprim-sulfamethoxazole, linezolid and colistin (p = 0.041, 0.038 and 0.021, respectively). Moreover, there were higher significant differences in the resistance prevalence among human and chicken Campylobacter isolates against cefepime, aztreonam (p = 0.009 and 0.008, respectively) and cefoxitin (p < 0.001). Meanwhile, there were no statistically significant differences in the resistance profiles among human and chicken Campylobacter isolates for the other tested antimicrobials (p > 0.05) ( Figure 2).

The Minimum Inhibitory Concentrations of Ciprofloxacin against Campylobacter Isolates
Thirty-eight ciprofloxacin-resistant Campylobacter isolates that were resistant to 21 or more antimicrobials were tested against ciprofloxacin antibiotic by the broth microdilution method for the determination of its minimum inhibitory concentrations (MICs). Those isolates were recovered from human (13) and chicken (25) sources. The 25 chicken isolates were obtained from the chicken liver (11) and cloacal swabs and breast meat (7 each) samples. Interestingly, all the 38 tested isolates were 100% resistant to ciprofloxacin (MIC ≥ 4 µg/mL) ( Table 4) and these results were 100% correlated with those of the disc diffusion method. Additionally, the minimum bactericidal concentration (MBC) values of ciprofloxacin ranged from 8 to ≥256 µg/mL.

Molecular Grouping of Campylobacter Isolates from Different Sources
All the 38 screened Campylobacter isolates (100%) were identified as genus Campylobacter ( Figure 4). Moreover, 29 isolates (76.3%) were positive for mapA gene and confirmed to be C. jejuni ( Figure 5A), while the remaining 9 isolates (23.7%) were positive for the ceuE gene and confirmed to be C. coli ( Figure 5B). These results were 100% correlated with those of the conventional identification methods. Of the 29 C. jejuni isolates, 11 (37.9%) were obtained from human and 18 (62.1%) from chicken sources. Moreover, nine C. coli isolates were obtained from seven chicken (77.8%) and two human (22.2%) samples. There were statistically significant differences in the prevalence of C. jejuni and C. coli isolates among human stool swabs and chicken breast meat samples (p = 0.001 and 0.029, respectively). Meanwhile, there were no statistically significant differences in the prevalence of C. jejuni and C. coli isolates among chicken liver and cloacal swabs samples (p = 0.395 and 0.286, respectively) ( Figure 6). (22.2%) samples. There were statistically significant differences in the prevalence of C. jejuni and C. coli isolates among human stool swabs and chicken breast meat samples (p = 0.001 and 0.029, respectively). Meanwhile, there were no statistically significant differences in the prevalence of C. jejuni and C. coli isolates among chicken liver and cloacal swabs samples (p = 0.395 and 0.286, respectively) ( Figure 6).   (22.2%) samples. There were statistically significant differences in the prevalence of C. jejuni and C. coli isolates among human stool swabs and chicken breast meat samples (p = 0.001 and 0.029, respectively). Meanwhile, there were no statistically significant differences in the prevalence of C. jejuni and C. coli isolates among chicken liver and cloacal swabs samples (p = 0.395 and 0.286, respectively) ( Figure 6).   (22.2%) samples. There were statistically significant differences in the prevalence of C. jejuni and C. coli isolates among human stool swabs and chicken breast meat samples (p = 0.001 and 0.029, respectively). Meanwhile, there were no statistically significant differences in the prevalence of C. jejuni and C. coli isolates among chicken liver and cloacal swabs samples (p = 0.395 and 0.286, respectively) ( Figure 6).

Determination of Fluoroquinolone Resistance by PCR-RFLP Technique
All the 38 tested Campylobacter isolates had the same RFLP fragments (179 bp) (Figure 7), which suggested having a chromosomal point mutation in the QRDR of the gyrA gene by substitution of threonine amino acid at position 86 to isoleucine (Thr-86-to-Ile). Additionally, there was a 100% correlation between the ciprofloxacin MIC values and the PCR-RFLP results.

Determination of Fluoroquinolone Resistance by PCR-RFLP Technique
All the 38 tested Campylobacter isolates had the same RFLP fragments (179 bp) ( Figure  7), which suggested having a chromosomal point mutation in the QRDR of the gyrA gene by substitution of threonine amino acid at position 86 to isoleucine (Thr-86-to-Ile). Additionally, there was a 100% correlation between the ciprofloxacin MIC values and the PCR-RFLP results.

Discussion
It was announced that several worldwide crises were developed due to the wide spreading of resistant fungi [18] and bacteria such as MRSA, VRSA and Klebsiella spp., in addition to zoonotic foodborne pathogens including Campylobacter spp., Salmonella enteritidis and Salmonella typhimurium [10,[19][20][21][22][23][24][25][26]. The increasing antimicrobial resistance of Campylobacter spp., especially to FQs, macrolides and tetracyclines, is of great importance to human health worldwide. Our results revealed a high prevalence of Campylobacter spp. (57.1%) in samples recovered from chicken and human origins at Sharkia Governorate, Egypt. This is partially similar to a previous study carried out in Poland (54.4%) [27], but the levels were higher than those obtained in previous studies carried out in Egypt; 32.8% [10], 27.3% [28] and 7.6% [29]. Herein, Campylobacter spp. were more prevalent among human samples, followed by chicken ones, which was in contrast with previous studies conducted in Egypt, where Campylobacter spp. were more prevalent among chicken samples, followed by human ones [29,30]. Among our chicken samples, Campylobacter spp. were more prevalent among cloacal swabs and liver (88.6% each), which is higher than prevalences reported in a previous study carried out in Egypt (54.3% and 34.1%, respectively) [10]. In the current study, chicken franks and luncheon meats were Campylobacter negative (0% each), which is in complete agreement with a previous study conducted in Egypt [31]. Of note, the most common Campylobacter spp. was C. jejuni (45.7%), which is in complete agreement with the results of other studies carried out in Tunisia (68.9%) [32] and South Korea (77.6%) [33]. Generally, the variations in the prevalence of Campylobacter spp. among various studies could be due to the type of the tested samples, hygienic measures, isolation and identification methods, environmental conditions and the geographical location [34].
Unsurprisingly, there is a variation in the antimicrobial resistance among and within different countries, which strongly correlated with the type of prescribed drugs alongside the variation in guidelines for the use of antimicrobial drugs. In this context, the high levels of ciprofloxacin and doxycycline resistance rates detected among our tested isolates (76.7% and 84.2%, respectively) were lower than those detected in a previous study carried out in Tunisia (99.2% and 100%, respectively) [35]. Alarmingly, there has been a global warning concerning the evolution of MDR strains; however, concrete steps are

Discussion
It was announced that several worldwide crises were developed due to the wide spreading of resistant fungi [18] and bacteria such as MRSA, VRSA and Klebsiella spp., in addition to zoonotic foodborne pathogens including Campylobacter spp., Salmonella enteritidis and Salmonella typhimurium [10,[19][20][21][22][23][24][25][26]. The increasing antimicrobial resistance of Campylobacter spp., especially to FQs, macrolides and tetracyclines, is of great importance to human health worldwide. Our results revealed a high prevalence of Campylobacter spp. (57.1%) in samples recovered from chicken and human origins at Sharkia Governorate, Egypt. This is partially similar to a previous study carried out in Poland (54.4%) [27], but the levels were higher than those obtained in previous studies carried out in Egypt; 32.8% [10], 27.3% [28] and 7.6% [29]. Herein, Campylobacter spp. were more prevalent among human samples, followed by chicken ones, which was in contrast with previous studies conducted in Egypt, where Campylobacter spp. were more prevalent among chicken samples, followed by human ones [29,30]. Among our chicken samples, Campylobacter spp. were more prevalent among cloacal swabs and liver (88.6% each), which is higher than prevalences reported in a previous study carried out in Egypt (54.3% and 34.1%, respectively) [10]. In the current study, chicken franks and luncheon meats were Campylobacter negative (0% each), which is in complete agreement with a previous study conducted in Egypt [31]. Of note, the most common Campylobacter spp. was C. jejuni (45.7%), which is in complete agreement with the results of other studies carried out in Tunisia (68.9%) [32] and South Korea (77.6%) [33]. Generally, the variations in the prevalence of Campylobacter spp. among various studies could be due to the type of the tested samples, hygienic measures, isolation and identification methods, environmental conditions and the geographical location [34].
Unsurprisingly, there is a variation in the antimicrobial resistance among and within different countries, which strongly correlated with the type of prescribed drugs alongside the variation in guidelines for the use of antimicrobial drugs. In this context, the high levels of ciprofloxacin and doxycycline resistance rates detected among our tested isolates (76.7% and 84.2%, respectively) were lower than those detected in a previous study carried out in Tunisia (99.2% and 100%, respectively) [35]. Alarmingly, there has been a global warning concerning the evolution of MDR strains; however, concrete steps are being taken against the spread of both XDR and PDR strains. Herein, 75%, 20.8% and 4.2% of the tested Campylobacter isolates were recognized as MDR, XDR and PDR, respectively. This is consistent with the results of a previous study conducted in Egypt, where the above-mentioned resistance criteria were observed among 28.5%, 69% and 2.5% of the tested isolates, respectively [31]. In the current study, all Campylobacter isolates had MAR indices of 0.5 or greater. These results were higher than those recorded in a previous study conducted in South Africa, where MAR indices of the tested isolates were 0.2 or lower [36]. The high resistance rates of Campylobacter isolates in developing countries could be due to the uncontrolled usage of antimicrobials in veterinary medicine as growth promoters and in human and animal treatments without any prescription. Additionally, the high resistance rates observed in the present study to erythromycin, ciprofloxacin and doxycycline are alarming as these antibiotics are the drugs of choice used for the treatment of human campylobacteriosis, which causes a fundamental problem, where antibiotic treatment become limited. Therefore, antimicrobial usage must be controlled in animals and humans. Additionally, there is an urgent need for the wide application of alternative drugs from medicinal plants [37,38] or drug repurposing [39].
One of the best drugs available for treating human campylobacteriosis is FQ antibiotics. Many physicians describe FQs as a first-line therapy [11,12,15,17]. Unfortunately, there is widespread FQ resistance due to mutations in the gyrase gene [15]. Clinically, it is very important to find a rapid genetic method for the detection of gyrase gene mutations instead of sequencing methods. In the present study, the PCR-RFLP results for all examined isolates revealed that they all had the same fragment pattern (179 bp), which confirmed the resistance to ciprofloxacin and suggested the presence of a chromosomal point mutation in the QRDR of the gyrA gene by substitution of threonine amino acid at position 86 to isoleucine (Thr-86-to-Ile). Accordingly, there was a strong direct correlation between the MIC values of ciprofloxacin and the PCR-RFLP results. This result was in complete agreement with a previous study conducted in Brazil [15], which can help in controlling the FQ resistance and allow the proper usage of antibiotics.

Ethical Statement
The sole aim of the specimen collection in this study was to care for patients and to perform antimicrobial susceptibility testing for proper diagnosis and treatment. Therefore, it was not necessary to take the ethical approval, but prior to starting the study, the participants provided informed consent.

Sample Collection
This study was carried out between March 2017 and September 2018 (18 months) at Zagazig city, Sharkia Governorate, Egypt. A total of 210 samples were collected from chicken (n = 175) and human (n = 35) sources. Ten samples from chickens of 6 weeks of age were collected per month from multiple retail outlets (n = 16, 11 samples from each outlet) including cloacal swabs, breast muscles, liver, chicken franks and chicken luncheon meats (n = 35 each), while human stool samples were obtained as swabs from diseased children with diarrhea from private laboratories at Zagazig city. Cloacal and stool swabs were transferred directly into a sterile tube containing 9 mL of Bolton broth with Bolton broth selective supplement (Oxoid, UK) with 20 mm space lift in the tube to achieve microaerophilic conditions, while other samples were transferred into ice boxes for laboratory isolation and identification of Campylobacter spp.

Isolation and Identification of Campylobacter Species
Isolation of Campylobacter spp. was achieved according to the International Standards Organization (ISO) guidelines [40]. Briefly, the enrichment broth containing samples was incubated for 48 h at 42 • C in darkness under microaerophilic conditions (5% O 2 , 10% CO 2 and 85% N 2 ) using CampyGen sachets (Oxoid, Cheshire, UK) and anaerobic jar (Oxoid, Cheshire, UK). After that, 10 µL of the enrichment broth was streaked onto the surface of the selective modified charcoal cefoperazone deoxycholate agar (mCCDA) plates with CCDA selective supplement (Oxoid, Cheshire, UK), and the plates were incubated for 48 h at 42 • C in darkness under microaerophilic conditions. For more purification, suspicious colonies were cultivated onto blood agar (Oxoid, Cheshire, UK) supplemented with 5% sterile defibrinated sheep blood, and the plates were incubated for 48 h under microaerophilic conditions. The Campylobacter isolates were presumptively confirmed via cultural characteristics on mCCDA, Gram's staining, motility, some biochemical tests such as catalase, oxidase and indoxyl acetate and sodium hippurate hydrolysis tests and finally susceptibility to nalidixic and cephalothin.

Disc Diffusion Method
Kirby-Bauer disc diffusion method [41] was used to determine the susceptibility of all Campylobacter isolates to 24 antimicrobials belonging to 10 different antimicrobial classes that were regularly used in human and veterinary medicine in Egypt using the following antimicrobial discs ( (3)(4)(5)(6)(7)(8)(9)(10) colonies were added to a tube containing 5 mL of sterile physiological saline (0.9%) to make the bacterial suspension, which was compared with 0.5 McFarland standard solution. After that, the prepared suspension was streaked on the surface of Mueller-Hinton agar (Oxoid, Cheshire, UK) supplemented with 5% sterile defibrinated sheep blood and then the discs were placed into the plates, which were inverted and incubated at 42 • C for 24-48 h in darkness under microaerophilic conditions. The degree of susceptibility was determined by measuring the visible inhibition zones, and the results were interpreted according to the breakpoints of the Clinical Laboratory Standard Institute (CLSI) to categorize the antimicrobial agents into resistant, intermediate or susceptible [42,43].
The MDR Campylobacter isolates were defined as isolates that showed resistance to one or more antimicrobials in at least three different classes, while the isolates resistant to one agent from all antimicrobial categories with the exception of one or two categories were termed as extensively drug resistant (XDR). Meanwhile, the isolates resistant to all agents in all antimicrobial categories were described as pan drug resistant (PDR) [44]. Finally, the multiple antibiotic resistance (MAR) indices of the isolates were determined by the following equation: number of antibiotics to which the isolates were resistant/the total number of antibiotics used [45].

Broth Microdilution Method for Determining Ciprofloxacin Minimal Inhibitory Concentrations
The double-fold broth microdilution method was used to determine the MIC values of ciprofloxacin (EPPI, Cairo, Egypt) using Mueller-Hinton broth (Oxoid, Cheshire, UK) as per the CLSI guidelines [42]. Double-fold serial dilution of ciprofloxacin was done in the 96-well microtiter plates (Techno Plastic Products, Trasadingen, Switzerland) with the initial concentration of 256 µg/mL. The microtiter plates were incubated at 42 • C in darkness under microaerobic conditions for 48 h. The MIC of ciprofloxacin was defined as the lowest concentration that prevented the visible growth of Campylobacter isolates after 48 h incubation [46], and the MBC of ciprofloxacin was defined as the lowest concentration that killed 99.9% of the bacterial population after 48 h incubation [47]. Positive control was included through preparing the wells with bacterial inoculum without ciprofloxacin, while negative control was done by preparing wells containing ciprofloxacin without bacterial inoculum. Results of the broth microdilution method were demonstrated according to the CLSI breakpoints for ciprofloxacin resistance (MIC ≥ 4 µg/mL) [42]. Additionally, MIC 50 and MIC 90 of ciprofloxacin were identified as the lowest concentrations, which inhibit 50% and 90% of the examined bacterial isolates, respectively [48].

Conventional PCR and PCR-RFLP Assays
Total DNA extraction was carried out using QIAamp DNA Mini Kit (Qiagen, Chatsworth, CA, USA) according to the manufacturer's instructions. Three pairs of oligonucleotide primers (Metabion, Bayern, Germany) targeting 23S rRNA, mapA and ceuE genes were used in conventional uniplex PCR assays for verification of the genera Campylobacter, C. jejuni and C. coli, respectively. Moreover, PCR-RFLP was used to determine FQ-resistant Campylobacter isolates by detecting the mutation in QRDR of gyrA gene using FastDigest RsaI restriction enzyme (Thermo Fisher, Bremen, Germany) [15,17]. All PCR reactions were done in triplicate, using Emerald Amp GT PCR Master Mix (Takara, Kyoto, Japan). according to the manufacturer's instructions. Sequences of oligonucleotide primers used in all PCR assays are shown in Table 5 [17,49,50]. Agarose gel electrophoresis for visualization of PCR products were done as previously described [51]. Reference strains of C. jejuni (NCTC11322) and C. coli (NCTC11366) were used as positive controls, and PCR-grade water (no template DNA) was used as a negative control in PCR assays. Table 5. Sequences of oligonucleotide primers targeting four genes of Campylobacter species and their PCR-amplified products.

Statistical Analysis
The data were analyzed using SPSS version 26 (IBM Corp, Armonk, NY, USA). The chi-square test was used to study the variations in the prevalence of Campylobacter spp. from different origins and to assess the differences in the antimicrobial resistance patterns of the recovered isolates from various sources and among C. jejuni and C. coli. The p values were considered statistically significant if they were less than 0.05.

Conclusions
This study revealed a high prevalence of MDR Campylobacter spp. in addition to the existence of both XDR and PDR strains among human and chicken isolates in Egypt. Therefore, there is an urgent need to control the antimicrobial usage in animal production and to improve the hygienic control strategies during slaughtering and carcass processing to reduce the occurrence of resistant Campylobacter strains. Interestingly, PCR-RFLP technique was found to be helpful in the detection of FQ-resistant Campylobacter isolates, which offer useful insights into the molecular mechanism involved in the FQ resistance in Campylobacter spp., which will help public health specialists in the management of Campylobacter infections. Furthermore, enhanced research efforts are required to illustrate the mechanisms of the transmission and persistence of FQ-resistant Campylobacter spp. alongside different hosts. Complementary to that, it is essential to provide a solid guideline for the administration of FQ antimicrobials especially in the veterinary field, which will decrease the wide spreading of human FQ-resistant Campylobacter isolates. A major limitation for our approach is the need for DNA sequencing to confirm the results of the mutation in QRDR of gyrA gene.