β-Lactamase Producing Escherichia coli Encoding blaCTX-M and blaCMY Genes in Chicken Carcasses from Egypt

Escherichia coli with multidrug resistance and β-lactamase genes may constitute a great public health hazard due to the potential for their transmission to humans through the food chain. This study determined the prevalence, antibiotic resistance profiles, phylogroups, and β-lactamase genes of E. coli isolates from chicken carcasses marketed in Mansoura, Egypt. Interestingly, E. coli was detected in 98% (98/100) of the chicken carcasses examined, which seemed among the highest contamination rates by E. coli worldwide. From the 425 genetically verified uidA gene-positive E. coli, 85 isolates were further studied for antimicrobial resistance profiles, phylogroups, and β-lactamase genes. Interestingly, 89.41% of E. coli (76/85) strains tested against 24 different antibiotics were multidrug-resistant. Of the examined 85 E. coli isolates, 22 (25.88%) isolates harbored blaCTX-M and were resistant to ampicillin, cefazoline, and ceftriaxone, while three of them were resistant to ceftazidime besides. Nine (10.59%) E. coli strains harbored AmpC- β-lactamase blaCMY and were resistant to ampicillin. One isolate co-carried blaCMY and blaCTX-M genes, though it was negative for the blaTEM gene. Of the 35 isolates that harbored either extended-spectrum β-lactamase (ESBL) and/or AmpC β-lactamase genes, six strains (17.14%) were assigned to pathogenic phylogroup F and one to phylogroup E, whereas 28 (80%) isolates belonged to commensal phylogenetic groups.


Introduction
Antimicrobial resistance is one of the universal threats which affect human and animal health especially in developing countries, including Egypt, due to a lack of rules controlling the usage of antimicrobials in different sectors of life (animal, agriculture, as well as human) [1,2]. Food-producing animals usually receive antimicrobials as a treatment or prophylactic courses to protect them from many bacterial infections [3,4]. Such overuse and misuse of antimicrobials results in the production of highly resistant bacteria; even commensal bacteria can acquire such resistance owing to chromosomal mutation or via gaining resistance traits from mobile genetic elements (e.g., plasmids, integrons, and transposons) [5]. and transferred in an icebox to the Laboratory of Food Hygiene and Control Department, Faculty of Veterinary Medicine, Mansoura University, wherein the microbiological analysis was performed.

Isolation and Identification of Escherichia coli
Each whole poultry carcass was separately rinsed with 225 mL of sterile tryptone soya broth (CM0989; Oxoid, Hampshire, UK) and the suspension obtained was incubated at 37 • C, for [18][19][20][21][22][23][24] h. Enriched cultures were plated onto sterile MacConkey agar (CM1169; Oxoid, Hampshire, UK), then the plates were incubated for 24 h at 37 • C. Up to five typical presumptive E. coli colonies were sub-cultured onto nutrient agar slopes and incubated for 24 h at 37 • C for further biochemical and molecular identification. A total of 600 typical colonies were subjected to various biochemical tests for further identification and confirmation of E. coli.

Molecular Identification of E. coli
For molecular confirmation, the biochemically positive E. coli isolates (n = 434) were subjected to PCR for amplification of the uidA gene, the gene-specific for pathogenic E coli [27]. For chromosomal DNA isolation, 5 mL from the bacterial culture in nutrient broth (CM0001; Oxoid, Hampshire, UK) was incubated overnight at 37 • C, followed by centrifugation at 3000 rpm for 15 min. The supernatant was removed and the pellet was resuspended in 1 mL nuclease-free water, homogenized, transferred to a 1.5-mL Eppendorf tube, vortexed, and placed in a heat block at 70 • C for 15 min. The tubes containing the heated lysate were then centrifuged and the resultant supernatant was used as a DNA template.
Escherichia coli strains isolated from examined chicken carcasses were screened for the presence of the uidA gene using a primer set for the amplification of the uidA gene with the following sequence 5 -ATGCCAGTCCAGCGTTTTTGC-3 and 5 -AAAGTGTGGGTCAATA ATCAGGAAGTG-3 for sense and antisense, respectively. The primer set can amplify a DNA of 1487 bp (Eurofins Genomics ® , St. Charles, MO, USA). PCR amplification was carried out in a 20-µL reaction mixture using a ready-to-use solution of GoTaq Green Master Mix, (Promega Corporation ® , Madison, WI, USA) supplied in 2× Green GoTaq reaction buffer (pH 8.5). PCR reaction mixture consisted of 10 µL GoTaq Master Mix 2×, 1 µL (10 pmol) from each of sense and antisense primers, 2 µL DNA template, and 6 µL autoclaved water. The mixture was subjected to 30 cycles of amplification in a Gene Amp PCR system 2700 (Applied Biosystems ® , Foster City, CA, USA). The first cycle was preceded by denaturation at 95 • C for 5 min. Each cycle performed for the uidA gene consisted of denaturation for 45 s at 95 • C, annealing for 45 s at 63 • C, and extension at 72 • C for 1 min, followed by a final extension at 72 • C for 5 min by the end of the last cycle, then cooling at 4 • C. The PCR products were separated through running agarose gel (1.5% agarose) electrophoresis after loading 5 µL of the PCR mixture for 35 min at 100 V. The amplified DNA was then stained in an ethidium bromide solution (Sigma-Aldrich ® Co., St. Louis, MO, USA), followed by visualization and photographing under ultraviolet light.
Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, Enterococcus faecalis ATCC 29212, and Staphylococcus aureus ATCC 29213 were used as the control for the determination of MIC.

Prevalence of E. coli in the Chicken Samples Examined
Escherichia coli was molecularly detected at a very high prevalence rate of 98% (98/100) among the 100 chicken carcasses tested. A total of 600 colonies (circular bright, pink-colored colonies) were selected from the 100 chicken carcasses, based on their morphological identification. Among these 600 colonies, 434 isolates were confirmed biochemically (production of indole from tryptophan, Methyl red test, Voges-Proskauer test, Citrate utilization test, Urease test) as E. coli and were further subjected to molecular confirmation by PCR for detection of the uidA gene: the E. coli marker gene. Interestingly, 97.92% (425/434) of the biochemically identified isolates were positive for the uidA gene (Figure 1), indicating that the biochemical tests can be used as an accurate tool for the identification of E. coli. A supplementary figure for the amplified DNA product of uidA gene for molecular verification of E. coli is presented in Figure S1.
Madison, WI, USA). A reaction mixture of 20-µL volume consisted of 10 µL GoTaq Green Master Mix, 2 µL of DNA template, and 1 µL (20 pmol) for each of the sense and antisense primers of the four genes. The PCR was performed using a Perkin-Elmer GeneAmp 9600 Thermal Cycler in which the mixture was subjected to 30 cycles of amplification. The annealing temperatures for the different amplified genes were described in Table 1. The electrophoresis of PCR products was mentioned before.

Prevalence of E. coli in the Chicken Samples Examined
Escherichia coli was molecularly detected at a very high prevalence rate of 98% (98/100) among the 100 chicken carcasses tested. A total of 600 colonies (circular bright, pink-colored colonies) were selected from the 100 chicken carcasses, based on their morphological identification. Among these 600 colonies, 434 isolates were confirmed biochemically (production of indole from tryptophan, Methyl red test, Voges-Proskauer test, Citrate utilization test, Urease test) as E. coli and were further subjected to molecular confirmation by PCR for detection of the uidA gene: the E. coli marker gene. Interestingly, 97.92% (425/434) of the biochemically identified isolates were positive for the uidA gene ( Figure 1), indicating that the biochemical tests can be used as an accurate tool for the identification of E. coli. A supplementary figure for the amplified DNA product of uidA gene for molecular verification of E. coli is presented in Figure S1. The uidA gene is very specific to E. coli, especially the pathogenic strains associated with foodborne infection outbreaks, and is frequently identified by the redundant detection of uidA by PCR. The occurrence of E. coli at a high percentage in 98% of poultry carcass samples indicates poor hygienic measures during poultry carcass processing. The uidA gene is very specific to E. coli, especially the pathogenic strains associated with foodborne infection outbreaks, and is frequently identified by the redundant detection of uidA by PCR. The occurrence of E. coli at a high percentage in 98% of poultry carcass samples indicates poor hygienic measures during poultry carcass processing.
In developing countries, including Egypt, the use of conventional methods of slaughtering and evisceration resulted in poultry exposure to high contamination levels with different types of bacteria such as E. coli, which is considered a normal inhabitant of the intestinal tract of the live birds [36]. The present study is considered among those which showed the highest prevalence rate of E. coli in poultry carcasses. Likewise, high E. coli prevalence rates of 93.7% [37] and 90% [38] were reported in raw poultry carcasses in Egypt, and also by rates of 89% in Qatar [39], 87.5% in Turkey [40], and 60% in Bangladesh [41]. Such a high prevalence of E. coli could result in the production of inferior quality poultry meat with subsequent economic losses and public health hazards of great concern.
On the contrary, lower contamination rates of poultry meat by E. coli organisms had been recorded in former studies performed Egypt with an incidence of 37.7% [42], 35% [43], and 11.7% [25]. Lower E. coli contamination rates in poultry meat were also recorded worldwide. For instance, prevalence rates of 39.76%, 38.7%, and 40.82% were reported in India [44], the USA [45], and Bangladesh [46], respectively. The variations in the E. coli contamination rates among different studies depends on the hygienic facilities available through the poultry processing system. Therefore, strict hygienic measures during poultry carcass processing should be adopted.
In the current study, the higher resistances rates of the E. coli strains isolated from poultry meat against tetracycline and ampicillin were in concordance with previous studies conducted in Egypt; in which 94.7% and 100% [47], as well as 80.9% and 71.4% [25], of E. coli isolates recovered from poultry meat were resistant to tetracycline and ampicillin, respectively. A similar pattern of antibiotic resistance of E. coli strains isolated from poultry meat had been recorded in previous studies in other countries [48,49], indicating the selective pressure of these antimicrobials in the treatment of E. coli infections in poultry due to the rational use of antimicrobial drugs in poultry production for treatment, subtherapeutic or prophylactic purposes, in addition to increasing productivity [3,50,51]. Therefore, a strict system should be established to control the use of antimicrobials not only for the animal, but also in the agricultural sectors.
Among the 85 E. coli isolates examined, 76 isolates (89.41%) were resistant to antimicrobials of three different classes or more, making them categorized as multidrug-resistant strains. Multidrug resistance (MDR) among E. coli strains isolated from poultry in previous studies exhibited a high prevalence of 69.1% [52] and 83% [53] of the E. coli isolates. The high prevalence of resistance patterns against three or more classes of antimicrobials could be related to the different antibiotic regimens used for the different livestock species [54,55]. In fact, it is nowadays accepted that the overuse of antibiotics in animals and poultry is the main driver for the dissemination of multi-drug resistance [56]. It is estimated that there is more extensive antibiotic use in livestock and poultry than in human medicine. The use of antimicrobials in animal production is associated with the emergence and spread of AMR in food-related bacteria and leads to the selection of antimicrobial resistance among pathogenic and commensal bacteria in the intestinal tract of food animals. Therefore, resistant bacteria can contaminate food products and colonize the human microbiota via the food chain by the handling and/or consumption of contaminated foods [57,58].
The present study indicated extensive contamination of chicken carcasses examined with multidrug-resistant E. coli that may constitute a tremendous threat to public health. The competent authorities should therefore enforce antimicrobial resistance (AMR) regulation to confirm the cautious use of antimicrobials to reduce the risk of transmission of antimicrobial-resistant organisms via the food chain. Farmers should be prevented from the unsystematic use of antimicrobials in poultry production and promoted to implement preventive measures by observing biosecurity in addition to good management practices.
Extended-spectrum β-lactamase (ESBL) and AmpC β-lactamase E. coli exhibited resistance against third-generation cephalosporins (ceftriaxone and ceftazidime) antibiotics detected in 35 (41.18%) E. coli isolates recovered from poultry meat in this study. Such prevalence of ESBL-producing E. coli is consistent with the prevalence rates reported in previous studies on ESBL E. coli in chicken meat in which ESBL-producing E. coli was recorded in Ghana with a prevalence rate of 52.8% [59], as well as in Germany, with a rate of 38.7% [60]. On the contrary to the findings of the current study, a lower prevalence of ESBL had been isolated from poultry meat in Tanzania at a rate of 20.1% [61] and in Pakistan by rates of 7.76% [62]. It has been indicated that among 52.9% AmpC-producing Enterobacteria, most isolates were identified as Escherichia coli [60]. It is noteworthy that many publications found that extended-spectrum β-lactamase (ESBL) and AmpC-producing E. coli present in humans and animals mostly shared identical sequence types (STs), suggesting the transmission of such resistant genes and human infections with ESBL-producing E. coli of animal source [63].

Determination of β-Lactamase Genes
The β-lactamase encoding gene bla TEM conferring resistance to penicillins was detected in 67.06% (57/85) of the E. coli isolated from the poultry carcasses. Twenty-six (25.88%) of the E. coli strains harbored bla CTX-M and were resistant to ampicillin, cefazoline, and ceftriaxone, while three of them were resistant to ceftazidime as well ( Table 2). Ten (11.76%) E. coli strains harbored AmpC-β-lactamase bla CMY and were resistant to ampicillin. Six of them exhibited resistance to cefazoline and ceftriaxone harboring bla TEM , while two among these ten strains were also resistant to ceftazidime (Table 3). Six strains of the bla CMY and the bla CTX-M harboring E. coli strains were holding bla TEM genes (Tables 2 and 3). One strain was holding both bla CMY and bla CTX-M genes and was negative for bla TEM . No E. coli strains were positive for bla SHV and bla OXA. The PCR-amplified DNA products of each of bla TEM , bla CTX-M , and bla CMY genes are presented in a supplementary figure ( Figure S2).  The minimum inhibitory concentration (MIC) breakpoints follow the CLSI [28] and EUCAST [29] guidelines. The phylogrouping was detected by the quadruplex phylotyping PCR method according to Clermont et al. [34]. R: resistant. ND: not detected. It is well known that among the β-lactamase genes, bla TEM is the most predominant one that is widely spread in Gram-negative bacteria, encoding enzymes that hydrolyze penicillin and first-generation cephalosporins. In accordance with our findings concerning the predominance (67.06%) of bla TEM , previous studies confirmed such predominancy with prevalence rates of 92% [64] and 80% [65] in E. coli isolated from poultry. On the contrary, bla CTX-M was the predominant (58.1%) ESBL gene detected in E. coli recovered from chicken meat in the Netherlands [66]. Several studies in Egypt confirmed the spreading of bla CTX-M in different sources including food [14,25]. The bla CTX-M gene was formerly identified in ESBL-producing E. coli isolates from chicken and beef samples examined in Egypt [52]. Likewise, five ESBL E. coli strains harboring bla CTX-M were isolated from raw and ready-to-eat beef products [67], with the same bla CTX-M gene prevalence recorded in the present study. Globally, bla CTX-M was revealed as the predominant ESBL gene in ESBL-producing isolates [21], as major public health importance pathogen [68].
The AmpC β-lactamase bla CMY had been previously detected in six E. coli strains isolated from colibacillosis-infected chickens in Egypt [69], although this gene could not be detected in E. coli recovered from sound poultry meat [25]. It has been demonstrated that the first isolation of E. coli strains encoding the AmpC-resistant gene bla CMY in Egypt was recovered from hospitalized patients with urinary tract infections in three Egyptian hospitals [70]. Our study is the first that recovered E. coli strains harboring the bla CMY gene from sound (non-infected) chicken meat in Egypt.
All the ESBL and AmpC-producing E. coli strains isolated from poultry carcasses in the current study showed resistance against further antibiotics of different antimicrobial classes other than β-lactam. In this context, previous studies indicated that ESBL-and AmpCproducing bacteria are frequently cross-resistant to other antimicrobials such as aminoglycosides, chloramphenicol, tetracyclines, sulfonamides, trimethoprim, or quinolones, usually due to the existence of different resistance genes on mobile genetic elements such as plasmids, transposons, or integrons [71][72][73].

Phylogroup Characterization of Isolated E. coli Strains
The frequency distributions of different E. coli phylogroups among the tested E. coli isolates from chicken carcasses are illustrated in Figure 3. The AmpC β-lactamase blaCMY had been previously detected in six E. coli strains isolated from colibacillosis-infected chickens in Egypt [69], although this gene could not be detected in E. coli recovered from sound poultry meat [25]. It has been demonstrated that the first isolation of E. coli strains encoding the AmpC-resistant gene blaCMY in Egypt was recovered from hospitalized patients with urinary tract infections in three Egyptian hospitals [70]. Our study is the first that recovered E. coli strains harboring the blaCMY gene from sound (non-infected) chicken meat in Egypt.
All the ESBL and AmpC-producing E. coli strains isolated from poultry carcasses in the current study showed resistance against further antibiotics of different antimicrobial classes other than β-lactam. In this context, previous studies indicated that ESBL-and AmpC-producing bacteria are frequently cross-resistant to other antimicrobials such as aminoglycosides, chloramphenicol, tetracyclines, sulfonamides, trimethoprim, or quinolones, usually due to the existence of different resistance genes on mobile genetic elements such as plasmids, transposons, or integrons [71][72][73].

Phylogroup Characterization of Isolated E. coli Strains
The frequency distributions of different E. coli phylogroups among the tested E. coli isolates from chicken carcasses are illustrated in Figure 3.  Among the 85 E. coli isolates, the majority (70/85; 82.35%) were identified as commensal strains with a phylogroup of B1, A, or C, divided into phylogroup B1 (34/85; 40%), followed by phylogroup A (26/85; 30.59%), and phylogroup C (10/85; 11.76%). The pathogenic phylogroup E. coli was detected in 13 (15.29%) of the 85 E. coli isolates, which were distributed as 7 (8.24%) phylogroup F, 3 (3.52%) for each of D and E phylogroups, and only 1% was Clade I or II, with an absence of phylotype B2. The amplification of the genes incorporated in the phylogroup typing of the confirmed E. coli strains is presented as a supplementary figure ( Figure S3) showing a representative agarose gel electrophoresis of the DNA for the arpA, chuA, yjaA, and trpA (quadruplex PCR), along with TspE4C2, trpA (group C) (singleplex), and arpA (group E).
Similar to the current study, the predominant phylogroup among the E. coli strains detected in previous studies in retail chicken samples worldwide were phylogroup B1 followed by phylogroup A, such as in Brazil [74,75], Italy [76], Ghana [51], and Pakistan [77], as well as in Egypt in poultry in which phylogroup B1 was the predominant phylogroup among the E. coli isolates (74%), followed by phylogroup C (20%), and then pathogenic phylogroup D (4%), while there were no E. coli strains possessing phylotypes A, B2, E and F [26].
Of the 35 E. coli which exhibited resistance against third-generation cephalosporin (ceftriaxone and ceftazidime) antibiotics in this study, six (17.14%) strains belonged to pathogenic phylogroup F and one to phylogroup E. However, the other 28 (80%) were considered as commensal E. coli strains belonging to commensal phylogenetic groups. Concerning our E. coli isolates that harbor bla CTX-M , 22.72% were phylogroup A, 50% belonged to phylogroup B1, while 13.63% belonged to each of phylogroup C and phylogroup F (Table 4). Similarly, 22.22% of bla CMY -positive Escherichia coli isolates were phylogroup A, while 33.33% were categorized as phylogroup B1 and F, whereas 11.11% only belonged to phylogroup C (Table 4). It has been confirmed that the predominant phylogenetic grouping of extended-spectrum β-lactamases and AmpC-producing E. coli, which exhibited resistance against third-generation cephalosporins and were isolated from broiler chickens, were mainly type B1 and A (commensal phylogroup), followed by type D and B2 (pathogenic phylogroup) [60,78]. However, all (100%) ESBL E. coli isolated from diseased poultry samples in a previous study conducted in Egypt, belonged to phylogroup D, while most (66%) of the AmpC strains belonged to phylogroup B1 [69], which is quite opposite to that reported in our study.
The results of the current study might lead to the assumption that resistant group F isolates from chicken carcasses may contribute to the dissemination of antimicrobial resistance in humans, resulting in limited treatment of E. coli infection. It has been reported that E. coli belonging to phylogroup F that were isolated from chicken samples were closely related to the extraintestinal pathogenic E. coli causing human infections [79].
Above and beyond the fact that urinary tract infections (UTIs) are among the most common infections both in community and hospitals, they are most frequently caused by multidrug-resistant E. coli that challenge UTI treatment. Although the wide spreading of the phylogroups B2, F, and D pathogenic E. coli strains cause extraintestinal infection, the commensal E. coli strains belonging to the most newly diverged phylogroups are also responsible for severe intestinal infections.

Conclusions
The present study reported high contamination rates of chicken carcasses with multidrugresistant E. coli contaminants, along with the isolation of pathogenic E. coli isolates harboring extended-spectrum β-lactamases and AmpC-β-lactamases-encoding genes bla CTX-M and bla CMY . To our knowledge, this is the first study isolating bla CMY from sound chicken meat for human consumption in Egypt. The occurrence of E. coli in almost all of chicken samples indicates poor hygiene and highlights the antimicrobial resistance problem in Egypt caused by the rational use of antimicrobials in animal husbandry and calls for a nationwide surveillance program to monitor antimicrobial resistance. These findings provide evidence that healthy broilers in Egypt could be a source for the dissemination of transmissible resistance mechanisms among foodborne pathogens brought from the unhygienic environment of the food chain during slaughtering. Although the majority of the isolated strains were commensal E. coli, about 16% of the strains were pathogenic, while 7% were pathogenic and categorized as a phylogroup F that harbored plasmid-mediated antibiotic-resistant genes, which may lead to infections via consumption of contaminated food with no treatment in humans. In addition, the commensal E. coli strains which harbor antibiotic-resistance genes, which are mainly plasmid-mediated strains, can be easily transmitted to further pathogenic bacteria that may lead to human infection with no treatment.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/foods12030598/s1, Figure S1: A representative agarose gel electrophoresis showing the amplified DNA product of the uidA gene (the marker gene for E. coli) at the expected molecular weight of 1487 bp. Genomic DNA from E. coli isolates was used as a template; Figure S2 Funding: This research paper is supported by the project "Increasing the impact of excellence research on the capacity for innovation and technology transfer within USAMVB Timis, oara" code 6PFE, submitted in the competition Program 1-Development of the national system of research-development, Subprogram 1.2-Institutional performance, Institutional development projects-Development projects of excellence in RDI.

Data Availability Statement:
Data is contained within the article or supplementary material.

Conflicts of Interest:
The authors declare no conflict of interest.