Prevalence of Multidrug-Resistant Salmonella enterica Serovars in Buffalo Meat in Egypt

The current study aimed to investigate the presence of Salmonella spp. prevalence in buffalo meat in Egypt, along with studying the antimicrobial susceptibility of the recovered isolates. Salmonella spp. was detected in 25% of tested buffalo meat. A total of 53 (100%) isolates were genetically verified by PCR as Salmonella, based on the detection of the invA gene. The stn and hilA genes were detected in 71.7% (38/53), and 83.0% (44/53) of the recovered isolates, respectively. Salmonella Enteritidis (11/53; 20.7%) was the most commonly isolated serovar, followed by S. Typhimurium (9/53; 17%), S. Montevideo (6/53; 11.3%), meanwhile, S. Chester, S. Derby, S. Papuana, and S. Saintpaul were the least commonly identified serovars (a single strain for each; 1.9%). Among the 16 antimicrobials tested, amikacin, imipenem, gentamicin, cefotaxime, meropenem, ciprofloxacin, and enrofloxacin were the most effective drugs, with bacterial susceptibility percentages of 98.1%, 94.3%, 92.5%, 86.8%, 83.0%, 73.6%, and 69.8%, respectively. Meanwhile, the least effective ones were erythromycin, streptomycin, clindamycin, cefepime, and nalidixic acid, with bacterial resistance percentages of 100%, 98.1%, 88.7%, 77.4%, and 66%, respectively. Interestingly, the high contamination level of Egyptian buffalo meat with multidrug-resistant Salmonella (79.2%; 42/53) can constitute a problem for public health. Therefore, programs to control Salmonella contamination are needed in Egypt.


Introduction
Buffalo is a good source of meat, especially in areas where the climate adversely affects the production efficiency of other animals. The demand for buffalo meat appears to be increasing, especially in arid regions because of its unique physiological characteristics strains, particularly S. Typhimurium and S. Enteritidis isolated from different animal origin foods worldwide [13,14]. The increased prevalence rates of antimicrobial-resistant microorganisms during the course of food production is possible due to the extensive use of the common antimicrobials (particularly overuse or misuse) as growth promoters, prophylactic, or therapeutics, in food-producing animals [15]. The emergence of MDR Salmonella has become a global public health problem, as it gravitates toward more virulence [9].
To date, limited information is available concerning the Salmonella spp. isolation from Egyptian buffalo meat [16,17]. Therefore, considering the increased marketing and consumption of buffalo meat in Egypt, and the global increase in the prevalence of MDR Salmonella isolates against the most ordinarily used antimicrobials, this work was planned to study the prevalence, serotyping, virulence genes existence and the phenotypic antimicrobial resistance profile of Salmonella isolates from buffalo meat distributed in Mansoura, Egypt.

Sample Collection
A total of 100 buffalo meat samples (each sample represented by 250 g) were collected from the retail butcher shops located in Mansoura city, Egypt, from November 2020 to June 2021. Each sample was individually wrapped into a sterile bag, labeled, and then transferred rapidly in a refrigerated box at ∼4 • C to the Laboratory of Food Hygiene and Control Department, from the Faculty of Veterinary Medicine, Mansoura University, Egypt, for bacteriological analyses. The specified mass of the test meat portion used for isolation of Salmonella spp. was represented by a quantity of 25 g, which was sampled aseptically, using a sterile scalpel, from each tested sample.

Isolation and Identification of Salmonella spp.
The detection of Salmonella spp. was performed according to ISO [18]. In brief, 25 g meat portion from each buffalo meat sample were aseptically excised and introduced into a sterile homogenizer flask containing 225 mL sterile buffered peptone water (CM0509B; Oxoid Ltd., Basingstoke, UK), and thoroughly stirred for one min, in a stomacher (Stomacher 400 Lab Blender; Seward Medical, London, UK). Each mixed meat sample was then incubated at 37 • C for 24 h. From pre-enriched culture, volumes of 0.1 mL and 1 mL, respectively were aseptically inoculated into 10 mL each of Rappaport-Vassiliadis Salmonella enrichment broth (RV; CM0669; Oxoid Ltd., Basingstoke, UK) and Muller-Kauffmann tetrathionate/novobiocin broth (MKTTn; CM1048; Oxoid Ltd., Basingstoke, UK), respectively, followed by incubation at 42 • C for 24 h, and 37 • C for 24 h, consecutively. A loopful from RV and MKTTn broths was streaked onto two selective solid media; Xylose-Lysine-Desoxycholate (XLD) agar (CM0469; Oxoid Ltd., Basingstoke, UK) and Brilliant Green Agar w/Sulfadiazine (BGA; NC7310; Neogen Ltd., Maharashtra, India) plates, which were incubated for 24 h at 37 • C, and at 35 • C, respectively. Approximately five typical suspected colonies of presumptive Salmonella (pink color, with or without black centers), were picked up and purified onto nutrient agar slopes, then incubated for 24 h at 37 • C, for biochemical and serological confirmation.
As has been previously indicated by several authors [3,9,10,12,16,17], it is important to mention that more than one Salmonella isolate was tested from a single positive meat sample for the following reasons: (i) meat samples can contain more than one Salmonella serovars, (ii) two or more of the same Salmonella serovars can exhibit different antimicrobial resistant patterns, (iii) more than one isolates of the same or different Salmonella serovar taken from a single positive sample can exhibit different existence profiles of the virulence genes tested.

Molecular Analysis
The chromosomal DNA was isolated according to Abd-Elghany and Sallam [19], using the QIAamp DNA Mini Purification kit (Qiagen, Germany), and following the producer's recommendations.
Genomic DNA of Salmonella Typhimurium (RIMD 1985009) and Escherichia coli K12DH5α acquired from National Research Centre (NRC), Dokki, Cairo, Egypt was served as positive and negative control reference strains, respectively, for the presence and absence of invA, stn, and hilA genes. Phenotypically-identified Salmonella isolates, detected from examined buffalo meat samples, were molecularly identified by using the multiplex polymerase chain reaction assays, based on the presence of invA marker gene targeting for invasion of Salmonella in the host organism, using specific oligonucleotide primer sequences constructed to amplify DNA amplicon of 275 bp for the invA gene [20]. The identified strains were then tested for the detection of stn and hilA virulence genes, using specific oligonucleotide primer sequences designed for PCR amplification of DNA fragments at a molecular size of 617 bp for the stn gene [21] and 854 bp for the hilA gene [22].
Multiplex PCR was carried out using a thermal cycler (Eppendorf Mastercycler, Hamburg, Germany), in a 50-µL volume, which includes 1 µL Salmonella genomic DNA template, 2 µL each of forward and reverse primers (10 pmol each), 25 µL DreamTaq Green Master Mix (Thermo Scientific, St. Leon Roth, Germany), and 20 µL of sterile distilled water. The denaturation step at 94 • C for 4 min, 35 cycles (94 • C, for 45 s, 56 • C, for 60 s, 72 • C, for 60 s) was followed by a final extension, at 72 • C, for 7 min. The PCR-amplified products of each reaction mixture were separated by subjecting 4 µL aliquots to agarose (1.5%) gel electrophoresis (Cleaver Scientific-Horizontal Electrophoresis System; Cleaver Scientific Ltd., Rugby, UK), for 30 min, at 100 V, followed by 25-min staining in 1% solution of ethidium bromide solution. Finally, the separated PCR products were visualized and photographed under UV illumination. Amplified genes were checked by DNA sequencing with the BigDye Terminator version 3.1 Cycle Sequencing Kit, following the producer's recommendations on an ABI Prism 3100 automated sequencer (Applied Biosystems), using the same primers as those used for gene amplification.
PCR-verified Salmonella isolates were then subjected to slide and tube agglutination technique for Salmonella serotyping based on the detection of somatic (O), and flagellar (H) antigens by using separated O and H Salmonella antisera (Denka Seiken Co., Tokyo, Japan) according to the White-Kauffmann-Le Minor scheme.

Antimicrobial Susceptibility Tests
The antimicrobial susceptibility of the Salmonella strains was evaluated using the agar disk diffusion standard method according to the Clinical and Laboratory Standards Institute guidelines [23]. Antimicrobial susceptibility discs (Difco Laboratories, and BioMerieux,  [23]. Escherichia coli ATCC 25922 was used as a positive control strain. Multiple antibiotic resistances (MARs) index for each resistance pattern was calculated using the formula MAR Index = Number of resistance antimicrobials/total number of antimicrobials tested; where a MAR index > 0.2 reveals high-risk contamination [24].

Prevalence of the Isolated Salmonella enterica Serovars in Buffalo Meat
The current study indicated that 25 (25%) of the 100 buffalo meat samples monitored were contaminated with Salmonella spp. based on the molecular confirmation of the existence of the Salmonella marker gene; the invA gene by the PCR technique, which was conducted on the genome of the presumptive Salmonella isolates.
Variable detection rates of Salmonella spp. in buffalo meat had been obtained by several research groups worldwide, using conventional microbiological methods. Our results are in close agreement with those published by Mezali and Hamdi [25], in Algeria, who could isolate Salmonella spp. from 23.61% of raw red meat and different meat products. Higher incidences of Salmonella spp., of 80%, and 46.67%, however, were recorded in buffalo meat in Laos [26], and Bangladesh [27], respectively. Conversely, lower incidences of 7% [28], 7.11% [29], 18% [30], 10.66% [31], and 7.4% [32] were reported for Salmonella in buffalo meat samples from Iran, Laos, Egypt, India, and Nepal, respectively.
The broad variation in Salmonella prevalence in the examined samples from different investigations could be related to several factors including the seasonal effects, geographical provenience, the used sampling, and bacteriological techniques, as well as the degree of the slaughter hygiene and the occurrence of products cross-contamination during different stages of buffalo dressing and preparation [33].

Distribution of Salmonella serovars Isolated from Buffalo Meat Samples
There were 14 different Salmonella serovars identified among the isolates (Figure 1). Out of them, Salmonella Enteritidis (11/53; 20.7%) was the most frequently encountered, followed by S. Typhimurium (9/53; 17%), S. Montevideo (6/53; 11.3%), meanwhile, S. Derby, S. Saintpaul, S. Papuana, and S. Chester were the least commonly identified serovars (a single strain for each 1.9%) ( Figure 1). The dominant occurrence of S. Enteritidis and S. Typhimurium serovars in the present investigation was also pointed out by the results of other several studies of foodborne salmonellosis outbreaks, in which S. Enteriditis and S. Typhimurium were isolated at high rates from the examined buffalo meat samples, from different countries [28,[34][35][36]. In agreement with the results of some previously conducted investigations [16,17], the current study strengthens the fact that Egyptian buffalo meat could be considered a major potential source of human salmonellosis.

PCR Confirmation of the Salmonella Isolates
Multiplex PCR (Figure 2) method was carried out to target the conserved regions of Salmonella spp., such as the invA, stn, and hilA genes [9,37,38]. PCR amplification results revealed that only 53 (65.4%) out of the 81 biochemically identified isolates from the examined buffalo meat specimens were confirmed as Salmonella spp. by the invAtargeted gene. The present study indicated that all (100%; 53/53) serologically identified Salmonella isolates (n = 53) examined were molecularly confirmed by multiplex PCR to be contaminated with Salmonella spp., based on the presence of the invA gene (275 bp   The distribution of the virulence genes detected among the different Salmonella serovars identified is fully described in Table 1.  The invA gene is one of the most widely used genetic markers for the detection of Salmonella spp. in food origin samples, which is responsible for invasion in the host cells [39]. Its presence is strongly associated with the occurrence of other virulence genes, such as the stn gene, with important contributions to the pathogenicity process, that causes an enterotoxin effect on host cells, leading to gastroenteritis with diarrhea [40], and the hilA gene, which participated in the adhesion and invasion processes of the pathogen to the host cells, and increases Salmonella pathogenicity [39]. The presence of these genes aids the organisms to interact with the host cells and may indicate the virulence potential of Salmonella spp. The PCR product specificity was evaluated by sequencing the amplified invA, stn, and hilA fragments. The sequenced amplicons had 100% similarity with the analogous regions of invA, stn, and hilA genes available in the GenBank ® . Our finding of the invA gene among Salmonella serovars was nearly similar to those obtained by Sallam et al. [40], in Egypt, Thung et al. [41], in Malaysia, and Dong et al. [42], in China, who reported that all (100%) of the tested Salmonella isolates were positive for invA gene-specific for Salmonella spp. Interestingly, Yanestria et al. (2019) reported the detection of the invA gene in only 12.5% of the tested milkfish-origin Salmonella isolates [43]. In the same context, the high identification rate of the hilA gene (78.5%) among the tested Salmonella isolates in the present study was compatible with that reported by Thung et al. [41], who reported the presence of hilA gene in 82.6% of recovered Salmonella isolates from beef samples in Malaysia. Meanwhile, the detection rate of the stn gene in our Salmonella isolates (71.4%) was lower than that recorded in a previously conducted investigation by Sallam et al. [40], who detected the stn gene in all (100%) examined Salmonella isolates from beef samples in Egypt.
Results of previously conducted studies suggested that the PCR technique, targeting the amplification of the invA gene, showed better specificity, sensitivity, and accuracy compared with conventional methods (e.g., cultural on XLD plates and biochemical methods), besides its capacity in identifying Salmonella isolates from buffalo meat specimens within few hours. Therefore, those characteristics make PCR techniques a suitable and complementary method for confirmation and identification of Salmonella isolates from the examined samples [39,40].

Antimicrobial Resistance of Salmonella Isolates
The results of the disc diffusion test conducted for antibiotic resistance of buffalo meat origin Salmonella isolates indicated that out of 16 tested antimicrobials, the most effective drugs were AMK, IPM, GEN, CTX, MEM, CIP, ENR, TE, and AMP, which exhibited bacterial sensitivity percentages of 98.1%, 94.3%, 92.5%, 86.8%, 83.0%, 73.6%, 69.8%, 58.5%, and 56.6%, respectively. The least effective antimicrobials against the tested Salmonella isolates were E, S, and CM, with bacterial resistance percentages of 100%, 98.1%, and 94.3%, respectively ( Table 2). The high resistance rates of Salmonella isolates against the aforementioned antimicrobials, which are widely used in human as well as veterinary medicine in Egypt are probably due to their over-usage as therapeutics, prophylactic agents, or growth promoters in the livestock veterinary medicine of the screened region [9,16,44,45]. Although restrictive measures are applied to reduce the prevalence of Salmonella Typhimurium and S. Enteritidis in different countries, they are not healthcare providers for reportable diseases in Egypt. Unfortunately, Salmonella infections are therapeutically treated in Egypt, and such a way is not sufficient to clean hosts from Salmonella spp., which can remain in live stocks, often in latent form. The highest resistance rate of Salmonella isolates against E (100%), and NA (66.0%) in our study is in accordance with that obtained by Sallam et al. [40], in Egypt, who found a high resistance percentage by Salmonella isolates against E (100%), and NA (70%). These results are also in accordance with those reported by Hassan et al. [25], who found that all buffalo meat origin Salmonella isolates were susceptible to CIP and GEN, in Bangladesh. On the other hand, the low rate of resistance against AMK, IPM, GEN, CTX, MEM, CIP, and ENR in the present investigation could be related to their non-use or low-frequency use in animal production; hence these antimicrobials are considered the drug of choice in the management of human Salmonella infections. The high susceptibility played by the Salmonella strains against CIP and ENR was not surprising, since the antimicrobial resistance toward these molecules is reported worldwide because of their large use in food-producing animals [36,44,45].

Antimicrobial Resistance Profiles of Salmonella Isolates
The expressed AMR profiles and MAR indexes of the tested isolates are summarily presented in Table 3. The 53 resistant isolates comprised 16 resistance profiles. The E + S + CM + CPM (n = 6), E + S + CM (n = 6), E + S + CM + CPM + NA (n = 6) profiles were the most frequently encountered, followed by E + S + CM + CPM + NA + SXT + CET + AMP + TE + ENR (n = 5), E + S (n = 5), E + S + CM + CPM + NA + SXT + CET (n = 5), E + S + CM + CPM + NA + SXT + CET + AMP + TE + ENR + CIP (n = 4) and E + S + CL + FEP + NA + SXT + CN + AMP (n = 4). The great majority (77.3%, 41/53) of the isolates showed resistance to four, or more tested antimicrobials, from different classes. Accordingly, isolates expressing resistance to at least one drug, in a minimum of three or more antimicrobial classes, were considered multidrug-resistant [44], in which the MAR index was > 0.2. These MDR strains comprised all S. Enteritidis (n = 11), S. Infantis (n = 5), S. Essen (n = 3), S. Tsevie (n = 2), and S. Dublin (n = 2) strains. In addition, seven strains of S. Typhimurium, five strains of S. Montevideo, five strains of S. Rissen, three strains of S. Virchow, and one strain each of S. Anatum, S. Derby, S. Papuana, and S. Saintpaul were also MDR (Table 4). Only six (11.3%) out of the 53 isolates expressed resistance to just one or two antimicrobials, and these included two strains of S. Chester and one strain each of S. Typhimurium, S. Montevideo, S. Virchow, and S. Anatum, in which MAR index was 0.125 or less, indicating that there was no MDR (Table 3). A MAR index value higher than 0.2 is considered high risk, while a value lower than 0.2 indicates low risk [45].     In the current investigation, the occurrence of MDR strains (77.3%) is almost similar to that published by Boonmar et al. [26], who stated that 73% of Salmonella isolates (44/60) recovered from meat samples in Laos showed MDR. Higher MDR (100%) was observed by Sallam et al. [40], in Egypt, while a lower incidence (32.2%) of MDR Salmonella isolates was recorded in Algeria by Mezali and Hamdi [25]. The appearance of MDR among Salmonella isolates is a concern for public health, which needs more caution to prevent the hazardous and unnecessary use of antimicrobials in food industries and veterinary fields. The differences in the resistance patterns found in different investigations might be due to variations between the sampled geographical areas, locally used drugs, farm management practices, and misuse or overuse of some antimicrobials.
Interestingly, the only one isolate that showed a pandrug-resistant pattern (resistant to all of the 16 antimicrobials tested) was S. Enteritidis, while the rest of S. Enteritidis serovar isolates (n = 9) besides seven (87.5%) of the eight S. Typhimurium identified in the present study were multidrug-resistant (resistant to more than 10 of the antimicrobials tested). Additionally, the four isolates, which showed an extensively drug-resistant pattern (resistant to 13 or more of the antimicrobials tested) were distributed as one from each of S. Enteritidis, S. Typhimurium, S. Montevideo, and S. Rissen (Tables 3 and 4). On the other, hand all of the identified S. Dublin, S. Anatum, S. Derby, S. Papuana, S. Saintpaul, and S. Chester were resistant to 10 or less of the antimicrobials tested. Our findings concerning the antimicrobial resistance pattern of Salmonella serovars, especially the predominant ones (S. Enteritidis and S. Typhimurium) substantiate what had been reported in various studies originated from different countries [46][47][48][49].
Salmonella strains isolated in the present survey indicated 16 different AMR patterns, with an average MARs index of 0.436, and 77.3% (41/53) of them showed MAR indices > 0.2. A value of MAR index > 0.2 indicated an overuse and/or misuse of antibiotics. The current results for the MAR index reveal higher resistance rates among Salmonella isolates. Therefore, establishing valuable national surveillance systems in order to monitor the rational use of antimicrobials in the field of veterinary medicine is crucial. Likewise, for a better understanding of the antimicrobial susceptibility pattern of the isolated Salmonella strains, further studies are recommended focusing on the monitoring of their genotypic resistance pattern.

Conclusions
This study concluded that a high percentage of buffalo meat, marketed in Mansoura City, Egypt, was contaminated with Salmonella spp., with a dominant occurrence of S. Enteritidis (11/53) and S. Typhimurium (9/53) serovars. The great majority of Salmonella (~89%, 47/53) isolates were multidrug-resistant with an average MAR index of 0.436, and 77.3% (41/53) of them showed MAR indices > 0.2. which indicated an overuse and/or misuse of the antibiotics. Additionally, the virulence of Salmonella isolates was determined by the existence of the invA, stn, and hilA genes, which were detected in 100%, 71.7%, and 83% of Salmonella recovered isolates, respectively. Therefore, buffalo meat can constitute a significant public health concern, emphasizing the necessity of the implementation of a better antimicrobial stewardship program in Egypt, in order to decrease the unnecessary use of antimicrobials in food-producing animals. In addition, continuous efforts are required to maintain the disease prevalence low, applying control measures based on the prevention of contamination of animal meat at slaughter and butcher shops level, as well as proper cooking of meat at private households.