Antimicrobial Resistance and Virulence of Non-Typhoidal Salmonella from Retail Foods Marketed in Bangkok, Thailand

Nontyphoidal-Salmonella bacteria cause foodborne gastroenteritis that may lead to fatal bacteremia, osteomyelitis, and meningitis if not treated properly. The emergence of multidrug-resistant Salmonella strains is a global public health threat. Regular monitoring of genotypes and phenotypes of Salmonella isolated from humans, animals, foods, and environments is mandatory for effective reduction and control of this food-borne pathogen. In this study, antimicrobial-resistant and virulent genotypes and phenotypes of Salmonella isolated from retail food samples in Bangkok, Thailand, were investigated. From 252 raw food samples, 58 Salmonella strains that belonged only to serotype Enteritidis were isolated. Disc diffusion method showed that all isolates were still sensitive to amikacin and carbapenems. More than 30% of the isolates were resistant to ampicillin, tetracycline, and ciprofloxacin. Twenty isolates resist at least three antibiotic classes. Minimum inhibitory concentration tests showed that 12.07% of the isolates produced extended-spectrum β-Lactamase. Polymerase chain reaction indicated that 32.76, 81.03, 39.66, and 5.17% of the isolates carried blaTEM-1, tetA, sul2, and dfrA7, respectively. All isolates were positive for invasion-associated genes. Effective prevention and control of Salmonella (as well as other food-borne pathogens) is possible by increasing public awareness and applying food hygienic practices. Active and well harmonised “One Health” co-operation is required to effectively control food-borne zoonosis.


Introduction
Salmonella causes food-borne gastroenteritis (salmonellosis) with high and increasing prevalence worldwide [1][2][3]. The bacteria are ubiquitously present in the environment and throughout the food chain, i.e., farm-to-folk. Humans become infected through the consumption of contaminated water or foods mainly of animal origins, such as poultry meat, eggs, pork, beef, dairy products, and ready-to-eat produce [4,5]. Salmonella serovars with human host preference include S. Typhimurium and S. Enteritidis [6,7]. Clinical symptoms of salmonellosis usually begin 6-8 h to 7 days after infection and are characterised by abdominal cramp, fever, and diarrhoea [8]. The diseases can be self-limited in healthy individuals but may be severe, which requires prompt medical attention and may also be life-threatening if the bacteria invade beyond the gastrointestinal tract [9]. According to the World Health Organization (WHO), Salmonella is one of the key causative agents of diarrheal disease, which inflicts not only huge medical intervention expenses but also loss of productivity [10].
Antibiotic resistance among bacteria is a global phenomenon. Regular monitoring of serotypes and drug-resistant phenotypes and genotypes of Salmonella that contaminate foods may help track the cause of the food-borne diseases and may lead to appropriate food safety policy for intervention, prevention, and/or effective treatment measures of food-borne illnesses. Therefore, in this study, we assessed the prevalence of antimicrobial phenotypes and drug resistance-associated and virulence genes in Salmonella isolated from retail food samples in the Bangkok metropolitan area.

Sample Collection and Bacterial Isolation and Identification
Five different food categories (chicken, n = 44; pork and beef, n = 28; seafood, n = 60; fruits and vegetables, n = 60; and dairy products, n = 60) comprising 252 samples were collected from 19 wet markets and 2 supermarkets between October and December 2017. All markets are located in the central and peripheral districts of the Bangkok Metropolitan area. Food samples were maintained in sterile bags on ice and transferred to the laboratory within 2 h.
Food samples were processed according to the international standard, five-step method of the ISO protocol: 6579: 2002 Microbiology of Food and Animal Feeding Stuffs-Horizontal Method for the Detection of Salmonella spp. [23,24]. Firstly, individual samples were pre-enriched in a non-selective medium. Twenty-five grams of each sample was placed in a sterile 500 mL flask containing 225 mL of Trypticase Soy Broth and incubated at 37 • C for 18-24 h. Then, 0.1 mL of each overnight culture was inoculated into 10 mL of selective enrichment medium, Rappaport-Vassiliadis Soya broth (Merck, Darmstadt, Germany), and incubated at 42 • C for 24 h. The cultures (0.1 mL aliquots) were spread onto selective agar plates, i.e., xylose lysine deoxycholate agar (XLD) and Salmonella-Shigella agar (SS) selective plates, and the plates were incubated at 37 • C for 18-24 h. Suspected Salmonella colonies (small red colonies with/without central black dots on XLD agar and translucent colourless colonies with/without central black dots on SS agar) were subjected to conventional biochemical assays for Salmonella verification, including triple sugar iron (TSI) agar utilisation, deamination of lysine, ornithine decarboxylation, citrate and urease productions, and indole formation, as well as motility testing [25].

Serotyping of the Salmonella Isolates
All Salmonella isolates were serotyped using polyvalent O and H antisera by slide agglutination technique (Kauffmann-White-Le Minor scheme) [26]. The assay was performed according to the manufacturer's instructions (Serosystem, Clinag, Bangkok, Thailand). Briefly, individual Salmonella colonies were suspended in normal saline solution on glass slides. They were mixed separately with 9 polyvalent Salmonella antisera reagents in a 1:1 Foods 2022, 11, 661 3 of 12 ratio, and the slides were rocked in a circular motion for 30 s. Bacterial agglutination was visually observed. Strains giving negative or positive agglutinations were recorded.

Determination of Intestinal Cell Invasion by Salmonella Isolates
The ability of the isolated Salmonella strains to invade human colon carcinoma cells (Caco-2 cell line) was investigated. Confluent Caco-2 cell monolayer was established in 24-well tissue culture plates (approximately 2 × 10 5 cells/well) containing Dulbecco's modified Eagle's medium (DMEM) (Gibco, NY, USA) supplemented with 10% fetal bovine serum and 50 µg/mL gentamicin at 37 • C in 5% CO 2 atmosphere. The monolayers were rinsed twice in phosphate-buffered saline, pH 7.4 (PBS). Cells were infected with individual Salmonella strains at a multiplicity of infection (MOI) 1:50 [27]. Plates were incubated at 37 • C in 5% CO 2 incubator for 4 h. The cells were rinsed to remove extracellular bacteria and replenished with DMEM containing gentamicin (50 µg/mL) for 1.5 h. Cells were then rinsed with PBS and stained with Giemsa reagent. Salmonella invasion into the Caco-2 cells was observed under inverted microscopy (200 and 400× magnifications) (Zeiss, Jena, Germany). Alternatively, the infected cells were lysed by adding 1% Triton X-100 (Sigma); the lysate was spread on an LB plate and incubated at 37 • C for 24 h. The presence of bacterial colonies on the cultured plate indicates the invasive ability of the bacterial isolate.

Polymerase Chain Reaction for Determination of Drug Resistance and Virulence Genes of the Salmonella Isolates
All Salmonella isolates were screened for the presence of virulence genes (invA and hilA) and antimicrobial resistance genes (tetA, tetC, bla TEM-1 , bla CMY-2 , sul2, and dfrA7) by using PCR. Genomic DNA was extracted from each Salmonella culture using the conventional boiling method [27]. Two millilitres of each bacterial culture were centrifuged at 14,000× g for 5 min. Sterile distilled water (600 µL) was added to the pellet and re-centrifuged. The supernatant was discarded, and 200 µL of sterile distilled water was added to the pellet. The sample was then placed in a 100 • C heat-block for 10 min, immediately cooled on ice for 5 min, and centrifuged at 14,000× g for 5 min. The supernatant was used as a PCR template.
PCR was conducted using primers listed in Table 1. The PCR reaction mixture (25 µL) contained 3 µL of DNA template, 2.5 µL of 10× Taq buffer, 2 mM MgCl 2 , 0.2 mM dNTP, 1 µM each primer, and 1 U of Taq DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA). The thermal cycles were initial denaturation at 94 • C for 5 min, 35 cycles of denaturation at 94 • C for 45 s, annealing at 52-60 • C for 40 s, extension at 72 • C for 40 s and a final extension at 72 • C for 7 min. Salmonella Enteritidis ATCC 13076 and constructed plasmids containing the antibiotic-resistant genes served as positive controls, while buffer alone (without DNA template) served as a negative control. The PCR products were electrophoresed on 1.5% (w/v) agarose gels in 100 mL of 1× TAE buffer and stained with ethidium bromide. DNA bands were visualised using the ChemiDoc MP imaging system (Bio-Rad, Hercules, CA, USA).

Statistical Analysis
The statistical analysis and data comparison were performed using one-way ANOVA in GraphPad Prism version 9 (La Jolla, CA, USA). The p-value < 0.05 was considered statistically significant.

Prevalence and Serotypes of Salmonella in Retail Food Samples
Fifty-eight Salmonella isolates (23%) were recovered from a total of 252 retail food samples. All of them belonged to serovar Enteritidis. The isolated bacteria were from chicken (36 isolates, 62.07%), pork (16 isolates, 27.59%), and beef (6 isolates, 10.34%). The comparative prevalence of S. Enteritidis isolated from chicken and pork, chicken and beef, chicken and fruits, chicken and vegetables, pork and fruits, and pork and vegetables were different (p < 0.001). The Salmonella prevalence in pork and beef samples was also different (p < 0.05). Nevertheless, no difference was found between samples of beef and fruits, beef and vegetables, and fruits and vegetables (p > 0.05). The isolates were further classified into six different groups, i.e., B (n = 17; 29.31%), C (n = 22; 37.93%), E (n = 15; 25.86%), G (n = 1; 1.72%), and I (n = 2; 3.45%), and non-A-I (n = 1; 1.72%). Group C was predominant in this study (Table 2).
chicken AMP, TE, CIP, and SXT chicken AMP, TE, CIP, and SXT chicken AMP, TE, and SXT chicken AMP, TE, and CIP E +

Antimicrobial and Virulence Genotypes of the Salmonella Isolates
PCR was used to determine drug resistance and virulence genes of the Salmonella isolates. The drug resistance and virulence genes that were detected included invA, hilA, tetA, bla TEM-1 , sul2, and dfrA7, of which their PCR amplicon sizes were 244, 296, 210, 504, 405, and 265 base pairs (bp), respectively (Figure 1). The invasion operon genes, invA and hilA, were detected in all isolates. The bla TEM-1 (n = 19; 32.76%), tetA (n = 47; 81.03%), sul2 (n = 23; 39.66%) and dfrA7 (n = 3; 5.17%) genes were carried by the resistance strains, a clear difference was noticed in the occurrence of these genes among the isolates. None of the isolates was positive for bla CMY-2 and tetC genes. The pork and chicken isolates were positive for at least one antimicrobial resistance-associated gene. The tetA was the most prevalent gene among the Salmonella isolated from pork and chicken, followed by sul2. None of the beef isolates carried the antimicrobial resistance-associated gene, and all of them were not resistant to any of the antibiotics tested ( Table 2).

Caco-2 Invasion Assay on Isolates
The ability of S. Enteritidis isolates to invade human intestinal epithelial (Caco-2) cells was determined. All 58 isolates, which carried invA and hilA genes, could invade the Caco-2 cells. The cell invasion of the representative isolate is shown in Figure 3.

Caco-2 Invasion Assay on Isolates
The ability of S. Enteritidis isolates to invade human intestinal epithelial (Caco-2) cells was determined. All 58 isolates, which carried invA and hilA genes, could invade the Caco-2 cells. The cell invasion of the representative isolate is shown in Figure 3.

Discussion
Regular monitoring of serotypes, antimicrobial-resistant characteristics, and virulence of food-borne pathogenic bacteria, particularly Salmonella enterica, can provide useful epidemiological information on food-borne bacterial infections in a locality [34]. In recent decades, S. Enteritidis has been identified as the predominant causative agent of salmonellosis in Thailand [35,36]. In this study, 23% of the raw food samples collected from open markets in the Bangkok metropolitan region were found to be contaminated with Salmonella. The contaminated food samples were solely meat (chicken > pork > beef), while seafood, fruits, vegetables, and dairy products were not contaminated. All contaminated Salmonella isolates belonged to serovar Enteritidis, of which group C was predominant. When compared with the prevalence of S. Enteritidis from raw foods in

Discussion
Regular monitoring of serotypes, antimicrobial-resistant characteristics, and virulence of food-borne pathogenic bacteria, particularly Salmonella enterica, can provide useful epidemiological information on food-borne bacterial infections in a locality [34]. In recent decades, S. Enteritidis has been identified as the predominant causative agent of salmonellosis in Thailand [35,36]. In this study, 23% of the raw food samples collected from open markets in the Bangkok metropolitan region were found to be contaminated with Salmonella. The contaminated food samples were solely meat (chicken > pork > beef), while seafood, fruits, vegetables, and dairy products were not contaminated. All contaminated Salmonella isolates belonged to serovar Enteritidis, of which group C was predominant. When compared with the prevalence of S. Enteritidis from raw foods in other countries, e.g., abattoirs in Iran and butcher shops and supermarkets in Pakistan where the prevalence rates were 43 and 37.5%, respectively, the bacterial prevalence in our study was less [37,38].
Drug susceptibility testing data revealed that even though the S. Enteritidis isolated in this study were resistant to many groups of antibiotics, including penicillin, combined β-lactam agents, cephalosporins, aminoglycosides, tetracyclines, fluoroquinolones, and folate pathway antagonists, most of these MDR Salmonella strains were still sensitive to amikacin and carbapenems. Even though the isolates of this study showed high resistance to ampicillin, tetracycline, and ciprofloxacin, the prevalence of resistant isolates was still less compared to those isolated in Brazil, Iran, and China [39][40][41].
Invasion into cultured epithelial cells has been routinely used for determining Salmonella virulence [42][43][44][45][46]. Genotypic and phenotypic analysis of the S. Enteritidis isolates of this study revealed that the bacteria carried invasion genes (invA and hilA). Nevertheless, they showed different degrees of invasiveness when tested by the invasion assay using intestinal epithelial (Caco-2) cells. The results conformed to those reported previously by others [47][48][49][50][51]. Most MDR Salmonella isolates were found to carry the antimicrobialassociated genes, namely, bla TEM-1 , tetA, sul2, and dfrA7 [28,52]. The prevalence of drug resistance genes was highest for tetA, followed by sul2, bla TEM-1 , and dfrA7. No isolate carried tetC and bla CMY-2 . Detail analysis of the entire genomes of the isolates by using next-generation sequencing should be performed further to provide the insight information for guiding appropriate treatment decisions and allow rapid tracking of transmission of the drug-resistant clones.
Epidemics of human salmonellosis are generally associated with a particular prevalent serovar and serotype of S. enterica. Epidemic tracking of the bacterial pathogens, e.g., through identification of the causative strain origin as well as the antimicrobial susceptibility pattern and their virulence characteristics in an outbreak, can be readily performed either phenotypically or genotypically, or both [29]. It is also noteworthy that retail food products undergo extensive processing and handling during production, which potentially enhance the risk of contamination [30]. Appropriate food hygienic education for end-consumers must be regularly implemented. Since the majority of food-borne diseases, including salmonellosis, are zoonotic, thus, improving food hygiene through health education and "One Health" approach should be practiced at all levels, i.e., from a locale to a nation-wide and global responsible practices.

Conclusions
In conclusion, the findings of this study supported the notion of the divergence of Salmonella serotypes isolated from a variety of raw food samples from the opened market and hypermarket in Bangkok and its periphery, Thailand. The findings also provided insight into the molecular characterisation of virulence-and drug-resistance traits, as well as the antimicrobial susceptibility pattern of the bacterial pathogen. The spread of MDR strains of Salmonella isolates with the cell invasion potential was become growing continuously. This requires good planning and effective control programs to prevent and manage infections for their spreading to community and public health.