Occurrence and antibiotic resistance of Salmonella spp. in raw beef from wet market and hypermarket in Malaysia

Tan, C.W., Noor Hazirah, M.Z., Shu’aibu, I., New, C.Y., Malcolm, T.T.H., Thung, T.Y., Lee, E., Wendy, R.D. Nuzul, N.J., Noor Azira, A.M., Ungku Fatimah, U.Z.A., Rukayadi, Y., Rinai, R. and Son, R. Department of Food Science, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor Darul Ehsan, Malaysia Department of Food Service and Management, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400 UPM Serdang, Serdang, Selangor Darul Ehsan, Malaysia Food Safety and Food Integrity, Institute of Tropical Agriculture and Food Security (ITAFoS), Universiti Putra Malaysia, 43400 UPM Serdang, Selangor Darul Ehsan, Malaysia Institute of Bioscience, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor Darul Ehsan, Malaysia


Introduction
Salmonellae are rod-like Gram-negative bacteria that cause salmonellosis, one of the most important foodborne illnesses worldwide. Salmonella is described as highly pathogenic bacteria as their infective dose is as low as 108 to 10 4 cells (Perreten, 2005). Salmonellosis is a self-limiting infection that does not normally require antibiotic medication. However, when the disease worsens, antibiotic therapy could be recommended for the patient (CDC, 2017). In Malaysia, Salmonellae have been recognized as the culprits for some of the food poisoning outbreaks (ISID, 2013;Karim et al., 2017). Outbreaks of food poisoning are probably due to crosscontamination, improper handling, inappropriate storage temperature and inadequate cooking which encourage the growth of these pathogens in food.
Meat contributes an important proportion in the human diet. It is classified into red and white meat depending on the concentration of myoglobin, a substance in the muscle fiber of animals that is responsible for the meat color. Oxymyoglobins are meat components responsible for the red color of animal flesh and are developed upon exposure of myoglobin to oxygen (Mancini and Hunt, 2005). However, the redness of flesh also depends on the animal species, ages, and fiber types. Generally, cow, sheep, buffalo, goat, pork, and horses are categorized as red meat animals.
Data on prevalence and outbreak involving Salmonella reported were usually those associated with poultry, meat, eggs and dairy product (Dominguez et al., 2009;Gantois et al., 2009;Minami et al., 2010;Chong et al., 2017). The innermost part of the meat of animals is considered sterile unless the animal is infected with pathogens. The contamination of meat by pathogenic microbes is likely due to cross-contamination at various processing steps in the slaughterhouse and marketplace, transportation of the meat, poor packaging and after cooking.
Antibiotics are widely used in the animal farm to treat, prevent and control infections and promote growth (Marshall and Levy, 2011). The excessive use and misuse of antibiotics in livestock are among the public health concerns as the antibiotic residues could be passed along the food chain which eventually makes the bacteria in the food chain resistant to them. Animals are the main reservoir for many zoonotic pathogens and the extensive use of antibiotics as growth promoters as well as therapeutic agents in the animal farm can promote the evolution of antibiotic-resistant strains. For instance, several food animals such as poultry and cattle have been reported to serve as reservoirs for multidrug-resistant Salmonellae (Yoke-Kqueen et al., 2008;Arslan and Eyi, 2010;Thung et al., 2016).
The purpose of this study was to determine the occurrence of Salmonella in raw beef samples obtained from various wet markets and hypermarkets in Serdang, Selangor, Malaysia using MPN-Plating and MPN-PCR. Antibiotic susceptibility testing was performed on the isolates recovered from the raw beef samples to determine their tendencies for multiple antibiotic resistance.

Sample collection
A total of eighty-two beef samples were purchased from different wet markets (n=41) and hypermarkets (n=41) in Serdang, Selangor, Malaysia. All samples were collected in a sterile plastic bag, placed into a cold storage box with ice packs and transported to the laboratory. Samples were analysed immediately upon arrival to the laboratory.

Sample processing
Each sample was processed individually on a sterilized cutting board. Beef samples were cut and weighed approximately 10 g into a stomacher bag and added with 90 mL of buffered peptone water (BPW; Oxoid, UK). The mixture was then homogenized for 60 s using a Stomacher Lab-Blender 400 (Seward Medical, UK).

Enumeration of Salmonella using the Most Probable Number (MPN) method
The homogenized sample in the stomacher bag (10 -1 dilution) was serially diluted by transferring 1 mL into a tube containing 9 mL of BPW to make a 10 -2 dilution and this action continued until 10 -4 dilution was reached. Each dilution was then transferred into a set of 3-tubes MPN. The MPN tubes were incubated at 37°C overnight. Following the incubation, MPN tubes were examined for growth and turbidity and proceeded for confirmation on plating and PCR.

MPN-plating
Positive MPN tubes were confirmed by streaking on Xylose Lysine Deoxycholate (XLD; Oxoid, UK) agar plates. The agar plates were incubated at 37°C overnight. Agar plates yielding presumptive Salmonella colonies (black centers) were confirmed positive. The presumptive colonies were picked from each agar plate and streaked on Tryptic Soy Agar (TSA; Oxoid, UK) for purification and later stored as working culture.

MPN-PCR
DNA extraction was carried out using boiled cell method (Pui et al., 2011). From the positive MPN tubes, 1 mL was transferred into a 1.5 mL microcentrifuge tube. The suspension was centrifuged at 12,000 x g for 3 mins. The supernatant was discarded, and the pellet was re-suspended in 500 μL of sterile distilled water. The suspension in the tube was heated on a hot dry bath (Labnet, USA) at 100°C for 10 mins and then cooled immediately at -20°C for another 10 mins. The suspension was centrifuged at 12,000 x g for 1 min prior to use. The supernatant was used as the DNA template.
PCR detection of Salmonella in raw beef samples was performed using a pair of primer (SIGMA, USA) with the forward sequence of 5'-ATC GCT GAC TTA TGC AAT CG-3' and reverse sequence of 5'-CGG GTT GCG TTA TAG GTC TG-3' specifically designed to amplify a 204 bp fragment of ompC genes in the Salmonella spp. (Kwang et al., 1996).
A total of 25 μL reaction mixture was prepared for the amplification and consisted of 7.5 μL of 5×PCR buffer, 2.0 μL of 25 mM MgCl 2 , 0.5 μL of 10 mM dNTPs, 0.25 μL of 10 μM ompC primers, 1.5 U of Taq polymerase, 2 μL of DNA template and finally made up to 25 µL with sterile distilled water. PCR was carried out in the thermocycler, GeneAmp ® PCR System 2700 (Applied Biosystem,USA) with the following conditions: initial denaturation at 95°C for 3 mins, 35 cycles each of eISSN: 2550-2166 © 2018 The Authors. Published by Rynnye Lyan Resources denaturation at 95°C for 1 min, annealing at 56°C for 1 min, extension at 72°C for 1 min, and a final extension at 72°C for 7 mins. In order to visualize the PCR products, 5 μL of the PCR products were loaded onto an ethidium bromide-stained 1.2% agarose gel prepared using 0.5× TBE buffer. The products were subjected to gel electrophoresis at 100 V for 30 mins and viewed under the UV-transilluminator using gel documentation system (SynGene, UK).

Antibiotic susceptibility testing
Antibiotic susceptibility testing was conducted on the confirmed isolates of Salmonella from the samples by Kirby-Bauer method (Bauer et al., 1966). Salmonella isolates kept in the stock culture were subcultured aerobically in Muller Hinton broth (MHB; Oxoid, UK) at 37°C for overnight. The cultures were swabbed evenly onto Mullen Hinton agar (MHA; Oxoid, UK) plate and left to dry for 5 mins. Antibiotic disks were pressed down lightly to ensure attachment on the agar surface. The plates were incubated in inverted positions at 37°C for 24 h. A total of twelve types of antibiotic susceptibility test disks, namely imipenem (10 μg), nalidixic acid (30 μg), cephalothin (30 μg), ciprofloxacin (5 μg), ceftriaxone (5 μg), sulphamethoxazole/ trimethoprim (25 μg), tetracycline (30 μg), ceftazidime (30 μg), gentamicin (10 μg), streptomycin (10 μg), chloramphenicol (30 μg), and kanamycin (30 μg) were used in this study. After incubation, the diameter of the inhibition zone was measured to the nearest whole millimetre. Antibiotic susceptibility profiles of the isolates were determined based on the Clinical and Laboratory Standards Institute (CLSI) guidelines. The value of multiple antibiotic resistance (MAR) index was calculated using the formula, a/b, where 'a' is the number of antibiotics to which the particular isolate was resistant and 'b' is the total number of antibiotics tested (Krumperman, 1983).

Occurrence of Salmonella in raw beef
The occurrence of Salmonella in raw beef obtained from the hypermarkets and wet markets are shown in Table 1. The enumeration of Salmonella in the raw beef samples through the MPN-plating showed a higher prevalence (64.63%) compared to the MPN-PCR (17.07%) approach. In the case of Salmonella concentration in the raw beef samples, the MPN-plating also showed a higher concentration of presumptive Salmonella in the samples (ranging from 3 to 4600 MPN/g) than the MPN-PCR method with a range of 3 to 30 MPN/g. The MPN-PCR results revealed that 9.76% (4/41) of the samples from the hypermarkets and 24.39% (10/41) of the samples from the wet markets were found to harbour Salmonella with the microbial load of 3-30 MPN/g and 3-9.2 MPN/g respectively. The prevalence and concentration of Salmonella in the raw beef samples obtained from the wet markets were found to be significantly lower (P<0.05) than those from the hypermarkets.

Antibiotic susceptibility test
Four confirmed Salmonella isolates were recovered from the samples and tested for antibiotic susceptibility test. Antibiotic susceptibility profiles of Salmonella isolates from raw beef samples are shown in Table 2. All the four (100%) isolates were susceptible to imipenem, gentamicin, kanamycin, and chloramphenicol. About 75% of the isolates were susceptible to ceftriaxone and ciprofloxacin. Half (50%) of the isolates were susceptible to nalidixic acid, sulphamethoxazole/ trimethoprim, tetracycline, ceftazidime, and streptomycin. However, 25% of the isolates showed intermediate resistance to nalidixic acid, ciprofloxacin, ceftazidime, and streptomycin and all the isolates were resistant to cephalothin. In addition, 50% of the isolates were resistant to both of sulphamethoxazole / trimethoprim and tetracycline and 25% resistant to nalidixic acid, ceftriaxone, ceftazidime, and streptomycin. All isolates showed resistance to at least two antibiotics used in this study. Multiple antimicrobial resistance (MAR) index values of the isolates are displayed in Table 3. Isolate A had a MAR index value of less than 0.2. Isolate B, C and D displayed multiple antibiotics resistance with MAR index values of 0.25, 0.25 and 0.33, respectively.

Discussion
The study of the prevalence of Salmonella in raw beef samples by MPN-plating on XLD agar recorded a higher level of detection rate and microbial concentration (MPN/g) than the MPN-PCR approach.  Pui et al. (2011) reported that MPN-PCR had higher sensitivity in detecting Salmonella from sliced fruits than MPN-plating method. This was further supported by the report of Maks and Fu (2013), in which the PCR approach had significantly better detection of Salmonella with fewer false-negative results than the bacteriological medium.
Several studies have also found out that the low specificity of XLD medium created a high number of false positive identification of Salmonella species (Rall et al., 2005;Hyeon et al., 2012;Yhiler et al., 2015). Rall et al. (2005) stated that XLD allows the growth of selective bacteria with Salmonella, Proteus, and Citrobacter that displayed similar characteristics as negative lactose fermentation and negative hydrogen sulfide (H 2 S) production on this classical medium. Thus, the colonies of Proteus and Citrobacter probably appear and identified as presumptive Salmonella colony on XLD medium. In addition, Hyeon et al. (2012) and Yhiler et al. (2015) concluded that XLD medium had high sensitivity but with low specificity for the detection of Salmonella.
The results of the MPN-PCR recorded the overall prevalence of Salmonella in the raw beef samples of 17.07%. Comparatively, Thung et al. (2017) reported 7.5% of raw beef samples obtained from the retail markets in Selangor Malaysia, positive for Salmonella with the microbial load of <3 to 15 MPN/g. Similarly, Chong et al. (2017) also reported 10% of beef carcasses collected from the local abattoirs in Selangor, Malaysia as positive for Salmonella. Modarressi and Thong (2011) reported that 27.2% of raw beef samples purchased from the retail markets in Kuala Lumpur, Malaysia were contaminated with Salmonella. In other countries, the occurrence of Salmonella in the raw beef sample collected from the retail markets in Vietnam was reported to be 39.9 to 48.6% (Phan et al., 2005;Thai et al., 2012).
Moreover, the occurrence of Salmonella in the wet market samples was found to be higher than in the hypermarket samples. This finding was similar to the results reported by Shafini et al. (2017) in Malaysia. Shafini et al. (2017) found out that 20.8%, 14.8% and 11.1% of the beef samples from the wet markets, hypermarkets, and butcher shops respectively were contaminated with Salmonella. This corresponds with a finding in China in which Yang et al., (2011) reported the prevalence of Salmonella on raw poultry with higher prevalence in the wet markets (54.4%) than in large markets (50.3%) and small markets (52.1%). The reason for the higher incidence of Salmonella in samples FULL PAPER  obtained from the wet markets than those from hypermarket could be due to inadequate food safety practices compared to the food safety measures and sanitation programmes implemented in the hypermarkets. Thung et al. (2016) mentioned that hypermarket workers generally practice good personal hygiene and sanitation in the food processing environment than those of the wet markets.
The microbial load of Salmonella from the wet market and hypermarket samples were 3-9.2 MPN/g and 3-30 MPN/g, respectively. The concentration of Salmonella in the hypermarket samples was found to be slightly higher than that of the samples from the wet market. This finding was in contrast with the study of Thung et al. (2016), who reported the microbial load of Salmonella in beef samples from the wet market with a higher concentration (<3 to 15 MPN/g) than supermarket samples (<3 to 3.6 MPN/g). The higher Salmonella concentration in the hypermarket samples might be due to the longer operating hours than in wet markets, in which the meat exposed to the environment and temperature abuse longer which facilitates bacterial multiplication. Conversely, many wet markets operation hours are usually half a day eventually providing quality fresh produce for the consumers.
The existence of multiple antibiotic resistant bacteria can lead to human infections and probably creates challenges for the treatment of Salmonella infections in humans and animals (Thung et al., 2016). The use of antibiotics for treatment and as growth promoters in farm animals may promote the emergence and spread of antibiotic-resistant microorganisms to the human population (Khoo et al., 2015). Bacterial isolate is classified as multiple antibiotic resistant if the MAR value is higher than 0.2. Three isolates were found with a MAR value of higher than 0.2 and more than 50% (2/4) of Salmonella isolates were resistant to cephalothin, sulphamethoxazole/ trimethoprim, and tetracycline in this study. Another study reported that Salmonella isolated from the raw beef, chicken meat, and street foods exhibited the highest resistance to tetracycline (73.8%), followed by sulfonamide (63.6%) and streptomycin (57.9%) . Minami et al. (2010) stated that most of the Salmonella isolated showed resistance to sulfisoxazole, tetracycline, and streptomycin. Yoke-Kqueen et al. (2008) reported that Salmonella isolated from the indigenous vegetables and poultry exhibited multidrug resistance to erythromycin, tetracycline, streptomycin and trimethoprim/sulfamethoxazole.

Conclusion
MPN-PCR was more reliable than MPN-plating with regards to the degree of specificity of identification of Salmonella species in raw beef samples. Using MPNplating method, 64.63% of the samples were detected with Salmonella with a concentration range of 3-4600 MPN/g in comparison to 17.07% of the samples detected with a concentration range of 3-30 MPN/g via MPN-PCR approach. Raw beef can become a common vehicle for Salmonella transmission and if inadequate practice of food hygiene and sanitation, Salmonella infection is possible. Multiple antibiotic resistant Salmonella should be regarded as an issue and the spread should be viewed as a challenge to combat the bacteria and conserving public health.

Conflict of Interest
The authors declare no conflict of interest.