Isolation of Salmonella from raw beef and chicken used in fast food restaurants in Abuja, Nigeria

¹Department of Veterinary and Pest Control Services, Federal Ministry of Agriculture and Rural Development, No 1 Kapital Street, Garki Area 11 Abuja, Nigeria, ²Department of Veterinary Public Health and Preventive Medicine, Faculty of Veterinary Medicine, Ahmadu Bello University Zaria, PMB 1096, Zaria, Nigeria, ³Department of Veterinary Public Health and Preventive Medicine, Faculty of Veterinary Medicine, Ahmadu Bello University Zaria, PMB 1096, Zaria, Nigeria, ⁴World Health Organization, Country Office, Abuja, Nigeria

^&Corresponding author
Salome Samuel Bawa, Department of Veterinary and Pest Control Services, Federal Ministry of Agriculture and Rural Development, No1 Kapital Street, Garki Area 11 Abuja, Nigeria

Introduction    

Salmonella is one of the most common causes of foodborne diarrheal diseases worldwide and most of these infections are zoonotic and transmitted from apparently healthy carrier animals to humans through contaminated foods. The main reservoirs of zoonotic Salmonella are food animals, and the main sources of infections in industrialized countries are animal-derived products, notably fresh meat products, poultry, and eggs [1]. In developing countries, however, contaminated water, vegetables, and human-to-human transmission contribute to a comparatively larger proportion of human cases than those in developed countries [2]. Microbial food safety is an increasing public health concern worldwide and the importance of food as a vehicle for the transmission of many diseases has been documented for a long time, especially in developing countries where hygienic standards are not strictly enforced and followed. These microorganisms have led to foodborne outbreaks and several countries have seen dramatic and steady increases in human outbreaks of salmonellosis, caused by infections in animals [3]. In addition to human health implications, Salmonella is a pathogen of significant importance in animal production especially in the context of the emergence of antibiotic-resistant strains. The therapeutic use of antibiotics in animals has led to a threat to animal and human health. Increasing attention has been centered on the control and prevention of Salmonella in animal production, as this is the most likely source of outbreaks in humans [4]. In most developing countries, there is paucity of reliable statistics on foodborne diseases due to poor or non-existent reporting systems. Biological contaminants largely bacteria, viruses, and parasites constitute the major cause of food-borne diseases[5]. Even though the restaurant industry plays an important role in the safety of the food supply chain, the proportion of illnesses that result from the consumption of food from restaurants is still unknown [6].

Although Salmonella being the second highest cause of food-borne illnesses in humans it has shown the varying prevalence rates in meat from many studies and its implication suggests the need for improved and strict food hygiene and safety management system in the abattoirs [7], during transportation, handling, and preparation of food. The rapid diagnosis of foodborne illness-causing pathogens is crucial for the food industry and public health. The invA gene is necessary for full virulence in Salmonella and has been suggested to trigger internalization required for invasion of deeper tissues [8]. It has been reported, that there is a considerable decrease in the number of false-positive results when invA primers specific for Salmonella were used; and amplification of the invA gene is now acknowledged as an international standard procedure for the detection of the Salmonella genus [8]. In recent times, several investigations continued to demonstrate the presence of Salmonella from different parts of Nigeria. Salmonella have been reported in chickens and turkey in Nsukka [9], in frozen poultry meat in Ibadan [10], poultry in Kwara [11], poultry and poultry sources from Maiduguri and Ibadan [12], raw meats sold in Lagos [13], retail beef and related meat products in Zaria [14] and raw beef in Jos [15]. Prevalence rates of 14.1% have been reported from commercial broiler chickens [16] and a prevalence of 15.4% from raw chicken meat in southern Nigeria [17]. This study was carried out to determine the prevalence of Salmonella from raw beef and chicken used to prepare meals in Fast- food restaurants in Abuja, Nigeria.

Methods     

Study area: Abuja is the Federal capital of Nigeria located in the central region of the country just north of the confluence of rivers Niger and Benue. It is situated within the Savannah region with moderate climatic conditions. The territory is currently made up of six local councils, comprising the Abuja Municipal Area Council (AMAC), Abaji, Gwagwalada, Kuje, Bwari, and Kwali.

Study design: we conducted a cross-sectional study using a systematic random sampling approach where each restaurant had an equal chance of being selected within the sampling frame. The sampling frame was made up of 284 fast-food restaurants, established based on the list of restaurants provided by the Association of Fast Food and Confectioneries in Nigeria (AFFCON) Abuja chapter. The study was conducted from November 2017 to July 2018 after obtaining Ethical approval from the Scientific and Ethical committee of the FCT Health Research Ethics Committee in July (Approval Number: FHREC/2016/01/87/ 10-11-16). Permission was sought from the management of each fast food restaurant selected before the commencement of the study.

Sampling technique: samples of raw meat were aseptically collected in sterile cellophane bags, labeled and then placed on ice packs in a Coleman flask and transported from point of collection to the laboratory for analysis. All collected samples were analyzed within 5 hr of collection. A total of 151 restaurants constituted the sampling sites. Two hundred and eighty-six (136 raw beef and 150 raw chicken meat) samples were collected and tested for Salmonella.

Procedure for examination of samples for Salmonella: Salmonella was isolated based on standard protocols [18].

Pre-enrichment ten (10) g of sample was weighed and aseptically transferred to a stomacher bag; 90 ml peptone water was added and briefly homogenized in a laboratory Stomacher for 1 minute. The homogenate was then poured into sterile conical flasks and incubated at 37°C for 24hr.

Enrichment 10ml of the incubated homogenate was transferred to sterile conical flasks containing 90ml of Rappaport Vassiliadis broth and incubated at 37°c for 24hrs.

Plating a loopful of the incubated homogenate was then streaked on the SSA plate to ensure isolated colonies which were then incubated at 37°C for 24hrs.

Preliminary identification one or more characteristic colonies appearing transparent or translucent colorless colonies, with or without black centers on SS agar were picked and inoculated into Triple Sugar Iron (TSI) agar and Urea agar. Colonies which gave reactions typical of Salmonella by showing Alkaline/Acid with or without gas and hydrogen sulfide on TSI and were urease negative were kept at 4°C on Nutrient agar (NA) slants until characterized.

Biochemical characterization: the biochemical characterization performed was based on standard techniques [18]. All isolates that gave reactions typical of Salmonella in all or most of the tests and substrates were considered to belong to the genus Salmonella. Microbact 12E Gram-negative bacillus (GNB) rapid identification system (Oxoid, Basingstoke UK). This is a miniature identification system that is computer-aided used for members of the family Enterobacteriaceae. Organism identification is based on substrate utilization and pH change. It has been reported to be, convenient, accurate, and simple to use alternative to the traditional time consuming conventional biochemical methods for identification [19]. A 24 hr culture of presumptive Salmonella colonies on selective media was obtained; next, an oxidase test was performed using oxidase test strips. One to three (1-3) isolated colonies of each culture, were picked and emulsified in 3ml sterile normal saline. The microplate was placed in a holding tray and the seal peeled back. Four (4) drops of bacterial suspension were added to each well, resealed and incubated at 37°C for 24hr. After 24hr of incubation, appropriate reagents were added to well 8, 10, and 12. Two(2) drops of Kovac's reagent was added in well 8 and observed for 2 minutes,1 drop of VP1(Voges Proskauer) and VP2 to well 10 and observed for 15-30 minutes, and 1 drop of TDA in well 12, which was interpreted immediately. Results were recorded in report forms containing the substrates that were tested. Twelve (12) substrates were tested; lysine, ornithine, hydrogen sulfide, glucose, mannitol, xylose, ONPG (ο-Nitrophenyl β-D-galactopyranoside), indole, urease, Voges- Proskauer, citrate, and TDA. Three substrates formed 1 group with each substrate assigned a number; when a substrate gave a positive result, the corresponding number for that group was summed up and recorded. A computer-identification software was used which permitted the input of a 4 digit code generated from the report forms; this promptly gave the probable identity of the organism tested in percentage. The Microbact software permits a 75% cut-off point for a probable identification. All tests that gave less than 75% were not accepted as Salmonella.

Antimicrobial susceptibility profiling: all the isolates identified as Salmonella species from the Microbact 12E identification were tested for their susceptibility to twenty (20) antimicrobial agents with the following disc contents; Tetracycline(30μg), Amoxicillin/clavulanic acid(30μg), Ampicillin(10μg), Chloramphenicol(30μg), Trimethoprim(5μg), Sulphamethoxazole/trimethoprim(25μg), Gentamicin(10μg), Ciprofloxacin(5μg), Nitrofurantoin(300μg), Amoxycillin(10), Cephalothin(30), Cefoxitin(30), Cefotaxime(30), Ceftazidime(30), Tobramycin(10), Amikacin(30), Norfloxacin(10), Nalidixic acid(30), Colistin Sulphate(10) and Imipenem(10) based on recommendations of CLSI performance standards for antimicrobial susceptibility testing [20]. Two to three (2-3) colonies of the appropriate culture were inoculated into 5ml tryptone soy broth and incubated at 37°C until the turbidity approximated 0.5 McFarland´s standard. Mueller Hinton agar plates were produced and used according to manufacturers´ instructions. (Oxoid, Basingstoke UK). Sterile swabs were dipped into the broth culture with the excess broth drained by pressing on the inner side of the tube; this was used to streak the Mueller Hinton agar in three directions at 180o until the entire surface was streaked. The plates were allowed to dry at room temperature for 10 minutes and the antimicrobial disks were dispensed unto the plates using the multiple disc dispenser (Oxoid, Basingstoke UK). The discs were further pressed with sterile forceps to ensure complete contact with the medium. The Petri dishes were then inverted and incubated at 37°C for 18hrs. After incubation, the zones of incubation were measured to the nearest millimeter and interpreted based on the interpretation of the zone diameter of test culture provided by the Clinical and Laboratory Standards Institute (CLSI) [20].

Determination of multiple antibiotic resistance (MAR) indexes: the MAR index points to the level of antibiotic resistance exhibited by an organism. This is calculated as described by Saba and colleagues [21]. MARI= a/b where "a" is the total number of antibiotics to which an organism is resistant and "b" is the total number of antibiotics against which the organisms were tested.

Determination of multiple antibiotic resistance (MAR) indexes: the MAR index points to the level of antibiotic resistance exhibited by an organism. This is calculated as described by Saba and colleagues [21]. MARI= a/b where "a" is the total number of antibiotics to which an organism is resistant and "b" is the total number of antibiotics against which the organisms were tested.

Determination of multidrug resistance (MDR): MDR was determined by ascertaining the drug class of each test antibiotic and registering those isolates with resistance to three or more classes [22].

Determination of virulence potentials of isolates by polymerase chain reaction

DNA extraction all isolates were inoculated into 5ml of tryptone soy broth (TSB) and incubated for 24hrs at 37°c. DNA extraction was carried out using the ZR Fungal/Bacterial quick DNA MiniPrep™ D3024 (Zymo research). All protocols were followed and ultra-pure DNA was eluted into 50µl DNA elution buffer. All preliminary isolates were subjected to the polymerase chain reaction, with the inv A being the targeted gene.

Primer primers used were obtained from INQABA™, South Africa and were synthesized based on the sequence of the inv A, with F5'GTG, AAA, TTA, TCG, CCA, CGT, TCGGGCAA-3' as the forward primer and R 5'-TCA, TCG, CAC, CGT, CAA, AGG, AAC, C-3' as a reverse primer to give an expected amplicon size of 284bp [23].

Salmonella PCR assay PCR was carried out in a total volume of 50µl containing 10µl template DNA, 2.5µl of the forward primer, 2.5µl of reverse primer, 25µl of DreamTaq PCR master mix containing dNTPs, Taq polymerase and 4mM MgCl₂, 10µl of nuclease-free water was also added. PCR was performed in a DNA thermal cycler (Applied Biosystems, Gene Amp PCR system 9700). After an initial denaturation step of 2 min at 95°c, 35 cycles of amplification were performed. Each cycle consisted of the following steps; 30 sec at 95°c (denaturation), 30 sec at 53°c (primer annealing), and 1 min at 72°c (extension) and 72°c for 7 min for the final extension. Ten microlitres of the reaction mixture were then resolved by electrophoresis on 2% agarose gels with the 100bp DNA ladder (Fermentas™, Germany), and the reaction products were visualized by staining with ethidium bromide.

Results     

Recovery frequency of samples: a total of 286 samples were examined for Salmonella. Salmonella was isolated from only two samples which were from two different wards, namely Garki (1.7%) and Central area (3.2%) (Table 1). None of the poultry samples was positive for Salmonella.

Identification and characterization of isolates: Identification with Microbact eleven suspected Salmonella were identified for further testing based on the Preliminary tests that were carried out. The 11 suspects were further tested using the Microbact rapid kit for Enterobacteriaceae, which identified 3 out of 11 that were tested as Salmonella, but one isolate identified as Salmonella had a percentage probability below 75% and was not considered as positive. Three (3) isolates were identified as Proteus mirabilis, two (2) as Citrobacter fruendii, two (2) as Acinetobacter baumanii, and one (1) as Serratia liquefaciens.

Serotyping of isolates serotyping was carried out at the WHO collaborating center for Antimicrobial resistance, Technical University Denmark, and isolates were confirmed to be the serotype Salmonella Enteritidis.

In vitro susceptibilities of the isolates to 20 antibiotics: Susceptibilities of Salmonella isolates to the 20 antimicrobials tested was noted; Both isolates were susceptible to Cefoxitin (FOX), Ceftazidime (CTZ), Amikacin (AK), Norfloxacin (NOR), Ciprofloxacin(CIP), chloramphenicol (C), Nalidixic acid(NA), Imipenem (IPM) and Amoxicillin/clavulanic acid (AMC). All the isolates(100%) were resistant to Ampicillin (AMP), Amoxicillin(AML), Cephalothin (KF), Tetracycline(TET), Trimethoprim (W), Sulphamethoxazole/trimethoprim (SXT) and Nitrofurantoin (F). One of the isolates was susceptible to Cefotaxime (CTX), Tobramycin (TOB), Gentamicin (CN), and Colistin sulfate (CT).

Resistance patterns of the Salmonella isolates: each isolate showed a distinct resistance pattern with isolate represented with code ‘C2B´ showing resistance to 9 antibiotics (AMP, AML, KF, CTX, TOB, TE, W, SXT, F )and isolate with code ‘G2CB2´ showed resistance to 10 antibiotics(AMP, AML, KF, TOB, CN, TE, W, SXT, F, CT), both displaying multiple drug resistance (MDR) phenotypes. Our study observed that all the isolates had multiple antibiotic resistance (MAR) index greater than 0.2(0.45 and 0.5 respectively).

Detection of invA among isolates: all 11 suspected isolates that were recovered from preliminary tests were screened for the invA gene. Only the 2 isolates identified as Salmonella by Microbact also showed the expected invA bands. (Figure 1).

Discussion     

Two Salmonella isolates, which included 1(1.7%) in raw beef from Garki ward and 1(3.2%) in raw beef from Central area wards of Abuja were isolated in this study giving a prevalence of 1.5% for the 136 raw beef samples analyzed, none was isolated from 150 raw chicken samples. An overall prevalence of 0.7% of the 286 samples investigated is observed to be lower than recent studies [15-17] which reported prevalences of 11% in raw beef, 14.1% from commercial broiler chicken, and 15.4% in raw chicken respectively. Our investigation observed multiple drug resistance (MDR) in the two Salmonella isolates recovered. All the isolates (100%) were resistant to ampicillin, amoxicillin, cephalothin, sulphamethoxazole, tetracycline, trimethoprim and nitrofurantoin. One (1) of the isolates was resistant to tobramycin, gentamicin, colistin sulphate and cefotaxime. The resistance of Salmonella isolates to cephalosporins is emerging to be a significant public health problem, cephalosporins are considered to be one of classes of drugs of choice used in the treatment of invasive non-typhoidal Salmonella infections in situations where trimethoprim/sulfamethoxazole or ampicillin is clinically ineffective [23-25]. Research has shown that there is a linear correlation of resistance to beta-lactam antimicrobials with the lactamase level over some time and resistance to beta-lactam can be attained by escalating enzyme levels [26]. Hence, the protracted use or misuse of cephalosporins selects for resistance over a period. The MAR index reported in this study (0.45-0.5) seems to be high, which signals the exposure of the animal product source (cattle) to high-risk contamination with antibiotic-resistant pathogens from high-risk environments, which increases the possibility of shedding microorganisms harmful to humans [27].

The invasion gene invA was detected in both isolates. This gene is essential for full virulence in Salmonella and is thought to trigger internalization required for invasion of deeper tissue [23]. This finding closely agrees with other studies that reported the detection of this gene in almost all Salmonella isolates, [8,23], and reported that PCR assay using invA primers specific for Salmonella considerably decreases the number of false-positive results. Amplification of the invA gene is now recognized as an international standard procedure for the detection of the Salmonella genus [8]. Several studies in Nigeria have reported recovery of Salmonella spp. from apparently healthy animals and animal products at varying prevalence; 6.4% from chicken, 14.1% in raw meat, 39% in apparently healthy slaughtered food animals and 2% in raw beef respectively [11,13,28-29]. Other studies carried out in other parts of Africa have continued to report Salmonella from animals and animal products. Salmonella was detected with a prevalence of 19% in beef carcasses in South Africa [30], 4% in sausage, 2% in spiced meat minced meat in Egypt [31] and 4.2% in slaughtered cattle and 12.1% in minced meat in Ethiopia [32]. This varying prevalence may be attributed to the Salmonella carriage among animals in different locations and countries, and may be influenced by multiple aspects such as the critical role of the slaughter process in contamination of carcasses, abattoir environment, poor slaughter techniques, wet and dry hides and skin, contaminated slaughter equipment, improper evisceration, faulty transportation, and retail procedures, storage conditions, sample types, culture methods and culture media used[33].

The storage temperature of the samples collected in this study could be a factor that may have contributed to the low prevalence rate observed, which is an important factor in the survival of pathogenic bacteria [34]. The temperature range for the growth of Salmonella spp. is 5.2-46.2°C, with the optimal temperature being 35-43°C. Although freezing affects Salmonella spp. survival, it does not guarantee the destruction of the organism. There is an initial sharp decrease in the number of viable organisms at conditions close to the freezing point as a result of the freezing damage, but Salmonella spp. in some situations have been seen to have the ability to survive long term frozen storage at lower temperatures as well [35]. Another contributing factor may be linked to a limitation observed in this study; samples for this study were collected from registered fast-food restaurants and during sampling, it was discovered that some fast-food restaurants were not registered with AFFCON. It was also observed that the 2 Salmonella isolates identified by Microbact were confirmed to be Salmonella based on PCR detection of the inv A gene; this suggests that the Microbact is a sensitive tool for rapid identification from a pure culture. The need for rapid identification of organisms from samples with or without outbreaks is necessary, and thus the convenience and efficiency demonstrated by the Microbact kit suggests a possibility for it to be used for rapid identification of foodborne organisms and also can be used to improve diagnosis in laboratory settings.

Our study was able to provide new information on an important foodborne pathogen in the food industry, specifically fast-food restaurants, which were hitherto not reported in Abuja. The MDR status of the isolates and the presence of a virulence gene reported in this study is also significant. The inappropriate application and control of a uniform food safety management system like the Hazard Analysis Critical Control Points (HACCP) by some of the fast-food restaurants sampled and the subsequent recovery of a pathogenic organism from animal products used in fast-food restaurants is a cause for concern. Food animals are the major reservoirs of Salmonella and the findings in this study suggest that meat sourced and kept in an unhygienic environment, and kept in inappropriate storage conditions could be a potential reservoir for Salmonella [36]. The consumption of contaminated food of animal origin can bring about the acquisition of antimicrobial-resistant foodborne pathogens [37] which is undesirable. Antimicrobial use and misuse have been contemplated to be the most vital selecting force for antimicrobial resistance of bacteria development and spread in both veterinary and human medicine [38]. Inappropriate antibiotic usage in food animals could predispose humans to risks of antibiotic-resistant bacterial infections with the situation additionally complicated by the potential of resistant bacteria to transfer their resistance determinants to resident human microflora and other pathogenic bacteria [37]. It is, therefore, necessary for all fast-food restaurants in Nigeria to properly implement a food safety management system. Lack of personal hygienic measures, poor hygiene practices in abattoirs, lack of disinfection during retailing/storage equipment, poor handling and processing methods coupled with poor environmental sanitation are all potential factors that can lead to contamination of meat. Hence, proper handling of raw meat is encouraged to prevent Salmonella contamination. Further studies should be carried out to assess risk factors involved in the spread of MDR Salmonella isolates along the food chain.

Conclusion     

Two (2) isolates of Salmonella were isolated from raw beef and all the isolates carried the virulence-associated gene invA and this suggests that they are potentially virulent. These isolates also exhibited multiple drug resistance (MDR) phenotypes. Overall it can be concluded that the prevalence of Salmonella in the Federal Capital Territory (FCT) seems to be lower than reported in recent studies.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Salome Samuel Bawa and Jacob Kwada Paghi Kwaga contributed to the conceptualization, data curation, formal analysis, investigation, methodology, resources, and writing of the original draft. Jacob Kwada Paghi Kwaga and Mohammed Kabiru Lawan contributed to data curation, project administration, supervision, validation, writing - review and editing while Samuel Bitrus Bawa contributed to interpreting results and investigation. All authors have read and agreed to the final manuscript.

Acknowledgments     

We are grateful to the Department of Public Health, Health, and Human Services Secretariat, Federal Capital Territory Administration, Garki Abuja for facilitating sample collection.

Table and figure     

Table 1: distribution of fast-food restaurants and recovery frequency of samples

Figure 1: detection of Salmonella by PCR

References