Prevalence of Salmonella contamination in processing chain of selected chicken-based side dishes

F U L L P A P E R Prevalence of Salmonella contamination in processing chain of selected chickenbased side dishes Novera, R., Rahayu, W.P., *Kusumaningrum, H.D., Indrotristanto, N. and Nikastri, E. Food Science Study Program, IPB University, PO BOX 220, Bogor 16002 Indonesia Department of Food Science and Technology, IPB University, PO BOX 220, Bogor 16002 Indonesia SEAFAST Center, IPB University, Kampus IPB Darmaga, Bogor 16680 Indonesia National Agency of Drug and Food Control, Jl. Percetakan Negara 23 Jakarta 10560 Indonesia


Introduction
Poultry and its products are among the most nutritious food, but also a well-known reservoir of Salmonella spp. Kang et al., 2017). Salmonella spp. is a very adaptive microorganism, which can cause human salmonellosis, an important foodborne disease (Giaouris and Nychas, 2006). In Indonesia, not many Salmonella spp. cases have been reported. A report from Chusniati et al. (2009) stated that domestic chicken eggs used in a traditional herbal drink (jamu) in Sidoardjo City, East Java of Indonesia, were contaminated by Salmonella spp. (5.56% from 36 samples). In 2011, the microbial examination conducted on 4808 food samples from the school canteens revealed that 13 (0.27%) samples contaminated by Salmonella (NADFC, 2012). In 2016 NAFDC examined 106 fried chicken samples using the Most Probable Number (MPN) method and confirmed by Polymerase Chain Reaction (PCR) method, found that 45 samples were contaminated by Salmonella spp. with 42.0% of prevalence, and 0.36-2.30 MPN/g of concentration (NADFC, 2016a). In 2017 NADFC reported 53 foodborne outbreaks in Indonesia, where 2041 people ill and 3 people died. About 13.21% of the outbreaks confirmed as microbial related, and 15 outbreaks (28.30%) happened in school (NADFC, 2017).
Salmonella on food has to be negative (NADFC, 2016b), but diseases related to Salmonella are still be found although not always foodborne related. Many studies showed that poultry or its products are the sources of Salmonella contamination. Barua et al. (2014) studied the sources of non-typhoidal Salmonella enterica serovars in human and found that nine serovars of Salmonella were isolated from human stool samples, and it's commonly found on poultry. Since 1981 the outbreaks related to salmonellosis had been reported by Cobet et al. (1981) who isolated 158 Salmonella Oranienburg from 150 hospitalized patients with diarrhoea. Punjabi et al. (2013) reported 296 laboratoryconfirmed enteric fever cases during the 7 months surveillance period in North Jakarta Indonesia, of which 221 (75%) were typhoid fever and 75 (25%) were the paratyphoid fever. Most of the cases occurred among children under five years old.
Chicken is one of the most popular food products in Indonesia. High level of protein and low-fat content contributed to the popularity of chicken. Furthermore, chicken is easy to prepare and widely used in restaurants, cafes, or canteens as a main meal or side dishes. Foods at the school canteens contributed to 15-20% of daily nutrition for the students. Most of the students (98.9%) who did not have breakfast will eat at the canteen. Those foods gave 31.06% of energy and 27.44% of protein every day (NADFC, 2012). Chicken based side dishes is a common food found in the school's canteen in Indonesia. This food has a risk of being contaminated by Salmonella. It is very important to ensure food safety at school considering the fact that the children are the future of the nation. Considering the severity and risk level of Salmonella and the regulation stated by NADFC (NADFC, 2012), the study of Salmonella contamination in chicken-based side dishes in school canteen is very important. This study aimed to identify the prevalence, the level, and the sources of Salmonella spp. contamination in selected chicken-based side dishes in the school's canteen.

Materials and methods
The chicken-based side dishes samples were collected from nine merchants of eight school canteens, namely fried chicken with precooking, fried chicken without precooking, breaded fried chicken, and sauced chicken. A total of 107 samples was collected, consisted of raw materials (chicken cuts, seasoning, egg, and water), precooked chicken, final product, swab sample from the hand of food handlers, and cutting boards.

Enumeration of Salmonella spp.
The enumeration method was conducted according to Pavic et al. (2009). A total of 25 g of samples were put into 225 mL of Buffer Peptone Water (BPW, Oxoid, UK) in stomacher bag and homogenized by stomacher (Seward™ Stomacher™ Model 400C Circulator Lab Blender, Fisher Scientific, USA) for 2 mins. A total of 10 mL of the initial suspension was transferred into each of 3 tubes containing 10 mL of BPW (double strength). An aliquot of the initial suspension was transferred into each of 3 tubes containing 9 mL of single strength BPW. Next, the initial suspension was diluted in BPW to provide 10 -2 , and 1 mL was transferred to each of 3 x 9 mL of BPW. All the tubes were incubated at 37 o C for 24 hrs and observed for the turbidity.
A 100 µL of the positive tubes were transferred into a petri dish containing Modified Semi-Solid Rappaport-Vassiliadis Agar (MSRVA, Oxoid, UK) and incubated at 42 o C. White colonies were observed after 24 and 48 h of incubation. The white colony was then subcultured on Xylose Lysine Deoxycholate Agar (XLDA, Oxoid, UK) and incubated at 37 o C for 24 hrs A single colony from specific colonies (red with or without black globule in the middle) was inoculated on Brain Heart Infusion Agar (BHIA, Oxoid, UK) medium and incubated at 37 o C for 24 hrs. Salmonella enterica serovar Typhimurium ATCC 14028 and E. coli ATCC 25922 were used as positive and negative controls.

Confirmation of Salmonella by PCR method
PCR method was conducted according to Rahn et al. (1992) with modification. One loop colony from BHIA was transferred into 500 µL of sterilized 0.85% Nalco (Merck, Germany) in 1.0-1.5 mL tube, homogenized and boiled in thermomixer (Eppendorf Thermomixer C, Germany) at 100 o C for 15 mins, then kept at -20 o C for 2 mins as a cold shock treatment. Then centrifuged (Hettich Zentrifugen Universal 320 R, Germany) at 12000 rpm for 5 mins. The DNA template was pipetted from the supernatant and transfer into 500 µL of sterilized microcentrifuge tubes.
PCR reaction (25 µL) using Go Taq Green Master Mix (Promega, US) kit with a composition of 2.5 µL DNA template from previous step, 0.5 µL each from forward and reversed primer which was F139 5'GTGAAATTATCGCCACGTTCGGGCAA-3' and R141 5'TCATCGCACCGTCAAAGGAACC-3' (Macrogen, Singapore), 12.5 µL Go Taq Green Master Mix, and 9 µL nuclease-free water (Promega, US) were prepared. The mixture of master mix and DNA template in PCR tube was homogenized and placed into thermal cycler (PCR) machine (Blue Ray Biotech, Taiwan) according to the shown in Table 1.

Detection of DNA amplification by electrophoresis
Tris-Borate EDTA (TBE) (1st base, Singapore) solution was prepared using miliQ water. Agarose 2% (1st base, Singapore) was prepared and 4-6 µL per 100 mL of flourosafe were added. The agarose was poured into the block to solidify. The remaining TBE solution was poured into the electrophoresis vessel (Biometra, Germany) then added the solid agarose. About 10 µL

Results and discussion
This study found that 30 of 107 samples were contaminated by Salmonella spp., indicated by the appearance of 284 bp band in PCR product (Figure 1). Invasive A gene (invA) was used as the target gene since it is the regulator for the invasive character. This character enabled Salmonella to penetrate the epithelium tissues of the intestine (Galán and Curtiss, 1989). A total of 4 of 30 end product samples were contaminated, but the level of contamination was significantly lower than chicken cuts (Table 2-5).
The number of contaminated end products was mainly associated with contamination in 8 of 21 (38.09%) chicken cut samples used as raw materials. Chicken cut samples were contaminated by Salmonella spp. with a range of 300-920 MPN/g. It was lower than the report from Kholifah et al. (2016) in Samarinda City, Indonesia, but higher than Naik et al. (2015) in India and Alali et al. (2012) in Russia. Hajrawati et al. (2016) reported that the intrinsic parameters of chicken carcasses in the Indonesian market were 0.84-0.85 for water activity and 6.00-6.37 for pH. This condition is suitable for Salmonella spp. which need water activity of 0.93 and pH of 4.5-9.0 to grow (Cary et al., 2000). El-Aziz (2013) reported that Salmonella enterica serovar Typhimurium was detected from 100 samples (each 25 frozen samples of chicken meat, liver, heart and gizzard) collected from Assiut city market, with a prevalence of 44% in chicken meat, 40% in liver, 48% in heart and none in the gizzard. The contamination on chicken cuts might occur during the slaughter process. In this study, chicken cuts and seasoning were brought from the traditional market. Maulita et al. (2017) reported that 8 of 10 samples of equipment used in the cutting process in the traditional market in Aceh, a province in Indonesia, were contaminated by Salmonella spp. with the highest concentration on the cutting board, 4.8 × 10⁴ MPN/g. A similar report came from Nidaullah et al. (2017) that found 161 Salmonella spp. isolated from 182 samples (88.46%) consisting of chicken carcasses, cages, knives, cutting boards, water, defeathering machines, and aprons collected Malaysian traditional markets, Two of four processing chain on the preparation of chicken-based side dishes in this study applied precooking process and was able to eliminate Salmonella spp. (Tables 2 and 5). The boiling process was conducted at 82-100 o C for 15-20 mins. The occurrence of Salmonella spp. in the end products may be due to crosscontamination after precooking process. The possible sources of contamination were seasoning, water, cutting board and the hand of food handlers (Table 2-5).
Two of 6 end products of sauced chicken were contaminated ( Table 5). The boiled chicken of sauced chicken was chopped and stir-fried with seasoning. Salmonella spp. was recovered from the seasoning, cutting board and hand of food handlers in this process (Tables 2 and 4). Swab samples of cutting boards and the hand of food handlers were conducted before the chopping process.  Groundwater and municipal water were used in the cooking process, but only groundwater was contaminated by Salmonella spp. The prevalence of Salmonella contamination in water was 23.5% (4 of 17 samples) and higher than the report from Momtaz et al. (2013) which was 0.89% in Iran. Groundwater was well known as a source of Salmonella Enteritidis (Kovacic et al., 2017) and also the carrier of typhoidal Salmonella serovar (Levantesi et al., 2012).
Another material used in chicken side dishes processing was eggs. Eggs were used in breaded fried chicken processing and contaminated by 0.92 MPN/g of Salmonella spp. Contamination in egg in Indonesia was reported by Chusniati et al. (2009) who found Salmonella spp. in 5.56% of 36 egg samples. It was higher than reported by Singh et al. (2010). The results showed that contamination of Salmonella spp. in eggs was 4.82%, with lower incidence in farms samples. Furthermore, they obtained 27 isolates that included S. Typhimurium (55.5%), S. Lagos, S. Africana, and Salmonella II. Kouam et al. (2018) studied Salmonella spp. contamination of 140 eggs from 19 farms. All farms sample contain at least one type serovar of Salmonella spp.
Another possible source of Salmonella spp. in sauced chicken processing was cutting board and hand of food handlers. Two of seven swab samples of hands and one of 4 swab samples of the cutting board was contaminated by Salmonella spp. The level of contamination by Salmonella spp. was 0.3 MPN/hand and 0.3 MPN/25 cm 2 respectively. Gorman et al. (2002) reported that Salmonella could be isolated from the counter's surface where food was processed and from aprons used by food handlers with a prevalence of 16.6% from total samples. Salmonella spp. can survive in a dry environment for a long time (Humphrey et al., 1994). Salmonella Enteritidis PT4 has also been reported showing the ability to form biofilms on the stainless-steel surface. These bacteria could grow in the presence of invisible leftover on the surface of the utensils as a result of inadequate washing process as reported by Giaouris and Nychas (2006).
In this study, all food handlers used frying method where only half of the chicken submerged in the oil. Although the temperature reached a high point, the possibility of uneven heat over the meat make Salmonella spp. could be recovered. Roccato et al. (2014) studied the resistance of Salmonella Typhimurium DT 104 in chicken-based foods with 3 cooking method, namely frying, grilling and baking. The results showed that 26 of 78 samples were contaminated and the highest prevalence came from frying method (12 positive samples), followed by grilling and baking. The average temperature of chicken meat during the frying process was 59-74 o C. He et al. (2011)  but the temperature of 90 o C could reduce the number into 5.5-7.1 log. Evans et al. (1996) reported that Salmonella Enteritidis PT 4 in beef rissoles was not eliminated during the frying process on 142-157 o C for 5-7 minutes. This bacterium came from eggs that were one of the ingredients. Beef rissoles were assumed as cooked when floating in the oil. It turns out the inside temperature was only 48-60 o C while the external temperature was 91-95 o C.
Heat resistance of Salmonella is different in each serovar, influenced by genetic factors, environmental conditions, and adaptability. Pre-exposure to thermal treatment on precooking could increase survival ability in thermal treatment such as frying. Chen et al. (2013) found Salmonella Senftenberg could survive in the chicken litter for 24 hrs at 80 o C. A similar report from de Jong et al. (2012), chicken breast fillet inoculated by Salmonella Typhimurium was boiled, and after one minute the temperature of the chicken became 85 o C. Extreme reduction on Salmonella Typhimurium has seen in 2.20 min, but even after 10 mins of boiling the bacteria could still be recovered. In this study, the decreasing of Salmonella was detected in all food processing chains. However, Salmonella spp. was still recovered in some of the end products, except in fried chicken with precooking.

Conclusion
The pre-cooking step in the selected processing chain of chicken-based side dishes reduced Salmonella spp. to an undetectable level. The occurrence of Salmonella spp. in 4 of 30 end product samples likely associated with cross-contamination. Inadequate deepfrying practices by all food handlers probably also contributed to the survival of Salmonella. This study suggested the need for education of the food handlers on proper cooking to improve the safety of the food.