Sink survey to investigate multidrug resistance pattern of common foodborne bacteria from wholesale chicken markets in Dhaka city of Bangladesh

Antimicrobial resistance (AMR) among foodborne bacteria is a well-known public health problem. A sink survey was conducted to determine the AMR pattern of common foodborne bacteria in cloacal swab of broiler chickens and sewage samples from five wholesale chicken markets of Dhaka city in Bangladesh. Bacteria were identified by culture-based and molecular methods, and subjected to antimicrobial susceptibility testing. Resistance genes were identified by multiplex PCR and sequencing. Multidrug resistance (MDR) was observed in 93.2% of E. coli, 100% of Salmonella spp., and 97.2% of S. aureus from cloacal swab samples. For sewage samples, 80% of E. coli, and 100% of Salmonella and S. aureus showed MDR. Noteworthy, 8.3% of S. aureus from cloacal swab samples showed possible extensively drug resistance. Antimicrobial resistance genes (beta-lactamase—blaTEM, blaSHV; quinolone resistance gene—qnrS) were detected in a number of E. coli and Salmonella isolates from cloacal swab and sewage samples. The methicillin resistance gene (mecA) was detected in 47.2% and 25% S. aureus from cloacal swab and sewage samples, respectively. The findings envisage the potential public health risk and environmental health hazard through spillover of common foodborne MDR bacteria.

In Bangladesh, wholesale markets are the important sources of live poultry supply to consumers, which are particularly located in larger cities, where wholesalers sell their poultry to retailers, hawkers and even directly to restaurants 27 . Different species of poultry are aggregated in wholesale markets, where multiple poultry species are housed together for sale that can facilitate spread of bacterial and viral infections among poultry and from poultry to humans. Furthermore, wholesale markets have been rendered as the potential reservoirs of poultry associated foodborne antimicrobial resistant bacteria and ARGs 28 . Several research works have been conducted regarding the prevalence, AMR pattern of E. coli and Salmonella spp. from cloacal swab samples of broiler chickens sold in live bird markets in Bangladesh 18,23,26,29 . However, the occurrence of AMR foodborne bacteria in poultry sold in wholesale market has not been examined. The present study was, therefore, conducted to determine the AMR pattern of common foodborne bacteria isolated from cloacal swab and sewage samples from wholesale chicken markets of Dhaka city in Bangladesh. We also focused on the determination of antimicrobial resistance genes in common foodborne bacteria.

Results
Demographic information and disposal practices of wholesale chicken markets. We observed that in all five wholesale chicken markets, poultry, especially chickens were mainly brought from different districts (second tier of regional administrative area) of Bangladesh. Approximately 5000-10,000 chickens were sold each day per wholesale chicken market except Mohammadpur Krishi Market where more than 10,000 chickens vended each day. During sampling, we observed that different types of poultry such as broiler, layer, cockerel, backyard chickens, turkey, duck, and pigeon were kept together in a very compact stall throughout the day for selling. These poultry were mainly distributed to various retail markets, supershops, hotel, restaurant, and community center throughout Dhaka city (Supplementary Table S1).
Liquid wastes such as blood, waste water mixed with droppings were washed out and drained into nearby Buriganga River, which flows past the southwest outskirts of the capital city Dhaka. Solid wastes such as poultry plumes, offales and leftover feed were disposed into drum, which was then collected by the waste collectors and disposed into the municipal corporation dustbin situated close to the road. However, in Karwan Bazar Kitchen Market, solid wastes were normally kept into the nearby bucket, which were offered to different merchants who purchased and sold these to fish ranchers.
Venn diagram displaying more than one bacteria (90%, 45/50) was co-isolated from cloacal swab samples from wholesale chicken markets. The occurrence of co-isolation with E. coli, Salmonella spp., and S. aureus was found highest (46%, n = 23), followed by E. coli and Salmonella spp. Single isolation with either E. coli, Salmonella spp. and S. aureus was found in 5 (10%) of cloacal swab samples (Fig. 2).
Prevalence of β-lactamase-, and PMQR-encoding genes in E. coli and Salmonella spp.. According to BSBL-encoding genes, all isolates of E. coli and Salmonella spp. from cloacal swab and sewage samples were positive for blaTEM gene. The blaSHV gene was detected in 20% (1/5) of E. coli isolates from sewage samples. None of the tested E. coli and Salmonella spp. isolates had ESBL-encoding gene in both types of samples. For PMQR genes, the qnrA gene was present in 9.1% (4/44) of the E. coli isolates from cloacal swab. The qnrS gene was simultaneously detected in 70.5% (31/44) and 80% (1/5) of E. coli isolates from cloacal swab and sewage samples, respectively. On the contrary, only one isolate (2.6%) of Salmonella spp. from cloacal swab samples contained the qnrS gene. The qnrB gene was completely absent in all isolates of E. coli and Salmonella spp. from both types of sample ( Table 2 Coincidence of resistance genes. Among E. coli isolates, the most frequent gene combinations were a two-gene pattern of blaTEM + qnrS in cloacal swab (n = 31) and sewage samples (n = 3), followed by blaTEM + qnrA (n = 4) in cloacal swab samples. However, one isolate of sewage samples carried three genes combination (blaTEM + blaSHV + qnrS). Furthermore, coincidence of blaTEM with qnrS was also identified as the most prevalent combination in case of Salmonella spp. recovered from cloacal swab samples. However, no coincidence of resistance genes in Salmonella spp. was found in case of sewage samples (Table 3).

Discussion
The pilot sink survey reports the baseline findings on the extent and distribution of foodborne bacteria such as E. coli, Salmonella spp., and S. aureus alongside their antimicrobial resistance pattern including resistance genes in cloacal swab and sewage samples from five wholesale chicken markets in Dhaka city of Bangladesh. This investigation indicated the high occurrence of E. coli (88% vs 100%), Salmonella spp. (76% vs 100%), and S. aureus (72% vs 80%) in cloacal swab and sewage samples, respectively. These findings are in accordance with  15 . In the current study, mixed isolation of E. coli, Salmonella spp. and S. aureus were observed in 90% of cloacal swab and 100% of sewage samples. These three foodborne bacteria are usually present in the gut of chickens, and faecal shedding allows these bacteria to be transmitted among chickens, and in the environment of the wholesale markets 7 . ESBL-producing microorganisms are far reaching around the world including Bangladesh 10,19 . Accordingly, antimicrobial resistance inferable from ESBLs is a significant general wellbeing concern. In the current investigation, the prevalence of ESBL-producing E. coli, and Salmonella spp. from cloacal swab samples was 98%, and 76%, individually. The rate of ESBL-positive E. coli and Salmonella spp. was higher than those found in feces of chickens, 30% ESBL-producing E. coli in Bangladesh 9 , and 13.1% ESBL-producing Salmonella spp. in Belgium 38 . Similarly, the isolation rate of ESBL-producing E. coli, and Salmonella spp. was also higher in sewage samples (100%, and 60%, respectively). Prevalence of 17.2-29.5% ESBL-producing E. coli from water samples has been reported previously in Bangladesh 10,24,25 . The high prevalence of ESBL-producing E. coli and Salmonella spp. in cloacal swab and sewage samples might be due to widespread use of broad-spectrum antimicrobials in poultry production, ESBL-producing bacteria have evolved and shown greater incidence owing to mutations, selection, and the spread of ARGs in chickens and environment 39 .
Methicillin-resistant S. aureus (MRSA), which is present in chickens and sewage samples may serve as a reservoir for MRSA, accordingly permitting this microorganism to continue and spread in the community. The consequences of this investigation demonstrated a high prevalence (47% vs 50%) of MRSA in cloacal swab and sewage samples, separately. This is inconsistent with the result of previous reports, in which 19.2% prevalence of MRSA was observed in cloacal swab from broiler chickens in Chattogram district of Bangladesh 40 . Prevalence of MRSA Table 2. Prevalence of β-lactamase-, and PMQR-encoding genes in E. coli and Salmonella spp. isolated from cloacal swab and sewage samples. n No. of isolates, BSBL broad-spectrum β-lactamase, ESBL extendedspectrum β-lactamase, PMQR plasmid-mediated quinolone resistance.  www.nature.com/scientificreports/ in cloacal swab and environmental samples were 27.3%, and 8.3% in Iraq and Netherlands, respectively 15,16 . The variation observed in this study might be attributed to differences in the regulation of antibiotics used during poultry production systems and hygiene management of wholesale markets 41 . An important finding of concern in this study is that 93% E. coli, 100% Salmonella spp., and 97% S. aureus isolates from cloacal swab samples were MDR. Results similar to ours were reported from Bangladesh, Egypt, and India where MDR E. coli isolates from cloacal swab samples were 100%, 98.2%, and 94%, respectively 18,20,42 , and MDR Salmonella spp. ranged from 91.5-100% in China 43,44 . Furthermore, a higher number of MDR E. coli, Salmonella spp., and S. aureus was also found from sewage samples. But findings contrasting to ours were reported from Bangladesh, India, and South Korea 25,34,45 . A significant proportion of the isolates among these three types of bacteria from both samples were resistant to 6-8 antimicrobial classes. A high occurrence of MDR E. coli, Salmonella spp., and S. aureus isolated from both samples was also observed in different markets. Moreover, the emergence of extensively drug resistance (XDR) in foodborne bacteria has become a noteworthy public health threat because of very few, or even sometimes no, antibiotics can be effective for infections caused by these bacteria. Prominently, the current examination additionally saw that 8.3% of S. aureus isolates from cloacal swab samples were possible extensively drug resistant (pXDR). A previous report from India identified XDR S. aureus in 15.1% of clinical samples in humans 46 . The reason behind the high MDR and existence of pXDR among these foodborne bacteria in this study might be attributed to the selective pressure because of unnecessary use of antimicrobials as feed added substances or prophylactic medicines in chickens and the environment where chickens are raised. Because of abuse of antimicrobials, microorganisms with MDR may at last supplant drug-touchy microorganisms in conditions soaked with antimicrobial agents 47 . In Bangladesh, antimicrobials are easily accessible, and can be bought without prescription by the veterinarians. In this way, exacting rules and comprehensive antimicrobial medication monitoring frameworks should be earnestly advocated and implemented particularly in developing countries like Bangladesh for the rational use of antimicrobial agents in poultry production to diminish the development of MDR and pXDR isolates.
Fluoroquinolones are considered as medications of decision for treatment of human infections caused by Gram-negative and Gram-positive microbes 48 . However, the development of resistance to fluoroquinolones has been arisen throughout the time because of abuse and additionally overdose of medications in human and veterinary practice thus of expanding general wellbeing concern 49 . Nalidixic acid, the first generation quinolone, has the ability to develop resistance quite rapidly. In this current investigation, nalidixic acid resistance was the most as often as possible noticed antimicrobial resistance in E. coli, Salmonella spp., and S. aureus from cloacal swab and sewage samples. Moreover, resistance to pefloxacin was found higher among E. coli and Salmonella spp. in both samples in this study. A few studies have likewise indicated that resistance to nalidixic acid and decreased susceptibility to fluoroquinolones have expanded among E. coli, Salmonella spp., and S. aureus from chickens and environment 10,23,26,34,37,50 .
A high percentages of E. coli, Salmonella spp., and S. aureus recovered from cloacal swab and sewage samples were resistant to ampicillin, trimethoprim-sulfamethoxazole, and doxycycline, which is as per past reports conducted in Bangladesh 18,23,50,51 , Egypt 8 , and India 36 . The high resistance to ampicillin, trimethoprim-sulfamethoxazole, and doxycycline indicates that these antimicrobials have been persistently utilized in enormous amounts in the poultry production in Bangladesh. These perceptions uphold the likelihood that chickens might be a possible source of antimicrobial-resistant foodborne infections in humans 52 .
Of note, about 60% of E. coli isolates were resistant to imipenem and meropenem, and 40% of Salmonella spp. to imipenem in cloacal swab and sewage samples though this two antimicrobials are not used in poultry practices in Bangladesh. Our findings expressed a higher percentage of carbapenem resistance of E. coli and Salmonella spp. contrasted with previous investigation on chickens and environmental samples, were found to be 1.8% in Egypt 20 , 2.9% in Philippines 53 , and 3.1-8.1% in India 45,54 . Development of carbapenem resistance in E. coli and Salmonella spp. may be because of horizontal gene exchange of carbapenemase-encoding genes together with co-and cross-selective mechanisms 55 . If once acquired carbapenem resistance, at that point this resistance can be transmitted from humans to poultry and from poultry to humans through the food chain. Global epidemiological observation of resistance to these "last resort" antimicrobials is needed to build up potential connections between reservoirs and to restrict the bidirectional exchange of the encoding genes between foodborne bacteria and other commensal bacteria 55 .
Colistin as antimicrobial substance has been widely used in poultry production in many countries including Bangladesh, especially for prevention and treatment of Enterobacteriaceae infections and as growth promotion purposes 56 . According to WHO, colistin has been reclassified as an antibiotic of highest priority critically importance to treat infections caused by MDR and carbapenem-resistant bacteria in human medicine 57 . The extensive use of colistin in humans and poultry is recognized as the reason for the emergence and dissemination of the colistin resistance. In the present study, almost all the isolates of Salmonella spp., and 50% isolates of S. aureus obtained from cloacal swab and sewage samples were resistant to colistin. It is much higher compared to the findings of Aditya 58 who reported 50% of the Salmonella spp. isolates of chickens were resistant to colistin. But, another study reported 75% of the S. aureus isolates from environmental samples showed resistant to this antimicrobial 50 , which is higher than our study. The long term prophylactic use of colistin in poultry production may be attributed to this high resistance rate, which can be spread by direct or indirect poultry-to-human contact.
In this investigation, it was noticed that the S. aureus isolates from cloacal swab and sewage samples had the highest antibiotic resistance to ceftazidime (100%), oxacillin (89%), and cefixime and cloxacillin (75-81%). The oxacillin-resistant S. aureus was previously isolated from 32.1% of chicken samples 15 , which was in conflict with our outcomes. The purpose behind the distinction in resistance rates may be a fast change in antimicrobial sensitivity patterns of bacteria inside a brief period 5 .
Among the predominant β-lactamase genes responsible for β-lactam antibiotic resistance, blaTEM, blaSHV, and blaCTX-M (blaCTX-M-1 and blaCTX-M-2) are viewed as generally assorted. The β-lactamase genes are www.nature.com/scientificreports/ generally located on plasmids, which could advance the dissemination of β-lactamase genes in Gram-negative bacteria 13 . The blaTEM gene was the most prevalent broad-spectrum β-lactamase (BSBL)-encoding gene found in all the isolates of E. coli and Salmonella spp. from cloacal swab and sewage samples, which is consistent with previous studies conducted in Bangladesh 18 , Portugal 19 , and China 43 . Another BSBL-encoding gene, the blaSHV, was detected in one E. coli isolate from cloacal swab samples. Very little data is accessible on the event of β-lactamase encoding genes in isolates from chickens in Bangladesh, yet comparative outcomes have been accounted for in studies somewhere else 20,53 . No blaCTX-M-1 and blaCTX-M-2 ESBL-encoding genes were detected in any isolates of E. coli and Salmonella spp. in this study. These findings are conflicting with prior investigations in Bangladesh, where over 90% of E. coli isolates from droppings of chickens and water samples harboured the blaCTX-M-1 gene 9,24 . Presence of β-lactamase genes on genetic mobile elements can encourage their exchange across bacterial species or genera 13 . Plasmid-mediated quinolone resistance (PMQR) spoke to by quinolone resistance (qnr) genes is generally distributed among Gram-negative bacteria including Bangladesh 10,20 . PMQR genes have been reported to be carried on portable gene elements, and can be easily moved among various bacterial strains and species. This potential intensifies the development of multidrug resistance in light of the fact that PMQR supposedly decreases microbial susceptibility to antimicrobials and supports the event of resistance-associated mutations on bacterial chromosomes, accordingly making Enterobacteriaceae infections significantly more hard to treat 59 . In the current study, qnrS gene was detected in 70.5% of E. coli and 80% of Salmonella spp. isolates from cloacal swab samples. In Bangladesh, qnrS gene was detected with similar percentage (72.2%) of E. coli isolates acquired from cloacal swab samples in Mymensingh district 26 . However, in Uganda, qnrS gene was identified with low percentage (18.8%) of Salmonella spp. recovered from chickens 60 . Furthermore, 2.6% of E. coli isolates from sewage samples also harboured the qnrS gene, which is comparatively lower than the study conducted earlier in Bangladesh 24 . Moreover, qnrA gene was detected only in 9.1% of the E. coli isolates from cloacal swab samples which is inconsistent with the previous report, who detailed none of the E. coli isolates from chickens harboured qnrA gene 26 . High presence of the PMQR gene qnrS hence shows the capability of horizontal transfer of resistance genes 13 .
In this investigation, the occurrence of mecA gene in S. aureus isolated from cloacal swab and sewage samples were 47.2% and 25%, individually, which is significantly higher than that reported beforehand in Bangladesh 21 . The high occurrence of mecA gene in cloacal swab and sewage samples in our investigation showed that the overuse of antibiotics in poultry production that eliminate methicillin-sensitive Staphylococcus and encourage MRSA colonization.
The coexistence of β-lactamase and PMQR genes in E. coli and Salmonella spp. emerges a genuine worry to people. Notably, thirty one E. coli isolates from cloacal swab samples carried blaTEM and qnrS gene, while four isolates carried blaTEM and qnrA gene. These outcomes are in concurrence with the findings of prior study in China, which exhibit that the presence of β-lactamase and PMQR genes in the similar E. coli strain 22 . Likewise, blaTEM, blaSHV, qnrS genes were coexisted in one E. coli isolate from sewage samples. Until now, this is the first time that blaTEM, blaSHV, qnrS have been found to coexist in an individual E. coli isolate in Bangladesh. One Salmonella spp. isolates from cloacal swab samples possessed a combination of blaTEM and qnrS gene. Further, our information associate with previously referenced studies concerning the coexistence of various resistance mechanisms in one isolate 61 . β-Lactamase genes are often harboured on plasmids containing other resistance genes, e.g., PMQR gene, accordingly, the use of β-lactam antimicrobials enables their co-selection 61 . Further molecular analyses could be performed to build up the relatedness of the foodborne bacteria from the chicken samples to human isolates since the antimicrobial resistance genes assessed in this study can be easily moved to poultry and human strains. Furthermore, further study on the isolates should be led to depict the association between the presence and level of expression of the selected genes. To prevent the spread of antimicrobial resistance, further research using large sample collections is needed to better understand the molecular genetic pathways involved in the spread of antimicrobial resistance genes from foodborne pathogens to people. The results of the occurrence of foodborne bacteria and their pattern of antimicrobial resistance should be utilized with caution because few samples were collected and examined at a time point though the samples were representative of the target population. To extrapolate the results of our pilot study to the broader population, a follow-up study with large sample size is warranted.
In conclusion, our baseline data indicate the presence of common foodborne MDR bacteria in cloacal swab of broiler chickens sold in the wholesale chicken markets of Dhaka city of Bangladesh and the respective sewage samples, along with the existence of β-lactamase, plasmid mediated quinolone resistance, and methicillin resistance genes. The findings envisage the potential public health risk and environmental health hazard through spillover of common foodborne MDR bacteria. The results highlight the need for rapid implementation of an integrated program for surveillance of antimicrobial resistance in order to monitor trends, raise awareness, and improve practices with special emphasis to sanitary sewage system to safeguard next generation antimicrobial agents.

Materials and methods
Study design and areas. A cross-sectional sink survey was conducted in five wholesale chicken markets www.nature.com/scientificreports/ from 10 wholesale poultry stalls per market and 10 swab samples were collected from each stall and then pooled. In addition, five pooled sewage samples from dedicated sewage systems for the wholesale market, having one sample (pooled from five individual samples) from each market were collected. Data on origin of chickens, type of chickens sold, market distribution, daily average sales size, waste disposal, and drainage facilities were also collected.
Sample processing. Cloacal swab samples were collected by using sterile swab sticks. The exterior of the cloaca of chickens was cleaned with cotton soaked with 70% alcohol, and a sterile swab stick was inserted into the cloaca. Then the swab stick was dipped directly into the sterile collection tube containing 1 mL of sterile buffered peptone water (BPW). Besides, sewage water sample (~ 150 mL) was taken aseptically in a sterile falcon tube. After collection, all samples were kept in a cool box with ice packs, and transported to the laboratory on the same day. On getting to the laboratory, bacteria in sewage water were pelleted by centrifugation at 600×g for 20 min, and re-suspended in 1 mL of BPW.  www.nature.com/scientificreports/ S. aureus were performed from the pure culture. Positive isolates were stored in nutrient broth containing 50% (v/v) buffered glycerol at − 20 °C for further study.

Molecular detection of common foodborne bacteria.
Biochemically positive isolates of the bacteria were confirmed by polymerase chain reaction (PCR) assay. For molecular detection, the pure isolates of the organisms were sub-cultured overnight in NB and genomic DNA was extracted by using the "boiling" method as described by Dashti et al. 63 . Two uniplex PCR targeting malB promoter gene and ITS gene were used for the confirmation of E. coli and Salmonella genus, respectively. On the other hand, a duplex PCR was carried out for the confirmation of S. aureus with two sets of genus-and species-specific primers. Primers used for E. coli: ECO-1 (5′-GAC CTC GGT TTA GTT CAC AGA-3′) and ECO-2 (5′-CAC ACG CTG ACG CTG ACC A-3′) for the amplification of 585 bp 64 68 , and in some cases when breakpoints of some antimicrobials in CLSI were unavailable, the guideline of European Committee on Antimicrobial Susceptibility Testing (EUCAST) was used 69 . Extended spectrum β-lactamases (ESBL)-producing E. coli and Salmonella spp., and methicillin-resistant S. aureus (MRSA) were detected by using the double-disk synergy method and cefoxitin disk diffusion method, respectively 68 .

Definition of MDR and pXDR.
Isolates resistant to at least one agent in three or more antimicrobial classes were defined as multidrug resistance (MDR) while isolates resistant to at least one agent in all but two or fewer antimicrobial classes i.e. bacterial isolates remain susceptible to three classes were defined as possible extensively drug resistant (pXDR) 70 .
Detection of antimicrobial resistance genes. Detection of β-lactamase and plasmid-mediated quinolone resistance (PMQR) genes. The presence of β-lactamase-encoding genes (broad-spectrum β-lactamases: blaTEM, blaSHV, and extended spectrum β-lactamases: blaCTX-M-1, blaCTX-M-2), and PMQR genes (qnrA, qnrB, and qnrS) in E. coli and Salmonella spp. were determined by two separate multiplex PCR with specific primers listed in Supplementary Table S6. Details of the PCR protocol and thermal profile used for detection of β-lactamase, and PMQR genes were described in our previous study 71 . A representative numbers (blaTEM-8, blaSHV-1, qnrA-2, and qnrS-7) of PCR products were sequenced from commercial service provider (Macrogen Inc., Seoul, Korea). The identification of sequences was confirmed by comparison with known sequences in GenBank by using the BLAST program (National Center for Biotechnology Information, USA).
Detection of methicillin resistance gene. A uniplex PCR targeting methicillin resistance gene (mecA) in S. aureus was standardized, and used in this study with specific primer listed in Supplementary Table S6 66 . Each PCR reaction mixture was constituted in a final reaction mixture of 25 µL made up of 12.5 µL PCR master mix (Thermo Fisher Scientific, Waltham, MA, USA), 1.5 µL (15 pmol) each of forward and reverse primers, 7.5 µL of nucleasefree water, and 2 µL of DNA template. Amplification was performed with this thermal profile: heating at 94 °C www.nature.com/scientificreports/ for 5 min, followed by 30 cycles of denaturation at 94 °C for 1 min, primer annealing at 55 °C for 1 min, extension at 72 °C for 2 min, and a final extension step for 10 min at 72 °C.
Data analyses. Data were entered into spread sheet (Microsoft Excel 2010) and transferred into SPSS software v22.0 (IBM Corp., Armonk, NY, USA) for statistical analysis. Descriptive statistics were used to compute the prevalence of bacteria and resistance percentage. The significant differences in prevalence of bacteria and resistance percentage among sample types, sampling area were determined using Chi-square test (Z-test for proportions) and Fisher's exact test (wherever appropriate). The level of significance was set at p < 0.05. An UpSet plot was constructed to show the antimicrobial resistance pattern of E. coli, Salmonella spp. and S. aureus, and a Venn diagram to find out the co-isolation of bacteria was drawn by using online tools 72,73 . Map was generated to show the sampling sites in Dhaka city of Bangladesh using ArcGIS 10.4.1 Software (ESRI Redlands, NY, USA).
Ethical consideration. This study was approved by the Animal Welfare and Experimentation Ethics Committee of Bangladesh Agricultural University, Mymensingh. The approval number was AWEEC/BAU/2017 (13).
No animal experimentation was done in this study. However, informed written consent was taken from the live bird shop owners before sampling from broiler chickens. Cloacal swab samples were collected from the broiler chickens causing minimal distress to the chickens as per OIE guidelines 74 .

Data availability
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.