Safety of ready‐to‐eat chicken in Burkina Faso: Microbiological quality, antibiotic resistance, and virulence genes in Escherichia coli isolated from chicken samples of Ouagadougou

Abstract In Burkina Faso, flamed/grilled chickens are very popular and well known to consumers. The aim of this study was to evaluate the microbiological quality, the antibiotic resistance, and the virulence gene from Escherichia coli isolated from these chickens in Ouagadougou. A total of 102 grilled, flamed, and fumed chickens were collected in Ouagadougou and analyzed, using standard microbiological methods. All E. coli isolates were checked with the antimicrobial test and also typed by 16‐plex PCR. The mean of aerobic mesophilic bacteria (AMB) and thermo‐tolerant coliforms (TTC) was found respectively between 6.90 ± 0.12 × 107 CFU/g to 2.76 ± 0.44 × 108 CFU/g and 2.4 ± 0.82 × 107 CFU/g to 1.27 ± 0.9 × 108 CFU/g. E. coli strains were found to 27.45%. Forty samples (38.24%) were unacceptable based on the AMB load. Fifty‐nine samples (57.85%) were contaminated with TTCs. Low resistance was observed with antibiotics of betalactamin family. Diarrheagenic E. coli strains were detected in 21.43% of all samples. This study showed that flamed/grilled chickens sold in Ouagadougou could pose health risks for the consumers. Need of hygienic practices or system and good manufacturing practices is necessary to improve the hygienic quality of flamed/grilled chickens. Our results highlight the need of control of good hygiene and production practices to contribute to the improvement of the safety of the products and also to avoid antibiotic resistance. Slaughter, scalding, evisceration, plucking, bleeding, washing, rinsing, preserving, grilling, and selling may be the ways of contamination.

and infections to Escherichia coli, shigellosis, brucellosis, hepatitis A, and salmonellosis. Among the most appreciated street foods, meat products occupied a very good position. However, the poor sanitary practices during cooking and sale multiply the risk of microbial contamination. High ambient temperatures especially in tropical environments have been described as the major factors responsible for facilitating the access and multiplication of bacterial contaminants in meat products (Barro et al., 2006;Mankee et al., 2003). E. coli which is the first germ of fecal contamination indicators are responsible for several infections. In developing countries, the majority of the population treat these infections by self-medication because of ignorance and refusal of cure.
This practice is one of the causes of the emergence of antibioticresistant bacteria (Rubin & Samore, 2002). In addition, the uncontrolled use of antibiotics in the livestock sector for animal diseases treatment and prevention as well as a growth promoters contribute to increase the spreading of antibiotic-resistant bacteria (Kagambèga et al., 2013). Antimicrobials are valuable means to treat clinical diseases and keep healthy and growth promotion.
However, the treatment of all herds and flocks with antimicrobials for increasing the growth and preventing illness has become an endless debate (Witte, 1998). Often whole flocks or herds of sick animals are treated at once, containing animals that are not sick. It is now generally accepted that the main risk factor for the increase in resistance to pathogenic bacteria is the anarchical use of antibiotics (Somda et al., 2017).
In Burkina Faso, more than 50,000 poultry are transported, killed, and consumed every day in Ouagadougou (DSMRA 2015).
The hygiene often fails during slaughtering, scalding, evisceration, plucking, bleeding, washing, and rinsing, and increase the health risk associated with the consumption of this meat (Coulibaly, Bakayoko, & Karou, 2010). The aims of this study were to assess (1) microbiological quality of ready-to-eat chicken borne, (2) antibiotic resistance, and virulence genes of E. coli isolated in these chickens.

| Samples collection
From September to December 2016, 102 chickens meats were collected from 102 producers in 59 sectors distributed on three crowns in Ouagadougou. These samples were composed of 66 grilled chickens (nine in crown 1, 28 in crown 2, and 27 in crown 3), 29 flamed chickens (two in crown 1, 18 in crown 2, and nine in crown 3), four fumed chickens (two in crown 1, two in crown 2, and 0 in crown 3), and three chickens prepared around fire (three in crown 1, 0 in crown 2, and 0 in crown 3). The samples were transported to the laboratory and kept at 4°C. The microbiological examination was started within the eight following hours.
All microbiological and molecular tests were carried out at Laboratoire National de Santé Pulique (LNSP) Ouagadougou, Burkina Faso.

| Microbiological analysis
Twenty-five grams of each meat sample (for chicken a mixture of meat from neck, breast, wings, and legs) was added into 225 ml of Buffered Peptone Water (Liofilchem, Teramo, Italy) and homogenized using a stomacher (400 Circulator; Seward, London, UK).
Dilutions of 10 to 10 were realized from the stock solution according to ISO methods (ISO6887-2 2004).

| Aerobic mesophilic bacteria's enumeration
These microorganisms were enumerated on the plat count agar (PCA) medium according to the standard (ISO4833 2003). Briefly, 1 ml of each dilution was plated on PCA and incubated at 30°C for 72 ± 3 hr. The enumeration of the colonies was carried out on two successive dilution dishes (less than 300 colonies and more than 10 colonies on a dish). The results were expressed according to the standard (NFV08-102 1998) using the following formula: Σ c = the total of colonies counted on all the retained dishes of two successive dilutions and of which at least one dish contains 10 colonies; V = volume of inoculum applied to each dish in milliliters; n1 = dish number retained at the first dilution; n2 = dish number retained at the second dilution; 0.1 = dilution factor; d = dilution ratio corresponding to the first dilution retained.
All the results were interpreted according to the standard (AFSSA 2007-SA-0174 2008). The threshold of detection of this germ was 10 8 CFU/g.

| Thermo-tolerant coliforms (E. coli) enumeration
Seeding was carried out in a double layer on violet red bile lactose agar and incubated at 44°C for 24 hr according to standard (NFV08-053 2002). Bacteria were enumerated according to the formula described above. The threshold of detection of this germ was 10 2 CFU/g (AFSSA 2007-SA-0174 2008). After enumeration, two or three colonies were subcultured on Bromocresol purple agar and eosin methylene blue agar and incubated at 37°C for 24 hr.
E. coli were characterized using API 20E system and isolated strains preserved at −30°C for antibiotic resistance test and molecular characterization.

| Coagulase-positive staphylococci enumeration
About 0.1 ml of the suspension was streaked on to the Baird Parker agar with egg yolk with potassium tellurite and carefully spread with a spreader (ISO6888-2/A1 2003). Incubation was carried out at 37°C for 24-48 hr. All the dishes where there was growth of small black, shiny colonies surrounded by a transparent halo and an opaque border of 2-5 mm in diameter were retained to enumerate. The enumeration was carried out according to the formula described above. The threshold of detection of this germ was 10 3 CFU/g according to AFSSA 2007-SA-0174 (2008).

| Antibiotic resistance test
All isolates were also tested for susceptibility to 14 different anti-

| 16-plex PCR assay
The presence of STEC, EPEC, ETEC, EIEC, and EAEC on grilled chicken meat samples was detected by 16-plex PCR for the genes uidA, pic,bfp,invE,hlyA,elt,ent,escV,eaeA,ipaH,aggR,stx1,stx2,estIa,estIb, and ast. The primers and PCR conditions were as previously described (Antikainen et al., 2009). The nucleotide sequences and predicted sizes of the amplified products for the specific oligonucleotide primers used in this study are shown in Table 1

| Microbiological analysis
The average loads of various microorganisms determined are summarized in   chickens around the fire) of the samples were unacceptable (superior to 10 2 CFU/g) according to their load of TTCs (

| Antimicrobial susceptibility testing of griller/ flamed chickens isolates
Globally, the E. coli isolated from fumed, grilled, and flamed chickens showed a low resistance to antibiotic teste. All isolates were sensitive to imipenem, ciprofloxacin, nalidixic acid, norfloxacin, gentamicin, aztreonam, and chloramphenicol. However, resistance was observed with cefotaxime (7.14%), ceftriaxone (10.71%), sulfamethoxazole/trimethoprim (32.14%), ticarcillin (39.3%), amoxicillin/clavulanic acid (46.43%), ampicillin (42.86%), and tetracycline (64.3%) (Figure 1). Twenty-one percent (6/28) of tested strains were extended-spectrum-beta-lactam (ESBL) positives.  meat. The contamination could be also at the level of the sale material, the seller himself, and also the quality of the grilling (insufficient time of grilling) as stated some researchers (Khallaf et al., 2014). Our survey results showed that during the grilling process, some others sellers used dirty water to reduce fire intensity when it is high. The charcoal used for grilling could also be the source of the contamination because the conditions of production of the charcoal do not obey the rules of hygiene. In addition, others sellers were alone and were at the same time responsible for the whole chain of meat production. They slaughtered the chickens, scalded them, eviscerated them, plucked them, washed them, preserved them, burned them, and even received the money. These practices could increase the risk of contamination of meat at the end of production. Barro et al. (2006) and Gedik, Voss, and Voss (2013) showed that money handling constitutes another risk factor of street foods contamination.

| D ISCUSS I ON
Money can get contaminated and may thus play a role in the transmission of microorganisms to other people, during the selling.
Our results showed that the great majority of E. coli isolates from grilled/fumed chicken were susceptible to ciprofloxacin, chloramphenicol, norfloxacin, nalidixic acid, aztreonam, imipenem, and gentamicin. However, resistance was observed to antibiotics of betalactamin family such as ampicillin, amoxicillin/clavulanic acid, ceftriaxone, ticarcillin cefotaxime, and others families such as tetracycline and trimethoprim/sulfamethoxazole. ESBL was found in six strains. Antimicrobials are used in the absence of illness to prevent diseases when animals are susceptible to infection (Turtura, Massa, & Ghazvinizadeh, 1990). This practice is very usual in developing countries where outbreak is caused by enteric pathogens which are the sources of farms poultry diseases. Our results were in agreement with previous studies realized on fresh and broiler meat chickens in Canada, Spain, in Iran (Cortés et al., 2010;Vincent et al., 2010 (Farooq, Hussain, Mir, Bhat, & Wani, 2009;Nzouankeu et al., 2010). E. coli strains as STEC can be transmitted via the fecal-oral route and contamination often occurs through the ingestion of contaminated food or water, direct contact with animals, interhumans, or contaminated objects, or more rarely by inhalation (Crump et al., 2002;Grant et al., 2008). Presence of the DEC in grilled/flamed chicken meat refect poor food hygiene and the common occurence of potential pathogens from human in Burkina Faso. Hygiene rules must be applied strictly to the slaughterhouse and in the sale outlets to avoid contamination of grilled/ flamed meat by DEC.
Our study showed that most of our samples analyzed contain amounts of germs that exceed the acceptable limits to the standards established for both mesophilic aerobic total flora, fecal coliforms, and diarrheagenic E. coli pathogroups. It is necessary to set up good hygienic practices in the production of grilled, fumed, and flamed chickens to ensure the quality of finished products. In addition, intervention strategies, such as promoting hand washing with soap and good hygienic practices at the slaughterhouses and sales outlets, can have a sound practical impact on public health.

| CON CLUS ION
This study shows also the emergence of betalactamin resistance to E. coli isolates in grilled, fumed, and flamed chickens. The resistance of E. coli to betalactamins highlights the need for the establishment of a network and continuous monitoring of antibiotic resistance.

ACK N OWLED G EM ENTS
The authors gratefully acknowledge the ARES for funding this study through the project Qualisani and the Laboratoire National de Santé Publique/Burkina Faso for laboratory facilities provided.