Escherichia coli commonly colonize the gastrointestinal tract of animals and some of its strains instigate gastroenteritis, urinary tract infections and meningitis.1 According to WHO, diarrheal diseases associated with E. coli account for over 4% of the total daily global disease burden and about 1.8 million deaths each year, of which 90% are children.2 Avian Pathogenic Escherichia coli (APEC) in poultry causes colibacillosis, septicemia and cellulitis and it may also link to the extra-intestinal pathogenic E. coli strains in humans. These extra-intestinal pathogenic E. coli strains cause diseases in outside the intestinal tract of humans.3 To control bacterial infections in humans and animals, huge amounts of antibiotics are used. A number of antibiotics like β-lactams and quinolones are commonly recommended for poultry diseases.4,5 In many Asian and African countries, large amounts of antibiotics are frequently added in broiler feed as antimicrobial growth promoters and results in antibiotic resistance.6 Enzymatic degradation of antimicrobial drugs, alteration in bacterial proteins and changes in membrane permeability to antibiotics are mechanisms of development of antibiotic resistance in bacteria. In intensively reared animals, antibiotics are administered to whole flock rather than the individual animal. Therefore, antibiotic selection pressure against resistance in bacteria in broiler is high and their feces are often loaded with antibiotic resistant microbiota.7 In Pakistan, for example, fecal E. coli of food animals were found resistant against many commonly prescribed veterinary medicines.8
At slaughter, resistant strains from gut readily spoil the poultry carcass and consequently, meat is often contaminated with multi-resistant E. coli.9 Antibiotics not only select the pathogenic bacteria for resistance but also induce resistance in microbiota of exposed individuals10 and thus it becomes difficult to treat human ailments.11 Resistant APEC may transfer their resistant genes to humans through food chain and cause complications in the treatment of urinary tract infections.3,12 Resistant fecal E. coli may become part of human intestinal microbiota directly by individual exposure or indirectly via food chain.10 Therefore, resistant E. coli from poultry may contaminate the environment and are disseminated throughout the ecosystem.13 Despite uncontrolled use of antibiotics, particularly in developing countries, at poultry farms and high prevalence rate of antibiotic resistance in E. coli isolates of broiler, there is no latest report on spread of resistant E. coli in the environment.
Tetracycline-resistance genes were identified in E. coli isolates of meat and fecal samples of broiler14 and such resistant genes may spread from poultry feces to humans.10,15 Transfer of CMY-2 AmpC β- lactamase plasmids of E. coli from food animals to humans has also been reported in Iowa, United States.16 Increase in resistance of Gram-negative bacteria is mainly due to presence of mobile genes that may spread through bacterial population. Furthermore, unprecedented human air travel and migration allowed resistant-bacterial plasmids and clones to transport between countries. Owing to transfer of resistant bacteria, antimicrobial resistance has become a major threat to global public health against effective prevention and treatment of an ever-increasing range of infections caused by E. coli.17-19
In many developing countries, including Pakistan, India, Bangladesh, and Afghanistan, more than 90% of human population consumes broiler meat purchased from local poultry processing markets located in open environments of streets, roads and populated areas of cities. Skinning and evisceration are done without taking care of contamination of meat with feces and blood. Unhygienic practices of using unclean knives, dirty containers, and unwashed cutting boards are the possible routes of contamination from poultry feces to poultry meat and surrounding environment. In Pakistan, these poultry shops are the major source of poultry meat.
Keeping in view the unhygienic conditions at the poultry shops, we speculated that these poultry shops might be spreading resistant E. coli originating from broiler feces. The key objective of our study was to evaluate spread of fecal E. coli and its resistance determinants through poultry slaughtering and processing practices adopted at local markets.
Materials and Methods
Fecal and meat samples (n=225 each) of broilers were randomly collected from different poultry shops located in four cities (Khanewal, Multan, Dera Ghazi Khan and Bahawalpur) of South Punjab, Pakistan. Not any single bird was specifically sacrificed for this study. Swabbing of slaughterers’ hands, consumers’ hands, slaughtering knives, cutting boards, canisters and broiler carcasses was done (n=50 each) from randomly selected poultry shops. A sterilized swab, premoistened with phosphate buffer saline (PBS), was used for each sample. To cover maximum area, the swab was rolled three times for each sample. Each swab was dipped and vortexed in 5 mL of brain heart fusion broth for 10 s.20
Samples of broiler feeds and drinking water were also randomly collected from these poultry processing shops. Swabbing of consumer’s hands was done before their entrance to the poultry shops (as mentioned above) and samples of tap water (unfiltered and untreated) were collected after running the tap for five minutes. Collected samples were placed in sterile plastic bags and stored at 4°C until further experimentation. All the chemicals and materials used in this study were purchased from Oxoid (UK) or Sigma- Aldrich (USA).
Isolation and colony counts
For isolation of E.coli, fecal, meat and feed samples were homogenized in Butterfield’s phosphate-buffered water. The diluted blends and water samples (tap/drinking) were inoculated in Lauryl Tryptose Soya (LTS) broth following incubation.21 All swabs were separately enriched in Lauryl Tryptose Soya (LTS) broth. Positive samples from LTS broth were identified with gas production and transferred on Levine’s Eosin Methylene Blue (L-EMB) agar.22 Dark centered and flat colonies with or without metallic sheen were isolated and confirmed as E. coli by IMViC and sugar biochemical reactions. Colonies of E. coli were counted and reported in CFU/ml, CFU/g or CFU/cm2.
Antimicrobial resistance pattern of isolated E. coli strains were carried out with disk diffusion method on Mueller Hinton agar.23 Ten drugs belonging to eight antimicrobial categories [1st generation cephalosporins (cephradine), 3rd generation cephalosporins (ceftriaxone), phenicols (chloramaphenicol), aminoglycosides (gentamycin), quinolone/ flouroquinolone (nalidixic acid, ciprofloxacin), penicillin (penicillin, amoxicillin), tetracycline (oxytetracycline) and macrolides (azithromycin)] were used in this study. Results on E. coli isolates were classified to sensitive and resistant groups. Antibiotic control strain E. coli ATCC 25922 was used for standardization.24
Resistant isolates were further classified to possible multi-drug resistant (MDR), confirmed MDR, possible extensively-drug resistant (XDR) and possible pandrug resistant (PDR).25 Possible MDR isolates were resistant to at least one drug in less than three out of the used eight antimicrobial categories. Confirmed MDR isolates were resistant to at least one drug in at least three antimicrobial categories. Any confirmed MDR isolate that was resistant to one drug in all but less than or equal to two antimicrobial categories was named as possible XDR and a possible XDR isolate resistant to all the used ten drugs was named as possible PDR.
All tetracycline-resistant isolates were further observed for presence of tetracyclineresistance genes through PCR. Bacterial genomic DNA was extracted by method of Seidavi et al.26 and confirmed by 16s rRNA gene using oligonucleotide primers ECO-F GACCTCGGTTTAGTTCACAGA and ECO-R CACACGCTGACGCTGACCA with product size 585bp.27Moreover, tetracycline resistant genes (tetA and tetB) were amplified with primers tetA-F GGTTCACTCGAACGACGTCA, tetA-R CTGTCCGACAAGTTGCATGA, tetB-F CCTCAGCTTCTCAACGCGTG and tetB-R GCACCTTGCTGATGACTCTT.28 PCR reactions were carried out in total volume of 25 μl containing 5 μL DNA, 2.5 μL buffer × 10, 2 μL MgCl2, 5 μL dNTP mix, 0.25 μL of each primer and 0.2 μL Taq polymerase.29 Denaturation process were carried out at 95°C for 5 min and amplification was performed at 94°C for 60 s, 50°C for 45 s, and extension at 72°C for 90 s, with further extension at 72°C for 300 s. PCR products were analyzed by electrophoresis in 1.5% agarose gel and stained by ethidium bromide.
Association of prevalence of E. coli and antibiotic resistance with sample categories was estimated by employing Chi-square (χ2) test on SPSS version 21 (IBM Corporation, USA).
Prevalence percentage and colony Counts
Greater than 80% isolates of all sample categories had E. coli except that of consumer hands (6%) and tap water (12%) (Table 1). All collected samples of feces, meat, canister and carcass were 100% positive for E. coli. Prevalence percentage of E. coli in samples of slaughterer hand, knife and cutting board was statistically similar (about 92%).
Observed load of E. coli was more than double in broiler feces (4.1×104 CFU/g) than meat (1.9×104 CFU/g) samples (Figure 1). Drinking water of broiler had about six times more E. coli load than tap water (1.5×102 CFU/mL). Similarly, slaughterer hands (1.4×104 CFU/cm2) had much greater load of E. coli than on consumer hands (11 CFU/cm2).
Antibiotic resistance and MDR
In present study, antibiotic resistance of E. coli isolates from all sample categories against the 10 antibiotics ranged from 0 to 100% (Table 2). Most E. coli isolates from feces, meat, carcass and canister were resistant against eight antibiotics. Surprisingly, level of resistance against oxytetracycline was >80% in all sample categories except samples of consumer hands (0%), tap water (4%) and drinking water (20%). Comparing the results of resistance of feces and meat, significant (P<0.05) difference in antibiotic resistance against gentamycin, nalidixic acid, penicillin, cephradine, amoxicillin, azithromycin and ciprofloxacin was found. E. coli isolates from consumer hands were completely sensitive to all antibiotics while isolates from tap water showed minimum resistance against the tested drugs. Majority of isolates from all sample categories were sensitive against ceftriaxone except isolates of feces and cutting board.
All isolates from slaughterer hands and knife were susceptible to azithromycin (100%) and no significant difference was observed for ceftriaxone, gentamycin, penicillin, cephradine, oxytetracycline and ciprofloxacin for both sample categories.
On average, about 94% of the isolates from various sample categories were resistant to one or more of the drugs from ≥3 antimicrobial categories tested. Alarmingly, 51 isolates from broiler feces and 24 isolates from broiler meat were resistant to all drugs except two and were classified as possible XDR. Moreover, 4 isolates from broiler feces and 1 isolate from canister were completely resistant to all the drugs tested.
PCR products for tetracycline-resistance genes were obtained for 95 faecal, 57 meat, 6 slaughterer hand, 7 slaughterer knife, 6 cutting board, 5 canister, 7 carcass, 8 feed and 2 drinking water isolates (Table 3). For all sample categories, prevalence percentage of genotypic results of tetracycline-resistance was lower than results of phenotypic percentage. The positive rate of tetA was greater in feces (44%) while almost same prevalence percentage for tetA and tetB was observed in all sample categories. Study demonstrated presence of tetA, tetB and tetA+tetB in all sample categories except in isolates of consumer hands and tap water.
Escherichia coli is the commensal bacterium and reside in the gut of animals including broiler.30 Similar to this study (Figure 1, Table 1), greater prevalence of E. coli in poultry feces was observed in previous study as compared to other sample categories. 10 Contaminated poultry feed and water are major sources of E. coli for broiler. 31,32 Prevalence of E. coli in drinking water (86%) and feed (84%) at poultry shops (Figure 1, Table 1) was greater than reported in water (19%) and feed (35%) samples of broiler farms.31-33 This is probably because of contamination of feed and water pots with broiler feces at poultry shops. Amin et al.34 observed E. coli in about 8% samples of tap water of Pakistan that was in line to present study (Table 1). At poultry shops, tap water is used for the washing of utensils, tables, cutting knives, and hands. Relatively lesser E. coli load in tap water means that it was not the major source of high E. coli prevalence in various samples collected from slaughtering and processing markets.
During slaughtering, carcass of poultry may be contaminated with the gut contents from which E. coli may spilled out as a contaminant. Seidavi and collaeuges,26 found that 88%, 38% and 25% samples respectively of cecum, ileum and duodenum (gut contents) yielded E. coli in healthy broiler chickens. Feces of broiler, gut contents and carcass of bird are considerably major sources of contamination of meat. High prevalence percentage of E. coli in meat (100%) was in line to study of India (98%) while higher than of Morroco (48%), Washington (39%) and Bangkok (25%).35-38 Presence of E. coli in meat poses serious health issues for consumers at large. Extremely high prevalence rate pinpointed precarious and dangerous unhygienic conditions of slaughtering at retail butcher shops.
We assumed that the screening and then colonization under selection of antibiotics from extraneous sources resulted in resistant E. coli colonization of the chicken gut. Work on poultry showed the presence of different antibiotics in reasonable amounts that could select E. coli transferred to poultry meat.39 Resistant bacteria from feces of bird may spread to meat40 and possibly to surrounding periphery. In most categories of samples, resistant E. coli were detected (Table 2). Resistance in broiler E. coli might have been developed due to extensive use of antibiotics at poultry farms.41,42 Majority of isolates from all sample categories were resistant to oxytetracycline (Table 2). Tetracycline resistant E. coli isolates from poultry meat, feces, water, carcass and poultry products was also examined in a number of previous studies.15,30,33,43,44 Elevated level of tetracycline resistance might suggest widespread and extensive use of tetracycline at poultry farms.10 Tetracycline is the first line drug used for prophylaxis and for growth promotion of livestock.45 Determined resistance of E. coli against oxytetracycline (84%) in poultry feed was much higher than the study of Portugal that showed maximum resistance of 41% against tetracycline.39 These differences in antibiotic resistance in E. coli among previous and the present studies might be due to the differences in purity, dosage, price, laws, access, availability and awareness regarding usage of antibiotics in different regions.46 In Switzerland, for example, where usage of antibiotic is monitored by the government agencies, low levels of antibiotic resistance against cephalosporins were observed in isolates of poultry faeces.47Quinolone is next to tetracycline that is commonly used at clinical sites.48 Frequency of resistance in Escherichia coli isolated from meat and faecal samples (63-72%) to nalidixic acid (Table 2) was lesser than findings of Miles15 who reported 85% resistance of nalidixic acid against E. coli. Greater resistance level of E. coli isolates against first generation cephradine (Table 2) suggested that the presence and routine use of antibiotics. This exerts selective pressure in poultry farms, allowing Cephradine-resistant pathogenic E. coli strains to dominate the intestinal microbiota of the birds. Compared with other drugs used in this and a previous study,49 sensitivity of E. coli isolates to ceftriaxone in poultry meat perhaps was due to low usage of this antibiotic at broiler farms. Complete resistance (100%) for two drugs importantly in fecal samples indicated routine use of these antibiotics in animal feed again driven up penicillin, tetracycline, quinolones and fluoroquinolone resistance rates. In present study, isolates exhibited alarmingly higher multi-drug resistance (Table 4). Prevalence of such highly resistant isolates poses a challenge not only for poultry industry but also in ailment of humans. Observed MDR, XDR and PDR probably evolved through excessive usage of multiple antibiotics in poultry for growth promoters and therapeutic purposes- a practice that needs to be modified greatly if poultry farmers want to control spread of pathogenic E. coli.
Reports on dissemination and amplification of resistant genes including that for tetracycline-resistant genes in the environment are available.50 Detection of tetA and tetB genes carried by all studied groups except control groups of consumer hands and tap water (Table 3) demonstrated the distribution of similar resistant determinants in diverse genetic background. This probably indicated the continuous exposure to tetracycline resulted in high percentage of tetracycline-resistant E. coli and these isolates may have diversity of resistance genes. Presence of tetracycline-resistance genes in above mentioned sample categories showed significant level of crossresistance among all routes relevant to broiler meat. This study was in agreement to the results of Miles et al.15 who determined tetracycline-resistance in diverse genetic background. This elevated level of dissemination of tetA and tetB genes in present study suggested limited therapeutic options for broiler colonized with tetracycline- resistant E. coli. It has already been documented that indiscriminate usage of antimicrobial agents in poultry industry had led to dissemination of antibiotic resistant genes from poultry to humans.40 It is further suggested that colonization of these resistant species will effect human gastrointestinal microbiota by transfer of their plasmids to endogenous microbiota.10 Therapeutic use of drugs, such as quinolones, in animal feed had resulted in resistance in Enterobacter species in humans.51 Therefore, in-feed antibiotics must be replaced with suitable alternatives.52
Prevalence of resistant E. coli in most sample categories clearly depicted the level of cross contamination of broiler meat from different routes during slaughtering and processing practices. Greater resistant isolates from all sample categories, specifically in feces and meat, against most antibiotics presented an alarming situation. Spread of tetracycline-resistance genes may precariously depict broiler meat as a danger for human health. These resistant determinants of E. coli may not only pose threat for slaughterers but also confront several risks to control most common diarrheal outbreaks in southern Punjab, Pakistan and similar areas. Greater MDR isolates revealed the possibility of indiscriminate use of antibiotics at poultry farms. For the provision of safe and healthy protein source for humans, regulatory authorities must strictly govern the use of antibiotics at poultry farms. There is a dire need to educate poultry farmers for the use of only permitted antibiotics within the recommended dosage. Moreover, slaughterers must be educated how to prevent cross contamination and adopt hygienic practices during slaughtering and processing of chicken meat for sale to prevail the situation. Contaminated raw meat loaded resistant E. coli is unfit for human consumption until proper cooking. To control disease burden, therefore, there is dire need to vigorously screen broiler meet for resistant E. coli.