Antibiotic Resistance of Campylobacter Recovered from Faeces and Carcasses of Healthy Livestock

Campylobacter is of major significance in food safety and human and veterinary medicine. This study highlighted resistance situation in the area of veterinary public health in Ghana. Using selective mCCDA agar, isolates were confirmed phenotypically on API CAMPY and genotypically by multiplex PCR of IpxA gene. The susceptibility profile of species to common and relevant antibiotics was determined by the Kirby-Bauer disk diffusion method. Cattle, sheep, goat, and pig faecal samples analysed, respectively, yielded 13.2% (16/121), 18.6% (22/102), 18.5% (25/135), and 28.7% (29/101) Campylobacter species while 34.5% (38/110), 35.9% (42/117), 23.9% (32/134), and 36.3% (37/102) were, respectively, recovered from the carcasses. Species identified in faeces were C. jejuni 35.8% (33/92), C. jejuni subsp. doylei 4.3% (4/92), C. coli 47.8% (44/92), and C. lari 12.0% (11/92). Species discovered in carcasses were C. jejuni 83.9% (125/149), C. jejuni subsp. doylei 2.0% (3/149), C. coli 6.0% (9/149), and C. lari 8.1% (12/149). Resistance ranged from 92 to 97% to the β-lactams, 7 to 69% to the quinolones, 0 to 44% to the aminoglycosides, 97 to 100% to erythromycin, 48 to 94% to tetracycline, 45 to 88% to chloramphenicol, and 42 to 86% to trimethoprim/sulfamethoxazole as 0% resistance was observed against imipenem.


Introduction
Campylobacter is a key zoonotic pathogen which causes foodborne enteritis with C. jejuni and C. coli being the most isolated species [1]. Campylobacter infections in humans are mainly associated with consumption of undercooked chicken as exposure to farm animals, consumption of pork, improperly cooked beef, raw milk, and untreated water are other transmission routes. The significant contribution of ruminants as important reservoirs of Campylobacter has been established through molecular epidemiological research [2]. Due to growing demand for meat and products of livestock, the possibility of disease transmission from these food animals' sources cannot be dismissed. Increasing antibiotic resistance in Campylobacter from animal sources is well reported globally [3][4][5][6]. Resistance has developed to nearly all antibiotics used in veterinary medicine [7].
Livestock production in Ghana currently is changing from free range to commercial productions with increased use of antimicrobials as growth promoters and therapeutic agents. Such practice could increase levels of resistant bacteria in the gut-intestinal flora of animals, such as Campylobacter, and subsequently increase resistance in foods due to faecal contamination during slaughter. In Ghana, regional reports of heightening drug resistance among several pathogens from human sources have been documented [8]; however research on resistance of bacteria from animal sources is very sketchy. The purpose of this study was to highlight resistance trends in the area of veterinary public health. Therefore isolation rate and level of antimicrobial resistance among Campylobacter species from faeces and carcasses of livestock were reported.

Materials and Method
2.1. Study Site. The study was conducted at the Kumasi Abattoir. Kumasi is the capital of Ashanti region and the second most populous Metropolis in Ghana. The abattoir supplies Ghanaian markets with slaughtered, processed, and packaged goats, cattle, pigs, and sheep. The daily slaughtering capacity is about 200 cattle, 100 pigs, and 250 sheep and goats. Animals intended for slaughter at the abattoir are transported from different regions within Ghana especially from Yeji in Brong Ahafo and the Northern regions and also from neighbouring countries like Burkina Faso, Mali, and Niger.

Sample Collection.
Fresh faecal samples were collected from individual animals (cattle, pigs, goats, and sheep) from unrelated herds at the Kumasi Abattoir slaughterhouse complex. A single animal was selected at random and about 5 g of faeces was aseptically removed from the bowel after evisceration at the slaughtering line. Samples were collected into sterile ziplock bags, kept on ice and returned to the laboratory for processing. Carcasses of cows, sheep, goats, and pigs being packaged at the abattoir for the markets were randomly swabbed using sterile swab sticks and inoculated into sterile Amies Transport Medium (Copan ESwab sticks, Italy) and transferred to the laboratory on ice. Sampling took place from March 2013 to February 2014.

2.3.
Processing, Isolation, and Identification. About 0.2 g of faecal material was plated directly onto mCCDA agar plates (Oxoid CM0698) using sterile swab stick and incubated at 42 ∘ C for 48 hrs. Carcass swabs together with the transport media were aseptically transferred into sterile bijou bottles and preenriched with 5 mL of blood-free Campylobacter broth (Oxoid CM0963) supplemented with CCDA selective supplement (Oxoid, SRO 155E) and incubated overnight at 37 ∘ C. The overnight enrichment culture was cultured onto mCCDA agar plate and incubated at 42 ∘ C for 48 hrs. CampyGen (Oxoid CN0025A) was introduced to provide microaerophilic condition. Colonies showing typical morphology of Campylobacter spp. were subcultured onto Nutrient agar, followed by biochemical tests including Gram stain, oxidase, and catalase. Isolates which were small, curved, catalase, and oxidase positive; Gram negative bacilli were presumed to be Campylobacter spp. The presumed isolates were further subjected to standard phenotypic tests using API CAMPY system (bioMérieux, Marcy l'Etoile, France) to identify to species level.

Genotyping of Campylobacter Species by Multiplex PCR
2.4.1. DNA Extraction. Genomic DNA was isolated from cultures grown on 5% sheep blood agar (Accumix, AM5014, India) for 24 to 48 h at 37 ∘ C under microaerophilic conditions. Cell lysates were prepared by suspending a 10 L loopful of growth in 100 L of sterile distilled water in a microcentrifuge tube. The tubes were heated at 100 ∘ C for 10 min and subsequently cooled to 4 ∘ C. The tubes were centrifuged at 13,000 rpm (Sigma, Germany) for 5 min, and the supernatant was stored at −20 ∘ C for further analysis.

Genus-Specific PCR Amplification.
The primers of the lpxA gene (DNA Technology, Denmark) were used in this study [9]. Forward primers complementary to the lpxA nucleotide sequence of C. coli (lpxA C. coli), C. jejuni (lpxA C. jejuni), C. lari (lpxA C. lari), and C. upsaliensis (lpxA C. upsaliensis) were used in combination with the reverse primer lpxARKK2m for confirmation of Campylobacter species by multiplex PCR with expected amplicon (fragment) sizes ( Table 1). The reaction mixture consisted of 2 L Taq buffer 10x, 0.7 L Taq polymerase (Fermentas, UK), forward primer 50 pmol (0.5 L), 50 pmol (0.5 L) reverse primer, dNTPs 5 Mm, 0.4 L, and nuclease free water 12.9 L, 0.5 L of each primer, and 1.5 L of the genomic DNA template. The reaction tubes (20 L) were placed in the thermal cycler (Thermo PCR, Sprint, USA). The cycle involved initial denaturing at 94 ∘ C for 5 min. The denatured DNA was maintained at 94 ∘ C for 1 min followed by annealing of primers to the DNA template at 50 ∘ C for 1 min; the annealed primers were extended at 72 ∘ C for 2 min. This cycle was repeated 30x and a final fillingin at 72 ∘ C was carried out for 10 min [9]. The PCR products were run on 1% agarose gel (Sigma, Germany) at 100 V for 45 minutes; the separated PCR products (bands) were visualized with UV imager (Syngene, USA). A 1 kb ladder was used as a molecular size standard. Campylobacter strains obtained from Statens Serum Institute (Denmark) were used as positive control and a negative control was included and examined.

Statistical Analysis.
Data were transferred to a Microsoft Excel spreadsheet for analysis. Descriptive analysis was carried out using percentages. Associations were determined using the Chi-square test at a significance level of < 0.05. All statistical tests were two-tailed. Stata 14.0 software was used for statistical analysis.

Resistance Profile of Faecal Isolates of Animals.
Cattle isolates' resistance to erythromycin, chloramphenicol, and trimethoprim/sulfamethoxazole was, respectively, 100%, 88%, 56%, and 81% each to ampicillin and tetracycline. Against the cephalosporins resistance was 75% to cefotaxime and 81% to cephalexin. Resistance to the quinolones was 50% to nalidixic acid, 44% to ciprofloxacin, and 19% to norfloxacin. Resistance to the aminoglycosides was 6% to kanamycin and 0% to gentamicin.   Sheep isolates' resistance was 100% to erythromycin, 86% to trimethoprim/sulfamethoxazole, 91% each to ampicillin and tetracycline, and 55% to chloramphenicol. Against the cephalosporins resistance was 86% to cefotaxime and 91% to cephalexin. Resistance to the quinolones was 50% to ciprofloxacin and 36% each to nalidixic acid and norfloxacin. Against the aminoglycosides, resistance was 18% to kanamycin and 14% to gentamicin.
No resistance (0%) was recorded against imipenem; however intermediate susceptibility was observed among strains from all the animals (Table 6).
Resistance of C. lari to the quinolones was 100% to nalidixic acid, 93.3% to ciprofloxacin, and 83.3% to norfloxacin. Against the cephalosporins, resistance was 93.3% to the cefotaxime and 100% to the cephalexin. Resistance to aminoglycosides was 26.7% to gentamicin and 36.7% to kanamycin. Resistance to erythromycin was 100%, to ampicillin and tetracycline 96.7% each, to trimethoprim/sulfamethoxazole 90%, and to chloramphenicol 66.7% (Table 7).

Multidrug Resistance (MDR) in Campylobacter Species
from Animals. Resistance to three (3) or more antibiotics was defined as multidrug resistance (MDR) in this study. Strains of C. jejuni, C. coli, and C. lari, respectively, showed multidrug resistance of 66.6% (156/234), 20.5% (48/234), and 12.8% (30/234). Multidrug resistance in C. jejuni strains from the animal carcasses was consistently higher than faecal strains across studied animals. Contrastingly, higher MDR was observed in C. coli strains obtained from faecal sources of the animals. With the exception of sheep, the differences in MDR of faecal and carcass strains from pigs ( = 0.000), goats ( = 0.003), and cattle ( = 0.012) were statistically significant.

Discussion
The fastidious nature of Campylobacter coupled with its susceptibility to environmental stresses such as heat, drying, and exposure to air often results in damaged cells which hamper their recovery to a greater extent than most bacteria. Again, a viable but nonculturable (VBNC) state exhibited by Campylobacter can result in underestimation or nondetection of the organism by culture-based techniques, yet cells in this state can still infect susceptible hosts [10]. Notwithstanding, Campylobacter have been isolated from animals in different countries at varying rates. Documented range of 5-49% has been reported in sheep and goats [11,12], 0-80% in cattle, and 50-100% in pigs [13,14]. In our study, Campylobacter prevalence ranged from 13.2 to 28.7% in faecal samples and from 23.9 to 36.3% in the carcasses and again Campylobacter recoveries from carcasses were more than from faeces. This could be due to contamination of carcasses with intestinal contents during manual skinning, evisceration, carcass washing, and processing at the abattoir. Similarly, A Mpalang et al. [15] recorded 50% Campylobacter prevalence in carcasses compared to 20% in faecal samples. The 13.2% recovery rate from cattle faeces in this study compares to the 12.7% in Ethiopia but is lower than studies in USA (26.7%-29.1%), Finland (31.1%), and Canada (76-95%) [16][17][18][19]. Campylobacter contamination in cattle carcasses are generally low; however, Noormohamed and Fakhr [20] reported 78% from beef livers which is higher than the 35.2% obtained in this study. Lower isolation rates have also been reported in different countries [18,21]. The 18.6% Campylobacter prevalence in sheep faeces is lower than the 23% and 38.0% rates reported in studies in Ghana and Ethiopia, respectively [17,22], but higher than the 4.5% reported in USA [23]. In sheep carcasses, rates of 11.0% have been reported in Ethiopia which is lower than the 35.9% obtained in this study although a higher rate of 72.2% has been described in Greece [24,25]. The 18.5% Campylobacter recovery in goat faeces from this study was lower than the 33% and 20% reported by Abrahams et al. [22] and Salihu et al. [11] in studies in Ghana and Nigeria, but higher than the 3.2% reported in USA [15,23]. Goat meat is not known to be a major source of campylobacteriosis; however, the high utilization of mutton in local Ghanaian dishes may contribute to the increased transmission source. In goat carcasses, Campylobacter contamination rates of 50% and 63.5% have been reported from Congo and Greece, with 9.4% in Ethiopia compared to the 23.9% in our study [15,24,25].
Campylobacter prevalence of 28.7% recovered from pig faeces was lower compared to 42.4%, 50%, and 71.4% in studies in Japan, Ethiopia, and Mexico [17,26,27]. Geographical distribution of Campylobacter contamination in pig carcasses is fairly low (2.0-25.3%) [20,27] which is still lower than the 36.3% in this study. Campylobacter contamination in pig carcasses has been documented in various studies [28]. It must be noted that there were no significant differences in contamination rates of the various food animal carcasses which may be as a result of the levels of colonization of slaughter animals, abattoir hygiene, slaughter, and dressing methods [29].
Campylobacter jejuni and C. coli were the most commonly identified species although C. coli were more in faecal samples and C. jejuni were more in carcasses. Similar findings were made by A Mpalang et al. [15] who also recovered more C. coli (26.1%) than C. jejuni (10.1%) from faecal isolates of goats in Ethiopia; 25.9% C. coli and 3.4% C. jejuni in nondiarrhoeic goat faeces have been documented in South Africa [30]. These findings suggest that C. coli are more common in Africa [31]. Campylobacter jejuni dominance in carcasses (83.9%) is comparable to work by Wieczorek et al. [4] and Noormohamed and Fakhr [20]. However, a number of studies have rather shown higher C. jejuni in faecal samples of animals [17,23,26].
Although C. coli is the most commonly identified species in pigs [32,33]; an interesting pattern was discovered in our study where C. jejuni and C. coli isolations were similar (48.2% each) in pig faeces and a rather higher C. jejuni (75.6%) than C. coli (10.8%) in the carcasses. Matthew-Belmar et al. [34] and Kramer et al. [35] recorded more C. jejuni (53.5%) than C. coli (46.5%) from pigs in Grenada and UK, respectively. However, higher C. coli than C. jejuni have been recorded in Nigeria [36], in Ethiopia [17], and in Poland [21]. Other studies do show that C. jejuni may coexist with C. coli in pigs but usually the C. jejuni are always present in 10-100fold lower numbers than C. coli [37]. In Ghana, most of the farms rear multiple animals and it could be that the pigs may have acquired the C. jejuni from poultry on the same farm [23,38].
BioMed Research International 7 Conventional culture methods and API biochemical reactions of isolates from food animals were 95.8% in agreement with the results of PCR for identification and differentiation of Campylobacter species which is comparable with data from other studies [9,39]. The bacteriological methods (culture and API) are as reliable as the molecular PCR (IpxA gene) method in detecting Campylobacters from animals.
The limitation in detecting antibiotic resistance in this study was as it is in all studies dependent on culture that the resistance rates were determined according to the bacteria species that were culturable at the time of analysis. Direct PCR on the specimens might be more sensitive but not detecting the actual species where the antibiotic resistance genes are present in.
High levels of resistance were expressed against most of the antibiotics. Faecal and carcass strains showed resistance range of 97-100% to erythromycin which is consistent with work in Nigeria and Spain, where resistance of 81-82.6% to erythromycin has been described [3,5], but lower rates have been reported in Ethiopia (60.3%) and USA (55%) [6,20].
Resistance to tetracycline was between 58 and 94% which is consistent with documented rates in Poland, USA, Ethiopia, and Thailand [4,6,20,41] as resistance to chloramphenicol was in the range of 45-88% higher than rates of 61.5% and 67.4% reported against chloramphenicol in Ethiopia and Nigeria, respectively [5,6]. No resistance was observed against imipenem although intermediate susceptibility was found in both faecal and carcass isolates. Generally, resistant strains were commonly found in cattle and sheep compared to goats and pigs and resistance was also higher in the carcasses than in the faecal isolates.
The extensive application of antibiotics in animal husbandry for therapy, prophylaxis, and growth promotion has often been associated with the spread of resistance. Another factor contributing to the increase and spread of resistance is intensive rearing which promotes clinical infections in animals leading to widespread prophylactic usage of drugs which may be unwarranted. Currently in Ghana livestock production is changing from free range to commercial productions and may have added to the high level resistance currently documented.
Antibiotic use in animal feed as growth promotants also plays a significant role in the spread of resistance. Worldwide, antibiotics are widely used in livestock and poultry for growth promotion to enhance feed utilization and production [14]. In Ghana, 98% of livestock farmers use antibiotics on their farms as growth promoters and in the management of diseases. The antibiotics used are mainly tetracyclines (oxytetracycline, doxycycline, remacycline, and chlortetracycline), sulphadimidine, dihydrostreptomycin, piperazine, albendazole, tylosin, ivermectin, and benzylpenicillin which can lead to possible cross-and co-resistance [42,43]. In a study in Kumasi, the knowledge of livestock farmers on antibiotics, withdrawal periods, and dosages was very low and farmers usually depended more on fellow farmers than veterinarians for antibiotic knowledge. Poor dosing practices especially when an antibiotic failed to resolve an infection were a common practice. In such cases, different antibiotics were tried and abused until the disease was treated [43].
Also the challenge of distinguishing different antibiotic brands of the same active ingredient resulted in the application of different antibiotic brands of the same active ingredient which saw no improvement in the disease condition [43]. In the Northern Region of Ghana, Addah et al. [44] reported of nonadherence to dosing and withdrawal periods among several livestock farmers. These practices ultimately increase antibiotic residues in faecal content and carcasses of these animals which is evident in the high level resistance established in this study.

Conclusion
The study has revealed multidrug resistant Campylobacter species in the faecal content and carcasses of healthy livestock animals in Ghana indicating possible risks of infection to people through consumption of contaminated animal products or by direct contact with animals. Moreover high levels of resistance observed among the Campylobacter species to the common and cheap antibiotics raise uncertainties about their effectiveness in the treatment of animal and human diseases in the study region. It is urgent that extensive education and training are given to livestock farmers on judicious application of antibiotics and a national antibiotic resistance management team setup to monitor and control antibiotic use in both human and veterinary medicine.