Isolation, characterization, and antimicrobial susceptibility pattern of Escherichia coli O157:H7 from foods of bovine origin in Mekelle, Tigray, Ethiopia

Escherichia coli O157:H7 is an emerging and major zoonotic foodborne pathogen. It has an increasing concern about the spread of antimicrobial-resistant strains. This study aimed to isolate and characterize Shiga toxin-producing E. coli O157:H7 from raw milk, yogurt, and meat of bovine origin and determine their antimicrobial susceptibility pattern. A cross-sectional study was conducted from December 2014 to June 2015, and a total of 284 milk and meat samples were collected from different sources in Mekelle. The collected samples were analyzed for the presence of E. coli and Shiga toxin-producing E. coli O157:H7 and the determination of their antimicrobial susceptibility pattern following the standard bacteriological and molecular techniques and procedures and antimicrobial sensitivity test. Out of the total 284 samples, 70 (24.6%) were bacteriologically positive for E. coli and 14.3% were found to be Shiga toxin-producing E. coli O157:H7. Of note, 100% of E. coli isolates carried the pal gene and 41.7% eaeA gene (EHEC). Of these EHEC isolates, 40% and 60% were positive for stx1 and stx2, respectively. E. coli isolates showed the highest level of susceptibility to gentamycin (91.7%) but the highest level of resistance to amoxicillin (95.8%). Of the tested isolates, 18 (75%) of E. coli showed multidrug-resistant. This study revealed the occurrence of Shiga toxin-producing E. coli O157:H7 in foods of bovine origin in the study area. In conclusion, a nationwide phenotypic and molecular characterization, in-depth typing, and drug-resistant gene identification of E. coli O157:H7 should be undertaken.


Introduction
In Ethiopia, both food shortage and lack of appropriate food safety assurance systems are problems that have become obstacles to the country's economic development and public health safety (1,2). Animal products are generally regarded as high-risk commodities in respect of pathogen contents, natural toxins, and other possible contaminants and adulterants (3). Foodborne microorganisms are major pathogens affecting food safety and cause human illness worldwide as a result of the consumption of foodstuff, mainly animal products contaminated with vegetative pathogens or their toxins. Most of these microbes have zoonotic importance, resulting in a significant impact on both public health and economic sectors (4). Moreover, the emergence of multidrug-resistant (MDR) pathogens presents a global challenge for treating and preventing disease spread through zoonotic transmission (5).
Data regarding foodborne diseases are extremely scarce at a national level, and a few studies conducted in different parts of the country showed the poor sanitary conditions of catering establishments and the presence of pathogenic organisms such as Campylobacter, Salmonella, Staphylococcus aureus, Bacillus cereus, and Escherichia coli (6)(7)(8)(9)(10).
Escherichia coli found in humans can be categorized on basis of genetic and clinical criteria into the following three main groups: commensal, pathogenic (enteric or diarrheagenic), and extraintestinal pathogenic E. coli (ExPEC) (11). The typical diarrheagenic strains include enterotoxigenic (ETEC), enterohemorrhagic (EHEC), enteroinvasive (EIEC), enteropathogenic (EPEC), and enteroaggregative (EAEC) E. coli (12,13). The E. coli that cause enteric disease have been divided into pathotypes, based on their virulence factors and mechanisms by which they cause disease (11,14). One of these pathotypes, called Shiga toxin-producing E. coli (STEC), refers to those strains of E. coli that produce at least 1 member of a class of potent cytotoxins called Shiga toxin. The STEC is also called verotoxin-producing E. coli (VTEC). The names Shiga toxin (Stx), derived from similarity to a cytotoxin produced by Shigella dysenteriae serotype 1 (15), and verotoxin (VT), based on cytotoxicity for Vero cells (16) (11). The ability to produce Shiga toxin was acquired from a bacteriophage presumably directly or indirectly from Shigella (20). Shiga toxin-producing E. coli is recognized as an important cause of food-borne disease in humans and causes large outbreaks worldwide (21). E. coli O157:H7 is the leading cause of hemorrhagic colitis and hemolytic-uremic syndrome (HUS), a life-threatening sequela characterized by a triad of acute renal failure, microangiopathic hemolytic anemia, and thrombocytopenia in humans (22, 23). People of all ages are susceptible to infection with STEC. However, the young and the elderly are more susceptible and are more likely to develop more serious symptoms (24). Domestic and wild animals are sources of EHEC O157:H7 but the major animal carriers are healthy domesticated ruminants, primarily cattle, and to a lesser extent, sheep, and possibly goats. Fresh meat and raw milk are, nevertheless, considered common vehicles for E. coli, particularly for the EHEC (O157:H7) strain. Contamination of meat usually occurs during animal slaughter, as a result of poor slaughter practices, abattoir hygiene, and animal handling practices (20, 25).
Foods of bovine origin are frequently implicated in human outbreaks of Shiga toxin-producing E. coli (STEC) O157 (26,27). Shiga-toxigenic E. coli is transmitted by the fecal-oral route by either consumption of contaminated food or water, from direct contact with infected animals, or via person-to-person contact. Moreover, a series of studies on the resistance of E. coli isolated from animals and humans have strongly suggested that those bacteria that are resistant to antimicrobials used in animals would also be resistant to antimicrobials used in humans (28-30). In general, antimicrobial resistance is a global public health problem, and growing scientific evidence indicates that it is negatively impacted by both human and animal antimicrobial usage (31). Although antimicrobial therapy is not the primary tool for treating infections caused by STEC O157:H7, MDR STEC O157:H7 is a public health issue as those strains participate in a reservoir of resistance genes that could be easily exchanged between Enterobacteriaceae in the host and the environment (32). However, less is currently known about the molecular basis of MDR in STEC O157:H7 isolates of food origin (33). Therefore, it is of paramount importance to systematically investigate and characterize this recurring food-borne disease. Thus, the current study was conducted to isolate and characterize Shiga toxinproducing E. coli O157:H7 from raw milk, yogurt, and meat of bovine origin from different sources in the study area Frontiers in Veterinary Science frontiersin.org . /fvets. . and to determine the antimicrobial susceptibility pattern of the isolates.

Study area
This study was conducted from December 2014 to June 2015 in Mekelle City. Mekelle is the capital city of Tigray Regional State, situated approximately 783 km North of Addis Ababa at 38.5 • East longitude and 13.5 • North latitude at an altitude of 2,300 above sea level. The city has a total population of 215,546 (34), 308 cafeterias, 292 restaurants, 258 supermarkets, and an active urban-rural exchange of goods which has 30,000 micro and small enterprises (35). The weather condition is hot and dry. The mean annual rainfall of the area is 628.8 mm and an annual average temperature of 21 • C (36).

Study design and study population
A cross-sectional study on Shiga toxin-producing E. coli O157:H7 was conducted from December 2014 to June 2015 on raw milk, yogurt, and meat samples collected from different sources and parts of Mekelle City, Tigray, Ethiopia. The study populations comprised purposively selected milking dairy cows and slaughtered cattle found in Mekelle City.

Sampling technique and sample collection
A total of 284 samples of bovine origin, comprised of raw milk (n = 145), yogurt (n = 48), and meat (n = 91), were collected using a purposive random sampling technique. These samples were collected based on the willingness of the owners until the required sample size was achieved. Raw milk samples were aseptically collected directly from the teats of lactating cows (n = 100), whole-sellers (n = 17), cafeterias (n = 28), and similarly the yogurt samples were collected from dairy farms (n = 26) and cafeterias (n = 22) using a sterile universal bottle. However, the raw meat samples were collected from abattoirs (n = 55), butchery shops (n = 16), and restaurants (n = 20) during slaughtering and selling. Then, the sections of meat samples were aseptically removed and placed in separate sterile plastic bags to prevent spilling and cross-contamination. Both samples were transported to the Veterinary Microbiology Laboratory of the College of Veterinary Sciences, Mekelle University using an icebox and stored at +4 • C until the laboratory work was conducted.

Isolation and identification of E. coli and E. coli O :H Microbiological procedures
Isolation of E. coli was attempted according to Quinn et al. (37) with slight modification. A part of each sample (10 ml or 10 g) was enriched in peptone water (HiMedia, Mumbai, India) (90 ml) and was incubated at 37 • C for 24 h. Enriched samples were inoculated on MacConkey Agar (MCA) (HiMedia, Mumbai, India) by four flame techniques, and plates were incubated at 37 • C for 24 h. Pink-colored colonies were considered presumptive of E. coli. Gram staining was performed as per procedures described by Merchant and Packer (38) to determine the size, shape, and arrangement of bacteria. The organisms revealed Gram-negative, pink-colored with rod-shaped appearance, and arranged in single or in pairs were suspected as E. coli.
A single well-isolated colony was picked from MCA and streaked on Eosin Methylene Blue Agar (EMB) (HiMedia, Mumbai, India) and incubated at 37 • C for 24 h. The characteristic green metallic sheen growth of colonies is a presumptive identification for E. coli. Colony morphology and colors on MCA and EMB agar plates together with the Gram-stain procedure were used as an initial identification of E. coli colonies (39). Such colonies were taken from EMB into nutrient broth and agar for further biochemical examination. Standard biochemical tests (catalase test, indole, methyl red, Voges-Proskauer test, nitrate reduction, citrate utilization, and urease production) were used as confirmation of identification (40-46). The Triple Sugar Iron test was performed according to Vanderzant and Splittstresser (47). Carbohydrate fermentation tests of the isolates were performed according to the method of Simmons (43). Colonies that were confirmed as E. coli were further subcultured onto MacConkey Agar with Sorbitol to differentiate E. coli O157: H7 from other strains, especially lactose fermenters E. coli. A bacterial strain that was used as a quality control organism in this study was a standard strain of E. coli ATCC 25922.
PCR amplification of the pal, eaeA stx , and sxt virulence genes of E. coli Escherichia coli genomic DNA extraction and purification were performed as per the protocol given by PureLink R Genomic DNA Purification Kit, USA, for Gram-negative organisms, and the total genomic DNA was checked by running on 1.0% agarose gel.
The DNA of all presumptive isolates was subjected to multiplex PCR for analyzing the presence of the pal (48), eaeA (49), stx1, and stx2 (50) genes and further modified by Fitzmaurice (51). The amplification was performed according to the protocol reported by Parekh and Subhash (52)

Antibiotic susceptibility testing
The isolates of E. coli were screened for in vitro antimicrobial susceptibility using the agar disk diffusion method described by Bauer et al. (55). For this, the following seven different antibiotic disks (Oxoid Ltd., Basingstoke, Hampshire, England) with their concentrations given in parentheses were used in the antibiograms: amoxicillin (AMX) (2 µg), doxycycline (DOX) (30 µg), erythromycin (ERY) (15 µg), gentamicin (GEN) (10 µg), penicillin (PEN) (10 µg), trimethoprimsulfamethoxazole (SXT) (20 µg), and tetracycline (TET) (30 µg). The selection of these antibiotics was based on the availability and frequent use of these antimicrobials in the study area both in veterinary and human medicine. Four to five well-isolated colonies from nutrient agar plates were transferred into tubes containing 5 ml of a normal saline solution until it achieved the 0.5 McFarland turbidity standards, and then a sterile cotton swab was dipped into the adjusted suspension and then spread evenly over the entire surface of the plate of Mueller-Hinton agar (Oxoid Ltd., Basingstoke, Hampshire, England) to obtain uniform inoculums. The plates were then allowed to dry for 3-5 min. Antibiotic-impregnated disks were then applied to the surface of the inoculated plates with sterile forceps. After 18-24 h of incubation at 37 • C, the plates were examined, and the clear zone (inhibition zones

Data management and analysis
All collected data were entered into Microsoft Excel Sheet and analyzed through the SPSS version 16.
Accordingly, descriptive statistics such as percentages and frequency distribution were used to determine the occurrence of the food items, and the chi-square (χ 2 ) test was applied to assess the association.

Results
The occurrence of E. coli and E. coli O :H Out of the total 284 samples collected from the different sample sources, 70 (24.6%) were found to be positive for E. coli. The sample-based positivity of meat, raw milk, and yogurt samples were 41 (45.1%), 18 (12.4%), and 11 (22.9%), respectively. There was a significant difference (χ ² = 32.1; p = 0.000) among the different sample types in the occurrence of E. coli (Table 1). Of the total isolated E. coli isolates, 14.29% were found to be Shiga toxin-producing E. coli O157:H7.
Occurrence rates of 56.3%, 40%, and 50% were recorded in meat samples collected from butchery, abattoir, and restaurant, respectively. The positivity of E. coli for raw milk samples collected from farms, whole-sellers, and cafeterias were 10, 17.6, and 17.9%, respectively, whereas yogurt samples collected from farms and cafeterias were found to be 23. 1

PCR amplification results of the virulence genes of E. coli
The PCR result indicated that all (100%) of the tested isolates of E. coli carried the pal gene and 41.67% eaeA gene (EHEC). Among the eaeA gene-carried E. coli isolates, 40% carried the stx1 gene and 60% of them carried the stx2 gene.

Antimicrobial susceptibility profile of E. coli isolates
The E. coli strains isolated from meat, raw milk, and yogurt showed the highest level of susceptibility to gentamycin (91.7%). However, the highest level of resistance was observed against amoxicillin (95.8%) ( Table 3). MDR was detected in 75% of the isolates.

Discussion
This study revealed an overall positivity rate of 24.6% for E. coli. This finding was in line with the reports of Bedasa et al.  (69%). The variation might be due to differences in sample types of food of bovine origin, sources of the samples, methodological approaches, diagnostic techniques, calculation/interpretation, geographical locations, hygienic conditions, handling and transportation of samples, and contamination rates from utensils and personnel. Generally, in the present findings, the positivity of E. coli was higher in the samples collected from butchery, restaurants, cafeterias, .
/fvets. . and whole sellers other than the other sources. This might be due to food establishments in the study area being found to have poor sanitation and were not maintained well, specifically, poor repair condition of kitchens, improperly managed toilet facilities, inappropriate solid waste receptacles, lack of standard dishwashing compartments, and lack of handwashing lavatories in a large number of establishments (8).
The antimicrobial susceptibility test results of E. coli strains isolated from meat, raw milk, and yogurt showed the highest level of susceptibility for gentamycin (

Conclusion
This study revealed a relatively high occurrence of E. coli as well as Shiga toxin-producing E. coli O157:H7 in food of bovine origin in Mekelle City and isolates developed a high level of MDR (a bacterium that is resistant to more than one antibiotic) to the antimicrobials tested. Hence, foods of bovine origin can serve as a potential vehicle for transmitting E. coli, in particular, Shiga toxin-producing E. coli O157:H7, and its presence indicates a serious public health hazard and gives a warning signal for the possible occurrence of a foodborne outbreak in humans through the consumption of raw or undercooked food items. As a result, both its presence and development of MDR should receive particular attention and is an alert for the concerned bodies. Therefore, a coordinated effort is needed to reduce or eliminate the risk posed by these pathogens at various stages in the food chains and on the appropriate use of antimicrobials both in veterinary and human treatment regimes. Moreover, awareness creation should be made on foodborne diseases caused by Shiga toxin-producing E. coli O157:H7 with due consideration to the safe handling and consumption of food of animal origin and selection and safe use of antimicrobials. In conclusion, a nationwide phenotypic and molecular characterization, in-depth typing, and drug-resistant gene identification of E. coli O157:H7 should be undertaken.

Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Ethics statement
The animal study was reviewed and approved by Institutional Ethical and Environmental Considerations Review Committee, College of Veterinary Sciences, Mekelle University. Written informed consent was obtained from the owners for the participation of their animals in this study.

Funding
This study was funded by the College of Veterinary Sciences, Mekelle University.

Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated . /fvets. .
organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.