ZPH Manuscript Proof
Isolation of multidrug-resistant Escherichia coli O157 from
goat feces and carcasses in the Somali region of Ethiopia
Zoonoses and Public Health
r
Fo
Journal:
Manuscript ID:
Manuscript Type:
Date Submitted by the Author:
n/a
Dulo, Fitsum; Addis Ababa University,
Feleke, Aklilu; Addis Ababa University,
Szonyi, Barbara; International Livestock Research Institute,
Fries, Reinhard; Free University of Berlin,
Baumann, Maximilian; Free University of Berlin,
Grace, D; ILRI, MGL
vi
Subject Area:
Original Article
Re
Complete List of Authors:
Draft
E.coli spp, Foodborne disease, Public health, Zoonoses, Livestock
ew
ly
On
Zoonoses and Public Health
Page 1 of 20
1
Isolation of multidrug-resistant Escherichia coli O157 from goat feces and carcasses
2
in the Somali region of Ethiopia
3
4
5
Fitsum Dulo1, Aklilu Feleke1, Barbara Szonyi2, Reinhard Fries3, Maximilian Baumann3,
6
and Delia Grace4
7
8
r
Fo
9
1
Addis Ababa University, Faculty of Veterinary Medicine, Debre Zeit, Ethiopia
10
2
International Livestock Research Institute, Box 5689, Addis Ababa, Ethiopia
11
3
Free University of Berlin, Faculty of Veterinary Medicine, Brümmer Str. 10, D-14195
12
Berlin
13
4
vi
Re
International Livestock Research Institute, Box 30709, Nairobi, Kenya
ew
14
15
16
Corresponding author:
17
Barbara Szonyi
18
International Livestock Research Institute, Box 5689, Addis Ababa, Ethiopia
19
Email: b.szonyi@cgiar.org
20
Tel: +251 11 617 2686
21
Fax: +251 11 617 2001
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
22
23
24
25
1
Zoonoses and Public Health
ZPH Manuscript Proof
26
27
SUMMARY
28
Toxigenic E. coli is an important cause of gastro-enteritis in developing countries. In
29
Ethiopia, gastro-enteritis due to food-borne disease is a leading cause of death. Yet, there is
30
no surveillance for E. coli O157 and little is known about the carriage of this pathogen in
31
Ethiopia’s livestock. Also, antimicrobial resistance patterns are not well known in this part
32
of the world. A cross-sectional study was conducted in 2014 to determine the prevalence
33
and antibiotic resistance of E. coli O157 in goat meat, feces, and environmental samples
34
collected at a large abattoir in the Somali region of Ethiopia. A total of 235 samples,
35
including 93 goat carcass swabs, 93 cecal contents, 14 water, 20 hand, and 15 knife swabs
36
were collected. The samples were enriched in modified tryptone broth containing
37
novobiocin, and plated onto sorbitol MacConkey agar. Isolates were confirmed using
38
indole test and latex agglutination. Antimicrobial susceptibility testing was conducted
39
using the disk diffusion method.
40
Overall, six (2.5%) samples were contaminated with E. coli O157 of which two (2.1%)
41
were isolated from cecal contents, three (3.2%) from carcass swabs, and one (7.1%) from
42
water. All isolates were resistant to at least two of the 18 antibiotics tested. Abattoir
43
facilities and slaughter technique were conducive to carcass contamination. This study
44
highlights how poor hygiene and slaughter practice can result in contaminated meat, which
45
is especially risky in Ethiopia because of the common practice of eating raw meat. We
46
detected multidrug-resistance to drugs not used in goats, indicating that livestock are not
47
the originators of antimicrobial resistance in this part of the world. The isolation of
48
multidrug-resistant E. coli O157 from goats from a remote pastoralist system is a wake-up
49
call for global action on regulating and monitoring antimicrobial use in both human and
50
animal populations.
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 2 of 20
2
Zoonoses and Public Health
Page 3 of 20
51
52
Keywords: Antimicrobial resistance, E. coli O157, Goat, Ethiopia; Pastoralist
53
54
55
IMPACTS
•
from a remote pastoralist system where veterinary inputs and drug use are limited.
56
57
The study describes the isolation of multidrug-resistant E. coli O157 from goats
•
We detected multidrug-resistance to drugs not used in goats, indicating that
58
livestock are not the originators but rather the victims of antimicrobial resistance in
59
this part of the world.
60
•
r
Fo
The study highlights how poor hygiene and slaughter technique can result in
Re
61
contaminated meat, which is especially risky in Ethiopia because of the common
62
practice of eating raw meat.
vi
63
ew
64
65
66
INTRODUCTION
67
Escherichia coli (E. coli) are a group of bacteria that form a normal part of the intestinal
68
micro-flora of healthy animals and humans. However, certain serotypes, including E. coli
69
O157, have acquired virulence factors that allow them to produce toxins known as
70
verotoxins or Shiga-like toxins. Verotoxigenic E. coli (VTEC) cause serious disease in
71
humans worldwide, including diarrhea, hemorrhagic colitis, hemolytic-uremic syndrome,
72
and sometimes death (Farrokh et al., 2012). Consumption of contaminated raw and
73
undercooked meat is the most common means of transmission to humans (Chapman et al.,
74
2001).
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
3
Zoonoses and Public Health
ZPH Manuscript Proof
75
Cattle are the major reservoir of E. coli O157, and, to some extent, sheep and goats (La
76
Ragione et al., 2009). The pathogen is carried in the intestinal tract and excreted in the
77
feces. During slaughter, the pathogen may be present on the skin or in the feces of the
78
animal, and may get transferred to the carcass during evisceration or skin removal.
79
Therefore, poor slaughter technique, particularly poor hygienic practices during slaughter
80
greatly increase the risk of meat contamination with E. coli O157. The risk of such
81
contamination is also related to the E. coli O157 carriage status of the pre-slaughter
82
population (McEvoy et al., 2003). Therefore, both the carriage status of the pre-slaughter
83
animal population, and slaughter hygiene, are essential in determining and mitigating the
84
risk of exposure of meat consumers to E. coli O157.
85
Gastro-enteritis due to food-borne disease is one of the most common illnesses in Ethiopia,
86
and it is a leading cause of death among people of all ages in the country (Institute for
87
Health Metrics and Evaluations, 2013). The lack of surveillance of food-borne pathogens,
88
poor hygienic conditions and sub-standard slaughter practices in the abattoirs, and the
89
widespread cultural practice of raw meat consumption, are all major factors contributing to
90
the high risk of exposure of Ethiopians to food-borne pathogens such as VTEC. In spite of
91
the high risk of exposure to VTEC in Ethiopia, there is no surveillance for this pathogen
92
and very little is known about the carriage rate of E. coli O157 in different livestock
93
populations.
94
Antimicrobial resistance in E. coli O157 is an increasing global public health concern. The
95
widespread administration of antimicrobials to animals and humans promotes the selection
96
of antimicrobial resistant strains, which complicates the treatment of bacterial infections.
97
Livestock in particular, are often incriminated as originators of antimicrobial resistance in
98
industrialized nations, where antibiotics are commonly used to improve productivity (Van
99
Boeckel et al., 2015). Human infections with multidrug-resistant bacteria are especially
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 4 of 20
4
Zoonoses and Public Health
Page 5 of 20
100
problematic in developing countries, where treatment options are often limited. In
101
Ethiopia, both veterinary and medical drugs are commonly misused and antimicrobial
102
resistance in E. coli O157 in raw meat, including goat meat, has already been reported
103
(Hiko et al., 2008). More information is needed on resistance patterns particularly in
104
developing countries, because resistant infections are an increasing problem worldwide
105
(Van Boeckel et al., 2015).
106
Goat meat is traditionally consumed in the low land areas of Ethiopia. However, in recent
107
years, there has been an increasing demand for goat meat for domestic consumption in
108
large urban areas throughout the country and also for export. The arid Somali zone,
109
covering 250,000 square kilometers of Ethiopia, is home to 4.4 million people, 38% of
110
whom are pastoralists. According to official statistics there are 3.1 million goats in this
111
area. Much of the zone is remote and isolated but it is a major supplier of goat meat to both
112
domestic markets and for export to Djibouti, Somalia and the Middle East (Hassen et al.,
113
2013). However, animal diseases are rampant, and little is known about the health status
114
of the goats and their carriage status of important food-borne pathogens including E. coli
115
O157. Furthermore, there is no information on the antimicrobial resistance patterns of
116
strains circulating in livestock in this remote area. Therefore, this study was carried out to
117
1) assess the pre-slaughter (i.e. carriage status) of E. coli O157 in goats originating in the
118
Somali region of Ethiopia; 2) to assess the hygienic practices and the level of carcass
119
contamination with E coli O157 during the slaughter process of goats and 3) to determine
120
the antimicrobial susceptibility pattern of the isolates.
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
121
122
MATERIALS AND METHODS
123
Study area
5
Zoonoses and Public Health
ZPH Manuscript Proof
124
The target population for this study was goats raised by pastoralists in the Somali zone of
125
Ethiopia. The goats kept by pastoralists are indigenous breeds (i.e. long-eared Somali
126
goats). The study was conducted in a major public abattoir in the Somali region of
127
Ethiopia. Sampling was carried out in two 3-day long visits over a period of 2 months.
128
Study design
129
A cross-sectional, abattoir-based study was conducted from January to April 2014 to
130
determine the prevalence and antibiotic resistance patterns of E. coli O157 in goat meat,
131
feces, and environmental samples. Sample size was determined using the formula by
132
Thrusfield (Thrusfield, 2005), based on expected prevalence of E. coli O157 in goat meat
133
and feces, which was estimated at 5% following Mersha et al. (Mersha et al., 2009). The
134
confidence level was 95% and the precision was 5%. Thus, the required sample size was
135
73; however 93 samples were taken deliberately in order to maximize the precision of the
136
study.
137
Sample collection
138
Carcass sampling
139
For each animal, four different sites of the carcass (thorax, brisket, flank and crutch) were
140
swabbed following the guidelines of the International Organization for Standardization
141
(International Organization for Standardization 2003). Each sampling area covered 100
142
cm2 by placing a sterile template (10 cm x 10 cm) on the carcass. For each sampling area, a
143
sterile cotton-tipped swab (2 X 3 cm), fitted with shaft, was first moistened in 10 ml of
144
buffered peptone water (Oxoid Ltd., Hampshire, UK) and then rubbed first horizontally
145
and then vertically several times across the carcass surface. On completion of the rubbing
146
process, the shaft was broken by pressing it against the inner wall of the test tube and
147
disposed leaving the cotton swab in the test tube. The four swabs were put into a screw-
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 6 of 20
6
Zoonoses and Public Health
Page 7 of 20
148
capped test tube containing 10 ml of sterile buffer peptone water, and were transported to
149
the laboratory in a cool box within twenty four hours of sampling.
150
Fecal sampling
151
The fecal samples were collected immediately after evisceration from cecal contents of
152
slaughtered goats; an aseptic incision was made with surgical blade in the cecum to obtain
153
a representative sample of 10 g of the cecal content. The fecal material was aseptically
154
compressed and the resultant liquid was decanted in sterile universal bottle. The samples
155
were labeled and transported on ice to the laboratory, where they were held in a cold
156
storage overnight and processed the following day.
157
Environmental sampling and assessment of slaughter hygiene
158
Environmental samples were collected by swabbing the hands of abattoir workers (both
159
hands, both sides) and swabbing the slaughter knives (blade and handle, both sides). In
160
addition, water samples (10 ml) were collected before and during operation from the
161
buckets. Slaughtering hygiene and the sanitary status of the abattoir were determined by
162
the use of a structured checklist and through direct observations of the premises and the
163
practices of abattoir workers. This study was approved by the Academic Commission of
164
the College of Veterinary Medicine and Agriculture, Addis Ababa University, and the
165
Institutional Research Committee of the International Livestock Research Institute with
166
reference number IREC-2013-03. Moreover, ethical clearance to use human subjects for
167
this study was obtained from the Ministry of Science and Technology. Human subjects
168
enrolled in this study were those who gave their consent after the purpose of the study had
169
been explained to them.
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
7
Zoonoses and Public Health
ZPH Manuscript Proof
170
Culture and isolation of E. coli O157
171
Fecal samples
172
Approximately 1ml/1g of fecal pellet (homogenized when possible) was suspended into 9
173
ml of modified tryptone soya broth supplemented with novobiocin (10 mg/l) (Oxoid Ltd,
174
Hampshire, UK). Samples were vortexed and incubated overnight at 37°C. After selective
175
enrichment, 50µl of product was streaked onto sorbitol MacConkey agar (Oxoid Ltd.,
176
Hampshire, UK) and the plates were incubated at 37°C for 24 hours. Up to six colorless
177
colonies (non- sorbitol fermenters) were picked and separately sub-cultured on
178
MacConkey agar for 24 hours at 37°C for purification. After overnight incubation, the
179
purified and intensely red colonies with a pale periphery were tested for indole production
180
and indole forming isolates were seeded onto nutrient agar for serological confirmation.
181
The indole test was carried out as follows. One colony was inoculated into 4 ml of tryptone
182
soya broth, using a straight inoculation wire. Incubation was done for overnight at 37°C.
183
After this, one drop of indole reagent was added to the tryptone soya broth culture to test
184
for indole production (red ring-positive). The microbiological analysis was carried out at
185
the Ethiopian Nutrition and Health Research Institute (EHNRI).
186
Carcass bacterial swabs and environmental samples
187
The carcass bacterial swabs and the environmental samples were incubated overnight at 37
188
0
189
(1:9), and subjected to similar tests for bacteriological analysis as fecal samples.
190
Confirmatory test by latex agglutination for E. coli O157 serogroup
191
Non-sorbitol fermenting (NSF) isolates were inoculated onto nutrient agar for testing.
192
Then, NSF and indole positive colonies were serotyped using the DrySpot E. coli O157
193
latex agglutination test (Oxoid Ltd., Hampshire, UK). One drop of saline was dispensed to
194
the small ring (at the bottom of each oval) in both the test and control reaction areas
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 8 of 20
C after being suspended into modified tryptone soya broth supplemented with novobiocin
8
Zoonoses and Public Health
Page 9 of 20
195
ensuring that the liquid did not mix with the dried latex reagents. Using a sterile single use
196
plastic loop, a portion of the colony to be tested was picked and carefully emulsified in the
197
saline drop until the suspension was smooth. Then, using paddle the suspension was mixed
198
into the dry latex spots until completely suspended and spread to cover the reaction area.
199
The test card was picked up and rocked for up to 60 seconds, and agglutination was
200
detected under normal lighting conditions. A result was positive if agglutination of the
201
latex particles occurred within 1 minute. This indicated the presence of E. coli serogroup
202
O157. A negative result was obtained if no agglutination occurred and a smooth blue
203
suspension remained after 60 seconds in the test area.
204
Antimicrobial susceptibility testing
205
Antimicrobial resistance tests were performed by standard disc diffusion technique
206
(Clinical and Laboratory Standards Institute, 2012). Resistance testing discs included 18
207
antimicrobial agents and contained ampicillin (10µg), amoxycillin-clavulanic acid
208
(20/10µg), cefotaxime (30µg), ceftriaxone (30µg), cefoxitin (30µg), cefuroxime sodium
209
(30µg), chloramphenicol (30µg), ciprofloxacin (5µg), erythromycin (15µg), gentamicin
210
(10µg), kanamycin (30µg), nalidixic acid (30µg), nitrofurantoin (50µg), norfloxacin
211
(10µg), streptomycin (10µg) , sulfamethoxazole-trimethoprim (25µg), sulfonamides
212
(300µg), tetracycline (10µg) (Oxoid Ltd., Hampshire, UK). The isolates were considered
213
resistant if the diameter of the inhibition zone was less than or equal to the resistance
214
breakpoint provided by guidelines (Clinical and Laboratory Standards Institute, 2012).
215
Each isolated bacterial colony from pure fresh culture was transferred into a test tube of 5
216
ml tryptone soya broth and incubated at 37oC for 6 hours. The turbidity of the culture broth
217
was adjusted using sterile saline solution, or more isolated colonies were added to obtain
218
turbidity that is usually comparable with that of 0.5 McFarland standards (approximately
219
3x108 CFU per ml). Mueller-Hinton agar (Becton, Dickinson and Company, New Jersey,
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
9
Zoonoses and Public Health
ZPH Manuscript Proof
220
USA) plates were prepared according to the manufacturer. A sterile cotton swab was
221
immersed into the suspension and rotated against the side of the tube to remove the excess
222
fluid and then swabbed in three directions uniformly on the surface of Mueller-Hinton agar
223
plates. After the plates dried, antibiotic disks were placed on the inoculated plates using
224
sterile forceps. The antibiotic disks were gently pressed onto the agar to ensure firm
225
contact with the agar surface, and incubated at 37oC for 24 hours. Following this, the
226
diameter of inhibition zone formed around each disk was measured using a black surface,
227
reflected light and transparent ruler by laying it over the plates. The results were classified
228
as sensitive, intermediate, and resistant according to the standardized table supplied by the
229
manufacturer (Clinical and Laboratory Standards Institute, 2012).
230
Data analysis
231
Data were transferred to a Microsoft Excel spread sheet (Microsoft Corp., Redmond, WA,
232
USA). The prevalence of E. coli O157 in cecal contents, carcass swab and environmental
233
samples was determined by dividing the number of positive samples by the total number of
234
samples examined.
r
Fo
ew
vi
Re
235
On
236
RESULTS
237
Isolation and prevalence of E. coli O157
238
The study was conducted on a total of 235 samples, comprising 93 goat carcass swabs, 93
239
cecal contents and 49 environmental samples consisting of 14 water samples, 20 hand
240
swabs and 15 butcher knife swabs. Out of the total of 235 different samples examined, six
241
(2.6%) were found to be contaminated with E. coli O157. Escherichia coli O157 was
242
isolated from two (2.2%) cecal contents, three (3.2%) carcass swabs, and one (7.1%) water
243
sample.
244
Susceptibility of isolates to antimicrobial agents
ly
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 10 of 20
10
Zoonoses and Public Health
Page 11 of 20
245
Antimicrobial susceptibility testing results showed that of the 6 isolates, all were resistant
246
to erythromycin, 5(83.3%) were resistant to ampicillin, 3(50%) were resistant to
247
nitrofurantoin, 2(33.3%) were resistant to cefoxitin, streptomycin, sulfamethoxazole-
248
trimethoprim, sulfonamides and tetracycline, and 1(16.7%) showed resistance to
249
amoxycillin-clavulanic acid. None of the isolates were resistant to cefotaxime, ceftriaxone,
250
cefuroxime sodium, chloramphenicol, ciprofloxacin, gentamicin, nalidixic acid and
251
norfloxacin (Figure 1).
252
All isolates were resistant to at least two antibiotics. Multidrug-resistance to more than two
253
antimicrobial agents was detected in 3(50%) of the isolates. Two isolates (33.3%) were
254
resistant to eight antimicrobials tested (Table 1).
255
Table 1: Antimicrobial resistance patterns of E. coli O157 isolates
r
Fo
Number of
a
Re
Drug patterns
Number
vi
antimicrobials
Proportion
of isolates of isolates (%)
2
AMC,E
1
16.7
2
AMP,E
2
33.3
3
AMP,E,F
1
16.7
>4
AMP,E,F,FOX,S,SXT,S3,TE 2
ew
On
256
33.3
ly
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
257
a
258
nitrofurantoin, S: streptomycin, SXT: sulfamethoxazole-trimethoprim, S3: sulfonamides, TE: tetracycline
259
Facilities and practices at the abattoir
260
The observational survey revealed that deficiencies in the abattoir’s facilities did not allow
261
for maintenance of hygiene. For example, tap water, hot water, soap, retention room
262
(cooling facilities), pest control, changing rooms, latrine and bathroom facilities were not
263
present in the abattoir. The entire slaughter process was done in the same area without
264
separation of dirty and clean zones. Water was only available in buckets, and the same
Key for Table 1: AMP: ampicillin, AMC: amoxycillin-clavulanic acid, FOX: cefoxitin, E: erythromycin, F:
11
Zoonoses and Public Health
ZPH Manuscript Proof
265
buckets of water were used for cleaning knives, washing hands, washing carcasses, and
266
washing the floor. All workers were wearing street clothes. The workers regularly placed
267
their equipment on dirty surfaces during their work. Other risky practices, such as “fisting”
268
were also observed. During “fisting”, the butcher punches his fist forcefully between the
269
skin and the carcass surface to detach the skin. Fisting should be done with utmost care to
270
avoid carrying dirt, debris and pathogens from the skin onto the sterile carcass surface, but
271
butchers were observed touching both the inside of the carcass and the soiled outside of the
272
animal’s skin while removing the skin.
273
r
Fo
274
DISCUSSION
275
To our knowledge, this was the first study to investigate the presence, prevalence, and
276
antimicrobial resistance patterns of E. coli O157 in goats slaughtered in a major abattoir in
277
the remote Somali region of Ethiopia. We isolated E. coli O157 from 3.2% of carcass swab
278
samples, 2.2% of cecal contents and 7.1% of water samples. The hygienic conditions and
279
slaughter practices in the abattoir were found to be conducive to cross-contamination of
280
goat carcasses with pathogens including E. coli O157, posing a risk to meat consumers.
281
The prevalence of E. coli O157 in carcass swabs in the current study is consistent with
282
findings of previous studies conducted in other regions of Ethiopia. Mersha et al. (Mersha
283
et al., 2009) in Modjo, and Hiko et al. (Hiko et al., 2008) in Debre Zeit, reported a
284
prevalence of E. coli O157 in 3% and 5% of goat meat samples, respectively. Current
285
international guidelines are that E. coli O157 should be absent from 10 g of meat.
286
Therefore any detection of E. coli O157 in meat is considered unacceptable in several
287
countries. However, there are no regulations in Ethiopia to protect meat consumers from
288
food-borne pathogens such as E. coli O157. This is especially problematic because of the
289
widespread practice of raw meat consumption throughout the country. Therefore, avoiding
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 12 of 20
12
Zoonoses and Public Health
Page 13 of 20
290
carcass contamination during the slaughter process should be of utmost importance in
291
order to protect the public from food-borne illness due to consumption of contaminated
292
meat.
293
In the present study, the Somali region of Ethiopia was chosen because of its large goat
294
population. Furthermore, the consumption of goat meat is much more frequent in the
295
Somali region compared to other areas of Ethiopia, with fresh goat meat reportedly
296
purchased by the predominantly Muslim population from retail shops 1-5 times a week.
297
Human E. coli O157 infections mostly originate from animal-source foods (Jo et al., 2004).
298
Domestic ruminants, including goats, have been established as natural reservoirs for E. coli
299
O157, and therefore they play a significant role in the epidemiology of human infections
300
(Griffin and Tauxe, 1991). Several recent reports have clearly identified goats as sources of
301
E. coli O157 infection (Chapman et al., 2000; La Ragione et al., 2009). Furthermore, due to
302
their inquisitive nature, goats are likely to be in regular contact with humans, increasing the
303
risk of direct fecal-oral transmission of zoonotic pathogens (La Ragione et al., 2009). In the
304
Somali region, goats are raised by pastoralists who have frequent, close contact with their
305
animals. Thus, the risk of transmission of zoonotic diseases in this system is high.
306
Furthermore, the deeply-rooted cultural practice of raw meat consumption throughout
307
Ethiopia increases the risk of food-borne illness. This, combined with our results suggests
308
that food-borne illness due to E. coli O157 could be a significant public health concern in
309
the Somali region of Ethiopia and possibly in other parts of the country as well. Clinical
310
data are needed to estimate the real impact of E. coli O157 on human health in Ethiopia.
311
In this study, E. coli O157 was isolated from both cecal contents and carcass swabs.
312
Contamination of meat with E. coli O157 can occur during the processing of carcasses,
313
when gut contents and fecal matter or contaminated hides come in contact with the sterile
314
meat surfaces (Elder et al., 2000). In the present study, presence of E. coli O157 on goat
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
13
Zoonoses and Public Health
ZPH Manuscript Proof
315
carcasses indicates transfer of faecal material onto the sterile carcass during the slaughter
316
process. Indeed, risky practices during the slaughtering process were observed and
317
documented in the present study. For instance, the practice of fisting, whereby the workers
318
touch the soiled outside of the animal’s skin while removing the skin, can facilitate transfer
319
of pathogens onto the sterile carcass surface. We also isolated E. coli O157 from abattoir
320
water. In the abattoir, water was stored in plastic buckets and the same bucket was used to
321
wash the floor, carcasses, hands and equipment. Therefore, wash water in this abattoir
322
could be a major source of contamination, and could contribute to carcass-to-carcass
323
spread of pathogenic bacteria across the slaughter line. The survey also revealed that
324
workers at the abattoir didn’t have adequate training in safe meat handling and proper
325
slaughtering hygiene. Also, they were not equipped and supplied with the necessary
326
materials that would enable them to maintain general hygiene. For instance, some of the
327
slaughter staff indicated that inadequate supply of clean water posed a challenge towards
328
maintaining hygiene. These are common problems in public abattoirs throughout Ethiopia.
329
Often, the resources available to public abattoirs are not sufficient to properly train staff
330
and to maintain equipment and facilities in hygienic conditions. To address some of these
331
issues, our team has been collaborating with the Ministry of Agriculture on providing
332
training to abattoir workers on proper practices and hygiene during the slaughter process.
333
All E. coli O157 isolates in the present study exhibited resistance to at least two or more of
334
the 18 antimicrobial agents tested. Resistance to erythromycin and ampicillin were the
335
most common resistance profiles identified among our study isolates. High resistance of E.
336
coli O157 to erythromycin has also been reported from the Middle East (Harakeh et al.,
337
2005; Osaili et al., 2013). Furthermore, half of the E. coli O157 isolates were resistant to
338
nitrofurantoin. Interestingly, there was only moderate resistance to tetracycline in this
339
study, although it is one of the most commonly available drugs for use among livestock in
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 14 of 20
14
Zoonoses and Public Health
Page 15 of 20
340
Ethiopia. It is readily available in different dosage forms and in combination with other
341
antibiotics and vitamins. To our knowledge, this is the only antimicrobial drug which is
342
most likely to be readily available in the study area for veterinary use. No resistance was
343
observed to the newer generation of antimicrobials such as ciprofloxacin and norfloxacin,
344
which are important in the treatment of human cases of gastroenteritis.
345
In Ethiopia, a previous study on the antimicrobial susceptibility of E. coli O157 was
346
conducted in 2006 in retail raw meat in the vicinity of Addis Ababa (Debre Zeit and Mojo
347
towns) (Hiko et al., 2008). The 2006 study analyzed E. coli O157 isolates from sheep,
348
cattle and goat meat, and reported that all isolates were susceptible to most of the 13
349
antibiotics tested. In contrast, the results of the present study indicated high levels of
350
antimicrobial resistance, including multi-drug resistance in E. coli O157 in goat meat
351
isolates. This discrepancy may indicate that antimicrobial resistance of E. coli O157 in the
352
Somali region of Ethiopia is more prevalent and severe than elsewhere, or that
353
antimicrobial resistance of E. coli O157 has been rapidly increasing in Ethiopia.
354
Indeed, emergence and dissemination of antimicrobial resistance is an increasing problem
355
in enteric bacteria worldwide. Antimicrobial resistance may arise due to misuse by
356
humans, or misuse in feeding or treatment of animals (Schroeder et al., 2002). Especially
357
sub-therapeutic and prophylactic administration of antibiotics in livestock, which is a
358
common practice in high income countries, creates ideal conditions for the development of
359
antimicrobial resistant strains (Van Boeckel et al., 2015). Thus, antibiotic use in animals is
360
of great concern because of its role in generating resistance in human pathogens. In the
361
Somali zone of Ethiopia, pastoralists rear goats in an extensive system without the use of
362
specialized inputs. Here, goats only feed on natural forages and shrubs, veterinary input is
363
minimal, and the availability of drugs is limited (Hassen et al., 2013). This study detected
364
multidrug-resistance to antibiotics not used in goats, suggesting that livestock are not the
r
Fo
ew
vi
Re
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
15
Zoonoses and Public Health
ZPH Manuscript Proof
365
originators but rather the victims of antimicrobial resistance in this remote area of the
366
world. Further investigations are needed to determine which factors are responsible for the
367
emergence of multidrug-resistant strains in this vast and extensive production system, but
368
it is plausible that the origin of resistance was drugs used to treat human infections.
369
The present study had some limitations in laboratory facility and therefore, other serotypes
370
and virulence factors for E .coli O157 were not investigated. Also, it has been shown that
371
the use of immune-magnetic separation (IMS) with enrichment in broth culture enhances
372
the isolation of E. coli O157 from samples with a low concentration of the bacteria
373
(Chapman et al., 1994). In this study, enrichment without IMS was employed for the
374
isolation of E. coli O157. Nonetheless, the present study revealed that multidrug-resistant
375
E. coli O157 were prevalent in the goat population of the remote Somali region of
376
Ethiopia, which emphasizes the need for more stringent monitoring of antibiotic use in
377
both human and animal populations.
r
Fo
378
ew
vi
Re
379
Conflict of Interest
380
The authors declare that they have no conflict of interest.
On
381
382
Acknowledgements
383
This study was conducted under the Safe Food Fair Food project of the International
384
Livestock Research Institute (ILRI), funded by International Agricultural Research, GTZ,
385
Germany (Project no: 11.7860.7-001.00). We thank the Ethiopian Nutrition and Health
386
Research Institute (EHNRI) and the study participants for their cooperation.
ly
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 16 of 20
387
388
389
16
Zoonoses and Public Health
Page 17 of 20
390
391
REFERENCES
392
Chapman, P. A., A. T. Cerdan Malo, M. Ellin, R. Ashton, and Harkin, 2001: Escherichia
393
coli O157 in cattle and sheep at slaughter, on beef and lamb carcasses and in raw
394
beef and lamb products in South Yorkshire, UK. Int. J. Food Microbiol. 64, 139-
395
150.
396
Chapman, P. A., J. Cornell, and C. Green, 2000: Infection with verocytotoxin-producing
397
Escherichia coli O157 during a visit to an inner city open farm. Epidemiol. Infect.
398
125: 531-536.
399
r
Fo
Chapman, P. A., D. J. Wright, and C. A. Siddons, 1994: A comparison of immunomagnetic
Re
400
separation and direct culture for the isolation of verocytotoxin-producing
401
Escherichia coli O157 from bovine faeces. J. Med. Microbiol. 40, 424-427.
402
vi
Clinical and Laboratory Standards Institute (CLSI), 2012: Performance standards for
ew
403
antimicrobial susceptibility testing; Twenty-second informational supplement.
404
CLSI, Wayne, PA.
405
Elder, R. O., J. E. Keen, G. R. Siragusa, G. A. Barkocy-Gallagher, M. Koohmaraie, and W.
406
W. Laegreid, 2000: Correlation of enterohemorrhagic Escherichia coli O157
407
prevalence in feces, hides, and carcasses of beef cattle during processing. Proc.
408
Natl. Acad. Sci. USA 97, 2999-3003.
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
409
Farrokh, C., K. Jordan, F. Auvray, K. Glass, H. Oppegaard, S. Raynaud, D. Thevenot, R.
410
Condron, K. De Reu, A. Govaris, K. Heggum, M. Heyndrickx, J. Hummerjohann,
411
D. Lindsay, S. Miszczycha, S. Moussiegt, K. Verstraete, and O. Cerf, 2012: Review
412
of Shiga-toxin-producing Escherichia coli (STEC) and their significance in dairy
413
production. Int. J. Food. Microbiol. 162, 190-212.
17
Zoonoses and Public Health
ZPH Manuscript Proof
414
Griffin, P. M., and R. V. Tauxe, 1991: The epidemiology of infections caused by
415
Escherichia coli O157:H7, other enterohemorrhagic E. coli, and the associated
416
hemolytic uremic syndrome. Epidemiol. Rev. 13, 60-98.
417
Harakeh, S., H. Yassine, M. Gharios, E. Barbour, S. Hajjar, M. El-Fadel, I. Toufeili, and R.
418
Tannous, 2005: Isolation, molecular characterization and antimicrobial resistance
419
patterns of Salmonella and Escherichia coli isolates from meat-based fast food in
420
Lebanon. Sci. Total. Environ. 341, 33-44.
421
422
423
425
raw meat products in Ethiopia. J. Infect. Dev. Ctries. 2, 389-93.
Institute for Health Metrics and Evaluations (IHME), 2013: Global burden of disease study
2010. University of Washington, Seattle, WA.
ew
427
Hiko, A., D. Asrat, and G. Zewde, 2008: Occurrence of Escherichia coli O157:H7 in retail
vi
426
chains in Shinille District, Somali Region, Ethiopia. ICARDA, Addis Ababa.
Re
424
Hassen, A., A. Ismail, A. Haile, and G. Legese, 2013. Analysis of sheep and goat value
r
Fo
International Organization for Standardization (ISO), 2003: Microbiology of food and
428
animal feeding stuffs -carcass sampling for microbiological analysis. ISO, Geneva,
429
Switzerland.
430
On
Jo, M. Y., J. H. Kim, J. H. Lim, M. Y. Kang, H. B. Koh, Y. H. Park, D. Y. Yoon, J. S.
431
Chae, S. K. Eo, and J. H. Lee, 2004: Prevalence and characteristics of Escherichia
432
coli O157 from major food animals in Korea. Int. J. Food. Microbiol. 95, 41-9.
433
La Ragione, R. M., A. Best, M. J. Woodward, and A. D. Wales, 2009: Escherichia coli
ly
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 18 of 20
434
O157:H7 colonization in small domestic ruminants. FEMS Microbiol. Rev. 33, 394-
435
410.
436
437
McEvoy, J. M., A. M. Doherty, J. J. Sheridan, F. M. Thomson-Carter, P. Garvey, L.
McGuire, I. S. Blair, and D. A. McDowell, 2003: The prevalence and spread of
18
Zoonoses and Public Health
Page 19 of 20
438
Escherichia coli O157:H7 at a commercial beef abattoir. J. Appl. Microbiol. 95,
439
256-66.
440
Mersha, G., D. Asrat, B. M. Zewde, and M. Kyule, 2009: Occurrence of Escherichia coli
441
O157:H7 in faeces, skin and carcasses from sheep and goats in Ethiopia. Lett. Appl.
442
Microbiol. 50, 71-6.
443
Osaili, T. M., A. R. Alaboudi, and M. Rahahlah, 2013: Prevalence and antimicrobial
444
susceptibility of Escherichia coli O157:H7 on beef cattle slaughtered in Amman
445
abattoir. Meat. Sci. 93, 463-8.
446
r
Fo
Schroeder, C. M., C. Zhao, C. DebRoy, J. Torcolini, S. Zhao, D. G. White, D. D. Wagner,
447
P. F. McDermott, R. D. Walker, and J. Meng, 2002: Antimicrobial resistance of
448
Escherichia coli O157 isolated from humans, cattle, swine, and food. Appl.
449
Environ. Microbiol. 68, 576-81.
452
Thrusfield, M. 2005. Veterinary Epidemiology. Blackwell Publishing Professional, Ames,
Iowa.
ew
451
vi
450
Re
Van Boeckel, T. P., C. Brower, M. Gilbert, B. T. Grenfell, S. A. Levin, T. P. Robinson, A.
453
Teillant, and R. Laxminarayan, 2015: Global trends in antimicrobial use in food
454
animals. Proc. Natl. Acad. Sci. USA.
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ZPH Manuscript Proof
19
Zoonoses and Public Health
ZPH Manuscript Proof
r
Fo
Figure 1: The proportion of E. coli O157 isolates resistant to 18 antimicrobial agents (n=6 isolates)
Key for Figure 1. AMP: ampicillin, AMC: amoxycillin-clavulanic acid, CTX: cefotaxime, CRO: ceftriaxone,
FOX: cefoxitin, CFX: cefuroxime sodium, C: chloramphenicol, CIP: ciprofloxacin, E: erythromycin CN:
gentamicin, K: kanamycin, NA: nalidixic acid, F: nitrofurantoin, NOR: norfloxacin, S: streptomycin, SXT:
sulfamethoxazole-trimethoprim, S3: sulfonamides, TE: tetracycline.
vi
Re
248x128mm (96 x 96 DPI)
ew
ly
On
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 20 of 20
Zoonoses and Public Health