VETERINARSKI ARHIV 69 (6), 335-347, 1999

ISSN 1331-8055 Published in Croatia




Prevalence of antimicrobial resistance and enteropathogenic serogroups in Escherichia coli isolates from wildlife in Trinidad and Tobago

Abiodun Adesiyun* and Michael Downes

School of Veterinary Medicine, Faculty of Medical Sciences, University of the West Indies, St. Augustine, Trinidad and Tobago




* Contact address:
Prof. A. A. Adesiyun,
School of Veterinary Medicine, Faculty of Medical Sciences, University of the West Indies, St. Augustine, Trinidad and Tobago.
Phone: 1 868 645-2640; Fax: 1 868 645-7428; E-mail: adesiyun@tstt.net.tt


ADESIYUN, A., M. DOWNES: Prevalence of antimicrobial resistance and enteropathogenic serogroups in Escherichia coli isolates from wildlife in Trinidad and Tobago. Vet. arhiv 69, 335-347, 1999.

ABSTRACT

Antibiograms of Escherichia coli isolates from free-ranging and captive wildlife (mammals, avian and reptiles) were determined using the disc diffusion method; the prevalence of enteropathogenic serogroups amongst isolates was also investigated. Out of a total of 472 isolates of E. coli tested using 8 antimicrobial agents, 451 (95.6%) exhibited resistance to one or more antimicrobial agents. Resistance was high to cephalothin (88.1%), moderate to ampicillin (34.3%), streptomycin (33.7%) and neomycin (22.1%) and relatively low to chloramphenicol (7.8%) and gentamycin (6.8%), regardless of source. Prevalence of resistance amongst E. coli isolates from captive mammals was statistically significantly (P<0.0; c2) higher than for isolates from free-ranging mammals to streptomycin, sulphamethoxazole/trimethoprim (SXT), and chloramphenicol. However, the prevalence of resistance amongst E. coli isolates from free-ranging mammals, was significantly (P<0.05; c2) higher than those from captive mammals for ampicillin and cephalothin. Escherichia coli isolates from racing pigeons were statistically significantly (P<0.01; c2) and more resistant than isolates from wild birds for ampicillin, streptomycin, nalidixic acid, sulphamethoxazole/trimethoprim and chloramphenicol. Prevalence of resistance amongst E. coli isolates from captive birds were, however, statistically significantly (P<0.05; c2) lower than those of isolates from wild birds for ampicillin, neomycin and gentamicin. Amongst 200 randomly selected E. coli isolates tested, 46 (23.0%) belong to enteropathogenic serogroups, with 22 (11.0%), 21 (10.5%) and 3 (1.5%) agglutinated by E. coli polyvalent antiserum serogroups A, B and C, respectively. The prevalence of enteropathogenic serogroups amongst E. coli isolates from free-ranging mammals (20.4%) was statistically significantly (P<0.05; c2) lower than those found in isolates from captive mammals (48.0%). Similarly, E. coli isolates from wild birds had a statistically significantly (P<0.05; c2) higher (50%) frequency of enteropathogenic serogroups than isolates from either racing pigeons (7.4%) or captive birds (12.5%). It was concluded that the high prevalence of resistance to antimicrobial agents amongst. E. coli isolates poses potential therapeutic problems to captive wildlife, while the occurrence of enteropathogenic serogroups amongst isolates could pose a health hazard to consumers of wildlife meat.

Key words: Escherichia coli, antibiograms, enteropathogenic serogroups, wildlife, Trinidad and Tobago



Introduction

Escherichia coli is a normal inhabitant of the gastrointestinal tracts of animals and human beings (DRASAR and HILL, 1974; HOWE and LINTON, 1976; GYLES, 1993). The microorganism has been responsible for both intestinal and extra-intestinal infections, thus making chemotherapy important for control (SUSSMAN, 1985; ROBINS-BROWNE, 1987; KARMALI, 1989; GYLES, 1993). Indiscriminate use of antimicrobial agents in humans and domestic animals in chemoprophylaxis, chemotherapy or as growth promoters has resulted in the development of resistance amongst bacteria, including E. coli (WALTON; 1971; SMITH, 1975; WASHINGTON, 1979). It is also well known that the transfer of resistance factors does occur amongst bacteria, particularly members of the family Enterobacteriaceae (ANDERSON, 1968; WIERUP, 1975; TOWNER, 1982). Exposure of animals to microorganisms from various sources, especially in foods and the environment, may then facilitate the transmission of resistant bacteria (ROLLAND et al., 1985). To date, however, there is very limited information available in the literature on the antibiograms of E. coli strains from wildlife, either free-ranging or captive (ROLLAND et al., 1985; ROUTMAN et al., 1985).

E. coli possess virulence markers such as O-antigens (ORSKOV et al., 1977; KUSECEK et al., 1984), haemolysin production (BALJER et al., 1986; SUTTORP et al., 1990), K-capsule production (WILLIAMS-SMITH and HUGGINS, 1980) and pilus production (YERUSHALMI et al., 1990), amongst others, which play significant role in the pathogenesis of human and animal strains of the micro organism. However, it is known that the majority of E. coli strains lead a symbiotic existence and are considered harmless (RADOSTITS et al., 1994).

Enteropathogenic E. coli (EPEC) strains are responsible for enteritic diseases in humans, principally associated with neonatal and infantile diarrhoea (ROBINS-BROWNE, 1987; DOYLE, 1990; ECHEVERIA et al., 1991). Although EPEC strains have been isolated to date, an animal reservoir has not been well established as is currently known for verocytotoxigenic E. coli (VTEC) where cattle serve as the major reservoir (MOHAMMED et al., 1986; KARMALI, 1989). Presently, there is a dearth of information on the prevalence of E. coli in wildlife. ADESIYUN (1999) recently found all strains of E. coli isolated from captive and free-ranging wildlife to be negative for haemolytic, mucoid and serogroup O157 strains.

In Trinidad and Tobago the efforts of the government to encourage wildlife farming has resulted in a dramatic increase in the number of backyard wildlife farmers, and hence contact between captive wildlife and their owners. Furthermore, wildlife meat constitutes a delicacy particularly during the annual hunting season. It is well known that during evisceration of animals, enteric micro-organisms contaminate carcases, if carried out in unsanitary conditions (ADESIYUN and KRISHNAN, 1995).

The specific objectives of the study were, therefore, to determine the prevalence of resistance to antimicrobial agents and enteropathogenic serogroups amongst E. coli isolates from free-ranging and captive wildlife (mammals, avian and reptiles) on Trinidad. The effect of management of wildlife on farms, and eating habits, on the characteristics of E. coli isolates was also investigated.

Materials and methods

The isolates of E. coli studies originated from hunted free-ranging mammalian, avian and reptilian wildlife during the October to February 1995-1996 and 1996-1997 hunting seasons in Trinidad and Tobago. Isolates of E. coli from captive wildlife were obtained from captive or farmed wildlife in cages or pens at government or private facilities or individual farms.

The procedure for sample collection from the various wildlife species was described earlier (ADESIYUN et al., 1998). Briefly, for hunted wildlife, participating licensed hunters tied up a 5-cm section of the small intestine of the respective specimen during evisceration. Tied segments of the intestines were placed in sterile plastic containers provided by the investigators. Samples were kept ice-cooled (4 °C) and transported to the laboratory within 24 h of collection. Intestinal contents which could not be sent ice-cooled to the laboratory within 24 h were stored frozen at -20 °C. For captive wildlife on farms, cloacal, rectal or faecal swabs were obtained directly from restrained animals or from freshly voided faeces in the enclosures of those animals which could not be easily restrained. Swabs of samples were dipped into 9 ml of Amies transport medium (ATM) (Difco, Detroit, Michigan, U.S.A.).

All wildlife (free-ranging or captive) samples were free of diarrhoea at the time of study, based on the consistency of faeces and history obtained from owners of captive wildlife. All captive wildlife were apparently healthy but it was not possible to determine the health status of hunted wildlife prior to their death.

Faecal, rectal or cloacal swabs of captive and free-ranging wildlife were inoculated directly onto eosin methylene blue (EMB) agar and streaked for isolation. Inoculated agar plates were incubated aerobically at 37 °C for 24 h. Colonies with characteristic metallic sheen were selected and subjected to biochemical tests using standard methods (MACFADDIN, 1980). In all, 2-5 colonies from each plate were selected for identification.

The disc diffusion method (ANONYmous, 1994) was used to determine the antibiograms of E. coli isolates. The following antimicrobial agents and concentrations were used: ampicillin (10 µg), cephalothin (30 µg), chloramphenicol (30 µg), gentamycin (10 µg), nalidixic acid (30 µg), neomycin (10 µg), streptomycin (10 µg) and sulphamethoxazole /trimethoprim (25 µg). The interpretation of sizes of zones of inhibition followed the recommendation of the disk manufacturer (Difco, Detroit, Michigan, U.S.A.).

The aim of the experiment was to test a total of 200 isolates of E. coli from all sources studied. Selection of isolates was by random sampling. Enteropathogenic E. coli (EPEC) antisera A, B and C (S.A. Scientific, Inc., Texas, U.S.A.) were used to type the selected isolates by slide agglutination. Each isolate was tested concurrently against the three antisera and only distinct agglutination visible within seconds of emulsifying the growth from blood agar medium in a drop of antiserum was considered positive. Positive and negative controls, kindly provided by the manufacturer, were used.

The c2 test for independence, with degree of freedom, where applicable, was used to compare the frequencies of resistance to antimicrobial agents and enteropathogenic serogroups amongst E. coli isolates from pairs of sources.

Results

From a total of 472 isolates from all sources tested, 451 (95.6%) were resistant to one or more antimicrobial agents (Table 1). The range of resistance was from 88.7% (94 of 106) for isolates from captive mammals to 100.0% (78 of 78) for isolates from free-ranging mammals. The difference was statistically significant (P<0.01; c2).

Table 1. Prevalence of resistance to antimicrobial agents amongst E. coli isolates from free-ranging and captive wildlife on Trinidad and Tobago

Animal source

Class of animal

N of isolated tested

N (%) of isolates resistanta

N (%) of E. coli isolates resistant to:

KFb

AMP

S

N

NA

SXT

C

CN

Hunted wildlife

Mammalsc

78

78 (100.0)

77 (98.7)

29 (37.2)

7
(9.0)

10 (12.8)

14 (17.0)

0
(0.0)

1
(1.3)

4
(5.1)

Captive wildlife

Mammalsd

106

94
(88.7)

79 (74.5)

23 (21.7)

43 (40.6)

16 (15.1)

16 (15.1)

16 (15.1)

12 (11.3)

1
(0.9)

Aviane

72

69
(95.8)

68 (88.9)

10 (13.9)

17 (23.6)

10 (13.9)

9 (12.5)

6
(8.3)

4
(5.6)

3
(4.2)

Reptilesf

13

12
(92.3)

10 (76.9)

3 (23.1)

2 (15.4)

1
(7.7)

2 (15.4)

0
(0.0)

0
(0.0)

0
(0.0)

Subtotal

191

175 (91.6)

153 (80.1)

36 (18.8)

62 (32.5)

27 (14.1)

27 (14.1)

22 (11.5)

16 (8.4)

4
(2.1)

Racing pigeons

Aviang

118

117 (99.2)

106 (89.8)

68 (56.7)

63 (53.4)

41 (34.7)

29 (24.6)

32 (27.1)

19 (16.1)

10 (8.5)

Wild birds

Avianh

85

81
(95.3)

80 (94.1)

29 (34.1)

27 (31.8)

26 (30.6)

4
(4.7)

1
(1.2)

1
(1.2)

14 (16.5)

Total

472

451 (95.6)

416 (88.1)

162 (34.3)

159 (33.7)

104 (22.0)

74 (15.7)

55 (11.7)

37 (7.8)

32 (6.8)

aResistant to one or more antimicrobial agents

bKF=cephalothin; AMP=ampicillin; S=streptomycin; N=neomycin; NA=nalidixic acid; SXT=sulphamethoxazole/trimethoprim; C=chloramphenicol; CN=gentamycin

cConsisted of agouti (Dacyprocta leporina)-60; opossum (Didelphis marsupialis)-9;
brocket deer (Mozama americana trinitalis)-4; lappe (Agouti paca)-3;
wild hog (Tayassa tajacu)-1; armadillo (Dasypus novemcinctus)-1

dMammals comprised agouti (Dacyprocta leporina)-68; brocket deer (Mozama americana trinitalis)-12; wild hog (Tayassa tajacu)-10; lappe (Agouti paca)-6; porcupine (Coendou prehensilis)-5; monkey (Cebus apella)-5

eComprised pigeons (Columba spp.)-8; orange-winged Amazon parrot (Amazon amazonica)-26; blue and gold or clue and yellow macaws (Ara araruara)-31; cacatoos (Cacatua spp.)-3; red-billed toucan (Ramphastos tucanas)-2; barn own (Tyto alba)-1; lovebird (Agapornis roseicollis)-1

fComprised savannah snakes (Amphisbaena alba)-5; red-eared sliders (Trachemys script elegens)-4; red-footed tortoise (Geochelone carbonaria)-3; spectacled caiman (Caiman crocodilus)-1

gColumba livia

hWild pigeons (Columba spp.)-49; ruddy ground doves or talpacoti dove (Columbigallina talpacoti)-10; yellow-headed black bird (Agelaius icterocephalus)-8; pied water tyrant (Fluvicola pica)-3; northern water thrush (Catharus spp.)-2 and other birds

The overall prevalence of resistance to cephalothin was very high (88.1%), moderate to ampicillin (34.3%), streptomycin (33.7%) and neomycin (22.0%) and comparatively low to chloramphenicol (7.8%) and gentamycin (6.8%). Seventy-four (15.7%) and 55 (11.7%) isolates of E. coli were resistant to nalidixic acid and sulphamethoxazole/trimethoprim, respectively. With the exception of cephalothin and gentamycin, E. coli isolates from racing pigeons exhibited the highest prevalence of resistance to the remaining six antimicrobial agents (ampicillin, streptomycin, neomycin, nalidixic acid, sulphamethoxazole/trimethoprim (SXT) and chloramphenicol) compared to isolates from other sources.

The prevalence of resistance in E. coli isolates from captive mammals was statistically significantly higher than was found in isolates from free-ranging mammals respectively, for streptomycin, 40.6% and 9.0% (P<0.001); sulphamethoxazole/trimethoprim, 15.1% and 0.0% (P<0.001); and chloramphenicol, 11.3% and 1.3% (P<0.01). Although the prevalence of resistance to neomycin in E. coli from captive wildlife (15.1%) was higher than was found in isolates from free-ranging mammals (12.8%), the difference was not statistically significant (P>0.05).

A comparison of the prevalence of resistance in E. coli isolates from racing pigeons to those from wild birds (i.e. free-ranging) revealed statistically significant higher resistance prevalence in isolates from racing pigeons to ampicillin, 56.7% and 34.1% (P<0,001); streptomycin, 53.4% and 31.8% (P<0.01); nalidixic acid, 24.6% and 4.7% (P<0.001); sulphamethoxazole/trimethoprim, 27.1% and 1.2% (P<0.001) and chloramphenicol, 16.1% and 1.2% (P<0.001).

Isolates of E. coli from wild birds were statistically significantly more resistant to antimicrobial agents than those from other captive birds to ampicillin, 34.1 and 13.9% (P<0.05); neomycin, 30.6% and 13.9% (P<0.05) and gentamycin, 16.5% and 4.2% (P<0.05). The prevalence of resistance amongst E. coli isolates from wild birds was also higher than was found in isolates from captive birds to cephalothin, 94.1% and 88.9%, and streptomycin, 32.6% and 23.6%, but the differences were not statistically significant (P>0.05).

Amongst captive reptilian isolates of E. coli, prevalence of resistance was 76.9%, 23.1%, 15.4%, 15.4% and 7.7% to cephalothin, ampicillin, streptomycin, nalidixic acid and neomycin, respectively. All reptilian isolates were sensitive to sulphamethoxazole/trimethoprim, chloramphenicol and gentamycin.

Out of a total of 451 isolates which exhibited resistance to one or more of the 8 antimicrobial agents used, 71 (15.7%) resistant patterns were observed (Table 2). The common resistance patterns were KF (30.8%), KF-S (12.9%), KF-AMP (12.0%), KF-S-AMP (4.4%) and KF-N (4.2%).

Table 2. Prevalence of resistance patterns amongst E. coli
isolates from wildlife on Trinidad and Tobago

Resistance patterns

N (%) of isolates

 

Resistance patterns

N (%) of isolates

KF

139 (30.8)

KF-SXT

5 (1.1)

KF-S

58 (12.9)

KF-S-N-CN

5 (1.1)

KF-AMP

54 (12.0)

KF-S-N-SXT-AMP

4 (0.9)

KF-S-AMP

20 (4.4)

KF-S-NA-N

4 (0.9)

KF-N

19 (4.2)

KF-S-NA

4 (0.9)

KF-AMP-N

13 (2.9)

KF-S-AMP-C

4 (0.9)

KF-AMP-NA

9 (2.0)

KF-S-N-SXT-NA-AMP-C

3 (0.7)

KF-S-AMP-N

8 (1.8)

KF-SXT-AMP

3 (0.7)

KF-S-N

7 (1.6)

KF-N-NA-AMP

3 (0.7)

KF-NA

6 (1.3)

KF-S-N-SXT-NA-AMP

3 (0.7)

KF-S-NA-AMP

6 (1.3)

KF-S-N-NA-AMP

3 (0.7)

KF-N-SXT-NA-AMP-C

5 (1.1)

KF-S-N-AMP-CN

3 (0.7)

KF-N-SXT-NA-AMP

5 (1.1)

Others

61 (13.5)

KF=cephalothin; S=streptomycin; AMP=ampicillin; N=neomycin; NA=nalidixic acid;
SXT=sulphamethoxazole/trimethoprim; C=chlramphenicol; CN=gentamycin

Table 3 shows the frequency of detection of enteropathogenic serogroups amongst E. coli isolates from captive and free-ranging wildlife. Overall, 46 (23.0%) of 200 isolates tested belonged to enteropathogenic serogroups with 22 (11.0%), 21 (10.5%) and 3 (1.5%) in serogroups A, B and C, respectively. The prevalence of enteropathogenic serogroups was lowest (7.7%) amongst captive reptilian isolates of E. coli and highest amongst isolates from wild birds (50.0%).

Table 3. Prevalence of E. coli isolates belonging to enteropathogenic serogroups in free-ranging and captive wildlife in Trinidad and Tobago

Source of isolates

Class of animal

N of isolates tested

N (%) isolates in enteropathogenic serogroupsa

N (%) of isolates agglutinated by:

Poly Ab

Poly Bb

Poly Cb

Hunted wildlife

Mammals

54

11 (20.4)

5 (9.3)

4 (7.4)

2 (3.7)

Captive wildlife

Mammals

25

12 (48.0)

5 (20.0)

7 (28.0)

0 (0.0)

Avian

24

3 (12.5)

3 (12.5)

0 (0.0)

0 (0.0)

Reptiles

13

1 (7.7)

0 (0.0)

1 (7.7)

0 (0.0)

Sub-total

62

16 (25.8)

8 (12.9)

8 (12.9)

0 (0.0)

Racing pigeons

Avian

54

4 (7.4)

1 (1.9)

3 (5.6)

0 (0.0)

Wildbirds

Avian

30

15 (50.0)

8 (2.7)

6 (20.0)

1 (3.3)

Total

200

46 (23.0)

22 (11.0)

21 (10.5)

3 (1.5)

aAgglutinated by polyvalent antiserum serogroups A, B, or C
bPolyvalent antisera A, B, or C

The prevalence of enteropathogenic serogroups in E. coli isolates from captive mammals (48.0%) was statistically significantly (P<0.05) higher than was found in isolates from free-ranging mammals (20.4%).

Enteropathogenic serogroups were detected in isolates of E. coli from wild birds (50.0%) at a significantly higher frequency than in isolates from racing pigeons, 7.4% (P<0.001) and from other captive birds, 12.5% (P<0.05).

Out of a total of 13 reptilian isolates of E. coli studied, only 1 (7.7%) belonged to an enteropathogenic serogroup, group C.

Discussion

It was evident that resistance to antimicrobial agents was generally high (95.4%) for E. coli isolates which originated from both captive and free-ranging wild life. The relatively high prevalence of resistance to antimicrobial agents amongst apparently healthy captive wildlife could have therapeutic implications. This is so because transfer of resistance factor is known to be common, particularly amongst enteric bacteria (ANDERSON, 1968; WIERUP, 1975; TOWNER, 1982).

It was surprising to detect that E. coli strains from free-ranging wildlife, especially mammals, exhibited moderate to high resistance to cephalothin, ampicillin, streptomycin, neomycin and nalidixic acid, while showing low resistance to chloramphenicol and gentamycin. These findings are at variance with the findings of ROUTMAN et al. (1985) who reported a very low prevalence of resistance in E. coli strains isolated from free-ranging African yellow baboons with no human contact. In that study, the prevalence of resistance to streptomycin, ampicillin and chloramphenicol was 6.7%, 6.0% and 0.7%, respectively. In contrast, ROLLAND et al. (1985) found higher resistance to tetracycline (94.1%), kanamycin (70.6%), ampicillin (47.1%) and cephalothin (17.6%) amongst coliform isolates from baboons which fed on human refuse compared to a corresponding prevalence of resistance of 7.5%, 47.5%, 2.5%, and 2.5%, respectively detected in isolates from baboons with no human contact. These findings in E. coli isolates from wildlife cannot be explained by reports that direct and repeated exposure to antibiotics produce strong selective pressures for the maintenance of antibiotic resistance in the enteric bacteria of humans and domestic animals (HIRSH et al., 1980; LEVY et al., 1981; GYLES, 1993). It is known, however, that colonization of the intestinal tract by resistant coliforms can occur even in the absence of such selection pressures, particularly as a result of contact with resistant bacteria in food or on environmental fomites (WRIGHT et al., 1976; LEVY, 1984). The very high prevalence of resistance found in free-ranging mammalian wildlife in the current study can therefore be partially explained by the fact that on some Trinidadian hunting grounds, occasional findings of grazing cattle occur. A similar reason may explain the high prevalence of multiple resistance found amongst E. coli isolates from wildlife. It was also of interest to note that E. coli strains isolated from cattle on Trinidad had comparable prevalence of resistance to streptomycin (83.8%), neomycin (33.3%), ampicillin (26.1%), chloramphenicol (3.6%) and gentamycin (3.2%) as reported by ADESIYUN and KAMINJOLO (1992). The fact that antimicrobial agents are readily available to livestock farmers in the country, and therefore often used unsupervised, may have been partially responsible for the resistance reported in E. coli isolates from livestock in Trinidad and Tobago.

The finding that E. coli isolates from racing pigeons had the highest prevalence of resistance to six of the eight antimicrobial agents tested, compared to isolates from other sources, is of management significance. This may be partly explained by the fact that a majority of racing pigeon owners are affluent in society and could readily afford the use of antimicrobial agents in their various lofts, particularly sulphamethoxazole/ trimethoprim, which was the drug choice the lofts visited. It was therefore no surprise to learn that E. coli isolates from racing pigeons had a significantly higher prevalence of resistance than isolates from either wild birds or other captive birds. Diet may, however, have played some role in the significantly higher prevalence of resistance found in E. coli isolates from free-ranging wild birds compared to other captive birds. This is because free-flying wild birds are more exposed to more diverse environments and diets. In fact, a majority of the wild birds sampled in the present study originated from a feed mill in Port of Spain, the nation's capital city, and from a solid sewage dump where exposure to E. coli strains from humans and livestock was probable. Wildlife has been documented to acquire resistant strains of aerobic bacteria from food or the environment (WRIGHT et al., 1976; LEVY, 1984; ROLLAND et al., 1985).

Regardless of the wildlife source of E. coli isolates, it appears that chloramphenicol, gentamycin and sulphamethoxazole/trimethoprim were most effective on the micro-organism. A similarly low prevalence of resistance was found in E. coli isolates from livestock in the same environment (ADESIYUN and KAMINJOLO, 1992). Gentamycin is not too commonly utilized by livestock farmers or captive wildlife farmers in the country because of high cost and the fact that it cannot be administered orally.

It was of potential clinical significance to consumers of wildlife meat that 23% of the 200 isolates of E. coli tested belonged to the enteropathogenic serogroups. A comparatively lower prevalence (7.8%) of E. coli isolates in enteropathogenic serogroups was detected amongst dairy cows faecal isolates in the same environment (ADESIYUN et al., 1997a). Enteropathogenic E. coli (EPEC) strains have been implicated in diarrhoeal diseases in humans (KNUTTON et al., 1987; KARMALI, 1989; DOYLE, 1990) and it is believed that they produce toxins (CLEARLY et al., 1985). The fact that wildlife meat has become a delicacy in Trinidad and Tobago may therefore pose a health risk to consumers. This is because hunted wildlife meat is often barbecued or cooked in the forest with a possibility of poor sanitary conditions prevailing, especially during evisceration and cooking. It is well known that E. coli constitutes the normal intestinal flora of both humans and animals (DRASAR and HILL, 1974; HOWE and LINTON, 1976; GYLES, 1993). It is recognized, however, that unlike the verocytotoxigenic E. coli (VTEC) strains for which, to date, cattle has been established as the major animal reservoir (MOHAMMED et al., 1986; KARMALI, 1989), no animal reservoir has been documented for EPEC strains. Furthermore, it is known that there is a marked species specificity for EPEC infections, for example, EPEC strains that cause disease in dogs, cattle, pigs and rabbits are different from those that cause disease in humans. However, the possibility of an interchange of EPEC strains between captive wildlife and their handlers cannot be ignored.

It was therefore of epidemiological relevance that a significantly higher prevalence of E. coli isolates from mammalian captive wildlife (48%) belonged to the enteropathogenic group(s) compared to what was found amongst E. coli isolates from free-ranging mammals (20%). This finding may again be explained, in part, by the close association of captive mammals with their handlers, or by exposure to feeds contaminated by EPEC strains, as was reported for resistant strains of E. coli isolated from wildlife exposed to human environments (ROLLAND et al., 1985; ROUTMAN et al., 1985). This finding was, however, reversed for free-flying wild birds and captive avian wildlife (caged birds and racing pigeons) where a significantly higher prevalence of E. coli isolates in enteropathogenic serogroups was found in free-flying wild birds (50%) compared to captive wild birds (13%) and racing pigeons (7%). A possible explanation for this apparent incongruity may be the locales from which the free-flying wild birds in the current study were sampled, i.e. feed mill and human solid sewage dump, as earlier stated. It is improbable that caged avian wildlife and racing pigeons will be exposed to human garbage in their diet.

It was concluded that the prevalence of resistance to antimicrobial agents amongst E. coli isolates was high and may pose therapeutic problems in captive wildlife, while EPEC strains may pose a health risk to consumers of wildlife meat.


Acknowledgements
We thank the Pan American Health Organization (PAHO) for funding the project. The laboratory assistance rendered by Nadira Seepersadsingh is also acknowledged, and we are grateful to Ms. Colette Marina Gall for typing the manuscript.


References

ADESIYUN, A. A. (1999): Absence of Escherichia coli 157 in a survey of wildlife from Trinidad and Tobago. J. Wildlife Dis. 35, 115-120.

ADESSIYUN, A. A., M. CAMPBELL, J. S. KAMINJOLO (1997a): Prevalence of bacterial enteropathogens in pet dogs in Trinidad. J. Vet. Med. B. 44, 19-27.

ADESIYUN, A. A., J. S. KAMINJOLO (1992): Susceptibility to antibiotics of Escherichia coli isolated from diarrhoeic and non-diarrhoeic livestock in Trinidad. Revue d'Elev. Med. Vet. des. Pays. Trop. 45, 260-262.

ADESIYUN, A. A., C. KRISHNAN (1995): Occurrence of Yersinia enterocolitica 3, Listeria monocytogenes and thermophilic Campylobacter spp. in slaughter pigs and caracasses in Trinidad. Food Microbiol. 12, 99-107.

ADESIYUN, A. A., N. SEEPERSADSINGH, L. INDER, K. CAESAR (1988): Some bacterial enteropathogenes in wildlife and racing pigeons in Trinidad. J. Wildlife Dis. 34, 73-80.

ADESIYUN, A. A., L. A. WEBB, H. ROMAIN, J. S. KAMINJOLO (1997b): Prevalence and characteristics of strains of Escherichia coli isolated from milk and faeces of cows on dairy farms in Trinidad. J. Food Prot. 60, 1174-1181.

ANDERSON, J. D. (1968): The ecology of transferable drug resistance in the Enterobacteria. Ann. Rev. Microbiol. 22, 131-180.

ANONYMOUS (1994): Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. 4th ed. Approved Standard. NCCLS M7-A4 17, 1-28.

BALJER, G., M. SAITO, A. MAYR (1986): Nachweis von enterotoxinbilden Escherichia coli stammen (ETEC) bei Hunden mit akuter gastroenteritis. Prakt. Tierarzt 67, 427-477.

BROES, A., M. DROLET, M. JACQUES, J. M. FAIRBROTHER, W. M. JOHNSON (1988): Natural infection with an attaching and effacting Escherichia coli in a diarrhoeic puppy. Can. J. Vet. Res. 52, 280-282.

CLEARY, T. G., MATHEWSON, E. FARIS, L. K. PICKERRING (1985): Shiga-like cytotoxin production by enteropathogenic Escherischia coli serogroups. Infect. Immun. 47, 335-337.

DRASAR, B. S., M. J. HILL (1974): The distribution of bacterial flora in the intestine. In: Human intestinal flora (Drasar, B. S., M. J. Hill, Eds.). Academic Press. London. pp. 36-43.

DOYLE, M. P. (1990): Pathogenic Eschereichia coli, Yersinia enterocolitica and Vibrio parahaemolyticus. Lancet 336, 1111-1115.

ECHEVERIA, P., F. ORSKOV, I. ORSKOV, I. S. KNUTTON, F. SCHENTZ, J. E. BROWN, U. LEXOMBOON (1991): Attaching and effacing enterropathogenic Escherichia coli as a cause of infantile diarrhoea in Bangkok. J. Infect. Dis. 164, 550-554.

GYLES, C. L. (1993): In: Pathogenesis of Bacterial Infections in Animals. Iowa University Press. Ames. pp. 164-189.

HIRSH, D. C., G. V. LING, A. L. RUBY (1980): Incidence of R-plasmids in fecal flora of healthy household dogs. Antimicrob. Agents. Chemoth. 17, 313-315.

HOWE, K., A. H. LINTON (1976): The distribution of O-antigen types of Escherichia coli in normal calves, compared with man, and their R-plasmid carriage. J. Appl. Bacteriol. 40, 317-330.

KARMALI, M. A. (1989): Infection by verocytotoxin producing Escherichia coli. Clin. Microbiol. Rev. 2, 15-38.

KNUTTON, S. L., D. R. LLOYD, A. McNEISH (1987): Adhesions of enteropathogenic Escherichia coli to human intestinal erythrocytes and cultured human intestinal mucosa. infect. Immun. 55, 69-77.

KUSECEK, B., A. WLOOCH, A. MERCER, V. VAINSANEN, G. PLUSCKHE, T. KORHONEN, M. ACHTMAN (1984): Lipopolysaccharide capsule and fimbriae and virulence factors among 01, 07, 016, 018, 075 and K1, K5 or K100 Escherichia coli. Infect. Immun. 43, 368-379.

LEVY, S. B. (1984): Antibiotic-resistant bacteria in food of man and animals. In: Antimicrobials and Agriculture (Woodbine, M., Ed.). Butterworths. London. pp. 525-532.

LEVY, S. B., R. C. CLOWES, E. L. KOENIG (1981): Molecular biology, pathogenicity and ecology of bacterial plasmids. Plenum Publishing Corp. New York.

MACFADDIN, J. F. (1980): Biochemical tests for identification of medical bacteria. Williams and Wilkins. Baltimore.

MOHAMMED, A., J. S. M. PEIRIS, E. A. WIJEWANTA (1986): Serotypes of verocytotoxigenic Escherichia coli from cattle and buffalo calf diarrhoea. FEMS Microbiol. Lett. 35, 261-265.

ORSKOV, I., F. ORSKOV, B. JANN, K. JANN (1977): Serology, chemistry and genetics of O and K antigens of Escherichia coli. Bacteriol. Rev. 41, 667-710.

PEETERS, J. E., R. GEEROMS, F. ORSKOV (1988): Biotype, serotype and pathogenicity of attaching and effacing enteropathogenic Escherichia coli strains isolated from diarrhoeic commercial rabbits. Infect. Immun. 56, 1442-1448.

RADOSTITS, O. M., D. C. BLOOD, C. C. GAY (1994): Veterinary Medicine, 8th ed. Bailere Tindall. London. pp. 618-627.

ROBBINS-BROWNE, R. M. (1987): Traditional enteropathogenic Escherichia coli infantile diarrhoea. Rev. Infect. Dis. 9, 28-53.

ROLLAND, R. M., G. HANSFATER, B. MARSHALL, S. D. LEVY (1985): Antibiotic-resistant bacteria in wild primates: increased prevalence in baboons feeding on human refuse. Appl. Environ. Microbiol. 49, 791-794.

ROUTMAN, E., R. D. MILLER, J. PHILLIPS-CONROY, D. HARTL (1985): Antibiotic resistance and population structure in Escherichia coli from free-ranging African yellow baboons. Appl. Environ. Microbiol. 50, 749-754.

SMITH, H. W. (1975): Persistence of tetracycline resistance in pig E. coli. Nature 258, 628-631.

SUSSMAN, M. (1985): Escherichia coli in human and animal disease. In: The Virulence of Escherichia coli (Sussman, M., Ed.). Academic Press Inc. New York. pp. 7-45.

SUTTORP, N., B. FLOER, H. SCHNITTLER, W. SEEGER, S. BHAKDI (1990): Effects of Escherichia coli haemolysin on endothelial cell function. Infect. Immun. 58, 3796-3801.

TOWNER, K. J. (1982): Resistance to trimethoprim among urinary tract isolates in the United Kingdom. Rev. Infect. Dis. 4, 456-460.

WALTON, J. R. (1971): The public health implications of drug resistant bacteria in farm animals. Annal. N. Y. Acad. Sci. 182, 358-361.

WASHINGTON, J. A. (1979): The effects and significance of subminimal inhibitory concentrations of antibiotics. Rev. infect. Dis. 1, 781-786.

WIERUP, M. (1975): Antibiotic resistance and transferable antibiotic resistance of Escherichia coli isolated from Swedish calves 5 and 30 days old. Nordisk Vet. 27, 77-84.

WILLIAMS-SMITH, H., M. B. HUGGINS (1980): The association of the O18, K1 and H7 antigens and the Col v plasmid of a strain of Escherichia coli with its virulence and immunogeneticity. J. Gen. Microbiol. 121, 387-400.

WRIGHT, C., S. D. KOMINSON, R. B. YEE (1976): Enterobacteriaceae and Pseudomonas aeruginosa recovered from vegetable salad. Appl. Environ. Microbiol. 31, 453-454.

YERUSHALMI, Z., N. I. SMORODINSKY, M. W. NAVEH, E. Z. RON (1990): Adherence of pili of avian strains of Escherichia coli O78. Infect. Immun. 58, 1129-1131.

Received: 21 July 1999
Accepted: 15 December 1999



ADESIYUN, A., M. DOWNES: Uspjesnost djelovanja antimikrobnih lijekova na enteropatogene seroskupine bakterije Escherichia coli izdvojene iz divljih zivotinja u Trinidadu i Tobagu. Vet. arhiv 69, 335-347, 1999.

SAZETAK

Upotrebom disk difuzijske metode odredeni su antibiogrami na bakteriju Escherichia coli izdvojenu iz divljih zivotinja (sisavaca, ptica i gmazova) iz slobodne prirode i zatocenistva u Trinidadu i Tobagu, kao i na njezine pojedine izdvojene enteropatogene seroskupine. Na ukupno 472 izolata bakterije E. coli istrazeno je djelovanje osam razlicitih antimikrobnih lijekova. Od svih izolata, 451 (95,6%) izolat bio je otporan na jedan ili veci broj antimikrobnih lijekova. Otpornost je bila visoka prema cefalotinu (88,1%), umjerena prema ampicilinu (34,3%), streptomicinu (33,7%) i neomicinu (22,1%), a relativno niska prema kloramfenikolu (7,8%) i gentamicinu (6.8%), bez obzira na podrijetlo izolata. Otpornost izolata bakterije E. coli podrijetla od zivotinja iz zatocenistva bila je statisticki znacajno (P<0,0; c2) visa nego izolata podrijetla od slobodno zivucih sisavaca na streptomicin, sulfametoksazol/trimetoprim i kloramfenikol. Ipak, otpornost izolata od slobodno zivucih sisavaca bila je znacajno (P<0,05; c2) visa nego onih izdvojenih iz sisavaca iz zatocenistva na ampicilin i cefalotin. Izolati bakterije Escherichia coli izdvojeni iz golubova pismonosa bili su statisticki znacajno (P<0,01; c2) otporniji nego izolati izdvojeni iz divljih ptica na ampicilin, streptomicin, nalidiksicnu kiselinu, sulfametoksazol/trimetoprim i kloramfenikol. Osjetljivost izolata bakterije E. coli izdvojenih iz ptica u zatocenistvu bila je ipak, statisticki znacajno (P<0,05; c2) slabija nego izolata izdvojenih iz divljih ptica na ampicilin, neomicin i gentamicin. Medu 200 nasumce odabranih i istrazenih izolata bakterije E. coli, 46 (23,0%) pripadalo je enteropatogenim seroskupinama, i to seroskupini A 22 (11,0%), seroskupini B 21 (10,5%) i seroskupini C 3 (1,5%). Otpornost enteropatogenih seroskupina bakterije E. coli izdvojenih iz slobodno zivucih sisavaca (20,4%) bila je statisticki znacajno (P<0,05; c2) slabija nego izdvojenih iz sisavaca iz zatocenistva (48,0%). Slicno su, izolati bakterije E. coli iz divljih ptica imali statisticki znacajno (P<0.05; c2) vecu (50%) ucestalost enteropatogenih seroskupina nego izolati izdvojeni iz golubova pismonosa (7,4%) ili ptica iz zatocenistva (12,5%). Iz navedenog se zakljucuje da veca otpornost izolata bakterije E. coli na antimikrobne lijekove moze predstavljati probleme u lijecenju divljih zivotinja u zatocenistvu, dok pojava enteropatogenih seroskupina medu izolatima moze predstavljati zdravstvenu opasnost za potrosace mesa divljih zivotinja.

Kljucne rijeci: Escherichia coli, antibiogrami, enteropatogene seroskupine, divlje zivotinje, Trinidad i Tobago


Back