key: cord-0010688-f68vilm8
authors: CID, D.; PÍRIZ, S.; RUIZ‐SANTA‐QUITERIA, J. A.; VALLE, J.; VADILLO, S.; de la FUENTE, R.
title: In vitro susceptibility of Escherichia coli strains isolated from diarrhoeic lambs and goat kids to 14 antimicrobial agents
date: 2008-04-17
journal: J Vet Pharmacol Ther
DOI: 10.1111/j.1365-2885.1996.tb00070.x
sha: dd818980fbdb85698d1576ee0d1e2305d0a4c4c2
doc_id: 10688
cord_uid: f68vilm8

The in vitro activities of 14 anti‐microbial agents were determined against 92 strains of E. coli isolated from lambs (60 strains) and kids (32 strains) affected by neonatal diarrhoea. The overall percentage of resistant strains to streptomycin, sulphadimethoxine and tetracycline was very high (above 70%). A high level of resistance (from 30% to 50%) to ampicillin, kanamycin, neomycin and chloramphenicol was also detected. The E. coli strains were highly susceptible to cephalosporins, polymyxin and quinolones. Most of the strains showed multiresistance: 77.2% of isolates were resistant to at least two antibiotics, 55.4% were resistant to at least four antibiotics and 33.7% were resistant to at least six antibiotics. A total of 34 antibiotypes could be distinguished.

Neonatal diarrhoea is one of the most common and economically devastating problems encountered in farm animals (Holland. 1990) . Certain E. coli strains are an important cause of diarrhoea in all farm animal species (Holland, 1 9 9 0 Wray et al.. 1993) . The role of enterotoxigenic E. coli (ETEC), which produces enterotoxins and expresses fibrial colonization factors in neonatal diarrhoea, has been well established. In addition, other non-enterotoxigenic E. coli strains that produce other toxins [verotoxigenic E. coli (VTEC) and necrotoxigenic E. coli (NTEC)] or cause a characteristic histological lesion [attaching and effacing E. coli (AEEC)] have also been associated with neonatal diarrhoea in domestic animals (Holland. 1990; De Rycke. 1991) . Although the etiological diagnosis in small ruminants is not determined for a large percentage of cases of neonatal diarrhoea, application of anti-microbial therapy is a common feature in clinical practice. However, high proportions of resistant E. coli strains in farm animals and even in animal feed lots have been reported (Coates & Hoopes, 1980 : Jackson, 1981 Wood et al.. 1983; Prescott et al.. 1984; Gonzalez & Blanco, 1985 . 1986 Aalbzk et al.. 1991; Adesiyun, 1992; Wray et al., 1993 : Blanco et aL.1993 . The aim of this study was to evaluate the susceptibility of E. coli strains isolated from diarrhoeic lambs and kids to different antimicrobial agents.

The study was performed with 92 strains of E. coli isolated from lambs (60 strains) and goat kids (32 strains) affected by neonatal diarrhoea. The bacterial strains were isolated in our laboratory between 1989 and 1992 from 51 outbreaks of neonatal diarrhoea on farms located in the central region of Spain. One strain per animal was selected. None of the strains studied produced F5 or F41 fimbrial antigens or heat-stable enterotoxin STa (Cid et al.. 1993) . Fifty-four of the strains were selected because they produce one of the following potential virulence factors: 21 strains produced F17 fimbrial antigen, 16 were VTEC, 12 were NTEC and five produced heat-labile enterotoxin. The remaining 38 strains, which did not produce any of the fimbrial antigens or toxins studied, were selected for comparison.

Faecal samples from all animals from which E. coli strains were selected were negative to rotavirus and coronavirus as assessed using a comercial ELISA kit (Tetravalent, VCtoquinol, Magny-Vernois, France).

The anti-microbial agents studied were provided by the manufacturers: ampicillin (Gema, Barcelona, Spain), cephalexin (Antibioticos, Madrid, Spain), cefazolin (Antibioticos), cefamandole (Idly. Alcobendas. Madrid, Spain), polymyxin B (Sigma Quimica, Alcobendas, Madrid, Spain), streptomycin (Antibibticos). kanamycin (Rasfler, Madrid, Spain), gentamicin (Indukern. Barcelona. Spain), neomycin (Indukern), nalidixic acid (Casens-Fisons. Zaragoza, Spain), oxolinic acid (Indukern), sulphadimethoxine (Indukern), chloramphenicol (Fluka Chemika-Biochemika. Alcobendas, Madrid, Spain), tetracycline (Indukern). The antibiotics were dissolved and diluted as recommended by the manufacturers. Fresh dilutions of all compounds were prepared daily. 0 1996 Blackwell Science Ltd 397 Anti-microbid susceptibility testing In vitro susceptibility tests were performed by the agar dilution method according to the recommendation of the National Committee for Clinical Laboratory Standards (NCCrS, 1990) . Mueller-Hinton agar (Difco) was used for the agar dilution procedure. The standarized inocula were prepared by inoculating each strain into 2 mL of brain heart infusion (BHI) broth (Difco). The tubes were incubated at 37°C until the turbidity reached a density corresponding to McFarland nephelometer standard no. 0.5. Bacterial cultures were diluted 1 : l O in 0.9% NaCl solution and 1 pL of each bacterial suspension was applied by use of a Steers replicator to the anti-microbial-containing agar plates, yielding an inoculum of approximately lo4 CW per spot. Plates were incubated at 37°C for 24 h. Minimal inhibitory concentration (MIC) was read as the lowest concentration of anti-microbial agent that suppressed visible bacterial growth. One colony or a slight haze in which no distinct colonies could be observed was ignored. Reference strain E. coli ATCC 25922 was included as an internal control in all parts of the study. The range of interpretative categories of susceptibility for each anti-microbial agent were those recommended by the National Committee for Clinical Laboratory Standards (NCCLS, 1990 ).

The significance of dilferences was determined by chi-square (x') test. A level of P < 0.05 was considered to be significant.

The results of in vitro susceptibilities to 14 anti-microbial agents of the 92 strains of E. Coli studied are shown in Table 1 . The range and MZC50 and MZC90 (the lowest concentration of anti-microbial agent at which there was no growth of 50% and 90% of strains) of each of the 14 agents, as well as the percentage of E. cob resistant strains were determined ( Table 1) .

The overall percentage of resistant strains to streptomycin, sulphadimethoxine and tetracycline was very high (above 70%). Moreover, a high level of resistance (from 30% to 50%) to ampicillin. kanamycin, neomycin and chloramphenicol was also detected. E. cob strains were highly susceptible to cephalosporins (only one ovine isolate was resistant to cefamandole and one caprine isolate was resistant to cephalexin), polymyxin and the two quinolone compounds tested (only one ovine isolate was resistant to nalidixic acid).

In general, the different groups of anti-microbial agents tested showed similar activities against the E. coli strains isolated from lambs and from kids. However, significantly higher rates of resistance (P<O.O5) were found in kids' isolates than in those from lambs to ampicillin (59.3% in strains from kids vs. 35% in strains from lambs), kanamycin (65.5% vs. 35%). gentamicin (28.1% vs. 5%). neomycin (53.1% vs. 18%) and sulphadimethoxine (90.6% vs. 66.6%). Higher MCs0 values were observed for caprine isolates than for ovine isolates for ampicillin (64 pg/mL vs. 4 pg/mL), streptomycin (128 pg/mL vs. 64 pg/ mL), kanamycin (256 pg/mL vs. 4 pg/mL), neomycin (64 pg/ mL vs. 1 pg/mL) and chloramphenicol (64 pg/mL vs. 8 pg/mL).

No significant digerences between the 54 strains producing a potential virulence factor and the 38 non-fimbriated nontoxigenic strains were found in the rates of resistance against any of the antimicrobials tested. However, VTEC strains were significantly less resistant to kanamycin and neomycin than F17 * strains. and significantly less resistant to streptomycin, kanamycin and neomycin than non-fimbriated non-toxigenic strains.

Fourteen strains (15.2%) were sensitive to all the antimicrobials tested. Only seven strains (7.6%) were resistant to one antibiotic. Thus, most of the strains showed multiiesistance: 77.2% of isolates were resistant to at least two antibiotics, 55.4% were resistant to at least four antibiotics and 33.7% were resistant to at least six antibiotics. About 16% of the strains were resistant to seven or eight antibiotics.

VTEC strains showed the lowest rate of multidrug resistance with an average of 2.1 resistances per strain, whereas F 1 7 + , NTEC, LT+ and non-fimbriated non-toxigenic strains showed rates of 4.6, 4. 4 and 4.2 respectively.

A total of 3 4 antibiotypes could be distinguished. The most frequently occurring are those shown in Table 2 . . The use of drugs does not induce resistance but rather provides an intense selection pressure which eliminates the susceptible bacteria in the host and spares the resistant ones (Hinton et al.. 1986 ). This is particularly true for bacteria that inhabit the intestinal tract because they are readily exposed to the actions of antibiotics administered orally either at prophylactic or therapeutic levels or when incorporated in animal feed as growth promoters (Hinton et al.. 1986 ). Studies of E. coli isolated from different animal species showed a relationship between the degree of anti-microbial drug use and extent of resistance (Linton, 1977) . Therefore, antibiotic resistance is usually prevalent in intensively reared calves, pigs and poultry, whereas adult grazing cattle and sheep have a very low incidence of resistance among their faecal E. coli (Linton, 1977 : Hinton et al., 1985 .

Most of the veterinary reports on resistance of E. coli relate to calves, pigs and poultry. Comparatively few reports on antimicrobial resistance of ovine and caprine E. coli strains are available. In general, the frequency of resistance to different anti-microbial classes used in this study was higher than those reported in other studies. Wray et al. (1993) reported that more than 50% of ovine E . coli strains were resistant to chlortetracycline, sulphonamides and ampicillin. The incidence of resistance to other drugs tested by these authors was lower: neomycin (38%), furazolidone ( < 1%0), chloramphenicol (18%), apramycin ( < 1%) and streptomycin (48%). and 30.2% of the isolates were sensitive to all the anti-microbials tested. Moreover, the percentage of sensitive E. coli strains isolated from newborn lambs by Corlu et al. (1990) to different anti-microbials was higher than those described in this study. On the other hand, the percentages of resistant strains from lambs and kids to streptomycin, tetracycline, kanamycin and neomycin reported by Adesiyun (1992) in Trinidad are similar to those found in this study. However, the incidence of resistance to ampicillin and chloramphenicol found by Adesiyun (1992) was very low in comparison with that found in this study.

The percentages of resistant strains to ampicillin, aminoglycosides, sulphonamides, tetracycline and chloramphenicol found in this study are similar to those previously reported for bovine E. coli (Coates & Hoopes, 1980 : Sherwood et al., 1983 : Gonzalez & Blanco, 1989 Blanco et al., 1993 : Wray et al., 1993 . However, higher rates of resistance to nalidixic acid and cephalothin (a fist generation cephalosporin) have been described for bovine E. coli isolates (Pohl et al., 1991 : Blanco et al., 1993 than those observed in this study.

Some reports suggest that pathogenic E. coli are more likely to be resistant than non-pathogenic strains (Gyles et al.. 1977) . However, there is no conclusive evidence for this suggestion, since in several studies ETEC strains have been found to be more sensitive to antibiotics than non-ETEC strains (Brunton et al., 1980; De Boy et al., 1980 : Gonzalez & Blanco, 1985 . Furthermore, no major differences were observed between the incidence of resistance recorded for pathogenic and nonpathogenic E. coli isolates from pigs (Franklin, 1984) . On the other hand, Blanco et al. (1993) found that E. coli strains isolated from diarrhoeic calves were more resistant to antibiotics and had a higher rate of multidrug resistance than those isolated from healthy calves. A similar finding has been reported for chickens (Chularisi & Suthienkul, 1989) . Nevertheless, Adesiyun (1992) found that the prevalence of resistance to antibiotics was similar for strains isolated from diarrhoeic and non-diarrhoeic lambs and kids.

The strains included in this study were all isolated from animals with diarrhoea, although their association with the disease could not be established. Thus, the classification of F17+, VTEC and NTEC strains as potentially pathogenic is tentative, since there is little information about the role of this fimbriae and toxins in disease of sheep and goats (Wray et al.. 1993) .

The results of this study for VTEC are in contrast to those previously described for strains isolated from diarrhoeic calves by Gonzalez and Blanco (1989) . who found that VTEC were significantly more resistant to different antibiotics than NTEC strains and non-VTEC non-NTEC strains, and that multiple drug resistance was higher in VTEC than in NTEC and non-VTEC non-NTEC strains. Burnens et 01. (1 995) . however, found that 4 1 of 50 VTEC strains isolated from cattle in herds with and without calf diarrhoea were sensitive to all the antibiotics tested.

The reason for the significantly high rates of resistance of kids' isolates for Merent anti-microbials is not clear, since the housing conditions, the management of animals and the use of antibiotics in sheep and goat farms in the central region of Spain are broadly similar. Nevertheless, Adesiyun (1992) also found higher rates of resistance in kid than in lamb isolates for streptomycin, tetracycline, kanamycin and neomycin.

In this study the highest rates of resistance have been detected against the anti-microbial drugs most commonly used by Spanish veterinary clinics for treatment of diarrhoea in farm animals (sulphonamides. tetracyclynes, streptomycin), while the E. coli strains studied were highly susceptible to rarely used antimicrobial agents (cephalosporins, quinolones, polymixins).

Neonatal diarrhoea in lambs and kids is often treated with antimicrobial drugs. However, antibiotic therapy is frequently ineffective, partly due to the presence of drug-resistant strains and the failure to identify drug sensitivity. The therapeutic implications of our findings in the treatment of neonatal diarrhoea in small ruminants should be emphasized.

In this study a very high rate of multi-resistant E. coli strains was found. More than 30% of the strains were resistant to at least six anti-microbial agents. Multiple drug-resistant isolates of E. coli strains from farm animals have been reported worldwide (Coates & Hoopes. 1980 : jakson. 1981 Prescott et al., 1984;  Gonzalez & Blanco. 1985 . 1986 : Aalbzk et d., 1991 : Adesiyun. 1992 : Wray et al., 1993 Blanco et d., 1993) . The seriousness of the problem arises from the fact that once multiresistant organisms develop they can persist in the host or in the environment in the absence of antibiotic selection (Hinton, 1986) . This may be of particular importance since cattle (Orskov et al., 1987; Borzyck et al., 1987 : Gonzalez & Blanco, 1989 ) and probably sheep and goats (Doyle & Schoeni. 1987; Beutin et al., 1993) might act as reservoirs of certain E. coli strains pathogenic for man. Nevertheless, as suggested by Hinton (1986) , the importance of animals as a source of resistant E. coli should not be taken out of context because the medical use of antibiotics is principally responsible for the presence of resistance among bacterial isolates from man.

Prevalence of antibiotic-resistant Escherichia coli in Danish pigs and cattle. Acta Pathologica. Microbiologica et Zmrnunologica Scandinavica

Susceptibility to antibiotics of Escherichia coli strains isolated from diarrhoeic and non-diarrhoeic livestock in Trinidad. Revue D'Elevage et de Medecine Vethinuire des Pays Tropicaux

Resistencia a antibioticos en E. coli de origen bovino

Bovine reservoir for verotoxin-producing Escherichia

99 3 ) Prevalence and some properties of verotoxin (Shiga-like Toxin)-producing E. coli in seven different species of healthy domestic animals

Fnterotoxigenic Escherichia coli in central Canada

Prevalence and clinical significance of vero-cytotoxin-producing Escherichia coli (VTEC) isolated from cattle in herds with and without calf diarrhoea

Antimicrobial resistance of Escherichia coli isolated from chickens

F17 fimbriae in Escherichia coli from lambs and kids

Sensitivities of E. coli isolated from bovine and porcine enteric infections to antimicrobial antibiotics

Konya bolgesindeki ishalli kuzulardan izole edilen E. coli suslarinin cesitli ozellikleri uzerinde incelemeler. Doga. Turk Veterinerlik ve Hayvancilik Dergisi

Antibiotic resistance in enterotoxigenic and non-enterotoxigenic Escherichia coli

Les colibacilles producteurs de cytotoxines: importance en m6decine vkterinaire et en santk publique. Annales de Recherches Vikrinaires

Isolation of Escherichia coli 0157:H7 from retail fresh meats and poultry

Antimicrobial drug resistance in porcine enterotoxigenic Escherichia coli of 0-group 149 and non-enterotoxigenic Escherichia coli. Veterinary Microbiology

Relation between antibiotic resistance and number of plasmids in enterotoxigenic and non-enterotoxigenic E. coli strains

Colonization antigens, antibiotic resistance and plasmid content of enterotoxigenic Escherichia coli isolated from piglets with diarrhoea in Galicia (north-western Spain). Veterinary Microbiology

Serotypes and antibiotic resistance of verotoxigenic (VTEC) and necrotizing (NTEC) E. coli strains isolated from calves with diarrhoea

Naturally occurring plasmid carrying genes for enterotoxin production and drug resistance

Antibiotic-resistant coliform in the waste of piggeries and dairy farms. Iapanese Iournal of Veterinary Science

The ecology of Escherichia coli in animals including man with particular reference to drug resistance

The ecology of Escherichia coli in calves reared as dairy-cow replacements

The ecology of drug resistance in enteric bacteria

Antimicrobial susceptibility of E. coli from diarrhoeic lambs and kids 401

Some infectious causes of diarrhea in young farm animals

1) A survey of antibiotic resistance of E. coli isolated from farm animals in Great Britain from 1971-1977

Antibiotic resistance: the present situation reviewed. Veterinary Record, 100. 354-360. National Committee for Clinical Laboratory Standards

Cattle as reservoir of verotoxinproducing Escherichia coli 0157:H7

Antibiogrames des entkrobacteries pathogenes pour les animaw d'elevage et les pigeons, isoltks en Belgique de 1986 a 1990. Annales de Mkdecine Vitirinaire

Antimicrobial drug susceptibility of bacteria isolated from disease processes in cattle, horses, dogs and cats

Prevalence of enterotodgenic E. coli in calves in Scotland and northern England

Escherichia coli resistant to tetracycline and to other antibiotics in the faeces of UK chickens and pigs in 1980

Antimicrobial resistance of gram-negative bacteria isolated from foods in Mexico

ACKNOWLEDGMENT This work was supported by the CICYT (grants Gan 598190 and AGF-9 5-0834).