Resistance to antibiotics in Gram-negative bacteria isolated from broiler carcasses

Resistência a antibióticos em bactérias Gram-negativas isoladas de carcaças de frangos

Abstracts

One hundred and ninety-seven isolates of Gram-negative bacteria, comprising 10 genera, were isolated from poultry carcasses at a processing plant in order to investigate resistance to low levels of antibiotics. The samples were taken just after evisceration and before inspection. Most of the isolates were of Samonella and Escherichia. Other genera present were Enterobacter, Serratia, Klebsiella, Kluyvera, Erwinia, Citrobacter, Pseudomonas and Aeromonas. Distinct profiles of antibiotic resistance were detected. Resistance to more than two antibiotics predominated and spanned several classes of antibiotics. Salmonellae and escherichiae were mainly resistant to the aminoglycosides, followed by tetracycline, nitrofuran, sulpha, macrolide, chloramphenicol, quinolones and beta-lactams. Most isolates were sensitive to 30mug/ml olaquindox, the growth promoter in use at the time of sampling. However, many were resistant to a level of 10mug/ml and 13mug/ml olaquindox, levels present in the gut due to the dilution in the feed. The results suggest a possible role of low level administration of antibiotics to broilers in selecting multi-resistant bacteria in vivo.

Poultry; antibiotic; multi-resistance; bacteria; growth promoter


Para verificar a resistência a baixos níveis de antibióticos foram obtidos e identificados 197 isolados de bactérias Gram negativas pertencentes a 10 gêneros. Os isolados foram coletados em abatedouro industrial, imediatamente após a evisceração e antes do serviço de inspeção. Salmonella e Escherichia foram os gêneros identificados com maior freqüência. Os demais gêneros foram Aeromonas, Enterobacter, Serratia, Klebsiella, Citrobacter, Erwinia, Kluyvera, Pseudomonas e Aeromonas. Diferentes perfis de resistência a antibióticos foram detectados. A resistência a mais de dois antibióticos foi freqüente na maioria dos isolados e incluiu diversas classes de antibióticos. As bactérias dos gêneros Salmonella e Escherichia apresentaram maior percentagem de isolados resistentes à classe dos aminoglicosídeos, seguida das classes de tetraciclina, nitrofurano, sulfa, macrolídeo, cloranfenicol, quinolona e â-lactâmico. A maioria dos isolados foi sensível a 30mig/ml de olaquindox, promotor de crescimento usado no período da coleta das amostras. Foram observados isolados com resistência ao olaquindox nas concentrações de 10mig/ml e 13mig/ml, níveis presentes no intestinos das aves devido à diluição da ração. Os resultados sugerem que os baixos níveis de antibióticos administrados aos frangos podem estar selecionando "in vivo" bactérias multirresistentes.

Frango; antibiótico; multirresistência; bactéria Gram-negativa; promotor de crescimento


Resistance to antibiotics in Gram-negative bacteria isolated from broiler carcasses

[Resistência a antibióticos em bactérias Gram-negativas isoladas de carcaças de frangos]

M.A.S. Moreira, C.A. Moraes

Departamento de Microbiologia

Universidade Federal de Viçosa – UFV

Rua P.H. Rolfs s/n

36571-000 - Viçosa, MG

Recebido para publicação em 24 de setembro de 2001.

E-mail: mmoreira@alunos.ufv.br Apoio financeiro: FAPEMIG e CAPES

ABSTRACT

One hundred and ninety-seven isolates of Gram-negative bacteria, comprising 10 genera, were isolated from poultry carcasses at a processing plant in order to investigate resistance to low levels of antibiotics. The samples were taken just after evisceration and before inspection. Most of the isolates were of Samonella and Escherichia. Other genera present were Enterobacter, Serratia, Klebsiella, Kluyvera, Erwinia, Citrobacter, Pseudomonas and Aeromonas. Distinct profiles of antibiotic resistance were detected. Resistance to more than two antibiotics predominated and spanned several classes of antibiotics. Salmonellae and escherichiae were mainly resistant to the aminoglycosides, followed by tetracycline, nitrofuran, sulpha, macrolide, chloramphenicol, quinolones and b-lactams. Most isolates were sensitive to 30mg/ml olaquindox, the growth promoter in use at the time of sampling. However, many were resistant to a level of 10mg/ml and 13mg/ml olaquindox, levels present in the gut due to the dilution in the feed. The results suggest a possible role of low level administration of antibiotics to broilers in selecting multi-resistant bacteria in vivo.

Keywords: Poultry, antibiotic, multi-resistance, bacteria, growth promoter

RESUMO

Para verificar a resistência a baixos níveis de antibióticos foram obtidos e identificados 197 isolados de bactérias Gram negativas pertencentes a 10 gêneros. Os isolados foram coletados em abatedouro industrial, imediatamente após a evisceração e antes do serviço de inspeção. Salmonella e Escherichia foram os gêneros identificados com maior freqüência. Os demais gêneros foram Aeromonas, Enterobacter, Serratia, Klebsiella, Citrobacter, Erwinia, Kluyvera, Pseudomonas e Aeromonas. Diferentes perfis de resistência a antibióticos foram detectados. A resistência a mais de dois antibióticos foi freqüente na maioria dos isolados e incluiu diversas classes de antibióticos. As bactérias dos gêneros Salmonella e Escherichia apresentaram maior percentagem de isolados resistentes à classe dos aminoglicosídeos, seguida das classes de tetraciclina, nitrofurano, sulfa, macrolídeo, cloranfenicol, quinolona e â-lactâmico. A maioria dos isolados foi sensível a 30mg/ml de olaquindox, promotor de crescimento usado no período da coleta das amostras. Foram observados isolados com resistência ao olaquindox nas concentrações de 10mg/ml e 13mg/ml, níveis presentes no intestinos das aves devido à diluição da ração. Os resultados sugerem que os baixos níveis de antibióticos administrados aos frangos podem estar selecionando “in vivo” bactérias multirresistentes.

Palavras-chave: Frango, antibiótico, multirresistência, bactéria Gram-negativa, promotor de crescimento

INTRODUCTION

Since 1950, sub-therapeutic doses of antibiotics have been added to poultry, cattle, turkey and swine feed (Feinman, 1998; Miltenburg, 2000). These antibiotics are used in the therapeutics and the prevention of bacterial infection, and are also added continuously to animal feeds to promote growth, to increase feed efficacy and to decrease waste production (Madigan et al., 1999; Van Den Bogaard & Stobberingh, 2000).

The extensive use of antibiotics as therapeutic agents and as growth promoting substances is considered an important factor in the emergence of drug resistant bacteria of animal origin (Levy, 1978; Khachatourians, 1998; Meng et al., 1998). This has been associated with the rise of antibiotics resistance in Enterobacteriaceae, leading to the expansion of the pool of such resistance genes in nature (Levy, 1978). Clinical and microbiological evidences indicate that resistant bacteria can be transmitted from animals to humans, resulting in infections that are difficult to treat (Levy, 1987). Smith et al. (1998) suggested that the increase of quinolone-resistant Campylobacter jejuni infections in humans is largely due to the acquisition of resistant isolates from poultry. The transfer of resistant bacteria can be accomplished through the food chain (Meng et al., 1998). Levy et al. (1976) observed the transfer of tetracycline resistant E. coli isolates from chicken to chicken and from chicken to humans. Resistance genes are being selected for as a direct consequence of the indiscriminate use of antibiotics in animals, and may increase the incidence of multi-resistant human pathogens of animal origin (Young, 1994). The antibiotic resistance spectra of these bacteria are broad and include both growth promoter antibiotics and those of therapeutic use. Cross-resistance can be developed for antibiotics of the same class due to structural similarities (Feinman, 1998). Avoparcin is a glicopeptide antibiotic used only as a growth promoter, while vancomycin, belonging to the same group, is a valuable therapeutic antibiotic. Vancomycin resistant enterococci (VRE) presented cross-resistance with avoparcin, in Denmark and Germany (Feinman, 1998; Khachatourians, 1998). There is evidence that the use of avoparcin is forcing the selection of VRE that are acquired by humans through the food chain (Khachatourians, 1998; McCormick, 1998; Miltenburg, 2000). Cross-resistance for antibiotics of different classes can also occur due to multi-resistance plasmids or to multi-drug efflux pumps (Davis, 1994; George, 1996; Paulsen et al., 1996). Non-pathogenic bacteria may be reservoirs of resistance genes that can be transferred to pathogens and are a potential public health risk (Watanabe, 1963; Van Deen Bogaard & Stobberingh, 2000).

The objective of this work was to isolate and identify Gram-negative bacteria, especially members of the Enterobacteriaceae family, from wholesome poultry carcasses and to determine their antibiotic resistance profiles, including the antibiotics used for animal feed and those used in animal and human therapeutics.

MATERIALS AND METHODS

Twenty carcasses from each of three lots of Hubbarb broilers were collected in a processing plant immediately after evisceration and just before post-mortem inspection. The lot was defined as the yield of one morning production. Isolation of enterobacteria and other Gram-negative bacteria was performed according to Bridson (1990), and Vanderzant & Splittstoesser (1992). Pre-enrichment was performed in buffered glucose-brilliant green-bile-EE broth (Oxoid, Basingstone, Hampshire, England), for two hours. Two selective media were used, violet red bile glucose agar-VRB and Mac Conkey no 3 Agar (Oxoid, Basingstone, Hampshire, England).

Resistance to 17 antibiotics was tested by the Kirby-Bauer method (Bauer et al., 1966). The criterion to choose the antibiotics tested was based on their use as growth promoters and in human and animal therapeutics. Representatives of several chemical classes of antibiotics active against Gram-negative bacteria were also included. The antibiotics1 and their concentrations per disk were: spectinomycin (10mg), cefadroxil (30mg), cefaclor (30mg), nalidixic acid (30mg), sulphamethoxazole/trimethoprin (25mg), streptomycin (10mg), neomycin (10mg), apramycin (15mg), cephalexin (30mg), colistin sulphate (10mg), furazolidone (15mg), nitrofurantoin (50mg), spiramycin (100mg), norfloxacin (2mg), pefloxacin (5mg), tetracycline (5mg), and chloramphenicol (10mg). The first five antibiotics were at the Kirby-Bauer levels and the others were the lowest levels commercially available.

The disks were 8mm in diameter and an inhibition halo of 10mm diameter, or less, was arbitrarily considered positive for resistance. Larger measures were considered an indication of some sensitivity. This measure was arbitrarily chosen because of the unavailability of standards for the low levels and because of the objective of this work. Two antibiotics to which Gram-negative bacteria are resistant were used as negative controls, penicillin G (1u) and bacitracin (10u).

The isolates were identified by the Bacteria and Yeast Identification System (Biolog, 1993) (Biolog, Hayward, California, USA). Serological confirmation of the genus Salmonella and species E. coli were made by the slide agglutination method (Edwards & Ewing, 1972), with anti-Salmonella polyvalent flagella and somatic anti-sera and anti-E. coli A, B and C anti-sera (Probac do Brasil, São Paulo, Brazil), respectively.

The antimicrobial growth-promoter used by the broiler industry at the sampling time was olaquindox, a compound included in the quinoxaline group. Olaquindox was mixed with Mueller Hinton agar (Oxoid, Basingstone, Hampshire, England) in concentrations similar to those used in poultry feed, i.e. 30mg/ml in the growing phase and 40 mg/ml in the finishing phase, according to the manufacturer’s instructions. Because of the dilution during feeding, 1 part feed: 2 parts water (Castro, 1994), the corresponding concentrations in the gut should be 10mg/ml and 13mg/ml, respectively. These concentrations were also tested. Fifty cultures were inoculated on each plate, oriented by a numbered grid and the presence or absence of growth was scored.

RESULTS

One hundred ninety-seven isolates of Gram-negative bacteria were isolated from poultry carcasses and identified. Salmonella was present at a frequency of 54.3 % and Escherichia 27%. Other genera were less frequently isolated: Aeromonas (5.6%), Enterobacter (4.1%), Serratia (3.5%), Klebsiella (3.0%), Kluyvera (1.0%), Pseudomonas, Citrobacter and Erwinia 0.5% each. Among all the isolates, 9.1% were sensitive to all the antibiotics tested and 90.9% were resistant to at least one. All were sensitive to colistin sulphate, an antibiotic used a growth promoter and therapeutic agent in animals (Miltenburg, 2000).

Resistance to two antibiotics, or more, was considered multiple resistance. Most multi-resistant isolates were found within the Salmonella, Escherichia, Klebsiella and Aeromonas genera. The greatest proportion of isolates resistant to spiramycin, nalidixic acid, tetracycline, chloramphenicol and sulphamethoxazole/trimethoprin was found among the Enterobacter species, which were all resistant to spiramycin, used as a growth promoter and therapeutic drug for animals (Miltenburg, 2000). Some of the Enterobacter isolates were resistant to up to six different antibiotics. Resistance to spiramycin was prevalent in 87.7% of the Serratia isolates. Some of the Serratia were resistant only to spiramycin.

Spectinomycin and nalidixic acid resistance was more frequently found among the isolates of Aeromonas. Spectinomycin, an aminoglycoside, is also an antibiotic used as a growth promoter and nalidixic acid, a quinolone antibiotic, is used as a therapeutic drug and growth promoter (Khachatourians, 1998).

Among the isolates of Salmonella and Escherichia, resistance profiles were very different, demonstrating their diversity, including up to 10 antibiotics for Salmonella (Tab. 1) and eight for Escherichia (Tab.1 and 2 ). Four Salmonella isolates showed resistance to the three quinolones tested: nalidixic acid, pefloxacin and norfloxacin.

Resistance to nalidixic acid was found mainly among multiresistant isolates of Salmonella and Escherichia. This was also true for resistance to tetracycline, sulphamethoxazole/trimethoprin and streptomycin, while fewer isolates were resistant to neomycin.

Resistance to aminoglycosides was the most prevalent among the isolates, while resistance to b-lactams was the least prevalent (Tab. 3). Among the multi-resistant isolates listed in Tab. 1 and 2 , some are resistant to at least three of the four aminoglicosydes tested. Few of the isolates of Salmonella were resistant to the first generation b-lactams, cefadroxil and cephalexin, and to cefaclor of the second generation b-lactams. Only two isolates of the multi-resistant escherichiae were resistant to the b-lactams tested (data not shown).

Among the 197 isolates, only 2.7% were resistant to 30mg/ml as well as to 40mg/ml olaquindox. The frequency of resistance increased to 57.7% and 27.9% when 10mg/ml and 13mg/ml were used, respectively. Of the salmonellae and escherichiae that were resistant to five antibiotics or more, 72% and 30%, respectively, were resistant to olaquindox.

DISCUSSION

Bacteria of Enterobacteriaceae are often antibiotic targets (Stecchini et al., 1992) and Salmonella is the main target in poultry. The different resistance profiles found for Salmonella and Escherichia comprehend drugs that are used also as growth promoting agents (Miltenburg, 2000; Chopra & Roberts, 2001). It has been found that the emerging resistance of E. coli O157: H7 to chloramphenicol, streptomycin, tetracycline, trimethoprin-sulphamethoxazole and other drugs could be related to the increased prevalence of this pathogen in animals that are fed antibiotics (Kim et al., 1994). There is also direct evidence that the use of antimicrobial drugs in animal feed selects for resistant serotypes of the nontyphoid salmonellae which can be transmitted to humans through the food chain or by direct contact (Feinman, 1998). It has been determined that the increase in isolation of quinolone resistant Campylobacter jejuni from humans is actually due to contamination with isolates from poultry that were fed with fluoroquinolones (Smith et al., 1998). The use of these drugs as growth promoter in the USA has possibly led to the development of a quinolone resistant C. jejuni reservoir in poultry (Smith et al., 1998). Isolates of Salmonella resistant to the three quinolones tested were found. This is important because these drugs are used in human and animal therapy (Barros et al., 1993; Bootle, 2000). Based on their resistance profiles, these are different isolates. Two of them were also resistant to olaquindox. Nalidixic acid is used for treating urinary tract diseases in humans (Uptodate, 2001). In animals, the fluoroquinolones are often used to treat infections by Gram-negative bacteria (Bootle, 2000).Salmonella isolated from animals are often resistant to nalidixic acid and have a decreased susceptibility to fluoroquinolones (Feinman, 1998). Many previously reported nalidixic acid resistant salmonellae appear to have emerged from turkey flocks that were routinely treated with a closely related antibiotic, enrofloxacin, or from poultry products from processors that obtain animals from these farms (Feinman, 1998).

A high percentage of isolates resistant to antibiotics within the classes most often used in Brazil as growth promoters was found. Such is the case for neomycin, spectinomycin and apramycin (aminoglycosides) and furazolidone and nitrofurantoin (nitrofuran). However, drugs such as penicillin (a b-lactam), chloramphenicol, tetracycline and streptomycin (an aminoglycoside) were used in animal feed in the past, but are now prohibited in most countries, including Brazil(Motta, 1996). These drugs are widely used in human and veterinary medicine.

Some of the multi-resistant isolates were resistant to at least three of the four aminoglycosides tested. Streptomycin, neomicyn and apramycin are used in animal feed and in veterinary and human therapy (Barros et al., 1993; Feinman, 1998; Miltenburg, 2000). In humans, streptomycin is used to treat infections by Gram-negative and Gram-positive bacteria (Barros et al., 1993).

The screening of olaquindox resistant isolates was performed because this was the growth promoter in use in the feed of the poultry supplied to the slaughter house at the time of sampling. Several isolates were resistant to 10mg/ml of olaquindox. As this is the concentration present in the gut, the growth promoter could be exerting selective pressure upon the gut microflora.

It was not attempted to prove that antibiotic resistance is due to the use of growth promoting substances. The aim was mainly to create a collection of isolates in which a diversity of resistance mechanisms might be detected. However, the results showed the presence of multi-resistant Gram-negative bacteria in poultry carcasses including isolates resistant to antibiotics currently used in poultry feed in Brazil. We are currently investigating the resistance mechanisms underlying multi-resistance in some of those isolates.

  • BARROS, E.; BITTENCOURT, H.; BRUNSTEIN, C. et al. Anti microbianos: consulta rápida. Porto Alegre: Artes Médicas, 1993. 163p.
  • BAUER, A.W.; KIRBY, W.M.M.; SHERRIS, J.C. et al. Antibiotics susceptibility testing by a standardized single disk method. Am. J. Clin. Pathol., v.45, p.493-496, 1966.
  • BIOLOG. MICROSTATIONTM SYSTEM RELEASE 3.50. Hayward, 1993.
  • BOOTLE, D.M. Fluorinated quinolones: use and misuses. In: BONAGURA, J.D. (Ed.). Kirks current veterinary therapy: XIII small animal practice. Philadelphia: Saunders, 2000. p.41-48.
  • BRIDSON, E.Y. The oxoid manual. 6.ed. Hampshire: Unipath, 1990.
  • CASTRO, A.G.M. Qualidade da água: manejo de frango. Campinas: Fundaçăo Apinco de Cięncias e Tecnologia Avícolas, 1994. 243p.
  • CHOPRA, I.; ROBERTS, M. Tetracycline antibiotic: mode of action, applications, molecular biology and epidemiology of bacterial resistance. Microbiol. Mol. Biol. Rev., v.65, p.232-260, 2001.
  • DAVIS, J. Inactivation of antibiotics and the dissemination of resistance genes. Science, v.264, p.375-381, 1994.
  • EDWARDS, P.R.; EWING, W.H. Identification of Enterobacteriaceae Minneapolis: Minnesota: Burgess, 1972.
  • FEINMAN, S.E. Antibiotics in animal feed-drug resistance revisited. Am. Soc. Microbiol. News, v.64, p.24-30, 1998.
  • GEORGE, A.M. Multidrug resistance in enteric and other Gram negative bacteria. FEMS Microbiol Lett., v.139, p.1-10, 1996.
  • KHACHATOURIANS, G.G. Agricultural use of antibiotics and the evolution and transfer of antibiotic-resistant bacteria. Can. Med. Assoc. J., v.159, p.1129-1136, 1998.
  • KIM, H.H.; SAMADPOUR, M.; GRIMM, L. et al. Characteristics of antibiotics-resistent Escherichia coli O157: H7 in Washington State, 1984-1991. J. Infect. Dis., v.6, p.1606-1609, 1994.
  • LEVY, S.B. Antibiotics use for growth promoter in animals: ecological and public health consequences. J. Food Protec., v.50, p.616-620, 1987.
  • LEVY, S.B. Emergence of antibiotic-resistant bacteria in the intestinal flora of farm inhabitants. J. Infec. Dis., v.135, p.688-690, 1978.
  • LEVY, S.B.; FITZGERALD, G.B.; MACONE, A.B. Spread of antibiotic-resistant plasmids from chicken to chicken and from chicken to man. Nature, v.260, p.40-42, 1976.
  • MADIGAN, M.T.; MARTINKO, J.M.; PARKER, J. Brocks biology of microorganisms. 9.ed. Englewood Cliffs: Prentice Hall, 1999. 1046p.
  • McCORMICK, J.B. Epidemiology of emerging/re-emerging antimicrobial-resistant bacterial pathogens. Curr. Opin. Microbiol., v.1, p.125-129, 1998.
  • MENG, J.; ZHAO, S.; DOYLE. M.P. et al. Antibiotic resistance of Escherichia coli O157:H7 and O157:NM isolated from animals, food, and humans. J. Food Prot., v.61, p.1511-1514, 1998.
  • MILTENBURG, G. Promotores e aditivos de crescimento em avicultura. In: CONFERĘNCIA APINCO DE CIĘNCIA E TECNOLOGIA AVÍCOLA, 2000, Campinas. Anais... Campinas: FACTA, 2000. p.206-215.
  • MOTTA, E.G. Restriçőes e uso de aditivos (promotores de crescimento) em raçőes de aves: revisăo do ministério. In: CONFERĘNCIA APINCO DE CIĘNCIA E TECNOLOGIA AVÍCOLA, 1996, Curitiba. Anais... Campinas: FACTA, 1996. p.51-57.
  • PAULSEN, I.T.; BROWN, M.H.; SKURRAY, R.A. Proton-dependent multidrug efflux systems. Microbiol. Rev., v.60, p.575-608, 1996.
  • SMITH, K.E.; BESSER, J.M.; HEDBERG, C.W. et al. Quinolona-resistant Campylobacter jejuni infections in Minnesota, 1992-1998. N. Engl. J. Med., v.340, p.1525-1532, 1998.
  • STECCHINI, M.L.; MANZANO, M.; SARAIS, I. Antibiotic and disinfectant susceptibility in Enterobacteriaceae isolated from minced meat. Int. J. Food Microbiol., v.16, p.79-85, 1992.
  • UPTODATE clinical reference library: release 9.1 windows. Wellesley: UPTODATE, 2001. (2 CD-ROM).
  • VAN DEN BOGAARD, A.E.; STOBBERINGH, E.E. Epidemiology of resistance to antibiotics links between animals and humans. Int. J. Antimicrob. Agents, v.14, p.327-335, 2000.
  • VANDERZANT, C.; SPLITTSTOESSER, D.F. (Eds.). Compendium of methods for the microbiological examination of foods. 3.ed. Washington: APHA, 1992. 1219p.
  • WATANABE, T. Infective heredity of multiple drug resistance in bacteria. Bacteriol. Rev., v.27, p.87-109, 1963.
  • YOUNG, H.K. Do nonclinical uses of antibiotics make a difference? Infect Control. Hosp. Epidemiol., v.15, p.484-487, 1994.

Publication Dates

  • Publication in this collection
    22 July 2002
  • Date of issue
    Feb 2002

History

  • Received
    24 Sept 2001
Universidade Federal de Minas Gerais, Escola de Veterinária Caixa Postal 567, 30123-970 Belo Horizonte MG - Brazil, Tel.: (55 31) 3409-2041, Tel.: (55 31) 3409-2042 - Belo Horizonte - MG - Brazil
E-mail: abmvz.artigo@abmvz.org.br