Antibiotic usage is possibly the most important factor that promotes the emergence, selection and dissemination of antibiotic-resistant microorganisms in both veterinary and human medicine [1, 2]. This acquired resistance occurs not only in pathogenic bacteria but also in the endogenous flora of exposed individuals (animals and humans) or populations [3–6]. In intensively reared food animals, antibiotics may be administered to whole flocks rather than individual animals, and antimicrobial agents may be continuously fed to food animals such as broilers and turkeys as antimicrobial growth promoters. Therefore the antibiotic selection pressure for resistance in bacteria in poultry is high and consequently their faecal flora contains a relatively high proportion of resistant bacteria [7].

In Jamaica, chicken is considered to be the most widely produced and consumed meat protein, five times more than beef and 10 times more than pork. Over the period 1998 – 2002, Jamaica's local poultry production yielded approximately 380 M kg with a monetary value of nearly $90 billion [8]. Chicken farms are widely distributed throughout the island with the majority being located on the south coast and large commercial operations accounting for 65–70% of total broiler production [8]. Thus, the high consumption of chicken meat warrants great care in safeguarding the industry against threatening factors.

At slaughter, resistant strains from the gut may contaminate poultry carcasses and as a result poultry meats are often associated with multiresistant E. coli [9–11]; likewise eggs become contaminated during laying [12]. Hence, antimicrobial resistant faecal E. coli from poultry can infect humans both directly and via food. Although rare, these resistant bacteria may colonize the human intestinal tract and may also contribute resistance genes to human endogenous flora.

However, the mechanism of spread of antibiotic resistance from food animals to humans remains controversial. Colonization of the intestinal tract with resistant E. coli from chicken has been shown in human volunteers [13] and there is historical evidence that animals are a reservoir for E. coli found in humans [14]. Further, spread of antibiotic resistance plasmids in E. coli from chickens to human handlers [15, 16] or of antibiotic-resistant microorganisms from poultry to humans in various countries [13, 17–19] has been reported. Resistance has been found in organisms common to both humans and animals, such as Salmonella spp., Campylobacter spp. and enterococci among others [20]. Due to the intricate balance of microflora of different habitats within the ecosystem, the transfer of resistance genes among bacteria occupying different habitats has the potential to occur frequently. Resistance genes may be transferred vertically among bacteria of different genera and families [21] or horizontally among different bacterial species within the same genus or family [22, 23].

Molecular tools have been used to correlate animal associated pathogens with similar pathogens affecting humans and to clearly demonstrate transferable resistant genes carried by plasmids common to both animals and humans [2, 21, 24, 25]. The possibility for transfer of antibiotic resistance genes among humans, animals and the environment [22] is a direct threat to public health. This threat prompts ongoing research into emerging resistance mechanisms, novel approaches to antimicrobial efficacy and stringent control measures in the prudent use of antimicrobials in animal and human medicine. As the possibility of transfer of tetracycline-resistant bacteria from animals to humans is controversial, tetracycline-resistant E. coli isolates from poultry and humans sources were analysed for plasmid diversity and the presence of specific genes associated with tetracycline resistance.

The aim of this study was to investigate the prevalence and genetic basis of tetracycline resistance in commensal E. coli strains isolated from healthy broiler chickens and compare these data with isolates obtained from hospitalized patients.

In this study we examined antimicrobial resistance in commensal E. coli isolates from broiler chickens and compared them to clinical isolates from hospitalized patients. Because of the geographic sampling of the avian isolates, this surveillance provides a representative sample of the resistance trends in the Jamaican poultry industry.

Compared with the other antimicrobial agents used in this study, avian isolates were susceptible to gentamicin, amoxicillin/clavulanate and chloramphenicol. However, increasing resistance frequencies were noted for the fluoroquinolones, in the apparent absence of selection pressure.

There is strong evidence that the use of antimicrobial agents can lead to the emergence and dissemination of resistant E. coli [13, 15, 16, 26–28], which can then be passed onto people via food or through direct contact with animals. However, there are increasing numbers of reports detailing circulation and amplification of antimicrobial resistance genes, including tetracycline resistance in the environment [22, 30, 29, 31, 32], which could facilitate the emergence and spread of antibiotic resistance in bacteria. The finding of multiple drug resistant commensal E. coli isolates from broiler chickens from farms that did not report antibiotic use concurs with these latter reports. Rysz and Alvarez [31] demonstrated that bacteria in the soil could acquire resistance to tetracycline from environmental exposure, possibly creating a reservoir of resistance factors generated outside host animals. Their finding suggests that collection of environmental samples from a farm is important for accessing exposure to resistance factors from farm runoff in watersheds, except in cases where soil is augmented with organic fertilizer. Sayah et al. [32] reported that farm environmental isolates showed reduced susceptibility (as measured by disc diffusion zone sizes) compared to faecal sample isolates to most agents studied. They suggested that non-sampled sources, e.g., farm workers and wildlife with access to the farm environment, could be sources of resistance factors. We concur with this as a possible source of resistance determinants in our study.

Antimicrobial resistance to tetracycline, kanamycin and nalidixic acid was noted among avian E. coli isolates (Table 1). The presence and frequency of tetracycline resistance in E. coli from chickens agree with findings of other studies on antibiotic resistance in E. coli [16, 32]. The patterns of resistance to tetracycline have been attributed in part to widespread and lengthy use of tetracycline in the poultry industry [3, 4, 6, 15, 16]. Since tetracycline is a naturally derived compound, bacteria can be exposed to these agents in nature and outside any human use for disease treatment, for prophylaxis, or for livestock growth promotion. Tetracycline is a commonly used first line antibiotic for many domestic animals and is often used before the antibiotic resistance profile of a pathogen has been determined. Resistance to tetracycline is plasmid mediated, with a wide variety of genetic determinants [33]. This makes it more possible for a susceptible bacterium to acquire resistance factors by conjugation, or by transformation, as was demonstrated in this study.

Low levels of resistance were observed for ofloxacin, ciprofloxacin and gentamicin. The observation that over 85% of avian E. coli isolates were resistant to nalidixic acid (concomitant with reduced susceptibilities to the fluoroquinolones) is important considering that the fluoroquinolones are used to treat a range of E. coli infections in humans [34]. This finding concurs with previous reports [16, 35], and underscores the need to monitor quinolone- and fluoroquinolone-resistant bacteria in poultry production as their emergence is an important health concern among most in the food safety community. This is against the background that nalidixic acid selects for low-level resistance to ciprofloxacin (and possibly other fluoroquinolones) [36]. While quinolone resistance involves chromosomal mutations that reduce membrane permeability and decrease drug accumulation or alter DNA topoisomerases, resistance to fluoroquinolones is most associated with mutations in DNA gyrase [36, 37].

Aminoglycoside resistance in E. coli most often occurs by aminoglycoside-modifying enzymes [38, 39] encoded on transmissible plasmids [40]. It is not surprising that more isolates were resistant to kanamycin when compared to gentamicin, as kanamycin is susceptible to the largest number of enzymes. Conversely, resistance to gentamicin is mediated by modifications at few sites on the molecule.

Multi-drug resistance (to at least three antimicrobials) was found in E. coli from both avian and human sources (Table 2; Fig. 1), but was higher in frequency and proportion in avian isolates. However, when these multi-drug resistant organisms were compared, it is clear that they do not have common sources of resistant bacteria. Most avian multi-drug resistant isolates exhibited resistance to a combination of antimicrobials that included kanamycin, tetracycline and nalidixic acid, while most human isolates were resistant to kanamycin and tetracycline or kanamycin and gentamicin. This suggests that avian isolates that were Tet-R are more likely to become resistant to additional antimicrobial agents. Resistance to tetracycline may be conserved in bacterial populations over time, regardless of selection pressure, which might result in an overall increase in resistance over time. In addition, it is likely that avian sources could act as reservoirs for Tet-R for environmental contamination and human colonization. Further, the difference in resistance between the two groups of isolates can be explained in terms of the interactions of the organisms (associated with the host) and potential horizontal gene transfer in their respective environments.

To deal with multi-drug resistant organisms, it is usually recommended that potentially synergistic antimicrobial combinations be used. We evaluated our isolates for cross-resistance between tetracycline and other agents. Isolates resistant to other antimicrobial agents, except ampicillin, were also resistant to tetracycline. Tetracycline-resistant strains from avian sources were susceptible to ampicillin, amoxicillin/clavulanate, chloramphenicol, ciprofloxacin and gentamicin, but resistant to kanamycin and nalidixic acid. This indicates limited cross-resistance and suggests that potential therapeutic options still exist for poultry colonized by Tet-R E. coli. On the other hand, a greater percentage of the Tet-R strains from human sources were resistant to other antibiotics, although 57 – 95% was susceptible to amoxicillin/clavulanate, gentamicin and the quinolones. Among these isolates, there was significant cross-resistance to the aminoglycosides, ampicillin, and the quinolones.

Our results indicate that the Tet-R plasmids carried by avian or human E. coli isolates demonstrate the distribution of similar resistance determinants in diverse genetic backgrounds. This probably indicates a low level of exposure as animal hosts with a continuous exposure to tetracycline have a higher percentage of Tet-R E. coli, and these isolates carry a greater diversity of resistance genes.

In gram negative organisms, including E. coli, tetracycline resistance is frequently regulated by efflux genes that are normally associated with large plasmids which are mostly conjugative. These plasmids often carry other antibiotic resistance genes, heavy metal resistance genes and/or other pathogenic factors such as toxins, hence the selection for any of these factors selects for the plasmid. The molecular investigations into the underlying tetracycline resistance mechanisms revealed that 69% of the tetracycline-resistant isolates possessed resistance plasmids based on successful transformations of a susceptible strain. It is likely that the remainder of isolates were using resistance mechanisms based on the chromosome or on integrons carried on plasmids [41, 42]. A mechanism for tetracycline resistance was investigated using universal primer pairs based on sequences belonging to the efflux classes, tetB and tetD [30]. The resulting products of approximately 1 kb would strongly suggest that tetracycline resistance among the E. coli isolates in this study is regulated mainly by genes encoding an efflux pump mechanism. These Tet-R strains contained either tetB or tetD efflux genes, which together with the absence of polymorphism in the plasmid RFLP data, suggest that there is a similarity of tetracycline resistance mechanism in these organisms. Since many human and poultry commensal bacteria carry the same tet genes, plasmids, transposons and integrons as disease-causing species [43], our results suggest that these agents could be transferred to disease-causing bacteria.

In conclusion, the results of this study provide evidence for significant antimicrobial resistance of E. coli isolates from broiler chickens raised on farms without recorded antimicrobial use, and hospitalized patients in Jamaica. We noted that tetracycline resistance was generally associated with plasmids, and the resistance was mediated by tetB or tetD efflux genes. The results suggest that tetracycline resistance mechanisms from avian and human sources may be similar. Long-term prospective studies examining isolates from defined geographic locations are required to more accurately detect temporal and spatial differences in antimicrobial resistance in strains of E. coli.