Antimicrobial resistant zoonotic pathogens present on food constitute a direct risk to public health. Antimicrobial resistance genes in commensal or pathogenic strains form an indirect risk to public health, as they increase the gene pool from which pathogenic bacteria can pick up resistance traits. Food can be contaminated with antimicrobial resistant bacteria and/or antimicrobial resistance genes in several ways. A first way is the presence of antibiotic resistant bacteria on food selected by the use of antibiotics during agricultural production. A second route is the possible presence of resistance genes in bacteria that are intentionally added during the processing of food (starter cultures, probiotics, bioconserving microorganisms and bacteriophages). A last way is through cross-contamination with antimicrobial resistant bacteria during food processing. Raw food products can be consumed without having undergone prior processing or preservation and therefore hold a substantial risk for transfer of antimicrobial resistance to humans, as the eventually present resistant bacteria are not killed. As a consequence, transfer of antimicrobial resistance genes between bacteria after ingestion by humans may occur. Under minimal processing or preservation treatment conditions, sublethally damaged or stressed cells can be maintained in the food, inducing antimicrobial resistance build-up and enhancing the risk of resistance transfer. Food processes that kill bacteria in food products, decrease the risk of transmission of antimicrobial resistance.

Cattle are considered to be the main reservoir for Shiga toxin-producing Escherichia coli (STEC) and are often the direct or indirect source of STEC outbreaks in humans. Accurate measurement of the concentration of shed STEC in cattle feces could be a key answer to questions concerning transmission of STEC, contamination sources and efficiency of treatments at farm level. Infected animals can be identified and the contamination level quantified by real-time quantitative PCR (qPCR), which has its specific limitations. Droplet digital PCR (ddPCR) has been proposed as a method to overcome many of the drawbacks of qPCR. This end-point amplification PCR is capable of absolute quantification independent from any reference material and is less prone to PCR inhibition than qPCR. In this study, the qPCR-based protocol described by Verstraete et al. (2014) for Shiga toxin genes stx1 and stx2 and the intimin gene eae quantification was optimized for ddPCR analysis. The properties of ddPCR and qPCR using two different mastermixes (EMM: TaqMan^® Environmental Master Mix 2.0; UMM: TaqMan^® Universal PCR Master Mix) were evaluated, using standard curves and both artificial and natural contaminated cattle fecal samples. In addition, the susceptibility of these assays to PCR-inhibitors was investigated. Evaluation of the standard curves and both artificial and natural contaminated cattle fecal samples suggested a very good agreement between qPCR using EMM and ddPCR. Furthermore, similar sensitivities and no PCR inhibition were recorded for both assays. On the other hand, qPCR using UMM was clearly prone to PCR inhibition. In conclusion, the ddPCR technique shows potential for the accurate absolute quantification of STEC on the farms, without relying on standardized reference material.

To select Lactobacillus strains from laying hens for potential use as probiotic to control Salmonella Enteritidis infection. One hundred and eighty-six lactobacilli were isolated from the cloaca and vagina of laying hens, and identified at the species level by a polyphasic taxonomic approach. All isolates belonged to the Lactobacillus acidophilus, Lactobacillus reuteri or Lactobacillus salivarius phylogenetic groups, with the L. reuteri group being the most predominant group. Based on genetic diversity, about 50 representative strains were selected and tested for in vitro properties that could be predictive for probiotic activity in laying hens. Salmonella inhibition was shown to be species dependent, and correlated to some extent with the production of lactic acid. A selection of strains was evaluated in a S. Enteritidis challenge experiment. Two strains, L. reuteri R-17485 and Lactobacillus johnsonii R-17504 significantly decreased the colonization of chicks by S. Enteritidis in caeca, liver and spleen. Lactobacilli isolated from laying hens were observed to inhibit Salmonella growth in vitro, most probably through production of lactic acid, and to decrease in vivo the S. Enteritidis colonization of chicks. The data demonstrate that Lactobacillus isolates from laying hens may have probiotic potential in reducing S. Enteritidis infection.

Farm disinfectants are widely used in primary production, but questions have been raised if their use can select for antimicrobial resistance. The present study examined the use of disinfectants in poultry and pig husbandry and its contribution to the antibiotic and disinfectant susceptibility of Escherichia coli (E. coli) strains obtained after cleaning and disinfection. On those field isolates antibiotic susceptibility was monitored and susceptibility to commonly used active components of farm disinfectants (i.e. glutaraldehyde, benzalkoniumchloride, formaldehyde, and a formulation of peracetic acid and hydrogen peroxide) was tested. This study showed a high resistance prevalence (> 50%) for ampicillin, sulfamethoxazole, trimethoprim and tetracycline for both production animal categories, while for ciprofloxacin only a high resistance prevalence was found in broiler houses. Disinfectant susceptibility results were homogenously distributed within a very small concentration range. Furthermore, all E. coli strains were susceptible to in-use concentrations of formaldehyde, benzalkoniumchloride and a formulation of peracetic acid and hydrogen peroxide, indicating that the practical use of disinfectants did not select for disinfectant resistance. Moreover, the results showed no indications for the selection of antibiotic resistant bacteria through the use of disinfectants in agricultural environments. Our study suggests that the proper use of disinfectants in agricultural environments does not promote antibiotic resistance nor reduce E. coli disinfectant susceptibility.