key: cord-1029829-pk9b8u61 authors: Bergšpica, I.; Kaprou, G.; Alexa, E. A.; Prieto‐Maradona, M.; Alvarez‐Ordóñez, A. title: Identification of risk factors and hotspots of antibiotic resistance along the food chain using next‐generation sequencing date: 2020-11-27 journal: EFSA J DOI: 10.2903/j.efsa.2020.e181107 sha: 7197f8a16a942bf84eec3545b2d6d7d68cf16943 doc_id: 1029829 cord_uid: pk9b8u61 Bacterial antimicrobial resistance (AMR) is considered to be very alarming following an upward trend and thus posing a primary threat to public health. AMR has tremendous adverse effects on humans, farm animals, healthcare, the environment, agriculture and, thus, on national economies. Several tools have been proposed and adopted by numerous countries after comprehending the need for antimicrobial stewardship and for a rational use of antibiotics. These tools include diagnostics for infections or AMR detection, for measuring and monitoring antibiotic consumption (e.g. surveillance tools) and for guiding medical doctors and veterinarians in selecting suitable antibiotics. In addition, it has been known that the food chain represents a leading vector for the transmission of pathogens to humans via various routes (direct or indirect). Considerable efforts have been made and are still in progress both at international and national levels in order to control and mitigate the spread of pathogens and thus ensure food safety. During the last decades, a new concern has risen regarding the food chain playing a potential major role in the transmission of resistant bacteria as well as resistance genes from the animal kingdom to humans. Several recent studies highlight the role of food processing environments as potential AMR hotspots contributing to this spread phenomenon. Next‐generation sequencing (NGS) technologies are becoming broadly used in the AMR field, since they allow the surveillance of resistant microorganisms, AMR determinants and mobile genetic elements. Moreover, NGS is capable of providing information on the mechanisms driving and spreading AMR throughout the food chain. In the current work programme, the aim was to acquire knowledge and skills to track AMR genes and mobile genetic elements in the food chain through NGS methodologies in order to implement a quantitative risk assessment and identify hotspots and routes of transmission of AMR along the food chain. According to the World Health Organization (WHO), antimicrobial agents are indispensable drugs for the protection of human and animal health. Yet, the persisting emergence and spread of pathogenic microorganisms that are resistant to antimicrobial agents constitute a growing global concern (WHO, xxxx). Antimicrobial resistance (AMR) is one of the major health issues of the 21st century. A major factor leading to the spread of AMR is the use of antimicrobial drugs in veterinary medicine and human medicine, where misuse and overuse of antibiotics have been detected. However, another hotspot for AMR can be the food production environment. Nevertheless, the relative drivers and the potential sources of AMR bacteria in the food chain are not well established yet. Overprescription of antimicrobial agents by clinicians is considered one the main sources of selection for AMR, and some antimicrobials are used in larger quantities in food production. These resistant bacteria can afterwards reach humans via the environment, food or other means (Hudson et al., 2017) . According to the Center for Disease Control and Prevention (CDC), antibiotic-resistant bacteria can spread through the food chain in various ways. For instance, in the process of animal slaughtering and meat processing, contamination of meat (or other animal products) as well as processing equipment can occur by resistant bacteria. In addition, animal waste can be a source of resistant bacteria which can contaminate the surrounding environment. Furthermore, vegetables and fruits can get contaminated while being in contact with water, soil or fertilisers containing animal waste contaminated with resistant bacteria (Centers for Disease Control and Prevention, online). In other words, food and food processing environments (FPEs) could be reservoirs and vehicles of transmission of AMR to humans causing a major public health impact. In order to bring antibiotic-resistant bacteria under control several actions could be implemented, such as in the fields of i) education and training, ii) surveillance, monitoring and record-keeping, iii) legislation and regulations, or iv) optimisation and reduction of irrational antibiotic use (Canic ßa et al., 2019). Among the surveillance and monitoring tools, omics technologies (e.g. genomics, metagenomics and transcriptomics), serve the purpose of monitoring and controlling AMR in various One Health settings, mainly in respect to the selection and distribution of AMR in food-related settings. In addition, omics could facilitate the unravelling of the associated AMR mechanisms (Canic ßa et al., 2019). Major improvements have been made in sequencing technologies during the last decade. These technologies are referred to as next-generation (or high-throughput) sequencing (NGS). NGS has considerably revolutionised the analysis not only of bacterial genomes but also of complex bacterial communities (i.e. intestinal or environmental microbiota). Regarding the AMR field, NGS facilitates both the identification of already known resistance genes and it also enables the prediction of novel ones. NGS has advanced the capacity of tracking bacterial clones (e.g. multidrug-resistant clones), and the identification of new antibiotic resistance genes (ARGs) as well as their genetic carriers (i.e. plasmids). Therefore, NGS paves the way for new perspectives in risk assessment and AMR surveillance while simultaneously it enables the comprehension of the AMR dynamics between pathogens and commensals stemming from various sources (e.g. environment, animals and humans) (Rupp e et al., 2019). The main advantages offered by NGS are: i) it facilitates clinical decision-making by providing various levels of information to guide the treatment with the appropriate antimicrobial; ii) it improves outbreak investigation and guides the interventions to control them; iii) it allows a retroactive analysis once new information appears (storing NGS data for future analysis); iv) it has the potential to link various fields such as clinics, food, environment and animals; v) it provides mechanistic information on the resistance. Regarding these latter advantages, unlike phenotypic testsproviding information only related to resistance/susceptibility to antimicrobials -NGS can shed light on the molecular basis for the resistance. Moreover, NGS has the potential to characterise novel resistance mechanisms once they arise (via the sequencing of isolates proved to be phenotypically resistant). This is an outstanding added value compared to conventional nucleic-acid based techniques (e.g. polymerase chain reaction, PCR) (Berendonk et al., 2017) . Description of work programme The aim of the work programme was to acquaint the fellows in the execution of assessments to identify risk factors present in the food chain that could allow occurrence and spread of antibioticresistant microorganisms, AMR genes and mobile genetic elements in FPEs. The main focus was to investigate the potential of NGS methodologies as a tool for the identification and characterisation of resistant bacteria and their AMR gene repertoire. The outcomes of the work programme may assist the design of knowledge-based interventions facilitating the reduction of the dispersal of multidrugresistant microorganisms in the food industry. The objectives of the work programme are listed below: Objective 1: Training of the fellows on the risk assessment methodologies routinely used by the mentor and other collaborators at the host institution. The research fellows gained in-depth training in specific tools commonly employed by laboratory members of the supervisor and collaborators. The fellows were familiarised with the use of several software tools such as: Monte Carlo simulation distribution software (@Risk, Crystal Ball), software programs for quantitative microbiology and risk assessment models (Combase, FSSP) and last but not least risk ranking tools (RiskRanger). Moreover, they received training in the use of the SPSS software program for processing the acquired data from the performed experiments in order to identify any correlations among the various parameters being analysed. In addition, several simulations were ran regarding various case studies employing the @Risk software allowing the fellows to obtain a more clear view and acquire hands-on experience on Monte Carlo simulations and the importance of distribution functions in the risk assessment process. Objective 2: Execution of a qualitative risk assessment approach on AMR in the food chain. A qualitative assessment of the risk posed by extended-spectrum b-lactamase (ESBL)-producing Enterobacteriaceae was undertaken, focused on hazard identification and exposure assessment. Exposure assessment modelling was carried out regarding the presence of ESBL-producing E. coli in pork meat products within the European Union (EU). A systematic literature review was conducted where the Scopus database was used for the article search. The extracted data were latter used for building an exposure assessment model. Software (@Risk and Crystal Ball) were used for the simulation part. Apart from the literature review conducted related to the prevalence of ESBLproducing E. coli on pork meat, the data from EFSA Comprehensive European Food Consumption Database (EFSA, 2020) were also taken into account in order to perform the exposure assessment. Objective 3: Phenotypic and genotypic characterisation of a collection of isolates from FPEs. The fellows characterised a wide collection of isolates from FPEs with classical microbiology tools and NGS (Figure 1 ). These included culture-based methods -pure culture isolation, determination of resistance to antibiotics (growth in media supplemented with antibiotics (chromogenic media)), and molecular-based methods (PCR). In more details, the two fellows isolated pure cultures of Enterobacteriaceae, Enterococcus spp. and Staphylococcus spp. from samples obtained from FPEs stemming from 30 industries from different food sectors. Over 500 pure isolates were screened on chromogenic media. Pure isolates were verified by matrix-assisted laser desorption/ionisation time-offlight mass spectrometry (MALDI-TOF). Pure isolates were obtained and preserved. DNA isolation was performed for all the isolates which exhibited resistance in the chromogenic media. PCR was performed for Enterobacteriaceae isolates to determine presence of bla CTX-M and bla SHV genes. Regarding the Enterococcus faecalis and Enterococcus faecium, the presence of the vanA and vanB genes was examined. 15 E. coli and 45 E. faecalis and E. faecium isolates were screened against a wide array of antimicrobials following the broth microdilution method. The most interesting isolates were analysed through whole genome sequencing (WGS), specifically using the Illumina HiSeq system. The acquired whole genome sequences were analysed by bioinformatic tools. In Figure 1 , the flow chart followed for the experimental part is depicted. Objective 4: Integration of metadata provided in Objective 3 into risk assessment models. Main hazards were identified along the food production chain in different food industries. Analyses of a questionnaire on hygiene and sanitation practices filled by food producers were performed. Correlations between different variables were assessed, binary logistic regression, as well as some statistical analyses were performed. Regression analyses were conducted in order to investigate any possible correlations between the industry status in relation to the occurrence of Enterobacteriaceae and Enterococcus spp., respectively and the metadata on the food safety practices employed by each industry. The FSSP software was used to perform the analyses. Objective 5: Communication, dissemination and outreach activities. The fellows participated in dissemination and outreach activities. Two review articles with the following titles 'A review on Extended Spectrum b-Lactamase (ESBL) producing Escherichia coli in pigs and pork meat in the European Union' and 'Rapid methods for antimicrobial resistance diagnostics' will be sent for publication during September in peer reviewed journals. The former review covers the analysis of the prevalence of ESBL-producing E. coli in pork meat and pigs in the EU, and dissemination pathways of these bacteria along the pork production chain. The latter review article epitomises the newly introduced rapid methods such as NGS, Fourier-transform infrared (FTIR) spectroscopy, MALDI-TOF and lab-on-a-chip (LoC) technologies addressing the detection of AMR in a fast and reliable manner. In Table 1 , the courses, conferences and seminars attended by the fellows are summarised. Fellows were also actively participating in outdoor activities hosted by members of the laboratory and by the work programme supervisors. Both fellows have successfully fulfilled the objectives and tasks of the proposed work programme. Activities performed allowed the fellows to acquire skills related to the execution of risk assessments and the communication and dissemination of results through various actions (seminars, seminars, publications). More specifically, one review article is going to be submitted in the peer-reviewed Journal of Antibiotics (impact factor 3.893) and the second one in the Journal of Microbiological Methods (Impact Factor: 1.707). Moreover, Fellows were familiarised with risk assessment and Monte Carlo simulation software for data processing. Furthermore, short trainings were performed regarding NGS analyses as well as MALDI-TOF apparatus use. Certain objectives were slightly amended due to the COVID-19 pandemic; however, the fellows successfully fulfilled the main objectives and tasks of the work programme. Apart from the two review papers which are going to be submitted during September, the fellows are planning to continue collaborating in the following months in order to publish the results stemming from the activities regarding the objectives 2 and 3; more specifically, the results of the analysis on the occurrence of AMR from FPEs and the WGS analyses of the most interesting isolates, as well as those from the exposure assessment modelling for ESBLs in pork products. The results of this ongoing effort will very likely lead to 2-3 more publications. The close cooperation between the fellows and the hosting site personnel will be maintained. Hopefully, research grant proposals between sending and hosting institutions will be drafted. Moreover, close collaboration between the hosting site and the home institution of the fellows is envisioned in order to further proceed with the outcomes of the work programme and finalise the remaining publications. The role and implementation of next-generation sequencing technologies in the coordinated action plan against antimicrobial resistance Antibiotic resistance in foodborne bacteria Centers for Disease Control and Prevention, online. Antibiotic Resistance, Food, and Food Animals. EFSA (European Food Safety Authority), 2020. Available online The agri-food chain and antimicrobial resistance: a review How next-generation sequencing can address the antimicrobial resistance challenge Acknowledgements: This report is funded by EFSA as part of the EU-FORA programme.