Significance: Our past understanding of pathogen evolution has been fragmented because of tendencies to study human clinical isolates. To understand the evolutionary trends of pathogenic bacteria though, we need the context of their nonpathogenic relatives. Our unique and detailed dataset allows description of the parallel evolution of two key human pathogens: the causative agents of plague and Yersinia diarrhea. The analysis reveals an emerging pattern where few virulence-related functions are found in all pathogenic lineages, representing key “foothold” moments that mark the emergence of these pathogens. Functional gene loss and metabolic streamlining are equally complementing the evolution of Yersinia across the pathogenic spectrum.

Summary: Multilocus sequence analysis of 417 strains of Yersinia pseudotuberculosis revealed that it is a complex of four populations, three of which have been previously assigned species status [ Y. pseudotuberculosis sensu stricto (s.s.), Yersinia pestis and Yersinia similis ] and a fourth population, which we refer to as the Korean group, which may be in the process of speciation. We detected clear signs of recombination within Y. pseudotuberculosis s.s. as well as imports from Y. similis and the Korean group. The sources of genetic diversification within Y. pseudotuberculosis s.s. were approximately equally divided between recombination and mutation, whereas recombination has not yet been demonstrated in Y. pestis , which is also much more genetically monomorphic than is Y. pseudotuberculosis s.s. Most Y. pseudotuberculosis s.s. belong to a diffuse group of sequence types lacking clear population structure, although this species contains a melibiose‐negative clade that is present globally in domesticated animals. Yersinia  similis corresponds to the previously identified Y. pseudotuberculosis genetic type G4, which is probably not pathogenic because it lacks the virulence factors that are typical for Y. pseudotuberculosis s.s. In contrast, Y. pseudotuberculosis s.s., the Korean group and Y. pestis can all cause disease in humans.

In response to the EFSA call New approaches in identifying and characterizing microbial and chemical hazards, the project INNUENDO (https://sites.google.com/site/theinnuendoproject/) aimed to design an analytical platform and standard procedures for the use of whole‐genome sequencing in surveillance and outbreak investigation of food‐borne pathogens. The project firstly attempted to identify existing flaws and needs, and then to provide applicable cross‐sectorial solutions. The project focused in developing a platform for small countries with limited economical and personnel resources. To achieve these goals, we applied a user‐centered design strategy involving the end‐users, such as microbiologists in public health and veterinary authorities, in every step of the design, development and implementation phases. As a result, we delivered the INNUENDO Platform V1.0 (https://innuendo.readthedocs.io/en/latest/), a stand‐alone, portable, open‐source, end‐to‐end system for the management, analysis, and sharing of bacterial genomic data. The platform uses Nextflow workflow manager to assemble analytical software modules in species‐specific protocols that can be run using a user‐friendly interface. The reproducibility of the process is ensured by using Docker containers and throught the annotation of the whole process using an ontology. Several modules, available at https://github.com/TheInnuendoProject, have been developed including: genome assembly and species confirmation; fast genome clustering; in silico typing; standardized species‐specific phylogenetic frameworks for Campylobacter jejuni, Yersinia enterocolitica, Salmonella enterica and Escherichia coli based on an innovative gene‐by‐gene methodology; quality control measures from raw reads to allele calling; reporting system; a built‐in communication protocols and a strain classification system enabling smooth communication during outbreak investigation. As proof‐of‐concepts, the proposed solutions have been thoroughly tested in simulated outbreak conditions by several public health and veterinary agencies across Europe. The results have been widely disseminated through several channels (web‐sites, scientific publications, organization of workshops). The INNUENDO Platform V1.0 is effectively one of the models for the usage of open‐source software in genomic epidemiology.

Pigs are considered as a major reservoir of human pathogenic Yersinia enterocolitica and a source of human yersiniosis. However, the transmission route of Y. enterocolitica from farm to pork is still unclear. The transmission of pathogenic Y. enterocolitica from pigs to carcasses and pluck sets was investigated by collecting samples from 364 individual ear-tagged pigs on the farm and at the slaughterhouse. In addition, isolated strains were analyzed, using pulsed-field gel electrophoresis. Isolation of similar genotypes of pathogenic Y. enterocolitica 4/O:3 in animals on the farm and at the slaughterhouse and in carcasses shows that carcass contamination originates from the strains a pig carries during the fattening period. Direct contamination from the carrier pig to its subsequent pluck set is also the primary contamination route for pluck sets, but cross-contamination appears to have a larger impact on pluck set contamination than on carcasses. In this study, the within-farm prevalence of pathogenic Y. enterocolitica varied from 0% to 100%, indicating specific farm factors affect the prevalence of Y. enterocolitica in pigs. The association of farm factors with the high prevalence of pathogenic Y. enterocolitica on farms was studied for the first time, using correlation and two-level logistic regression analyses. Specific farm factors, i.e. drinking from a nipple, absence of coarse feed or bedding for slaughter pigs, and no access of pest animals to pig house, were associated with a high prevalence of pathogenic Y. enterocolitica 4/O:3.

The transmission of Yersinia pseudotuberculosis in the pork production chain was followed from farm to slaughterhouse by studying the same 364 pigs from different production systems at farm and slaughterhouse levels. In all, 1,785 samples were collected, and the isolated Y. pseudotuberculosis strains were analyzed by pulsed-field gel electrophoresis. The results of microbial sampling were combined with data from an on-farm observation and questionnaire study to elucidate the associations between farm factors and the prevalence of Y. pseudotuberculosis . Following the same pigs in the production chain from farm to slaughterhouse, we were able to show similar Y. pseudotuberculosis genotypes in live animals, pluck sets (containing tongue, tonsils, esophagus, trachea, heart, lungs, diaphragm, liver, and kidneys), and carcasses and to conclude that Y. pseudotuberculosis contamination originates from the farms, is transported to slaughterhouses with pigs, and transfers to pluck sets and carcasses in the slaughter process. The study also showed that the high prevalence of Y. pseudotuberculosis in live pigs predisposes carcasses and pluck sets to contamination. When production types and capacities were compared, the prevalence of Y. pseudotuberculosis was higher in organic production than in conventional production and on conventional farms with high rather than low production capacity. We were also able to associate specific farm factors with the prevalence of Y. pseudotuberculosis by using a questionnaire and on-farm observations. On farms, contact with pest animals and the outside environment and a rise in the number of pigs on the farm appear to increase the prevalence of Y. pseudotuberculosis .

The presence of Listeria monocytogenes in the pork production chain was followed from farm to slaughterhouse by examining the farm and slaughterhouse levels in the same 364 pigs, and finally by analyzing the cut meats from the same pig lots. Both organic and conventional farms were included in the study. Altogether, 1,962 samples were collected, and the 424 L. monocytogenes isolates were analyzed by pulsed-field gel electrophoresis. The results from microbial analyses were combined with data from an on-farm observation and a questionnaire to clarify the associations between farm factors and prevalence of L. monocytogenes. The prevalence of L. monocytogenes was 11, 1, 1, 24, 5, 1, and 4% in feed and litter, rectal swabs, intestinal contents, tonsils, pluck sets (including lungs, heart, liver, and kidney), carcasses, and meat cuts, respectively. The prevalence was significantly higher in organic than in conventional pig production at the farm and slaughterhouse level, but not in meat cuts. Similar L. monocytogenes genotypes were recovered in different steps of the production chain in pigs originating from the same farm. Specific farm management factors, i.e., large group size, contact with pet and pest animals, manure treatment, use of coarse feed, access to outdoor area, hygiene practices, and drinking from the trough, influenced the presence of L. monocytogenes in pigs. L. monocytogenes was present in the production chain, and transmission of the pathogen was possible throughout the chain, from the farm to pork. Good farm-level practices can therefore be utilized to reduce the prevalence of this pathogen.