Epigenetic modifications are heritable changes in gene expression that occur without changes in the DNA sequence. They are responsible of important phenotypic trait variations in all kingdoms of life. DNA methylation is the most common form of epigenetic modification in prokaryotes and eukaryotes and is catalyzed by DNA methyltransferases (MTases). The 5mC (5-methylcytosine) modification can be detected using treatment of DNA with bisulfite and various sequencing technologies. Recently, the development of the PacBio sequencing technology has enabled sequencing of single molecules in real time (SMRT) and detecting other methylated bases (e.g. 6mA, 4mC). In bacteria, the biological consequences of DNA methylation have not been extensively investigated yet. The impact of DNA methylation in virulence has been reported in some human and animal pathogens. However, the involvement of DNA methylation in host-pathogen interaction remains largely unexplored for most of the bacterial plant pathogens. The main purpose of the EPI-PATH project is to investigate the role of DNA methylation in the plant pathogenic bacterium Ralstonia solanacearum during host jump and adaptation. This bacterium is a Species Complex (named RSSC) responsible of the bacterial wilt, one of the most destructive plant bacterial diseases affecting more than 250 plant species including many important crops (potato, tomato, banana, peanuts). For that purpose, we will first study the MTases in the GMI1000 strain. In vitro and in planta expression of the MTases will be compared. The role of each GMI1000 MTase in virulence will be investigated by generating single-deleted mutants. The impact of MTase deletion on the GMI1000 methylome will be determined using both SMRT and bisulfite/Illumina sequencing. Here, protocols for library construction and DNA treatment will be optimized in order to detect the 6mA, 4mC and 5mC modifications. For data analysis, existing bioinformatics tools will be used and adapted to high GC content genomes, such as RSSC genomes. The MTase repertoires and expression, and the methylomes will then be investigated in 20 wild-type strains representative of the RSSC genetic diversity. This should provide a global overview of the methylome diversity in RSSC and potential correlation with host specificity. In a second time, we will address the methylome diversification in 30 experimentally evolved clones, derived from the GMI1000 strain after serial passages during 300 generations in a given host. These clones show a fitness gain in their experimental host (tomato, eggplant, bean, cabbage) but their genomes reveal only little even any mutation compared to the ancestral clone. The methylomes of the ancestral and the evolved clones will be compared in order to identify differential methylation marks. We will focus on differential methylation marks found in several independent evolved clones. The transcriptome profiles of the evolved clones will also be investigated in order to connect differential methylation marks in gene promoters with transcriptome variation. For the most interesting candidates, the effect of individual methylation marks on the in planta fitness will be characterized by conducting site-directed mutagenesis in the ancestral GMI1000 clone and then measure the fitness of the mutants in planta. This should validate the impact of DNA methylation in the expression of the adaptive phenotype of GMI1000 on different host plants. The EPI-PATH project will provide a first step toward understanding the biological consequences of DNA methylation in RSSC, which will represent one of the first plant pathogen investigated for epigenetic modification impact. The knowledge on the MTases and the differential methylation marks within RSSC strains together with the tools optimized in this project will constitute solid roots for future studies aiming at identifying the role of epigenetic modifications in host adaptation in other plant pathogenic bacteria.

The aim of the Path2Bos project is to reconstruct the evolutionary path of cattle from its domestication starting about 10,000 years ago in Anatolia, during its later spread into Europe and Africa and up to now, through a paleogenomic analysis of fossils, the direct witnesses of evolution. The goal is to identify genomic regions that were selected at the early stages of domestication, corresponding to basic phenotypes that should be preserved in ongoing genetic selection schemes. The project is based on a previous paleogenetic characterization of a large number (~ 700) of 9,000- to 1,000-year-old archaeological bones of ancient domesticated cattle and their wild ancestors, the aurochs. We have genotyped the mitochondrial genomes and sequenced the hypervariable regions of almost 200 of these ancient bones, allowing us to assign reliably their mitochondrial haplogroups and to follow the evolution of populations from their initial domestication in Anatolia during the Neolithic as well as their spread and evolution in Europe and North Africa until the Middle Ages. Using sequence capture, we obtained complete mitogenomes from 40 of these samples representing the various clades, reconstructed the evolution and timing of radiation of aurochs’ populations, and untangled the impacts on population diversity of both climate changes during the Pleistocene-Holocene transition and initial domestication. We have sequenced the genome of a 9,000-year-old aurochs from the domestication centre in Anatolia that will serve as a reference genome to follow the genomic changes and selective sweeps during the domestication process. We propose to sequence about 30 of these ancient genomes and to compare them with genomes and phenotypic records from modern domestic animals to reconstruct many aspects of the selection pressure exerted during different prehistoric and historic periods. We will also sequence several individuals from modern hardy breeds to generate reference genetic data from breeds that have escaped recent selection schemes or that were selected for alternative phenotypes. Our data will be used in combination with modern genomic data from the 1,000 Bull Genomes consortium in various complementary ways to identify and to date signatures of selection during the cattle domestication process. We will screen for selection events that are either recent or old, complete or ongoing, acting on new variants or on standing variation. Using powerful tools to detect selective sweeps in genomes, ancient genomic data will provide the ability to date the various selection events, to identify the population(s) of origins onto which selection was exerted and to explore the validity of the various demographic models used to detect selective sweeps from modern genomic data. We will also use extensive GWAS data, produced by one of us using modern cattle, to reconstruct the past evolution of complex multigenic traits. Path2Bos will (1) improve the power and accuracy of the identification of genomic regions under selection, (2) estimate the strength of selection and date the origin of the corresponding selective events, (3) identify variants that were selected in the past and that have been lost in modern selection schemes, thereby pinpointing the genetic bases of phenotypic traits that might be useful to preserve for the long-term sustainability of cattle husbandry. Thus, it will provide an original and very useful cattle genome annotation data source to complement the genomic characterization efforts of modern cattle breeds and enrich the current selection strategies. A strong point of Path2Bos is the complementarity of the expertise and resources, including preliminary data, of the consortium partners, in particular paleogenomics and a large collection of characterized archeological samples, involvement in the 1000 Bull Genomes consortium and GWAS, selective sweep method developments and analyses of genomes and a high-throughput sequencing facility.