Exploring the role of downy mildew susceptibility genes in Vitis by studying the genetic variation and the in vivo function

Viticulture is worldly threatened by fungal pathogens. Management of vineyards entails the use of great amounts of fungicides in several applications each year, generating a harmful environmental impact on human health and local biodiversity. In the last decades, resistance (R) loci scouting and deployment have been the main strategy to cope with some of the most aggressive grapevine pathogens: Erysiphe necator, the causal agent of powdery mildew (PM), and Plasmopara viticola, the causal agent of downy mildew (DM). However, R loci-based resistance can be easily overcome by pathogens within a few years from their introgression. Recently, susceptibility (S) genes have been used as a new source of durable and broad-spectrum resistance in many crops and tree species. In grapevine, S genes like VvMLO7 were associated with susceptibility to PM and their role has been studied, also using knock down mutants. On the other hand, S genes associated to susceptibility to DM are not yet available in grapevine. Loss of function mutations in AtDMR6 and AtDLO provide resistance against DM in Arabidopsis, therefore their putative grapevine orthologs VvDMR6.1, VvDMR6.2, VvDLO1 and VvDLO2 are candidate S genes, likely associated to DM in grapevine .In this thesis, we explored the role of the aforementioned grapevine S genes starting from a broad genetic diversity analysis to a functional characterization study on dmr6.1 grapevine plants. Regarding the survey on genetic variability, VvDMR6.1, VvDMR6.2, VvDLO1 and VvDLO2 were investigated in 190 grapevine genotypes belonging to Vitis vinifera spp., wild species, hybrids and the so-called hybrid/wild species, in order to find S genes natural mutants. The scouted genes were deep-sequenced and reads were mapped on PN40024 12× V2 reference genome. A bottleneck analysis was carried out in order to, firstly, identify SNPs (Single Nucleotide Polymorphisms) impacting on the coding sequence, then investigate the potential disrupting role of impacting SNPs on codons and the amino acid sequence and therefore on the protein folding and function. A representative handful of disrupting SNPs were chosen for confirmation by Sanger sequencing. The disrupting impact of amino acid mutations caused by the validated SNPs was then checked on a protein three-dimensional model. DM resistance phenotypic data were collected and compared to the frequency of the reconstructed haplotypes per each gene. Two of them, in VvDMR6.2, were found significantly more represented in DM resistant genotypes. VvMLO7 was sequenced in the 190 grapevine accessions as well, and the resulting data were subjected to the same bottleneck analysis. Once amino acid sequence-disrupting mutations were scouted, we took advantage of the known MLO protein model to identify those mutations that were changing conserved amino acids. Ten mutations were predicted to impact protein function, but no association with phenotypic data was possible since all SNPs were at the heterozygous state. This broad survey provided a resource for grapevine and plant genetics and could corroborate genomic-assisted breeding programs as well as tailored gene-editing approaches for resistance to biotic stresses. Functional characterizations of VvDMR6.1, VvDMR6.2, VvDLO1 and VvDLO2 were carried out taking advantage of dmr6.1 grapevine plants previously obtained by the CRISPR/Cas9 genome editing technique. Plants of four different edited lines (each line originated from a separate transformation event) were subjected to P. viticola inoculation assay. Leaves were sampled at 0, 24, 96 hours post-inoculation (hpi) and 8 days post-inoculation (dpi). Samples were collected for several purposes. A gene expression analysis of VvDMR6 and VvDLO genes and pathogenesis-related genes was carried out on 0, 24 and 96 hpi samples. Samples taken at 8 dpi were used for symptom assessment via visual, digital, and histological observation. Since the salicylic acid (SA)-inactive forms 2,5-dihydrobenzoic acid (2,5-DHBA) and 2,3-DHBA are products of S5H (SA-5- hydroxylases) and S3H enzymes, putatively encoded by VvDMR6 and VvDLO, these metabolites were quantified through LC-MS in the same samples collected at 0, 24 and 96 hpi. This overview on the many aspects related to VvDMR6 and VvDLO gene function, made it clear that no uniform effect of VvDMR6.1 knock-out was detectable among different edited lines and further investigations are needed to define the role of the single genes and the relationship among them.