Powdery mildew (PM) is a major fungal disease that threatens thousands of plant species. PM is caused by Podosphaera leucotricha in apple and Erysiphe necator in grapevine. Powdery mildew is controlled by frequent applications of fungicides, having negative effects on the environment, and leading to additional costs for growers. To reduce the amount of chemicals required to control this pathogen, the development of resistant apple and grapevine varieties should become a priority. PM pathogenesis is associated with up-regulation of specific MLO genes during early stages of infection, causing down-regulation of plant defense pathways. These up-regulated genes are responsible for PM susceptibility (S-genes) and their knock-out causes durable and broad-spectrum resistance. All MLO S-genes of dicots belong to the phylogenetic clade V. In grapevine, four genes belong to clade V. VvMLO7, 11 and 13 are up-regulated during PM infection, while VvMLO6 is not. Chapter 2 reports the genome-wide characterization and sequence analysis of the MLO gene family in apple, peach and woodland strawberry, and the isolation of apricot MLO homologs. Twenty-one homologues were found in apple, 19 in peach and 17 in woodland strawberry. Evolutionary relationships between MLO homologs were studied and syntenic blocks constructed. Candidate genes for causing PM susceptibility were inferred by phylogenetic relationships with functionally characterized MLO genes and, in apple, by monitoring their expression following inoculation with the PM causal pathogen P. leucotricha. In apple, clade V genes MdMLO11 and 19 were up-regulated, whereas the two other members of clade V, MdMLO5 and 7, were not up-regulated. The clade VII gene MdMLO18 was also up-regulated upon P. leucotricha infection. Chapter 3 reports the knock-down, through RNA interference, of MdMLO11 and 19, as well as complementation of the mutant phenotype by expression of the MdMLO18 gene in the Arabidopsis thaliana triple mlo mutant Atmlo2/6/12. The knock-down of MdMLO19 resulted in a reduction of PM disease severity up to 75%, whereas the knock-down of MdMLO11, alone or combined with MdMLO19, did not cause any reduction or additional reduction of susceptibility compared to MdMLO19 alone. Complementation by MdMLO18 did not restore susceptibility. Cell wall appositions (papillae), a response to PM infection, were found in both susceptible plants and PM resistant plants where MdMLO19 was knocked-down, but were larger in resistant lines. The expression analysis of 17 genes related to plant defense, and quantification of phenolic metabolites in resistant lines revealed line-specific changes compared to the control. Chapter 4 evaluates the presence of non-functional alleles of the MdMLO19 S-gene in apple germplasm. The screening of the re-sequencing data of 63 apple genotypes led to the identification of 627 SNP in five MLO genes (MdMLO5, MdMLO7, MdMLO11, MdMLO18 and MdMLO19). Insertion T-1201 in MdMLO19 caused the formation of an early stop codon, resulting in a truncated protein lacking 185 amino-acids and the calmodulin-binding domain. The presence of the insertion was evaluated in a collection of 159 apple genotypes: it was homozygous in 53 genotypes, 45 of which were resistant or very resistant to PM, four partially susceptible and four not assessed. These results strongly suggest that this insertion is causative for the observed PM resistance. The absence of a clear fitness cost associated to the loss-of-function of MdMLO19, might have contributed to the high frequency of the mutation in breeding germplasm and cultivars. Among the genotypes containing the homozygous insertion, ‘McIntosh’ and ‘Fuji’ are commonly used in apple breeding. After barley and tomato, apple is the third species with a reported natural non-functional mlo allele in its germplasm, with the important difference that the allele is present in a relatively large number of apple genotypes, most of which not related to each other. Chapter 5 reports the knock-down through RNA interference of four grapevine MLO genes, all members of clade V. VvMLO7, 11 and 13 are up-regulated in early stages of infection, whereas VvMLO6 is not responsive to the pathogen. Knock-down of VvMLO6, 11 and 13, alone or combined, did not decrease PM severity, whereas the knock-down of VvMLO7, alone or in combination with VvMLO6 and VvMLO11, caused a reduction of severity of 77%. Cell wall appositions (papillae), a response to PM attack, were present in both resistant and susceptible lines, but were larger in resistant lines. Thirteen genes involved in defense were less up-regulated in resistant plants, highlighting the reduction of PM disease severity. In Chapter 6 we discuss the results presented in this thesis. The pivotal role of MLO genes in the interaction of PM pathogens with apple and grapevine is described and further experiments aimed at addressing open questions are proposed. The results described in this thesis open interesting avenues in MLO genes research, particularly the finding that a natural mlo mutation in apple appeared to be more common than expected. This mutation is directly applicable in marker assisted breeding for durable PM resistance in apple.