Small changes fuel a rapidly evolving threat: The genomic basis of resistance breaking in spinach downy mildew

Abstract

The cultivation of spinach (Spinacia oleracea L.) faces a severe global threat from downy mildew, a destructive disease caused by the obligate biotrophic oomycete Peronospora effusa. High-density monocultures of resistant varieties exert intense selective pressure on pathogen populations, leading to frequent resistance breakdowns. Newly deployed host resistance genes are rendered obsolete within a few growing seasons. By 2025, the number of officially recognised resistance-breaking isolates, or races, had reached 20, reflecting the alarming evolutionary pace of P. effusa. To uncover the molecular mechanisms driving this adaptation, this thesis establishes a comparative pangenomic framework spanning all known races. Chapter 2 investigates the population structure and reproductive strategies of 26 isolates. Mitochondrial genomes grouped isolates into two haplogroups. Nuclear comparisons, however, identified three highly distinct clonal clusters, while the remaining isolates exhibited complex mosaic genomes. This discordance between mitochondrial and nuclear phylogenies provides definitive genomic evidence for active sexual recombination in the field, where parental alleles are reshuffled to generate highly adaptable genotypes. Chapter 3 shifts to high-resolution comparative genomics by constructing the first sequence-resolved pangenome graph for an oomycete. Telomere-to-telomere assemblies of six diverse isolates revealed a 58 Mb genome organised in 17 core chromosomes. A highly repetitive 2.1 Mb accessory chromosome (chromosome 18) was also identified in a subset of isolates and shown to degrade during sexual recombination. While core chromosomes were remarkably stable, transposable elements comprised over 50% of the genome and drove 80% of structural variation. The pangenome graph also revealed highly variable virulence-associated effector genes, especially in large physical clusters that act as evolutionary hotspots where TE expansions and duplications continuously alter copy number. In Chapter 4, this phased diploid framework was extended to all 19 officially denominated races to connect genomic variation with host-resistance breakdown. Phased assemblies defined distinct haplogroups and exposed recombination and heterozygosity across the collection, including up to fourfold differences in genome-wide heterozygosity. Alternating histories of sexual crossing and prolonged clonal propagation, often marked by complete loss of heterozygosity, were common. Screening phased, unclustered RXLR effectors against spinach differential phenotypes linked candidate effectors to evasion of major resistance loci through independent mechanisms. Evasion of RPF3 is likely caused by a precise 12.7 kb deletion on chromosome 15 removing a candidate RXLR effector, probably mediated by flanking TE repeats and arising independently in separate lineages. In contrast, virulence at RPF4 and RPF11 is associated with local mutations and pseudogenization, including frameshifts that truncate chromosome 15 and chromosome 13 effectors. Ultimately, this work shows that P. effusa adapts with remarkable agility, using transposable-element activity, genetic reshuffling, and precise mutations to outpace single-gene host defences. Because this obligate biotroph is unculturable, managing spinach downy mildew requires a shift from reactive race identification to proactive, predictive surveillance. Systematic sequencing of field isolates should expand the current pangenome, while coupling this framework with structural modelling tools such as AlphaFold can improve effector discovery and clarify molecular interfaces with spinach immune receptors. Together, these strategies can guide breeding of multigenic cultivars and help protect the spinach supply chain.