Genome-wide signatures of selection in C. kahawae reveal candidate genes potentially involved in host-specialization. Genome-wide signatures of selection in C. kahawae reveal candidate genes potentially involved in host-specialization

Plants and their pathogens are engaged in continuous evolutionary battles, with pathogens evolving tocircumvent plant defense mechanisms and plants responding through enhanced protection to prevent ormitigate damage induced by pathogen attack. Artificial ecosystems are composed of genetically identicalpopulations of crop plants with little changes from year to year. These environments are highly conduciveto the emergence and dissemination of pathogens and they exert selective pressure for both quantitativetraits linked to pathogen fitness, such as aggressiveness, and qualitative virulence factors responsible forfungi pathogenicity. In this study, we used a comparative genome-wide approach to investigate thegenomic basis underlying the pathogenicity and aggressiveness of the fungal plant pathogenColletotrichum kahawae infecting green coffee berries. The pathogenicity was investigated by comparinggenomic variation between C. kahawae and its non-pathogenic sibling species, while the aggressivenesswas studied by a genome-wide association approach with groups of isolates with different phenotypicprofiles. High genetic differentiation was observed between C. kahawae and the most closely relatedspecies with 5 560 diagnostic SNPs identified, in which a significant enrichment of non-synonymousmutations was detected. Functional annotation of these non-synonymous mutations revealed a significantenrichment mainly in two gene ontology categories, “oxidation-reduction process” and “integralcomponent of membrane”. Finally, the annotation of several genes potentially under-selection revealedthat C. kahawae’s pathogenicity, may be a complex biological process in which important biologicalfunctions such as, detoxification and transport, regulation of host and pathogen gene expression, andsignaling are involved. On the other hand, the genome-wide association analyses for aggressivenesswere able to identify 10 SNPs and 15 SNPs of small effect in single and multi-association analysis,respectively, from which 7 were common. The annotation of these genomic regions allowed theidentification of four candidate genes (“F-box domain-containing”, “nitrosoguanidine resistance”, “Fungalspecific transcription factor domain-containing” and “C6 transcription factor”) that could be associated withaggressiveness. This study provides, for the first time, lights on the mechanisms and loci putativelyinvolved in C. kahawae aggressiveness and pathogenicity.