Unravelling the genetic architecture of fungicide resistance and cultivars linked aggressiveness in a collection of isolates from the wheat fungal pathogen Zymoseptoria tritici, collected in France in 2018

Zymoseptoria tritici is responsible of septoria tritici blotch (STB), which is among the most economically detrimental diseases of wheat worldwide. This disease can be controlled using resistant varieties and/or fungicides, but fungal populations are known to adapt to those control methods which can therefore become less effective. Previous studies have suggested the existence of a positive correlation between increasing fungicide resistance in pathogen populations and increasing aggressiveness (i.e. the quantitative component of pathogenicity). This link may be explained by the co-selection of both traits in fungal populations, or by direct mechanistic relationship with identical genes being involved in the adaptation to both traits. Our goal was to validate or infirm this correlation and, if occurs, check these two assumptions. To this end, we first collected Z. tritici isolates from two locations in France (Bignan in Bretagne, Rots in Normandie), on four bread wheat cultivars having different degrees of resistance to STB, and from fields treated and untreated with fungicides. We selected a panel of 126 Z. tritici isolates that we evaluated for their aggressiveness (on the four sampled wheat cultivars grown under controlled conditions), and for their level of resistance (based on EC50 assessed in vitro) to 4 DMI and 2 SDHI fungicides. We established a moderate negative relationship between resistance to some DMIs and aggressiveness on two cultivars. Then, we performed whole-genome sequencing of the 126 isolates on Illumina NovaSeq 6000 (i.e. paired-end 150bp sequences). Mapping these sequences on the genome of the reference strain IPO323, we obtained a matrix of 1,003,343 high-quality variants, including 902,503 SNPs and 100,840 short InDels. These data were used to perform a genome wide association study (GWAS) in order to identify candidate genes involved in fungicide resistance and aggressiveness, followed by a multivariate meta-analysis to identify potential shared variants associated with both traits. Following this approach, two candidate loci were identified. Both could have a pleiotropic effect on fungicide resistance and aggressiveness, and explain the negative correlation observed between both traits. The first locus corresponds to a gene encoding a phosphodiesterase type 1 (PDE1), and the second locus is intergenic and could have an effect on the expression of flanking genes.