Genome-wide expression QTL mapping reveals the highly dynamic regulatory landscape of a major wheat pathogen

Abstract

In agricultural ecosystems, outbreaks of diseases are frequent and pose a significant threat to food security. A successful pathogen undergoes a complex and well-timed sequence of regulatory changes to avoid detection by the host immune system, hence well-tuned gene regulation is essential for survival. However, the extent to which the regulatory polymorphisms in a pathogen population provide an adaptive advantage is poorly understood. We used Zymoseptoria tritici, one of the most important pathogens of wheat, to generate a genome-wide map of regulatory polymorphism governing gene expression. We investigated genome-wide transcription levels of 146 strains grown under nutrient starvation and performed expression quantitative trait loci (eQTL) mapping. We identified cis-eQTLs for 65.3% of all genes and the majority of all eQTL loci are within 2kb upstream and downstream of the transcription start site (TSS). We also show that polymorphism in different gene elements contributes disproportionally to gene expression variation. Investigating regulatory polymorphism in gene categories, we found an enrichment of regulatory variants for genes predicted to be important for fungal pathogenesis but with comparatively low effect size, suggesting a separate layer of gene regulation involving epigenetics. We also show that previously reported trait-associated SNPs in pathogen populations are frequently cis-regulatory variants of neighboring genes with implications for the trait architecture. Overall, our study provides extensive evidence that single populations segregate large-scale regulatory variation and are likely to fuel rapid adaptation to resistant hosts and environmental change.

Population-level transposable element expression dynamics influence trait evolution in a fungal crop pathogen

“…In Z. tritici , RIP induces C→T transitions at CpA sites ( 45 ). We analyzed all SNPs in the genome associated either with TE transcription or gene transcription ( 51 ). Then, we assessed whether the polymorphism is likely RIP associated.…”