Modulating the gene networks underlying phenotypic robustness in plant quantitative immunity – PRobIty

To document the global regulation of S. sclerotiorum transcripts during the colonization of A. thaliana and gains insights into regulators of known virulence-associated pathways, we have collected multiple S. sclerotiorum genome-scale expression datasets and formatted them to make them directly comparable. To this date, our collection includes 269 samples, among which 103 correspond to interaction with the model plant A. thaliana. we have set up a pipeline for regulatory network reconstruction and used it on a core set of fungal genes highly and stably expressed in planta. In this pipeline, network edges were identified based on a meta-analysis of LASSO regression and Random Forest inference strategies.

We completed the analysis of the global transcriptome of S. sclerotiorum during the colonization of hosts from six botanical families initiated in the frame of previous projects. This work revealed that 52% of S. sclerotiorum genes upregulated in planta were host-specific. Genes related to detoxification of host defense compounds were enriched in the specialized transcriptomes, while the core transcriptome overrepresented functions associated with carbohydrate catabolism and sugar transport. Focusing on camalexin, a defense compound specific for plants in the Brassicaceae family, we provide evidence that cis-regulatory variation contributes to the evolution of camalexin responsiveness in Sclerotinia. Using promoter region analyses, we identified a motif enriched in the cis-elements of S. sclerotiorum genes but not their orthologs in the closely-related S. trifoliorum. In baker’s yeast, the motif is recognized by zinc finger transcriptional regulators. We have initiated the functional characterization of these regulators in S. sclerotiorum and we expect to uncover pathways controlled by these important transcriptional regulators.

Future efforts will aim at:
- Generating a detailed infection time course on A. thaliana Col-0, and global expression data for S. sclerotiorum natural and mutant strains and A. thaliana mutant genotypes, to assess the robustness of gene expression to genetic changes.
- Building a hybrid model connecting regulatory network and metabolic network, building a met-genome model including plant and fungal gene exploiting dual RNA-seq data in which global gene expression for S. sclerotiorum interacting with A. thaliana was assessed.

1. Kusch S, Larrouy J, Ibrahim HMM, Mounichetty S, Gasset N, Navaud O, Mbengue M, Zanchetta C, Lopez-Roques C, Donnadieu C, Godiard L, Raffaele S. Transcriptional response to host chemical cues underpins the expansion of host range in a fungal plant pathogen lineage. ISME J. 2021 Jul 19. doi: 10.1038/s41396-021-01058-x.

Interaction with fungi is a key determinant of plant growth in natural and in agricultural settings, with fungal pathogens causing severe yield losses worldwide. Quantitative disease resistance (QDR) is a form of plant immunity involving complex molecular bases that remain poorly characterized. It is widespread in nature and is often broad spectrum and durable. Our recent work revealed genes from the model plant Arabidopsis thaliana with diverse molecular functions mediating QDR to the fungal pathogen Sclerotinia sclerotiorum. How these genes contribute to the QDR phenotype remains however unclear. We hypothesize that the robustness of the QDR to pathogen diversity emerges at the global scale from the architecture of plant and pathogen gene networks, involving numerous conserved genes.

We propose to combine global transcriptome analysis, genome scale modeling and genome editing to provide a systematic characterization of the topology of gene networks associated with QDR in Arabidopsis and study the link between network topology and phenotypic robustness. To this end, we will characterize plant and fungal transcriptome reprogramming over time in wild type and in Arabidopsis mutant lines impaired in QDR. These data will be integrated into a genome-scale dynamic modeling framework enabling predictive approaches for plant disease management. Using Machine Learning approaches, model inversion and optimization, and network motif analysis we will characterize sets of genes associated with QDR. Next, we will use multiplex genome editing in Arabidopsis to validate experimentally synergy, redundancy, compensation and tradeoff phenomena revealed by the modeling approach. Finally, we will apply genome-scale dynamic models to design Arabidopsis genotypes with QDR resilient to unfavorable climatic conditions. Our preliminary work indicated that small variations in climatic condition impair QDR to Sclerotinia in Arabidopsis. Combining model predictions and multiplex genome editing, we will aim at improving QDR in unfavorable realistic global climate projections.

Studying plant and fungal gene networks underlying QDR phenotypic responses robust to internal and environmental perturbations will lead to conceptual advances on the molecular bases of durability, and provide novel opportunities for engineering disease resistance in crops reducing the use of fungicides.