Transcriptome-based host epistasis and pathogen co-expression in barley-powdery mildew interactions

Authors: VELASQUEZ-ZAPATA, VALERIA - Iowa State University; SMITH, SCHUYLER - Iowa State University; PRIYANKA, SURANA - Sanger Institute; CHAPMAN, ANTONY V. - Rothamsted Research; Wise, Roger

Submitted to: Genetics
Publication Type: Pre-print Publication
Publication Acceptance Date: 9/23/2023
Publication Date: 9/23/2023
Citation: Velasquez-Zapata, V., Smith, S., Priyanka, S., Chapman, A.E., Wise, R.P. 2023. Transcriptome-based host epistasis and pathogen co-expression in barley-powdery mildew interactions. Genetics. https://doi.org/10.1101/2023.09.18.558274.
DOI: https://doi.org/10.1101/2023.09.18.558274

Interpretive Summary: Plant pathogens are among the greatest threats to crop production worldwide, resulting in yield losses of 10 to 20% (= $100 to 200 billion) each year. Obligate biotrophic fungi, for example, mildews and rusts, require a living host to survive and cause some of the most destructive epidemics. A general view of the regulatory programs that render a plant resistant to pathogens is beginning to emerge in model organisms, but is still in its infancy in large-genome temperate cereals that are vital to feeding the world's growing population. The interaction between barley, and the powdery mildew fungus, is central to address this challenge. The nucleotide-binding leucine-rich-repeat (NLR) immune receptor encoded by Mildew locus a (MLA) is an ancestral protein required for protection against destructive cereal diseases, including powdery mildew, Ug99 stem rust, stripe rust, and rice blast. This research reports the use of time-course gene expression of barley infected with the powdery mildew pathogen to infer diverse gene effects governed by the MLA immune receptor and two other host factors critical to disease defense, Blufensin1 (Bln1) and Required for Mla6 resistance3 (Rar3). Gene effect models revealed epistatic interactions between Mla6 and Bln1 (a situation where the expression of one gene is modified by the expression of other genes) and the impact of rar3 on the barley and powdery mildew transcriptomes. For the first time, we classified 115 novel NLRs under unique gene effect models, which clustered at chromosome hotspots, suggesting genome accessibility and recruitment of transcriptional machinery as a contributing factor to resistance activation. Impact: Most plant resistance genes deployed in agriculture encode NLRs. However, the mechanisms by which NLR receptors impart critical functions to plant cells are often targets of pathogen effectors, thus, these discoveries provide a foundation for further research into the complex molecular interactions that control disease resistance in crops. These results impact modern breeding approaches to expand resistance to new and emerging pathogens.

Technical Abstract: Mildew locus a (Mla) from the Triticeae grain crop barley (Hordeum vulgare L.) encodes a multi-allelic series of nucleotide-binding leucine-rich repeat (NLR) immune receptors. These variable NLRs recognize complementary secreted effectors from the powdery mildew fungus, Blumeria hordei (Bh), to block disease progression. We used a dynamic time-course transcriptome of barley infected with Bh to infer gene effects and epistatic relationships governed by the Mla6 NLR, two other host loci critical to the interaction, Blufensin1 (Bln1) and Required for Mla6 resistance3 (Rar3), and genes that interact with them. Bln1 is an R-gene independent regulator of immunity and the resistant bln1 mutant exhibits enhanced basal defense to compatible Bh. Rar3 is required for MLA6-mediated generation of H2O2 and the hypersensitive reaction; the rar3 mutant contains an in-frame Lys-Leu deletion in the SGT1-specific domain that compromises immunity by a subset of Mla alleles. Interactions of Mla6 and Bln1 resulted in symmetric, suppression and masked epistasis on the Bh-induced barley transcriptome. Likewise, dominant or equal effects were caused by Mla6 and Sgt1. Of a total of 468 barley NLRs, 366 were expressed in our dataset and 115 of those were grouped under different gene effect models, which localized to several chromosome hotspots. The corresponding Bh infection transcriptome was classified into nine co-expressed modules, linking differential expression with pathogen development. Expression of most of 517 Bh effectors exhibited dependence on disease phenotype and was associated with appressorial or haustorial structures, suggesting that disease is regulated by a host-pathogen intercommunication network that diversifies the response.