The effect of the allelic diversity in AvrSr35 on Sr35-based resistance in wheat

Abstract

Ug99 is the most devastating race group of wheat stem rust caused by the fungal pathogen Puccinia graminis f. sp. tritici (Pgt) and was originally found in Uganda in 1999 (Pretorius et al., 2000). Ug99 has since rapidly evolved giving rise to new virulent races that spread in parts of Africa and the Middle East, causing significant yield loss and food insecurity. The expansion of the Ug99 lineage provided evidence that this fungus is capable of evolving quickly to overcome resistance of a large number of wheat varieties with known wheat stem rust resistance (R) genes. The emergence of the Ug99 race group highlights the importance of studies aimed at understanding the evolutionary dynamic of R genes and their corresponding avirulence (Avr) factors. Here, I used a stem rust-wheat pathosystem to investigate the impact of intra-species variation in a recently identified fungal effector AvrSr35 on the ability of stem rust resistance gene Sr35 to recognize it and trigger defense response. We have identified the allelic variants of AvrSr35 inducing reduced immune response in heterologous (tobacco leaves) and homologous (wheat protoplasts) transient expression systems. Missense mutations in the coding region of AvrSr35 with potential to interfere with AvrSr35-Sr35 interaction were detected by comparing allelic variants of AvrSr35 in a diverse collection of Sr35-virulent and Sr35-avirulent Pgt isolates. To facilitate molecular analyses of wheat-rust pathosystem and characterization of additional R-Avr pairs, we assembled the reference genome of U.S. Pgt isolate 99KS76A-1 (race RKQQC). Advances in long-read sequencing technologies provide the opportunity to improve the quality of genome assemblies of many pathogens, including the complex cereal rust genomes. The assembly of the dikaryotic Pgt genome is challenged by the presence of two haplotypes in the dikaryon (Spatafora et al., 2017). Using Oxford Nanopore long-read sequencing in combination with error correction based on Illumina short reads, we generated a haplotype-resolved assembly of the 99KS76A-1 isolate. We demonstrate that our assembly contains fewer gaps and shows higher N50 value than currently available reference genomes.