Skip to main content

Mechanisms of Immune Receptor Diversification in Cereals

Periodic Reporting for period 1 - MIREDI (Mechanisms of Immune Receptor Diversification in Cereals)

Reporting period: 2017-09-01 to 2019-02-28

The plant immune system is innate – it is encoded in the germline. In natural pathosystems, plants effectively deploy hundreds of immune receptors to detect and disarm rapidly evolving pathogens including viruses, bacteria, nematodes, insects, fungi and oomycetes. Exactly how such receptor diversity can evolve is an elusive question with important practical ramifications. A central class of plant immune receptors, called Nucleotide Binding Leucine Rich Repeats proteins (NLR), has been implicated in recognition of vastly diverse pathogen-derived effector molecules. An emerging paradigm of NLR diversification involves new gene fusions to host proteins that are normally targeted by pathogen effectors. Such NLR fusions ‘bait’ pathogens by integrating proteins that are targeted by the pathogen and therefore were named ‘integrated domains’ (NLR-IDs). We study NLR-IDs in grasses, a plant family that includes the three most important crops: maize, rice and wheat. We combine state of the art genome sequencing techniques and bioinformatics, molecular biology and novel reverse genetics approaches to determine how NLRs have evolved in grasses, what limits diversification of immunity in domesticated crops, and how to apply this knowledge to engineer novel NLR receptors in cereals.
In the past year, we have successfully completed and published our analysis of NLRs in 9 publicly available grass genomes that identified NLR clade prone to new exogenous domain fusions (Bailey et al Genome Biology 2018). In order to further test the effects of speciation, domestication and inbreeding on the number of NLRs in grasses, we acquired diverse germplasm of 5 species of Panicoideae. We designed an enrichment capture for targeted resequencing of NLRs, domestication genes and genome markers for wild, landraces and elite inbred members of the economically important crops of Maize, Sorghum bicolor and Setaria Italica. We obtained the seeds, extracted DNA from 80 samples across and Illumina libraries were prepared for capture and sequencing by Arbor bioscience using long read PacBio platform. In additiona, we prepared a set of 60 tetraploid wheat lines that include both domesticated and wild species for capture-based sequencing of NLRs. For wheat, we have analysed newly available genomes and prepared sequences to supplement existing wheat capture design. We already secured wheat DNA samples from our collaborators and are preparing them for sequencing. In addition, we are generating RNA sequencing data using newest long read technologies for a subset of species that represent independent polyploidization and hybridization events.

Protein domains integrated into NLRs are considered putative, homologous decoys of plant proteins, targeted by pathogen effectors. We undertook a new approach to build on the previous NLR-ID pipelines developed in our lab.This new approach incorporated a reciprocal BLAST analysis comparing the entire NLR complement from the plant genomes on public databases, including Phytozome, Plant Ensemble, and Refseq against all of the predicted proteins from any plant species.This analysis allows to identify putative plant proteins targeted by pathogens, which represent potential susceptibility molecules that pathogens exploit to establish disease. Using the A. thaliana protein accessions, we mapped NLR-IDs to plant metabolic and signalling pathways. This analysis successfully enabled identification of a series of pathways represented by multiple NLR-ID homologues. We have cloned nine NLR-IDs that we are currently challenging with putative corresponding effectors from economically important bacterial, fungal, and insect pathogens of wheat to identify functional NLRs for future study.

Finally, we have identified and cloned a candidate NLR platform that was able to signal independently of other paired NLRs in heterologous Nicotiana benthamiana. We are currently performing deletions and domain swap analysis to test whether this platform can generate new functional fusions.
Because several NLR/NLR-IDs have been shown to function in pairs and are genetically linked in head to head orientation, we extended our initial analyses and wrote the tools and tested co-occurrence of NLRs in pairs. The results of our ‘tandem’ analyses are published in Bailey et al 2018 and scripts are freely available at group github page. In brief, we identified that NLR-IDs from our previously identified major integration locus were significantly enriched in ‘paired’ NLR genes and were paired with NLRs from another specific clade.

To improve our yeast two hybrid screens beyong the state of the art, we designed a bait capture to target effectors predicted from Puccinia graminis f. sp. tritici, Puccinia striiformis f. sp. tritici (provided by Diane Saunders at JIC) and wheat blast Magnaporthe oryzae (provided by Nick Talbot, Exeter). DNA sequence of predicted effectors was extracted, conserved effectors with over 90% similarity were collapsed to prevent unbalanced enrichment of different effectors. Effector sequence with low complexity was masked to prevent design of baits with off target hits. Then filtering steps where carried out to remove any which were cross hitting to the wheat nuclear, chloroplast or mitochondrial genome. Baits which had low specificity to target sequence were also removed. Finally baits where manually designed to effectors which had no baits after filtering.