Summary

In plants, intracellular immune receptors of the NLR (nucleotide-binding, leucine-rich repeat) family are modular proteins that monitor translocated pathogen effector proteins and activate immune responses typified by the hypersensitive cell death. One of these NLR proteins is MLA, a member of a group of proteins that carry a coiled-coil (CC) domain prior to the nucleotide-binding (NB) and leucine-rich repeat (LRR) domains. In this paper, the authors mined the transcriptome of 50 different accession of wild barley, generating a rich library of natural variants of the MLA immune receptor. They found a pattern of diversification in the CC domain, which they argue might be related to functional diversification of these receptors. Furthermore, they detected positive selection signals in the LRR region of MLA, which is thought to confer recognition specificity to the pathogen.
The findings represent an excellent example of molecular evolution in plant NLRs and the new receptor variants uncovered by this study have the potential to help in the quest for durable resistance against different pathogen strains.
There are two parts in this study: (1) the generation of a sequence library of natural variants of the MLA gene and the analyses to identify signatures of selection and diversification; (2) a combination of secondary structure prediction and functional analyses of the CC domains to complement the first part.
Our general view is that the first part is interesting and has yielded exciting molecular evolution findings. Figure S2, for instance, is truly beautiful with MLA being fairly conserved in its general structure yet so diverse in terms of amino acid sequences. However, the second part would benefit from revising the methodology related to the structural analyses of the CC domains. Also, the functional analyses are limited to autoactivity assays of the CC-domains. It’s not surprising that all of the assayed domains trigger cell death given that the CC domains of wheat MLA-like genes Sr33 and Sr50, which group outside the two clades described here, are also autoactive. Given that all of the assayed domains are autoactive, and appear to be conserved across the broader MLA family including wheat homologs, it is difficult to accept several of the conclusions proposed in this manuscript. For example, the title (“subfamily-specific functionalization”) might be misleading given that no pathogen or effector related functional data on subfamily 2 has been reported.