Rapid divergence and high diversity of miRNAs and miRNA targets in the Camelineae
Lisa M. Smith, Hernan A. Burbano, Xi Wang, Joffrey Fitz, George Wang, Yonca Ural-Blimke and Detlef Weigel
Department of Animal and Plant Sciences, University of Sheffield, Western Bank, Sheffield S10 2TN, UK
Department of Molecular Biology, Max Planck Institute for Developmental Biology, Spemannstrasse 35, 72076 Tubingen
This paper is from the most recent issue of The Plant Journal, and I thought it made some rather interesting points. The paper focused on small RNA seq of several tissues from Capsella rubella, a member of the Camelineae tribe and frequent outgroup to the Arabidopsis genus. With sRNA annotations from A. thaliana and A. lyrata, the authors look at the evolution of miRNAs within closely related species.
First of all, I was interested in the bioinformatics suite that this group chose to perform their annotation. After aligning unique reads with bowtie, they used several clustering softwares (miR-deep 1.3, DSAP, UEA sRNA toolkit), resulting in some wildly different loci and annotations. Within figure 1c, it appears that only half of known miRNA loci annotated by DSAP or miR-deep are found in common with each other, though this is higher when looking at miRNA families. Is this because of failings within these softwares (the authors mention a high false negative rate in miR-deep)?
The article goes on to look at the variation in miRNAs in relation to their target genes between the 3 species. They found that unique miRNA-target pairs were highly species divergent, with most pairings being unique to the different species. Of the non-divergent pairs, almost all are more ancient pairings that are present outside of brassicaceae, leading the authors to hypothesize that there are two differentially evolving subsets of miRNAs: “young, evolutionarily dynamic miRNAs, and older miRNAs with a conserved subset of mRNA targets”.
The authors go on to look at the levels of polymorphism in miRNAs and their targets throughout A. thaliana. This analysis ultimately lead to higher mutation rate in miRNA sequences themselves, forcing the authors to conclude that the target sites are undergoing stronger selection. I thought this was a bit confusing, as you might expect higher conservation in target sites which could be in the CDS of genes, a point mentioned by the authors but not elaborated upon.