Abstract

Extensive adenosine-to-inosine (A-to-I) editing of nuclear-transcribed RNAs is the hallmark of metazoan transcriptional regulation, and is fundamental to numerous biochemical processes. Here we explore the origin and evolution of this regulatory innovation, by quantifying its prevalence in 22 species that represent all major transitions in metazoan evolution. We provide substantial evidence that extensive RNA editing emerged in the common ancestor of extant metazoans. We find the frequency of RNA editing varies across taxa in a manner independent of metazoan complexity. Nevertheless, cis-acting features that guide A-to-I editing are under strong constraint across all metazoans. RNA editing seems to preserve an ancient mechanism for suppressing the more recently evolved repetitive elements, and is generally nonadaptive in protein-coding regions across metazoans, except for Drosophila and cephalopods. Interestingly, RNA editing preferentially target genes involved in neurotransmission, cellular communication and cytoskeleton, and recodes identical amino acid positions in several conserved genes across diverse taxa, emphasizing broad roles of RNA editing in cellular functions during metazoan evolution that have been previously underappreciated.