Although numerous studies have emphasized the role of microRNAs (miRNAs) in the control of many different cellular processes, they might also exert a profound effect on the macroevolution of animal body plans. It has been hypothesized that, because miRNAs increase genic precision and are continuously being added to metazoan genomes through geologic time, miRNAs might be instrumental for canalization of development and morphological evolution. Nonetheless, an outstanding question remains: how are new miRNAs constantly evolving? To address this question, we assessed the miRNA complements of four deuterostome species, chosen because of their sequenced genomes and well-resolved phylogeny. Our comparative analysis shows that each of these four species is characterized by a unique repertoire of miRNAs, with few instances of miRNA loss. Moreover, we find that almost half of the miRNAs identified in this study are located in intronic regions of protein coding genes, suggesting that new miRNAs might arise from intronic regions in a process we term intronic exaptation. We also show that miRNAs often occur within cotranscribed clusters, and describe the biological function of one of these conserved clusters, the miR-1/miR-133 cluster. Taken together, our work shows that miRNAs can easily emerge within already transcribed regions of DNA, whether it be introns or preexisting clusters of miRNAs and/or miRNAs and protein coding genes, and because of their regulatory roles, these novel players change the structure of gene regulatory networks, with potential macroevolutionary results.

Summary: Retinoic acid (RA) is a vitamin A-derived morphogen important for axial patterning and organ formation in developing vertebrates and invertebrate chordates (tunicates and cephalochordates). Recent analyses of genomic data have revealed that the molecular components of the RA signaling cascade are also present in other invertebrate groups, such as hemichordates and sea urchins. In this review, we reassess the evolutionary origins of the RA signaling pathway by examining the presence of key factors of this signaling cascade in different metazoan genomes and by comparing tissue-specific roles for RA during development of different animals. This discussion of genomic and developmental data suggests that RA signaling might have originated earlier in metazoan evolution than previously thought. On the basis of this hypothesis, we conclude by proposing a scenario for the evolution of RA functions during development, which highlights functional gains and lineage-specific losses during metazoan diversification. genesis 46:640-656,