Conservation and Evolutionary Divergence in the Activity of Receptor-regulated Smads

Abstract

Activity of the TGFΒ pathway is essential to the establishment of body axes and tissue differentiation in bilaterians. Orthologs for core pathway members have been found in all metazoans. Uncertain homology of the body axes and tissues patterned by this pathway raises questions about the activities of these molecules across the metazoan tree. We focus on the principal canonical transduction proteins (R-Smads) of the TGFΒ pathway, which instruct both axial patterning and tissue differentiation in the developing embryo. We compare the activity of R-Smads from a cnidarian (Nematostella vectensis), an arthropod (Drosophila melanogaster), and a vertebrate (Xenopus laevis) in Xenopus embryonic assays. NvSmad1/5 ventralized Xenopus embryos when expressed in dorsal blastomeres, similar to the effects of XSmad1. However, NvSmad1/5 was less potent than XSmad1 in its ability to activate downstream target genes in Xenopus animal cap assays. NvSmad2/3 strongly induced general mesendoderm markers, but weakly induced genes involved in specifying the Spemann organizer. Furthermore, NvSmad2/3 was unable to induce a secondary axis in Xenopus embryos, whereas the orthologs from Xenopus (XSmad2 and XSmad2) and Drosophila (dSmad2) were capable of doing so. Replacement of the NvSmad2/3 MH2 domain with the Xenopus counterpart led to a slight increase in inductive capability, but it could not generate a secondary body axis. We conclude that the activities of Smad1/5 orthologs have been largely conserved across Metazoa, but the activity of Smad2/3 orthologs has undergone more evolutionary divergence. Given the high level of sequence identity among R-Smad orthologs, we compared the protein sequences of Smad2/3 orthologs from 30 different metazoan taxa to locate regions of variation among taxa. Functional regions showed striking conservation, with most of the amino acid variation located in regions that are not well-described in the literature at present. We recommend further chimeric and mutagenic experimentation with Smad2/3 and present candidate sites. Our data demonstrate that large-scale morphological variation can be caused by fine-scale molecular divergence.