Moreover, our study raises the possibility that in those vertebrates that kept Dact4, this protein may inhibit the function of the other Dacts. Our study provides the basis for structural and molecular biologists to systematically test the function of the shared and divergent Dact protein motifs, and for cell and developmental sellekchem biologists to explore the combinatorial aspects of Dact function. Methods Database searches Genomes of humans, mouse, cattle, dog, African elephant, opossum, platypus, chicken, turkey, zebrafinch, duck, budgerigar, Anole lizard, Western painted turtle, Chinese soft shield turtle, Xenopus tropicalis, coelacanth, spotted gar, zebrafish, Atlantic cod, Medaka, Fugu, Tetraodon, stickleback, Nile Tilapia, Southern platyfish, sea lamprey, Ciona intestinalis, Ciona savignyi, Drosophila melanogaster, Caenorhabditis elegans, Saccharomyces cerevisiae were searched using the Ensembl browser.

Genomes of the Burmese python, Oikopleura dioica, Branchiostoma floridae, Saccoglossus kowalevskii, Strongylocentrotus purpuratus, Aplysia californica, Tribolium castaneum, Bombyx mori,Caenorhabditis briggsae, Loa loa and of the groups Kinetoplastida including Trypanosoma and Fungi were searched using the NCBI browser. The genomes of the elephant shark and the Japanese lamprey were searched at the respective genome pro ject portals. EST data bases for the above species and for Xenopus laevis, and for the taxonomical groups lungfish, chondrosts, holosts, teleosts, chondrichthyans, cyclostomes, ascidians, proto stomes and for protists were performed, using the NCBI browser.

The first round of TBLASTN searches were per formed using the human and mouse Dact1,2,3. chicken Dact1,2. Xenopus laevis dact1a,1b and zebrafish dact1,2 protein sequences as queries. Subsequently, we also used the newly identified zebrafish, lizard and turtle Dact3 and Dact4 sequences, the lamprey and the Branchiostoma sequences. Moreover, we performed searches with protein sequences encoded by individual exons and with protein motifs. Fgenesh was used to predict the exon struc ture for sequences where no annotation was available. Molecular phylogenetic analyses For molecular phylogenetic analyses, protein sequences were aligned using ClustalW and T Coffee. The alignment was optimized manually using BioEdit, using information from pairwise alignments and the position of functionally significant amino acids.

The resulting alignment had large gaps, and many regions outside identifiable conserved motifs could not be aligned unambiguously. Using the automated1 and strict settings of trimAl as a AV-951 guide, non significant residues were removed manually. The most suitable evolution model for the alignment was determined by using ProtTest3 as JTT G F. The JTT model was used in all subsequent analyses. Phylogenetic tree reconstruction was carried out employing a variety of methods.