Evolution of the main skeleton-forming genes in sponges (phylum Porifera) with special focus on the marine Haplosclerida (class Demospongiae).

The skeletons of sponges (Phylum Porifera) are comprised of collagen, often embedded with small siliceous structures (spicules) arranged in various forms to provide strength and flexibility. The main proteins responsible for the formation of the spicules in demosponges are the silicateins, which are related to the cathepsins L of other animals. While the silicatein active site, necessary for the formation of biosilica crystals, is characterized by the amino acids SHN, different variants of the silicatein genes have been found, some that retain SHN at the active site and some that don't. As part of an effort to further understand skeleton formation in marine sponges of the order Haplosclerida, a search for all silicatein variants were made in Irish species representing the main clades of this large sponge group. For this task, transcriptomes were sequenced and de novo assembled from Haliclona oculata, H. simulans and H. indistincta. Silicatein genes were identified from these and all available genomes and transcriptomes from Porifera. These were analysed along with all complete silicateins from GenBank. Silicateins were only found in species belonging to the class Demospongiae but excluding Keratosa and Verongimorpha and there was significant duplication and diversity of these genes. Silicateins showing SHN at the active site were polyphyletic. Indeed silicatein sequences were divided into six major clades (CHNI, CHNII, CHNIII, SHNI, SHNII and C/SQN). In those clades where haplosclerids were well represented the silicatein phylogeny reflected previous ribosomal and mitochondrial topologies. The most basal silicatein clade (CHNI) contained sequences only from marine haplosclerids and freshwater sponges while one silicatein from H. indistincta was more related to cathepsins L (outgroup) than to the overall silicatein clade indicating the presence of an old silicatein or an intermediary form. This data could suggest that marine haplosclerids were one of the first groups of extant demosponges to acquire silicatein genes. Furthermore, we suggest that the paucity of spicule types in this group may be due to their single copy of SHNI variants, and the lack of a silintaphin gene.