The dispanins: a novel gene family of ancient origin that contains 14 human members

Abstract

The Interferon induced transmembrane proteins (IFITM) are a family of transmembrane proteins that is known to inhibit cell invasion of viruses such as HIV-1 and influenza. We show that the IFITM genes are a subfamily in a larger family of transmembrane (TM) proteins that we call Dispanins, which refers to a common 2TM structure. We mined the Dispanins in 36 eukaryotic species, covering all major eukaryotic groups, and investigated their evolutionary history using Bayesian and maximum likelihood approaches to infer a phylogenetic tree. We identified ten human genes that together with the known IFITM genes form the Dispanin family. We show that the Dispanins first emerged in eukaryotes in a common ancestor of choanoflagellates and metazoa, and that the family later expanded in vertebrates where it forms four subfamilies (A-D). Interestingly, we also find that the family is found in several different phyla of bacteria and propose that it was horizontally transferred to eukaryotes from bacteria in the common ancestor of choanoflagellates and metazoa. The bacterial and eukaryotic sequences have a considerably conserved protein structure. In conclusion, we introduce a novel family, the Dispanins, together with a nomenclature based on the evolutionary origin.

Figure 1. The phylogeny of the vertebrate Dispanins.

The phylogenetic tree shows the propsed hierarchy of the eukaryotic Dispanin family. The four subfamilies (A–D) are marked with colors. Node support values are given as posterior probabilities (all nodes >0.5) and corresponding bootstrap values are given in parenthesis in percentage of 1000 bootstraps. Human sequences and sequences discussed in the manuscript are written in bold.

Figure 2. The evolutionary history of the Dispanins.

The figure shows a schematic representation of the evolutionary history of the Dispanin family and its subfamilies (DSPA–D) and is based on the mining, phylogenetic analysis and evolutionary relationship of the species. The numbers at the branch ends represents the number of genes within each subfamily.

Figure 3. The protein features and topology of the Dispanin subfamilies.

The picture shows the membrane topology and sequences features of a representative human member of each subfamily. Conserved motifs and residues are shown and those which have a sequence identity of more than 90% are framed in black and those with 80–90% sequence similarity are framed in blue. Predicted phosphorylation (Green) and glycosylation (orange) sites are shown.

Figure 4. Multiple sequence alignment of the Dispanins.

The multiple sequence alignment shows the conserved regions around the two transmembrane helices in the Dispanins. Consensus protein sequences of each vertebrate subfamily together with invertebrates, M. brevicollis (Mb) and a bacterial consensus sequences are included Conserved residues are coloured according to the Zappo scheme in Jalview and well conserved motifs are marked with a square. The conserved splice site between the two transmembrane helices is marked with a red line.