
Image of extended ataxia network courtesy of Huda Y. Zoghbi, Baylor College of Medicine, Houston.
The game Six Degrees of Kevin Bacon is a study in network science. This game is a humorous take on the small world phenomenon, which proposes that any two people are separated by only a small number of acquaintances. Biological network analysis uses the same rationale to provide insight into large-scale genomics and proteomics data. Using this technique, Lim et al. report interactions between proteins involved in distinct Purkinje cell disorders with similar phenotypes in a recent article in Cell.
Ataxias are neurodegenerative disorders with similar phenotypes and dissimilar genotypes. All ataxias involve Purkinje cell degeneration and result in loss of coordination. Researchers have identified over 23 genes with mutations that are associated with different ataxias.
The authors used yeast two-hybrid screens to identify proteins that interact with ataxia-related proteins. Proteins involved in dominant and recessive ataxias as well as their paralogs (duplicate genes) and known binding partners comprised the 'ataxia-associated' bait. In the yeast two-hybrid fishing analogy, the bait proteins are used to catch their prey, or protein binding partners. Here, the prey was brain cDNA library and the human ORFeome, including the coding sequences and the full complement of splice variants for over 7000 genes. In all, the authors identified 770 protein–protein interactions between 42 ataxia-associated bait and 561 interacting prey. Roughly 96% of the interactions were previously unknown.
The authors validated a subset of their findings using coaffinity purification. They then did bioinformatic analysis using well defined terms from the Gene Ontology Consortium. Roughly 72% of the protein pairs identified in the yeast two-hybrid screen localized to the same cellular domain, and 98% of the protein pairs shared at least one Gene Ontology functional annotation.
To their original interactome, the authors then added interactions with ataxia-associated proteins reported in the literature and 'interologs,' proteins whose orthologs interact with ataxia-associated proteins in other species. In the 'expanded' dataset, the authors found 6972 protein–protein interactions between 3607 proteins. They reported that 18 of the 23 ataxia-causing proteins interacted directly or indirectly (via 1 intermediate protein) with each other.
The ataxia interactome showed a high degree of connectivity. The authors compared their interactome of proteins involved in phenotypically related ataxias to a control interactome of proteins involved in unrelated diseases. The mean shortest path length between proteins was 3.11 for the ataxia network and 3.67 for the control network. The authors also found more 'triples,' with two ataxia-associated proteins separated by one intermediate, interacting protein, in the ataxia relative to the control network.
The authors identified several proteins that interacted with many ataxia-associated proteins, forming 'hubs' in the network. Three of these proteins, RBM9, A2BP1 and RBPMS, regulate RNA binding and splicing. These data indicate that ataxias may be caused by RNA deregulation.
Together, these data suggest that although different mutations are responsible for each ataxia, their general etiology may be similar. Because dominant spinocerebellar ataxias are polyglutamine expansion diseases, perhaps proteins involved in other polyglutamine expansion diseases, including Huntington disease, may also show high connectivity with ataxia-associated proteins.
