Poster Presentation 2014 International Biophysics Congress

Improvements of a method to predict interfaces for GPCR oligomerization (#595)

Akira Saito 1 , Hiroyuki Toh 2 , Wataru Nemoto 1
  1. Tokyo Denki University, Saitama, Japan
  2. Computational Biology Research Center (CBRC), National Institute Advanced Industrial Science and Technology (AIST), Koto-ku, Tokyo, Japan

G-protein-coupled receptors (GPCRs) interact with one another to form a homo- and/or heterodimer or higher-order molecular complexes: oligomers. Several types of GPCR oligomers are associated with some diseases. For example, hetero-oligomerization of cannabinoid receptor 1 (Cbr1) and angiotensin receptor 1 (Agtr1) is considered to enhance angiotensin II-mediated signaling. The same process may be involved in the coupling of Agtr1 with multiple trimeric G-proteins in hepatic stellate cells from ethanol-treated rats where Cbr1 is upregulated. Therefore, the Cbr1–Agtr1 heterodimer in liver fibrosis that is caused by alcohol consumption is believed to be a novel target for antifibrotic compounds. In order to control GPCR oligomerization, identifications of the interfaces are informative. This knowledge can also facilitate peptide inhibition experiments and the development of guidelines for mutation experiments that are designed to elucidate the functional differences between a monomer and oligomers. However, identifications of those interfaces experimentally are difficult challenges. Therefore, accurate prediction of the residues for the oligomerization would increase our understanding of signal transduction and GPCR-associated diseases. Many methods for the prediction of interfaces of soluble proteins have been proposed, but these methods cannot be applied to the interfaces for the GPCR oligomerizations. Hence, we developed a method to predict interfaces for GPCR oligomerization: GRIP (Nemoto & Toh, Proteins 2005). In our present study, execution speed and prediction accuracy of GRIP were improved. On the new server, a user can execute GRIP using only structure data in a new format, whereas the old server required both structure and sequence data.