MULTIPLE TEMPLATE BASED HOMOLOGY MODEL OF ACTIVE STATE D2 DOPAMINE RECEPTOR

LOBON, IGNACIO EXEQUIEL; BOGADO, MARIA LUCRECIA; ANGELINA, EMILIO; LUCHI, ADRIANO; SOSA, GLADYS LAURA; PERUCHENA, NELIDA

Buenos Aires

Congreso; Reunión Conjunta de Sociedades de Biociencias; 2017

Dopamine is an essential neurotransmitter in the central nervoussystem and exerts its effects through the activation of five subtypesof G protein coupled receptors (D1 to D5). Among subtypes, D2 hashigh therapeutic relevance for treatment of Parkinson?s disease,schizophrenia and other disorders in the central nervous system.While D2 structure has not been solved yet, an homology model(HM) of D2 based on the known structure of the closely related D3was reported. In the solved structure D3 has been captured in itsinactive state, therefore the resulting models also would be in theinactive state.In this work we constructed an HM of D2 in the activated, G proteincoupled state. To build the model two templates were considered:D3 (PDB-ID: 3PBL) due to its high degree of sequence identity toD2 (78% in the TM helices) and the fully activated b 2 adrenergicreceptor (PDB-ID: 3SN6) to model mostly the intracellular region ofthe receptor in the activated, G protein coupled state.Sequence of D2 in its long form (NP_000786.1) and of Gi proteininhibitory alpha subunit (PDB-ID: 1GP2) were aligned with the tem-plate sequences. The alignment was then used to construct several3-D models of D2 with the program Modeller 9.15. The model withthe lowest value of the Modeller objective function was subjected toquality assessment with PROCHECK. The model was then insertedinto an heterogeneous biological membrane using CHARMM-GUIserver and subjected to MD simulations with Amber14 for furtherrefinement, in order to get a 3-D modelcloser to the native form ofthe protein.HM based on a single model requires the choice of a crystallizedstructure highly similar to the receptor under study, while the strat-egy used in this work, of alignment with multiple models, involvesthe excision of the receptor in several domains and the subsequentselection of the most appropriate template for each of these do-mains, this strategy being very useful to increase the accuracy ofthe models.