Evolutionary association of receptor-wide amino acids with G protein-coupling selectivity in aminergic GPCRs

Abstract

G protein-coupled receptors (GPCRs) induce signal transduction pathways through coupling to four main subtypes of G proteins (G_s, G_i, G_q, and G_12/13), selectively. However, G protein selective activation mechanisms and residual determinants in GPCRs have remained obscure. Herein, we performed extensive phylogenetic analysis and identified specifically conserved residues for the aminergic receptors having similar coupling profiles. By integrating our methodology of differential evolutionary conservation of G protein-specific amino acids with structural analyses, we identified specific activation networks for G_s, G_i1, G_o, and G_q To validate that these networks could determine coupling selectivity we further analyzed G_s-specific activation network and its association with G_s selectivity. Through molecular dynamics simulations, we showed that previously uncharacterized Glycine at position 7x41 plays an important role in receptor activation and it may determine G_s coupling selectivity by facilitating a larger TM6 movement. Finally, we gathered our results into a comprehensive model of G protein selectivity called "sequential switches of activation" describing three main molecular switches controlling GPCR activation: ligand binding, G protein selective activation mechanisms, and G protein contact.

Figure 1.. Selectivity-determining residues for each Gα subtype.

(A) The formula for specific residue identification. (B) The schema describes the comparisons between paralogous human receptors to find the specifically conserved residues for each Gα. Arrows represent a single comparison. (C) The distribution of specifically conserved residues for each Gα subtype and hierarchical clustering of them (complete linkage). (D) Possible variants of G_s specific residues that are observed in non-coupler receptors are compared with the WT activity score. (E) Maximum-likelihood phylogenetic tree of aminergic receptors including coupling profiles, conservation information of selected specifically conserved residues (I, Inoue; A, Avet), The background color scale for each consensus amino acid correlates with their conservation (identity).

Figure 2.. Structural analysis of molecular pathways that are observed upon coupling with a heteromeric G protein complex.

(A) The most common molecular signal transduction pathways from ligand-binding pocket to G protein–coupling interface. The arrows represent a contact change upon coupling to a G protein. The network is summarized and divided into different layers based on their functional relevance. (B) Projection of main chains of specifically conserved and consensus residues in different layers of activation on inactive ADRB2 structure (PDB ID 2RH1). (C) The distribution of specifically conserved residues for each analyzed Gα subtype.

Figure 3.. Specific activation networks for G_s, G_i1, G_o and G_q.

(A) TM6 tilt comparison between the active receptors we used. Red: G_s couplers, Orange: G_o, G_i1, and G_q, Blue: 5HT1B G_o coupler as an exception. (B) Interactions within the receptor that are specific (P < 0.01) to G_s. Red: increasing contact, blue: decreasing contact, orange circle: present in common activation mechanism, red fill: uniquely identified specific residue for G_s, grey fill: G_α specific residue. The width of the lines correlates with statistical significance. A group of residues that possibly facilitate in TM6 movement for G_s coupling was shown on inactive (blue) and active (red) structures. (C, D, E) Specific interaction networks for G_i1, G_o, and G_q. P < 0.1 is used for G_i1. *: This interaction is identified only if 5HT1B is neglected from the comparison because of its larger TM6 movement.

Figure 4.. Analysis of molecular dynamics simulations reveals functional importance of glycine at 7x41.

(A) Four different MD simulation systems were shown in their initial conformation. (B) For each simulation, distribution of frames with respect to their state of activation was shown, distance in Angstrom. (C) The common pathway representing the impact of the mutations at 7x41. (D, E) The common pathway was represented on average structures that were obtained in all MD trajectories for every mutation and WT. The movements of residues were represented with arrows.

Figure 5.. Sequential switches of activation model for G protein selectivity.

The model describes that all switches in different layers of receptors must be turned off for receptor activation and coupling of the G protein. If switches at upper layers are halted due to a mutation, the following switches become turned off which inhibits G protein coupling eventually.

Figure S1.. Selective pathways for activation containing all conserved residues.

This network contains complete list of conserved residues for each layer we demonstrated in Fig 2B. Each layer from top to bottom is represented with a different color. Some of the nodes are lack an edge due to the filtration step that we applied based on frequency of the information change between two residues.