GPCR activation and GRK2 assembly by a biased intracellular agonist

Abstract

Phosphorylation of G protein-coupled receptors (GPCR) by GPCR kinases (GRKs) desensitizes G protein signaling and promotes arrestin signaling, which is also modulated by biased ligands 1-6 . Molecular assembly of GRKs to GPCRs and the basis of GRK-mediated biased signaling remain largely unknown due to the weak GPCR-GRK interactions. Here we report the complex structure of neurotensin receptor 1 (NTSR1) bound to GRK2, Gαq, and an arrestin-biased ligand, SBI-553 7 , at a resolution of 2.92 Å. The high-quality density map reveals the clear arrangement of the intact GRK2 with the receptor, with the N-terminal helix of GRK2 docking into the open cytoplasmic pocket formed by the outward movement of the receptor TM6, analogous of the binding of G protein to the receptor. Strikingly, the arrestin-biased ligand is found at the interface between GRK2 and NTSR1 to enhance GRK2 binding. The binding mode of the biased ligand is compatible with arrestin binding but is clashed with the binding of a G protein, thus provide an unambiguous mechanism for its arrestin-biased signaling capability. Together, our structure provides a solid model for understanding the details of GPCR-GRK interactions and biased signaling.GPCRs comprise the largest family of cell surface receptors whose signaling is primarily mediated by two types of downstream effectors: G-proteins and arrestins.The switch of GPCR signaling from G-protein pathways to arrestin pathways is controlled by a small family of GPCR kinases, GRKs, which phosphorylate either the receptor C-terminal tail or the third intracellular loop (ICL3) 1-3 . Phosphorylation of GPCRs promotes recruitment of arrestin, which blocks G-protein binding and desensitizes G-protein signaling 3 . Because drugs that selectively activate either Gprotein pathways or arrestin pathways (biased signaling) are proposed to have better therapeutic and safety index 4,5 , the mechanism of GPCR biased signaling has been a subject of intensive research over the past two decades.

Biased Signaling in Mutated Variants of β2-Adrenergic Receptor: Insights from Molecular Dynamics Simulations

“…Furthermore, ICL2 and ICL3 that were known to interact with G protein and β-arrestin, ,,, have also been shown to involve in GRK binding using an integrated approach of cross-linking, hydrogen–deuterium exchange mass spectrometry (MS), and several other experiments . We also note that the H8 is involved in receptor–kinase interactions of rhodopsin and neurotensin receptor 1 . In a cryo-EM study of the rhodopsin–GRK1 complex, the interactions of H8 residues at positions 8.48 and 8.51 with a serine residue in GRK1 (S5) were suggested to play a critical role in the phosphorylation of class A GPCRs including β 2 AR by GRKs .…”

“…99 We also note that the H8 is involved in receptor−kinase interactions of rhodopsin 98 and neurotensin receptor 1. 108 In a cryo-EM study of the rhodopsin−GRK1 complex, the interactions of H8 residues at positions 8.48 and 8.51 with a serine residue in GRK1 (S5) were suggested to play a critical role in the phosphorylation of class A GPCRs including β 2 AR by GRKs. 98 In our simulations, the residue P330 8.48 shows a pronounced reduction in n SOP for B2RY compared with B2RWT, consistent with its inability to bind GRK.…”

“…In addition to the TM helices and H8, intracellular loop 3 (ICL3) that connects TM5 and TM6 is reportedly involved in transducer binding to GPCRs and biased signaling. ,,,,− A recent study on β 2 AR using fluorescent resonance energy transfer (FRET) experiments and MD simulations has demonstrated that ICL3 acts as an autoregulator of GPCR signaling by sampling conformations, which are in a dynamic equilibrium between states that block and expose the transducer-binding cavity . We construct a two-dimensional potential of mean force (PMF) plot as a function of the ICL3–H8 distance (L258 ICL3 C α –L340 8.58 C α ) that measures the opening of the intracellular cavity and TM6 displacement from the inactive β 2 AR to characterize the conformational states (Figures a–c).…”

Development of a Fluorescent Ligand for the Intracellular Allosteric Binding Site of the Neurotensin Receptor 1

“…Design of fluorescent NTSR1 ligand 14 . (A) Docking of 3 (light green) to the NTSR1 cryo-EM structure (PDB 8JPF) resulted in a binding pose highly similar to the experimentally determined binding mode of 1 (beige) close to the intracellular opening of the receptor. (B) According to docking, an enlargement of the ethanolamine-substituent as in 8 (cyan) can be well accommodated by NTSR1.…”

“…In addition to the functional in vitro investigations of the modified quinazolines, we performed docking experiments with a recently released NTSR1 cryo-EM structure in complex with 1 to further guide the design of our fluorescent molecular probe. Docking of our internal reference compound 3 into the human NTSR1 obtained from the complex with GRK2 and Gq (PDB 8JPF, Figure A) resulted in a highly similar pose compared to the experimentally observed binding mode of 1 , with the quinazoline’s hydroxyethylamino-substituent pointing toward the intracellular opening of the receptor. In agreement with the functional data, the enlargement of this side chain by an additional hydroxyethyl unit as in 8 can be well accommodated by the receptor (Figure B).…”