Discrete approaches for leveraging g protein-coupled receptors as therapeutic tools

Abstract

Listen

Translate article

<p dir="ltr">G protein-coupled receptors (GPCRs) represent the largest and most therapeutically valuable family of human membrane proteins, serving as targets for over a third of FDA- approved drugs. Despite their clinical success, detailed mechanistic, structural, and interactome-level knowledge of GPCRs remains limited. This thesis uses multiple approaches to explore these aspects of GPCR biology, with a focus on brain-expressed aminergic GPCRs.</p><p dir="ltr">First, G protein recruitment and G protein inwardly rectifying K+ (GIRK) channel activation assays were employed to investigate Ulotaront, a novel antipsychotic, at three aminergic receptors: the trace amine-associated receptor 1 (TAAR1), the dopamine D2 receptor (D2R), and the serotonin receptor 1A (5-HT1AR). Ulotaront was found to be a full agonist at TAAR1, with modest, a partial, weaker agonist at 5-HT1AR and a weak partial agonist with low potency at D2R. Using TAAR1- knockout (KO) mice, we demonstrated that TAAR1 is required for Ulotaront's effects on lowering core body temperature (CBT), baseline locomotion, and reversing MK-801-induced pre-pulse inhibition (PPI) disruption.</p><p dir="ltr">Virtual screening is a powerful method to discover GPCR ligands and typically requires high-resolution protein structures. Recent advances in artificial intelligence (AI)-based protein structure prediction prompted us to evaluate the performance of AlphaFold2 and homology modelling in virtual screening to discover new TAAR1 ligands. After enriching TAAR1 models generated by both methods, docking screens identified 25 TAAR1 agonists, which were evaluated using a TAAR1-Gas recruitment assay. The hit- rate achieved with the AlphaFold screen was more than double that of the homology model screen, and the most potent agonist discovered was further tested for selectivity. The lead compound from the AlphaFold screen showed favourable pharmacokinetics and TAAR1-dependent antipsychotic-like effects in mouse behavioural assays.</p><p dir="ltr">While virtual docking has proven successful in the identification of GPCR ligands, computational approaches to predict quantitative ligand parameters are less established. To address this, molecular dynamics (MD) simulations, G protein recruitment, and radioligand competition assays were conducted to improve our understanding of D2R activation mechanics. This led to the development of a computational pipeline that accurately predicted D2R ligand efficacy and affinity (R2: 0.82 and 0.7 goodness-of-fit with experimental results). These studies also provided insights into specific residues involved in ligand specificity at D2R and B2AR.</p><p dir="ltr">Finally, we systematically explored the interactomes of human GPCRs with accessory proteins using state-of-the-art suspension-bead arrays (SBAs), focusing on receptor-activity modifying proteins (RAMPs) and p11. The RAMP screen revealed over 50 new GPCR-RAMP interactions and validated endogenous GPCR-RAMP complexes in SK-N- MC cells using quantitative in situ proximity assays. Similarly, the p11 screen identified over 20 GPCR-p11 interactions, showing that these complexes increased p11 protein levels and that p11 inhibited TAAR1 expression and signalling.</p><p dir="ltr">In conclusion, this thesis uses a variety of strategies to improve our understanding of GPCR biology. The findings clarify how existing GPCR-targeted drugs work, present methods for identifying and optimizing GPCR ligands, and highlight the role of accessory proteins in modulating GPCR responses, potentially leading to new directions in research and drug discovery.</p><h3>List of scientific papers</h3><p dir="ltr">I. <b>Saarinen M#</b>, Mantas I, Flais I, Ågren R, Sahlholm K, Millan MJ, Svenningsson P#. TAAR1 dependent and independent actions of the potential antipsychotic and dual TAAR1/5-HT1A receptor agonist SEP- 363856. Neuropsychopharmacology. 2022 Dec;47(13):2319-2329. <a href="https://doi.org/10.1038/s41386-022-01421-2" rel="noreferrer" target="_blank">https://doi.org/10.1038/s41386-022-01421-2</a></p><p dir="ltr">II. Díaz-Holguín A*, <b>Saarinen M*</b>, Vo DD, Sturchio A, Branzell N, Cabeza de Vaca I, Hu H, Mitjavila-Domenech N, Lindqvist A, Baranczewski P, Millan MJ, Yang Y, Carlsson J#, Svenningsson P#. AlphaFold accelerated discovery of psychotropic agonists targeting the trace amine-associated receptor 1. Sci Adv. 2024 Aug 9. <a href="https://doi.org/10.1126/sciadv.adn1524" rel="noreferrer" target="_blank">https://doi.org/10.1126/sciadv.adn1524</a></p><p dir="ltr">III. Chen Y*, Kahlous NA*, <b>Saarinen M,</b> Pérez-Conesa S, Svenningsson P, Delemotte L#, Carlsson J#. Revealing molecular determinants of ligand efficacy and affinity at the D2 dopamine receptor through molecular dynamics simulations. [Manuscript]</p><p dir="ltr">IV. Kotliar IB*, Bendes A*, Dahl L*, Chen Y, <b>Saarinen M,</b> Ceraudo E, Dodig- Crnković T, Uhlen M, Svenningsson P, Schwenk JM#, Sakmar TP#. Multiplexed mapping of the interactome of GPCRs with receptor activity-modifying proteins. Sci Adv. 2024 Aug 2. <a href="https://doi.org/10.1126/sciadv.ado9959" rel="noreferrer" target="_blank">https://doi.org/10.1126/sciadv.ado9959</a></p><p dir="ltr">V. <b>Saarinen M#</b>, Kotliar IB, Hoffkes I, Bowin CF, Bendes A, Dahl L, Glaros V, Abney A, Bendes A, Kreslavsky T, Schwenk JM, Sakmar TP, Svenningsson P#. Receptor activity drives the GPCR-p11 interactome. [Manuscript]</p><p dir="ltr">*Equal contribution # Corresponding author</p>