Molecular Simulations and Drug Discovery of Adenosine Receptors

Abstract

G protein-coupled receptors (GPCRs) represent the largest family of human membrane proteins. Four subtypes of adenosine receptors (ARs), the A₁AR, A_2AAR, A_2BAR and A₃AR, each with a unique pharmacological profile and distribution within the tissues in the human body, mediate many physiological functions and serve as critical drug targets for treating numerous human diseases including cancer, neuropathic pain, cardiac ischemia, stroke and diabetes. The A₁AR and A₃AR preferentially couple to the G_i/o proteins, while the A_2AAR and A_2BAR prefer coupling to the G_s proteins. Adenosine receptors were the first subclass of GPCRs that had experimental structures determined in complex with distinct G proteins. Here, we will review recent studies in molecular simulations and computer-aided drug discovery of the adenosine receptors and also highlight their future research opportunities.

Keywords: G protein-coupled receptors; adenosine receptors; drug discovery; mechanisms; molecular simulations.

Figure 1

Comparison of experimental structures of the inactive antagonist-bound (PDB: 5UEN) and active agonist-G protein-bound (PDB: 6D9H) A₁AR. (A) Outward movement of transmembrane helix 6 (TM6) in the active agonist-bound receptor (red) induces opening of the intracellular pocket for binding with G protein as compared to the inactive antagonist-bound conformation of the receptor (blue). (B) The receptor activation is accompanied by adjustments of the ligand binding pocket, extracellular loops (ECLs) and the W^6.48 “rotamer toggle switch”. Antagonist DU172 (blue) and agonist adenosine (ADO, red) occupy different regions of the orthosteric pocket in the inactive and active conformations of the A₁AR receptor, respectively. (C) “Microswitches” play a critical role in the activation of adenosine receptors, including the R^3.50-E^6.30 salt bridge and the Y^7.53 “tyrosine toggle switch”.

Figure 2

GaMD simulations revealed mechanism of specific G protein coupling to adenosine receptors: 2D potential of mean force (PMF) profiles of the (A) A₁AR-G_i, (B) A_2AAR-G_s, (C) A_2AAR-G_i and (D) A₁AR-G_s complex systems regarding the agonist RMSD relative to the cryo-EM conformation and AR:NPxxY-G:α5 distance. The white triangles indicate the cryo-EM or simulation starting structures. Summary of specific AR-G protein interactions: (E) the ADO-bound A₁AR prefers to bind the G_i protein to the G_s. The latter could not stabilize agonist ADO binding in the A₁AR and tended to dissociate from the receptor. (F) The A_2AAR could bind both the G_s and G_i proteins, which adopted distinct conformations in the complexes. Adapted from reference [99] with permission from American Chemical Society. Further permissions related to the material excerpted should be directed to the American Chemical Society.

Figure 3

The “allosteric” binding sites (pocket 1 and pocket 2) in the A₁AR. The cryo-EM structure of A₁AR–Gi2 complex bound by the PAM MIPS521 (pocket 2) was shown. Another allosteric binding site (pocket 1) was suggested by mutation experiments and MD simulations.

Figure 4

MIPS521 PAM molecule stabilized the ADO agonist binding and the A₁AR-G_i2 protein complex, as well as slowing deactivation of A₁AR upon removal of the G_i2 protein. RMSD (Å) of ADO from GaMD simulations completed in the absence (a) or presence (b) of MIPS521, G_i2 (c) or both (d). (e–h) Distance between the intracellular ends of TM3 and TM6 (measured as the distance in Å between charge centers of residues R^3.50 and E^6.30) in the absence (e) or presence (f) of MIPS521, G_i2 (g) or both (h). Each condition represents three GaMD simulations, with each simulation trace displayed in a different color (black, red, blue). The lines depict the running average over 2 ns. Reprinted from reference [39] with permission from Springer Nature.

Figure 5

Binding and dissociation pathways of caffeine (CFF) from the A_2AAR were revealed from GaMD simulations. (A–C) Time courses of the N6.55:ND2-CFF:N1 distance calculated from GaMD equilibration, Sim2 and Sim3 GaMD production simulations. (D–F) Trace of CFF (orange and red) in the A_2AAR observed in the GaMD equilibration, Sim2 and Sim3 GaMD production simulations. The seven transmembrane helices are labeled I to VII, and extracellular loops 1–3 are labeled ECL1–ECL3. Reprinted from reference [123] with permission from Frontiers.

Figure 6

Differentiated order parameters of lipid molecules were found in simulation systems of the inactive and active A₁AR: (A) inactive A₁AR using dihedral-boost GaMD, (B) active A₁AR using dihedral-boost GaMD, (C) inactive A₁AR using dual-boost GaMD and (D) active A₁AR using dual-boost GaMD. Red diamond lines represent the average -S_CD order parameters for the cytoplasmic lower leaflet and blue diamond lines for the extracellular upper leaflet. Reprinted from reference [128] with permission from Wiley.

Figure 7

Overview flowchart for retrospective docking of positive allosteric modulators (PAMs) in the A₁AR: Starting from the cryo-EM structure of the active ADO-Gi-bound A₁AR (6D9H) and docking model of PAM VCP171-bound A₁AR (ADO-A₁AR-Gi-VCP171), GaMD simulations were carried out to construct structural ensembles to account for the receptor flexibility. Meanwhile, a compound library was prepared for 25 known PAMs of the A₁AR and 2475 decoys obtained from the DUD-E with openbabel 2.4.1. Ensemble docking was then performed to identify the PAMs for which the AUC and enrichment factors were calculated to evaluate docking performance. Both rigid-body and flexible docking were tested using AutoDock. Reprinted from reference [161] with permission from Elsevier.