G-protein-coupled receptors (GPCR) represent the largest and most diverse receptor family and the most common target of currently marketed therapeutics. Common drugs targeting GPCRs include ranitidine (an H2 receptor antagonist), albuterol (a β2-adrenergic receptor antagonist), atenolol (a β1-adrenergic receptor antagonist), ondansetron (a 5HT3 receptor antagonist), and caffeine (an adenosine receptor antagonist). Because of their importance in biology and medicine, high-resolution analysis of the structures of GPCRs is needed, but such information has been slow to materialize because of technical problems affecting expression and crystallization of the proteins. After initial success in solving the structure of bovine rhodopsin in 2000,1 there was a seven-year hiatus before the structure of the human β2-adrenergic receptor was solved.2 Over the past five years, however, advances in technology combined with the concerted efforts of leading structural laboratories have led to the solution of the structure of over a dozen GPCRs with the solution of many more on the way. The latest success story is that of protease-activated receptor 1 (PAR1).
PAR1 is an unusual GPCR. The PAR1 structure (blue) is superimposed on the structure of the ß2 adrenergic receptor (pink) or the M2 muscarinic receptor (yellow). The ligand-binding pocket is substantially shallower in PAR1, as demonstrated by the binding of vorapaxar (green) on the surface of PAR1. By comparison, binding of the inverse agonist carazolol (magenta) on the ß2 adrenergic receptor or the antagonist QNB (orange) on the M2 muscarinic receptor occurs in the deep ligand binding pockets of these GPCRs.Adapted by permission from Macmillan Publishers Ltd: Nature (Zhang et al. Nature. 492:387- 392), Copyright 2012.
PAR1 is an unusual GPCR. The PAR1 structure (blue) is superimposed on the structure of the ß2 adrenergic receptor (pink) or the M2 muscarinic receptor (yellow). The ligand-binding pocket is substantially shallower in PAR1, as demonstrated by the binding of vorapaxar (green) on the surface of PAR1. By comparison, binding of the inverse agonist carazolol (magenta) on the ß2 adrenergic receptor or the antagonist QNB (orange) on the M2 muscarinic receptor occurs in the deep ligand binding pockets of these GPCRs.Adapted by permission from Macmillan Publishers Ltd: Nature (Zhang et al. Nature. 492:387- 392), Copyright 2012.
PAR1 is a widely expressed receptor that responds with high sensitivity to protease activity in the extracellular environment. The activation mechanism is unusual in that it involves an intramolecular process in which thrombin (or other thrombin-like proteases) cleaves an N-terminal proteolytic site, generating a tethered peptide that interacts with the extracellular face of the receptor (thus both receptor and ligand are part of the same structure). Binding of the tethered ligand causes a conformational change in PAR1 that results in the activation of the heterotrimeric G protein coupled complex on the cytoplasmic surface of the plasma membrane. PAR1 is the most abundant GPCR on platelets and connects, through thrombin activation, the coagulation cascade to platelet-mediated thrombus formation. As a result of its importance in thrombosis, PAR1 has been the target of several drug development programs focused on generation of novel antiplatelet agents. The most advanced compound in clinical development is vorapaxar, which is a highly specific PAR1 inhibitor. In phase III trials, vorapaxar was found to be protective against recurrent myocardial infarction, but at the cost of increased bleeding, including intracranial hemorrhage.3 One key to solving the structure of GPCRs is identifying small molecules capable of stabilizing the tertiary structure of the receptors. Vorapaxar served this purpose for PAR1, allowing the crystal structure of the PAR1-vorapaxar complex to be solved at a resolution of 2.2 Å (Figure).
Characterization of the PAR1-vorapaxar structure was a collaboration between the laboratories of Brian Kobilka, MD, co-recipient (along with Robert Lefkowitz) of the 2012 Nobel Prize in Chemistry for his work on GPCRs, and Shaun Coughlin, MD, PhD, who originally described the activation mechanism of PAR1. In the current work, Zhang et al. identify several features of PAR1 that distinguish it from previously resolved GPCR structures. Like other GPCRs, PAR1 has seven transmembrane (TM) spanning domains. Differences in the packing structure of internal amino acids of TM domains 3, 5, and 6 suggest that signal propagation in PAR1 from the extracellular peptide-binding interface to cytosolic domains differs from that of other members of the A family of GPCRs. Owing to a strong hydrogen-bonding network, TM7 is displaced inwardly, toward TM2 (Figure). In contrast to GPCRs that are activated by small ligands (e.g., hormones), PAR1 does not possess a deep ligand-binding pocket. Rather PAR1 has a shallow, closed pocket (Figure) that becomes even more tightly closed once vorapaxar is removed. Based on mutational studies of xenopus PAR1, activation of the receptor by its tethered ligand appears to involve superficial interactions with extracellular loops rather than deep interactions within a ligand-binding pocket. Such an activation mechanism is unusual, even among other GPCRs activated by peptide ligands.
In Brief
The crystal structure of PAR1 complexed with vorapaxar provides the first high-resolution depiction of this unusual receptor and demonstrates how a small molecule (vorapaxar) can block receptor activation by binding to a relatively superficial site on the protein (Figure). The PAR1 structure elucidated by Zhang and colleagues will inform additional structure-function studies aimed at understanding how engagement of the tethered ligand of PAR1 transmits a signal through the receptor. That and other information generated from the crystal structure will enable more rational design of improved PAR1 inhibitors.
