Protease-activated receptor 1 as an endogenous model of peptidergic Gαq-Gα12-biased G protein signaling

G protein-coupled receptors (GPCRs) are the most widely targeted class of signaling proteins, comprising ∼30% of FDA-approved drugs. Their therapeutic potential arises from their ability to translate diverse extracellular cues into intracellular signals via G proteins, arrestins, and other effectors. This signaling versatility creates opportunities for functional selectivity, where ligands preferentially engage particular pathways. However, few endogenous receptor systems display well-defined ligand-dependent divergence across multiple signaling levels. Protease-activated receptor 1 (PAR1) is a candidate model. Thrombin canonically cleaves PAR1 at Arg41, whereas activated protein C (aPC) has been reported to cleave PAR1 at Arg46 in endothelial and co-receptor-supported settings, creating distinct tethered peptide ligands. Thrombin cleavage drives canonical Gαq- and Gα12/13-dependent prothrombotic and barrier-disruptive signaling, whereas aPC cleavage has been associated with anticoagulant, cytoprotective, and anti-inflammatory signaling. However, PAR1's transducer-wide coupling profile, transcriptional consequences, and physiological outputs remain incompletely characterized.

We integrated transducer-wide biosensor assays (TRUPATH, TGFα shedding, PRESTO-Tango), analysis of a PAR1 ± thrombin TRE-MPRA dataset followed by targeted TRE dual-luciferase validation, and platelet activation and calcium flux studies in primary human platelets to define how protease identity reshapes signaling from proximal transducer engagement to physiological output.

Thrombin produced robust PAR1 coupling to Gαq and Gα12, whereas aPC produced detectable coupling only to Gα12. Neither protease generated detectable β-arrestin-2 recruitment in PRESTO-Tango. Both proteases supported dose-dependent TGFα shedding that was insensitive to FR900359. A PAR1 ± thrombin TRE-MPRA dataset identified thrombin-responsive transcriptional response elements and nominated NFκB1 and THRB for targeted follow-up. Luciferase assays showed NFκB1 was thrombin-induced and FR900359-sensitive, whereas THRB was induced by both thrombin and aPC and was FR900359-insensitive. In primary human platelets, thrombin, but not aPC, induced P-selectin expression and a calcium response. Thrombin responses were suppressed by FR900359, supporting a requirement for Gαq in these platelet activation markers.

Together, these findings support PAR1 as an endogenous model of protease-dependent functional selectivity that, in our heterologous assay systems, separates signaling along a Gαq-versus-Gα12 axis, thus providing a framework for future technologies, such as high-throughput tethered-peptide evolution platforms, to understand the principles of G protein selectivity across GPCRs.