Proteolysis of the exodomain of recombinant protease-activated receptors: Prediction of receptor activation or inactivation by MALDI mass spectrometry

Protease-activated receptors (PARs) mediate cell activation after proteolyt ic cleavage of their extracellular amino terminus. Thrombin selectively cle aves PAR1, PAR3, and PAR4 to induce activation of platelets and vascular ce lls, while PAR2 is preferentially cleaved by trypsin. In pathological situa tions, other proteolytic enzymes may be generated in the circulation and co uld modify the responses of PARs by cleaving their extracellular domains. T o assess the ability of such proteases to activate or inactivate PARs, we d esigned a strategy for locating cleavage sites on the exofacial NH2-termina l fragments of the receptors. The first extracellular segments of PAR1 (PAR 1E) and PAR2 (PAR2E) expressed as recombinant proteins in Escherichia coli were incubated with a series of proteases likely to be encountered in the c irculation during thrombosis or inflammation. Kinetic and dose-response stu dies were performed, and the cleavage products were analyzed by MALDI-TOF m ass spectrometry. Thrombin cleaved PAR1E at the Arg41-Ser42 activation site at concentrations known to induce cellular activation, supporting a native conformation of the recombinant polypeptide. Plasmin, calpain and leukocyt e elastase, cathepsin G, and proteinase 3 cleaved at multiple sites and wou ld be expected to disable PAR1 by cleaving COOH-terminal to the activation site. Cleavage specificities were further confirmed using activation site d efective PAR1E S42P mutant polypeptides. Surface plasmon resonance studies on immobilized PAR1E or PAR1E S42P were consistent with cleavage results ob tained in solution and allowed us to determine affinities of PAR1E-thrombin binding. FAGS analyses of intact platelets confined the cleavage of PAR1 d ownstream of the Arg41-Ser42 site. Mass spectrometry studies of PAR2E predi cted activation of PAR2 by trypsin through cleavage at the Arg36-Ser37 site , no effect of thrombin, and inactivation of the receptor by plasmin, calpa in and leukocyte elastase, cathepsin G, and proteinase 3. The inhibitory ef fect of elastase was confirmed on native PAR1 and PAR2 on the basis of Ca2 signaling studies in endothelial cells. It was concluded that none of the main proteases generated during fibrinolysis or inflammation appears to be able to signal through PAR1 or PAR2. This strategy provides results which c an be extended to the native receptor to predict its activation or inactiva tion, and it could likewise be used to study other PARs or protease-depende nt processes.