Characterization of Protease-Activated Receptor (PAR) ligands: Parmodulins are reversible allosteric inhibitors of PAR1-driven calcium mobilization in endothelial cells

Abstract

Several classes of ligands for Protease-Activated Receptors (PARs) have shown impressive anti-inflammatory and cytoprotective activities, including PAR2 antagonists and the PAR1-targeting parmodulins. In order to support medicinal chemistry studies with hundreds of compounds and to perform detailed mode-of-action studies, it became important to develop a reliable PAR assay that is operational with endothelial cells, which mediate the cytoprotective effects of interest. We report a detailed protocol for an intracellular calcium mobilization assay with adherent endothelial cells in multiwell plates that was used to study a number of known and new PAR1 and PAR2 ligands, including an alkynylated version of the PAR1 antagonist RWJ-58259 that is suitable for the preparation of tagged or conjugate compounds. Using the cell line EA.hy926, it was necessary to perform media exchanges with automated liquid handling equipment in order to obtain optimal and reproducible antagonist concentration-response curves. The assay is also suitable for study of PAR2 ligands; a peptide antagonist reported by Fairlie was synthesized and found to inhibit PAR2 in a manner consistent with reports using epithelial cells. The assay was used to confirm that vorapaxar acts as an irreversible antagonist of PAR1 in endothelium, and parmodulin 2 (ML161) and the related parmodulin RR-90 were found to inhibit PAR1 reversibly, in a manner consistent with negative allosteric modulation.

Keywords: Calcium mobilization; GPCR; ML161; Negative allosteric modulator; PAR1; PAR2; Parmodulin; Protease-Activated Receptor; RWJ-58259; Vorapaxar.

Fig. 1

Thrombin-mediated PAR1 signaling.

Fig. 2

Previously-reported PAR antagonists used in these studies.

Fig. 3

A: Adherent Ea.hy926 cells containing Fluo-4 dye. Cells were imaged in a clear-bottom, black wall 96-well plate, using an Evos Fl inverted microscope (20X) with GFP setting. B: Representative Fluo-4-Ca²⁺ emission curve (525 nm) in response to TFLLRN-NH₂ (3.16 μM). C: Concentration–response curve of a typical (unsuccessful) assay with the PAR1 antagonist ML161 and 3.16 μM TFLLRN-NH₂ before optimization (manual media exchanges), n = 3. D: Representative concentration–response curve with the PAR1 antagonist ML161 and 5 μM TFLLRN-NH₂ after assay optimization (automated liquid handling).

Fig. 4

A: Calcium mobilization responses with PAR1 agonist 5 μM TFLLRN-NH₂ (column A) plus optional PAR1 antagonists 10 μM ML161 (column C) or 0.316 μM vorapaxar (column D). B: Table of calculated Z′-factors.

Fig. 5

PAR agonist concentration–response with EA.hy926 cells: A) TFLLRN-NH₂ (PAR1)-mediated iCa²⁺ mobilization; B) SLIGKV-NH₂ (PAR2)-mediated iCa^2+.

Fig. 6

Concentration-response for PAR1 antagonists in the TFLLRN-NH₂-mediated iCa²⁺ mobilization assay (5 μM) with EA.hy926 cells: A) vorapaxar, B) atopaxar, (C) RWJ-58259 D) DG-207.

Fig. 7

Concentration-response for parmodulins (ML161 analogs) in the TFLLRN-NH₂-mediated (5 μM) iCa²⁺ mobilization assay.

Fig. 8

Concentration-response for the PAR2 antagonist TJF5 (Fairlie’s PAR2 antagonist, 32) in the SLIGKV-NH₂-mediated (5 μM) iCa²⁺ mobilization assay.

Fig. 9

Selectivity data of antagonists in PAR1 (TFLLRN-NH₂)- and PAR2 (SLIGKV-NH₂)-driven iCa²⁺ mobilization_{a a}ML161, RR90, Q94, TJF5 (Fairlie’s PAR2 antagonist) were used at 10 μM; Vorapaxar, Atopaxar, and RWJ-58259 were at 0.316 μM. TFLLRN-NH₂ and SLIGKV-NH₂ were used at 3.16 μM; Vehicle (V) = 10% DMSO-HBSS/HEPES.

Fig. 10

iCa²⁺ concentration–response of the PAR1 agonist TFLLRN-NH₂ in the presence of increasing concentrations of (A) vorapaxar, (B) RWJ 58259, (C) ML161, (D) RR-90.

Fig. 11

Reversibility studies of the PAR1 antagonists vorapaxar, RWJ-58259, ML161, and RR-90 in the iCa²⁺ assay. Cells containing antagonists were optionally washed with buffer prior to treatment with the PAR1 agonist TFLLRN-NH₂ (5 μM). Vorapaxar and RWJ-58259 were used at 0.316 μM; ML161, CJD125, RR-90 were used at 10 μM.

Scheme 1

Synthesis of PAR1 antagonist RWJ-58259 and alkynylated analog DG-207.

Scheme 2

General synthesis of parmodulins with west side modifications.

Scheme 3

Synthesis of azido analog DG-5.

Scheme 4

Solution-phase synthesis of TJF-5 (Fairlie’s PAR2 antagonist).