Isoform-specific antagonists of exchange proteins directly activated by cAMP

Abstract

The major physiological effects of cAMP in mammalian cells are transduced by two ubiquitously expressed intracellular cAMP receptors, protein kinase A (PKA) and exchange protein directly activated by cAMP (EPAC), as well as cyclic nucleotide-gated ion channels in certain tissues. Although a large number of PKA inhibitors are available, there are no reported EPAC-specific antagonists, despite extensive research efforts. Here we report the identification and characterization of noncyclic nucleotide EPAC antagonists that are exclusively specific for the EPAC2 isoform. These EAPC2-specific antagonists, designated as ESI-05 and ESI-07, inhibit Rap1 activation mediated by EAPC2, but not EPAC1, with high potency in vitro. Moreover, ESI-05 and ESI-07 are capable of suppressing the cAMP-mediated activation of EPAC2, but not EPAC1 and PKA, as monitored in living cells through the use of EPAC- and PKA-based FRET reporters, or by the use of Rap1-GTP pull-down assays. Deuterium exchange mass spectroscopy analysis further reveals that EPAC2-specific inhibitors exert their isoform selectivity through a unique mechanism by binding to a previously undescribed allosteric site: the interface of the two cAMP binding domains, which is not present in the EPAC1 isoform. Isoform-specific EPAC pharmacological probes are highly desired and will be valuable tools for dissecting the biological functions of EPAC proteins and their roles in various disease states.

Fig. 1.

Relative potency of EPAC2 specific antagonists. (A) EPAC GEF activity was monitored as a decrease in fluorescence of a reaction mixture containing 100 nM EPAC, 200 nM Rap-mantGDP, 25 µM GDP, and 25 µM cAMP. The K_GEF activities at individual ESI concentrations were calculated by fitting the kinetic traces to a single exponential decay. Apparent IC₅₀ values were obtained by plotting individual reaction rates of EPAC1 (open symbols) or EPAC2 (filled symbols) against the ESI-05 (triangles) and ESI-07 (squares) concentrations. (B) Dose-dependent competition of ESI-05 (triangles), ESI-07 (squares), and cAMP (circles) with 8-NBD-cAMP in binding to EPAC2. Similar results were obtained from three independent titrations. Apparent IC₅₀ values were obtained by fitting of the individual titration curves and expressed in the format of means and SDs.

Fig. 2.

Effects of EPAC2-specific antagonists on 007-AM–mediated cellular activation of Rap1. Serum-starved HEK293/EPAC2 cells or HEK293/EPAC1 cells with or without pretreatment of ESI-05 or ESI-07 for 5 min were stimulated with 10 µM 007-AM for 10 min. GTP-bound Rap1 (Rap1GTP) obtained by a Ral-GDS-RBD-GST pull-down assay and total cellular Rap1 were detected by immunoblotting with Rap1-specific antibody. (A) HEK293/EPAC2 cells treated with ESI-05. (B) HEK293/EPAC2 cells treated with ESI-07. (C) HEK293/EPAC1 cells treated with ESI-05 or ESI-07. Similar results were obtained with three independent experiments for each panel of Fig. 2. A t test was used to determine statistical significance (*P < 0.05).

Fig. 3.

Suppressing of cellular activation of EPAC2-FL but not EPAC1-FL or EPAC1-camps by ESI-05. For HEK293 cells stably expressing EPAC2-FL (A–C), EPAC1-FL (D–F), or EPAC1-camps (G, H), it was demonstrated that 10 μM ESI-05 blocked EPAC2-FL but not EPAC1-FL or EPAC1-camps activation in response to 3 μM 007-AM or 2 μM forskolin and 100 μM isobutylmethylxanthine, respectively. Similar results were obtained with three independent experiments for each panel of Fig. 3. A t test was used to determine statistical significance (*P < 0.05).

Fig. 4.

Changes in hydrogen/deuterium exchange rates of EPAC2 induced by binding of ESI-07. Differences in deuteration levels in the free and ESI-07–bound EPAC2 at various time points (from top to bottom: 10, 100, 1,000, 10,000, and 100,000 s) are shown in color-coded bars ranging from blue (−50%) to red (50%), as indicated at the bottom right corner of the figure.

Fig. 5.

Mechanism of ESI-07–mediated isoform-specific EPAC2 antagonism. (A) Proposed model of ESI-07 binding to EPAC2. Areas within CBD-A (cyan) and CBD-B (green) of EPAC2 (2BYV) with reduced H/D exchange rates in response to ESI-07 binding are shown in dark blue. (B) Proposed model of cAMP binding to EPAC2. Areas within CBD-A (cyan) and CBD-B (green) of EPAC2 (2BYV) with reduced H/D exchange rates in response to cAMP binding are shown in dark blue; the area with increased H/D exchange rates in response to cAMP binding is shown in red (29).