Synthetic FXR agonist GW4064 is a modulator of multiple G protein-coupled receptors

Abstract

The synthetic nuclear bile acid receptor (farnesoid X receptor [FXR]) agonist GW4064 is extensively used as a specific pharmacological tool to illustrate FXR functions. We noticed that GW4064 activated empty luciferase reporters in FXR-deficient HEK-293T cells. We postulated that this activity of GW4064 might be routed through as yet unknown cellular targets and undertook an unbiased exploratory approach to identify these targets. Investigations revealed that GW4064 activated cAMP and nuclear factor for activated T-cell response elements (CRE and NFAT-RE, respectively) present on these empty reporters. Whereas GW4064-induced NFAT-RE activation involved rapid intracellular Ca(2+) accumulation and NFAT nuclear translocation, CRE activation involved soluble adenylyl cyclase-dependent cAMP accumulation and Ca(2+)-calcineurin-dependent nuclear translocation of transducers of regulated CRE-binding protein 2. Use of dominant negative heterotrimeric G-protein minigenes revealed that GW4064 caused activation of Gαi/o and Gq/11 G proteins. Sequential pharmacological inhibitor-based screening and radioligand-binding studies revealed that GW4064 interacted with multiple G protein-coupled receptors. Functional studies demonstrated that GW4064 robustly activated H1 and H4 and inhibited H2 histamine receptor signaling events. We also found that MCF-7 breast cancer cells, reported to undergo GW4064-induced apoptosis in an FXR-dependent manner, did not express FXR, and the GW4064-mediated apoptosis, also apparent in HEK-293T cells, could be blocked by selective histamine receptor regulators. Taken together, our results demonstrate identification of histamine receptors as alternate targets for GW4064, which not only necessitates cautious interpretation of the biological functions attributed to FXR using GW4064 as a pharmacological tool but also provides a basis for the rational designing of new pharmacophores for histamine receptor modulation.

Figure 1.

FXR-independent activation of luciferase reporters by GW4064. A, FXR mRNA and protein expression in commonly used cell lines. mRNA and total protein were isolated from the indicated cell lines and were analyzed by RT-PCR (top panel) or immunoblot analysis (bottom panel). P.C, plasmid control; N.C, no RT control. GAPDH and β-actin were used as loading controls for RT-PCR and immunoblots, respectively. “Transfected FXR” is lysate from Flag-FXR–transfected HEK cells. B, HEK cells in 24-well plates were transfected with the indicated reporters (0.2 μg), expression plasmids, or empty vector (0.1 μg) and internal control pEGFPC1 (0.1 μg). At 12 hours after transfection, cells were treated with GW4064 (GW) for 24 hours and harvested, and GFP fluorescence and luminescence were measured and the GFP-normalized luciferase activity was plotted as fold luciferase activity over that for the vehicle (0.1% DMSO)–treated empty expression vector–transfected respective reporter controls. One representative of 9 independent experiments performed in duplicate and showing similar patterns but different fold activities is shown. C, Normalized luciferase activity from HEK cells transfected and treated as above was plotted as fold activity over that for the respective vehicle (V)-treated reporter controls. Data are means ± SEM from 4 independent experiments performed in duplicate. D–I, HEK (D–G and I) or MCF-7 and HepG2 cells (H) were transfected as above with the indicated plasmids. Data are means ± SEM of 3 independent experiments performed in duplicate. PC, positive control ligands. The GW4064 concentration used was 1 μM (B–E and G–I) or as indicated (F). Where indicated, 30 minutes before GW4064 treatment, cells were pretreated with 5 μM XCT-790 (D and G). The CDCA concentration used was 50 μM (I).

Figure 2.

GW4064 induces NFAT translocation and calcineurin activation by a PI-PLC–dependent mechanism. A, GW4064 (GW) induction of intracellular Ca²⁺ accumulation was assessed by flow cytometry using Fluo-4 AM as described previously (27) in a FACSCalibur flow cytometer. Data are representative of 3 independent experiments showing identical patterns. FITC, fluorescein isothiocyanate; V, vehicle. B, HEK cells in 10-cm dishes were treated with 1 μM GW4064, ionomycin (1 μM), or FK506 (10 μM) for 30 minutes and then were assayed for calcineurin activity using a colorimetric calcineurin assay kit. Recombinant human calcineurin (CaN, 40 U) was used as a positive control. Data represent nanomoles of PO₄ released (values are normalized to control) and are means ± SEM from 3 representative experiments performed in triplicate. C, HEK cells in chamber slides were treated with 1 μM GW4064 or ionomycin (I, 1 μM) for 30 minutes, and endogenous NFATc1 was detected by immunocytochemical analysis. The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Magnification, ×40; bars, 50 μm. Data are representative of 3 independent experiments. D, HEK cells in 24-well plates were transfected with 200 ng of NFAT-RE-Luc, and 100 ng of EGFPC1. Cells were then treated with the indicated compounds for 30 minutes followed by treatment with 1 μM GW4064 for 24 hours. Normalized luciferase values were then plotted as percent inhibition of GW4064 response. Data are means ± SEM of 3 independent experiments performed in duplicate. *, P < .05; *, P < .01. 2-APB, 2-aminoethoxydiphenyl borate.

Figure 3.

GW4064 activation of CRE requires involvement of both cAMP and Ca²⁺ signaling. A, HEK cells in 6-well plates were treated with the indicated doses of GW4064 (GW) or forskolin (FSK) for 30 minutes as indicated. Cells were then lysed, and the cAMP level was measured using an HTRF-based kit and normalized with protein content. The cAMP concentration was plotted. Data are means ± SEM of 3 independent experiments performed in triplicate. B, HEK cells in 10-cm dishes were treated with 1 μM GW4064 for the indicated time periods and were immunoblotted for phospho (P)-CREB S133. Total ERK1/2 was used as a loading control. Data are representative of 3 independent experiments. C, HEK cells were pretreated with 10 μM FSK for 30 minutes followed by treatment with the indicated concentrations of GW4064 for another 30 minutes. D, HEK cells were pretreated with KH7 or BAPTA-AM for 30 minutes followed by 5 minutes of treatment with 1 μM GW4064. cAMP was measured and plotted as in panel A. For C and D, data are means ± SEM of 3 independent experiments performed in triplicate. E, HEK cells in 24-well plates were transfected with 200 ng of NFAT-RE-Luc and 100 ng of EGFPC1. Cells were then treated with the indicated compounds for 30 minutes followed by treatment with 1 μM GW4064 for 24 hours. Normalized luciferase values were then plotted as percent inhibition of the GW4064 response. Data are means ± SEM of 3 independent experiments performed in duplicate. F, HEK cells in 10-cm dishes were treated as indicated, and nuclear accumulation of endogenous TORC2 was determined by Western blotting of cytosolic or nuclear extracts. α/β-Tubulin and histone H3 were used as markers for cytosolic and nuclear fractions, respectively. G and H, HEK cells in 24-well plates were transfected with CRE luciferase and the expression plasmids as above and were treated as indicated. Normalized luciferase values were plotted as fold activity over that for reporter plasmid transfected vehicle (V, 0.1% DMSO)–treated cells. Data are means ± SEM of 3 independent experiments performed in duplicate. PKA-cat, PKA catalytic subunit; Stauro, staurosporine. *, P < .05; **, P < .01; ***, P < .001.

Figure 4.

GW4064 activation of NFAT-RE is blocked by G_q/11, Gα_i, and G_o G protein minigenes. HEK cells in 24-well plates were transfected with 200 ng of NFAT-RE-Luc, 100 ng of EGFPC1, and 200 ng of the indicated G protein minigenes. Cells were then treated with 1 μM GW4064 for 24 hours. Normalized luciferase values were then plotted as percent inhibition of the GW4064 response. Data are means ± SEM of 3 independent experiments performed in duplicate.

Figure 5.

GW4064 binds to and modulates the activities of histamine receptors. A–D, Membrane fractions from CHO-K1 cells overexpressing the indicated receptors were incubated with cognate radioligands (1.20 nM [³H]pyrilamine for H1 [K_d = 1.10 nM], 0.10 nM [¹²⁵I]aminopotentidine for H2 [K_d = 0.45 nM], 0.40 nM [³H]-N-α-methylhistamine for H3 [K_d = 0.38 nM] and 8.20 nM [³H]histamine for H4 [K_d = 5.70 nM]) for 1.5 to 3 hours in the presence of the indicated concentrations of unlabeled GW4064 (GW) or ligands, and data are plotted as percent binding of radioligands with values in the absence of cold competitors set as 100. Nonspecific binding in each case was determined by coincubation with 1.0 μM pyrilamine for H1, 3.0 μM tiotidine for H2, 1.0 μM (R)-(−)-α-methylhistamine (RAMH) for H3, and 1.0 μM histamine for H4. Data represent means ± SEM from 1 experiment performed in duplicate. E and F, NFAT-RE-luciferase activity was determined in HEK cells transfected with pcDNA3.1 or H1 and treated with the indicated concentrations of histamine or GW4064 for 3 hours. G and H, NFAT-RE-luciferase activity was determined in H1-transfected HEK cells after the indicated cotreatments. I–K, HEK cells overexpressing H2, H3, or H4 and cAMP GloSensor plasmids were treated with GW4064 or histamine receptor ligands for 15 minutes. For the H2 assay (I), cells were treated with GW4064 alone for 15 minutes or were pretreated with 1 μM amthamine for 15 minutes. For H3 and H4 assays (J and K), cells were pretreated with 10 μM forskolin for 15 minutes and were then treated with imetit, clobenpropit, or GW4064 for 15 minutes further. For the Imetit + GW4064– or Clobenpropit + GW4064–treated groups, imetit or clobenpropit was added 15 minutes before the GW4064 treatment. The cAMP level was determined using a luminometer and is plotted as relative light units (RLU). Data are means ± SEM of 3 independent experiments performed in duplicate.

Figure 6.

GW4064 induces FXR-independent cell death in MCF-7 and HEK cells. A, MCF-7 or HEK cells were plated in 96-well plates (10 000 cells/well) and were maintained in medium supplemented with the indicated serum concentrations for 24 hours. The cells were then treated with the indicated concentrations of GW4064 (GW) in medium with identical serum concentrations and were further cultured for 24 or 48 hours after which cell viability was assessed by the MTT assay and plotted as percent cell death compared with that for vehicle-treated controls (for which cell death was set as 0). B, Effects of 5 μM GW4064 on cell viability of untransfected (UT), empty vector, or FXR-transfected cells were assessed by MTT. V, vehicle. C, Comparison of PGL3 promotor or 3X-FXRE-Luc activities in HepG2, HEK, or MCF-7 cells transfected with empty (Vector) or FXR expression plasmids. Normalized luciferase activities were plotted as fold activity over values from vehicle-treated control groups. Data are means ± SEM of 3 independent experiments performed in duplicate. *, P < .05.

Figure 7.

All 4 histamine receptors and HDC are expressed in HEK and MCF-7 cells and selective histamine receptor modulators block GW4064-induced cell death in these cells. A, mRNA expression as determined by RT-PCR. PC, plasmid control; NC, no RT control. Data are representative of 3 independent experiments. B and C, MTT assay in HEK (B) and MCF-7 (C) cells treated with the indicated concentrations of selective histamine receptor modulators with or without 5 μM GW4064 for 48 hours. Data are plotted as percent cell death where cell death obtained with 5 μM GW4064 alone was set as 100%. C and D, Data are means ± SEM of 3 independent experiments performed in duplicate. *, P < .05; **, P < .01; ***, P < .001.