Development and implementation of a cell-based assay to discover agonists of the nuclear receptor REV-ERBα

Abstract

The nuclear receptors are transcription factors involved in the regulation of a variety of physiological processes whose activity can be modulated by binding to relevant small molecule ligands. Their dysfunction has been shown to play a role in disease states such as diabetes, cancer, inflammatory diseases, and hormonal resistance ailments, which makes them interesting targets for drug discovery. The nuclear receptor REV-ERBα is involved in regulating the circadian rhythm and metabolism. Its natural ligand is heme and there is significant interest in identifying novel synthetic modulators to serve as tools to characterize its function and to serve as drugs in treating metabolic disorders. To do so, we established a mammalian cell-based two-hybrid assay system capable of measuring the interaction between REV-ERBα and its co-repressor, nuclear co-repressor 1. This assay was validated to industry standard criteria and was used to screen a subset of the LOPAC^®1280 library and 29568 compounds from a diverse compound library. Profiling of the primary hits in a panel of counter and selectivity assays confirmed that REV-ERBα activity can be modulated pharmacologically and chemical scaffolds have been identified for optimization.

Keywords: REV-ERBα; assay development; drug discovery; high throughput screening; luciferase reporter; nuclear receptor.

Figure 1.. Structural formulae of the synthetic ligands GSK4112, SR9011 and SR9009.

Figure 2.. Schematic depiction of the mammalian cell-based two-hybrid system measuring the interaction between nuclear receptor REV-ERBα and corepressor NCoR.

Plasmids were co-transfected into HEK-293T cells, and then the Renilla and firefly luciferase signals were quantified using the Dual-Glo^® Luciferase System 24 h post-transfection. In the mammalian cell-based two-hybrid system, the DNA-binding domain is co-expressed with human REV-ERBα and the transcription activation domain is co-expressed with NCoR. The interaction of REV-ERBα and NCoR leads to transcription of firefly luciferase. GSK4112 is a REV-ERBα agonist binding to the heme binding site resulting in recruitment of the co-repressor NCoR.

Figure 3.. Characterization of the REV-ERBα mammalian cell-based two-hybrid assay.

A. Determination of the appropriate DNA to transfection reagent ratio (N/P) in the REV-ERBα mammalian cell-based two-hybrid assay. HEK-293T cells (10000 cells/well) were transiently transfected with different amount of DNA (0.05–0.3 μg/well) in combination with different jetPEI concentrations (0.1–0.6 μl/well). The Renilla luciferase signal was measured 24 h post transfection using the Dual-Glo^® Luciferase System. B. Effect of complex formation incubation time on transfection efficiency with jetPEI. HEK-293T cells were transfected in suspension with the plasmids in batch mode. The jetPEI reagent was added to DNA sample and the complex was incubated for 5, 15, 30, 45 and 60 min. The Renilla luciferase signal was measured 24 h post transfection using the Dual-Glo^® Luciferase System. C. Dose-response of the GSK4112 agonist in the assay. HEK-293T cells were transiently transfected with four plasmids for 8 h followed by stimulation by GSK4112 for 16 h. The data were normalized by ratio (firefly luciferase signal/Renilla luciferase signal) and fitted using GraphPad Prism. Calculation of GSK4112 EC₅₀ (5.5 µM) and Hill slope (1.21) was performed using GraphPad Prism. Results represent normalized mean values per well (triplicates). Error bars show the STD.

Figure 4.. Screening of a subset of the LOPAC®1280 and diverse Enamine compound libraries.

A. Correlation of activation (%) from two independent experiments by compounds from the LOPAC®1280 library. The firefly and Renilla luciferase signals were measured using the Dual-Glo® Luciferase System. The data were normalized by ratio (firefly luciferase signal/Renilla luciferase signal). The average activation (%) based upon the ratio firefly luciferase signal/Renilla luciferase signal measured on day 1 and 2 are plotted on the y and x-axes respectively [green: negative control (0.5% v/v DMSO); light blue: positive control (100 μM GSK4112); black: test compounds (inactive compounds); purple: hits (active compounds); orange: toxic compounds]. B. Dose-response of chelidamic acid in the assay. The data were normalized by ratio (firefly luciferase signal/Renilla luciferase signal) and fitted using GraphPad Prism. Results represent normalized mean values per well (triplicates). Error bars show the STD. Calculation of chelidamic acid EC50 (0.36 µM) and Hill slope (1.23) was performed using GraphPad Prism. C. Scatter-plot for the screening of the Enamine compound library against the REV-ERBα mammalian cell-based two-hybrid assay.

Figure 5.. Secondary screening of selected hits from diverse Enamine compound screen.

A. Counter screen using mammalian two hybrid assay. HEK-293T cell were transfected with Gal4-REV-ERBα and VP16-NCoR as described in Figure 1. B and C. Secondary screen assay using a mammalian single hybrid assay (without any VP16 construct). B. Cells were transfected with Gal4-hPXR. C. Cells were transfected with Gal4-hCAR. Transfected cells were treated for 24 h with DMSO (0.1% v/v) or 10 µM of selected compounds and luciferase assay were performed as described above. Results are expressed as mean ± SEM and statistical significance were assessed using GraphPad prism software by a 1-Way ANOVA followed by a Tuckey post-hoc test, *P < 0.05, **P < 0.01, ***P < 0.001.