Induction of cre recombinase activity using modified androgen receptor ligand binding domains: a sensitive assay for ligand-receptor interactions

Abstract

Novel systems of inducible gene expression are presented in which CRE-M, an altered form of cre recombinase (cre), is fused to and activated by ligand binding to two forms of the androgen receptor (AR) ligand binding domain (LBD). Selective activation or inactivation of gene transcription is induced upon the addition of appropriate ligand. The coupling of this cre-LBD system with our previously reported highly sensitive assay to measure cre activity in vitro using a dual fluorescent gene switch reporter provides a novel, high-throughput assay system for identifying compounds that bind to and activate various forms of the LBD of androgen receptor. This method can similarly be applied to screen compounds for their activating properties on other steroid hormone LBDs. Three different forms of the AR-LBD were fused to CRE-M, including the wild-type AR-LBD (wt), a non-ligand binding truncated form, LBD (T), and a mutated form (Thr-->Ala substitution) identified in the LNCaP prostate cancer cell line, LBD (LNCaP). We demonstrate a 10-fold induction of cre activity by the addition of androgen agonists to the CRE-M-AR-LBD(wt) fusion protein, but not in the presence of the anti-androgen, flutamide. However, cre activity can be induced by flutamide with the CRE-M-AR-LBD(LNCaP) fusion protein. Similar activation properties were obtained when these fusion proteins were expressed using adenoviral vectors. When combined with our previously reported cre-lox gene switch system, the CRE-M-AR-LBD system can be utilized in gene therapy systems in which a therapeutic product may be initially expressed, replaced by a second product, or turned-off following exposure to ligand. This provides an important, additional level of regulation to gene therapy systems.

Figure 1

Generation of ligand inducible cre recombinase. Strategy to generate variants of ligand inducible cre recombinase by fusing various forms of the AR-LBD to CRE-M (see Materials and Methods).

Figure 2

Functional assay of inducible cre recombinase activity induced by ligand. CHO-K1 cells were transiently transfected with reporter plasmid pCMV-EGFP/RFP (0.95 µg) and co-transfected with 0.75 µg of pCMV-CRE-M-AR(WT) (A–F), pCMV-CRE-M-AR(LNCaP) (G–L) or pCMV-CRE-M-AR(T) (M–R) in absence of ligand (A, D, G, J, M, P), in presence of 100 nM mibolerone (B, E, H, K, N, Q) or 100 nM OH-flutamide (C, F, I, L, O, R). Fluorescence microscopy using GFP filter (A, B, C, G, H, I, M, N, O), RFP filter (D, E, F, J, K, L, P, Q, R). Assay was performed 48 h after exposure to ligand (magnification 100×).

Figure 3

Relative induction of cre recombinase activity by different ligands. (A) Representation of luciferase reporter construct to assay cre recombinase activity. (B) PC-3 cells were co-transfected with reporter plasmid pCMVe-βAc-STOP-luc (0.1 pmol) and plasmids (0.1 pmol) coding for modified inducible cre recombinase in the absence or presence of ligand for a period of 48 h. Transfection efficiency was normalized with pTK-Rl. (C) PC-3 cells were co-transfected with reporter plasmid pCMVe-βAc-STOP-luc and pCMV-CRE-M-AR(LNCaP) (0.001 pmol) in presence of selected ligands at indicated ligand concentration (48 h exposure to ligand). Transfection efficiency was normalized with pTK-Rl which expresses renilla luciferase.

Figure 4

Functional assay for inducible cre recombinase activity generated by adenovirus. (A) Schematic representation of Adeno-Cre and Adeno-CMV-CRE-M-AR(LNCaP). (B and C) CHO-K1 cells transfected with 0.95 µg reporter plasmid pCMV-EGFP/RFP and infected with Adeno-Cre. Fluorescence microscopy using GFP filter (B) or RFP filter (C). Assay performed 48 h post-infection (magnification 400×). (D–G) Cell line QBI-293A was transfected with 0.95 µg of reporter plasmid pCMV-EGFP/RFP and 2 h later was infected with Adeno-CMV-CRE-M-AR(LNCaP) (E–G) in the absence of ligand (D, F) or in the presence of 100 nM OH- flutamide (E, G). Assay was performed 48 h post-infection. Fluorescence microscopy using GFP filter (D, E) or using RFP filter (F, G) (magnification 100×).

Figure 4

Functional assay for inducible cre recombinase activity generated by adenovirus. (A) Schematic representation of Adeno-Cre and Adeno-CMV-CRE-M-AR(LNCaP). (B and C) CHO-K1 cells transfected with 0.95 µg reporter plasmid pCMV-EGFP/RFP and infected with Adeno-Cre. Fluorescence microscopy using GFP filter (B) or RFP filter (C). Assay performed 48 h post-infection (magnification 400×). (D–G) Cell line QBI-293A was transfected with 0.95 µg of reporter plasmid pCMV-EGFP/RFP and 2 h later was infected with Adeno-CMV-CRE-M-AR(LNCaP) (E–G) in the absence of ligand (D, F) or in the presence of 100 nM OH- flutamide (E, G). Assay was performed 48 h post-infection. Fluorescence microscopy using GFP filter (D, E) or using RFP filter (F, G) (magnification 100×).