Androgen receptor degradation by the proteolysis-targeting chimera ARCC-4 outperforms enzalutamide in cellular models of prostate cancer drug resistance

Abstract

The androgen receptor is a major driver of prostate cancer and inhibition of its transcriptional activity using competitive antagonists, such as enzalutamide remains a frontline therapy for prostate cancer management. However, the majority of patients eventually develop drug resistance. We propose that targeting the androgen receptor for degradation via Proteolysis Targeting Chimeras (PROTACs) will be a better therapeutic strategy for targeting androgen receptor signaling in prostate cancer cells. Here we perform a head-to-head comparison between a currently approved androgen receptor antagonist enzalutamide, and its PROTAC derivative, ARCC-4, across different cellular models of prostate cancer drug resistance. ARCC-4 is a low-nanomolar androgen receptor degrader able to degrade about 95% of cellular androgen receptors. ARCC-4 inhibits prostate tumor cell proliferation, degrades clinically relevant androgen receptor point mutants and unlike enzalutamide, retains antiproliferative effect in a high androgen environment. Thus, ARCC-4 exemplifies how protein degradation can address the drug resistance hurdles of enzalutamide.

Fig. 1

Structures of AR-targeting PROTACs. a Chemical structures of enzalutamide. b–d Chemical structures of AR PROTACs 2a, 2b, and 2c. e Chemical structure of inactive epimer, which does not bind VHL. f Chemical structure of VHL component of AR PROTACs and epimer (R and R’, respectively), indicating the corresponding stereochemistry. g Chemical structure of selective androgen receptor degrader compound, SARD279

Fig. 2

Activity of AR-targeting PROTACs. VCaP cells cultured in charcoal-stripped serum (CSS) were treated with enzalutamide (enza), AR PROTACs (2a, 2b, and 2c) and SARD279 for 20 h. AR levels were determined by western blots. The western blot shows biological replicates and is representative of two independent experiments (n = 2). See Supplementary Figure 7 for full blot images

Fig. 3

PROTAC ARCC-4 potently degrades AR in prostate cancer cells. a Time course serum-free treatment of VCaP and LNCaP cells with 100 nM of ARCC-4 shows over 90% AR depletion by 6 h as measured by quantified western blot. AR levels were normalized to tubulin as a loading control. See Supplementary Figures 2, 3 for western blot images. Experiment was performed in duplicate. The data represent mean values ± SEM and the plot represents two independent experiments (n = 2). b Treatment of 22Rv1 and LNCaP cells for 24 h with vehicle (Veh), ARCC-4 or epimer (100 nM) in charcoal-stripped serum (CSS) media shows AR depletion as measured by western blot. AR-FL: full length AR; AR-V: splice variants of AR. Experiments were performed in biological duplicates and the plot represents two independent experiments (n = 2). See Supplementary Figure 8 for full blot images. c Treatment of T47D breast cancer cells for 24 h with 50 nM enzalutamide (Enza) or ARCC-4 shows no loss of glucocorticoid, estrogen, and progesterone receptors. The western blot shows biological replicates and represents two independent experiments (n = 2). See Supplementary Figure 8 for full blot images. d Transfection of VCaP cells with endo-ribonuclease prepared siRNA (esiRNA) targeting Firefly Luciferase (FLUC) as a negative control, or VHL shows a loss of ARCC-4 mediated AR degradation upon VHL knockdown. VCaP cells were treated with 100 nM of the indicated compounds for 12 h after 48 h post-transfection. Data represents two independent experiments. See Supplementary Figure 9 for full blot images. e Tandem Ubiquitin Binding Elements (TUBE1) pull-down assay in VCaP cells shows enrichment of polyubiquitin chains on AR upon a 2.5-hour treatment with 1 µM ARCC-4. This polyubiquitination of AR is not seen with the other compounds tested. Data represents two independent experiments. See Supplementary Figure 9 for full blot images. f ARCC-4-mediated AR degradation (1 µM) at 4 h is blocked by 1-h pretreatment with proteasome inhibitor epoxomicin (Epox, 2 µM) and is unaffected by 1-h pretreatment with lysosomal inhibitor bafilomycin (Baf, 100 nM). The western blot shows biological replicates and represents two independent experiments (n = 2). See Supplementary Figure 9 for full blot images

Fig. 4

ARCC-4 is better than enzalutamide in inducing apoptosis and inhibiting AR signaling in prostate cancer cells overexpressing AR. a Pretreatment of VCaP cells with ARCC-4, enzalutamide or epimer for 8 h in charcoal-stripped serum (CSS) media followed by a 48-h stimulation with 0.2 nM R1881 shows blockage of PSA upregulation as measured by western blot. Experiments were performed in duplicate. The results are representative of two independent experiments (n = 2). b Treatment of VCaP cells for 48 h with enzalutamide, ARCC-4, or epimer leads to varying apoptosis induction as measured by caspase-3 and caspase-7 activation. Experiments were performed in triplicate and the plot is representative of two independent experiments (n = 2). c, d Treatment of CRPC cells (VCaP or LNCaP/AR) with enzalutamide, ARCC-4, or epimer for 6 days demonstrates antiproliferative effects. Experiments were performed in triplicate and the plot is representative of two independent experiments (n = 2). e Treatment of AR-negative PC3 prostate cancer cells with enzalutamide, ARCC-4, or epimer for 6 days shows no antiproliferative effect. Epoxomicin treatment was a positive control for toxicity in these cells. Experiments were performed in triplicate and the plot is representative of two independent experiments (n = 2). f Treatment of VCaP cells with enzalutamide, ARCC-4, or epimer for 6 days in 0.1 nM R1881, following a 48-h FLUC or VHL esiRNA transfection. Data shows a substantial reduction in ARCC-4 inhibition of cell proliferation upon VHL knockdown. Experiments were performed in triplicate. All data represent mean values ± SEM

Fig. 5

ARCC-4 shows efficacy against clinically relevant AR mutations. a Treatment of LNCaP/F876L AR cells with indicated concentrations of enzalutamide or ARCC-4 for 48 h shows a greater increase in PSA levels with enzalutamide as measured by western blot. Results are representative of two independent experiments (n = 2). veh refers to vehicle-treated samples. See Supplementary Figure 10 for full blot images. b Treatment of HEK293T cells overexpressing wild-type (WT) or different AR mutants for 20 h (after 50 ng per ml doxycycline induction) with ARCC-4 in charcoal-stripped serum (CSS) media shows AR depletion of WT and all mutants tested as measured by quantified western blot. AR levels were normalized to GAPDH as a loading control. See Supplementary Figure 6 for western blot images. Experiments were performed in duplicate and the plot is representative of two independent experiments (n = 2). c AR degradation in HEK293T cells overexpressing WT or different AR mutants treated for 20 h (after 50 ng per ml doxycycline induction) with vehicle, ARCC-4, or epimer (500 nM) in CSS media. Experiment was performed in biological duplicates (n = 1). See Supplementary Figure 10 for full blot images. All data represent mean values ± SEM

Fig. 6

ARCC-4 retains activity and outperforms enzalutamide in the presence of higher androgen levels. a Treatment of VCaP cells with ARCC-4 (1 µM) in charcoal-stripped serum (CSS) media for 6 h shows reversal of AR degradation with increasing amounts of synthetic androgen R1881. This experiment is representative of two independent experiments (n = 2). See Supplementary Figure 11 for full blot images. b AR PROTAC promotes apoptosis at concentrations of R1881 that overcome enzalutamide-induced apoptosis. VCaP cells were treated for 48 h with increasing amounts of R1881 in the presence of vehicle, enzalutamide, ARCC-4, or epimer (all 1 µM); apoptosis induction was measured by activation of caspase 3/7. Experiments were performed in triplicate and the plot is representative of two independent experiments (n = 2). c AR PROTAC blocks prostate cancer cell proliferation at concentrations of R1881 that overcome enzalutamide-induced growth arrest. VCaP cells were treated for 6 days with increasing amounts of R1881 in the presence of vehicle, enzalutamide, ARCC-4, or epimer (all 1 µM). Experiments were performed in triplicate and the plot is representative of two independent experiments (n = 2). All data represent mean values + SEM