Niclosamide: CRL4AMBRA1 mediated degradation of cyclin D1 following mitochondrial membrane depolarization

Abstract

Targeted protein degradation has emerged as a promising approach in drug discovery, utilizing small molecules like molecular glue degraders to harness the ubiquitin-proteasome pathway for selective degradation of disease-driving proteins. Based on results from proteomics screens we investigated the potential of niclosamide, an FDA-approved anthelmintic drug with a 50 year history in treating tapeworm infections, as a molecular glue degrader targeting the proto-oncogene cyclin D1. Proteomics screens in HCT116 colon carcinoma and KELLY neuroblastoma cells, found that niclosamide induces rapid cyclin D1 degradation through a mechanism involving the ubiquitin-proteasome pathway. A genetic CRISPR screen identified the E3 ligase CRL4^AMBRA1 as a key player in this process. Structure-activity relationship studies highlighted critical features of niclosamide necessary for cyclin D1 degradation, demonstrating a correlation between mitochondrial membrane potential (MMP) disruption and cyclin D1 downregulation. Notably, various mitochondrial uncouplers and other compounds with similar drug sensitivity profiles share this correlation suggesting that MMP disruption can trigger cyclin D1 degradation, and that the cellular signal driving the degradation differs from previously described mechanism involving CRL4^AMBRA1. Our findings underscore the complexities of proteostatic mechanisms and the multitude of mechanisms that contribute to degrader drug action.

Fig. 1. Niclosamide induces proteostatic degradation of CCND1 in HCT116 cells. A) Change in protein levels relative to vehicle treatment (DMSO) in HCT116 cells treated with niclosamide at 5 μM for 3 h quantified by TMT labelling and LC MS/MS analysis versus p-value. B) (top) Western blot of anti-CCND1 and anti-tubulin of HCT116 cells treated with vehicle (DMSO), niclosamide at different 5 μM for different time points and (bottom) pre- and co-treated with bortezomib (5 μM), MLN4924 (5 μM) and TAK243 (1 μM). High cytotoxicity observed at 48 h treatment time. C) Relative quantification (RQ) amplicon levels from primer pair a normalized to vehicle treatment (DMSO) in HCT116 cells treated with the niclosamide at 5 μM for the indicated times. Values represent mean −ΔΔCt values ± SEM from triplicates (n = 3). D) Structure of niclosamide.

Fig. 2. CCND1 reporter cell lines capture degradation activity. A) Set-up towards establishing the fluorescence-based reporter system for quantifying CCND1 protein levels in Z138 cells. B) The population of Z138 reporter cells (Z138 CCND1-GFP.IRES.mCherry) gated by flow cytometry and used for quantification of CCND1. C) Dose response curve of CCND1 levels relative to vehicle (DMSO) treatment in the reporter cells treated with niclosamide for 3 h. Values represent the ratio of the geometric mean of GFP and mCherry values in the gated population. D) Workflow of setting up a CCND1 reporter in Hep3B cell with Cas9 activity and performing a FACS based genetic screen of members of the UPS which can rescue drug induced CCND1 degradation. E) Genes enriched in a FACS based genetic CRISPR screen (Bison library) rescuing niclosamide induced CCND1 degradation in Hep3B.Cas9.CCND1-GFP.IRES.mCherry reporter cells. The gates used are shown in D) and each contain 5% of the population shown. Gate D contains cells with gene KOs that stabilize CCND1, whereas gate A contains cells with gene KOs that destabilize CCND1.

Fig. 3. Structure activity study of niclosamide towards CCND1 degradation. A) Molecules probing the nitroarene ring of niclosamide. Change in CCND1 levels relative to vehicle (DMSO) treatment in the reporter cells treated at a concentration of 5 μM for 3 h. Values represent the ratio of the geometric mean of GFP and mCherry values in the gated population. B) As for A) but for molecules probing the salicylamide ring.

Fig. 4. Niclosamide induced CCND1 degradation correlates with mitochondria depolarization. A) Mitochondrial membrane potential (MMP) quantified using flow cytometry in HCT116^AID-AMBRA1 cells stained with JC-1. B) Change in MMP relative to vehicle (DMSO) treatment in HCT116^AID-AMBRA1 cells treated at a concentration of 5 μM for 3 h. Values represent the ratio of the geometric mean of GFP and DsRed values of all the measures cells. C) Change in CCND1 levels relative to vehicle (DMSO) quantified in reporter cells (Z138.CCND1-GFP.IRES.mCherry) treated with the compounds listed in B) and plotted against the change in MMP measured in HCT116^AID-AMBRA1 cells treated with the same compounds at a concentration of 5 μM for 3 h.

Fig. 5. Mitochondria depolarizers and other molecules show same correlation between CCND1 and mitochondrial membrane potential. A) Western blot of anti-CCND1 and anti-tubulin of HCT116 and AMBRA1 KO HCT116^{AMBRA1−/−} cells treated with vehicle (DMSO), niclosamide, CCCP, SR4, BAM15, C12TPP, FCCP and sevoflurane at 5 μM for 2 h. B) Change in CCND1 levels relative to vehicle (DMSO) quantified in reporter cells (Z138.CCND1-GFP.IRES.mCherry) treated with the compounds listed in B) and plotted against the change in mitochondrial membrane potential measured in HCT116^AID-AMBRA1 cells treated with the same compounds at a concentration of 5 μM for 3 h. C) Structure of the protonophore CCCP and molecules that correlated with niclosamide drug sensitivity according to the AUC secondary screen in PRISM dataset. D) Structures of niclosamide, tizoxanide and other closely structurally related molecules.

Fig. 6. Polypharmacology of niclosamide.