Allosteric inhibitor of β-catenin selectively targets oncogenic Wnt signaling in colon cancer

Abstract

Abnormal regulation of β-catenin initiates an oncogenic program that serves as a main driver of many cancers. Albeit challenging, β-catenin is an attractive drug target due to its role in maintenance of cancer stem cells and potential to eliminate cancer relapse. We have identified C2, a novel β-catenin inhibitor, which is a small molecule that binds to a novel allosteric site on the surface of β-catenin. C2 selectively inhibits β-catenin, lowers its cellular load and significantly reduces viability of β-catenin-driven cancer cells. Through direct binding to β-catenin, C2 renders the target inactive that eventually activates proteasome system for its removal. Here we report a novel pharmacologic approach for selective inhibition of β-catenin via targeting a cryptic allosteric modulation site. Our findings may provide a new perspective for therapeutic targeting of β-catenin.

Figure 1

Biophysical characterization of C2. (A) Molecular structure of C2. Molecular weight: 484 [g/mol]; molecular formula: C19H20Br2N2OS; SMILES: N(C(=S)NC=Cc1cc(Br)cc(Br)c1O)C2CC3CC2C4C=CCC34. (B) Thermal melting profile of 2.5 μM purified wild-type human β-catenin protein in complex with 20 μM C2. (C) Binding sensorgram of C2 to β-catenin as measured with surface plasmon resonance (SPR). Five doses of C2 (12, 37, 110, 330 and 1000 nM) were injected over immobilized β-catenin. Association and dissociation rate are indicated. Jagged lines represent sensorgram data for injections of C2. Smooth lines represent fitting to Langmuir kinetic model. The estimated kinetics constants: K_on = 42.37 × 10⁵ M/s; K_off = 6.84 × 10⁻³ 1/s; K_d = 2.89 × 10⁻⁸ M. (D) Binding kinetics data of β-catenin with C2 measured with Micro-scale thermophoresis (MST). Various concentrations of C2 ranging from 0 to 20 µM with 2-fold dilution series were tested in duplicate. The binding signal from a reference flow capillary containing no protein was subtracted to account for detection of specific interaction between the compound and β-catenin. (E) Domain deletion reporter for β-catenin. 10 signaling pathways (WNT, Notch, p53/DNA damage, TGFβ, cell cycle, NFκB, Myc/Max, HIF-1, MAPK/ERK and MAPK/JNK) along with positive and negative controls were tested. All measurements were done in triplicates (n = 3).

Figure 2

Selectivity of C2 for Wnt pathway. (A) Screening of β-catenin-dependent cell lines. (B) Effect of C2 on viability of colon cancer cell lines (24 hour treatment). IC50 values are shown for respective cell lines. (C) Colony assay for WNT-dependent vs independent cells (7 day treatment). The number of colonies in each well was counted after 7 days of incubation. (D) Cancer 10-pathway selectivity assay (24 hour treatment). (E) Western blot analysis of Wnt activity in DLD1 and SW480 cells (24 hour treatment). All measurements were done in triplicates (n = 3).

Figure 3

Mechanism of action of C2. (A) Time-course analysis of phosphorylation of β-catenin (1 µM C2). (B) Effect of C2 on ubiquitination of β-catenin (3 hour treatment). MG132 was used as proteasome inhibitor. (C) Combination of C2 (1 µM) with GSK3β inhibitor CHIR99021 (1 µM) (3 hour treatment). (D) Cellular localization of β-catenin as measured by confocal microscopy (24 hour treatment). Co-localization factor represents the level of β-catenin overlapping with nuclear DNA. Co-localization factor was calculated as 0.62 ± 0.03 and 0.25 ± 0.15 for DMSO and C2 treatment, respectively. (E) Nuclear-cytoplasmic fractionation of cells treated with C2. All measurements were done in triplicates (n = 3).

Figure 4

Effect of C2 on tumor growth. (A) Mouse xenograft test. Tumor volume (P = 0.00036), tumor growth (P = 0.0017) and actual images of tumors are shown. Error bars are in SEM.