Simultaneous Multi-Organ Metastases from Chemo-Resistant Triple-Negative Breast Cancer Are Prevented by Interfering with WNT-Signaling

Abstract

Triple-negative breast cancers (TNBCs), which lack specific targeted therapy options, evolve into highly chemo-resistant tumors that metastasize to multiple organs simultaneously. We have previously shown that TNBCs maintain an activated WNT10B-driven network that drives metastasis. Pharmacologic inhibition by ICG-001 decreases β-catenin-mediated proliferation of multiple TNBC cell lines and TNBC patient-derived xenograft (PDX)-derived cell lines. In vitro, ICG-001 was effective in combination with the conventional cytotoxic chemotherapeutics, cisplatin and doxorubicin, to decrease the proliferation of MDA-MB-231 cells. In contrast, in TNBC PDX-derived cells doxorubicin plus ICG-001 was synergistic, while pairing with cisplatin was not as effective. Mechanistically, cytotoxicity induced by doxorubicin, but not cisplatin, with ICG-001 was associated with increased cleavage of PARP-1 in the PDX cells only. In vivo, MDA-MB-231 and TNBC PDX orthotopic primary tumors initiated de novo simultaneous multi-organ metastases, including bone metastases. WNT monotherapy blocked multi-organ metastases as measured by luciferase imaging and histology. The loss of expression of the WNT10B/β-catenin direct targets HMGA2, EZH2, AXIN2, MYC, PCNA, CCND1, transcriptionally active β-catenin, SNAIL and vimentin both in vitro and in vivo in the primary tumors mechanistically explains loss of multi-organ metastases. WNT monotherapy induced VEGFA expression in both tumor model systems, whereas increased CD31 was observed only in the MDA-MB-231 tumors. Moreover, WNT-inhibition sensitized the anticancer response of the TNBC PDX model to doxorubicin, preventing simultaneous metastases to the liver and ovaries, as well as to bone. Our data demonstrate that WNT-inhibition sensitizes TNBC to anthracyclines and treats multi-organ metastases of TNBC.

Keywords: inhibiting WNT/β-catenin signaling; metastasis; triple-negative breast cancer.

Figure 1

Determination of differential IC₅₀ (DIC₅₀) of ICG-001 on multiple TNBC cell lines. (A) Results of WST-1 proliferation assays following 48 h of treatment with ICG-001 at various dosages ranging from 0.2 μM–30 μM using MDA-MB-231, MDA-MB-157, HCC38, MDA-MB-468 and TNBC PDX patient-derived cHCI-2 and cHCI-10 cells. All experiments in panels (B–E) used the cells’ specific IC50s as shown in Figure S1C. (B,C) qPCR for AXIN2, HMGA2, CCND1, MYC and PCNA in some of the same cells from panel A. (D,E) Immunoblot analysis for AXIN2, HMGA2, MYC, CCND1 and PCNA. (F) Immunoblot for non-phosphorylated Active-β-CATENIN (ABC) and total-β-CATENIN and are shown. β-ACTIN serves as the loading control. Results are expressed as mean ± SE, n = 3; unpaired Student’s t-test, two-tailed unequal variance used to calculate p values; *** p = 0.001, ** p ≤ 0.01 and * p ≤ 0.05 vs. control. Details of western blot can be viewed at the supplementary materials.

Figure 2

ICG-001 is able to synergize with doxorubicin, but not cisplatin, to repress tumor cell proliferation and to increase cytotoxicity in the doxorubicin chemoresistant HCI-10 PDX TNBC cells. MDA-MB-231Luc (Ai,ii) and cHCI-10Luc (Bi,ii) cells were analyzed by WST-1 assays at 96 h, following exposure to ICG-001 (1 µM and 5 µM concentration, MDA-MB-231 or 5 µM or 10 µM concentrations for cHCI-10 cells) in combination with cisplatin or DOX at various increasing concentrations, demonstrating inhibition of tumor cell proliferation. IncuCyte^® Cytox Green reagent was used to measure cytotoxicity in MDA-MB-231 (Ci,ii) and cHCI-10 (Di,ii) cells at various combinatorial concentrations over 48 h, showing an increased cytotoxicity response with ICG plus DOX. (E) MDA-MB-231 and cHCI-10 cells were exposed to ICG-001 (MDA−MB−231 = 5 µM; cHCI-10 at 10 µM) alone or in combination with DOX (0.5 µM) or with cisplatin (0.5 µM) for 48 h. Immunoblotting for total PARP, cleaved PARP, and BAX was conducted. ACTIN and TUBULIN served as loading controls. Results are expressed as mean ± SEM, n = 3. Details of western blot can be viewed at the supplementary materials.

Figure 3

WNT inhibition interferes with simultaneous multi-organ and bone metastases in vivo in MDA-MB-231 cells. MDA-MB-231 stably transduced with lentivector-luciferase was used to track cells by bio-imaging after surgical transplantation into the mammary fat pad of NSG mice, beginning one week after initiation of ICG-001 therapy (200 mg/kg, IP every other day for two weeks). (Ai) Tumor volume was tracked over time using calipers (n = 10 mice). (Aii) Representative images of tumors from the vehicle and ICG-001-treated mice. (Bi) Total bioluminescence flux (photons/sec, p/s) was quantified longitudinally in the primary tumors. Standard deviation is shown. (Bii) Ex vivo bioluminescence images of tumors. (C) Kaplan–Meier Survival curve from the vehicle and ICG-001-treated mice, n = 9/group. (D) Immunoblot analysis of HMGA2, EZH2, AXIN2, PCNA, CCND1 and MYC (i) and EMT markers VIMENTIN and SNAI from the primary tumors and (ii), β-ACTIN serves as the loading control. (E) Representative whole-body luciferase reporter images from vehicle or ICG-001 treated mice. (F) Quantification of whole-body metastasis as measured by the total bioluminescence flux in (photons/sec, p/s’ n = 6 mice/group). (G) Ex vivo bioluminescence of the organs. P-values were generated by one-way ANOVA followed by pairwise Student’s t-tests (* p < 0.05, ** p < 0.01 and *** p < 0.001). Total light flux was compared in the whole-body of the mice after therapy. Details of western blot can be viewed at the supplementary materials.

Figure 4

WNT inhibition interferes with de novo whole-body metastasis in highly chemoresistant TNBC PDX model. The TNBC PDX tumor model HCI-10, which is stably transduced with luciferase to track cells in vivo was bilaterally transplanted into the mammary fat pad of NSG mice. Three weeks after transplantation, ICG-001 therapy was initiated at a dose of either 100 or 200 mg/kg (IP, every other day for two weeks). (A) Bioluminescence images of both the primary tumors and either the lungs or lymph nodes at 6-weeks post-transplant. (B) Representative images of mice with whole-body metastases at 8 weeks after transplantation. (C) Ex vivo bioluminescence of lung, liver, ovary, kidney bone and brain harvested from the vehicle and ICG-001-treated mice. (D) Total light flux was compared in the whole body of the mice after therapy. (E) Immunoblot analysis of AXIN2, MYC (i) and VIMENTIN and SNAIL (ii) in primary tumors (ICG-001 at 200 mg/kg). β-ACTIN serves as the loading control. (F) Anti-metastatic effects of ICG-001 are shown, along with micrographs of H&E staining and anti-human mitochondria antibody staining of harvested tissue to confirm the presence of human tumor cells in rodent organs. (G) IHC was performed for CD31 and/or VEGF-A in the vehicle or ICG-001 treated cohorts for both the cHCI-10 or MDA-MB-231 primary tumors. p-Values were generated by one-way ANOVA followed by pairwise Student’s t-tests (** p < 0.01). Details of western blot can be viewed at the supplementary materials.

Figure 5

Impact of bone metastases on the bones as analyzed by high-resolution micro-computed tomography (μCT). (A) 3D images of a femur and a tibia from a control non-tumor-baring NSG mouse at 12 weeks of age, showing normal bone phenotype in both femoral and proximal tibial trabeculae bone. (B) Bones bearing metastases derived from MDA-MD-231Luc cells show the femoral and proximal tibial trabeculae; two representative mice are shown from the vehicle and ICG-001-treated mice. Red arrows (the pores) highlight loss of bone mass only in the tibial trabeculae. (C) Representative ex vivo bioluminescence images of bones from the vehicle and ICG-001-treated mice (i) and the quantification of flux-units (ii; n = 5 mice) at study endpoint. Inserts show the ex vivo bioluminescence images obtained for the same bones that were analyzed by μCT. p-Values generated by one-way ANOVA followed by pairwise Student’s t-tests (* p = 0.0225). Scale Bar of 1 mm or 100 μm (A,B).

Figure 6

ICG-001 sensitizes chemoresistant TNBC PDX tumor cells to doxorubicin, preventing liver, bone and ovarian metastasis. cHCI-10 cells (1.25 × 10⁶) that were freshly isolated from primary PDX HCI-10Luc2 tumors were tail vein injected into NSG females. One day after tail vein injection, mice were treated with either DOX alone (1.4 mg/kg, IP) or DOX in combination with ICG-001 (50 mg/kg, IP) using the dosing schedules outlined in the materials and methods. Total flux (p/s) was quantified by ex vivo bioluminescence imaging of the liver (A), bone (B) and ovaries (C); p-Values generated by unpaired Student’s t-test; two-tailed unequal variance (* p < 0.018).