Cripto-1 as a novel therapeutic target for triple negative breast cancer

Abstract

Triple-negative breast cancer (TNBC) presents the poorest prognosis among the breast cancer subtypes and no current standard therapy. Here, we performed an in-depth molecular analysis of a mouse model that establishes spontaneous lung metastasis from JygMC(A) cells. These primary tumors resembled the triple-negative breast cancer (TNBC) both phenotypically and molecularly. Morphologically, primary tumors presented both epithelial and spindle-like cells but displayed only adenocarcinoma-like features in lung parenchyma. The use of laser-capture microdissection combined with Nanostring mRNA and microRNA analysis revealed overexpression of either epithelial and miRNA-200 family or mesenchymal markers in adenocarcinoma and mesenchymal regions, respectively. Cripto-1, an embryonic stem cell marker, was present in spindle-like areas and its promoter showed activity in primary tumors. Cripto-1 knockout by the CRISPR-Cas9 system inhibited tumor growth and pulmonary metastasis. Our findings show characterization of a novel mouse model that mimics the TNBC and reveal Cripto-1 as a TNBC target hence may offer alternative treatment strategies for TNBC.

Trial registration: ClinicalTrials.gov NCT01151449.

Keywords: Cripto-1; Notch4; epithelial-mesenchymal plasticity; mouse model; triple-negative breast cancer.

Conflict of interest statement

CONFLICTS OF INTEREST

No potential conflicts of interest disclosed

Figures

Figure 1. Orthotopic metastasis of JygMC(A) cells and epithelial mesenchymal characterization in vitro

Lung metastasis by orthotopic cell injection in Balb/C nude mice. B. Representation of bioluminescent imaging of animals. Animals imaged at day 15, 22, 29, 42 and 52 post JygMC(A)-GFP/Luc cell injection. Ex-vivo imaging of lymph nodes and lung metastases. C. Immunofluorescence of luminal epithelial and basal mesenchymal markers in JygMC(A) parental cells. Epithelial marker: CDH1 (Alexa 488), luminal marker: CK18 (Alexa 488), basal marker: K5 and K14 (Alexa 488) and mesenchymal marker: Vimentin (Alexa 594). Nuclear staining in blue (DAPI). Scale bars: 200 μm.

Figure 2. JygMC(A) mammary primary tumor molecular phenotype, histology and EMT-MET plasticity

Analysis of normal adjacent tissue by immunohistochemistry of ERα A. PR B. and HER2/neu staining of a wild-type MMTV-Neu/Her2 tumor C. Analysis of primary mammary tumor tissue of ERα D. PR E. and HER2/neu F. Positive antibody signals are shown in brown, and the hematoxylin counterstain is shown in blue. Scale bars: 200μm. G. Histological tumor classification. Primary tumor possesses two types of cells: undifferentiated adenocarcinoma type (green line) and areas composed of atypical spindle-shaped cells suggesting EMT (yellow line). Lung metastasis morphology typical of the epithelial primary adenocarcinoma suggesting MET is present in the lung parenchyma. Lymph node tissue (red line) showing areas of atypical spindle-shaped cells (yellow line). Shown are hematoxylin and eosin staining. Scale bars: 100 μm.

Figure 3. Gene expression profiles of laser capture microdissected tumor progression samples through unsupervised hierarchical clustering

Primary tumor adenocarcinoma versus EMT-like areas; B. Primary tumor adenocarcinoma versus lung metastasis; C. Primary tumor EMT-like areas versus lung metastasis; D. Primary tumor, lung metastasis, normal mammary gland and normal lung parenchyma. Each row represents a single gene and each column a single sample. The red color indicates up-regulation, the green color indicates down-regulation, and the black color indicates no change in expression level compared with the reference sample. The gray color indicates that no intensity was detected. EMT: epithelial-to-mesenchymal transition.

Figure 4. Relevant pathways, gene and protein expression in primary tumors and lung metastasis and similarities with human breast cancer

Enriched pathways in primary mammary tumors. B. Enriched pathways in lung metastasis. C. Common seven genes between Nanostring and microarray analysis D. BMP4 and CDH1 protein expression. E. Differentially expressed genes between JygMC(A) primary mammary tumors and/or lung metastasis and human breast cancer subtypes and/or metastasis datasets.

Figure 5. miRNA regulation and Notch4 expression in vitro and in vivo

mRNA and microRNA network in JygMC(A) primary mammary tumor model B. Immunofluorescence staining of Notch4 in adherent JygMC(A) parental cell line C. Immunohistochemistry of Notch4 in primary tumor derived from JygMC(A)-GFP/Luc cells and lung metastasis. Brown staining represents positivity. Arrows indicate nuclear staining. D.Notch4 gene expression throughout microarray samples.

Figure 6. In vitro and In vivo effects of the gamma-secretase inhibitor RO4929097 in the JygMC(A) cell line and mouse model

Proliferation assay. JygMC(A)-GFP/Luc cells were seeded in 12-well dishes in triplicate at 5×103cells/well and cultured for 24, 48 and 72 hrs. Cells were then harvested and counted. Data are representative of two independent experiments in triplicate ±SD, *P < 0.0001, as compared to control vehicle-treated cells. B. Colony formation assay in soft agar. A total of 1.5×104cells were cultured for 10 days. Colonies were stained with NitroBlue Tetrazolium and quantified using Gelcount. Data are representative of two independent experiments in triplicate ±SD, *P < 0.005, as compared to control vehicle-treated cells. C. Boyden chamber migration and D. invasion assays. A total of 4×104cells in serum-free medium were seeded on the top chambers, and 2% or 5% FBS-containing medium was placed in the bottom wells as a chemoattractant. Cells were incubated for 24hrs. Cells that migrated/invaded through the membrane were stained and counted using a light microscope. Data are expressed as the average of cells in ten fields from each membrane (20X). Data are representative of two independent experiments in duplicate ±SD, *P < 0.01 and **P < 0.001, as compared to control vehicle-treated cells. E. Sphere formation assay. A total of 10,000 cells were cultured for 10 days on non-adherent plates and treated with different concentrations of RO4929097. Data are representative of two independent experiments in sixplicate ±SD, *P < 0.001, as compared to control vehicle-treated cells. F. Nude mice bearing mammary tumors (n = 10/group) were dosed orally following the schedule of 60mg/Kg every other week (7days on/7days off) or vehicle for 4 weeks and primary tumor volumes are represented in the graph, ±SD, *P < 0.05, as compared to control vehicle-treated animals. G. Number of pulmonary nodules per animal in control vehicle-treated and RO4929097-treated animals (*P < 0.05, one-sided values; Wilcoxon rank-sum test).

Figure 7. Mouse Cripto-1 promoter and Cripto-1 knockout in JygMC(A) cell line

Representation of bioluminescent imaging on animals injected bilaterally into the fourth mammary gland with JygMC(A) cells containing the mouse Cripto-1 promoter. Animals were imaged at day 8, 16, 19 and 23-post cell injection. B.In situ detection of Cripto-1 promoter activity in primary tumor tissue sections. C. H&E of the primary tumor tissue sections depicted in B. showing carcinoma areas (green) and EMT-like areas (yellow) and high magnification (20X). Scale bars: 3mm and 200μm. D. Proliferation assay. JygMC(A) cells were seeded in 12-well dishes in triplicate at 5x104cells/well and cultured for 24, 48 and 72 hrs. Cells were then harvested and counted. Data are representative of two independent experiments in triplicate ±SD, *P < 0.0002, as compared to control cells. E. Average of primary tumor volumes are represented in the graph on JygCr-1KO and JygMC(A) parental cells (n = 7 animals/group ±SD, *P < 0.01 and **P < 0.0001, as compared to control animals). F. Number of pulmonary nodules per animal in JygCr-1KO and JygMC(A) parental animals (*P < 0.05, one-sided values; Wilcoxon rank-sum test).