Vitamin D3-dependent VDR signaling delays ron-mediated breast tumorigenesis through suppression of β-catenin activity

Abstract

The Ron receptor is upregulated in human breast cancers and correlates with enhanced metastasis and reduced patient survival. Ron overexpression drives mammary tumorigenesis through direct β-catenin activation and augmented tumor cell proliferation, migration and invasion. Ron and β-catenin are also coordinately elevated in breast cancers. The vitamin D receptor (VDR) antagonizes β-catenin signaling. Herein, we examined mammary tumor onset and progression using a Ron-driven murine model of breast tumorigenesis crossed with VDR deficient mice. VDR ablation accelerated mammary tumor onset and led to tumors that exhibited a desmoplastic phenotype and enhanced metastases. Tumor levels of active β-catenin were markedly increased in the absence of VDR. In vitro, VDR activation in breast cancer cells reduced β-catenin activation and transcriptional activity leading to elevated expression of the extracellular Wnt inhibitor dickkopf-related protein 1, and a reduction in the interaction of β-catenin with the cyclin D1 promoter. Expression of a stabilized form or β-catenin ablated the protective effects of VDR activation.Collectively, these studies delineate a protective role for VDR signaling in Ron-induced mammary tumorigenesis through disruption of β-catenin activation.

Keywords: breast cancer; ron; vitamin D3 receptor; β-catenin.

Figure 1. VDR signaling delays Ron-mediated mammary gland hyperplasia

A. Western blot of mammary lysates from MMTV-Ron VDR+/+ and VDR−/− mice depicting Ron expression levels. B. qRT-PCR analysis of Ron and VDR mRNA expression in mammary glands (MG) and tumors from MMTV-Ron VDR+/+ and VDR−/− mice. Data represent mean values from three independent experiments ± SE. C. Representative mammary whole mounts (upper panels) and H&E-stained tissue sections (lower panels) from 10 week-old MMTV-Ron VDR+/+ and VDR−/− mammary glands (n = 8-11). D. Incidence of mice with hyperplastic mammary glands at 2.5, 4 and 6 months of age from MMTV-Ron VDR+/+ and VDR−/− animals. *P < 0.05.

Figure 2. VDR signaling delays mammary tumor formation and alters disease progression in MMTV-Ron mice

A. Mammary tumor kinetics from MMTV-Ron VDR+/+, VDR+/−, and VDR−/− mice. Between 22 and 33 female mice per genotype were examined temporally for palpable tumor formation. The percent tumor-free mice is plotted as a function of age and is statistically significant with complete VDR ablation compared to VDR haploinsufficiency and VDR wild type mice. Inset: The mean time-to-tumor onset was plotted for each genotype. B. Representative mammary glands (upper panels) and tumors (lower panels) from 8 month-old MMTV-Ron VDR+/+ and VDR−/− mice showing enhanced ductal dilation and desmoplasia with VDR ablation (n = 9-11 per genotype). C. Representative immunohistochemical staining for Ron in lung (upper panels) and liver (lower panels) sections from 10 month-old MMTV-Ron VDR+/+ and VDR−/− mice showing more metastases with loss of VDR (n = 5-6 per genotype). D. The average number of metastases per lung (n = 24-25) and liver (n = 14-19) per genotype as assessed by histological analysis is plotted. Data represent mean values ± SE. *P < 0.05.

Figure 3. Enhanced downstream β-catenin signaling in Ron-mediated mammary tumorigenesis with VDR ablation

A. Representative immunohistochemical staining for β-catenin in MMTV-Ron VDR+/+ and VDR−/− mammary glands from 4 month-old mice demonstrating enhanced expression in the absence of VDR (n = 6 per genotype). B. qRT-PCR mRNA expression of β-catenin target genes Cyclin D1, c-Myc, TCF-7, VEGF, and MMP7 in mammary tumors from 8 month-old MMTV-Ron VDR+/+ and VDR−/− mice. Data represent mean values from three independent experiments ± SE. C. Western analysis demonstrating enhanced expression of active β-catenin (ABC) levels in two representative MMTV-Ron VDR−/− tumor lysates compared to two MMTV-Ron VDR+/+ tumors that correlates with increased expression of β-catenin target genes cyclin D1 and c-Myc. *P < 0.05.

Figure 4. Ron receptor status determines epithelial cell sensitivity to vitamin D₃-mediated growth inhibition and delays in migration and invasion

A. Western analysis of levels of VDR and Ron ± 100 nM 1,25D₃ in R7 cells. B. R7 cells were treated with the designated concentrations of 1,25D₃ for 72 hours and cell viability/number was determined by crystal violet assays. Data are normalized to vehicle treated cells set at 1 and represent mean values from three independent experiments ± SE. C. R7 cells treated with the designated concentrations of 1,25D₃ starting at the 0 hour time point were examined in scratch assays for the percent of gap closure after 4 and 8 hours. Data represent mean values from three independent experiments ± SE. D. Matrigel invasion assay with R7 cells in the presence or absence of 100 nM 1,25D₃. Data represent mean values from four independent experiments ± SE. E. Western blot demonstrating the reduction in Ron expression in T47D cells stably transfected with a shRon lentivirus. F. qRT-PCR analysis of VDR, Ron and β-catenin mRNA expression in control T47D cells (shNT) and Ron knockdown T47D cells (shRon). Data represent mean values from three independent experiments ± SE. G. MTT assays of T47D control (shNT) and shRon cells treated with the designated concentrations of 1,25D₃ for 48 hours. Data represent mean values from 3-4 independent experiments per cell line ± SE. H. Scratch assays in T47D control (shNT) and shRon cells treated with 100 nM 1,25D₃ for 24 hours. Data represent mean values normalized to the vehicle-treated control group of each cell type and are from three independent experiments ± SE. *P < 0.05.

Figure 5. Vitamin D₃-dependent VDR signaling reduces β-catenin transcriptional activity

A. Western analysis demonstrating a dose-dependent reduction in active β-catenin (ABC) levels in R7 cells treated with the designated concentrations of 1,25D₃ for 48 hours, but no change in total β-catenin. B. qRT-PCR analysis of β-catenin target genes cyclin D1, TCF-7, VEGF, and MMP7 in R7 cells treated with the designated concentrations of 1,25D₃ for 72 hours. Data represent mean values from three independent experiments ± SE. C. Dual luciferase assays in R7 cells co-transfected with Topflash and pRLTX (expressing Renilla) plasmids for 48 hours, then treated with the designated concentrations of 1,25D₃ for 24 hours. Luciferase activities were normalized for transfection efficiency to Renilla activity. Data represent mean values from three independent experiments ± SE. D. Crystal violet staining of R7 cells transfected with a stabilized form of β-catenin (ΔN) or empty vector and treated with 100 nM 1,25D₃ or vehicle control (EtOH). Data represent mean values from four independent experiments ± SE. *P < 0.05.

Figure 6. Vitamin D₃-dependent VDR signaling induces DKK-1 expression and binds to β-catenin to disrupt interaction at consensus sequences within promoters of TCF/LEF target genes

A. qRT-PCR mRNA expression of DKK-1 in R7 cells treated with the designated concentrations of 1,25D₃ for 72 hours. Data represent mean values from three independent experiments ± SE. B. qRT-PCR mRNA expression of DKK-1 in tumors and mammary glands (MG) from 8 month-old MMTV-Ron VDR+/+ and VDR−/− mice demonstrating a reduction in DKK-1 levels in tumors with loss of VDR. Data represent mean values from three independent experiments ± SE. C. Western blot demonstrating loss of DKK-1 protein expression with siRNA-mediated silencing in R7 cells. D. R7 cells transfected with siRNA against DKK-1 were treated with the designated concentrations of 1,25D₃ for 72 hours and cell viability/number was determined by crystal violet assays. Data are normalized to the respective vehicle treated cells set at 1 and represent mean values from two independent experiments performed in quadruplicate ± SE. E. Chromatin immunoprecipitation (ChIP) assays with a mouse IgG isotype control and an anti-active β-catenin (ABC) antibody. ChIP-ABC quantitative real time PCR (qRT-PCR) analysis of R7 cells treated with 100 nM 1,25D₃ for 72 hours. The graph shows qRT-PCR on DNA purified from ChIP-ABC, using primers designed to the LEF-1 binding sequence within the mouse cyclin D1 promoter and relative to the respective input controls. Vitamin D₃ treatment significantly reduces enrichment compared to the vehicle control (EtOH). Data represent mean values from three independent experiments ± SE. F. Representative agarose gel from ChIP-ABC PCR showing reduced cyclin D1 promoter enrichment with vitamin D₃ treatment in R7 cells. Negative control primers designed to a site 4000 bp upstream of the LEF-1 binding sequence within the cyclin D1 promoter (Off-Target) verify specificity of cyclin D1 primers to the sheared DNA product. G. Densitometry analysis of PCR products from three separate ChIP-ABC experiments supporting loss of ABC interaction with the cyclin D1 promoter. Error bars represent SE. H. Re-ChIP qRT-PCR analysis of R7 cells treated with 100 nM 1,25D₃ for 72 hours and sequentially immunoprecipitated with anti-ABC then anti-VDR antibodies. The graph shows qRT-PCR of DNA purified from ChIP-ABC-VDR, using primers designed to the LEF-1 binding sequence within the mouse cyclin D1 promoter, and showing less interaction of the ABC-VDR complex at the cyclin D1 promoter. Data represent the relative mean CT values from two experiments ± standard deviation. *P < 0.05.

Figure 7. A working model of VDR signaling in Ron-mediated breast cancer

Binding of the Ron ligand, HGFL, to Ron induces β-catenin activation promoting breast cancer growth and metastasis when aberrantly expressed. VDR signaling in breast cancer cells antagonizes β-catenin activity, delaying tumor initiation and progression through two mechanisms: 1) Vitamin D₃-bound VDR can directly interact with β-catenin, preventing its association with TCF/LEF transcription factors thus inhibiting transcriptional activity; and, 2) ligand-dependent VDR signaling can regulate the transcription of various VDR target genes. One such VDR target includes DKK-1, which inhibits canonical Wnt signaling, thereby reducing cytosolic accumulation of β-catenin and nuclear localization for transcriptional activity.