Activation of Bone Morphogenetic Protein Signaling by a Gemini Vitamin D₃ Analogue Is Mediated by Ras/Protein Kinase Cα

The pivotal role of 1α,25-dihydroxyvitamin D₃ [1α,25(OH)₂D₃], the hormonally active form of vitamin D₃, comprises the maintenance of calcium/phosphate homeostasis and bone biology (1). In addition to its well-established actions, many studies have shown that 1α,25(OH)₂D₃ exerts antiproliferative, proapoptotic, and prodifferentiating effects on many different cell types (1, 2). Furthermore, 1α,25(OH)₂D₃ and certain vitamin D₃ analogues have been shown to inhibit invasion, angiogenesis, and metastasis associated with certain tumors (3, 4). We previously reported that 1α,25(OH)₂D₃ and a novel Gemini vitamin D₃ analogue, Ro-438-3582 [Ro3582; 1α,25-dihydroxy-20S,21(3-hydroxy-3-methylbutyl)-23-yne-26,27-hexafluorocholecalciferol] (Fig. 1), inhibited the proliferation of breast epithelial cells and activated the bone morphogenetic protein (BMP) signaling pathway (5, 6).

BMPs are members of the transforming growth factor-β (TGF-β) superfamily, which are now generally considered as multifunctional cytokines that affect inflammation, immune response, cell proliferation, differentiation, and apoptosis (7). As do TGF-βs, BMPs regulate their biological cell responses through binding to two types of serine/threonine kinase receptors, which trigger phosphorylation of the receptor-mediated Smad (R-Smad) at the COOH-terminal of the MH2 domain (8). The activated R-Smads recruit the common Smad, Smad4, and translocate to the nucleus to mediate the transcription of BMP-dependent target genes, primarily during embryonic development and bone formation (8).

Interestingly, there is an accumulation of data suggesting the possible role of BMP as a tumor suppressor (9–14). For example, activation of the BMP signaling pathway inhibited the growth of epithelial cells by inducing the cyclin dependent kinase (CDK) inhibitor, p21 (9, 10), and in vivo tumor growth of androgen-insensitive prostate carcinoma cells was suppressed by BMP signals (11). In addition, cancer-associated stromal cells expressed the BMP antagonist, gremlin 1, which can promote tumor cell growth (12). Mutation or loss of expression of molecules in the BMP signaling pathway led to the enhancement of tumor progression (13, 14), indicating that BMP signaling may be important for inhibition of tumorigenesis. Therefore, modulating BMP signaling during the formation of breast cancer by pharmacologic agents, such as 1α,25(OH)₂D₃ and vitamin D₃ analogues, may be important for the prevention and possible treatment of breast cancer.

We recently reported that 1α,25(OH)₂D₃ and a vitamin D₃ analogue Ro3582 activated the BMP signaling pathway, as shown by enhanced phosphorylation of the MH2 domain (Ser^463/465) of Smad1/5 in breast epithelial cells (5, 6). We have now further explored the upstream kinase signaling pathways responsible for activation of BMP signaling by Ro3582 and for its role in growth inhibition of breast epithelial cells. The effects of 1α,25(OH)₂D₃ and its analogues are mainly mediated through the vitamin D receptor (VDR) or through the membrane-associated signaling pathway (1, 15). Membrane-associated responses to 1α,25(OH)₂D₃ [nongenomic, rapid response to 1α,25(OH)₂D₃], where the mechanism is still unclear, is now considered an essential type of action involved in calcium/phosphate transport, activation of protein kinase C (PKC), and/or the mitogen-activated protein kinase (MAPK) cascade (15–18).

Among many kinases, PKC has been shown to be regulated by 1α,25(OH)₂D₃ and certain several vitamin D₃ analogues (17, 19, 20). The PKC family is a group of serine/threonine kinases known to regulate cell growth, apoptosis, differentiation, cell migration, and carcinogenesis in different types of cells and models (21, 22). Boyan et al. (19, 20) suggest that a caveolar environment may play an important role in mediating the PKC activation by 1α,25(OH)₂D₃. Studies investigating the effects of 1α,25(OH)₂D₃ and vitamin D₃ analogues on the PKC family have shown that regulation of the MAPK pathway by 1α,25(OH)₂D₃ and vitamin D₃ analogues was mediated by the activation of RAF-1/Ras/PKC in muscle cells and myeloid leukemic cells (17, 18).

Interestingly, several studies have reported that PKC interacts with the TGF-β/BMP signaling pathway (23–25). PKC-dependent phosphorylation of the MH1 domain of TGF-β–specific Smads (Smad2/3) led to down-regulation of the growth-inhibitory and apoptotic action of TGF-β (23). In contrast, BMP-2 enhances apoptosis by using the PKC-dependent signaling pathway in human osteoblasts (24). It was also reported that Smad6 regulates TGF-β and plasminogen activator inhibitor-1 through a PKC-β–dependent mechanism (25), and Smad3 and PKCδ mediate TGF-β1–induced collagen I expression in human mesangial cells (26). These studies suggest a new mechanism for the regulation of PKC by 1α,25(OH)₂D₃ and vitamin D₃ analogues that may affect the TGF-β/BMP signaling pathway.

In the present study, we investigated whether PKC is involved in the activation of BMP/Smad signaling by a potent Gemini vitamin D₃ analogue Ro3582, and whether this is important for the inhibition of cell proliferation of breast epithelial cells. We report here that Ro3582 activates Smad signaling and inhibits the proliferation of MCF10AT1 cells through a Ras/PKCα pathway.

Reagents. 1α,25-Dihydroxyvitamin D₃ [1α,25(OH)₂D₃] and a Gemini vitamin D₃ analogue Ro3582 (>95% purity; Fig. 1) were provided by Hoffmann-La Roche, Inc. PKC inhibitors (Go6976, Go6983, and PKCβ C2-4 inhibitor), MAP/extracellular signal-regulated kinase (ERK) kinase (MEK) inhibitors (PD98059 and U0126), p38 inhibitor (SB203580), c-Jun-NH₂-kinase (JNK) inhibitor (SP600125), PKA inhibitor (H-89), and the AKT inhibitor [a phosphatidylinositol ether analogue, 1L6-hydroxymethyl-chiro-inositol-2-(R)-2-O-methyl-3-O-octadecyl-sn-glycerocarbonate] were obtained from Calbiochem. Other chemicals, including TGF-β receptor I kinase inhibitor (SB431542) and phosphoinositide-3 kinase (PI3K) inhibitor (LY294002·HCl), were from Sigma. Ras farnesyltransferase inhibitor (L-744832) was from Biomol. Vitamin D₃ analogues and inhibitors were dissolved in DMSO before addition to cell cultures, and final concentrations of DMSO were 0.1% or less. Controls with DMSO alone were run in all cases.

Cell culture. Human MCF10 breast epithelial cell lines (MCF10A and MCF10AT1) were provided by Dr. Fred Miller's group (Barbara Ann Karmanos Cancer Institute, Detroit, MI). The MCF10AT1 breast epithelial cell line was developed by transfecting Ha-ras oncogene into MCF10A normal immortalized breast epithelial cells, and the cell line was then passaged in mice to select more aggressive and malignant cells (27, 28). MCF10A and MCF10AT1 cells were maintained in DMEM/F12 medium supplemented with 5% horse serum, 1% penicillin/streptomycin, 10 μg/mL insulin, 20 ng/mL epidermal growth factor (EGF), 0.5 μg/mL hydrocortisone, and 100 ng/mL cholera toxin at 37°C, 5% CO₂.

Western blot analysis. MCF10A and MCF10AT1 cells were plated and starved for 24 h in serum-free DMEM/F12 medium. Cells were then incubated with compounds in 0.1% bovine serum albumin (BSA)/DMEM/F12 medium for the indicated times, as described in the figure legends. The proteins were extracted by cell lysis with radioimmunoprecipitation assay buffer (10 mmol/L Tris-HCl, 5 mmol/L EDTA, 150 mmol/L NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 0.1 mmol/L Na₃VO₄, 1% phenylmethylsulfonyl fluoride, 1% aprotinin, and 0.1% leupeptin). The same amount of protein was run in 4% to 15% gradient gel (Bio-Rad) and transferred to the polyvinylidene difluoride membrane (PALL). The primary antibodies against phospho-Smad1/5, phospho-Smad3, PKCα (Cell Signaling Technology, Inc.), hemagglutinin (Covance), actin (Sigma), and secondary antibodies (Santa Cruz Biotech, Santa Cruz, CA) were used.

Quantitative PCR analysis. Total RNA was isolated from cultured cells using the Trizol method from Invitrogen. One microgram of total RNA was reverse transcribed to cDNA using the random primers and Applied Biosystem High Capacity cDNA Archive Kit in a 96-well format Mastercycler Gradient from Eppendorf North America. Quantitative PCR was performed using Applied Biosystems Taqman Gene Expression Assay reagents on an ABI Prism 7000 Sequence Detection System. The thermal conditions were as follows: one cycle of 50°C for 2 min, one cycle of 95°C for 10 min, 40 cycles of 95°C for 15 s, and 60°C for 1 min. Labeled primers, including glyceraldehyde-3-phosphate dehydrogenase (GAPDH), BMP2, BMP6, and CYP24A1, were obtained from Applied Biosystems. GAPDH was used as an internal control. The relative changes in gene expression were calculated by the following formula: fold change = 2^(−ΔΔCt) = 2^{−[ΔCt (treated samples) − ΔCt (vehicle control)]}, where, ΔC_t = C_t (detected gene) − C_t (GAPDH) and C_t is the threshold cycle number.

Transient transfection of PKC isoforms. The vectors containing PKCα, PKCδ, PKCε, or PKCζ isoforms linked to hemagglutinin were kindly provided by Drs. Bernard Weinstein (Department of Medicine, Columbia University, New York, NY) and Jae-Won Soh (Department of Chemistry, Inha University, Incheon, Korea; ref. 29), and we subcloned the fusion protein of green fluorescent protein (GFP)–PKCα using pEGFP-C1 vector (BD Bioscience Clontech). For the transient transfection of PKCα, we incubated DNA with jetPEI transfecting agent (Poly-plus Transfection) in 150 mmol/L NaCl for 20 min, and this was directly added to the cells in MCF10AT1 culture medium. After 24 h, the cells were starved in serum-free DMEM/F12 overnight and then treated with experimental test compounds in 0.1% BSA/DMEM/F12.

Fluorescence microscopy. The immunofluorescence procedure was described previously (6). Briefly, MCF10AT1 cells were incubated with compounds in a poly-l-lysine–coated chamber slide (Nunc) or in a glass-bottomed dish (MatTek) for the indicated times. Then, cells were fixed in 4% paraformaldehyde [1× PBS (pH 7.4)] for 20 min, and blocked for 1 h with 10% BSA/0.5% Triton-X/1× PBS solution. The primary antibody (1:100 dilution for PKCα) and fluorophore-conjugated secondary antibody (Molecular Probes) were probed in 10% BSA/PBS solution. The cells were irradiated with a green laser (488 nm), and UV light (364 nm) was used for 4′,6-diamidino-2-phenylindole (DAPI) staining. GFP-PKCα fusion protein was directly detected by green fluorescence irradiation at 488 nm after transient transfection.

[³H]thymidine uptake assay. MCF10A and MCF10AT1 cells were incubated with compounds for 3 days. [³H]thymidine (1 μCi) was added to each well 3 h before the harvest. The cells were precipitated with 10% trichloroacetic acid, washed, and solubilized (30), and the incorporation of [³H]thymidine into the cells was analyzed with a liquid scintillation counter (Beckman Coulter).

Statistical analysis. Statistical significance was evaluated using the Student's t test.

The activation of Smad signaling by Ro3582 is blocked by PKCα/PKCβI inhibitor Go6976 in MCF10AT1 breast epithelial cells. In previous studies, we showed that 1α,25(OH)₂D₃ and Ro3582 inhibited the proliferation and activated Smad signaling, and Ro3582 exerted a much stronger effect than 1α,25(OH)₂D₃ (5, 6), as determined by the phosphorylation of Smad1/5 in MCF10AT1 breast epithelial cells. Here, we tested seven different serine/threonine kinase and PI3K inhibitors to investigate the upstream cell signaling pathways that may be responsible for the activation of Smad1/5 signaling. Among the inhibitors tested, the PKC inhibitor (Go6976, an inhibitor of the classic Ca²⁺-dependent PKCα and PKCβI isoforms) blocked the phosphorylation of Smad1/5 induced by Ro3582, whereas the PI3K inhibitor (LY294002·HCl), the MEK inhibitors (PD98059 and U0126), the p38 inhibitor (SB203580), the JNK inhibitor (SP600125), the PKA inhibitor (H-89), the AKT inhibitor (a phosphatidylinositol ether analogue), and the TGF-β type I receptor inhibitor (SB431542) showed little or no effect on the level of phospho-Smad1/5 induced by Ro3582 (Fig. 2A).

PKCα/PKCβI inhibitor Go6976 inhibits BMP2/6 and CYP24A1 mRNA synthesis induced by Ro3582. Next, we tested whether the PKCα/PKCβI inhibitor Go6976 regulates the target genes of BMP and VDR signaling induced by Ro3582. MCF10AT1 cells were incubated with Ro3582 with or without Go6976, and mRNA productions of BMP-2, BMP-6, and CYP24A1 were measured by quantitative PCR. BMP-2 and BMP-6 are ligands of the BMP signaling pathway. CYP24A1 (vitamin D 24-hydroxylase) is a prominent vitamin D response gene and it is known to inactivate 1α,25(OH)₂D₃ by hydroxylation in the 24-position. As shown in Fig. 2B, mRNA levels for BMP-2, BMP-6, and CYP24A1 were increased by Ro3582 at 12 h and more markedly at 24 h confirming our earlier studies (5, 6), and these inductions were significantly inhibited by Go6976.

Growth inhibition by Ro3582 is reversed by PKCα/PKCβI inhibitor Go6976 in MCF10AT1 breast epithelial cells. We previously reported that Ro3582 activates Smad signaling and inhibits cell growth in MCF10AT1 cells (5, 6). Here, we used the PKCα/PKCβI inhibitor Go6976 to determine whether blocking Smad signaling by Go6976 would affect growth inhibition by Ro3582 in MCF10AT1 cells. As shown in Fig. 2C, Ro3582 exerted ∼60% growth inhibition at 1 nmol/L, whereas Go6976 itself has little effect on growth inhibition at 0.1 μmol/L. When cells were treated with Ro3582 together with Go6976, the PKCα/PKCβI inhibitor Go6976 significantly reversed the growth inhibition induced by Ro3582 (P < 0.001).

PKCα is a crucial mediator for the phosphorylation of Smad1/5 by Ro3582. Because the PKC family has multiple isoforms, we next determined which PKC isoforms may be involved in the activation of Smad signaling by Ro3582. We first tested the different types of PKC inhibitors, such as Go6976 (PKCα/PKCβI inhibitor), Go6983 (PKCα, PKCβ, PKCγ, PKCδ, PKCζ inhibitor) and PKCβ C2-4 inhibitor (PKCα/PKCβ inhibitor), at several concentrations. The phosphorylation of Smad1/5 induced by Ro3582 was inhibited by all of these PKC inhibitors, suggesting that PKCα and/or PKCβ might be involved in Smad activation by Ro3582 (Fig. 3A). In addition, we tested four different PKC isoforms: PKCα (a classic PKC isoform), PKCε and PKCδ (novel PKC isoforms), and PKCζ (an atypical PKC isoform). These vectors were linked to a hemagglutinin tag and the transfection of vectors was confirmed by expression of hemagglutinin (Fig. 3B). Transfection with the control vector (pcDNA) showed a lower level of pSmad1/5 induction by Ro3582 compared with that without transfection (data not shown). Overexpression of PKCα enhanced the phosphorylation of Smad1/5 induced by Ro3582. However, overexpression of PKCε, PKCδ, or PKCζ showed little or no effect on the Ro3582-mediated increase in the level of pSmad1/5 (Fig. 3B).

Ro3582 activates PKCα in MCF10AT1 breast epithelial cells. PKCα is translocalized to the plasma membrane after calcium and diacylglycerol binding to C2 and C1 domain, respectively, which can lead to enzymatic activation (22). Using confocal microscopy, the cellular distribution of PKCα was determined after treatment with Ro3582 or 12-O-tetradecanoylphorbol-13-acetate (TPA), a well-known PKC activator, in MCF10AT1 cells. DAPI staining was used to recognize the nuclear morphology in cells. We first used PKCα antibody to detect the location of endogenous PKCα in MCF10AT1 cells. When the cells were treated with TPA (10 nmol/L), PKCα translocated to the membrane within 1 h. However, Ro3582 (10 nmol/L) did not change the cellular location of PKCα at 1 h. Interestingly, after 24 h of Ro3582 treatment, PKCα was markedly translocated to the membrane (Fig. 3C). We next transfected GFP-PKCα vector (PKCα linked to a fluorescent marker, GFP) and PKCα was detected directly using green fluorescence (GFP) at 488 nm. We confirmed that both Ro3582 and TPA triggered the translocation of PKCα to the membrane (Fig. 3D).

Ras is necessary for the induction of pSmad1/5 by Ro3582. Because studies of the interregulation between Ras and TGF-β/BMP signaling were reported earlier (31–35), we investigated whether the transfected Ras in MCF10AT1 cells may affect the regulation of Smad signaling induced by Ro3582. MCF10AT1 cells were established by transfecting Ha-ras into MCF10A normal breast epithelial cells and passaging in animals (27, 28). As shown in Fig. 4A, both MCF10A and MCF10AT1 cell lines showed the same level of response to TGF-β or BMP-2 treatment, determined by the phosphorylation of Smad3 by TGF-β and the phosphorylation of Smad1/5 by BMP-2. However, Ro3582 increased pSmad1/5 in Ha-ras–transfected MCF10AT1 cells, but not in the parent MCF10A cells. Furthermore, Ro3582 inhibited cell proliferation in MCF10AT1 cells, but not in the parent MCF10A cells (Fig. 4B), suggesting the critical role of Ras in Smad activation and growth inhibition by Ro3582. In addition, inhibiting Ras activity by a Ras farnesyltransferase inhibitor (Ras inhibitor, L-744832) blocked the phosphorylation of Smad1/5 induced by Ro3582 in a dose-dependent manner in MCF10AT1 cells (Fig. 4C), although the Ras inhibitor did not reverse growth inhibition induced by Ro3582 (Fig. 4D), which may be due to the growth-inhibitory effect of the Ras inhibitor by itself (data not shown).

PKCα activation by Ro3582 is blocked by a Ras farnesyltransferase inhibitor in MCF10AT1 cells. The integration between Ras/MAPK and PKC has been reported (36–38) and we found that both PKCα and Ha-ras regulated the phosphorylation of Smad1/5 that was induced by Ro3582. Therefore, we examined whether Ras signaling is necessary to regulate PKCα, which in turn enhances Smad signaling by Ro3582. Here, the treatment with the Ras inhibitor did not affect PKCα distribution in MCF10AT1 cells. However, Ro3582 induced the cellular translocation of PKCα to the membrane, and when treated together with the Ras inhibitor, the translocation of PKCα to the membrane induced by Ro3582 was markedly inhibited (Fig. 5).

BMPs, members of the TGF-β superfamily, have been identified as multifunctional regulators of development, bone formation, and tissue remodeling. Recently, the role of BMPs in proliferation, differentiation, apoptosis, and angiogenesis has drawn much attention (9–14, 39, 40). Although BMPs are known to regulate growth, differentiation, and apoptosis in many different cell types and models, the role of BMPs as tumor suppressors in breast cancer is still unsettled (41, 42).

We recently reported that a novel Gemini vitamin D₃ analogue, Ro3582, strongly inhibited the growth of human MCF10AT1 breast epithelial cells (5, 6). Mechanistic studies indicated that this compound activated BMP/Smad signaling, as determined by increased phosphorylation of Smad1/5, translocation of pSmad1/5 to the nucleus, and enhancement of the BMP/Smad transcriptional activity (6). In the present report, we identified key upstream signaling pathways responsible for BMP/Smad signaling that are activated by Ro3582. We found that the vitamin D₃ analogue Ro3582 activated PKCα, which led to the phosphorylation of Smad1/5 and inhibition of the growth of MCF10AT1 cells. Furthermore, Ras was involved in the activation of PKCα and Smad signaling by Ro3582, suggesting that both Ras and PKCα may act as crucial mediators of Ro3582 effects on MCF10AT1 breast epithelial cells.

Many studies of cross-talk between Ras and TGF-β/Smad signaling have been described during the last decade (31–35). Yue et al. (31) reported that TGF-β activated the Ras/MAPK pathway required for the autocrine TGF-β production and Smad1 regulation. Ras was also suggested as a mediator of pleiotropic TGF-β1 signaling in developing neurons (32). The activation of Ras/MAPK or the presence of oncogenic Ras was shown to enhance TGF-β–induced epithelial-mesenchymal transition (33, 34). More importantly, it has been shown that 1α,25(OH)₂D₃ regulates the MAPK pathway by activating Ras/RAF-1 signaling in muscle cells and in myeloid leukemic cells (17, 18). We showed in this report that the vitamin D₃ analogue Ro3582 increased phosphorylation of Smad1/5 and inhibited cell proliferation in MCF10AT1 cells transfected with Ha-ras, but not in the parent MCF10A cells that lack Ras (Fig. 4), suggesting that Ras is critical for the activation of Smad signaling and growth inhibition by Ro3582.

The kinase pathways, such as Ras/ERK/MAPK, MEKK, JNK, p38 MAPK, CDK, and PKC, have been shown to regulate Smad signaling (43–46). MEKK1 or JNK enhanced Smad phosphorylation, nuclear localization, and Smad-mediated transcription (44, 45). In contrast, the phosphorylation of R-Smads at the linker domain or MH-1 domain induced by Ras/Erk/MAPK, CDK2/4, and PKC inhibited the activation of Smad signaling (23, 43, 46, 47). Yakymovych et al. (23) reported that PKC activation resulted in the phosphorylation of the MH1 domains of Smad2 and Smad3 to abrogate DNA binding of Smad3. Although this study suggests that PKC may play a negative regulatory role in TGF-β/Smad–mediated transcription (23), our previous studies indicated that Ro3582 induced the phosphorylation of Smad1/5 in the MH2 domain (Ser^463/465) and enhanced the BMP-specific signaling pathway (5, 6). In addition, our present study showed that this was blocked by PKC inhibitors (Figs. 2 and 3), suggesting a role of PKC for the activation of BMP/Smad signaling by Ro3582.

PKCs were originally thought to be promitogenic kinases, but this effect may be PKC isoform dependent and cell-type dependent, as many PKCs can also inhibit cell cycle progression (22). Members of the PKC family are mainly classified into three groups: classic (calcium and diacylglycerol dependent kinase; α, βI, βII, and γ), novel (calcium insensitive and diacylglycerol dependent kinase; δ, ε, μ, and 𝛉), and atypical (calcium and diacylglycerol insensitive kinase; η and λ/ι) PKC isoforms (21). Among many different PKC isoforms, PKCα is known to inhibit cell proliferation via p21 induction and suppress tumor formation in vivo (48, 49). Several studies have previously shown that 1α,25(OH)₂D₃ activates PKCα, and activation of this isoform acts as an antiproliferative signal (17, 20, 48, 49). Our results also indicate that PKCα mediates the activation of Smad signaling, which is important for growth inhibition by the vitamin D₃ analogue Ro3582. As shown in Fig. 6, we have depicted a schematic diagram for our proposed mechanism of the vitamin D₃ analogue Ro3582 in MCF10AT1 breast epithelial cells. Many growth factors, such as EGF or vascular endothelial growth factor, can stimulate growth in epithelial cells via Ras-MEK-ERK signaling. Based on our data, we postulate that the vitamin D₃ analogue Ro3582 may require Ras to activate PKCα and to initiate Smad signaling, and PKCα signaling may be a key downstream target pathway to mediate growth inhibition by Ro3582 (Fig. 6).

It was suggested that PKCα activation by 1α,25(OH)₂D₃ was through the 1α,25(OH)₂D₃ membrane-associated rapid response steroid binding protein (16, 20). The classic VDR, which is known to translocate to the nucleus after ligand binding, has been proposed to be associated with caveolae in the plasma membrane and is responsible for the rapid response to vitamin D ligands (15, 50). The structural flexibility of 1α,25(OH)₂D₃ and vitamin D₃ analogues may determine their preferences for binding to different locations on the VDR and for selective responses between genomic and membrane-mediated effects (15, 50). In this study, we showed that Ro3582 activated PKCα, which led to the enhancement of Smad signaling in MCF10AT1 cells. However, we do not understand yet, and it needs to be determined, whether either membrane-associated rapid response steroid binding protein or VDR associated with caveolae mediates the activation of BMP/Smad signaling by Ro3582 in MCF10AT1 breast epithelial cells.

To the best of our knowledge, the present study shows for the first time that a Gemini vitamin D₃ analogue activates the BMP/Smad signaling through a Ras/PKCα pathway, which leads to the inhibition of cell proliferation in MCF10 human breast epithelial cells. Further investigations are needed to understand the interactions between the Ras/PKCα pathway and the regulation of BMP/Smad signaling by vitamin D₃ analogues and their interactions with the nuclear or membrane-bound VDR.

Grant support: NIH grants K22 CA 99990 and R03 CA112642, National Institute of Environmental Health Sciences grant P30 ES005022, and a Cancer Institute of New Jersey new investigator award (N. Suh).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Drs. Bernard Weinstein and Jae-Won Soh for providing the vectors of PKC isoforms, Dr. Fred Miller for the MCF10 cell lines, Dr. Allan Conney for helpful advice on our work, and the Department of Chemical Biology for technical help with this project.