The active metabolite of vitamin D 1α,25-dihydroxycholecalciferol (1,25D3) has exhibited broad-spectrum antitumor activity in xenograft animal models. However, its activity against metastatic disease has not been extensively investigated.
Squamous cell carcinoma (SCC) or 1,25D3-resistant variant SCC-DR cells were treated with 1,25D3. Actin organization was examined by immunofluorescence assay. Cell migration was assessed by “wound” healing and chemotactic migration assays. Cell invasion was assessed by a Matrigel-based invasion assay and in situ zymography. Matrix metalloproteinase 2 (MMP-2) and MMP-9 expression and secretion were examined by immunoblot analysis and an enzyme-linked immunosorbent assay, respectively. E-cadherin expression was assessed by flow cytometry, immunoblot analysis, and immunohistochemistry. Knockdown of E-cadherin was achieved by small interfering RNA. An experimental metastasis mouse model was created by intravenous injection of tumor cells; and lung tumor development in the mice was assessed by magnetic resonance imaging, gross observation, and histology.
SCC cellular morphology and actin organization were altered by 10 nM 1,25D3. 1,25D3 inhibited SCC cell motility and invasion, which were associated with reduced expression and secretion of MMP-2 and MMP-9, and 1,25D3 promoted the expression of E-cadherin. These findings were not observed in SCC-DR cells. Knock down of E-cadherin rescued 1,25D3-inhibited cell migration. Intravenous injection of SCC or SCC-DR cells resulted in the establishment of extensive pulmonary lesions in saline-treated C3H mice. Treatment with 1,25D3 resulted in a marked reduction in the formation of lung tumor colonies in mice that were injected with SCC cells, but not in mice that were injected with SCC-DR cells.
Vitamin D is a secosteroid hormone that regulates calcium homeostasis and bone mineralization as well as several other physiologic activities.1 1α,25-Dihydroxyvitamin D (1,25D3), the active metabolite of vitamin D, has demonstrated broad spectrum antitumor activities in numerous preclinical studies.2-7 1,25D3 exerts growth-inhibitory effects by a variety of mechanisms, including the induction of apoptosis, cell cycle arrest, and differentiation in cancer cells. We previously demonstrated that 1,25D3 inhibits the growth of murine squamous cell carcinoma (SCC) cell line SCCVII/SF in vitro and in vivo.5, 8, 9 Apoptosis contributes to this growth inhibition, because 1,25D3 induces caspase 3 activation and poly(adenosine diphosphate-ribose) polymerase (PARP) cleavage in SCC cells.9 In addition, we have demonstrated that 1,25D3 enhances paclitaxel, gemcitabine, cisplatin, and carboplatin-mediated antitumor activities.10-12 However, the activity of 1,25D3 against SCC metastasis has not been extensively studied.
Metastasis is a complex process that involves a series of steps, including the initial detachment of cells from the primary tumor; local invasion of the basement membrane; intravasation and survival in the circulation; extravasation, invasion, and survival; and proliferation in a distant site.13 Failure of any single step will lead to the suppression of systemic metastasis formation. 1,25D3 or its analogs may affect these crucial steps. Sung and Feldman demonstrated that 1,25D3 inhibits prostate cancer cell migration and adhesion by down-regulation of integrins.14 Bao et al reported that 1,25D3 inhibits prostate cancer cell invasion by suppressing the expression of matrix metalloproteinase 9 (MMP-9) and cathepsins.15 However, the mechanisms for the regulation of cell motility and invasion remain unclear. Conversely, vitamin D deficiency promotes cancer cell growth in several metastasis models. In diet-induced, vitamin D-deficient mice, the growth of human breast cancer cells, injected into the tibia of mice, was enhanced; and larger osteolytic lesions were observed compared with the lesions observed in vitamin D-sufficient mice.16, 17 Similarly, prostate cancer cell growth in bone was enhanced in vitamin D-deficient mice.18 These observations suggest that 1,25D3 may play a role in tumor metastasis by multiple regulatory pathways and prompted our current investigation.
In the current study, we investigated the role of 1,25D3 in the modulation of the morphology and behavior of SCC cells and cells from 1,25D3-resistant SCC variant (SCC-DR cells). The effect of 1,25D3 on cell motility and invasion was evaluated. We studied the role of E-cadherin–mediated cell-cell adhesion in the regulation of SCC cell motility by 1,25D3. Furthermore, the in vivo activity of 1,25D3 in suppressing lung colony formation after intravenous injection was evaluated.
MATERIALS AND METHODS
1,25D3 (Hoffmann-LaRoche, Nutley, NJ) was reconstituted in 100% ethanol (ETOH) and stored, protected from light, under nitrogen gas at −70°C. Anti-MMP-2 and anti-MMP-9 were purchased from Biomol (Farmingdale, NY). Anti-E-cadherin was purchased from Cell Signaling Technology (Beverly, Mass). Anti-actin was purchased from Calbiochem (San Diego, Calif).
Cell Culture and Tumor Model System
A murine SCC cell (SCCVII-SF) tumor model and SCC cells were used as described previously.19 1,25D3-resistant SCC-DR cells were generated by continuously culturing SCC cells in media containing 10 nM 1,25D3, as described previously.20 SCC cells were maintained in female C3H/HeJ mice ages 6 to 10 weeks (purchased from Jackson Laboratory; Bar Harbor, Me). The mice protocols used were approved by the Roswell Park Cancer Institutional Animal Care and Use Committee.
Indirect Immunofluorescence Assay
SCC or SCC-DR cells were plated on glass coverslips and treated with ETOH or 10 nM 1,25D3 for 48 hours, washed with phosphate-buffered saline (PBS), fixed with 60% acetone/3.7% paraformaldehyde in PBS, and blocked with 25% normal goat serum at room temperature. Actin filaments were stained with rhodamine-labeled phalloidin (1:500 dilution; Sigma Chemical Company, St. Louis, Mo), and nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (1:1000 dilution; Invitrogen, Carlsbad, Calif) for 1 hour. Fluorescent images were captured using a Nikon TE2000-E inverted microscope (Nikon Corporation, Tokyo, Japan) equipped with Roper CoolSnap HQ charged-coupled device camera (Photometrics, Tucson, Ariz).
”Wound” Healing Assay
A confluent monolayer of SCC cells was cultured overnight, and a scratch was introduced with a pipette tip. Then, images of cell migration into the wound were captured at 0 hours, 24 hours, and 48 hours using a light microscope.
Chemotaxis Migration Assay
Chemotactic migration activity was measured by Boyden-chamber assay using BioCoat Control Inserts (BD Biosciences, East Rutherford, NJ). SCC or SCC-DR cells were plated in insert chambers in serum-free RPMI 1640. The lower chambers were filled with RPMI 1640 with 5% fetal bovine serum. After 16 hours of incubation, cells that did not migrate were removed from the upper chambers with a cotton swab, and cells that had migrated through the pore membrane were identified by Diff-Quik Stain Set (Dade Behring, Newark, Del), examined, and counted under brightfield microscopy.
Invasion activity was measured by using a Boyden-chamber assay with BD BioCoat Matrigel Invasion Chambers, as in the chemotaxis migration assay, except that a longer incubation time of 48 hours was used. The results are expressed as follows: percentage invasion index = (the number of cells migrating through the collagen-coated membrane/the number of cells migrating through the uncoated control membrane).
In Situ Zymography
Glass coverslips were coated with 0.2 mg/mL Oregon green 488-conjugated gelatin (Invitrogen), cross-linked in 0.5% glutaraldehyde for 15 minutes at 4°C, and incubated with 5 mg/mL NaBH4 for 3 minutes. The coverslips were then sterilized with 70% ETOH for 15 minutes and incubated in serum-free media for 1 hour at 37°C. SCC or SCC-DR cells were plated on coated coverslips, treated with ETOH or 10 nM 1,25D3, incubated at 37°C for 24 hours, and processed by fluorescence microscopy procedures.
SCC or SCC-DR cells that had been treated with either ETOH or 10 nM 1,25D3 for 48 hours were harvested with Trypsin-ethylene diamine tetra acetic acid, blocked with 3% bovine serum albumin/PBS for 1 hour, then incubated either with an immunoglobulin G isotype control or with rabbit anti-E-cadherin 5 μg/mL for 1 hour at room temperature, and washed twice with PBS. Samples were stained with phycoerythrin-conjugated goat-antirabbit secondary antibody for 1 hour. Flow cytometric analysis was performed on a Becton Dickinson FACScan flow cytometer (Becton Dickinson, Franklin Lakes, NJ), and data were analyzed with FCS Express (De Novo Software, Los Angeles, Calif).
Cell lysates were prepared and immunoblot analyses were performed as described previously.5
Enzyme-Linked Immunosorbent Assay
The levels of MMP-2 (AbCam, Cambridge, Mass) and MMP-9 (R&D Systems, Minneapolis, Minn) secreted into the culture media were assessed using enzyme-linked immunosorbent assay kits according to the manufacturers' instructions.
For immunohistochemistry studies, SCC cells (4.5 × 105) were inoculated subcutaneously into the flank of C3H mice. Mice were treated for 3 days with single, daily, intraperitoneal injections of either saline or 0.625 μg/mouse of 1,25D3 when tumors were palpable. On day 4, tumor tissues were harvested, fixed in 10% formalin, embedded in paraffin, and sectioned at 5 microns. Slides were deparafinized in xylene, rehydrated, and incubated with 10% normal goat serum followed by avidin/biotin block (Vector Laboratories, Burlingame, Calif) to block nonspecific binding. Anti-E-cadherin antibody (1:500 dilution; catalog no. 610181; BD Biosciences) was incubated for 30 minutes at room temperature, followed by biotinylated horse antimouse secondary antibody (Vector Laboratories) for 15 minutes. ABC reagent (Vector Laboratories) was then applied for 30 minutes. To reveal endogenous peroxidase activity, slides were incubated with 3,3′-diaminobenzidine substrate (Dako, Carpinteria, Calif) for 5 minutes and counterstained with modified Harris hematoxylin (Richard-Allan Scientific, Kalamazoo, Mich) for 20 seconds. The slides were dehydrated, and the coverslips were mounted with Permount (Fisher Scientific, Pittsburgh, Pa).
Small Interfering RNA Transfection
siGENOME small interfering RNAs (siRNAs) specific for E-cadherin, nonspecific siRNA (siRNA-NS), and DharmaFECT 1 transfection reagent were purchased from Dharmacon (Lafayette, Colo). SCC cells were transfected with 50 nM siRNA-NS or siRNA-E-cadherin for 24 hours using DharmaFECT 1 according to the manufacturer's protocol.
Magnetic Resonance Imaging of Lung Metastasis
Experimental MRI studies were performed using 4.7 T/33-cm horizontal bore magnet (GE NMR Instruments, Fremont, Calif) that incorporated a removable gradient coil insert (G060; Bruker Medical Inc., Billerica, Mass) and generated a maximum field strength of 950 mT/m with a custom-designed, 35-mm radiofrequency transmit-receive coil. Induction and maintenance of anesthesia during imaging were achieved by inhalation of 2% to 3% Isoflurane in oxygen (Abbott Laboratories, Abbot Park, Ill). Animal body temperature was maintained at 37°C during imaging using an air heater system (SA Instruments Inc., Stony Brook, NY). Animals were placed prone on a magnetic resonance-compatible sled (Dazai Research Instruments, Toronto, Ontario, Canada) within a carrier tube and positioned to ensure placement of the thoracic region of the mice in the isocenter of the magnet. Global shimming was performed at the beginning of imaging to ensure optimal field homogeneity. Preliminary scout images were acquired on the sagittal plane for localization and for determination of subsequent slice prescriptions. Coronal T2-weighted (T2W) images were acquired with the following parameters: field of view (FOV), 4.8 × 3.2 cm; matrix, 256 × 192; slice thickness, 1.00 mm; effective echo time (TEeff), 41 msec; repetition time (TR), 2424 msec; number of averages, 4. In total, 21 slices were acquired across the thoracic region to cover the entire lung area. Axial T2W images were acquired using the same sequence (FOV, 3.2 × 3.2; 25 slices). The duration of the imaging session for each animal, including induction of anesthesia, positioning, and set up, was approximately 20 minutes. After image acquisition, raw image sets were transferred to a processing workstation and processed using the medical imaging software Analyze (AnalyzeDirect, version 7.0; Overland Park, Kan). Tumor volumes (mm3) were calculated from the manually traced tumor area in each acquired slice and the slice thickness.
Ex Vivo Quantification of Experimental Lung Metastasis
For experimental metastasis, 105 SCC cells or 3 × 105 SCC-DR cells were injected into tail veins of C3H/HeJ mice (6 per group). Lung tumor development was monitored by MRI weekly. The mice with SCC tumors were killed at 2 weeks after transplantation, and those with SCC-DR tumors were killed 5 to 7 weeks after transplantation. The upper trachea was sealed using 4-0 silk, and the lungs were removed and stained with intrathecally administered 5% (volume/volume) India ink using a blunt-end, 21-gauge needle; then, the lungs were fixed in Fekete solution (100 mL 70% alcohol, 10 mL formalin, and 5 mL glacial acetic acid), and metastases were scored as surface lesions excluding ink.
Sample Sizes and Statistical Considerations
For in vitro studies, differences between control and treatment groups were analyzed for statistical significance using 2-tailed Student t tests. For in vivo imaging studies, a total of 29 animals underwent experimental MRI examinations for the quantification of pulmonary metastasis. In the first set of studies, 14 animals were injected with SCC cells and assigned to 1 of 3 groups: saline (n = 8), 0.3 μg 1,25D3 (n = 3), or 0.6 μg 1,25D3 (n = 3). The MRI examinations presented in the figures were performed approximately 11 to 15 days after intravenous injection of tumor cells to quantify pulmonary tumor burden. In a second set of studies, 15 animals were injected with SCC-DR cells and treated with saline (n = 4), or 0.3 μg 1,25D3 (n = 6), or 0.6 μg 1,25D3 (n = 5), and MRI examinations were performed approximately 40 to 50 days after injection. Differences in MRI-based tumor volume measurements were analyzed by 1-way analysis of variance using the Bonferroni multiple comparisons test.
To investigate the impact of 1,25D3 on the cellular processes involved in metastasis, we first characterized the phenotype of SCC cells by light microscopy. SCC cells normally present a polygonal appearance, as indicated in vehicle control ETOH-treated cells (Fig. 1A). However, 1,25D3 induced a different phenotype after 48 hours. SCC cells were larger and had a flat morphology (Fig. 1A). 1,25D3 did not change the morphology of 1,25D3-resistant SCC-DR cells, which were generated through continuous culture of SCC cells in 1,25D3-containing media20 (Fig. 1A). Cytoskeletal reorganization is essential for changing cell shapes, cell migration, and other processes.21 Actin staining revealed a major reorganization of cell cytoskeleton: Numerous actin stress fibers were observed throughout cytoplasm of 1,25D3-treated SCC cells, whereas control cells displayed disorganized actin cytoskeleton (Fig. 1B). In contrast, 1,25D3 did not change the morphology or actin staining pattern of SCC-DR cells (Fig. 1), indicating the critical role of 1,25D3 in cytoskeletal reorganization.
Actin polymerization is a major factor involved in cell migration.22 Therefore, next, we examined the ability of 1,25D3 to modify cell motility using the scratch “wound” healing and Boyden chamber-based chemotaxis assays. In the wound healing assay, ETOH-treated SCC cells covered at least 50% of the gap at 24 hours and covered almost the entire damaged area by 48 hours (Fig. 2A). In contrast, the scratch was still largely uncovered at 48 hours in 1,25D3-treated SCC cells (Fig. 2A). SCC-DR cells migrated more slowly than SCC cells, and 1,25D3 did not affect the migration rate (Fig. 2A). In the Boyden chamber assay, 1,25D3 markedly (P < .00001) inhibited the chemotactic migration activity of SCC cells. It is noteworthy that 1,25D3 inhibited the migration of SCC-DR cells (P < .01), but the inhibition was substantially less than that observed in SCC cells (Fig. 2B). These results indicate that 1,25D3 suppresses the motility of SCC cells.
To explore the effect of 1,25D3 on the invasiveness of SCC cells, the Matrigel-based invasion chamber assay and an in situ gelatin-degradation assay were performed. 1,25D3 significantly (P < .01) reduced the invasion of SCC cells through Matrigel membrane (Fig. 3A). Gelatinase activity (indicating the activity of MMP-2 and MMP-9) was assessed by in situ zymography. The ability of SCC cells to degrade matrix in situ was markedly suppressed by 1,25D3 treatment, as indicated by the reduced Oregon green-labeled gelatin-degradation areas that appeared as black holes (Fig. 3B). In contrast, neither the invasiveness nor the matrix-degradation potential of SCC-DR cells was affected by 1,25D3 (Fig. 3A,B). The expression levels of gelatinases MMP-2 and MMP-9 were assessed by immunoblot analysis, and markedly reduced expression of both was observed in SCC cells that were treated with 1,25D3 (Fig. 3C). Supporting these findings, the levels of MMP-2 and MMP-9 secreted into cell culture media were reduced by the treatment of 10 nM or 100 nM 1,25D3 in SCC cells (Fig. 3D).
1α,25-Dihydroxycholecalciferol Induces E-Cadherin Expression in Squamous Cell Carcinoma Cells
E-cadherin is involved in cell polarity, structure, and invasion. Therefore, next, we investigated whether 1,25D3-induced morphology and motility changes in SCC cells were associated with changes in E-cadherin expression. 1,25D3 markedly induced the expression of membrane E-cadherin in SCC cells as assessed by flow cytometry (Fig. 4A). The expression of E-cadherin was not modified in SCC-DR cells (Fig. 4A). In line with this finding, the results from immunoblot analysis indicated enhanced E-cadherin expression upon 1,25D3 treatment in SCC cells (Fig. 4B). To examine whether E-cadherin was modulated in vivo, SCC or SCC-DR tumor-bearing mice were treated with 1,25D3 for 3 days. Tumors were collected, and E-cadherin levels were assessed by immunohistochemistry. The treatment with 1,25D3 markedly promoted the expression of E-cadherin in SCC tumors but not in SCC-DR tumors (Fig. 4C).
E-Cadherin Contributes to 1α,25-Dihydroxycholecalciferol-Reduced Squamous Cell Carcinoma Cell Motility
To study whether E-cadherin plays a role in 1,25D3-regulated cell migration, siRNA/E-cadherin was used to knockdown E-cadherin expression in SCC cells, as confirmed by flow cytometry (Fig. 5A) and immunoblot analysis (Fig. 5B). Knockdown of E-cadherin promoted SCC cell migration, indicating that E-cadherin contributes to SCC motility (Fig. 5C). 1,25D3 treatment suppressed the migration of control siRNA-NS–transfected SCC cells but to a lesser degree than it suppressed the migration of siRNA/E-cadherin–transfected SCC cells (Fig. 5C). This result indicated that siRNA-E-cadherin rescued SCC cell migration inhibited by 1,25D3, which could be observed more easily when migration was normalized to siRNA-NS control (Fig. 5D). Together, these findings suggest that the 1,25D3-suppressed SCC cell migration may be caused at least in part by increased expression of E-cadherin.
1α,25-Dihydroxycholecalciferol Inhibits the Formation of Lung Metastases in Squamous Cell Carcinoma Tumors
Because 1,25D3 markedly inhibited SCC cell migration and invasion, we next examined the in vivo effect of 1,25D3 on metastasis using a lung tumor colony-formation mouse model. Noninvasive MRI was used to quantify tumor burden in the lungs after intravenous injection of SCC or SCC-DR cells. Coronal and axial T2W images provided adequate contrast for detecting the lesions in relation to the healthy lung parenchyma. In saline-treated control animals that were injected with SCC cells, multiple bilateral lesions were visible on multislice T2W images (Fig. 6A, top left). In contrast, treatment with 1,25D3 resulted in a marked reduction (P < .0001) in pulmonary tumor nodules (Fig. 6A). MRI studies of the animals that were injected with SCC-DR cells revealed a distinct pattern of tumor formation with large masses visible in the lungs (Fig. 6A, bottom left). Treatment with 1,25D3 did not appear to result in any inhibition of tumor growth in the SCC-DR model system. Ex vivo macroscopic analysis of lungs was performed after intrathecal instillation of India ink to permit the visualization of lung tumor colonies (Fig. 6B) and to validate imaging findings. Consistent with the MRI data, a marked reduction in tumor colonies was observed in 1,25D3-treated animals injected with SCC cells compared with saline-treated controls. Histologic examination of lung tissue sections obtained from saline-treated animals injected with SCC cells revealed extensive regions of normal lung parenchyma replaced by tumor tissue (Fig. 6C). By comparison, lungs excised from animals that had been treated with 1,25D3 revealed the presence of healthy lung tissue with fewer areas of malignant transformation. Conversely, 1,25D3 did not alter the histology of lung tissues obtained from animals that were injected with SCC-DR cells (Fig. 6C). These results indicate that 1,25D3 strongly inhibits the formation of lung tumor colonies after the intravenous introduction of SCC cells.
There is growing evidence that 1,25D3 may have antitumor activity in several cancer types, including colorectal, breast, prostate, ovarian, and skin cancers. In addition to suppressing cancer cell proliferation and apoptosis, 1,25D3 affects angiogenesis and tumor metastasis.5, 23-25 However, the mechanisms for these anticancer effects remain unclear, and further investigations are needed.
In the current study, we demonstrated that 1,25D3 modulates the morphology of SCC cells from polygonal to a larger and more flattened shape. Pendas-Franco et al demonstrated that MDA-MB-453 human breast cancer cells underwent different morphology changes upon 1,25D3 treatment, from a rounded shape to polygonal and flattened.26 The assembly of actin stress fibers was observed in both studies, indicating a role for 1,25D3 in regulating organization of the cytoskeleton, which may lead to changes in cell shape and motility.
Modulation of cell cytoskeleton organization may affect cell motility. Further studies using “wound” healing assay and chemotactic migration assay have demonstrated that 1,25D3 inhibits the migration and chemotaxis of SCC cells. The invasiveness of SCC cells also is markedly suppressed by 1,25D3, as demonstrated in the Matrigel Boyden chamber assay and by in situ zymography, which is a unique and important technique for revealing proteolytic activity at specific sites in tissues or cell cultures. This is associated with reduced expression and secretion of MMP-2 and MMP-9 in SCC cells, as indicated in immunoblot analysis and enzyme-linked immunosorbent assay analysis, respectively.
The mechanisms for 1,25D3-mediated inhibition of tumor cell motility are not well understood. 1,25D3 reduced the invasion of the Lewis lung carcinoma cancer cell line LLC-LN7 through matrix-coated membranes, and this effect was associated with decreased protein kinase A activity.27 Young et al studied LLC-LN7 cells and demonstrated that 1,25D3-suppressed tumor cell motility was associated with the inhibition of granulocyte-macrophage–colony-stimulating factor production.28 1,25D3 also inhibited invasion of prostate cancer cell lines LNCaP, PC-3 and DU145, as assessed by Matrigel invasion assay, and that inhibition was associated with decreased MMP-9 and cathepsin activity and increased activity of tissue inhibitors of metalloproteinase 1.15 1,25D3 may play an opposite role in motility in normal cells. Rebsamen et al noted the enhanced cell motility of benign vascular smooth muscle cells in association with phosphatidylinositol 3-kinase activation.29 The possible differential regulation of cell motility may be beneficial in cancer treatment, because it may spare normal cells from adverse effects.
E-cadherin mediates cell-cell adhesion, and the disruption of strong cell-cell adhesion leads to the invasion and metastasis of tumor cells.30 Loss of E-cadherin not only causes deregulation of cytoskeleton and loss of cell polarity but also leads to increased cell motility and invasiveness,31 possibly because E-cadherin–mediated cell-cell adhesion promotes cell clusters and, thus, restricts cell motility toward the extracellular matrix. Cell-cell adhesion mediated by E-cadherin often is lost with the development of invasiveness in epithelial cancer.32 E-cadherin loss may be because of several mechanisms. MicroRNA 9 (miR-9) inhibits E-cadherin expression and enhances mammary epithelial cell motility and invasiveness, and the silencing of miR-9 in breast cancer cells suppressed lung metastasis in an orthotopic implantation model.33 E-cadherin may be silenced by DNA hypermethylation around the promoter region in premalignant situations.34 MMPs can proteolytically cleave and disrupt the function of E-cadherin.31 Therefore, restoring the function of E-cadherin may inhibit tumor metastasis.
In the current study, we have identified a novel link between 1,25D3-induced E-cadherin expression and decreased cell migration. 1,25D3 promotes the expression of E-cadherin in SCC cells and tumors, as demonstrated by flow cytometry, immunoblot analysis, and immunohistochemistry. E-cadherin induction depends on the vitamin D receptor gene (VDR), because 1,25D3 did not affect E-cadherin levels in SCC-DR cells, which have defective VDR signaling.20 In line with our data, another study revealed that 1,25D3 induced the expression of E-cadherin in colon carcinoma cells.35 We have demonstrated for the first time that the silencing of E-cadherin by siRNA rescues 1,25D3-inhibited SCC cell migration. These data indicate that 1,25D3-enhanced E-cadherin expression contributes to 1,25D3-mediated inhibition of SCC cell motility.
With the observation that 1,25D3 strongly inhibits SCC cell motility and invasion in vitro, next, we investigated the effect of 1,25D3 in vivo in an experimental mouse model, in which SCC cells were injected intravenously into mice and tumor growth in lungs was monitored. By using in vivo MRI studies and ex vivo macroscopic and microscopic analyses, we demonstrated that 1,25D3 markedly suppresses the ability of SCC cells to establish pulmonary metastases. It is noteworthy that, compared with SCC cells, it takes much longer for SCC-DR cells to form tumor colonies in lungs after being introduced intravenously in mice, and they form fewer colonies than SCC cells. This may be because SCC-DR cells express higher levels of endogenous E-cadherin compared with SCC cells, as indicated by our flow cytometry and immunoblot analyses, and thus have decreased ability to migrate and invade. However, our experimental metastasis model has intrinsic limitations, because cancer cells are introduced directly into the circulation; therefore, the model cannot evaluate the initial metastasis processes when cancer cells need to escape from the primary tumor site, invade through the basement membrane, migrate and invade through the surrounding tissue, and then enter the circulation. Spontaneous metastasis models should be used if feasible.
In support of our current findings, others have demonstrated that 1,25D3 inhibits metastatic growth in several tumor models. 1,25D3 suppressed spontaneous and experimental pulmonary metastasis of B16 mouse melanoma.25 22-oxa-1,25D3 reduced lung colony formation in an intravenous injection model and reduced angiogenesis in a basic fibroblast growth factor-induced angiogenesis model.36 The vitamin D analog EB1089 suppressed the occurrence of bone metastasis in a breast cancer metastasis model.37 1,25D3 or 22-oxa-1,25D3 treatment suppressed lung metastasis of Lewis lung carcinoma cells in an intravenous injection mouse model.36 Although preclinical models support an antimetastatic role for 1,25D3, further studies are needed to assess the clinical relevance.
In conclusion, we demonstrate that 1,25D3 modulates SCC cell morphology and actin arrangement. In addition, 1,25D3 inhibits SCC cell motility, possibly through the promotion of E-cadherin expression. 1,25D3 inhibits the invasiveness of SCC cells, which is associated with decreased expression and secretion of MMP-2 and MMP-9. In vivo, 1,25D3 suppressed SCC lung metastasis in an experimental model approximating metastasis. Although further mechanistic investigation into the regulation of metastasis by 1,25D3 is necessary, the observed antimetastatic activity of 1,25D3 in multiple preclinical model systems supports its evaluation in the clinical setting.
We thank Mrs. Rui-Xian Kong for her excellent technical support, Mr. Steve G. Turowski for assistance with magnetic resonance imaging studies, Ms. Ellen Karasik for her excellent technical assistance in immunohistochemistry study, and Dr. Pamela A. Hershberger for her critical review of the article.
This study was supported by National Institutes of Health/National Cancer Institute grants CA067267, CA085142 (C. S. Johnson), and CA095045 (D. L. Trump) and used core resources supported by National Cancer Institute grant P30CA16056 (D. L. Trump).