Progesterone potentiates the growth inhibitory effects of calcitriol in endometrial cancer via suppression of CYP24A1

Here, we evaluated the expression of CYP24A1, a protein that inactivates vitamin D in tissues. CYP24A1 expression was increased in advanced-stage endometrial tumors compared to normal tissues. Similarly, endometrial cancer cells expressed higher levels of CYP24A1 than immortalized endometrial epithelial cells. RT-PCR and Western blotting were used to examine CYP24A1 mRNA and protein levels in endometrial cancer cells after 8, 24, 72, and 120 h of exposure to progesterone, progestin derivatives and calcitriol, either alone or in combination. Progestins inhibited calcitriol-induced expression of CYP24A1 and splice variant CYP24SV mRNA and protein in cancer cells. Furthermore, actinomycin D, but not cycloheximide, blocked calcitriol-induced CYP24A1 splicing. siRNA-induced knockdown of CYP24A1 expression sensitized endometrial cancer cells to calcitriol-induced growth inhibition. These data suggest that CYP24A1 overexpression reduces the antitumor effects of calcitriol in cancer cells and that progestins may be beneficial for maintaining calcitriol's anti-endometrial cancer activity.


INTRODUCTION
Endometrial cancer is the fifth leading cancer in women, and more than 10,170 women will die of this disease this year in the U.S. [1]. Surgery and subsequent chemo-or radiation therapy are the standard treatments for endometrial tumors. Uterine tumors diagnosed at early stages (70-80%) typically have lower grades and are hormone sensitive, while tumors diagnosed at later stages (15-20%) are usually high-grade, hormone-independent, and associated with higher recurrence rates [2,3]. The incidence of endometrial cancer is increasing, and a better understanding of the mechanisms underlying endometrial cancer is pivotal for improving therapeutic strategies.
Progesterone and synthetic progestins decrease the risk of developing endometrial cancer [4]. However, high doses of progestins are accompanied by undesirable side effects, such as edema and weight gain. Combining another preventative agent with lower doses of progestins might help to maximize the latter's antitumorogenic effects while also reducing side effects. Numerous studies have demonstrated a role for vitamin D 3 in cancer prevention, but its dose-limiting calcemic effects make it unacceptable as a chemopreventive agent [5,6]. Combining a low dose of progesterone with a low dose of calcitriol, a vitamin D metabolite, might be more effective for treating endometrial cancer than higher doses of either agent alone. In vitro and in vivo studies from our laboratory and others Research Paper www.impactjournals.com/oncotarget have shown that progesterone and other chemopreventive agents enhance the antitumor effects of calcitriol [7][8][9][10].
We previously reported that progesterone-mediated upregulation of vitamin D receptor (VDR) levels increases calcitriol-induced growth inhibition in endometrial cancer cells [9,10]. Here, we expand upon our previous work by examining the effects of calcitriol and progesterone, both alone and in combination, on CYP24A1. We provide evidence that progesterone enhances the anti-tumorigenic effects of calcitriol by inhibiting CYP24A1 in endometrial cancer cells.

CYP24A1 expression increased as tumors progressed
CYP24A1 expression was analyzed in vivo by immunohistochemistry in tissue microarrays (TMAs) (US Biomax Inc.). TMAs consisted of 24 normal and 72 malignant tissues, 22 of which were from grade I, 26 from grade II, and 16 from grade III cancers. TMA staining was correlated with patient clinicopathological parameters ( Figure 1). In normal endometrial tissues, CYP24A1 expression was low or undetectable in epithelial cells, glands, and stromal cells. CYP24A1 expression increased markedly as tumor grades increased ( Figure 1 and Table 1). These data suggest that increased CYP24A1 expression is associated with endometrial carcinogenesis.

CYP24A1 expression in normal and malignant cell lines
CYP24A1 expression was examined in cellular extracts from a panel of normal and endometrial cancer cell lines by Western blot analysis. Using a CYP24A1 antibody which detects a single band at 65 kDa, we found higher levels of CYP24A1 in HEC-1B, Ishikawa, and RL-95 cells compared to EM-E6/E7-TERT immortalized endometrial epithelial cells ( Figure 2). This antibody does not recognize the CYP24A1 splice variant. These results are consistent with the CYP24A1 expression data.

Effects of progesterone, MPA, NORG, and NOR on CYP24A1 mRNA expression in endometrial cancer cells
Because progesterone and progesterone derivatives are used to treat endometrial cancer, we examined the effects of progesterone (PROG), medroxyprogesterone acetate (MPA), norestrel (NORG), norethindrone (NOR), and calcitriol, either alone or in combination, on CYP24A1 transcript expression in HEC-1B, Ishikawa, and RL-95 cells. Calcitriol increased CYP24A1 expression in all cancer cell lines ( Figure 3). Progesterone and MPA inhibited the calcitriol-induced increase in CYP24A1 expression at all time points in HEC-1B, Ishikawa, and RL-95 cells ( Figure 3). In HEC-1B and RL-95 cells, NORG did not affect calcitriol-induced CYP24A1 expression in the first 24 hours, but reduced its expression at all subsequent time points. NORG inhibited the calcitriol-induced increase in CYP24A1 expression in Ishikawa cells at all time points except 8 hours ( Figure 3). NOR inhibited the calcitriol-induced increase in CYP24A1 expression at 72 and 120 h in HEC-1B and RL-95 cells and at 24, 72, and 120 h in Ishikawa cells ( Figure 3). immunohistochemistry. CYP24A1 levels were higher in Grade III tumors than in normal endometrial tissues. Negative control for CYP24A1 is shown in Grade III tumor tissue. Original magnification, 400x.

Effects of progesterone, MPA, NORG, and NOR on CYP24A1SV mRNA expression in endometrial cancer cells
We then used specific primers to determine the effects of calcitriol and progesterone derivatives on CYP24A1SV expression in endometrial cancer cells [22]. As shown in Figure 3, progesterone and MPA markedly inhibited calcitriol-induced increases in CYP24A1SV expression at all time points in the three cell lines tested. NORG and NOR suppressed CYP24A1SV expression only at 72 and 120 h in all cell lines ( Figure 3).

Effects of progesterone, MPA, NORG, and NOR on CYP24A1 protein levels in endometrial cancer cells
High constitutive CYP24A1 levels have been reported in cancer cells; in normal cells, CYP24A1 activity is usually low, but can be induced by vitamin D [14,17]. CYP24A1 protein levels were examined by Western blotting after 8, 24, 72, and 120 h of exposure to progesterone, progestin derivatives, and calcitriol, either alone or in combination. None of the hormones alone or in combination affected CYP24A1 levels in HEC-1B, Ishikawa, or RL-95 endometrial cancer cells after 8 h ( Figure 4). Progesterone, MPA, and NORG inhibited calcitriol-induced increases in CYP24A1 levels in HEC-1B, Ishikawa, and RL-95 cells after 24, 72, and 120 h of treatment. Progesterone and MPA attenuated basal CYP24A1 levels after 72 and 120 h of treatment in all cell lines ( Figure 4). NORG markedly decreased calcitriol-induced increases in CYP24A1 levels in HEC-1B, Ishikawa, and RL-95 cells after 24, 72, and 120 h ( Figure 4). NOR inhibited CYP24A1 levels after 72 and 120 h in HEC-1B cells and after 24, 72, and 120 h in Ishikawa and RL-95 cells ( Figure 4).

Effects of an RNA synthesis inhibitor on calcitriol-induced CYP24A1 splicing
To ascertain whether RNA synthesis is required for calcitriol-induced CYP24A1 splicing, HEC-1B and Ishikawa cells were pretreated with actinomycin-D for 1 h to inhibit RNA synthesis and then treated with calcitriol, progesterone, or both for 30 min, 2 h, 8 h, or 24 h to induce CYP24A1 splicing. As shown in Figure 5, calcitriol did not induce CYP24A1 splicing after 30 min of culture. However, calcitriol induced splicing after 2, 8, and 24 h in both cell lines. Progesterone reduced calcitriolinduced CYP24A1 splicing in cancer cells. Pretreatment with actinomycin-D followed by treatment with calcitriol for 2, 8, or 24 h blocked CYP24A1 splicing. These results demonstrate that de novo RNA synthesis may be required for calcitriol-induced CYP24A1 splicing.

Effects of a protein synthesis inhibitor on calcitriol-induced CYP24A1 splicing
Treatment with calcitriol alone increased CYP24A1 mRNA expression in endometrial cancer cells. In contrast, treatment with progesterone and calcitriol together suppressed the calcitriol-induced increase in CYP24A1 expression. The induction of CYP24A1 might be a result of both direct and indirect responses to calcitriol. To investigate this possibility, we applied the same treatments described above in the presence of the protein synthesis inhibitor cycloheximide. Pre-treatment with cycloheximide reduced CYP24A1 splice variant expression in HEC-1B and Ishikawa cells treated with calcitriol compared to cells treated with calcitriol alone after 2, 8, and 24 h of culture ( Figure 5). These results indicate that de novo protein synthesis is not required for calcitriol-induced CYP24A1 splicing and that ongoing protein synthesis may be involved in calcitriol-induced CYP24A1 splicing.

siRNA-mediated suppression of CYP24A1 expression augmented the antiproliferative effects of calcitriol in endometrial cancer cells
Increased CYP24A1 expression has been reported in many cancers and reduces the anti-tumorigenic activity of calcitriol. To assess whether suppressing CYP24A1 expression increases the anti-cancer effects of calcitriol, we transfected HEC-1B and Ishikawa cells with siRNA that targets CYP24A1. In both cell lines, CYP24A1 protein levels decreased markedly in CYP24A1 siRNAtransfected cells compared to control siRNA-transfected cells ( Figure 6A). After confirming CYP24A1 knockdown in endometrial cancer cells, we examined proliferative potential in control siRNA-and CYP24A1 siRNAtransfected cells in the presence or absence of progesterone and calcitriol, either alone or in combination. Cell proliferation decreased by 15-20% (p<0.05) in CYP24A1 knockdown cells compared to control siRNA-transfected cells. CYP24A1 knockdown cells were more responsive to progesterone exposure, as demonstrated by a 70-75% decrease in proliferation compared to the 55-60% decrease observed in control siRNA-transfected cells ( Figure  6B). However, calcitriol treatment, which decreased cell viability by 5-10% in control siRNA-transfected cells, decreased viability to a greater extent (35-45%) in CYP24A1 knockdown cells. Combined treatment with progesterone and calcitriol decreased cell viability more than either agent alone. However, the decrease in cell viability induced by combined treatment was comparable to the decrease observed in both CYP24A1 knockdown and control siRNA-transfected cells. These findings indicate that decreased CYP24A1 expression potentiates calcitriol-induced growth inhibition in these cells. Co-administration of progesterone and its antagonist (Mifepristone) attenuated progesterone-induced growth inhibition, suggesting that the progesterone receptor mediates the anti-proliferative effects of progesterone in all of the cell lines examined ( Figure 6B).

CYP24A1 knockdown enhanced apoptosis in endometrial cancer cells by altering the expression of calcitriol-regulated genes
To investigate the mechanism by which CYP24A1 knockdown enhanced the anti-proliferative effects of calcitriol in endometrial cancer cells, we analyzed the effect of CYP24A1 knockdown on apoptosis and levels of the protein products of vitamin D-regulated genes. VDR levels were markedly decreased in untreated CYP24A1 knockdown cells compared to control siRNA-transfected cells. However, progesterone alone or in combination with calcitriol upregulated VDR levels in CYP24A1 knockdown and control siRNA-transfected cells. Progesterone and calcitriol, both alone and in combination, markedly reduced cyclin D1 and increased p21 levels in CYP24A1 knockdown cells. A similar pattern was observed in control siRNA-transfected cells, although the effect of calcitriol was smaller. Knockdown of CYP24A1 slightly, but significantly, increased caspase-3 levels compared to CYP24A1-expressing cells. Caspase-3 levels increased in both CYP24A1 knockdown and control cells following exposure to progesterone and calcitriol ( Figure 7). To examine apoptosis in progesterone-and calcitriol-treated cells, we used a cell death detection ELISA, which measures cellular cytoplasmic-histoneassociated DNA fragment (mono-and oligonucleosomes) levels. Nucleosome production increased in CYP24A1 siRNA-transfected cells compared to control siRNAtransfected cells (p<0.05). Calcitriol increased nucleosome production only in CYP24A1 knockdown cells; no effect was observed in control siRNA-transfected cells. Progesterone alone or together with calcitriol increased nucleosome production in both control and CYP24A1 knockdown cells; this increase was larger in CYP24A1 knockdown cells ( Figure 7B).

CYP24A1 knockdown enhanced calcitriolinduced anti-tumorigenic effects by decreasing MMP expression
The effects of CYP24A1 silencing on HEC-1B and Ishikawa cell invasiveness were determined using a Boyden chamber invasion assay. Progesterone treatment decreased cancer cell invasiveness compared to vehicle control-treated CYP24A1-expressing and CYP24A1 knockdown cells. Calcitriol decreased the invasive potential of CYP24A1-expressing cells by 15-20%.
However, in CYP24A1 knockdown cells, calcitriol decreased cell invasiveness by 25-30%. Combined treatment was more effective than single treatment in inhibiting cell invasive potential ( Figure 8A). Increased MMP expression is associated with increases in cell invasiveness and metastatic potential. Expression of MMP9 and MMP2 was assessed in CYP24A1-expressing and CYP24A1 knockdown endometrial cancer cells after progesterone and calcitriol treatment. Expression of both MMPs was higher in CYP24A1-expressing cells than in CYP24A1 knockdown cells. Calcitriol markedly inhibited MMP9 and MMP2 expression in CYP24A1 knockdown cells compared to CYP24A1-expressing cells ( Figure 8B).

DISCUSSION
CYP24A1 is an oncogene that inhibits the anticancer effects of calcitriol [17,18,[23][24][25], which induces the expression of CYP24A1 and its splice variant [26,27]. Overexpression of CYP24A1 has been reported in several types of advanced-stage tumors (colon, cervical, ovarian, breast, esophageal, lung, prostate) and is correlated with poor prognosis in cancer patients [18,[28][29][30][31][32][33]. The present study is the first to compare both CYP24A1 mRNA expression and protein levels in normal endometrium and in endometrial cancer cell lines, and to examine correlations between CYP24A1 expression and tumor cell proliferation. Consistent with previous reports, we found that CYP24A1 was increased in advanced-stage endometrial tumors compared to normal endometrial tissues. Our results suggest that CYP24A1 may serve as a tissue biomarker of endometrial tumorigenesis. Assessment of CYP24A1 expression in endometrial cancer patient biopsies would likely aid in developing treatment strategies by helping to indicate whether patients would respond to calcitriol treatment.
We confirmed CYP24A1 expression patterns in vitro in an immortalized endometrial epithelial cell line (EM-E6/E7-TERT) and in endometrial cancer cell lines (HEC-1B, Ishikawa, and RL-95). These results confirmed the endometrial cancer tissue microarray results and are in agreement with other studies showing that CYP24A1 is overexpressed in cancer cell lines [8,[32][33][34][35]. Increased CYP24A1 expression also contributes to the reduced antitumorigenic effects of calcitriol in these cancer cells, and inhibition of CYP24A1 may block the degradation of calcitriol. Studies using CYP24A1 inhibitors demonstrate that high CYP24A1 levels markedly increase the antiproliferative effects of calcitriol in cancer cells [9,[36][37][38].
Progestins are effective in treating endometrial cancer patients [39,40]. Synthetic progestins are generally derivatives of 17-alpha-hydroxy-progesterone Figure 5: Effects of actinomycin D and cycloheximide on calcitriol-induced CYP24A1 splicing. HEC-1B and Ishikawa cells were pre-treated with actinomycin D (5 µg/mL) or cycloheximide (10 µg/mL) for 1 h to inhibit RNA or protein synthesis. Cells were then treated with progesterone (PROG, 20 µM), calcitriol (100 nM), or both for 30 min, 2, 8, or 24 h, followed by RNA extraction. CYP24 splicing was analyzed by RT-PCR. 18S served as the loading control. www.impactjournals.com/oncotarget or 19-nortestosterone and can be categorized as estrane, pregnane, or gonane derivatives. These three progestin groups vary with regard to progestogenic potency and androgenic, estrogenic, and glucocorticoid activity, and thus may differ in their biologic effects. The synthetic progestin MPA has a methyl group at carbon 10 and belongs to the pregnane progesterone derivative group. MPA exhibits high progestational activity and is widely used in the treatment of endometrial carcinoma, with a response rate of 60-75%. Norgestrel is classified as a gonane because it contains a 4-ring carbon nucleus (gonane) in addition to an ethyl group at carbon 13. The progestin norethindrone is derived from estrane and has an 18-carbon steroid nucleus. The availability of oral progestins, and their high response rates and low toxicity profiles contribute to their popularity for the treatment of advanced-stage or recurrent endometrial carcinomas [39][40][41][42]. Synthetic progestins may have varying effects on endometrial tumor growth. In this study, we therefore compared the effects of three commonly used progestins, medroxyprogesterone-17 acetate (MPA), norgestrel, and norethindrone, as well as natural progesterone, on CYP24A1 expression.  were transfected with CYP24A1 siRNA (25 nM) or control siRNA (25 nM) using Dharma FECT 1 (GE Dharmacon). After 48 hours, cells were treated with progesterone, calcitriol, or both for 3 days. A. Expression of VDR, cyclin D1, p21, and caspase-3 was analyzed by Western blot. Representative Western blots are shown on top of the histograms. Relative VDR, cyclin D1, p21, and caspase-3 protein levels after normalization to β-actin are presented in the histograms. Protein levels in untreated control siRNA-transfected cells were arbitrarily designated 100; changes in protein levels in siRNA control and CYP24A1 knockdown non-treated and treated cells are shown as percentages of control cell levels. *p<0.05 compared to respective control cells. ---@--p<0.05 compared to control siRNA-transfected and CYP24A1 knockdown non-treated cells. B. Nucleosome production in control and CYP24A1 knockdown cells exposed to progesterone, calcitriol, or both was assessed by cell death detection ELISA. The data are presented as means ± SEM of three independent experiments. *p<0.05 compared to respective control cells. #p<0.05 in CYP24A1 siRNA-transfected non-treated compared to control siRNA-transfected non-treated cells. ns denotes non-significant changes in control siRNA-transfected non-treated cells compared to control siRNA-transfected calcitriol-treated cells. www.impactjournals.com/oncotarget

Figure 8: Effects of CYP24A1 knockdown on invasive potential and MMP expression in endometrial cancer cells. A.
Invasive potential in HEC-1B and Ishikawa cells transfected with CYP24A1 siRNA (25 nM) or control siRNA (25 nM) and treated with progesterone, calcitriol, or both for 3 days was evaluated using BD BioCoat Matrigel Invasion Chambers. Data are shown as means ± SEM from three independent experiments. *p<0.05 in control siRNA-transfected and CYP24A1 knockdown treated cells. B. MMP-9 and MMP-2 levels were measured by Western blot analysis in scrambled siRNA-transfected and CYP24A1 knockdown cells after treatment with progesterone, calcitriol, or both. Densitometric scan values were used to quantify changes in MMP levels.
Our time-course studies demonstrated that calcitriol increases CYP24A1 mRNA expression, protein levels, and splicing in all three endometrial cancer cell lines examined. Progesterone and the three progestins markedly decreased calcitriol-induced CYP24A1 and CYP24A1 splice variant expression. Progesterone and MPA inhibited CYP24A1 expression to a greater degree than norgestrel and norethindrone. MPA and megestrol acetate (MA, Megace) are the most widely used progestins in hormonal therapy (HT), contraception, and treatment of advanced or recurrent endometrial cancer. Other progestins, including norgestrel and norethindrone, also strongly inhibit endometrial cancer growth and progression [39][40][41][42][43]. Our results strongly suggest that progestins in general enhance calcitriol's antitumor effects by downregulating CYP24A1 expression.
The inconsistent antitumor effects of vitamin D are attributed to post-transcriptional modifications of CYP24A1 mRNA in cancer cell lines [13,14,22,26]. In endometrial cancer cells, CYP24A1 RNA splicing is rapid and efficient, and can be regulated by calcitriol treatment. Our results showed that calcitriol-induced CYP24A1 and CYP24-SV mRNA expression was inhibited in cancer cells pretreated with actinomycin-D, suggesting that de novo RNA synthesis may be required for calcitriol-induced CYP24A1 splicing. These findings are in agreement with a study of colon cancer in which actinomycin-D treatment blocked CYP24A1 splicing, while cycloheximide did not affect calcitriol-induced CYP24A1 splicing [44]. CYP24A1 and CYP24A1 splice variant proteins appear to be important for the complex control of calcitriol responses, and CYP24-SV is implicated in various clinical conditions. CYP24-SV suppresses the function of the enzyme 1α-OHase (25-OHD-1α-hydroxylase), which catalyzes the production of active vitamin D, by sequestering the 25-OHD (25-hydroxyvitamin D3) substrate. CYP24-SV also tightly controls the effects of calcitriol by binding to and inhibiting the catabolism of 1,25(OH)2D, thus attenuating its nuclear signaling.
Here, we show that siRNA-or progesteroneinduced CYP24A1 knockdown in endometrial cancer cells enhances the anti-proliferative effects of calcitriol. High CYP24A1 expression reduces calcitriol-induced growth inhibition. Our results are in agreement with in vivo studies in which orthotopic prostate cancer xenograft growth was markedly reduced in mice exposed to calcitriol and CYP24A1 shRNA or to a combination of calcitriol and ketoconazole/dexamethasone [18,32]. These studies suggest that the administration of calcitriol in combination with CYP24A1 inhibitors increases calcitriol's antitumorigenic effects. Our results demonstrated that CYP24A1 protein levels and cell proliferation markedly decreased after progesterone and CYP24A1 siRNA administration. Because cell proliferation was similarly reduced in control siRNA-and CYP24A1 siRNAtransfected cells after combined progesterone and calcitriol administration, this effect may be due to progesteroneinduced inhibition of CYP24A1 splicing, which may not be blocked by CYP24A1-siRNA.
In summary, this is the first study to evaluate CYP24A1 expression in endometrial cancer, and to examine the correlation between its expression and tumor progression. We also found that progesterone increased the anti-tumorigenic effects of calcitriol by decreasing CYP24A1 expression. These observations indicate that co-administration of calcitriol and progesterone may be therapeutically beneficial for cancer patients.

Tissue microarrays and immunohistochemical detection of CYP24A1
EMC962 endometrial tissue arrays, consisting of normal (12 cases per 24 cores) and cancerous (36 cases per 72 cores) tissues in duplicate, were obtained from US Biomax. Immunohistochemistry was performed using formalin fixed and paraffin-embedded arrays. The array slides were deparaffinized and antigen retrieval was carried out using Retriever 2100 (Electron Microscopy Sciences) in 10 mM sodium citrate buffer (pH 6). Vectastain Elite ABC kits and DAB were used for staining (Vector Laboratories Inc.). The primary antibody was CYP24A1 (1:100, SC 66851, Santa Cruz Biotechnology, CA). Images were captured using a Nikon Eclipse E-800 microscope. CYP24A1 levels in tumor tissues were quantified across cores as a product of the staining intensity (0=negative, 1=weak, 2=moderate, and 3=strong) and percentage of cells stained. Staining intensity was quantified under high magnification (x20). CYP24A1 levels were correlated with a range of clinical parameters, such as tumor classification (tumor stage and grade).

Cell culture and treatment
The Ishikawa human endometrial cancer cell line was obtained from Sigma (St Louis, MO), and HEC-1B and RL-95 cell lines were obtained from the American Type Culture Collection (ATCC, Manassas, VA). The cell lines were authenticated by DNA short-tandem repeat analysis performed by ATCC and Sigma. All three cell lines were initially expanded and cryopreserved within 1 month of receipt. Cells were typically used for 3 months, after which a fresh vial of cryopreserved cells was used. The culture conditions used in this study have been described previously [9,10]. In brief, Ishikawa cells were grown in DMEM supplemented with insulin (0.005 mg/mL, Invitrogen Carlsbad, California), HEC-1B cells were cultured in Eagle minimum essential medium (Invitrogen), and RL-95 cells were grown in DMEM:F12 medium supplemented with insulin (0.005 mg/mL). All three media were supplemented with 10% (v/v) fetal bovine serum (FBS, Invitrogen), 100 U/ml penicillin, and 100 U/ml streptomycin. Cells were cultured at 37°C in a humidified atmosphere containing 5% CO 2 . Forty-eight hours later, media were replaced with the same media with the addition of charcoal-stripped FBS. Cells were treated with progesterone (20 µM PROG, 99.9% pure; Sigma), medroxyprogesterone-17 acetate (MPA, 20 µM; Sigma), D(-)-norgestrel (NORG, 20 µM; Sigma), 19-norethindrone (NOR, 20 µM; Sigma), calcitriol (100 nM; Sigma), or a combination of progestins and calcitriol for 8, 24, 72, or 120 hours, and collected for RNA or protein extraction. To determine whether CYP24A1 inhibition enhanced the anti-cancer effects of calcitriol, cells were transfected with CYP24A1 siRNA (GE Dharmacon, Lafayette, CO) or scrambled siRNA and treated with progesterone, calcitriol, or both in the absence or presence of the progesterone receptor antagonist Mifepristone (0.1 mM) for 72 hours; cell viability was then assessed. In another set of experiments, HEC-1B and Ishikawa cells were pretreated with actinomycin D (5 µg/mL) or cycloheximide (10 µg/mL) for 1 h to inhibit RNA or protein synthesis, respectively. These cells were then treated with progesterone (PROG, 20 µM), calcitriol (100 nM), or both for 30 min, 2, 8, or 24 hours, followed by RNA extraction.

Western blot analysis
Extracts from endometrial cancer cells treated with vehicle or with PROG, MPA, NORG, NOR, or calcitriol, either alone or in combination, were analyzed using antibodies against CYP24A1 (SC 66851) and β-actin (Sigma, St Louis, MO). MMP2, MMP9, VDR, cyclin D1, P27, caspase-3 (all Santa Cruz Biotechnology), and CYP24A1 levels were determined in extracts from scrambled siRNA-and CYP24A1 siRNA-transfected, hormone-treated endometrial cancer cells. Equal amounts of protein were subjected to sodium dodecyl sulfatepolyacrylamide gel electrophoresis, and intracellular β-actin levels were analyzed as the loading control. An enhanced chemiluminescence system was used to visualize protein bands according to the manufacturer's recommendations (Pierce, Piscataway, NJ). Experiments were repeated at least three times. Bands were quantified by densitometry using ImageJ software, and protein band intensities were normalized to β-actin.

RNA isolation and RT-PCR for CYP24A1 and CYP24A1-SV
Cells treated with PROG, MPA, NORG, NOR, or calcitriol alone or in combination were harvested after 8, 24, 72, or 120 hours of treatment. Total RNA was extracted using TRIzol® Reagent (Invitrogen). Reverse transcription of mRNA and PCR was performed using the protocol for the Superscript™ First-Strand Synthesis System for RT-PCR (Invitrogen). The following primers were used for CYP24A1,

Cell viability assay
The viability of CYP24A1-expressing and knockdown endometrial cancer cells treated with progesterone, calcitriol, or both in the presence or absence of the progesterone receptor antagonist was evaluated using CellTiter 96 AQueous One Solution as previously reported [9,10]. Briefly, CellTiter 96 AQueous One Solution reagent was added into each well of the 96-well assay plate containing the samples in 100 µL of culture medium. Absorbance was measured at 490 nm using an ELX800 microtiter Reader (Winooski, VT). Relative cell viability was expressed as % change in treated cells over that in vehicle treated cells.

Cell invasion assay
Scrambled siRNA-or CYP24A1 siRNA-transfected endometrial cancer cells treated with progesterone, calcitriol, or both for 72 hours were detached using trypsin and resuspended in serum-free medium. Medium containing 10% FBS was applied to the lower chambers of BD BioCoat Matrigel Invasion Chambers (BD Biosciences) as a chemoattractant, and cells were then seeded in the upper chambers at a density of 2.5x10 4 cells per well in 100 mL of serum-free medium without progesterone or calcitriol. The chambers were incubated for 16 to 18 hours at 37°C. At the end of incubation, noninvading cells were removed from the upper surface of the membrane by scrubbing. The cells on the lower surface of the membrane were fixed for 2 minutes in 100% methanol and stained with 1% toluidine blue in 1% sodium borate for 2 minutes. Excess stain was removed by rinsing the inserts with water. Each membrane was removed from the insert and placed on a microscope slide. Cover slips were placed on the slides and cells were counted in five random fields per slide. These assays were run in triplicate.

Enzyme-linked immunosorbent assay (ELISA) for histone-associated DNA fragments
Endometrial cancer cell death was evaluated after treatment with progesterone, calcitriol, or both using a cell death detection ELISA kit (Roche Molecular Biochemicals) according to the manufacturer's instructions. This assay quantifies apoptosis by measuring mono-and oligonucleosomes in apoptotic cell lysates. Cells were harvested in lysis buffer, and the cytoplasmic and nuclear fractions were separated by centrifugation at 200 x g. The supernatant (cytoplasmic fraction) was added to a streptavidin-coated microtiter plate and incubated with biotin-labeled anti-histone antibody and HRP-conjugated anti-DNA antibody. After 2 h, immune complexes were detected by incubating with substrate (2,2'-azino-di[3ethyl-benzthiazolin-sulfonate]) for 10 min, and absorbance was measured at 405 nm.

Statistical analysis
Data are presented as the means of three independent experiments with SEM. Statistically significant differences between control and treatment groups were identified using two-way ANOVAs followed by Tukey posthoc tests. A p value of less than 0.05 was considered statistically significant.