|Home | About | Journals | Submit | Contact Us | Français|
Human lung tumors aberrantly express the 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3)-catabolizing enzyme, CYP24. We hypothesized that CYP24 reduces 1,25(OH)2D3-mediated transcription and allows lung cancer cells to escape its growth-inhibitory action. To test this, H292 lung cancer cells and the CYP24-selective inhibitor CTA091 were utilized. In H292 cells, CTA091 reduces 1,25(OH)2D3 catabolism, significantly increases 1,25(OH)2D3-mediated growth inhibition, and increases 1,25(OH)2D3 effects on induced and repressed genesin gene expression profiling studies. Pathway mapping of repressed genes uncovered cell cycle as a predominant 1,25(OH)2D3 target. In H292 cells, 1,25(OH)2D3 significantly decreases cyclin E2 levels and induces G0/G1 arrest. A broader set of cyclins is down-regulated when 1,25(OH)2D3 is combined with CTA091, and cell cycle arrest further increases. Effects of CTA091 on 1,25(OH)2D3 signaling are vitamin D receptor-dependent. These data provide evidence that CYP24 limits 1,25(OH)2D3 anti-proliferative signaling in cancer cells, and suggest that CTA091 may be beneficial in preserving 1,25(OH)2D3 action in lung cancer.
The active metabolite of vitamin D3, 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3)1 inhibits carcinogenesis and suppresses the growth of established tumors (Eisman et al., 1987; McElwain et al., 1995; Banach-Petrosky et al., 2006; Mernitz et al., 2007; Welsh, 2007). The effects of 1,25(OH)2D3 are mediated by binding to the vitamin D receptor (VDR), a member of the nuclear steroid hormone receptor superfamily (Evans, 1998). Upon binding 1,25(OH)2D3, VDR forms a heterodimer with the retinoid-X-receptor (RXR) and modulates the expression of genes whose promoters contain a vitamin D response element (VDRE). The VDREs of genes that are positively regulated by 1,25(OH)2D3 are comprised of direct repeats of two hexameric-core-binding motifs separated by 3 or 4 nucleotides. Known inducible 1,25(OH)2D3 targets in tumor cells include genes implicated in G0/G1 cell cycle arrest (such as the cyclin-dependent kinase inhibitor, p21Waf1/Cip1), differentiation, and apoptosis (Swami et al., 2003; Krishnan et al., 2004; White, 2004). It is possible that mechanisms interfering with 1,25(OH)2D3 signaling in tumor cells could enhance properties of tumor growth and progression.
1α,25-dihydroxyvitamin D324-hydroxylase (CYP24) is the primary enzyme responsible for the catabolic inactivation of 1,25(OH)2D3 and has been implicated as an oncogene (Prosser and Jones, 2004; Albertson et al., 2000). When bound by ligand, the VDR induces CYP24transcription through a series of VDREs located in the CYP24 promoter and by binding to a distal enhancer region (Chen and DeLuca, 1995; Ohyama et al.,1996; Meyer et al., 2010). This auto-regulatory loop is thought to limit 1,25(OH)2D3 signaling and the potential for hypercalcemia toxicity. We and others have demonstrated that CYP24expression is dysregulated in lung cancer: It is frequently expressed in non-small cell lung cancer (NSCLC) cell lines under basal growth conditions, andCYP24 mRNA and protein are over-expressed in primary human lung tumors(Beer et al., 2002; Anderson et al., 2006; Parise et al., 2006). In patients diagnosed with adenocarcinoma of the lung, CYP24 expression is independently prognostic of survival: Five-year survival rates were 81% in individuals whose tumors expressed low levels of CYP24mRNA and 42% for those with high CYP24mRNA expression (Chen et al., 2011).
We hypothesized that CYP24 promotes tumor growth by restricting 1,25(OH)2D3-mediated transcriptional regulation, thus allowing lung cancer cells to escape its inhibitory action. The studies outlined in this report were designed to test this hypothesis. In the experiments described, CYP24 inhibition was achieved using the highly selective inhibitor CTA091. CTA091 is a 24-sulfoximine analogue of 1,25(OH)2D3 which binds to the substrate binding pocket of CYP24 (Kahraman et al., 2004). It inhibits CYP24 at low nanomolar concentrations (IC50, 7.4 nM), in comparison with the high nanomolar concentrations which are required to inhibit enzymes involved in the biosynthesis of 1,25(OH)2D3 (CYP27A1 (IC 50 > 1000 nM) or CYP27B (IC 50, 554 nM) (Kahraman et al., 2004). CTA091 has low calcemic activity and does not itself regulate gene expression through the VDR (Kahraman et al., 2004; Posner et al., 2010). The ability of 1,25(OH)2D3 to regulate gene expression and induce cell cycle arrest was significantly increased in the presence of CTA091, indicating that CYP24 restricts anti-proliferative 1,25(OH)2D3 signaling in lung cancer cells.
1, 25 (OH) 2 D3was purchased from Sigma-Aldrich (St. Louis, MO, USA). The CYP24 inhibitor CTA091 was generously provided by Cytochroma, Inc. (Markham, Ontario, Canada). Mouse anti-CYP24 antibody (Clone 1F8) was purchased from Sigma-Aldrich. Rat anti-VDR antibody (MA1-710) was from Thermofisher(Pittsburgh, PA, USA). Mouse anti-Cyclin A (4656), mouse anti-Cyclin D1 (2978) and rabbit anti-Cyclin E2 (4132) antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). Rabbit anti-actin antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Donkey anti-rabbit IgG (NA934) and sheep anti- mouse IgG (NA931) horseradish perioxdase (HRP)-linked species-specific antibodies were purchased from Amersham Pharmacia Biotech (Little Chalfort, UK). HRP-conjugated donkey anti-rat IgG (712-035-150) was purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA, USA).
H292, HCC827, and A549human lung cancer cells were obtained from the American Type Culture Collection (Manassas, VA). H292 cells were maintained in RPMI 1640 supplemented with 10% fetal bovine serum (FBS) (HyClone Laboratories, Inc., Logan, UT), 2 mM L-glutamine, 10 mM HEPES, 1 mM sodium pyruvate, and 100 U/ml penicillin-streptomycin. HCC827 cells were cultured in RPMI 1640 supplemented with 10% FBS,2 mM L-glutamine, and 100 U/ml penicillin-streptomycin. A549 cells were maintained in BME supplementedwith 10% FBS,2 mM L-glutamine, and 100 U/ml penicillin-streptomycin. Cells were maintained at 37°C in a humidified atmosphere containing 5% CO2. The cell lines were authenticated by RADIL (St. Louis, MO) and determined to be mycoplasma-free using the VenorGeM mycoplasma detection kit (Sigma Aldrich).
H292 cells were seeded at a density of 6 ×105 cells per T25 flask in growth medium containing 10% charcoal-stripped serum (CSS). The next day, the medium was removed, the cells were washed with PBS, and 5.0 mL of fresh medium containing 100 nM 1,25(OH)2D3 or 100 nM 1,25(OH)2D3 plus CTA091 was added. At various times post-treatment (0h or 24h), cell homogenates were prepared by scraping the cells up into the tissue culture medium. Homogenates were flash frozen in liquid nitrogen and stored at −80°C until analysis by LC-MS/MS.
A standard curve for the 1,25(OH)2D3 assay was prepared by placing duplicate, 1 ml samples of tissue culture medium containing 1,3, 5, 10, 30, 50 and 100 ng/ml 1,25(OH)2D3into 13 × 100 mm glass screw top tubes. Ten microliters of a 5 μg/mL solution of internal standard D6-1,25(OH)2D3 in methanol were added to each tube, followed by 1 ml of methyl-tert-butyl ether. The samples were vortexed at a setting of 8 on the vortex genie for 1 minute. The samples were then centrifuged at 2,000 × g for 5 minutes. The tubes were then placed into a dry ice cold methanol bath for 10 minutes. The resulting supernatants were evaporated to dryness under a stream of nitrogen at 27°C. Each dried residue was dissolved in 100 μL of mobile phase, vortexed briefly, transferred to HPLC autosampler vials, and injected into the LC-MS/MS system. Experimental samples (1.0 mL each homogenate) were processed similarly.
An Agilent (Palo Alto, CA) 1100 thermostatedautosamplerand binary pump were used as the LC system. The column was a Phenomenex (Torrance, CA, USA) SYNERGI 4 micron Hydro-RP 80 2×100 mm. The HPLC mobile phase was isocratic consisting of methanol-water (85:15) with 2 mM ammonium acetate (by volume). The flow rate was 0.3 ml/min. Sample injection was 50μL. Mass detection was carried out using a Waters (Milford, MA, USA) Quattromicro triple stage bench top mass spectrometer with electrospray ionization in positive ion mode and multiple reaction mode (MRM). The settings of the mass spectrometer were as follows: The Capillary voltage was 4.0 kV. The Cone voltage was 15 V. The source temperatureand desolvation temperatures were 120°C and 400°C, respectively. The cone and desolvation gas flow were 110 and 550 L/h, respectively. The collision voltage was set at 12 V. The MRM m/z for 1,25(OH)2D3 was 434>399 and 440>405m/z for [D6]-1,25(OH)2D3 (internal standard). Waters (Milford, MA, USA) Masslynx version 4.0 softwarewas used to control the LC system and mass spectrometer and collect data. The IS ratio was calculated for each standard by dividing the analyte peak area by the peak area of the internal standard. Standard curves of 1,25(OH)2D3 were constructed by plotting the IS ratio versus the known concentration of analyte in each sample. Standard curves were fit by linear regression with weighting by 1/y2, followed by back calculation of concentrations.
Cells were seeded in 6-well plates at a density of 250 cells/well in complete growth medium. The next day, cells were treated with fresh medium (controls) or medium containing the indicated concentrations of 1,25(OH)2D3 ± CTA091. Treatments were repeated every 3 days. After 7d, colonies were fixed with methanol and stained with crystal violet, as described by us previously (Owonikoko et al. 2010). To enumerate colonies, grids were scored onto the back of each plate. Colonies in each section of the grid were inspected using a microscope, and those containing ≥ 50 cells were counted. The percent remaining colonies was calculated using the equation: % Remaining Colonies = 100 × [number colonies for treatment group/average number colonies for control group].
H292 cells (5× 105/well) were seeded into 6-well plates in complete growth medium. When still sub-confluent, the cells were treated with vehicle (control), CTA091 (50 nM), 1, 25(OH)2D3 (10 nM or 100 nM) or 1,25(OH)2D3 plus CTA091 for 24 h. RNA was extracted using the PerfectPure RNA Cultured Cell Kit (5 Prime, Gaithersburg, MD, USA) in accordance with the manufacturer’s instructions. The RNA concentrations were determined by NanoDrop, and RNA quality was assessed by RNA Integrity Number (RIN) using the Agilent Bioanalyzer. Only RNA preparations having a RIN greater than 7.0 were used for gene expression profiling. Whole genome gene expression analysis was carried out using the Illumina HT-12 Expression BeadChip which targets more than 25,000 annotated genes with more than 48,000 probes derived from the RefSeq (Build 36.2, Rel 22) and UniGene databases. Image processing and raw data extraction were performed using IlluminaGenomeStudio software. The data were normalized for background correction and plate scaling using GenomeStudio, and differential expression analyses were carried out using the Mann-Whitney error model. Differential expressionscores (diff scores) were used to identify genes that were significantly induced (diff score ≥ +13) or repressed (≤ -13) by each treatment compared to vehicle control. To identify candidate cellular pathways that were modulated by treatment, lists of significantly regulated genes were imported into MetaCore software (St. Joseph, MI).
RNA was prepared as outlined above (for gene expression analysis). RNA (500 ng) was converted into cDNA using the First Strand cDNA Synthesis kit from Origene (Rockville, MD), according to manufacturer’s instructions. A fixed volume of cDNA (1.3 μL) was used for each PCR reaction. The reactions were run on a BioRadiCycler (initial incubation at 95°C for 2 min followed by 30 cycles of 95°C for 30 sec, 60°C for 45 sec, 72°C for 1 min). Primer sequences were: CD14 Forward-5′-GCCGCTGTGTAGGAAAGAAG-3′; CD14 Reverse-5′-TAGGTCCTCGAGCGTCAGTT-3′; LL37 Forward-5′-CCAGGTCCTCAGCTACAAGG-3′; LL37 Reverse-5′-GGGTAGGGCACACACTAGGA-3′; CDH5 Forward-5′-ACCGGATGACCAAGTACAGC-3′; CDH5 Reverse-5′-CGGATGGAGTATCCAATGCT-3′. PCR products were resolved on 1.2%agarose gels, visualized by staining with ethidium bromide, and photographed. Densitometry was used to quantify each PCR product, and the resulting intensity values were normalized to GAPDH. GAPDH primers were purchased from Clontech (Mountainview, CA).
H292 cells (3× 105/well) were seeded into 6- well plates in RPMI 1640 complete growth medium without antibiotics. When cells reached approximately 50% confluence, they were tranfected with VDR siRNA (L003448-00, Dharmacon, Lafayette, CO) or control non-targeting siRNA (Dharmacon, D-001810-10-05) using the DharmaFECT Duo Transfection reagent in accordance with the manufacturer’s instructions. At 5h after transfection, the cells were rinsed with PBS, and then treated with vehicle, CTA091, or 1,25(OH)2D3 ± CTA091 in medium containing CSS. Whole cell extracts were prepared 3d post-treatment.
Procedures were conducted essentially as described previously (Parise et al., 2006). Briefly, cells were lysed in TX-100/SDS lysis buffer supplemented with protease and phosphatase inhibitors at 4°C for 30 min. Protein extracts were clarified by centrifugation at 4°C at 12,000 × g for 10 min. Protein concentrations were determined using the BCA protein assay (Pierce, Rockford, IL). Proteins (20 to 30 μg) were separated on 10% Precast SDS gels (BioRad, Hercules, CA)and then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore Corp., Bedford, MA). Blots were incubated with primary antibodies at 4°C overnight. The blots were washed in TTBS (0.05 % Tween 20, 150 mMNaCl, 10 mMTris-HCl, pH 7.6) and then incubated with a HRP-conjugated secondary antibody for 1 h. Blots were developed using ECL or supersignalchemiluminescent reagents. Densitometry was performed on resulting films using a hpscanjet 3970 scanner and UN-SCAN-IT 6.1 software (Silk Scientific Corporation, Orem, Utah).
H292 cells were seeded in 6-well plate at a density of 2 × 105/well. The next day, the cells were treated with vehicle, CTA091, or 1,25(OH)2D3 ± CTA091 in medium containing CSS. Forty-eight h post-treatment, cells were harvested for cell cycle analysis. To do this, cells were collected after trypsinization and washed twice with phosphate-buffered saline (PBS) containing 0.1% BSA. The cells were then fixed for at least 1 h with 70% ethanol at 4°C, washed twice with PBS, centrifuged, and resuspended in 1 ml PBS containing 50 μg propidium iodide (PI) (Invitrogen, Carlsbad, CA) and 3.8 mM sodium citrate. RNase A(Sigma-Aldrich) was added to a final concentration of 20 μg/mL, and the cell suspensions were incubated 3 h at 4°C. Immediately after RNase digestion, the cells were analyzed on anAccuri C6 flow cytometer (Ann Arbor, MI). A minimum of 5,000 events was collected for each sample. Data were analyzed using CFlowPlus Analysis software.
All statistical analyses were conducted using SAS (version 9.2; SAS Institute Inc., Cary, NC). For the catabolism assay, a t-test was used to compare the control group separately with each treatment group. For all other assays, analysis of variance (ANOVA; GLM procedure in SAS) was used to compare the means of the different treatment groups. This was followed by pairwise comparisons, specifically: between vehicle (control) and each treatment group, between 1,25(OH)2D3 treatment dose levels, and between 1,25(OH)2D3 and 1,25(OH)2D3 + CTA091 treatment groups. We controlled for batch/experiment effect when appropriate by adding batch/experiment as a categorical variable to the model. P-values < 0.05 were considered statistically significant.
Our prior data indicate that, when expressed in lung cancer cells, CYP24 catabolizes 1,25(OH)2D3 and significantly reduces its anti-proliferative activity (Parise et al., 2006). We hypothesized that this loss of activity is due to a CYP24-dependent diminution in 1,25(OH)2D3 signaling. The following studies using the human lung cancer cell line, H292, were undertaken to test this hypothesis. H292 cells express the VDR and low levels of CYP24 under basal growth conditions. When treated with 1,25(OH)2D3, H292 cells are induced to express CYP24 (Fig. 1A). This is expected as CYP24 is a known transcriptional target of 1,25(OH)2D3(Chen 1995). CYP24 induction is associated with a significant decrease in 1,25(OH)2D3 levels in H292 tissue culture homogenates (Fig. 1B, compare the 1,25(OH)2D3 concentration in the t=0 sample with the 1,25(OH)2D3 concentration in the 24h/0 nM CTA091 sample). In contrast, when H292 cells are exposed to 1,25(OH)2D3 in the presence of the CYP24-selective inhibitor CTA091 (at concentrations ≥ 10nM), there is no significant loss of 1,25(OH)2D3 from the culture system (Fig. 1B). 1,25(OH)2D3 stabilization occurs despite the fact that CYP24 protein levels are consistently higher in cells treated with 1,25(OH)2D3 plus CTA091 compared to 1,25(OH)2D3 alone (Fig. 1A). This latter observation is attributed to the fact that when 1,25(OH)2D3 catabolism is inhibited by CTA091, the level of VDR-dependent transcription of CYP24 exceeds that which is induced by 1,25(OH)2D3 alone (data not shown). The effects of CTA091 on 1,25(OH)2D3 catabolism are not specific to H292 cells. CTA091 also inhibits catabolism of 1,25(OH)2D3 in 128.88T lung cancer cells that express high basal levels of CYP24 (Parise et al., 2006).
To examine the effects of CYP24 inhibition on cell growth, clonogenic assays were conducted. When used as a single agent, CTA091 had no significant effect on the growth of H292 cells at any of the concentrations tested (Fig. 1C). In contrast, 1,25(OH)2D3 inhibited growth in a dose-dependent manner (Fig. 1D; compare cells treated with 10 nM 1,25(OH)2D3vs. 100 nM 1,25(OH)2D3). The anti-proliferative activity of 1,25(OH)2D3 in H292 cells was significantly increased by CTA091 (Fig. 1D; compare cells treated with 10 nM 1,25(OH)2D3plus CTA091 vs. 10 nM 1,25(OH)2D3 alone). The CYP24 inhibitor also significantly increased the anti-proliferative effects of 1,25(OH)2D3 in HCC827 cells, which express high levels of CYP24 upon 1,25(OH)2D3 treatment, and in A549 cells, which express high basal levels of CYP24 (Fig. 1D). Therefore, the effects of CTA091 on 1,25(OH)2D3 anti-proliferative activity are not cell line specific.
To determine whether CYP24 impacts transcriptional regulation by 1,25(OH)2D3 in H292 cells, a whole-genome, gene expression profiling study was undertaken. This approach was selected (instead of using a VDRE-reporter construct to monitor 1,25(OH)2D3-mediated gene regulation) because it afforded the opportunity to examine effects on a broad array of genes and to gain insights into the anti-proliferative actions of 1,25(OH)2D3 in lung cancer cells. H292 cells were treated for 24 h with vehicle (control), CTA091 alone, 10 nM 1,25(OH)2D3 ± CTA091, or 100 nM 1,25(OH)2D3. CTA091 was used at a final concentration of 50 nM, which was the lowest concentration at which we observed consistent inhibition of 1,25(OH)2D3 catabolism (Fig 1B). RNA was extracted and profiled using the IlluminaHT-12BeadChip. Compared to vehicle control, CTA091 alone significantly increased the expression of 161 genes (maximal induction 2-fold) and decreased the expression of 120 genes (maximal repression, 2-fold). However, these genes did not map strongly to specific gene ontology pathways (data not shown). The highest ranking induced pathway was apoptosis and survival (P = 4.19 × 10−4), and the highest ranking repressed pathway was galactose metabolism (P = 1.76 × 10−3). 1,25(OH)2D3 (100 nM) significantly increased the expression of 378 genes (1.5-fold to 105-fold induction, compared to control) and decreased the expression of 431 genes (1.4-fold to 5-fold suppression, compared to control). Genes that were significantly induced by 1,25(OH)2D3 included those whose promoters are known to contain VDREs such as CYP24 (105-fold increase), CD14 (12-fold increase), LL37 (4.6-fold increase), cytidinedeaminase (2.9-fold increase), and TRPV6 (2-fold increase) (Wang et al., 2005). To validate the array data, 2 genes that were induced by 100 nM 1,25(OH)2D3 (CD14 and LL37) and 1 gene that was repressed (CDH5) were selected for independent verification by RT-PCR. As shown in Fig. 2, the 3 genes were regulated by 1,25(OH)2D3 in the manner predicted by the array results.
Consistent with prior reports showing that 1,25(OH)2D3 inhibits progression through the cell cycle in many tumor cell types (reviewed in (Samuel and Sitrin, 2008), we found that cell cycle pathways were featured prominently among the top 5 gene ontology pathways that were repressed by 1,25(OH)2D3 (100 nM) and by 1,25(OH)2D3 (10 nM) + CTA091 in H292 cells (Table 1). Specific effects of 1,25(OH)2D3 on cell cycle regulatory proteins will be discussed in more detail below.
To establish the effect of CYP24 on 1,25(OH)2D3-mediated gene regulation in H292 cells, we compared the potency for gene regulation by 10 nM 1,25(OH)2D3 (where CYP24 is active) with that of 10 nM 1,25(OH)2D3 plus CTA091 (where CYP24 is inactive). We selected as our data set for this analysis only those genes that were significantly induced or repressed by 100 nM 1,25(OH)2D3 (identified above). The majority of 1,25(OH)2D3-induced genes (92%) were expressed at a greater level when CYP24 was inhibited with CTA091 (see CD14 and LL37 as representative examples, Fig. 2). This proportion is significantly higher than random chance (P< 0.0001). Conversely, most of the 1,25(OH)2D3-repressed genes (93%) were expressed at a lower level in the presence of CTA091 (see CDH5 as a representative example, Fig. 2). This proportion is also significantly higher than random chance (P< 0.0001). We conclude from these data that the capacity of 1,25(OH)2D3 to induce or repress the expression of specific genes is increased when its CYP24-mediated catabolism is inhibited.
Little is known about the mechanisms by which 1,25(OH)2D3 regulates the growth of lung cancer cells. Our array data suggest that a predominant effect of 1,25(OH)2D3 is to block cell cycle progression. To test this, we examined the effects of 1,25(OH)2D3 ± CTA091 on cyclin D1, cyclin A, and cyclin E2 protein levels by immunoblot (Fig. 3). We focused on the expression of these cyclins initially because they promote G1/S progression, and the G1/S transition was significantly repressed by 1,25(OH)2D3 treatment in H292 cells (Table 1). CTA091 alone did not modulate cyclin expression. Compared to vehicle, treatment with 100 nM 1,25(OH)2D3 was only associated with a significant reduction in cyclin E2 expression, while treatment with 1,25(OH)2D3 plus CTA091 resulted in a significant decrease in each of the cyclins examined. Thus, a broader set of cyclins is affected when 1,25(OH)2D3 catabolism is inhibited with CTA091.
To determine whether the effects of CTA091 on cyclin expression were dependent upon 1,25(OH)2D3 signaling through the VDR, we suppressed receptor expression using siRNA. Cells were transfected with siRNA, and 5 h later cells were treated as indicated in Fig. 4. Whole cell extracts were prepared 3 d post-treatment and analyzed for expression of VDR, CYP24, and cyclins by immunoblot. This time point was chosen based on manufacturer’s recommendations for achieving maximum siRNA-mediated down-modulation of protein expression. The VDR siRNA effectively reduced VDR protein levels and resulted in a functional loss of signaling (Fig. 4). This was evidenced by our finding that CYP24 was induced by 1,25(OH)2D3 + CTA091 in cells transfected with control siRNA but not in VDR-deficient cells. CTA091 alone had no significant effect on cyclin expression in cells transfected with either the control or VDR siRNAs. Compared to vehicle, cyclins A and E2 were significantly decreased in control siRNA-transfected cells treated with 1,25(OH)2D3 + CTA091, as expected. In contrast, no significant decreases in cyclin expression were observed in the VDR-deficient cells treated with either 1,25(OH)2D3 or 1,25(OH)2D3 + CTA091. These results support the conclusion that 1,25(OH)2D3down-modulates cyclin expression via a VDR-dependent process, and that cyclin down-modulation is compromised by CYP24 activity.
A 1,25(OH)2D3-mediated decrease in expression of cyclins required for S-phase entry is predicted to result in accumulation of cells in the G1 phase of the cell cycle. To determine if 1,25(OH)2D3 mediated cell cycle arrest, and whether such arrest was affected by CYP24, we treated cells for 48 h, stained them with propidium iodide (PI), and examined them by flow cytometry(Fig.5). CTA091 alone had no effect on the cell cycle distribution. 1,25(OH) 2D3 treatment resulted in a dose-dependent increase in cells in G0/G1 and a decrease in cells in G2/M compared to vehicle controls. The percentage of cells in G0/G1 increased significantly following treatment with 10 nM 1,25(OH)2D3 plus CTA091 compared to 10 nM 1,25(OH)2D3 alone, consistent with the conclusion that CYP24 activity restricts the induction of cell cycle arrest by 1,25(OH)2D3.
The vitamin D3 catabolizing enzyme CYP24 is frequently over-expressed in lung cancer (Beer et al., 2002; Anderson et al., 2006; Parise et al., 2006), and it is independently associated with poor survival in patients diagnosed with adenocarcinoma of the lung (Chen et al., 2011). We hypothesized that CYP24 contributes to the growth or progression of lung cancer by diminishing 1,25(OH)2D3 signaling. In support of this hypothesis, we now provide direct evidence that transcriptional regulation by 1,25(OH)2D3 is compromised in lung cancer cells in which CYP24 is active. Moreover, we show that 1,25(OH)2D3 mediates G0/G1 arrest in lung cancer cells, and that arrest can be achieved at ten-fold lower 1,25(OH)2D3 concentrations when CYP24 is inactivated. These data provide further support for the idea that, when over-expressed, CYP24 allows tumor cells to bypass the anti-proliferative effects of endogenous 1,25(OH)2D3.
Consistent with the notion that 1,25(OH)2D3 signaling suppresses lung cancer growth, we recently demonstrated that high nuclear VDR expression is associated with better overall survival in NSCLC. The 5 year OS rates were 59% (95% CI: 39–80%) for patients with high nuclear VDR expression versus 27% (95% CI: 12–41%) for low nuclear VDR expression (Srinivasan et al., 2011). Data presented in this manuscript provide a mechanistic explanation for these prior findings: 1,25(OH)2D3signaling through the VDR results in down-modulation of cyclins which promote S phase entry (Fig. 4). Although cyclin down-modulation was VDR-dependent in H292 cells, we cannot conclude from our studies that cyclins are direct transcriptional targets of 1,25(OH)2D3. Rather, cyclin suppression (observed at 24 h post-1,25(OH)2D3 treatment) may have occurred as a downstream consequence of the prior modulation of direct target genes. For example, as recently reported in ovarian cancer cells (Shen et al. 2011), 1,25(OH)2D3 may directly repress transcription of a growth factor receptor, which results in the subsequent cyclin down-modulation.
A corollary to our hypothesis that CYP24 allows tumor cells to bypass the anti-proliferative effects of endogenous 1,25(OH)2D3 is that therapeutic approaches to block CYP24 will restore anti-proliferative 1,25(OH)2D3 signaling in lung cancer. The CYP24-selective inhibitor employed in this study (CTA091) increased 1,25(OH)2D3 stability, transcriptional regulation, and growth inhibition in H292 cells. Thus, CTA091, represents one possible agent that could be used to increase 1,25(OH)2D3 signaling therapeutically. CTA091 has no effects on the growth of H292 cells when administered as a single agent (Fig.1). However, it was not devoid of biological activity in H292 cells as it did induce the expression of 161 genes and decrease the expression of 120 genes upon 24 h exposure. Of these genes, 7 were also induced and 16 were also repressed by 1,25(OH)2D3 (100 nM). One possible explanation for these findings is that single agent CTA091 regulated gene expression by stabilizing low levels of endogenous vitamin D3 metabolites that were contained in our tissue culture medium, and the genes that were regulated at these hormone concentrations differed from those that were modulated at the nanomolar concentrations of 1,25(OH)2D3 used in our gene expression profiling experiments. Alternatively, CTA091 may regulate gene expression by a VDR-independent mechanism, which remains to be elucidated. Future experiments will be required to distinguish between these possibilities.
With regard to its potential in vivo application, CTA091 has been delivered to normal rats, where it had the expected effect of increasing 1,25(OH)2D3 exposure (Posner et al., 2010). No acute toxicities were reported in these short term studies. We have also administered CTA091 to mice in combination with 1,25(OH)2D3 on a once weekly schedule for 3 weeks and observed no toxicity (as assessed by weight loss or increase in serum calcium) (data not shown). Cumulatively, these data suggest that CTA091 administration may represent a feasible approach to increase transcriptional regulation by 1,25(OH)2D3 in cancer.
Anti-proliferative effects of 1,25(OH)2D3 in lung cancer models have been previously reported by us and others (Parise et al. 2006; Mernitz et al. 2007; Chen et al. 2011). However, little information exists as to the basis for this activity. Our gene expression profiling studies revealed a predominant effect of 1,25(OH)2D3 on cell cycle regulation in H292 cells (Table 1). Consistent with this result, we were able to confirm in independent assays that 1,25(OH)2D3 decreases expression of cyclins that govern S phase entry and mediates G0/G1 arrestin H292 cells. While a limitation of our functional studies is the use of single cell line, G1 arrest is commonly observed upon 1,25(OH)2D3 treatment in other epithelial tumor cell lines (i.e. MCF-7, LNCaP, and SCCVII SF) (McElwain et al., 1995; Zhuang and Burnstein, 1998; Jensen et al., 2001). Extensive biochemical studies in MCF-7 breast cancer cells led to a model in which (1) 1,25(OH)2D3 inhibits cyclin D-dependent kinases; (2) This leads to Rbhyperphosphorylation and sequestration/inhibition of the transcription factor, E2F; and (3) Continued E2F inhibition results in suppression of cyclin A and cyclin E protein expression and a G1 phase block (Jensen et al., 2001). The observed 1,25(OH)2D3-mediated changes in cyclin expression in H292 cells would support the existence of a similar mechanism for growth inhibition.
We focused in this report on those genes/pathways that were repressed in response to 1,25(OH)2D3 treatment. It is important to note that 1,25(OH)2D3 also induced the expression of more than 400 genes, and that these could positively or negatively influence its anti-tumor activity. A detailed functional investigation of 1,25(OH)2D3-induced genes in lung cancer growth control, and the influence of CYP24 on these genes, is ongoing in our laboratory.
The authors thank Mr. Michael C. Gorry and Dr. Richard Steinman for many helpful discussions during the course of this work and members of the University of Pittsburgh Cancer Institute (UPCI) writing group for their thoughtful review of a draft of this manuscript. This work was supported by R01 CA132844 and P50 CA090440. This project used the UPCI Clinical Pharmacology Analytical Facility and was supported in part by award P30 CA047904.
Disclosures: The authors have nothing to disclose.
1Abbreviations: 1,25(OH)2D3, 1α,25-dihydroxyvitamin D3; CSS, Charcoal-stripped serum; CYP24, 1α,25-dihydroxyvitamin D324-hydroxylase; CTA091, MK-24(S)-S(O)(NH)-Ph-1; PBS, Phosphate buffered saline; VDR, Vitamin D Receptor; VDRE, Vitamin D Response Element
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.