CCAAT/Enhancer-Binding Protein D: a Molecular Target of 1,25-Dihydroxyvitamin D3 in Androgen-Responsive Prostate Cancer Lncap Ce

Research Article

CCAAT/Enhancer-Binding Protein D: A Molecular Target of

1,25-Dihydroxyvitamin D3 in Androgen-Responsive Prostate Cancer LNCaP Cells

Takayuki Ikezoe,1,2 Sigal Gery,1 Dong Yin,1 James O’Kelly,1 Lise Binderup,3 Nathan Lemp,1 Hirokuni Taguchi,2 and H. Phillip Koeffler1

1Division of Hematology/Oncology, Cedars-Sinai Medical Center, University of California at Los Angeles School of Medicine, Los Angeles, California; 2Department of Internal Medicine, Kochi Medical School, Kochi, Japan; and 3Leo Pharmaceuticals, Ballerup, Denmark

Abstract down-regulation of the antiapoptotic protein Bcl-2 and Bcl-XL 1,25-Dihydroxyvitamin D [1,25(OH) D ], the active metabolite (11). 3 2 3 CCAAT/enhancer binding proteins (C/EBP) are a highly of vitamin D3, inhibits the proliferation of prostate cancer cells. However, the molecular mechanisms by which 1,25(OH) D conserved family of leucine zipper type (bZIP) DNA-binding 2 3 a inhibits the proliferation of these cells remain to be fully proteins which are composed of six isoforms including C/EBP- , h g y q elucidated. In this study, we used microarray technology to C/EBP- ,C/EBP-,C/EBP-,C/EBP- as well as C/EBP identify target genes of 1,25(OH) D in androgen-responsive homologous protein (12). All C/EBPs share conserved COOH- 2 3 terminal regions that contain leucine zipper dimerization motifs prostate cancer LNCaP cells. 1,25(OH)2D3 up-regulated CCAAT/ enhancer-binding protein D (C/EBPD)byf5-fold in these cells. adjacent to basic DNA-binding domains. Their NH2-terminal Knockdown of C/EBPD expression by RNA interference showed regions are more diverse and contain transcriptional activation that C/EBPD is essential for the significant growth inhibition of domains (12). C/EBPs form homodimers and heterodimers with other C/EBP family members as well as other transcription LNCaP cells in response to 1,25(OH)2D3 treatment. Moreover, we found that 1,25(OH) D induced C/EBPD in other cancer factors such as Fos, cyclic AMP–responsive element binding 2 3 n cells, including the estrogen receptor (ER)–expressing MCF-7 protein/activating transcription factor, and nuclear factor B n and T47D breast cancer cells that are sensitive to the growth (NF- B; refs. 10, 13–15). C/EBPs are implicated in the regulation of growth and differentiation of a wide variety of cells such as inhibitory effects of 1,25(OH)2D3. On the other hand, 1,25(OH) D was not able to induce C/EBPD in either androgen hepatocytes, adipocytes, pneumocytes, as well as hematopoietic 2 3 q receptor–negative PC-3 and DU145 or ER-negative breast cells. For example, C/EBP plays an important role in the cancer MDA-MB-231 cells that were relatively resistant to differentiation of hematopoietic progenitor cells toward the granulocytic lineage; indeed, forced expression of C/EBPq in growth inhibition by 1,25(OH)2D3. Furthermore, forced expression of C/EBPD in prostate cancer LNCaP as well as breast U937 leukemic myeloblasts can induce these cells to differentiate q cancer MCF-7 and T47D cells dramatically reduced their clonal towards granulocytes (16). In addition, C/EBP deletional mice growth. Taken together, forced expression of C/EBPD in cancer have incomplete maturation of their granulocytes, and these cells cells may be a promising therapeutic strategy. (Cancer Res 2005; lack specific granule proteins (17). Furthermore, we have recently q 65(11): 4762-8) found that C/EBP interacted with cell cycle regulatory molecules, retinoblastoma and E2F during granulocytic differen- Introduction tiation in human and murine myeloid precursor cells (18). C/EBPa is also pivotal in terminal differentiation of granulocytes The 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] is a member of the as well as pneumocytes. Targeted inactivation of C/EBPa in mice seco-steroid hormone family, which controls calcium homeostasis showed hyperproliferation of type II pneumocytes and abnormal (1). The effects of 1,25(OH)2D3 are mediated mainly via alveolar structure, and histopathology of the liver displayed a interaction with a specific nuclear vitamin D3 receptor (VDR), structure resembling regenerative changes or hepatocellular which forms heterodimers with retinoid X receptor and binds to carcinoma (19). Others and ourselves have identified mutations the VDR response element, resulting in activation of target genes in the C/EBPa gene in individuals with acute myeloid leukemia, (1). 1,25(OH)2D3 induces growth arrest, differentiation, and myelodysplastic syndrome and non–small cell lung cancer apoptosis of a wide variety of cancer types, including those from (NSCLC); and these mutations disrupt the normal function of prostate (2–5), breast (6, 7), as well as leukemia (8, 9). For C/EBPa (20, 21). Recently, expression of C/EBPa was shown example, 1,25(OH)2D3 induced G0-G1 cell cycle arrest in down-regulated in NSCLC cells compared with normal lung association with up-regulation of cyclin-dependent kinase inhibi- a waf1 tissues, and forced expression of the p42 isoform of C/EBP tor (CDKI) p21 in androgen-responsive prostate cancer LNCaP reduced proliferation of NSCLC cells (22). cells (10). In addition, it induced apoptosis of LNCaP cells with C/EBPy is induced in murine mammary epithelial cells as their cell growth slows by either withdrawal of serum or exposure to oncostatin M, an interleukin 6–type cytokine (23–25). In addition, we have found that forced expression of C/EBPy in KCL-22 myeloid Note: T. Ikezoe and S. Gery contributed equally to the article. blast cells (cell line derived from chronic myeloid leukemia in Requests for reprints: Takayuki Ikezoe, Department of Internal Medicine, Kochi myeloid blast crisis) induced G0-G1 cell cycle arrest and Medical School, Nankoku, Kochi 783-8505, Japan. Phone: 81-88-880-2345; Fax: 81-88- 880-2348; E-mail: [email protected]. granulocytic differentiation of these cells in association with up- KIP1 I2005 American Association for Cancer Research. regulation of p27 protein (26).

Cancer Res 2005; 65: (11). June 1, 2005 4762 www.aacrjournals.org

In this study, we used microarray technology to identify the temperature of the product for 20 seconds. Specificity of PCR products was checked on agarose gel. All reactions were done in triplicates in an iCycler 1,25(OH)2D3 target genes in androgen-responsive LNCaP cells. 1,25(OH) D increased the level of C/EBPy in LNCaP cells; and iQ system (Bio-Rad Laboratories). For each sample, the amount of the 2 3 h forced expression of C/EBPy in these cells inhibited their clonal target gene and -actin (for internal control) was determined from a standard curve. The results are expressed in arbitrary units as a ratio of the growth. target gene transcripts/h-actin transcripts (each value represent the mean of three measurements of the sample). Materials and Methods Western blot analysis. Lysates were made by standard methods as Cell lines. Cell lines established from prostate cancer (LNCaP, PC-3, and previously described (29). Protein concentrations were quantitated using a DU145) and breast cancer (MCF-7, T47D, and MDA-MB-231) were obtained Bio-Rad assay (Bio-Rad Laboratories). Proteins were resolved on a 4% to 15% from American Type Culture Collection (Rockville, MD). LNCaP, PC-3, T47D, SDS polyacrylamide gel, transferred to an immobilon polyvinylidene difuride and MDA-MB-231 cells were cultured in RPMI 1640 (Life Technologies, membrane (Amersham Corp., Arlington Heights, IL), and probed sequen- y Grand Islands, NY) with 10% FCS. DU145 and MCF-7 cells were maintained tially with antibodies. Anti-C/EBP (Santa Cruz Biotechnology, Santa Cruz, h in DMEM with 10% FCS. CA), anti– -actin (Santa Cruz Biotechnology), and anti–glyceraldehyde-3- phosphate dehydrogenase (Research Diagnostics, Flanders, NJ) antibodies Chemicals. 1,25(OH)2D3 was obtained from Hoffmann-La Roche (Nutley, NJ) and was dissolved in absolute ethanol at 10À3 mol/L as a stock solution, were used. stored at À20jC, and protected from light. All-trans retinoic acid (ATRA) Small interference RNA. Primers were designed using the RNA was obtained from Sigma (St. Louis, MO), dissolved in absolute ethanol as a interference oligo retriever web site (http://katahdin.cshl.org:9331/RNAi/ stock concentration of 10À2 mol/L, stored at À20jC, and protected from html/rnai.html). The following small interfering (siRNA) sequence was light. Cycloheximide was obtained from Sigma, dissolved as a stock used: 5V-GCTGTCGGCTGAGAACGAGAAGCTGCACC. Scrambled siRNA was concentration of 10 mg/mL, and stored at À20jC. designed by the same method for control. The siRNA primers together with Soft agar colony assay. Cells were cultured in a two-layer soft agar the U6 promoter were cloned into pCR2.1 and confirmed by sequencing. y system for 14 days as previously described (9). Washed, single-cell LNCaP cells were cotransfected with either C/EBP siRNA or control siRNA suspension of cells were enumerated and plated into 24-well flat-bottomed along with pMSCVpuro vector (Clontech, Palo Alto, CA) and selected with 3 plates with a total of 500 cells per well in a volume of 400 AL per well. The puromycin. Equal numbers of surviving cells (3 Â 10 cell per well) were feeder layer was prepared with agar that had been equilibrated at 42jC. plated in 96-well plates and 24 hours later, treated either with control À7 Before this step, 1,25(OH)2D3 was pipetted into the wells. After incubation, diluent or with 1,25(OH)2D3 (10 mol/L). Cell proliferation was measured colonies were counted. Experiments were done thrice using triplicate plates after 5 days using MTT assays (Roche Diagnostics, Mannheim, Germany) per experimental point. according to the manufacturer’s protocol. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide Colony formation assay. Cells were split evenly into 6-well plates. After 4 f assays. Cells (10 /mL) were incubated with a variety of concentrations of growing to 60% confluence, cells were transfected with either pcDNA3.1- À9 À6 y 1,25(OH)2D3 (10 to 10 mol/L) for 6 days in 96-well plates (Flow C/EBP (26) or pcDNA3.1 empty vector (both contain the Neo-resistant Laboratories, Irvine, CA). After culture, cell number and viability were evaluated by measuring the mitochondrial-dependent conversion of the tetrazolium salt, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma), to a colored formazan product. MTT (0.5 mg/mL in PBS) was added to each well and incubated for 4 hours at 37jC. The medium was then carefully aspirated, and DMSO (Burdick & Jackson, Muskegon, MI) was added to solubilize the colored formazan product. Absorbance was read at 540 nm on a scanning multiwell spectrophotometer (Bio-Rad Laborato- ries, Hercules, CA) after agitating the plates for 5 minutes on a shaker. RNA extraction. LNCaP (5 Â 105/mL) cells were plated on 100-mm plates. On the following day, the medium was replaced with fresh RPMI À7 1640 containing 10% FCS with either 1,25(OH)2D3 (10 mol/L) or control diluent. After 18 hours of incubation, total RNA was extracted as previously described using TRIzol (Life Technologies; ref. 27) followed by a purification step using the RNeasy cleanup kit (QIAGEN, Inc., Valencia, CA) according to the manufacturer’s protocol. The RNA quality was confirmed on an agarose gel. Oligonucleotide microarrays. The HuGeneFL Array (Affymetrix, Inc., Santa Clara, CA) provides gene expression data for 5,600 full-length human sequences. The set of matched samples [LNCaP cells treated either with or without 1,25(OH)2D3] was studied independently twice. The preparation and microarray processing was done per standard Affymetrix protocol as previously published (28). The raw data images were analyzed using the GeneChip Analysis Suite (Affymetrix), and the data for each microarray were normalized by global scaling to a target value of 2,500 as we have previously reported (27, 28). The average background noise was in the range of 500. Real-time reverse transcription-PCR. One microgram of DNase I– treated RNA was reverse transcribed by using Moloney murine leukemia virus reverse transcriptase (Life Technologies), and 50 ng of the resulting cDNAs were used as templates for PCR. Real-time PCR was done using specific primers (sequences will be provided upon request), HotMaster Taq DNA Polymerase (Eppendorf, Hamburg, Germany) and SYBRGreen I Figure 1. (Molecular Probes, Eugene, OR). PCR conditions were as follows: 2 minutes Dose-response effects of 1,25(OH)2D3 on clonal proliferation of j j j prostate (A) and breast (B) cancer cells. Results are expressed as a mean at 94 C followed by 45 cycles of 94 C for 20 seconds, 60 C for 10 seconds, percentage of control plates containing no 1,25(OH)2D3. Point, mean of three 65jC for 25 seconds, and fluorescence determination at the melting independent experiments with triplicate dishes; bars, FSD. www.aacrjournals.org 4763 Cancer Res 2005; 65: (11). June 1, 2005

1,25(OH) D induces C/EBPD in LNCaP cells. Table 1. Inhibition of clonal proliferation of tumor cells 2 3 The gene showing the highest fold induction following 1,25(OH)2D3 by 1,25(OH)2D3 treatment was C/EBPy (4.7-fold). To confirm further that 1,25(OH) D induces C/EBPy expression in LNCaP, we did Cell line ED (mol/L) 2 3 50 Western blot analysis. As shown in Fig. 2A, the level of C/EBPy Â À10 protein was negligible in untreated LNCaP cells and treatment LNCaP 6 10 À9 À7 PC-3 1 Â 10À8 with 1,25(OH)2D3 (10 to 10 mol/L, 18 hours) markedly up- DU145 Resistant regulated its expression in a dose-dependent manner. ATRA MCF-7 7 Â 10À9 binds to retinoic acid receptor (RAR) and induces growth arrest T47D 8 Â 10À9 and differentiation of LNCaP cells (30); however, ATRA MDA-MB-231 Resistant (10À6 mol/L, 18 hours) was not able to induce expression of C/EBPy in LNCaP cells under similar culture conditions (Fig. 2B), suggesting that induction of C/EBPy seems specific to ligand- NOTE: The concentration of 1,25(OH)2D3 that induced 50% growth activated VDR not RAR. inhibition (ED50) was calculated from the dose-response curves shown Experiments were done to determine if the 1,25(OH)2D3- in Fig. 1. dependent increase in C/EBPy mRNA levels is a primary response or requires ongoing protein synthesis. LNCaP cells were treated with cycloheximide for 30 minutes followed by 1,25(OH)2D3 (10À7 mol/L) for 1, 4, and 6 hours. Real-time RT-PCR showed that gene). After 48 hours, medium was replaced with fresh medium containing the expression profile of C/EBPy following 1,25(OH) D treatment G418. Two weeks later, the cells were stained with 0.1% crystal violet to 2 3 was not perturbed by pretreating the cells with cycloheximide, assess colony formation. Colonies containing >40 cells were counted. This study was done twice in triplicate. indicating that de novo protein synthesis was not required for Statistical analysis. Statistical analysis was done by Student’s t test. 1,25(OH)2D3 to exert its effect (Fig. 2C). Suppression of C/EBPD reduces 1,25(OH)2D3-induced growth inhibition in LNCaP cells. To evaluate the role of y Results C/EBP in 1,25(OH)2D3-mediated growth inhibition, we applied y y Effect of 1,25(OH)2D3 on clonogenic growth of prostate and siRNA to silence C/EBP expression. Inhibition of C/EBP breast cancer cells. The LNCaP, PC-3, and DU145 prostate expression by the C/EBPy siRNA construct was shown in 293T cancer cells and MCF-7, T47D, and MDA-MB231 breast cancer cells cotransfected with a C/EBPy expression vector (pcDNA3.1- cells were cloned in soft agar in the presence of various C/EBPy) and either control siRNA (scrambled siRNA) or C/EBPy À10 À7 y concentrations (10 to 10 mol/L) of 1,25(OH)2D3. Dose- siRNA (Fig. 3A). The C/EBP siRNA also inhibited 1,25(OH)2D3- response curves were drawn, and the effective dose that inhibited dependent induction of C/EBPy protein in LNCaP cells (Fig. 3B). y 50% colony formation (ED50) was determined. 1,25(OH)2D3 To test the effect of C/EBP suppression on 1,25(OH)2D3-induced inhibited clonal proliferation of LNCaP, PC-3, MCF-7, and T47D growth inhibition, LNCaP cells were transfected with either y cells in a dose-dependent manner (Fig. 1A and B) with an ED50 of control siRNA or C/EBP siRNA and selected for 2 days with 6 Â 10À10,1Â 10À8,7Â 10À9, and 8 Â 10À9 mol/L, respectively puromycin. Equal numbers of surviving cells were grown either À7 (Table 1). On the other hand, both DU145 and MDA-MB-123 cells in the presence or absence of 1,25(OH)2D3 (10 mol/L) for 5 days were resistant to growth inhibition by 1,25(OH)2D3 (Fig. 1A and B). We also assessed the ability of 1,25(OH)2D3 to inhibit the proliferation of LNCaP cells by MTT assay on day 6 of culture; Table 2. Genes either induced or repressed after the effective dose that inhibited growth of LNCaP cells by 50% exposure of LNCaP cells to 1,25(OH)2D3 compared with diluent-treated cells was f10À6 mol/L (figure not shown). Gene name Accession no. Fold change Microarray analysis of 1,25(OH)2D3-inducible genes in LNCaP cells. To do oligonucleotide array analysis, LNCaP cells Array PCR À7 were cultured either without or with 1,25(OH)2D3 (10 mol/L, 18 hours). Total RNA was extracted and subjected to microarray C/EBPd M8366 4.7 5.8 analysis using HuGeneFL Array. A short exposure duration IjBa M69043 2.6 3.9 (18 hours) was chosen to enhance the identification of direct target MRP X78338 2.6 5.1 FK506-binding protein 5 U42031 2.7 5.8 genes that are responsible for growth inhibition and differentiation SUMO 1 U67122 2.6 2.6 induced by ligand-activated VDR. By using GeneChip Analysis Suite CD33 M23197 2.3 2.2 (Affymetrix), we sorted the genes that were called ‘‘present’’ in PMP70 X58528 2.0 3.2 control LNCaP cells and were either up-regulated or down- Renal cell carcinoma 1 D00591 0.32 1.1 z regulated by 2-fold after treatment with 1,25(OH)2D3 in two Neuronal cell adhesion molecule AB002314 0.4 0.43 independent experiments (Table 2). Six genes were up-regulated Paternally expressed gene 3 U90336 0.45 0.48 and three were down-regulated by 1,25(OH)2D3 in both experiments. The induction/repression of the genes identified by the microarray analysis was confirmed by quantitative real-time reverse Abbreviations: InBa, inhibitor of n light chain gene enhancer in B cells transcription-PCR (RT-PCR; Table 2). In eight of the nine genes a; MRP, multidrug resistance–associated protein; PMP 70, 70-kDa examined, the real-time RT-PCR data were consistent with the chip peroxisomal membrane protein. data, demonstrating that our microarray analysis was reliable.

Cancer Res 2005; 65: (11). June 1, 2005 4764 www.aacrjournals.org

amount of pcDNA3.1 empty vector (379 F 23 colonies per well; Fig. 4A). Western blot analysis showed that level of C/EBPy in the transiently transfected LNCaP cells, was comparable with levels in the 1,25(OH)2D3-treated LNCaP cells (Fig. 4B). 1,25(OH)2D3 up-regulates the expression of C/EBPy in both androgen receptor– and estrogen receptor–positive cancer cells. Among the prostate cancer lines, 1,25(OH)2D3 significantly inhibited growth of androgen receptor (AR)–positive LNCaP cells, had an intermediate effect on AR-negative PC-3 cells, and had minimally inhibitory activity on AR-negative DU145 cells (Fig. 1A). y We explored whether 1,25(OH)2D3 induced C/EBP in these AR- negative cells. Real-time RT-PCR analysis showed that in contrast À7 to LNCaP cells, 1,25(OH)2D3 (10 mol/L, 18 hours) was unable to increase the expression of C/EBPy in the AR-negative cells (Fig. 5). Some breast cancer cell lines, including MCF-7 and T47D cells, express estrogen receptor (ER) whose signaling plays an important role in carcinogenesis and progression of breast cancer cells (31). These ER-expressing breast cancer cells are growth inhibited when exposed to vitamin D3 (Fig. 1B). We therefore explored whether 1,25(OH)2D3 induced C/EBPy in these cells. Real-time RT-PCR analysis showed that 1,25(OH)2D3 (10À7 mol/L, 18 hours) increased the levels of C/EBPy by f5.7- and 3.5-fold, in MCF-7 and T47D, respectively (Fig. 5). On

Figure 2. 1,25(OH)2D3 induces C/EBPy in LNCaP cells. A, Western blot analysis of C/EBPy expression in LNCaP cells treated either with control diluent or 1,25(OH)2D3 for 18 hours at the indicated concentrations. B, Western blot analysis of C/EBPy expression in LNCaP cells treated with either control diluent, À7 À6 1,25(OH)2D3 (10 mol/L), or ATRA (10 mol/L) for 18 hours. The membranes were probed sequentially with anti-C/EBPy and either h-actin (A) or GAPDH (B) antibodies. C, real-time PCR analysis of RNA from LNCaP cells treated À7 either with control diluent or 1,25(OH)2D3 (10 mol/L) for 1, 4, and 6 hours either in the absence (ÀCHX) or presence (+CHX) of cycloheximide (10 Ag/mL). The results are expressed in arbitrary units as a ratio of C/EBPy transcripts/h-actin transcripts. Columns, means of triplicate samples; bars, FSD. and their growth rate was determine by MTT assays. As expected, 1,25(OH)2D3 treatment resulted in a marked inhibition (64%) in cell proliferation in the control-transfected cells (Fig. 3C). On the other hand, the responsiveness to 1,25(OH)2D3, although not completely extinguished, was significantly reduced (34%) in LNCaP cells transfected with the C/EBPy siRNA (Fig. 3C). Figure 3. Suppression of C/EBPy by siRNA reduces the antiproliferative effects y y of 1,25(OH)2D3 in LNCaP cells. Inhibition of C/EBP by siRNA in 293T (A) and These results indicate that C/EBP plays a major role in LNCaP (B) cells. 293T cells were cotransfected with pcDNA3.1-C/EBPy and y 1,25(OH)2D3-mediated growth inhibition of LNCaP cells. either control (scrambled) siRNA or C/EBP siRNA. LNCaP cells were y transfected with either control siRNA or C/EBPy siRNA and treated with either Forced expression of C/EBP inhibits the growth of LNCaP À7 control diluent or 1,25(OH)2D3 (10 mol/L, 18 hours). C/EBPy expression was cells. Further studies explored the biological function of C/EBPy examined by Western blotting. GAPDH was used to control for equal loading and in cancer cells. We transiently transfected the C/EBPy expression siRNA specificity. C, cell proliferation. LNCaP cells were cotransfected with y either control siRNA or C/EBPy siRNA, along with a vector containing the vector (pcDNA3.1-C/EBP ) into LNCaP cells. Colony formation puromycin resistance gene (pMSCVpuro) and selected with puromycin for assays showed that LNCaP cells transfected with pcDNA3.1-C/ 2 days. Equal numbers of transfected cells were plated and 24 hours later were À7 EBPy formed about 90% fewer colonies [mean 38 F 19 (SD) treated with either control diluent or 1,25(OH)2D3 (10 mol/L). After 5 days, cell proliferation was determined by MTT assays. Columns, means of quadruplicate colonies per well] compared with the number of colonies samples; bars, FSD. sic, control siRNA; siy, C/EBPy siRNA; À, 1,25(OH)2D3 developing from control LNCaP cells transfected with the same absent; +, 1,25(OH)2D3 present. www.aacrjournals.org 4765 Cancer Res 2005; 65: (11). June 1, 2005

Discussion In the present study, we used microarray analysis to identify potential 1,25(OH)2D3 target genes under conditions where 1,25(OH)2D3 causes growth inhibition in the AR-expressing prostate cancer cell line, LNCaP. Our study identified C/EBPy as being the gene maximally induced by 1,25(OH)2D3. Knockdown of C/EBPy expression by siRNA significantly reduced the ability of 1,25(OH)2D3 to inhibit growth in LNCaP cells. Furthermore, forced expression of C/EBPy in LNCaP cells inhibited their clonal growth, suggesting that C/EBPy behaves as a tumor suppressive molecule in these prostate cancer cells. The ER-expressing breast cancer cell lines MCF-7 and T47D also exhibited growth inhibition in response to 1,25(OH)2D3. As we have shown for LNCaP cells, 1,25(OH)2D3 induced C/EBPy in these cancer cells as well; and forced expression of C/EBPy in MCF-7 and T47D cells inhibited their clonal growth. Among LNCaP MCF-7 and T47D cells, LNCaP was most sensitive to growth inhibition by 1,25(OH)2D3 followed by MCF-7 and then T47D. The degree of 1,25(OH)2D3-mediated growth inhibitory activity correlated with y the induction levels of C/EBP by 1,25(OH)2D3 and with the growth inhibitory effects of C/EBPy itself when overexpressed in these cell lines. These results further support our hypothesis that C/EBPy is a key mediator of 1,25(OH)2D3 effects in AR-positive prostate and ER-positive breast cancer cells. On the other hand, 1,25(OH)2D3 did not increase the levels of C/EBPy in the AR- and ER-negative DU145 and MDA-MB-231 cells, respectively; likewise, these cells have been shown resistant to growth inhibition after treatment with 1,25(OH)2D3. These data suggest that the induction of C/EBPy mediated by 1,25(OH)2D3 in prostate and breast cancer cells correlates with either a functionally intact AR or ER and the responsiveness of the cells y Figure 4. C/EBP inhibits the clonal proliferation of LNCaP cells. LNCaP cells to 1,25(OH)2D3. were transfected with either pcDNA3.1 empty vector or pcDNA3.1-C/EBPy expression vector. A, colony formation assay. Transfected cells were treated for Our results together with earlier studies suggest that AR- and 2 weeks with G418, fixed, stained, and photographed. Colonies with >40 cells ER-positive cancer cells are generally more sensitive to were counted. Columns, means of three experiments done in triplicate plates; 1,25(OH)2D3. Still, 1,25(OH)2D3 exerts antiproliferative effects on bars, FSD. B, Western blot analysis. Transfected cells were harvested 2 days after transfection and analyzed for C/EBPy expression. C/EBPy expression in AR- and ER-negative cells as well (5, 31). In the present study, À7 LNCaP cells treated with either control diluent or 1,25(OH)2D3 (10 mol/L, we show that 1,25(OH)2D3 slows the growth of the AR-negative 18 hours) is shown for comparison. GAPDH was used to control for equal prostate cancer cells, PC-3, although the level of inhibition is loading. modest compared with that observed in the AR-positive prostate (LNCaP) and ER-positive breast (MCF-7 and T47D) cancer cells. the other hand, 1,25(OH)2D3 failed to increase the level of C/EBPy in ER-negative MDA-MB231 cells (Fig. 5) that are resistant to growth inhibition by 1,25(OH)2D3 (Fig. 1B). Overexpression of C/EBPy inhibits growth of MCF-7 and T47D cells and does not affect the growth of PC-3 cells. Whereas 1,25(OH)2D3 inhibited proliferation of PC-3, MCF-7, and T47D cells (Fig. 1), induction of C/EBPy following 1,25(OH)2D3 treatment was only observed in the ER-positive MCF-7 and T47D breast cancer cells but not in the AR-negative PC-3 prostate cells (Fig. 4). Colony formation assays showed that in MCF-7 cells, forced expression of C/EBPy (pcDNA3.1-C/EBPy) resulted in a significant (68%) growth reduction (486 F 28 colonies per well pcDNA3.1; 154 F 14 colonies per well pcDNA3.1-C/EBPy; Fig. 6). The growth of T47D cells was also inhibited by C/EBPy although to a lesser extent (43%, 98 F 9 colonies per well pcDNA3.1; 56 F 6 colonies per well pcDNA3.1-C/EBPy; Fig. 6). In contrast, C/EBPy y F Figure 5. 1,25(OH)2D3 up-regulates the expression of C/EBP in both AR- and did not significantly inhibit the growth rate of PC-3 cells (47 4 ER-positive cancer cells. The cancer cells were cultured with either control F À7 colonies per well pcDNA3.1; 42 7 colonies per well pcDNA3.1-C/ diluent or 1,25(OH)2D3 (10 mol/L). After 18 hours, RNA was extracted and y levels of C/EBPy transcripts were measured by real-time RT-PCR. The results EBP ; Fig. 6). These results suggest that 1,25(OH)2D3-mediated are expressed in arbitrary units as a ratio of C/EBPy transcripts/h-actin growth suppression in PC-3 cells occurs via a C/EBPy-independent transcripts. Columns, means of triplicate samples; bars, FSD. À, 1,25(OH)2D3 pathway. absent; +, 1,25(OH)2D3 present.

Cancer Res 2005; 65: (11). June 1, 2005 4766 www.aacrjournals.org

Figure 6. C/EBPy inhibits growth of MCF-7 and T47D cells and does not affect the growth of PC-3 cells. Colony formation assays (A) and Western blot analysis (B) were done as described in Fig. 3 legend.

Interestingly, levels of C/EBPy are not induced in PC-3 cells substrates for degradation by the ubiquitin-proteasome pathway following 1,25(OH)2D3 treatment and forced expression of (38). Up-regulation of SUMO 1 in LNCaP cells mediated by vitamin C/EBPy does not lead to their growth inhibition. These results D could protect CDKIs from degradation by the ubiquitin- suggest that the signaling pathways of 1,25(OH)2D3 in AR- and proteasome pathway stabilizing these proteins, resulting in ER-negative cells are, at least partly, different from the ones retardation of their proliferation. occurring in AR- and ER-positive cells. Previous studies done in The 1,25(OH)2D3 down-regulated by 60% the expression of other human cells including squamous carcinoma cells (32) and neuronal cell adhesion molecule (NRCAM) in LNCaP cells breast cancer cells (31) have also suggested that the molecular compared with control cells. Neuroblastoma cells express NRCAM mechanisms underlying 1,25(OH)2D3 growth inhibitory effects are and their levels of this protein paralleled their growth rate (39). cell type specific. However, the relationship between prostate cancer and NRCAM This study also found that 1,25(OH)2D3 increased the level of has not been studied. InBa in LNCaP cells. InBa forms a binding complex with NF-nBin Recently, other investigators did microarray analysis to identify the cytoplasm, preventing the latter from entering the nucleus and target genes of 1,25(OH) D in adenocarcinoma cells from the behaving as a progrowth transcription factor by activating target 2 3 prostate including LNCaP cells (40, 41). They used arrays carrying genes such as Bcl-2 and cyclin D (33). The level of InBa protein is 20,000 genes and identified IGFBP-3 as an up-regulated gene. regulated by the ubiquitin-proteasome pathway. Another gene Expression of IGFBP-3 in LNCaP cells under the conditions that we whose expression was up-regulated by 1,25(OH)2D3 is the small ubiquitin-related modifier 1 (SUMO 1; Table 2). This protein used, was extremely low as measured by real-time PCR (data not stabilizes InBa and prevents 26S proteasome–mediated degrada- shown); therefore, it was not identified as ‘‘present’’ in our tion of InBa (34). Therefore, overexpression of SUMO 1 also inhibits microarray analysis. We note that with the exception of one gene the activity of NF-nB via up-regulation of InBa (34). A gene that (FK506-binding protein 5), the gene expression profiles generated by the two studies do not overlap. It is likely that the dissimilarities was down-regulated by 1,25(OH)2D3 was the paternally expressed gene 3 (PEG 3), which is also implicated in the regulation of NF-nB are the result of different experimental conditions as well as activity; PEG 3 activates NF-nB by aiding the dissociation of the different criteria used to analyze the data and that each study InBa/NF-nB complex (35). Thus, several genes revolving around complements the other. NF-nB were modulated by 1,25(OH)2D3 in LNCaP cells. Further Taken together, we have found that 1,25(OH)2D3 stimulated the y studies are warranted to explore the effects of 1,25(OH)2D3 on expression of C/EBP in LNCaP cells, which contributed to their NF-nB activity in cancer cells. growth arrest mediated by 1,25(OH)2D3. Further studies are The 1,25(OH)2D3 up-regulates the levels of CDKIs, including required to clarify the molecular mechanisms by which ligand- p21waf1 and p27kip1 in cancer cells (36, 37). These CDKIs are also activated VDR enhances the expression of C/EBPy. www.aacrjournals.org 4767 Cancer Res 2005; 65: (11). June 1, 2005

Acknowledgments Comprehensive Cancer Center and Molecular Biology Institute membership (H.P. Koeffler), and Mark Goodson Chair of Oncology Research at Cedars-Sinai Medical Received 11/18/2003; revised 2/9/2005; accepted 3/24/2005. Center (H.P. Koeffler). Grant support: NIH, AT00151, the University of California at Los Angeles Center The costs of publication of this article were defrayed in part by the payment of page for Dietary Supplements Research: Botanicals, Leo Pharmaceuticals, the Parker charges. This article must therefore be hereby marked advertisement in accordance Hughes Fund, Aaron Eschman Fund, University of California-Los Angeles Jonsson with 18 U.S.C. Section 1734 solely to indicate this fact.

References 15. Cao Z, Umek RM, McKnight SL. Regulated expression 28. Hofmann WK, de Vos S, Tsukasaki K, et al. Altered of three C/EBP isoforms during adipose conversion of apoptosis pathways in mantle cell lymphoma detected 1. Minghetti PP, Norman AW. 1,25(OH)2-vitamin D3 3T3-L1 cells. Genes Dev 1991;5:1538–52. by oligonucleotide microarray. Blood 2001;98:787–94. receptors: gene regulation and genetic circuitry. FASEB 16. Park DJ, Chumakov AM, Vuong PT, et al. CCAAT/ 29. Ikezoe T, Miller CW, Kawano S, et al. Mutational J 1988;2:3043–53. enhancer binding protein q is a potential retinoid target analysis of the peroxisome proliferator-activated recep- 2. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey gene in acute promyelocytic leukemia treatment. J Clin tor g gene in human malignancies. Cancer Res 2001;61: TA, Feldman D. Antiproliferative effects of 1,25- Invest 1999;103:1399–408. 5307–10. dihydroxyvitamin D3 on primary cultures of human 17. Yamanaka R, Barlow C, Lekstrom-Himes J, et al. 30. Esquenet M, Swinnen JV, Heyns W, Verhoeven G. prostatic cells. Cancer Res 1994;54:805–10. Impaired granulopoiesis, myelodysplasia, and early Control of LNCaP proliferation and differentiation: 3. Skowronski RJ, Peehl DM, Feldman D. Actions of lethality in CCAAT/enhancer binding protein epsilon- actions and interactions of androgens, 1a,25-dihydrox- vitamin D3, analogs on human prostate cancer cell deficient mice. Proc Natl Acad Sci U S A 1997;94: ycholecalciferol, all-trans retinoic acid, 9-cis retinoic lines: comparison with 1,25-dihydroxyvitamin D3. En- 13187–92. acid, and phenylacetate. Prostate 1996;28:182–94. docrinology 1995;136:20–6. 18. Gery S, Gombart AF, Fung YK, Koeffler HP. C/EBPq 31. Swami S, Raghavachari N, Muller UR, Bao YP, 4. Danielpour D, Kadomatsu K, Anzano MA, Smith JM, interacts with Retinoblastoma and E2F1 during gran- Feldman D. Vitamin D growth inhibition of breast Sporn MB. Development and characterization of non- ulopoiesis. Blood 2004;103:828–35. cancer cells: gene expression patterns assessed by tumorigenic and tumorigenic epithelial cell lines from 19. Flodby P, Barlow C, Kylefjord H, Ahrlund-Richter L, cDNA microarray. Breast Cancer Res Treat 2003;80: rat dorsal-lateral prostate. Cancer Res 1994;54:3413–21. Xanthopoulos KG. Increased hepatic cell proliferation 49–62. 5. Campbell MJ, Reddy GS, Koeffler HP. Vitamin D3 and lung abnormalities in mice deficient in CCAAT/ 32. Akutsu N, Lin R, Bastien Y, et al. Regulation of gene analogs and their 24-oxo metabolites equally inhibit enhancer binding protein a. J Biol Chem 1996;271: expression by 1a,25-dihydroxyvitamin D3 and its clonal proliferation of a variety of cancer cells but 24753–60. analog EB1089 under growth-inhibitory conditions in have differing molecular effects. J Cell Biochem 1997;66: 20. Pabst T, Mueller BU, Zhang P, et al. Dominant- squamous carcinoma. Cell Mol Endocrinol 2001;7: 413–25. negative mutations of CEBPA, encoding CCAAT/ 1127–39. 6. Frampton RJ, Omond SA, Eisman JA. Inhibition of enhancer binding protein-a (C/EBPa), in acute myeloid 33. Karin M, Ben-Neriah Y. Phosphorylation meets human cancer cell growth by 1,25-dihydroxyvitamin D3 leukemia. Nat Genet 2001;27:263–70. ubiquitination: the control of NF-nB activity. Annu metabolites. Cancer Res 1983;43:4443–7. 21. Gombart AF, Hofmann WK, Kawano S, et al. Rev Immunol 2000;18:621–63. 7. Bower M, Colston KW, Stein RC, et al. Topical Mutations in the gene encoding the transcription factor 34. Desterro JM, Rodriguez MS, Hay RT. SUMO-1 calcipotriol treatment in advanced breast cancer. CCAAT/enhancer binding protein a in myelodysplastic modification of InBa inhibits NF-nB activation. Mol Lancet 1991;337:701–2. syndromes and acute myeloid leukemias. Blood 2002; Cell 1998;2:233–9. 8. Zhou JY, Norman AW, Chen DL, Sun GW, Uskokovic M, 99:1332–40. 35. Relaix F, Wei XJ, Wu X, Sassoon DA. Peg3/Pw1 is an Koeffler HP. 1,25-Dihydroxy-16-ene-23-yne-vitamin D3 22. Halmos B, Huettner CS, Kocher O, Ferenczi K, imprinted gene involved in the TNF-NFnB signal prolongs survival time of leukemic mice. Proc Natl Acad Karp DD, Tenen DG. Down-regulation and antiprolifer- transduction pathway. Nat Genet 1998;18:287–91. Sci U S A 1990;87:3929–32. ative role of C/EBPa in lung cancer. Cancer Res 2002;62: 36. Campbell MJ, Elstner E, Holden S, Uskokovic M, 9. Norman AW, Zhou JY, Henry HL, Uskokovic MR, 528–34. Koeffler HP. Inhibition of proliferation of prostate Koeffler HP. Structure-function studies on analogues of 23. Hutt JA, DeWille JW. Oncostatin M induces growth cancer cells by a 19-nor-hexafluoride vitamin D3 1a,25-dihydroxyvitamin D3: differential effects on leu- arrest of mammary epithelium via a CCAAT/enhancer- analogue involves the induction of p21waf1, p27kip1 kemic cell growth, differentiation, and intestinal binding protein y-dependent pathway. Mol Cancer Ther and E-cadherin. J Mol Endocrinol 1997;19:15–27. calcium absorption. Cancer Res 1990;50:6857–64. 2002;1:601–10. 37. Wang QM, Jones JB, Studzinski GP. Cyclin-dependent 10. Zhuang SH, Burnstein KL. Antiproliferative effect of 24. Hutt JA, O’Rourke JP, DeWille J. Signal transducer and kinase inhibitor p27 as a mediator of the G1-S phase 1a,25-dihydroxyvitamin D3 in human prostate cancer activator of transcription 3 activates CCAAT enhancer- block induced by 1,25-dihydroxyvitamin D3 in HL60 cell line LNCaP involves reduction of cyclin-dependent binding protein y gene transcription in G0 growth- cells. Cancer Res 1996;56:264–7. kinase 2 activity and persistent G1 accumulation. arrested mouse mammary epithelial cells and in 38. Pagano M. Cell cycle regulation by the ubiquitin Endocrinology 1998;139:1197–207. involuting mouse mammary gland. J Biol Chem 2002; pathway. FASEB J 1997;11:1067–75. 11. Blutt SE, McDonnell TJ, Polek TC, Weigel NL. 275:29123–31. 39. Hildebrandt H, Becker C, Gluer S, Rosner H, Gerardy- Calcitriol-induced apoptosis in LNCaP cells is blocked 25. O’Rourke J, Yuan R, DeWille J. CCAAT/enhancer- Schahn R, Rahmann H. Polysialic acid on the neural cell by overexpression of Bcl-2. Endocrinology 2000;141: binding protein-y (C/EBP-y) is induced in growth- adhesion molecule correlates with expression of poly- 10–7. arrested mouse mammary epithelial cells. J Biol Chem sialyltransferases and promotes neuroblastoma cell 12. Lekstrom-Himes J, Xanthopoulos KG. Biological role 1997;272:6291–6. growth. Cancer Res 1998;58:779–84. of the CCAAT/enhancer-binding protein family of 26. Tanosaki S, Gombart AF, Koeffler HP. C/EBP y causes 40. Krishnan AV, Peehl DM, Feldman D. Inhibition of transcription factors. J Biol Chem 1998;273:28545–8. G0/G1 cell cycle arrest with induction of expression of prostate cancer growth by vitamin D: regulation of 13. Hurst HC. Transcription factors. 1. bZIP proteins. p27 and activation of FKHR in CML blast cell. Abstract. target gene expression. J Cell Biochem 2003;88:363–71. Protein Profile 1994;1:123–68. Blood 2002;734a. 41. Krishnan AV, Shinghal R, Raghavachari N, Brooks JD, 14. Lamb P, McKnight SL. Diversity and specificity in 27. de Vos S, Hofmann WK, Grogan TM, et al. Gene Peehl DM, Feldman D. Analysis of vitamin D-regulated transcriptional regulation: the benefits of heterotypic expression profile of serial samples of transformed gene expression in LNCaP human prostate cancer cells dimerization. Trends Biochem Sci 1991;16:417–22. B-cell lymphomas. Lab Invest 2003;83:271–85. using cDNA microarrays. Prostate 2004;59:243–51.

Cancer Res 2005; 65: (11). June 1, 2005 4768 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 2005 American Association for Cancer Research. CCAAT/Enhancer-Binding Protein δ: A Molecular Target of 1,25-Dihydroxyvitamin D 3 in Androgen-Responsive Prostate Cancer LNCaP Cells

Takayuki Ikezoe, Sigal Gery, Dong Yin, et al.

Cancer Res 2005;65:4762-4768.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/65/11/4762

Cited articles This article cites 39 articles, 19 of which you can access for free at: http://cancerres.aacrjournals.org/content/65/11/4762.full#ref-list-1

Citing articles This article has been cited by 15 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/65/11/4762.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/65/11/4762. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.