|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
Molecular Mechanism of Vitamin D Action and its Implications in Ovarian Cancer Prevention and Therapy by Feng Jiang A dissertation submitted in partial fufillment of the requirements for the degree of Doctor of Philosophy Department of Pathol ogy College of Medicine University of South Florida Major Professor: Wenlong Bai, Ph.D. Santo V. Nicosia, M.D. Richard Jove, Ph.D. John C.M. Tsibris, Ph.D. Jin Q. Cheng, M.D., Ph.D. Douglas W. Cress, Ph.D. Date of Approval: May, 2004 Keywords: Vita min D receptor, GADD45, telomerase, G2/M arrest, apoptosis Copyright 2004, Feng Jiang
ACKNOWLEDGMENTS Thanks to Dr. Wenlong Bai, my advisor, for his considerate mentorship, insight and support. Thanks to committee members: Drs. Nicosia, Jove, Tsibris, Cheng and Cress for their invaluable guidance and encouragement. Thanks to all the faculty, staff and students in Department of Pathology, office of graduate affairs for helping me along the way. Thanks to my mom, dad, grandmother and all my family, particularly my wonderful husband, Zhaohui Fan who have been especially supportive and u nderstanding. Thanks to my friends, special thanks to my colleagues for sharing the hard study life with joy and friendship.
i Table of Contents List of Figures v Abstract viii INTRODUCTION 1 Nuclear receptor superfamily 1 Reg ulation of vitamin D receptor 12 Effect of Vitamin D in cancer 23 GADD45 and OCa 3 0 Telomerase and OCa 31 STUDY OBJECTIVES 34 RESULTS 35 Effect o f 1,25VD in OCa cells 35 1. 1,25VD suppresses OCa cell growth and induces cell cycle arrest at G1/S and G2/M 35 2. 1,25VD induces apoptosis in OCa c e lls 38 3. 1,25VD treated OVCAR3 cells recover slowly after 1,25VD withdrawal 38 G2/M arrest by 1,25VD in ovarian cancer cells mediated through the induction of GADD4 5 via an exonic enhancer 41
ii 1. GADD45 is a prim ary and immediate early response gene for 1,25VD in OCa cells 41 2. A novel VDRE in the 3 un translated region of GADD45 mRNA mediates the transcriptional up reg ulation of GADD45 by 1,25VD 42 3. Upregulation of GADD45 protein is required for the hormone induced cell cycle arrest at the G2/M but not the G1/S checkpoint 56 4. GADD45 mediates 1,25VD induced decrease in Cdc2 kinase activity in OCa cells 64 1,25VD induced apoptosis is mediated by destabilizati on of hTERT mRNA and decrease in telomerase activity 65 1. 1,25VD down regulates telomerase activity in OCa cells 65 2. hTERT is a VD target ed gene in OCa cells 6 9 3. The stability of hTERT mRNA is decreased by 1,25VD 73 4. Telomerase stably transfected cells have increased telomerase activity and prolonged telomere length 75 5. Overexpression of telomerase partially blocks 1,25VD induced apoptosis and increases the ability to recover after 1,25VD withdrawal 79 EB1089 is more potent than 1,25VD in suppressing OCa cell growth in vitro and in nude mice 84 VDR and RXR are expressed in precursors for epithelial OCa and normal ov arian surface epithelial ce lls are responsive to 1,25VD 90
iii DISCUSSION 9 3 Upregulation of GADD45 by 1,25VD is mediated through a novel exonic Enhancer 95 GADD45 protein upregulation by 1,25VD is required for the hormone induced cell cycle ar rest at the G2/M but not the G1/S checkpoint 96 Down regulation of telomerase by 1,25VD is mediated through destabilization of hTERT mRNA 98 Over expression of hTERT allows the recovery of OCa cells from 1,25VD treatment and partially relieves 1,25VD indu ced apoptosis 101 1,25VD analogue EB1089 has potent anti tumor activity in vitro and in vivo and has strong implications to can cer treatment and prevention 103 SUMMARY AND PERSPECTIVES FOR FUTURE STUDIES 106 MATERIALS AND MET HODS 113 Materials 113 Colorimetric methylthiazole tetrazolium (MTT) assays and statistical analysis 115 Cell cycle and apoptosi s analysis by flow cytometry 115 Northern blot analysis 11 5 Gel mobility shift assay (EMSA) 116 Construction o f luciferase reporter plasmids, deletion a nd site directed mutagenesis 116 Transcriptional assays 117 Chromatin immunoprecipitation (ChIP) assays 118
iv Immunoblotting analysis 119 Establishment of sta ble clones from OVCAR3 cells 119 In vitro immunocomplex assays 120 Telomerase activity assay 12 0 RT PCR analysis 121 Primers, probes fo r Real time PCR 121 Real time PCR 122 Telomere length assay 122 Immuno histochemical analysis 123 Nude mouse tumor studies 123 REFERENCES 125 About the Author End Page
v List of Figures Figure 1. Structural and functional organization of nuclear receptors 3 Figure 2. Metabolism and biological response of Vitamin D 13 Figure 3. Molecular structure of the VDR 16 Figure 4. Ovarian cancer mortality rates in the US (1970 1994) 24 Figure 5. 1,25VD inhibits OCa cell growth and induces cell cycle arrest at G1/S and G2/M checkpoints 36 Figure 6. 1,25VD induces apoptosis in OCa cells 39 Figure 7. OCa cells recover slowly from 1,25VD treatment 40 Fig ure 8. 1,25VD increases transcr iptio n of GADD45 mRNA in OCa cells 43 Figure 9. Multiple putative VDREs are present in GADD45 genome, which interact with VDR/RXR in vitro 46 Figure 10. 1,25VD induces GADD45 reporter activity through endog enous receptors in OCa cells 49 Figure 11. Induction of GADD45 reporter activit y by 1,25VD is VDR dependent 52 Figure 12. The VDRE in the fourth exon of GADD45 genome is the functional VDRE that mediates the transcriptional induction of G ADD45 by 1,25VD in OCa cells 54 Figure 13. Anti sense GADD45 blocks 1,25VD induced cell cycle arrest at G2/M, but not G1/S che ckpoint 58
vi Figure 14. 1,25VD induces G2/M arrest in wild type but not in GADD45 null MEFs 61 Figure 15. 1,25VD decreases cdc2 kinase activity in OVCAR3 tran sfected with control vector but not in cells stably transfected with t he anti sense cDNA of GADD45 63 Figure 16. 1,25VD down regulates telomerase activity in OCa cells 66 Figure 17. 1,25VD down regulates hTERT mRNA expression in OCa cells 67 Figure 1 8. Putative hTERT VDRE specifically binds to VDR/RXR heterodimer 70 Figure 19. The putative VDRE is not a functional VDRE in OVCAR3 cells 71 Figure 20. The stability of hTERT mRNA is decreased by 1,25VD 74 Figure 21. Telomerase stably transfected c ells have increased telomerase activity and prolonged telomere length 76 Figure 22. Overexpression of telomerase blocks 1,25VD induced down regulation of telomerase activity 77 Figure 23. Telomerase OVCAR3 cells are less responsive to 1,25VD induced gr owth inhibition 8 0 Figure 24. Telomerase OVCAR3 cells recover quickly from 1,25VD treatment 8 1 Figure 25. Overexpression of telomerase partially blocks 1,25VD induced apoptosis 82 Figure 26. EB1089 is more potent than 1,25VD in suppressing OCa cel l growth 85 Figure 27. EB1089 is more effective than 1,25VD in inducing GADD45 reporter activity 86 Figure 28. EB1089 inhibits OCa xenograft tumor growth 88
vii Figure 29. Normal ovarian epithelial cells and benign adenomas express VDR and RXR 91 Figu re 30. 1,25VD suppresses human primary ovarian cell growth 92 Figure 31. Model for the integrated cellular pathways for 1,25VD action in OCa cells 94
viii Molecular Mechanism of Vitamin D Action and its Implications in Ovarian Cancer Preventi on and Therapy Feng Jiang ABSTRACT 1,25 dihydroxyvitamin D3 (1,25VD), the active form of vitamin D (VD), suppresses the growth of numer o us human cancer cell lines by inhibiting cell cycle progression and inducing cell death. Genes that mediate e ach of these activities remain largely unidentified and there are no preclinical data for 1,25VD analogues in ovarian cancer (OCa). We hypothesize that 1,25VD and its analogues inhibit the development of OCa. In this study, we demonstrated, (a) 1,25VD caus es cell cycle arrest at the G1/S and G2/M transition and induces apoptosis in OCa cells. (b) We also found that gadd45 is one of primary target genes for 1,25VD mediated G2/M arrest. A direct repeat 3 (DR3) vitamin D response element (VDRE) is identified i n the fourth exon of gadd45. This exonic VDRE forms a complex with the vitamin D receptor (VDR)/retinoid X receptor (RXR) heterodimer in vitro and mediates the induction of reporter activity by 1,25VD in vivo VDR is recruited in a ligand dependent manner to the exonic enhancer but not to the gadd45 promoter regions. In OCa cells expressing GADD45 anti sense cDNA or GADD45 null mouse embryo fibroblasts, 1,25VD fails to induce G2/M arrest, suggesting that G2/M arrest induced by 1,25VD is mediated through GAD D45. Further study showed that GADD45 mediates the effect of 1,25VD by decreasing cdc2 kinase activity. (c) hTERT, the catalytic subunit of telomerase, is identified as a primary target for
ix 1,25VD. 1,25VD decreases telomerase activity and hTERT mRNA expres sion. The down regulation of hTERT mRNA is due to decreased mRNA stability by 1,25VD, rather than decreased transcription of hTERT through VDRE. Clones stably transfected with hTERT showed higher telomerase activity and longer telomere length than parental cells. Moreover, hTERT clones resist 1,25VD induced apoptosis and growth inhibition. In contrast to parental cells which do not recover from prolonged treatment with 1,25VD, hTERT clones re grew rapidly after 1,25VD withdrawal. (d) We demonstrated that th e 1,25VD analogue EB1089 inhibits OCa cells in vitro and OCa xenograft in vivo without inducing hypercalcemia. We also demonstrated precursors for epithelial OCa express VDR and human primary ovarian surface epithelial cells respond to 1,25VD. Taken togeth er, these results strongly suggest that 1,25VD analogues may be effective in the chemoprevention and chemotherapy of OCa.
1 INTRODUCTION Nuclear receptor superfamily The nuclear receptor superfamily consists of two classes of genes, nuclear hormone receptors and orphan receptors, which represent one of the most abundant classes of transcriptional regulators in animals (Robinson Rechavi, 2003). It affects a wide variety of functions, including homeostasis, reproductive development, fatty acid metabolism and detoxification of foreign substances. The cognate ligands have been identified for nuclear hormone receptors, but f or orphan receptors, the physiologically relevant ligand(s) remain unkown. The discovery of ligands for orphan receptors and additional receptors has become a very active research field. It has now been almost 20 years since the isolation of cDNAs encoding the glucocorticoid, estrogen and thyroid receptors (Hollenberg, 1985; Green, 1986; Weinberger, 1986), which were the first several receptors identified. Novel signaling pathways controlled by nuclear receptors have opened up new aspects of nuclear recepto rs action with respect to homeostasis, reproduction, development and metabolism in various organisms. Nuclear receptors form a superfamily of phylogenetically related proteins, with 21 genes in the complete genome of the fly Drosophila melanogaster 48 in humans and, unexpectedly, more than 270 genes in the nematode worm Caenorhabditis elegans (nuclear receptor nomenclature committee,1999).
2 Nuclear hormone receptors are a class of molecules that function as both signal transducers and transcrip tion factors. The hormonal ligands for these receptors are hydrophobic molecules such as the sex steroids (androgens, estrogens, progesterone), glucocorticoids, thyroid hormones, mineralocorticoids, 1,25VD and retinoids (all trans retinoic acid and 9 cis retinoic acid) etc. Other ligands, such as ecdysone, oxysterols, bile acids, leukotrienes and prostaglandins have been recently identified and characterized to have receptors that are structurally related to steroid receptor hormones and appear to act thro ugh similar mechanisms as the steroid receptors. Unlike the water soluble peptide hormones and growth factors, which bind to cell surface receptors, the fat soluble steroid hormones can pass through the lipid bilayer of the cell membrane and interact with their cognate receptors in the cell. These lipophilic hormones are potent regulators of development, cell differentiation, and organ physiology. Classification of nuclear receptors The nuclear receptor superfamily can be broadly divided into four classes (class I, II, III and IV), as proposed by Manglesdorf et al. (1995) based on their dimerization and DNA binding properties (Fig. 1). Class I receptors include the known classical steroid hormone receptors, which function as ligand induced homodimers and b ind to palindromic response element. Class I receptors include receptors for steroids such as estrogens (ER), progesterones (PR), glucocorticoids (GR), mineralocorticoids (MR), and androgens (AR), These molecules have widespread effects on the development and control of the reproductive system (Beato, 1995; Cole, 1995; Couse, 1998 and 1999).
3 Fig. 1. Structural and functional organization of nuclear receptors. C A/B C D E F N ER estrogen PR progesterone AR androgen GR Glucocorticoid MR mineralocorticoid T 3 R thyroid hormone RAR all trans RA VDR 1,25VD PPAR fatty acids EcR ecdysone FXR bile acids CAR androstane LXR oxysterol ERE ER ER ER ER VDRE RXR VDR Class I : Steroid receptors Class I I : RXR heterodimers RXRE RXR RXR RXR RXR RE NGFI B Class I I I : Dimeric Orphan Receptors Class I V: Monomeric Orphan Receptors RXR (9 cis RA) COUP HNF 4 TR2 TLX GCNF NGFI B ELP/SF 1 Rev erb ERR ROR C A/B C D E F N ER estrogen PR progesterone AR androgen GR Glucocorticoid MR mineralocorticoid T 3 R thyroid hormone RAR all trans RA VDR 1,25VD PPAR fatty acids EcR ecdysone FXR bile acids CAR androstane LXR oxysterol ERE ER ER ER ER VDRE RXR VDR Class I : Steroid receptors Class I I : RXR heterodimers RXRE RXR RXR RXR RXR RE NGFI B Class I I I : Dimeric Orphan Receptors Class I V: Monomeric Orphan Receptors RXR (9 cis RA) COUP HNF 4 TR2 TLX GCNF NGFI B ELP/SF 1 Rev erb ERR ROR
4 Fig. 1. Structural and functional organization of nuclear receptors. The six domains (A F) of nuclear receptors comprise regions of conserved function and sequences. All of nuclear receptors contain a central DNA binding domain (DBD), region C, which is the most highly conserved domain and includes two zinc finger modules. A ligand binding domai n (LBD, region E) is located in the C terminal half of the receptor, which contains activation function 2 (AF 2). Between the DBD and LBD is a hinge region (region D) that has variable length. In the amino terminus, the variable N terminal region (A/B) con tains transcriptional activation function 1 (AF 1). Some receptors also contain the F trail, the function of which is poorly understood.
5 Steroid receptors are believed to be synthesized in cytoplasms from single messenger RNAs (mRNAs). In contrast to othe r nuclear receptors, unliganded steroid receptors are associated with a large multi protein complex of chaperones, including Hsp90 and the immunophilin Hsp56, which maintain the receptors in an inactive but conformation. When complexed with Hsp90, through amino acid residues at the receptors C terminus, steroid receptors are unable to bind DNA (Pratt, 1993). The ER has two subtypes: ER a and ER b Considerable divergence between ER a and ER b is apparent in the N terminus (18 % homology). ER a is the predominan t subtype expressed in the major female organs such as ovary, uterus, vagina, mammary gland and certain areas of central nervous system especially in the hypothalamus. ER b exhibits a more limited expression patterns and is primarily detected in the ovary, prostate, testis, spleen, lung, hypothalamus and thymus (Couse, 1999; Kuiper, 1997). The differential distribution of each receptor in a certain tissues may reflect their distinct receptor functions. The PR is unique among steroid receptors in t hat it is composed of two naturally occurring hormone binding forms of proteins, A and B. The A isoform is an N terminally truncated version of the full length B isoform. The molecular weights of the human receptors as deduced from cDNA sequence are 98K an d 86K, respectively. The A and B forms of the progesterone receptor differ in their ability to activate target genes and are regulated differently in various types of cells (Conneely, 2002). The glucocorticoid receptor was the first transcription factor t o be isolated and studied in detail (Muller, 1991). There are 3 isoforms of GR, a b and g GR a modulates the expression of glucocorticoid responsive genes by binding to a specific glucocorticoid
6 response element (GRE) DNA sequence. In contrast, GR b and GR g inhibit the effects of hormone activated GR a (Bamberger, 1995; Ray, 1996; Rivers, 1999), suggesting that GR b and GR g may be a physiologically and pathophysiologically relevant endogenous inhibitor of glucocorticoid action and may participate in defining the sensitivity of tissues to glucocorticoids. The MR is a 107 kD protein. MR was cloned by Arriza et al. (1987) and plays important roles in regulation of hydroelectrolytic homeostasis MR activation by aldosterone raises renal salt reabsorption by incre asing the activity of the epithelial sodium channel in the distal nephron. Although the distal nephron is recognized as the major site of action of mineralocorticoid s, expression of MR in hippocampus, heart, and endothelium has suggested extrarenal activit y (Le Menuet, 2000). The AR gene is more than 90 kb long and codes for a 76 kD protein. AR mediates the biological action of androgens, principally testosterone and 5 alpha dihydrotestosterone, that play critical roles in the development and growth of the male reproductive and nonreproductive systems. AR is found in a variety of tissues and changes throughout development, aging and malignant transformation (Chang, 1988; Lubahn, 1988). Class II receptors heterodimerize with the receptor for 9 ci s retinoic acid (RXR) and characteristically bind to direct repeats. Exclusive of the classical steroid hormones, this group includes almost all other known ligand dependent receptors. The list of RXR dimer partners includes receptors for thyroid hormone ( TR), vitamin D (VDR), retinoic acid (RAR), 9 cis retinoic acid (RXR), and ecdysone (EcR). These receptors are necessary for normal development, differentiation or organ morphogenesis (Kastner,
7 1995; Sambrook,1994). Moreover, its heterodimeric partner, RXRs has been thought to exert multiple functions via their partner. The detection of an abnormal phenotype may raise the question as to which of these pathways has been affected. The class II receptors also include former orphan receptors for which endogeno us ligands have been recently discovered, such as receptors for polyunsaturated long chain fatty acids and their metabolites (peroxisome proliferator activated receptor s, PPARs), oxysterols (LXRs), bile acids (FXR), xenobiotics (xenobiotic receptor /pregnan e X receptor SXR/PXR and constitutive androstane receptor, CAR). These receptors have been adopted when they were shown to bind a physiological ligand (Chawla, 2001, Giguere, 1999, Blumberg B, 1998) and are now in the class II receptor category due to t heir ability to form heterodimers with RXR. Many of these adopted orphan receptors function as lipid sensors that respond to cellular lipid levels and elicit gene expression changes to ultimately protect cells from lipid overload. Three peroxisome prolif erator activated receptors (PPARs) isoforms have been characterized: PPAR a b / d and g PPARs play a critical role in lipid and glucose homeostasis. Lately they have been shown to interfere with different steps of the inflammatory response by modulating the expression of chemokines, chemokine receptors and adhesion molecules in endothelial cells, smooth muscle cells, monocytes/macrophages and T cells (Blanquart, 2003). LXR are commonly known as cholesterol sensors that respond to elevated cholesterol concen trations. Oxysterols are natural ligands for LXR. LXR is abundantly expressed in the liver and other tissues that are associated with lipid metabolism (Peet, 1998). Of particular interests are two xenobiotic receptors, SXR/PXR (human steroid
8 xenobiotic rec eptor/pregnane X receptor) and CAR (constitutive androstane receptor), which regulate the metabolic cascade of toxic endogenous lipids to protect the body from foreign chemicals. Whereas CAR mediates the response to phenobarbital like inducers, SXR/PXR res pond to many prescription drugs, steroids and toxic bile acids. These receptors seems to increase the clearance of foreign chemicals and provide an important feedforward loop for the xenosensors to initiate another round of signaling (Chawla, 2001; Xie, 20 01). Collectively, unlike the classical nuclear hormone receptors of which ligands were well known before the receptors were cloned and the role of which was well established in reproduction and homeostasis via feedback regulation, aformentioned lipid se nsors establish a new role for nuclear receptors in activating feedforward metabolic cascades that maintain lipid homeostasis by governing the transcription of genes involved in lipid metabolism, storage, transport and elimination. Most of the orphan rece ptors fall into the class III and IV categories. They represent a diverse and ancient component of the nuclear receptor superfamily, being found in nearly all animal species examined (Blumberg, 1998). Class III receptors, such as RXR, bind to direct repeat s as homodimers. RXR is involved in the transduction of retinoid signaling pathway and it is activated in vitro by the vitamin A metabolite 9 cis retinoic acid but little is known of the natural activators of RXR. Although RXR forms ( a b g ) can function as homodimers, they also serve as the dimerization partner of other nuclear receptors including RAR, TR, VDR, PPARs, LXR, FXR. Thus, as a heterodimerization partner, RXR is involved in the regulation of multiple cellular pathways. Heterodimers PPAR RXR, LX R RXR, and FXR RXR are permissive because
9 both heterodimeric partners can bind their cognate ligands and induce transcription. Ligand binding by RXR has been demonstrated in the context of RAR and TR heterodimers, suggesting the possibility of permissivene ss for RXR RAR and TR RXR heterodimers. Recently, allosteric modification of RXR by liganded VDR has been demonstrated. RXR acquires the holoreceptor conformation in the absence of its cognate ligand and gains the ability to recruit coactivators. These res ults provide evidence for RXR as a functionally active, nonsilent partner in 1,25VD mediated RXR VDR dependent gene expression (Bettoun, 2003). Class IV receptors typically bind to extended core sites as monomers, such as RevTR a (EAR1) and NGFI b which rec ognize the AGGTCA sequence without repetition. Although structurally related to the known receptors, no physiological ligands are known for these orphan receptors that are more numerous than receptors with known ligands (Blumberg, 1998). Taken together wi th the future identification of their definite ligand, orphan receptors will also provide an opportunity to identify potentially new functions in physiology. Domain structure of nuclear receptors Despite their evolutionary and functional differences, mem bers of the nuclear receptor superfamily generally share a similar structural organization. Typical nuclear receptor consists of multiple functionally distinct domains, including domains involved in DNA or ligand binding, dimerization, and transcriptional activation. Once they have bound their ligands, dimerized and achieved high affinity association with specific
10 responsive elements in DNA, their transactivation domains function together at the molecular level, resulting in gene activation (Issa, 1998; Kum ar, 1999). The N terminus of the nuclear receptor, A/B domain, has ligand independent transactivation activity, termed activation function 1 (AF 1). The sequence and length of the A/B domain are highly variable among receptors (Weinberger, 198 5). In addition, this region is the most frequent site of alternative splicing and secondary start sites, also contains a variety of kinase recognition sequences. It is thought that A/B domain may be responsible for the receptor species and cell type s pecific effects, as well as promoter context dependent properties of nuclear receptor transactivation. Nuclear receptors are characterized by a central DNA binding domain (DBD), which targets the receptor to specific DNA sequences known as horm one response elements. DBD, which is the most conserved region among various members of the superfamily, sets the nuclear receptors apart from other DNA binding proteins. There are nine cysteine residues within the DBD that are strictly conserved throughou t the superfamily of receptor proteins. The first eight cysteines in the N terminus coordinate two zinc atoms to form the so called zinc finger DNA binding motifs that are responsible for high affinity interaction with specific DNA sequences in the genomic region of hormone target genes. The first zinc finger contains the P (proximal) box region, an alpha helix that is responsible for high affinity recognition of the core half site of the response element. Located within the second zinc finger, D (distal) box, an alpha helix, determines the spacing between half sites and mediates receptor dimerization. The second zinc finger contains a six residue region, referred to as the T box, which has been suggested to form a dimerization interface for the interactio n with the RXR DBD.
11 Adjacent to the DBD is the D or hinge domain. This region has an poorly defined function. The D domain appears to allow for conformational changes in the protein structure following ligand binding. Also, this region may contain nuclear locolization signals and protein protein interaction sites. The C terminal half of the receptor encompasses the ligand binding domain (LBD), which possesses the essential property of hormone recognition and ensures both specificity and selectivity of the p hysiologic response. Although this domain is highly variable among family members, all nuclear receptors share a common structure of 10 to 13 alpha helixes organized around a hydrophobic binding pocket. A ligand dependent activation function (AF2) is loca ted at the extreme C terminus (helix 12). LBD also contains nuclear localization signals, dimerization motifs interaction surfaces for heat shock proteins, coregulators and for other transcription factors. The AF 2 domain provides an interactive surface fo r transcriptional corepressors and coactivators which links nuclear receptor activity with the preinitiation complexs (PIC). Nuclear receptors act in three steps: repression, derepression and transcriptional activation (MacDonald, 2001). Repression is cha racteristic of the apo (unliganded) nuclear receptor, which recruits a corepressor complex that possesses histone deacetylase activity (HDAC). Derepression occurs following ligand binding, which dissociates this complex and recruits a first coactivator com plex that has histone acetyltransferase (HAT) activity, resulting in chromatin condensation. In the third step, the HAT complex dissociates and a second coactivator complex is assembled (TRAP/DRIP/ARC), which is able to establish an interaction with the ba sal transcription machinery leading to transcriptional activation of the target gene.
12 Nuclear receptors may or may not contain a final domain at the C terminus of the E domain, the F domain. The sequence of the F domain is extremely variable among differen t receptors and their function is largely unknown. Regulation of vitamin D receptor VD and VDR VD was discovered nearly a century ago as the nutrient that prevented rickets, a devastating skeletal disease characterized by undermineralized bones (Brown, 1 999). VD is a lipophilic hormone essential for normal bone structure and to maintain serum calcium levels. VD acts in concert with parathyroid hormone (PTH) to tightly regulate the concentration of serum calcium and phosphate, thereby assuring proper skele tal mineralization. In addition to its classic function in mineral homeostasis, the present concept of VD had been broadened to that of a hormone involved in a complex system that regulates proliferation and differentiation of a variety of cell types and t issues (Sutton, 2003) (Fig. 2). 1,25VD is the biologically active form of VD. In the skin, 7 dehydrocholesterol is photo converted to pre VD upon exposure to ultraviolet light, is then isomerized to VD. Subcequently, VD is metabolized in the liver to 25 h ydroxy VD3 and, primarily in the kidney, to 1,25VD. The major source of VD is through sunlight (UV B) induced photobiosythesis in the skin instead of food consumption (Sutton, 2003) (Fig. 2). VD action is mediated through the VDR, a member of class II nuclear receptors. VD dependent stabilization of the VDR protein appears to occur in virtually all cell types. VDR heterodimerizes predominantly with RXR. Although VDR also forms
13 Fig. 2. Metabolism and biological response of Vitamin D. calcium homeostasis Non genomic mechanisms Classic Vitamin D responsive tissues: Intestine Skeleton Parathyroid glands Kidney Nonclassic Vitamin D responsive tissues: Hematopietic tissues Immune system Pancreas Skin and hair development Nervous system Reproductive system Endocrine pathway: Paracrine autocrine pathway: Cellular regulation of proliferation, differentiation and apoptosis UV light Skin 7 dehydroxcholesterol Vitamin D3 Diet Liver 25(OH)D3 1,25(OH) 2 D 3 (1,25VD) ( Active form) Kidney 25 OHase 1 a OHase 24 OHase 24,25(OH)2D3 1,24,25(OH)3D3 Excreted metabolites PTH VDRE RXR VDR Target gene calcium homeostasis Non genomic mechanisms Classic Vitamin D responsive tissues: Intestine Skeleton Parathyroid glands Kidney Nonclassic Vitamin D responsive tissues: tissues Immune system Pancreas Skin and hair development Nervous system Reproductive system Endocrine pathway: Paracrine autocrine pathway: Cellular regulation of proliferation, differentiation and apoptosis UV light Skin 7 dehydroxcholesterol Vitamin D3 Diet Liver 25(OH)D3 1,25(OH) 2 D 3 (1,25VD) ( Active form) Kidney 25 OHase 1 a OHase 24 OHase 24,25(OH)2D3 1,24,25(OH)3D3 Excreted metabolites PTH VDRE RXR VDR Target gene calcium homeostasis Non genomic mechanisms Classic Vitamin D responsive tissues: Intestine Skeleton Parathyroid glands Kidney Nonclassic Vitamin D responsive tissues: Hematopietic tissues Immune system Pancreas Skin and hair development Nervous system Reproductive system Endocrine pathway: Paracrine autocrine pathway: Cellular regulation of proliferation, differentiation and apoptosis UV light Skin 7 dehydroxcholesterol Vitamin D3 Diet Liver 25(OH)D3 1,25(OH) 2 D 3 (1,25VD) ( Active form) Kidney 25 OHase 1 a OHase 24 OHase 24,25(OH)2D3 1,24,25(OH)3D3 Excreted metabolites PTH VDRE RXR VDR Target gene calcium homeostasis Non genomic mechanisms Classic Vitamin D responsive tissues: Intestine Skeleton Parathyroid glands Kidney Nonclassic Vitamin D responsive tissues: tissues Immune system Pancreas Skin and hair development Nervous system Reproductive system Endocrine pathway: Paracrine autocrine pathway: Cellular regulation of proliferation, differentiation and apoptosis UV light Skin 7 dehydroxcholesterol Vitamin D3 Diet Liver 25(OH)D3 1,25(OH) 2 D 3 (1,25VD) ( Active form) Kidney 25 OHase 1 a OHase 24 OHase 24,25(OH)2D3 1,24,25(OH)3D3 Excreted metabolites PTH VDRE RXR VDR Target gene
14 Fig. 2. Metabolism and biological response of Vitamin D. Bioactive 1,25VD is generated by sequential hydroxylations of its precursor Vitamin D 3 in the liver and kidney. PTH (parathyroid hormone) stimulates 1 a OHase expression in the kidney and promotes calcium mobilization from the bone and reabsorption from the kidney. 1,25VD, in turn, induces calcium absorption in the intestine and calcium release from the skeleton. The action of 1,25VD is mediated through VDR or non genomic pathway. The distinct endocrine and paracrine/autocrine roles of 1,25VD are emphasized. The classic and nonclassic Vitamin D responsive tissues are also listed.
15 homodimers, these homodimers may not be transcriptionally active (Nishikawa, 1994; Issa, 1998). The human VDR (hVDR) gene has been localized to chromosome 12q13 14. The hVDR cDNA was cloned from a human jejunal poly(A)+ RNA library using avian VDR cDNA probe. hVDR gene contains 11 exons that span more than 75kb (Miyamoto, 1997). The noncoding 5 prime end of t he VDR gene includes exons 1A, 1B, and 1C, while its translated product is encoded by 8 additional exons (2 to 9). Three mRNA isoforms are produced as a result of the differential splicing of exons 1B and 1C. A 4.6 kb human transcript, contains a 1281 nuc leotide open reading frame that codes for the full length VDR protein of 427 amino acids. The hVDR coding sequence is highly homologous to the avian, amphibian, mouse and rat sequences, particularly in the highly conserved nine cysteine residues of the DBD Mammalian forms of VDR protein range in molecular weight between 52 60 by biochemical analysis, although the calculated molecular weight deduced from the amino acids sequence is 48.3 kD. The promoter of hVDR is GC rich and does not contain an apparent TA TA box, but has multiple Sp 1 recognition sites and an array of putative binding sites for transcription factors. VDR shares structural homology with other nuclear receptors. However, AF1 is absent in the short A/B region at the N terminus in VDR since re moval of the A/B domain does not affect ligand binding, DNA binding or transactivation (Fig. 3). Two highly conserved zinc finger DNA binding motifs constitute the DBD, which also contains the nuclear localization signal. The D domain or hinge region regul ates the receptors flexibility for conformational changes. Most of the natural mutations found in human VDR are located in the zinc finger region, resulting in defective DNA binding and
16 Fig. 3. Molecular structure of the VDR. The DNA binding do main contains two Zinc finger structures. The ligand binding domain contains putative heterodimerization interfaces. Also shown are two phosphorylation sites at Ser 51and Ser 208. A TIFIIB interactive interface and a coactivator binding domain are also sho wn. DNA binding Ligand binding coactivator interaction 22 114 166 402 427 A B C D E Ser51 Ser208 N C p p p p p p Zn Zn AF 2 DNA binding Ligand binding coactivator interaction 22 114 166 402 427 A B C D E Ser51 Ser208 N C p p p p p p p p Zn Zn p p Zn Zn AF 2
17 the most severe clinical phenotype of VD resistance (Haussler, 1998). The VDR LBD contains nine heptad repeats that form hydrophobic surfaces thought to act as dimerization interfaces. Coregulators of VDR In response to VD activation, VDR recruits m ultiple co activators, including members of the p160 SRC family (Gill, 1998) and CBP/p300 family (Castillo, 1999) that either have or can or recruit histone acetyl transferase activity. These coactivators are essential for the formation of the initial tran scription complex with RNA polymerase II. Recently, another coactivator complex has been identified as DRIP/TRAP complex (Rachez, 1999) that has no HAT activity and serves as a mediator between the VDR and RNA polymerase II complex (Pol II). DRIP/TRAP and SRC/p160 exist as distinct complexes but act cooperatively, as suggested in recent studies where both complexes were shown to act during the early stages of kerotinocyte differentiation. In the later stages of differentiation, DRIP/TRAP levels decrease an d SRC/p160 assumes a predominant role (Oda, 2003). In addition to SRC and DRIP, NcoA 62/ski interacting protein (skip) can augment VDR transcriptional activity (Baudino, 1998; MacDonald, 2001). Skip lacks LXXLL motifs, selectively associates with the VDR R XR through the LBD, but through a domain that is distinct from the Helix3 Helix5/Helix12 interaction surface. In contrast to coactivators, corepressors, NCoR (nuclear receptor corepressor) and SMRT (silencing mediator of retinoid and thyroid hormone recept or s) have been found to associate with ligand unbound VDR, TR and RAR to repress ligand induced transactivation functions.
18 Recently, Kitagawa et al. (2003) identified a multiprotein complex ( W STF I ncluding N ucleosome A ssembly C omplex, WINAC); This is a no vel ATP dependent chromatin remodeling complex that directly interacts with VDR through the Williams syndrome transcription factor (WSTF). WINAC, like other SWI/SNF and ISWI/complexes, can reorganize the chromatin structure either by opening it to allow tr anscription or compacting it to repress gene expression. Manipulation of WSTF expression levels established that it is essential for VDR activity by either stimulating transcription from promoters regulated by positive response elements or repressing trans cription from promoters containing negative response elements. Overexpression of WSTF can restore the impaired recruitment of VDR to VD regulated promoters in fibroblasts obtained from Williams syndrome patients. This suggests that WINAC dysfunction contri butes to Williams syndrome. This discovery has shed new light on the molecular mechanisms by which VDR controls gene expression with unexpected clinical implications (Belandia, 2003). Phosphorylation of VDR Hormone induced phosphorylation has been found in most nuclear receptors and may be involved in the regulation of receptor function. VDR contains two phosphorylation sites through which its activity can be modulated. Casein kinase positively regulates VDR activity while protein kinase C or protein kinase A negatively regulate it (Hsieh, 1991; Jurutka, 1996). VDR is phosphorylated by PKC at Ser 51 in vitro and in vivo resulting in retardation of specific interactions of VDR with VDREs. Ser 208 is the preferred site phosphorylated by casein kinase II (CKII ). Phosphorylation of Ser 51 or Ser 208 is not essential for VDR function but may reflect kinase specific
19 inputs that can be either positive or negative. The functional consequence of phosphorylation on VDR activity depend on both the cellular context and the signal transduction pathway or specific kinase involved. VDR phosphorylation represent a mechanism by which signals from the cell membrane in response to growth factor stimulation can modulate nuclear hormone receptor function. VDRE and VDR VDR forms h eterodimers with RXR. Isoforms of the RXR ( a b g ) serve as dimeric partners for VDR. In most cases, the VDR/RXR heterodimer binds VD response elements (VDREs) to mediate the biological activities of VD through transcriptional activation or repression of ta rget genes containing VDREs in their regulatory sequence. Almost all VDREs were identified in the promoter region of VD regulated genes (Issa, 1998). Although there is considerable variations between natural VDREs, a consensus DR3 type VDRE can be defined as a direct repeat of two six base half elements of the sequence, separated by a spacer of three nucleotides. The VDRE sequence directs the VDR RXR heterodimer where RXR binds the 5 half site and VDR occupies the 3 half site (Haussler, 1998). The orie ntation and spacing of the response elements directs the polarity of heterodimer binding. The fifth nucleotide position of the 3 half site is thought to be necessary for heterodimer binding. Although VDR RXR complexes that interact with DR3 VDREs predomin ate the transcriptional response to 1,25VD, alternative routes appear to be possible both with respect of dimmer formation and target gene VDRE structure (direct repeat 6, DR6 and inverted repeat 9, IR9). The RXR dimerization surfaces on VDR are located in the first zinc finger and in a structural motif designated as
20 heptad repeats in the LBD. Asn37 in the P box of the first zinc finger, Lys91 and Glu92 situated in the T box COOH terminal of the second zinc finger, and two of the heptad repeats within the L BD are critical in determining selective association between the VDR and its protein partner, RXR. Heterodimerization of the ligand activated VDR with RXR induces a VDR conformation that is essential for VDR transactivating function (Brown, 1999). VDR homo dimers showed binding activity for DR6 type elements in both ligand dependent and independent manner, while the functional significance of a VDR homodimer is yet unclear. Several negatively regulated VDREs were also identified in some genes such as, avian and human PTH (Liu, 1996) and protein kinase A (PKA) inhibitor (Rowland Goldsmith, 1999). Interestingly, by changing the two 3 terminal bases GT of the avian PTH VDRE to the concensus CA, the VDRE reversed from a negative to a positive VDRE (Koszewski, 1999). Nongenomic effect of VD Most of the biological actions of VD are thought to occur through the nuclear VDR mediated expression of target genes. Because these responses require transcription and translation of target genes, they are typically delayed by at least 30 min. However, more rapid effects (within seconds to minutes) in response to steroid hormones are also observed. For example, 1,25VD can stimulate the rapid formation of second messengers including ceramides, cAMP, inositols, calcium and to a ctivate a variety of protein kinases such as protein kinase C, protein kinase A, raf, MAPK, and src kinases family (Gniadecki, 1996; Gniadecki, 1998; Marcinkowska, 1997). Although many in vitro studies have shown rapid effects by 1,25VD and numerous other steroids, the field is
21 hindered by the inablility to identify the putative membrane receptors that trigger these nongenomic effects. The study in VDR knock out (VDRKO) mice showed that nongenomic effects of VD in osteoblasts are abrogated in the absence of nuclear VDR, suggesting that some nongenomic response require a functional nuclear VDR (Erben, 2002). VDRKO mice are important tools to decipher the molecular requirements of classical VDR that mediate the genomic and nonegenomic effects of 1,25VD in viv o Biological actions of VD The genomic and nongenomic actions of VD produce a multitude of responses. VDR has been found in classical VD target organs such as the intestine, bone, kidney, and the parathyroid glands as well as a host of target tissues not involved in calcium homeostasis, such as skin, muscle, pancreas, reproductive organs, and hematopoietic, immune and nervous systems (Berger, 1988; Clemens, 1988). The diversity and the growing number of VD regulated genes are a reflection of the pleiotropi c effects of 1,25VD on calcium and phosphate homeostasis, bone turnover, and proliferation and differentiation of a wide variety of cells, including keratinocytes, cancer cells and immune cells. Therefore the diversity and importance of VDR function goes f ar beyond mineral metabolism. Moreover, locally produced 1,25VD may serve as a paracrine modulator of cell growth and differentiation. To investigate the functional role of VDR, Yoshizawa et al. (1997) generated mice deficient in VDR. VDRKO mic e are viable and develop normally until the weaning period irrespective of reduced expression of VD target genes. Perhaps, the higher calcium content of murine milk, compared to human milk, keeps serum calcium normal, assuring normal growth of VDR null mut ant mice before weaning. Not surprisingly, after
22 weaning, VDRKO mice displayed typical features of hereditary vitamin D resistant rickets (HVDRR), a rare genetic disorder caused by mutations in the VDR genes, with symptoms such as severe bone formation, h ypocalcemia and alopecia. Although uterine hypoplasia, infertility and early lethality are not pronounced in patients with VD dependent rickets type II, possibly because of therapy with calcium supplements, most VDR null mutant mice died within 15 weeks af ter birth; uterine hypoplasia with impaired folliculogenesis were observed. Uterine hypoplasia was caused by impaired estrogen synthesis in the mutant ovaries. However, the uterus of these animals responded normally to administration of estrogen. Male repr oductive organs appeared normal in VDRKO mice. The fact that VDRKO mouse not only exhibits all features of human rickets, but also has marked growth retardation after weaning and uterine hypoplasia, implicates a role for VDR and calcium homeostasis during reproductive development and growth. Classical target tissues of VD related to calcium homeostatis are the intestines (calcium and phosphate absorption), kidney (phosphate and calcium reabsorption), parathyroid glands (suppression of PTH) and bone (osteocl asteogenesis, osteoblasts and mineralization). Target genes in these organs with prominent VDR in their promoters are the calcium binding proteins calbindin D9K (intestine), calbindin D28K (kidney and other tissues), the non collagenous bone specific mat rix protein osteocalcin and osteopontin (extracellular matrix protein in bone and other tissues) (Haussler, 1998). Serum 1,25VD is held constant in the normal state and is strictly regulated in response to factors controlling calcium homeostasis. 1,25 VD r egulates its own metabolism and biosynthesis by stimulation of 24 hydroxylase activity and inhibition of 1 a hydroxylase (Sutton, 2003).
23 In addition to its role in calcium homeostasis and bone metabolism, 1,25VD exhibits anti inflammatory and immunosuppr essive properties (Bouillon, 1995; Issa, 1998). 1,25VD interacts with mature monocytes and macrophages, enhancing their immune function and improving host defense against both bacterial infection and tumor cell growth. In contrast to the stimulatory effect s of the hormone on monocytes and macrophages, the principal action of 1,25VD in lymphocytes is as an immunosuppressive agent. It does so by decreasing both the rate of proliferation and the activity of T cells and B cells. 1,25VD and its analogues may pro ve beneficial as potential therapeutics in autoimmune diseases, such as psoriasis, multiple sclerosis, rheumatoid arthritis, diabetes and in transplantation. Effect of vitamin D in cancer Epidemiological data suggest that low VD levels increase the risk a nd mortality (Fig. 4) of prostate, breast and colon cancers (Waterhouse, 1976; Devasa, 1999). 1,25VD modulates cellular proliferation and differentiation of both normal and malignant cells. 1,25VD and its synthetic analogues inhibit carcinogenesis in mouse skin (Chida, 1985; Webb, 1988), decrease the size of transplanted sarcomas, reduce lung metastasis in mice (Sato, 1982), suppress the growth of human colonic, prostate and pancreatic cancer xenografts in vivo (Eisman, 1987; Blutt, 2 000; Colston, 1997), increase cell differentiation and decrease proliferation of leukemia (Dodd, 1983), breast (James, 1994) and prostate cancer (Blutt, 1997) cells. Studies in cancer cell lines have shown that 1,25VD causes cancer cells to accumulate in t he G1 phase of the cell cycle (Blutt, 1997), in the G2 phase ( Eisman, 1989 ) or undergo apoptosis ( Blutt 2000; Simboli Campbell, 1996). The list of VD induced proteins (Segaert, 1998) that pertain to cell growth and
24 Fig. 4. Ovarian cancer mortal ity rates in US, 1970 1994 (Devasa DJ et al. 1999, Atlas of cancer mortality in the United States. NIH publication No. 99 4564: pp 226 230).
25 differentiation is rapidly growing with HoxA10 (homeobox protein causing G1 arrest), Mad1 (differentiation relate d transcriptional repressor), TNF a insulin like growth factor binding proteins 3 and 5, apolipoprotein D, tumor suppressors BRCA1 and E cadherin. VDREs of numerous VD responsive genes have been identified and characterized. Some VDREs are simple direct re peats while others are complex with often overlapping multimeric structures. The induction of DNA protein kinase links 1,25VD to DNA repair and cancer chemoprevention mechanisms. 1,25VD and its synthetic analogues appear to exert their growth inhibitory e ffects via regulation of cell cycle progression. Typically, treatment of cells with 1,25VD causes cell arrest in the G 1 phase, resulting in a decreased number of cells in S phase, and increase in G 0 /G 1 This change is associated with alterations in the exp ression of cell cycle regulators. Although a number of genes involved in cell cycle control, apoptosis and cell differentiation have been identified, the exact mechanisms underlying the growth regulatory actions of 1,25VD and its analogues have not been co mpletely defined. In prostate model systems, VD has significant anti tumor activity in vitro and in vivo (Blutt, 1997 and 2000; Johnson, 2002). The effect of 1,25VD is associated with an increase in cell cycle arrest, apoptosis, differentiation and modulat ion of growth factor receptors. VD induces G0/G1 arrest and modulates cyclin dependent kinase inhibitors, p21 and p27. VD induces PARP cleavage, increases bax/bcl 2 ratio, reduces levels of p MAPKs, and p AKT, induces caspase dependent MEK cleavage and up regulation of MEKK 1, increases annexin V binding, which are markers of the apoptosis pathway (Johnson, 2002). Microarray studies in prostate cancer cells reveal many biologically relevant molecular targets of 1,25VD. For example, in LNCaP cells, 1,25VD ca uses
26 growth arrest through the induction of insulin like growth factor binding protein 3 (IGFBP3). These results provide a starting point for additional investigations to fully elucidate the mechanism of 1,25VD action in the prostate. 1,25VD can also signi ficantly increase the efficacy of drug mediated cytotoxicity. Phase I and II trials of 1,25VD either alone or in combination with carboplatin, paclitaxel or dexamethasone have been initiated in patients with prostate cancer. Data from these studies indicat e that high dose 1,25VD is feasible on an intermittent schedule, and provides proof of the concept that 1,25VD or its analogs are clinically effective (Johnson, 2002). 1,25VD inhibits the proliferation and induces the differentiation of normal and leuke mic myeloid cells into monocytes (Abe, 1981 ). 1,25VD regulates numerous genes such as c fos, c myc, IL 1, IL 6, TNF a p21, p27 and Mad1, a pro differentiating gene. These findings suggest that 1,25VD action on leukemic cells involves cell cycle control and differentiation. The response of hematopoietic cells to 1,25VD treatment depends on the cell type, differentiation state and dose of 1,25VD used. In most cell types, growth inhibition and maturation accompany each other. Both antiproliferative and differentiating effects of 1,25VD suggest a therapeutic role for the drug in hematological malignancies. In keratinocytes (Kitano, 1991), colonic adenocarcinoma cell lines (Scaglione Sewell, 2000) and squamous cancer cells (Hager, 2001), 1,25VD also blocks cell cycle progression in the G1 phase, preceded by the induction of the cyclin dependent kinase inhibitors p21 and p27. This induction i nvolves direct activation of p21 promoter through VDRE, as well as indirect mechanisms involving induction of growth inhibitory cytokine TGF b 1 (transforming growth factor beta 1) and its type II receptor. In squamous cancer
27 cells, increased levels of CDK i nhibitors prevent phosphorylation of the retinoblastoma protein and subsequent release of transcription factors of the E2F family. Therefore, upregulation of S phase related genes, such as the proto oncogene c myc, is inhibited. Non classical positive VDRE s were also demonstrated in the promoter of c fos and phospholipase C g 1, that both play a role in growth regulatory pathways. In human osteoblastic cells, cell cycle arrest at G0 G1 by a VD analogue is accompanied by hypophosphorylation of Rb followed by strong inhibition of Cdk2 activity (Maenpaa, 2001). These effects also correlated with increased level of p27, decreased level of Cdk2 and cyclin E, but p21 and cyclinD1 were not affected. G2/M arrested by VD or a VD analogue was also detected in Keratinoc ytes (Kobayashi, 1998). 1,25VD has also been shown to induce apoptosis in breast (Welsh, 1994), prostate (Hsieh, 1997), colon (Diaz, 2000) and glioma (Baudet, 1996) cell lines. 1,25VD and its analogues induce apoptosis in MCF 7 breast cancer cells; induct ion of apoptosis was associated with upregulation of p53 and Bax. 1,25VD and its analogues are also capable of inducing apoptosis in T47 D breast cancer cells, which possess a mutated p53. It appears that induction of apoptosis is independent of p53 status (Mathiasen, 1999). This study suggests that sensitivity to 1,25VD mediated apoptosis may be determined by the relative expression or subcellular distribution of pro and anti apoptotic members of the bcl 2 family rather than activation of any known caspas e (Mathiasen, 1999). Upregulation of apoptosis related proteins such as clusterin, cathepsin B and TGF b has been reported in MCF 7 cells undergoing apoptosis in response to 1,25VD and its analogues (James, 1996). In prostate cancer cells, 1,25VD induced d ecreases in the levels of antiapoptotic proteins Bcl 2, Bcl XL, and Mcl 1, BAG1L, XIAP, cIAP1, and
28 cIAP2 (without altering proapoptotic Bax and Bak) in association with increases in apoptosis (Guzey, 2002). Moreover, induction of apoptosis by 1,25VD was su ppressed by overexpressing Bcl 2, a known blocker of cytochrome c release (Blutt, 2000; Guzey, 2002). At this time the molecular mechanisms by which 1,25VD may induce apoptosis are not fully understood. Vitamin D analogue in cancer prevention and therapy Since the potential use of 1,25VD in the treatment or prevention of cancers is limited by the tendency of 1,25VD to cause hypercalcemia, recent research has focused on the development of analogues with less calcemic and/or greater antineoplastic activity t han 1,25VD. Numerous VD analogues have been synthesized and many of them have modifications in the C 17 side chain of VD. Among them, Seocalcitol (EB1089) is the analog that has been widely administrated to patients in Europe in a Phase I trial resulting in stabilization of the disease in patients with advanced breast and colorectal cancer (Gulliford, 1998). EB1089 contains a conjugated double bond system and is approximately 50 times more potent than 1,25VD in vitro while the actions of EB1089 in calcium metabolism in vivo are markedly reduced. Phase II trials with hepatomas showed reduction in tumor dimensions in patients with an advanced bulky, solid tumor (Dalhoff, 2003). Their studies suggest that 1,25VD or its analogue have an effect in the treatment of multiple human cancers. Either 1 ,25VD or EB1089 in combination with 9 cis RA act cooperatively to inhibit growth of breast, prostate, ovarian or small cell lung cancer cells (James, 1995; Guzey, 1998). 9 cis RA is a metabolite of vitamin A that has potent influences on cell differentiation, proliferation, homeostasis, and development. 1,25VD acts synergistically
29 with dexamethasone to suppress growth of breast, OCa and prostate cell lines (Johnson, 2002; Saunders, 1995). These findings have i mportant implications for the use of retinoids or dexamethasone with 1,25VD in cancer therapy. 1,25VD treatment enhances the sensitivity of prostate cancer cells to a number of anti cancer drugs, such as paclitaxel (Taxol) and cisplatin (cis diamminodichl oroplatinum), indicating that VD compounds may be useful if used in combination with conventional chemotherapy (Johnson, 2002). Vitamin D and ovarian cacer (OCa) Similar to breast and prostate cancers, OCa mortality and incident rates are lower in countr ies within 20 degrees of the equator (Waterhouse, 1976) where there are high amounts of sunlight. In the US women between the ages of 45 54 living in the North have 5 times the OCa mortality rate than women living in Southern states (Devasa, 1999; Waterhou se, 1987). The inverse correlation between sunlight exposure and OCa mortality indicates that in American women decreased synthesis of 1,25VD may contribute to OCa initiation and/or progression (Lefkowitz, 1994). OCa has the worst prognosis and remains th e most challenging among gynecological malignancies, since it is not diagnosed at an early stage and has severely progressed when diagnosed. Although surgical dissection of tumors and intense chemotherapy is routinly used to treat OCa, severe drug resistan ce results in only 20 30% survival rates. Poor response of advanced OCa to current treatments necessitates the development of novel therapeutic strategies to fight this deadly disease. In recent years studies have been initiated to develop synthetic VD a nalogs as therapeutic agents for a variety of human cancers including breast, prostate and colon cancers, while the similar studies in ovary are very limited.
30 VDR has been found in both normal and cancerous human ovarian cells by imm unohistochemistry; 83.3% of normal surface epithelium shows weak to moderate VDR immunoreactivity. Moderate to strong nuclear immunoreactivity for VDR was detected in almost all ovarian carcinomas tested (Villena Heinsen, 2002). VDR was also found in rat o varies by immunohistochemistry (Johnson, 1996) and in hen ovaries by ligand binding assays (Dokoh, 1983), indicating that ovarian cells can respond to VD. VDR expression in gynecologic neoplasms, including OCa (Ahonen, 2000; Saunders, 1992; Villena Heinsen, 2002), has also been described, indicating that VD could be an effective agent in OCa treatment and/or chemoprevention. The presence of VDR was demonstrated in the ovary and the VD induced decrease in cell number in CHO and OVCAR3 was also described. Specifically, Ahonen et al. (2000) showed that a 9 day treatment of OVCAR3 cells with 100 nM 1,25VD resulted in 73% inhibition of growth. However, the mechanism of VD action in OCa cells, as well as the potential of 1,25VD and its analo gues in OCa treatment remains unknown. GADD45 and ovarian cancer The growth arrest and DNA damage inducibe (GADD) gene GADD45 codes for a multifunctional protein (19 kD) that binds numerous proteins and plays a role in cell cycle progression as well as t he maintenance of genomic stability. Both genotoxic (i.e., UVR, IR, cisplatin, and adriamycin) and nongenotoxic stresses (i.e., apoptotic and/or growth inhibitory cytokines, serum starvation) induce GADD45 activation. GADD45 a (GADD45) belongs to the GADD45 family which also contains GADD45 beta and gamma. GADD45a is the only member that is induced by p53; the p53 binding site has
31 been identified in intron 3 (Hollander, 1993). p53 independent induction of GADD45 may also be achieved depending on the insult. GADD45 interacts with the products of two other p53 regulated genes, p21 and PCNA (proliferating cell nuclear antigen) (Vairapandi, 1996; Smith, 1994). PCNA impedes Gadd45 mediated negative growth control (Vairapandi, 2000). Zhan et al (1999) have shown that Gadd45, through its association with cdc2, appears to disrupt interactions between cdc2 and Cyclin B1 and thus may induce arrest at G 2 /M. GADD45 null mice have increased sensitivity to dimethylbenzanthracene (DMBA) induced carcinogenesis (Hollander, 2001). It is worth to note that in Gadd45a null mice, DMBA induces a dramatic increase in female ovarian tumors compared to the wild type (Hollander, 2001). Therefore, GADD45 may protect the ovary from carcinogenesis, while the role of GADD45 in OCa is not well defined. Although the upregulation of GADD45 by 1,25VD has been found in squamous cancer cells (Akutsu, 2001), the mechanism of the upregulation and the role of GADD45 in 1,25VD induced growth inhibition remain unknown. Telomerase and ovarian cancer The ends of chromosomes, the telomeres, are subject to progressive shortening in normal somatic cells, leading ultimately to irreversible growth arrest. In contrast, telomeres in all cancer cells are stabilized in length and effectively immortalized by th e enzyme telomerase, which catalyzes the synthesis of telomeric DNA repeats. Telomerase is a ribonucleoprotein complex that is made up of three components: 1. RNA template human telomerase RNA (hTR) contains a sequence complementary to the telomeric TTAGGG repeat; Telomeres act as protective caps, stabilizing the chromosomes by
32 preventing their degradation and aberrant recombination during cell division. 2. Protein component human telomerase associated protein 1 (hTEP1). 3. Catalytic subunit, known as hTERT, which is a type of reverse transcriptase able to synthesize TTAGGG repeats from the RNA template. The cellular activity of telomerase is determined by the presence or absence of hTERT. All human somatic cells constitutively express hTR (Cech, 2004; Newbold, 2002). Telomerase is responsible for the replication of chromosome end structures and is strongly upregulated in most human cancers. RT PCR analysis revealed that hTR and TP1 mRNA were expressed in more than 80% of OCa and even in normal ovari es. There was a significant correlation of telomerase activity with hTERT mRNA expression but not with TP1 or hTR. Repression of telomerase activity is associated with hTERT mRNA, but the expression of hTR and TP1 remained unchanged (Park, 1999; Kyo, 1999) The rate limiting step for telomerase activity seems to be the expression of the hTERT gene. The precise mechanism of how hTERT is regulated has not been elucidated yet. It has been shown that OCa cells containing wild type hTERT readily produ ce tumors while cells with dominant negative hTERT failed to form tumor in nude mice. This studya confirms the hypothesis that inhibition of telomerase will decrease tumorigenicity of OCa cells in vivo and supports the concept that hTERT is a potential sit e for anti cancer drug design for OCa (White, 2001). The protective effect of telomerase from apoptosis has been proposed for human fibroblasts and neuron cells (Ren, 2001; Gorbunova, 2002; Fu, 2000 ). A study of epidermoid tumor cells indicated that telom erase inhibition in cells with short telomeres leads to chromosomal damage,
33 which in turn triggers apoptotic cell death (Zhang, 1999). These data indicated that i nhibition of telomerase activity is a potential approach for the treatment of human malignancy Until now, little is known about the mechanism and significance of telomerase repression by VD in solid tumors including OCa. We observed growth inhibition by VD in OCa cells. Consequently, the mechanism and role of telomerase in VD induced growth inhibi tion was investigated in OCa cells.
34 STUDY OBJECTIVES We hypothesize that 1,25VD through VDR mediated gene regulation inhibits the development of OCa; We further hypothesize that 1,25VD and its synthetic analogues might be used for OCa prevention and t herapy. Based on hypothesis, the study objectives are: 1. Determine the biological responses of OCa cells to 1,25VD treatment; 2. Examine molecular mechanism of 1,25VD action by identifying target genes which mediate the different biological response; 3. Test the antitumor activity of 1,25VD analogues in vitro and in vivo.
35 RESULTS The work presented here provides evidence that GADD45 and telomerase are regulated by 1,25VD and defines a mechanism for 1,25VD in OCa cells. Our studies identify GADD45 and hTERT as important mediators of the tumor suppressing activity of 1,25VD in OCa cells. Our study also provides preclinical data indicating the effectiveness of synthetic 1,25VD analogues in OCa prevention and therapy. Effect of 1,25VD in OCa cells 1. 1,25VD suppresses OCa cell growth and induces cell cycle arrest at G1/S and G2/M To better understand the molecular mechanism of 1,25VD action in OCa cells, we tested the response of OCa cells to 1,25VD in proliferation assays. Cells were treated w ith vehicle (ethanol, ETOH) or 10 7 M 1,25VD for 6 or 9 days and cell growth was determined by MTT assays. As shown in Figure 5, panel A, OVCAR3 cell growth decreased in the presence of 10 7 M 1,25VD in a time dependent manner, confirming the sensitivity o f OVCAR3 to VD (Saunders, 1992). Since cell growth was not significantly affected by 1,25VD at concentrations of 10 8 M or lower (data not shown), it appears that there is a threshold to the action of 1,25VD. To determine the mechanism underlying the 1,2 5VD induced growth suppression, OVCAR3 cells were treated with vehicle or 10 7 M 1,25VD for 9 days and analyzed by
36 Fig. 5. 1,25VD inhibits OCa cell growth and induces cell cycle arrest at G1/S and G2/M checkpoints. 0.0 0.2 0.4 0.6 0.8 1.0 ETOH VD A B ETOH VD MTT (OD 595 ) 0 6 9 Days of Treatment 1 0 10 20 30 1 50 60 70 80 1 0 5 10 15 20 C G0/G1 S G2/M ETOH VD ETOH VD ETOH VD * * Percentage of cells (%) 0.0 0.2 0.4 0.6 0.8 1.0 ETOH VD A B ETOH VD MTT (OD 595 ) 0 6 9 Days of Treatment 1 0 10 20 30 1 50 60 70 80 1 0 5 10 15 20 C G0/G1 S G2/M ETOH VD ETOH VD ETOH VD * * Percentage of cells (%) A B ETOH VD MTT (OD 595 ) 0 6 9 Days of Treatment 1 0 10 20 30 1 50 60 70 80 1 0 5 10 15 20 C G0/G1 S G2/M ETOH VD ETOH VD ETOH VD * * Percentage of cells (%)
37 Fig. 5. 1,25VD in hibits OCa cell growth and induces cell cycle arrest at G1/S and G2/M checkpoints. ( A ) Suppression of cell growth by 1,25VD. OVCAR3 cells were plated in 96 wells and treated with 10 7 M 1,25VD (VD) or ethanol (ETOH) as a vehicle. Cell numbers were determin ed by the MTT assay. Eight samples were analyzed for each data point, and the data were reproduced three times, *P<0.01 (versus ETOH treatment) ( B ) Induction of cell cycle arrest at G1/S and G2/M checkpoints by 1,25VD. OVCAR3 cells were treated with ETOH or 10 7 M 1,25VD for 9 days. Treated cells were subjected to flow cytometry analysis. Data were reproduced three times. The cell cycle profile of a representative experiment is shown. G2/M peak (red color) is indicated by the arrow. ( C ) Bar graphs show th e average percentages of cells at G0/G1, S and G2/M in OVCAR3 cells treated with 1,25VD or ETOH for 9 days, P<0.05 (versus ETOH treatment).
38 flow cytometry. Figure 5, panel B shows that 1,25VD decreased the percentage of cells in the S phase, which was accompanied by an accumulation of cells in G0/G1 and G2/M. This suggests that 1,25VD causes cell cycle arrest at both G1/S and G2/M checkpoints. The increase in G0/G1 was estimated as 13% while the percentage of cells at G2/M increased by 8% (Fig. 5C), su ggesting that cell cycle arrest at both checkpoints contributed roughly equally to the growth suppressing activity of 1,25VD. Contrary to drugs used in conventional cancer chemotherapy, inhibition of cancer cell growth by 1,25VD is a chronic process, expla ining why the effect on the cell cycle is modest and requires treatment for a longer time. 2. 1,25VD induces apoptosis in OCa cells To determine whether 1,25VD treatment induces apoptosis in OVCAR3 cells, OVCAR3 cells were treated with vehicl e or 1,25VD for 9 days and fragmented DNA was detected by the Annexin V apoptosis assay kit. Apoptosis was induced to more than 60% in OVCAR3 cells after 9 days treatment of 1,25VD, as shown in Figure 6. 3. 1,25VD treated OVCAR3 cells recover slowly after 1,25VD withdrawal. To determine whether changes induced by 1,25VD are sustainable, we tested whether growth inhibition of OVCAR3 cells by 1,25VD can be reversed by removal of the hormone. OVCAR3 cells are treated with 1,25VD for 9 days, and the n 1,25VD was removed and cells were plated in 96 well plates with fresh medium. Cells were subsequently grown for an additional 12 days in the absence of the hormone. Consistent with studies in prostate cancer cells (Blutt, 2000), Figure 7 demonstrated that OVCAR3 cells pretreated with 1,25VD recover slowly from the treatment when
39 Fig. 6. 1,25VD induces apoptosis in OCa cells. (A) OVCAR3 cells were treated with ETOH or 10 7 M 1,25VD (VD) for 9 days. Apoptosis index was determined b y DNA fragmentation assay and a representative profile is shown. (B) Bar graphs show percentage of apoptotic cells in two experiments. ETOH VD A B Apoptotic cells (%) 0 20 40 60 80 0 20 40 60 80 ETOH VD Exp#2 Exp#1
40 Fig. 7. OCa cells recover slowly after 1,25VD treatment. OVCAR3 cells were pretreated with 10 7 M 1,25VD (VD) or ETOH for 9 days. Recovery of treated cells was assessed at indicated times after plating pretreated cells in 96 well plates. Cell numbers were determined with the MTT assay. Eight samples were analyzed for each data point, and the data were reproduced t hree times. Growth Rate 0 3 6 9 12 Days of treatment 0 2 4 6 8 10 12 14 VD pretreatment ETOH pretreatment
41 1,25VD is removed from the medium compared with vehicle treatment. The poor recovery of OVCAR3 cells from 1,25VD treatment further confirmed that extensive growth inhibition was induced by 1,25VD. G2/M arrest by 1,25VD in OCa cells is mediate d through the induction of GADD45 via an exonic enhancer To identify the genes that mediate the inhibitory effects of 1,25VD on cell cycle progression, a microarray analysis was performed to screen for genes that are differentially expressed in OCa cells t reated with vehicle vs. 10 7 M 1,25VD. OVCAR3 cells treated with 1,25VD showed a significant induction of the 1,25VD dependent 24 hydroxylase gene, a gene is known to be upregulated by 1,25VD In OVCAR3 cells, 1,25VD regulated genes encoded growth factors/ modulators, cytokines, kinases and transcription factors. Some of them were implicated in cell cycle regulation and apoptosis (data not shown). GADD45 is a nuclear protein with well known roles in G2 control and it has been shown to be upregulated by 1,25 VD in our microarray analysis. In order to explore the molecular mechanism of 1,25VD induced G2/M arrest, the regulation of GADD45 was investigated in OCa cells. 1. GADD45 is a primary and immediate early response gene for 1,25VD in OCa cells Among the ma ny differentially expressed genes from the microarray experiment, GADD45 was one of the genes that were upregulated by 1,25VD (data not shown). Since GADD45 has a well established role in cell cycle control (Wang, 1999; Zhan, 1999) in ovarian tumorigenesis (Hollander, 2001), we used Northern blots to confirm GADD45
42 regulation by 1,25VD. Compared to cells treated with vehicle, 10 7 M 1,25VD significantly increased GADD45 mRNA levels; 10 8 M 1,25VD caused a barely detectable increase (Fig. 8A), showing that 1 ,25VD induction of GADD45 mRNA is dose dependent. The induction was detectable as early as 2 hours following 1,25VD treatment with maximum induction detected at 8 hours (Fig. 8B), suggesting that GADD45 is an immediate early responsive gene for 1,25VD. Tr eatment for times longer than 8 hours maintained the induction but did not further enhance it. mRNA levels of glyceraldehyde 3 phosphate dehydrogenase (GAPDH) were not affected by 1,25VD, showing some specificity for the effect of 1,25VD on GADD45. The sta bility of GADD45 mRNA, as measured in the presence of actinomycin D, an inhibitor of RNA synthesis, was not different between cells treated with vehicle and 1,25VD (Fig. 8C), suggesting that the regulation of GADD45 by 1,25VD is transcriptional. Although t he fold induction of GADD45 was decreased (Fig. 8D), it persisted in the presence of an inhibitor of protein synthesis, cycloheximide. This shows that GADD45 induction by 1,25VD does not require new protein synthesis, thus identifying GADD45 as a primary 1 ,25VD target gene in OCa cells. This is consistent with the data from squamous carcinoma cells (Prudencio, 2001; Akutsu, 2001). 2. A novel VDRE in the 3 untranslated region of GADD45 mRNA mediates the transcriptional up regulation of GADD45 by 1,25VD To define the specific DNA elements that mediate the induction of GADD45 by 1,25VD, the genomic sequence of GADD45 was examined for the presence of putative VDREs. Based on similarity to the consensus VDRE sequence (Toell, 2000)
43 Fig. 8. 1,25VD increas es transcription of GADD45 mRNA in OCa cells. 0 2 4 6 8 10 0 20 40 60 80 100 120 ETOH VD C ETOH 0 2 4 8 ActD (hours) GADD45 GAPDH VD 0 2 4 8 ActD (hours) GADD45 GAPDH C ETOH 0 2 4 8 ActD (hours) GADD45 GAPDH VD 0 2 4 8 ActD (hours) GADD45 GAPDH Exposure to ActD (hours) Relative Level of mRNA (%) 0.0 0.5 1.0 1.5 2.0 2.5 ETOH VD Fold Induction + CHX CHX CHX ETOH VD ETOH VD D GAPDH GADD45 10 8 M 10 7 M ETOH VD ETOH VD GADD45 GAPDH A B GADD45 GAPDH 0 30 min 2 h 8 h 24 h 0 1 3 6 VD (days) GADD45 GAPDH VD B GADD45 GAPDH 0 30 min 2 h 8 h 24 h 0 1 3 6 VD (days) GADD45 GAPDH 0 1 3 6 VD (days) GADD45 GAPDH VD
44 Fig. 8. 1,25VD increases transcription of GADD45 mRNA in OCa cells. ( A ) Dose dependent induction of GADD45 mRNA by 1,25VD. Total RNA was isolated from OVCAR3 cells treated with ETOH or 1,25VD (VD) at indica ted dosages for 24 h. 20 m g RNA was used for Northern blot analysis with radio labeled GADD45 and GAPDH probes. ( B ) Time course of the induction of GADD45 mRNA by 1,25VD. OVCAR3 cells were treated with ETOH or 10 7 M 1,25VD for the indicated times. Total R NA was isolated and Northern blot was performed as in panel A ( C ) Lack of 1,25VD effect on GADD45 mRNA stability. OVCAR3 cells were treated with ETOH or 10 7 M 1,25VD for 24 h. The cells were washed and subsequently treated with 5 m g/ml actinomycin D (Act D) for the indicated times. Northern blot analysis was performed as in panel A. The signals on the Northern blots were quantified using Scion image Beta 4.02 software. The GADD45 signal was normalized with the corresponding GAPDH signal and presented in t he graph as percentage of the GADD45 mRNA level at time 0. ( D ) Effect of cycloheximide (CHX) on the induction of GADD45 mRNA by 1,25VD. OVCAR3 cells were exposed to ETOH or 10 7 M 1,25VD for 24 h in the absence or presence of 25 m M CHX. Northern blot was performed as in panel A. The GADD45 signal was quantified and normalized with the corresponding GAPDH signal as in panel C and presented in the bar graph as fold induction.
45 and the VDRE sequence of known VD target genes, such as osteopontin (OPN) (Noda, 1 990) and osteocalcin (OC) (Morrison, 1989), five putative DR3 type VDREs were identified (Fig. 9A). Four were in introns, one in exon 4 but none in the 5 promoter region. To test which of the five putative VDREs bind the receptors, gel mobility shift ass ays (EMSAs) were performed with recombinant VDR and RXR proteins. As shown in Figure 9, panel B, our conditions allow the detection of a specific VDR/RXR complex with the OC VDRE. The complex is up shifted by an RXR antibody and decreased by excess cold O C probe (upper panel). Under the same conditions, all putative VDREs except VDRE C bound RXR/VDR with an affinity comparable to OC VDRE (Fig. 9B, lower panel). The binding is specific for the VDR/RXR heterodimer since neither RXR nor VDR alone formed a de tectable complex with the VDRE probes. The VDREs were displaced from the complexes with an excess amount of cold OC VDRE but were not affected by the putative VDRE C that did not bind VDR/RXR (Fig. 9C). This shows that the binding is specific. The RXR anti body up shifted the complex while a non related antibody did not, confirming the presence of RXR in the complex. To determine whether VDREs binding the receptors in OVCAR3 cells mediate the induction of GADD45 by 1,25VD, a reporter gene was constructed with a 2.6 kb genomic DNA fragment of GADD45 containing all the putative VDREs located upstream of SV40 promoter and the cDNA of firefly luciferase gene (Fig. 10A). In OVCAR3 cells transiently transfected with this reporter, 1,25VD induced the luciferase a ctivity in a dose (Fig. 10B) and ti me dependent manner (Fig. 10C).
46 Fig. 9. Multiple putative VDREs are present in GADD45 genome, which interact with VDR/RXR in vitro Consensus DR3 type VDRE (R: A or G, K: G or T, S: C or G) RGKTSA nnn RGKTSA mOPN GGTTCA cga GGTTCA hOC GGGTGA acg GGG G CA Putative VDREs in GADD45 genome A 659 GGGTCA tgg GGG GTG B 1482 GGGTCA gga GGGTG G C 2129 G T TTCA ctc AGGTCA D 2233 GGTTG C atg GGTTCA E 2694 GG C TGA gtg AGTTCA Consensus DR3 type VDRE (R: A or G, K: G or T, S: C or G) RGKTSA nnn RGKTSA mOPN GGTTCA cga GGTTCA hOC GGGTGA acg GGG G CA Putative VDREs in GADD45 genome A 659 GGGTCA tgg GGG GTG B 1482 GGGTCA gga GGGTG G C 2129 G T TTCA ctc AGGTCA D 2233 GGTTG C atg GGTTCA E 2694 GG C TGA gtg AGTTCA B VDR RXR RXR Ab Cold hOC hOC A B C D E Putative VDREs VDR/RXR VDR/RXR Ab Free probe VDR/RXR Free probe + + + + + + + + + + + + + + + + A Exon4 2462~3122 Exon3 1151~1388 Exon1 1~339 Exon2 826~927 promoter A B D E C 2256 Putative VDRE 296 ATG 2573 3 UTR A Exon4 2462~3122 Exon3 1151~1388 Exon1 1~339 Exon2 826~927 promoter A B D E C 2256 Putative VDRE 296 ATG 2573 3 UTR
47 Fig. 9. Multiple putative VDREs are present in GADD45 genome, which interact with VDR/RXR in vitro VDR RXR Cold hOC VDRE C RXR Ab M2 C VDRE A + + + + + + + + + + + + + + + + VDRE B VDRE D VDRE E VDR/RXR VDR/RXR Ab Free probe VDR/RXR VDR/RXR Ab Free probe Free probe VDR/RXR VDR/RXR Ab VDR/RXR VDR/RXR Ab Free probe VDR RXR Cold hOC VDRE C RXR Ab M2 C VDRE A + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + VDRE B VDRE D VDRE E VDR/RXR VDR/RXR Ab Free probe VDR/RXR VDR/RXR Ab Free probe Free probe VDR/RXR VDR/RXR Ab VDR/RXR VDR/RXR Ab Free probe
48 Fig. 9. Multiple putative VDREs are present in GADD45 genome, which interact with VDR/RXR in vitro ( A ) Schematic representation of the human GADD45 genome and the position of the putative VDREs. The sequences of consensus DR3, mo use OPN (mOPN) VDRE, human OC (hOC) VDRE and the five putative GADD45 VDREs are listed. Hexameric VDRE half sites are shown in bold capital letters. The 3 bp space is shown in small letters. Deviations from the consensus sequence RGKTSA are underlined. ( B ) In vitro interaction of VDR/RXR heterodimer with putative GADD45 VDREs. EMSAs were performed in the presence of 10 7 M 1,25VD using hOC (upper panel) or putative GADD45 (lower panel) VDRE probes. Pre incubation with 2 m g anti RXR b (RXR Ab) or 100 fold mol ar excess of cold hOC VDRE (Cold hOC) was performed for the super shift and competition experiments, respectively. ( C ) Specificity of the interaction between VDR/RXR heterodimer and the putative GADD45 VDREs. EMSAs were performed as in panel B. Specificity of the interaction was demonstrated by competition with 100 fold molar excess of unlabeled hOC VDRE oligos as a specific competitor and the lack of competition with VDRE C oligos as nonspecific competitor. 2 m g anti Flag M2 monoclonal antibody was used as a non specific antibody control for the super shifting with anti RXR antibody.
49 Fig. 10 1,25VD induces GADD45 reporter activity through endogenous receptors in OCa cells 0 2 4 6 8 10 ETOH VD A VDRE SV40 promoter A B D E 366 2926 Luc 0 1 2 3 4 ETOH VD 10 -9 M VD 10 -8 M VD 10 -7 M 0 2 4 6 8 10 EOH VD D B C Fold Induction #1 #2 E RLU x 10 6 RLU x 10 6 + + + + VDR RXR RLU x 10 6 0 2 4 6 8 ETOH VD 12 h VD 24 h VD 36 h
50 Fig. 10 1,25VD induces GADD45 reporter activity through endogenous receptor s in OCa cells (A) Schematic representation of GADDLuc construct. (B) Dose dependent induction of GADD45 VDRE reporter activity by 1,25VD. OVCAR3 cells were transfected with 0.2 m g GADDLuc, 0.05 m g pCMVgal, 0.05 m g p91023B VDR and 0.05 m g pCMX RXR b and tr eated with ETOH or 1,25VD at the indicated concentrations. Luciferase activity was determined and normalized with cognate b gal activity. Each data point was analyzed in duplicate and reproduced three times. ( C ) Time dependent induction of GADDLuc lucifer ase activity by 1,25VD. OVCAR3 cells were transfected as in panel A and treated with ETOH or 10 7 M 1,25VD for the indicated times. Luciferase activity was determined as in panel B. ( D ) Induction of reporter activity by 1,25VD in cells stably transfected with the GADDLuc. OVCAR3 cells stably transfected with GADD45 reporter were transfected with 0.05 m g pCMVgal, 0.05 m g p91023B VDR and 0.05 m g pCMX RXR b and treated with ETOH or 10 7 M 1,25VD for 36 h. Luciferase activity from two stable clones was determin ed and shown as fold induction. ( E ) Endogenous VDR/RXR in OVCAR3 cells is sufficient for 1,25VD induction of GADD45 reporter. OVCAR3 cells were transfected with GADDLuc and pCMVgal with or without p91023B VDR (VDR) and pCMX RXR b (RXR). The total amount of plasmid DNA was balanced with empty vectors. Luciferase activity was determined as in panel B.
51 This induction was also detected in OVCAR3 cells in which the reporter was stably integrated into the genome (Fig. 10D), showing that it is not an artifact of the transient transfection. Expression of additional VDR, RXR or both did not further increase the induction in OVCAR3 cells (Fig. 10E), suggesting that endogenous RXR and VDR in OVCAR3 cells are sufficient. To test whether 1,25VD regulates the GADD45 reporter through VDR, the activation of the reporter was tested in HeLa cells that lack functional VDR. As shown in Figure 11, panel A, 1,25VD did not induce the activity of a known VDR reporter, p23, that was constructed with the promoter of 24 hydroxylas e (Arbour, 1998). The induction of the p23 reporter was restored by the ectopic expression of VDR in HeLa cells. Similar to the p23 reporter, 1,25VD did not cause measurable induction of luciferase activity in the GADD45 reporter in HeLa cells (Fig. 11B). Co transfection with VDR restored the 1,25VD induction in a manner that is dependent on the dose of the transfected receptor (Fig. 11B). Transfection with RXR alone did not affect reporter activity but its co transfection with VDR enhanced the induction co mpared to cells transfected with VDR alone. Similar to the results with OVCAR3 cells, 1,25VD induced reporter activity in HeLa cells in a dose and time dependent manner after receptor transfection (data not shown). These experiments demonstrate that the r egulation of the reporter is VDR dependent and involves the RXR. To determine which of the putative VDREs is functional, OVCAR3 cells were transfected with reporter constructs where VDREs, either individually or in
52 Fig. 11. Induction of GADD45 report er activity by 1,25VD is VDR dependent ( A ) The lack of functional VDR in Hela cells. Cells were transfected with 0.2 ? m g p23 and pCMVgal with or without p91023B VDR. Transfected cells were treated and luciferase activity was determined as in Fig. 10B. ( B ) VDR dependent induction of GADD45 report er activity by 1,25VD. Hela cells were transfected with GADDLuc and pCMVgal with the indicated amounts of p91023B VDR, pCMX RXR b or both. The total amount of plasmid DNA was balanced with empty vectors. Luciferase activity was determined as in Fig. 10B. 0 2 4 6 8 10 ETOH VD VDR + A RLU x 10 5 B 0 2 4 6 8 10 ETOH VD 4 40 4 40 4 4 40 40 VDR ( ng ) RXR ( ng ) 4 40 4 40 4 4 40 40 VDR ( ng ) RXR ( ng ) RLU x 10 6
53 combination, were deleted or mutated by altering key nucleotides known to be essential for receptor interaction. As shown in Figure 12, panel A, deletion of the VDRE A or VDRE A plus VDRE B region did not affect 1,25VD induction. Deletion into the VDRE D region caused a significant decrease while further deletion into the VDRE E region eliminated 1,25VD induction. These analyses suggest that regions around VDRE D and VDRE E are essential for 1,25VD regulation of GADD45. Site directed mutation in VDRE E el iminated the induction, while single or multiple mutations of key nucleotides in the intronic VDREs had no effect. This suggests that only VDRE E is essential for the induction. This conclusion is consistent with the lack of VD induction in mutant reporter s containing only VDRE D or both VDRE A and VDRE B (Fig. 12A). It is also consistent with the induction of the reporter that contains only VDRE E (Luc3) and with the loss of this induction by site directed mutation of VDRE E (MTELuc3) (Fig. 12A). Individu al mutation of the first two VDREs or combined mutation of all first three VDREs actually increased VD induction, indicating that some VDREs may function in a negative fashion. The deletion of the VDRE D region, not the mutation of the VDRE D sequence, cau sed a decrease in VD induction suggesting that DNA elements in the VDRE D region for transcription factors other than VDR may cooperate with VDR/RXR binding to VDRE E to mediate the up regulation of GADD45. The lack of a VD effect on the activity of pGL3 b asic, pGL3 promoter and the pGL3 control vectors showed that the regulation of the reporters by 1,25VD is specific to the GADD45 sequence.
54 Fig. 12. The VDRE in the fourth exon of GADD45 genome is the functional VDRE that mediates the transcriptional induction of GADD45 by 1,25VD in OCa cells Mock Input Rat IgG VDR VDRE E Promoter 1 Promoter 2 B + + + VD Mock Input Rat IgG VDR VDRE E Promoter 1 Promoter 2 B + + + VD 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc A 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc Luc Luc A 0 2 4 6 8 10 12 14 16 E TO H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc Luc Luc A 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc 0 2 4 6 8 10 12 14 16 E O H V D 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc Luc Luc A 0 2 4 6 8 10 12 14 16 E O H V D 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc Luc Luc 0 2 4 6 8 10 12 14 16 E O H V D 0 2 4 6 8 10 12 14 16 E O H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc Luc Luc A 0 2 4 6 8 10 12 14 16 E TO H V D 0 2 4 6 8 10 12 14 16 E TO H V D Wild Type (WT) Putative VDRE Mutant (MT) Putative VDRE Luc Luc A B D E 366 749 Luc 1718 RLU x 10 6 Luc Luc Luc Luc SV40 Enhancer Luc Luc Luc Luc Luc Luc 366 1723 1718 2489 Luc 1718 2486 Luc SV40 promoter 2486 MTELuc2 basic promoter control MTELuc1 MTALuc MTBLuc MTDLuc MTELuc MTABDLuc Luc2 Luc4 Luc3 Luc1 Luc5 MTELuc3 GADDLuc 749 366 366 366 366 366 Luc Luc Luc Luc Luc Luc A Luc Luc Luc A
55 Fig. 12. The VDRE in the fourth exon of GADD45 genome is the functional VDRE that mediates the transcriptional induction of GADD45 by 1,25VD in OCa cells ( A ) Mutational analysis of the GADD45 reporter. OVCAR3 cells were transfected with 0.2 m g of the reporter constructs together with pCMVgal, p91023B VDR and pCMX RXR b and treated with ETOH or 10 7 M 1,25VD for 36 h. Luciferase activity was determined as in Fig. 10B and shown on the right. Schematic representation of the different reporter con structs is shown on the left. pGL3 basic, pGL3 promoter and pGL3 control vectors were used as controls. ( B ) ChIP assays. Soluble chromatin was prepared from OVCAR3 cells treated with ETOH or 10 7 M 1,25VD for 60 min. ChIP assays were performed with contro l (rat IgG) or anti VDR antibody. The mock control is performed with immunoprecipitates from buffer that contains no soluble chromatin from OVCAR3 cells.
56 To demonstrate that the VDRE E interacts with VDR in vivo OVCAR3 cells were treated with or without 1,25VD and ChIP assays were performed with anti VDR antibodies (Fig. 12B). From soluble chromatin prepared from OVCAR3 cells treated with 1,25VD, the anti VDR antibody precipitated GADD45 DNA fragments containing the VDRE E but not the promoter regions. T he specificity of the ChIP assay was demonstrated by the lack of VDRE E DNA in the mock as well as in the immunoprecipitates of rat IgG control. The data shows that VDR is recruited to the VDRE E in vivo. More importantly, the recruitment of VDR to the r esponse element is apparently ligand dependent since the anti VDR antibody did not precipitate GADD45 DNA fragments from soluble chromatin prepared from cells treated with the vehicle. 3. Upregulation of GADD45 protein is required for the hormone induced cell cycle arrest at the G2/M but not the G1/S checkpoint Our studies have established GADD45 as one of the immediate early response genes for 1,25VD, but questions remain whether regulation at the RNA level extends to the protein level and whether GADD4 5 mediates the growth suppressing activity of 1,25VD in OCa cells. To address these questions, OVCAR3 cells were treated with 1,25VD for up to 6 days and GADD45 protein expression was examined by immunoblotting. As shown in Figure 13, panel A, 1,25VD incre ased the level of GADD45 protein in a time dependent manner in OVCAR3 cells whereas the level of b actin remained constant during treatment. Compared to the data on GADD45 mRNA (Fig. 8B), 1,25VD induced accumulation of GADD45 protein is a much slower proce ss. This suggests the presence of additional regulations for GADD45 expression at post
57 transcriptional steps. The slower accumulation of GADD45 protein in response to 1,25VD may explain why the hormonal effect on the cell cycle requires longer treatment. To test whether GADD45 mediates the inhibitory effect of 1,25VD on OCa cell cycle progression, OVCAR3 cells were stably transfected with an expression vector containing GADD45 cDNA in the anti sense orientation (Zhan, 1994) or an empty vector. Immunoblot analysis of GADD45 was used to select stable clones with significantly reduced level of GADD45 protein by the anti sense GADD45 as compared to the control clones (Fig. 13B). Although the anti sense GADD45 did not completely eliminate GADD45 protein (Fig. 1 3B), it decreased the expression of GADD45 protein in the presence of 1,25VD to a level lower than basal level of control clones (Fig. 13B). In other words, the anti sense clones represent a functional knock out of GADD45 in terms of VD induction. Flow c ytometry showed that a decrease of GADD45 protein in the anti sense clones was associated with an increase of the proportion of cells in G2/M phase and a decrease of those in S phase (Figs. 13C and 13D). In the control clones, 1,25VD decreased the percenta ge of cells in S phase and increased the cells in G0/G1 and G2/M, showing that the hormone induced a similar cell cycle arrest at both G1/S and G2/M transitions similar to that in the parental OVCAR3 cells (Figs.5B and 5C). In the anti sense clones, 1,25VD induced G2/M accumulation was blocked whereas 1,25VD induced decrease in S phase and increase in G0/G1 still occurred (Figs. 13C and 13D). The data strongly suggest that GADD45 mediates the
58 Fig. 13. Anti sense GADD45 blocks 1,25VD induced cell cycle arrest at G2/M, but not G1/S checkpoint. A Vector OVCAR3 C AS45 OVCAR3 ETOH VD 60 70 80 90 ETOH VD G0/G1 0 10 20 30 ETOH VD S D 0 5 10 15 20 25 ETOH VD Vector #1 #2 OVCAR3 AS45 OVCAR3 Percentage of cells (%) G2/M # * * * # B + + + VD b actin GADD45 Vector #1 #2 OVCAR3 AS45 OVCAR3 GADD45 b actin OVCAR3 pCMV45 Hela + 0 1 3 6 VD(days)
59 Fig. 13. Anti sense GADD45 blocks 1,25VD induced cell cycle arrest at G2/M, but not G1/S checkpoint. ( A ) Induction of GADD45 protein expression by 1,25VD in VD sensitive human ovarian cancer cells. OVCAR3 cells we re treated with ETOH or 10 7 M 1,25VD for indicated times and the level of GADD45 protein was analyzed by immunoblotting. Hela cells transfected with pCMV45 plasmid was included as a positive control. b Actin was used to show the equal amount of total prot ein is present in each lane. ( B ) Suppression of GADD45 protein expression by stable expression of GADD45 anti sense cDNA. OVCAR3 cells stably transfected with control vector (Vector OVCAR3) or the anti sense cDNA of GADD45 (AS45 OVCAR3) were treated with E TOH or 10 7 M 1,25VD for 24 h. The level of GADD45 and b actin protein was determined as (A). ( C ) Abrogation of 1,25VD induced G2/M arrest in AS45 OVCAR3 clones. Vector OVCAR3 and AS45 OVCAR3 clones were treated with ETOH or 10 7 M 1,25VD for 9 days. Cell cycle distribution was determined by flow cytometry. Three independent experiments were performed and the profile of a representative experiment is shown. G2/M peak is indicated by arrow. ( D ) Bar graphs show percentage of cells at G2/M, G0/G1, and S phas es. Each data point was analyzed in duplicate, *P<0.05; #P>0.05 (versus ETOH treatment).
60 inhibitory effect of 1,25VD on the G2/M transition in OVCAR3 cells. Similar to the data in colon cancer cells (Wang, 1999), the two untreated anti sense GADD45 clone s had slightly increased G0/G1 and G2/M fractions and slightly lower S fractions, compared with vector control OVCAR3 cells. To firmly establish GADD45 as the mediator for the 1,25VD induced cell cycle arrest at the G2/M transition, we tested the effect of 1,25VD on the cell cycle progression of mouse embryo fibroblasts (MEFs) established from either wild type or GADD45 null mice (Hollander, 1999). Both MEFs expressed similar levels of VDR protein as determined on immunoblots (Fig. 14A). It is known that t he MEFs from this strain of mice are a mixture of diploid and tetraploid cells (Hollander, 1999). This complicates the flow cytometry analysis of diploid cells, largely due to the inability to distinguish diploid cells at G2/M from tetraploid cells at G0/G 1 phases. Therefore, we compared the changes induced by 1,25VD in the cell cycle distribution of tetraploid cells. As shown in Figure 14, panels B and C, 1,25VD induced a consistent increase in the percentage of cells in G2/M and a decrease of that in S ph ase of wild type MEFs, although the magnitude of the response was less than in OVCAR3 cells. In GADD45 null MEFs, no induction of G2/M arrest by 1,25VD was observed, confirming the conclusion reached in OVCAR3 cells with the anti sense approach that GADD45 is required for 1,25VD induced cell cycle arrest at the G2/M transition. In the wild type and GADD45 null MEFs, 1,25VD did not induce G1/S arrest (data not shown). Since G1/S arrest by 1,25VD was observed in the MEFs derived from another strain of mice (u npublished data for a separate
61 Fig. 14. 1,25VD induces G2/M arrest in wild type but not in GADD45 null MEFs. A VDR b actin WT Null C 0 5 10 15 20 25 ETOH VD Percentage of Cells in G2/M WT Null WT VD ETOH null B WT VD ETOH null B
62 Fig. 14. 1,25VD induces G2/M arrest in wild type but not in GADD45 null MEFs. ( A ) Expression of VDR protein in wild type (WT) and GADD45 nu ll MEFs. The level of VDR protein was determined by immunoblotting with anti VDR antibody and equal loading was shown by immunoblotting with anti b actin antibody. ( B ) Induction of G2/M arrest in WT but not GADD45 null MEFs by 1,25VD. WT and GADD45 null ME Fs were treated with ETOH or 10 7 M 1,25VD for 9 days. The cells were subjected to flow cytometry analysis. Three independent experiments were performed and the profile of a representative experiment is shown. The G2/M peak for tetraploid cells is indicate d by the arrow. ( C ) Bar graph show the percentage of tetraploid cells at G2/M. Each data point was analyzed in duplicate *P<0.01 (versus ETOH treatment).
63 Fig. 15 1,25VD decreases cdc2 kinase activity in OVCAR3 transfected with control vector but not in cells stably transfected with the anti sense cDNA of GADD45. Vector OVCAR3 and AS45 OVCAR3 cells were treated with ETOH or 10 7 M 1,25VD (VD) for 9 days. Cellular extracts were immunoprecipitated with anti cyclin B1 antibody. The activity of cdc2 was assayed using Histone H1 as a substrate. The level of cdc2 and cyclin B1 protein was determined by immunoblotting. b Actin blot was included to show the equal loading. VD + + + Histone H1 cdc2 cyclinB1 b actin Histone H1 cdc2 cyclinB1 b actin Vector #1 #2 OVCAR3 AS45 OVCAR3
64 study in our lab), the response to 1,25VD appea rs to vary among MEFs from different strains. 4. GADD45 mediates 1,25VD induced decrease in cdc2 kinase activity in OCa cells It is well established that the G2/M transition of mammalian cells is controlled by the M phase promoting factor, a heterodimeric complex between cdc2 and cyclin B. Studies in recent years suggest that GADD45 may directly regulate the activation of cdc2 kinase (Zhan, 1999). Therefore, the effect of 1,25VD on cdc2 kinase activity in control and GADD45 anti sense clones was measured by in vitro immunocomplex kinase assays. As shown in Figure 15, 1,25VD decreased the kinase activity of cdc2 in the control clone. In the GADD45 anti sense clones, there is an increase in cdc2 kinase activity (Fig. 15), presumably due to the decrease in the level of GADD45 protein. More importantly, 1,25VD did not decrease the activity to a level below the basal activity of control clones. Immunoblot analysis showed that 1,25VD decreased the level of cyclin B protein and that the decrease was not observed in the anti sense clones (Fig. 15), suggesting that GADD45 mediates the VD effect on cdc2 activity by regulating the cyclin B level in OCa cells. The data clearly suggest that G2/M arrest induced by 1,25VD in OCa cells is mediated through GADD45 and the subse quent decrease in cdc2 kinase activity. Collectively, our data demonstrated that G2/M arrest by 1,25VD in OCa was mediated through the induction of GADD45 via an exonic enhancer. Next, to further demonstrate the mechanism of 1,25VD induced growth inhibiti on, we investigated the regulation of telomerase by 1,25VD.
65 1,25VD induced apoptosis is mediated by destabilization of hTERT mRNA and a decrease in telomerase activity Telomerase activity is increased in most human cancers. It has been shown that 1,2 5VD reduced telomerase activity in HL 60 human acute myeloblastic leukemia cells correlated with 1,25VD induced differentiation (Xu, 1996). In an effort to understand the mechanism underlying 1,25VD induced apoptosis, our study provides evidence that telom erase is decreased by 1,25VD and its decrease is responsible for the apoptosis induced by 1,25VD. 1. 1,25VD down regulates telomerase activity in OCa cells Cells undergoing apoptosis are associated with the decrease in telomerase activity ( Saretzki, 200 3 ). As shown in Figure 6, apoptosis was induced more than 3 fold in OVCAR3 cells by 1.25VD. To fully understand the role of 1,25VD in OCa cells and determine telomerase function in 1,25VD induced apoptosis, telomerase activity was determined in OVCAR3 cell s using PCR based telomeric repeat amplification protocol (PCR TRAP) assay OVCAR3 cells constitutively expressed significant telomerase activity and 1,25VD decreased telomerase activity in a time dependent manner (Fig. 16). PCR TRAP assay performed with various ammounts of protein indicated that 200 ng of protein extract provided a quantitative and reproducible assay for telomerase activity on OVCAR3 cells (data not shown). The activity was not detectable when the extract was heated to 65 o C to abolish the enzymatic activity of protein components of telomerase. These results confirmed the specificity of the telomerase signal measured in this assay. Within 3 days of 1,25VD treatment, telomerase activity remained ~ 50% of the control, whereas at day 6 and day 9
66 Fig. 16. 1,25VD down regulates telomerase activity in OCa cells. The OVCAR3 cells were treated with 10 7 M 1,25VD for the indicated times. The protein was extracted and subjected to the PCR TRAP assay, as described in Materials and Metho ds 0 1 3 6 9 Days of treatment Telomerase Activity OD450 x 10 0.0 0.2 0.4 0.6 0.8 1.0 1.2
67 Fig. 17. 1,25VD down regulates hTERT mRNA expression in OCa cells. hTERT GAPDH 1uM 200nM CHX 0 6 0 6 0 6 VD + pBABhTERT A C hTERT GAPDH B 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 3 9 Days of treatment 0 1 3 6 9 Days of treatment
68 Fig. 17. 1,25VD down regulates hTERT mRNA expression in OCa cells. Dose dependent inhibition of hTERT mRNA by 1,25VD. Total RNA was isolated from OVCAR3 cells treated with ETOH or 10 7 M 1,25VD for indicated times. 2 m g RNA was used for RT PCR analysis ( A ) and Real time PCR analysis ( B ) as described in Materials and Methods. ( C ) Effect of cycloheximide (CHX) on the inhibition of hTERT mRNA by 1,25VD. OVCAR3 cells were exposed to E TOH or 10 7 M 1,25VD for 6 days in the absence or presence of CHX at the indicated concentrations. RT PCR was performed as shown on panel A. RNA purified from pBABhTERT transfected OVCAR3 cells as positive control. RT reaction without reverse transcriptase was the negative control.
69 telomerase activity fell to only 15% of the control cell. Combined with data in Figure 5, these experiments showed that inhibition of telomerase activity by 1,25VD was correlated with the growth inhibition and apoptosis induced by 1,25VD. These data support ed that 1,25VD induced apoptosis was associated with decreased telomerase activity. 2. hTERT is a VD targeted gene in OCa cells To investigate the mechanism of VD action on telomerase, RT PCR was used to determine the mRNA level of hTERT. Compared to cells treated with vehicle, 10 7 M 1,25VD decreased mRNA levels of hTERT in a time dependent manner (Fig. 17A). Consistent with decreased telomerase activity by 1,25VD, inhibition began after 1 day of 1,25VD treatment, with further decrease detected after treatment for 3 6 and 9 days (Fig. 17A). The mRNA levels of GAPDH were not affected by 1,25VD, showing some specificity for 1,25VD. In order to quantitatively determine the decrease of hTERT mRNA, real time PCR wa s performed using total RNA purified from OVCAR3 cells treated with vehicle or 10 7 M 1,25VD. The results confirmed the decrease of hTERT mRNA after 3 days of 1,25VD treatment and further decrease after 9 days of treatment (Fig. 17B). These data indicate t hat the mechanism of telomerase down regulation involves the change in the mRNA of hTERT. The down regulation of hTERT mRNA by 1,25VD persisted in the presence of cycloheximide, an inhibitor of protein synthesis (Fig. 17C), suggesting that hTERT inhibition by 1,25VD does not require new protein synthesis, and that hTERT is a primary responsive gene for 1,25VD in OCa cells.
70 Fig. 18. Putative hTERT VDRE specifically binds to VDR/RXR heterodimer. EMSAs were performed in the presence of 10 7 M 1,25VD using putative hTERT VDRE probes. Pre incubation with 2 m g anti RXR b (RXR Ab) or 100 fold molar excess of cold hOC VDRE (Cold hOC) was performed for the super shift and competition experiments, respectively. Specificity of the interaction was demo nstrated by competition with 100 fold molar excess of unlabeled hOC VDRE oligos as a specific competitor and the lack of competition with Gadd45 VDRE C oligos as nonspecific competitor. 2 m g anti Flag M2 monoclonal antibody was used as a non specific antib ody control for the super shifting with anti RXR antibody. VD VDR RXR Cold hOC VDRE C RXR Ab M2 VDR/RXR VDR/RXR Ab Free probe + + + + + + + + + + + + + + + + + + + + + + +
71 Fig. 19. The putative VDRE is not a functional VDRE in OVCAR3 cells. The OVCAR3 cells were transfected with 0.2 m g pGL3 3328Luc ( A B ) together with 0.1 m g CMV b gal and 0.1 m g VDR. The trans fected cells were treated with 10 7 M 1,25VD, 10 6 M 9 cis RA or ETOH as indicated. Luciferase activity was determined and normalized by b gal activity. ( C ) 3328Luc was stably transfected into OVCAR3 cells. The 3328Luc OVCAR3 cells was treated with 10 7 M 1,25VD for 6 days and luciferase activity was determined. ( D ) ChIP assays. Soluble chromatin was prepared from OVCAR3 cells treated with ETOH or 10 7 M 1,25VD for 60 min. ChIP assays were performed with control (rat IgG) or anti VDR antibody. 0.0 0.5 1.0 1.5 2.0 2.5 0 2 4 6 8 10 12 ETOH VD ETOH 9 cis RA+ VD A B Luciferase Activity RLU x 10 6 Luciferase Activity RLU x 10 6 Luciferase Activity RLU x 10 5 C ETOH VD VDR Rat IgG Input + + + VD GADD45 VDRE E hTERT VDRE D 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 0 2 4 6 8 10 12 0 2 4 6 8 10 12 ETOH VD ETOH 9 cis RA+ VD A B Luciferase Activity RLU x 10 6 Luciferase Activity RLU x 10 6 Luciferase Activity RLU x 10 5 C ETOH VD VDR Rat IgG Input + + + VD GADD45 VDRE E hTERT VDRE D 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5
72 One putat ive DR3 type VDRE was identified in the hTERT promoter in a recent report ( Ikeda, 2003). To determine if this putative VDRE is functional in 1,25VD mediated inhibition of telomerase activity in OCa cells, EMSAs were performed and the results showed that th is VDRE bound recombinant VDR and RXR proteins (Fig. 18). The binding is specific for the VDR/RXR heterodimer since neither VDR nor RXR alone formed a detectable complex with the VDRE probes. The VDRE was displaced from the complexes with an excess amount of cold hOC VDRE but was not affected by the non specific sequence VDRE C. VDRE C has been shown not to bind VDR/RXR in our previous study (Fig. 9B). To determine whether binding the receptors to VDRE in vitro mediates the down regulation of hT ERT by 1,25VD in OVCAR3 cells, a reporter gene with a 3.3 kb promoter region of hTERT containing the putative VDRE located upstream of the cDNA of firefly luciferase gene was analyzed. In OVCAR3 cells transiently transfected with this reporter, 1,25VD did not decrease the luciferase activity (Fig. 19A). This is consistent with the study in prostate cancer cells showing that this VDRE mediated inhibition of telomerase activity required 1,25VD and 9 cis RA, not 1,25VD alone (Ikeda, 2003). However, in OVCAR3 c ells the combination of 1,25VD and 9 cis RA did not decrease the activity of telomerase reporter (Fig.19B). There is also no detectable decrease in reporter activity by prolonging the treatment of 1,25VD to 6 days in OVCAR3 cells in which the reporter was stably integrated into the genome (Fig. 19C). To find out if the putative VDRE interacts with VDR in vivo ChIP assays were performed with anti VDR antibodies. In the soluble chromatin prepared from
73 OVCAR3 cells treated with 1,25VD, the anti VDR antib ody precipitated the GADD45 exonic region containing the functional VDRE (Fig. 12B), but could not precipitate hTERT promoter fragments containing the putative VDRE (Fig. 19D). These data indicate that VDR is not recruited to the putative VDRE in vivo. I t is concluded that in OVCAR3 cells the putative VDRE in hTERT promoter is not a functional VDRE. 3. The stability of hTERT mRNA is decreased by 1,25VD The down regulation of hTERT mRNA after 1,25VD treatment could be due to the sho rt half life of the hTERT message. To examine whether 1,25VD mediated changes in hTERT mRNA stability contribute to its decreased expression, we performed Real time PCR analysis using RNA from actinomycin D treated cells. After OVCAR3 cells were treated fo r 3 days with 1,25VD or vehicle, transcription was inhibited by adding actinomycin D. At different time intervals, total RNA was isolated and hTERT mRNA was determined by Real time PCR analysis. As shown in Figure 20 A, the level of hTERT mRNA, after norma lization with GAPDH, decreased within approximately 10 hr in 1,25VD treated cells, whereas hTERT mRNA from vehicle treated cells was comparatively stable over this period of time. The half life of hTERT mRNA in 1,25VD treated and control cells is approcim ately10 hr and 29 hr, respectively. The same results were obtained using a sencond probe and primers designed for another region of hTERT cDNA (Fig. 20B). The rate of degradation of the hTERT mRNA was increased by the addition of 1,25VD, which demonstrates that the decrease of hTERT mRNA by 1,25VD is due to its decreased stability.
74 Fig. 20. The stability of hTERT mRNA is decreased by 1,25VD. OVCAR3 cells were treated with 10 7 M 1,25VD (VD) or ETOH for 3 days, followed by treatment with 5 m g/ml actin omycin D. Total RNA was extracted at 0, 2, 8, 12 hrs and subjected to Real time PCR using probe 1 ( A ) and probe 2 ( B ), as described in Meterials and Methods. 0 2 4 6 8 10 12 14 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 ETOH VD Probe 1 0 2 4 6 8 10 12 14 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 ETOH VD Probe 2 A B
75 4. Telomerase stably transfected cells have increased telomerase activity and prolonged telomer e length. Ectopic expression of hTERT in telomerase negative human fibroblasts and endothelial cells resulted in substantial telomerase activity and telomere maintenance. Furthermore, the ectopic expression of hTERT circumvented senescence and enabled thes e cells to be immortalized, i.e, proliferate indefinitely in culture (Hahn, 1999). To determine the role of telomerase in 1,25VD induced growth inhibition in OVCAR3 cells, we tested whether the overexpression of telomerase was sufficient to abolish 1,25VD induced growth inhibition. hTERT cDNA, under the control of a viral promoter insensitive to 1,25VD, was stably transfected into OVCAR3 cells. As shown in Figure 21A, the telomerase OVCAR3 cell lines express 2 4 folds higher levels of telomerase activity th an parental OVCAR3 cells. Telomere length in OVCAR3 cells is ~3 kb as shown in Figure 21B. This is consistent with results from previous studies showing that telomere teminal restriction fragments (TRF) are ~3 kb in ovarian carcinoma cells (Villa, 2000; B raunstein, 2001). It has been known that the majority of human cancers have much shorter telomeres than the corresponding normal tissues; in many cases approaching that is associated with crisis in normal cells (2 4 kb) (Wynford Thomas, 1999; Liu, 1999). Telomerase plays an important role in maintaining stable telomere length. Telomere length in telomerase OVCAR3 clones was prolonged to 6 10 kb, as shown in Figure 21B. The ectopic expression of hTERT in OVCAR3 cells dramatically increased telomerase acti vity and prolonged telomere length.
76 Fig. 21. Telomerase stably transfected cells have increased telomerase activity and prolonged telomere length. ( A ) Proteins were extracted from OVCAR3 cells and telomerase OVCAR3 cells and subjected to PCR TRAP assay, as described in Materials and Methods. ( B ) Genomic DNA extracted from OVCAR3 and Telomerase OVCAR3 were used to measure telomere length. Southern blot hybridization was performed as described in Materials and Methods. 0 1 2 3 4 5 Fold induction A #1 #2 #3 OVCAR3 Telomerase OVCAR3 21.2kb 6.1kb 5.0kb 3.6kb 2.7kb 1.9kb 4.2kb 8.6kb 7.4kb B #1 # 2 #3 OVCAR3 Telomerase OVCAR3
77 Fig. 22. Overexpre ssion of telomerase blocks 1,25VD induced down regulation of telomerase activity. ( A ) OVCAR3 cells and telomerase clones were treated with ETOH or 10 7 M 1,25VD for 6 days (6 d) or 9 days (9 d). The protein was extracted and subjected to PCR TRAP assay, a s described in Materials and Methods. ( B ) Genomic DNA extracted fr om OVCAR3 and Telomerase OVCAR3 treated with 10 7 M 1,25VD (V) or ETOH (E) for 9 days to measure telomere length. Southern blot hybridization was performed, as described in Materials and Met hods. 0 1 2 3 4 5 ETOH VD 6d VD 9d #1 #2 OVCAR3 Telomerase OVCAR3 A Fold induction B E V E V E V #1 #2 OVCAR3 Telomerase OVCAR3
78 In contrast to parental cells, no dramatic decrease of telomerase activity or telomere length was observed in telomerase OVCAR3 cells after treatment with 1,25VD for 6 or 9 days compared with vehicle treatment (Fig. 22 A and B). Cells with ectopic h TERT expression maintain high telomerase activity after 1,25VD treatment compared with parental OVCAR3 cells (Fig. 22A). These results demonstrate that ectopic expression of hTERT is sufficient to protect 1,25VD inhibition of telomerase activity. Therefore these cell lines can be used to determine whether telomerase plays any role with respect to 1,25VD. Although 1,25VD dramatically decreased telomerase activity, no remarkable loss of telomeric repeats was detected after 9 days of treatment with 1,25VD (F ig. 22B). No decrease in telomere length could be due to the fact that cells undergo only a few doublings in 9 days, the potential base pair loss in 9 days are estimated to be ~200 bp, the loss of telomeric repeats may not be measurable by conventional Sou thern Blotting. A recent study demonstrated that telomere dysfunction rather than the mean telomere length was modulated when glutathione (GSH) levels influenced the c myc dependent cell death (Biroccio, 2003). Telomerase and telomere structure are dynamic ally regulated in normal human cells and disruption of telomerse activity alters the maintenance of the 3 single stranded telomeric overhang without changing the rate of overall telomere shortening (Masutomi, 2003), suggesting that modulating telomere int egrity maybe essential regardless of telomere length. It is highly possible that although maintaining telomere length is important, 1,25VD down regulated telomerase activity could regulate OCa cell growth through modulation of telomere integrity.
79 5. Overex pression of telomerase partially blocks 1,25VD induced apoptosis and increases the ability to recover after 1,25VD withdrawal. To address the question if telomerase participates role in growth suppressing activity of 1,25VD in OCa cells, we compared the re sponse of telomerase OVCAR3 clones to 1,25VD to that of paretal OVCAR3 cells. The effect of 1,25VD on cell growth as well as the ability to recover from treatment after removal of the hormone was investigated. Similar to parental cells, growth of telomeras e OVCAR3 cells (Fig. 23) were also inhibited by 1,25VD, but to a lesser extent. Cells overexpressing hTERT keep growing in the presence of 1,25VD while the growth of parental cells halted after 6 days treatment with 1,25VD. Telomerase clones are more resis tant to 1,25VD induced growth inhibition. Additionally, after 9 days of 1,25VD treatment, the recovery of telomerase OVCAR3 cells was rapid compared to the recovery of the parental cell line (Fig. 24). Our observation suggests that ectopic expression of hT ERT in OCa cells protects about 50% of 1,25VD reduced growth, although other mechanisms independent of telomerase may also exist. Since overexpression of telomerase altered cell response to 1,25VD in the recovery assay, we next assessed whether alteration s in apoptotic index occurred. Parental and telomerase OVCAR3 cells were treated with 1,25VD and then analyzed by flow cytometry. As shown in Figure 25, in contrast to the massive apoptosis (70%) induced in parental cells after 9 days treatment with 1,25VD telomerase clones showed fewer apoptotic cells (<40%). The ectopic expression of hTERT partially rescued long term 1,25VD treated OVCAR3 cells from cell death. This may contribute to
80 Fig. 23. Telomerase OVCAR3 cells are less sensitive to g rowth inhibition by 1,25VD. OVCAR3 and telomerase OVCAR3 cells were plated in 96 wells and treated with 10 7 M 1,25VD (VD) or ETOH as a vehicle. Cell numbers were determined with the MTT assays. Eight samples were analyzed for each data point. *P<0.05 (ver sus ETOH treatment). 0 2 4 6 8 10 ETOH VD 0 2 4 6 8 10 ETOH VD 0 2 4 6 8 10 12 ETOH VD OVCAR3 Telomerase OVCAR3#1 Telomerase OVCAR3#2 0 3 6 9 Days of treatment *
81 Fig. 24. Telomerase OVCAR3 cells recover quickly from 1,25VD treatment. OVCAR3 and Telomerase OVCAR3 cells were pretreated with 10 7 M 1,25VD or ETOH for 9 days. Recovery of pretreated cells was assessed by plating pretreated cel ls in 96 well plates. Cell numbers were determined with the MTT assay. Eight samples were analyzed for each data point. OVCAR3 Telomerase OVCAR3 #1 0 1 2 3 4 5 6 7 0 2 4 6 8 10 0 2 4 6 8 10 12 14 VD pretreatment ETOH pretreatment 0 3 6 9 12 Days of treatment Telomerase OVCAR3 #2
82 Fig. 25. Overexpression of telomerase partially blocks 1,25VD induced apoptosis. OVCAR3 Telomerase OVCAR3#1 Telomerase OVCAR3#2 A Apoptotic cells (%) B 0 20 40 60 80 ETOH VD Exp #1 0 20 40 60 80 ETOH VD #1 #2 OVCAR3 hTERT OVCAR3 Exp #2
83 Fig. 25. Overexpression of telomerase partially blocks 1,25VD induced apoptosis. ( A ) OVCAR3 and Telomerase OVCAR3 clones were treated with ETOH or 10 7 M 1,25VD for 9 days. Apoptotic index was determined by flow cytometry and a representative profile is presented. ( B ) Bar graphs show percentage of apo ptotic cells in two experiments.
84 the reduced growth inhibition of telomerase clones by 1,25VD treatment (Fig. 23). Collectively, these studies show that in OVCAR3 cells down regulation of telomerase is correlated to 1,25VD induced growth inhibition and apoptosis. The enhancement of telomerase function in OVCAR3 cells, as measured by telomerase activity, allows the cells to recover from 1,25VD treatment and rescues them from 1,25VD induced apoptosis. Telomerase activity remains a key parameter that determ ines long term cell survival in OCa cells. This work supports our hypothesis that telomerase inhibition by 1,25VD may serve as an effective tool to eliminate OCa cells that have short telomeres. EB1089 is more potent than 1,25VD in suppressing OCa cell gr owth in vitro and in nude mice. The effective concentration of 1,25VD (10 7 M) is pharmacological, which is expected to induce hypercalcemia in vivo. EB1089, a 1,25VD analogue, is less calcemic and was shown to be more effective against breast (Colston, 19 92), pancreatic (Colston, 1997), colon (Akhter, 1996) and prostate (Blutt, 2000) cancers. To test whether EB1089 can be used for long term treatment and/or chemoprevention of OCa, our study investigated VDR expression in human ovarian tissues, the growth o f OVCAR3 cells and tumor xenografts in nude mice. In MTT assays, EB1089 is effective against OCa cells at concentrations 10 times lower than 1,25VD, making it a promising candidate for in vivo treatment of OCa (Fig. 26 A and B). As expected, EB1089 in duced GADD45 reporter activity effectively
85 Fig. 26. EB1089 is more potent than 1,25VD in suppressing OCa cell growth. OVCAR3 cells were plated in 96 well plates and treated 1,25VD (VD) ( A ) or EB1089 (EB) ( B ) at the indicated concentrations. Cell numbers were determined with MTT assay. Eight samples were analyzed for each data point, and the data were reproduced three times. Cell number X 10 3 0 1 2 3 4 ETOH VD 10 -9 M VD 10 -8 M VD 10 -7 M 0 9 Days of Treatment A Cell number X 10 3 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 ETOH EB 10 -9 M EB 10 -8 M EB 10 -7 M 0 9 Days of Treatment B
86 Fig. 27. EB1089 is more effective than 1,25VD in inducing GADD45 reporter activity. OVCAR3 cells were transfected wi th 0.2 m g GADDLuc, 0.05 m g pCMVgal, 0.05 m g p91023B VDR and 0.05 m g pCMX RXR b and treated with ETOH, 1,25VD or EB1089 at the indicated concentrations. Luciferase activity was determined and normalized with cognate b gal activity. 0 1 2 3 4 ETOH 10 -10 M 10 -9 M 10 -8 M 10 -7 M VD EB1089
87 at concentrations 10 ti mes lower than 1,25VD (Fig. 27), suggesting a mechanism of EB1089 action in OCa cells. To determine whether the effect of 1,25VD and EB1089 on OCa cell growth can be directly translated into tumor suppression in the whole animal, we tested the effect of EB 1089 in OVCAR3 tumor xenografts. As shown in Figure 28A, EB1089 at 1 m g/day/kg totally inhibited the growth of tumors while a dose of 0.3 m g/day/kg partially inhibited the growth. Based on statistical analysis of independent samples with the T test, the dif ference between placebo control (6 tumors) and 1.0 m g/day/kg (9 tumors) is significant (p=0.003). The difference between the 0.3 (6 tumors) and 1.0 m g/kg/day group is also significant (p=0.04). The difference between placebo control and 0.3 m g/kg was not s ignificant (p=0.281). This is probably due to the fact that the number of groups was small and the size of tumors was heterogeneous at the start of the study. In our experiments, each individual mouse was marked and the tumor volume was recorded separately ; the difference between 0.3 m g/kg/day and placebo is obvious when the growth of each individual tumor was examined separately. To determine if EB1089 is less calcemic, serum calcium levels were determined in treated nude mice (Fig. 28B). Blood tests at 30 days did not reveal significant increase in blood calcium. The levels in placebo (~9 mg/dl) and both of EB 1089 groups (9~10.5 mg/dl) were within the normal range (7 11 mg/dl). As expected, 1,25VD treatment for 1.0 m g/kg/day induced higher levels of ser um calcium (11~12 mg/dl at day 15 and day 30), which is outside the normal range. Importantly, no abnormality was seen in mice treated with EB1089 at 0.3 or 1.0 m g/kg/day for 4 weeks and normal body weight was maintained (data not shown).
88 Fig. 28. EB1089 inhibits OCa xenograft growth. 0% 50% 100% 150% 200% 250% 300% 350% 400% 450% 0 5 10 15 20 25 30 Days of treatment Tumor Growth Placebo 0.3 ug/Kg/Day 1 ug/Kg/Day A 0 2 4 6 8 10 12 14 Placebo EB1089 0.3 ug/Kg EB1089 1.0ug/Kg VD 0.3ug/Kg VD 1.0ug/Kg Calcium level (mg/dL) Day 0 Day 15 Day 30 B
89 Fig. 28. EB1089 inhibits OCa xenograft growth. ( A ) 2x10 6 OVCAR3 cells were injected s.c. into one site in the dorsal side of nude mice. After the tumors grew to 150 nm 3 mice were randomly divided into three groups a nd treated with placebo or EB1089 at indicated doses daily as described in Materials and Methods. The tomor size was measured every five days and the growth status of each tumor was recorded separately. *, p<0.01 compared with placebo group. ( B ) Average se rum calcium levels for each treatment group.
90 VDR and RXR are expressed in precursors for epithelial OCa and normal ovarian surface epithelial cells are responsive to 1,25VD. If 1,25VD is useful for OCa prevention, its receptors should be expressed in OSE cells and benign ovarian tumors, the precursors for malignant epithelial OCa, which gives rise to more than 90% of human OCa. Indeed, our immunohistochemical studies showed that both VDR and RXR are expressed strongly in human OSE cells and in benign tumor s. Skin tissues which express VDR and RXR were used as positive control and the primary antibody was substituted with IgG as negative control (Fig. 29). To test whether normal ovarian surface epithelial (OSE) cells respond to 1,25VDs growth suppression, primary human OSE cell cultures were treated with ETOH or 1,25VD and their growth was analyzed by the MTT assays. As shown in Figure 30, the growth of human OSE cells was suppressed by 1,25VD in a similar way as seen with OVCAR3 cells. These data clearly d emonstrate that human primary ovarian cells are sensitive to 1,25VD. Collectively, the expression of VDR in human ovary and the growth inhibition of human primary OSE by 1,25VD indicate that 1,25VD analogue could also be used for chemoprevention.
91 Fig. 29. Normal ovarian epithelial cells and benign adenomas express VDR and RXR. Paraffin sections of tissues were stained with anti VDR (rat monoclonal 9A7) or anti RXR (rabbit polyclonal) antibodies. The anti RXR antibody was generated against RXR a but also cross reacts with RXR b and RXR g IgG controls were performed with normal ovarian sections (rat IgG for VDR and rabbit serum for RXR). 2 2 1 1 4 4 3 3 5 5 6 6 7 7 8 8 Skin Normal Ovary Cystadenoma IgG Control VDR RXR
92 Fig. 30. 1,25VD suppresses human primary ovarian cell growth. Human primary ovarian epithelial cells were plated in 96 well plates and treated with 10 7 M 1,25VD (VD) or ETOH. Cell numbers were determined at the indicated times by the MTT assays. Eight samples were analyzed for each data point. P<0.05 (versus ETOH treatment). 0 5 10 15 20 25 30 ETOH VD * 0 3 6 9 Days of treatment MTT (OD 595 ) x 10
93 DISC USSION To demonstrate the role of 1,25VD and it analogues in OCa treatment and prevention, the mechanism by which 1,25VD mediated its antiproliferative activity was explored using a VD sensitive OCa cell line (OVCAR3) as a model system. The effect of the 1,25VD analogue EB1089 on OCa cell growth was also investigated in vitro and in vivo Cell cycle studies showed that 1,25VD increased the proportion of OVCAR3 cells in the G0/G1 and G2/M phase and decreased those in the S phase. 1,25VD also induced apopto sis in OVCAR3 cells. Furthermore, we have identified GADD45 and telomerase as target genes for 1,25VD and suggest a model of VD action in OCa cells (Fig. 31). 1,25VD induces binding of the VDR/RXR heterodimer to the VDRE located in the fourth exon of GADD 45 gene at a position downstream from the termination codon for protein translation. Presumably through the recruitment of co activator complexes, the activated VDR/RXR interacts with the Pol II complex bound to the promoter and increases the rate of GADD4 5 transcription. This leads to an increase in the amount of GADD45 protein that, through a yet unknown undefined mechanism, decreases the level of cyclin B, the regulatory subunit of cdc2 kinase. The resulting decrease in cdc2 activity is then responsible for disturbed cell cycle progression to M phase. In addition, 1,25VD inhibits
94 Fig. 31. Model for the integrated cellular pathways of 1,25VD action in OCa cells (See text for details). VDRE promoter GADD45 TFIIB 5 RNA pol II Co activator Complexes 3 GADD45 mRNA GADD45 protein cdc2 Cyclin B Cyclin B cdc2 G2 M hTERT 5 hTERT mRNA VD degradation hTERT activity 3 apoptosis x nucleus VD VD 3UTR RXR VDR Cyclin B Cyclin B promoter
95 telomerase activity by decreasing the stability of hTERT m RNA, not by the putative VDRE in the hTERT promoter which is not functional in vivo Importantly, hTERT clones are more resistant to 1,25VD induced apoptosis and growth inhibition. In contrast to parental cells that recover slowly from prolonged treatment with 1,25VD, hTERT clones re grow quickly after 1,25VD withdrawal. Lastly, our investigation of the antiproliferative effects of EB1089 on OVCAR3 cells and xenografts without inducing hypercalcemia may indicate a novel preventive and therapeutic option fo r the treatment of OCa. Upregulation of GADD45 by 1,25VD is mediated through a novel exonic enhancer Induction of GADD45 can be detected in cells within 2 hrs of treatment with 1,25VD (Fig. 8B). With the lack of a VD effect on mRNA stability (Fig. 8C) an d the identification of functional VDREs in the genome (Figs. 9 12), our data establish GADD45 as a primary and immediately early response gene for 1,25VD. Furthermore, our data suggest that GADD45 regulation is mediated through a VDR/RXR heterodimer, inst ead of a VDR/VDR homodimer. This conclusion is reached based on the inability of VDR to bind to all putative VDREs in EMSAs until the addition of RXR and the up shifting of the complexes by RXR antibody (Fig. 9). At variance with the belief that nuclear re ceptors forming heterodimers with RXRs bind DNA in the absence of ligand, our ChIP assays (Fig. 12B) clearly show that in vivo binding of VDR to the exonic VDRE is ligand dependent. Recently, Yamamoto et al (2003) reported a similar observation on VDRE lo cated in the promoter of the OPN gene, suggesting that the
96 ligand dependency is common to VDREs located inside and outside the promoter regions. Since p53 in OVCAR3 cells is mutated (Yaginuma, 1992), the regulation of GADD45 transcription by 1,25VD obvious ly occurs independently of p53 activity. This is consistent with the identification of the VDRE in the fourth exon that is distant from the p53 binding site which is located in the third intron (Hollander, 1993). This is also consistent with the conclusion reached by a study of squamous cell carcinoma (Prudencio, 2001). Furthermore, the VD regulation of GADD45 is also likely to be independent of BRCA1 since the DNA element for this tumor suppressor is located in the promoter region (Jin, 2000). It is striki ng that four of the five putative VDREs bound VDR/RXR equally well in EMSAs, but only VDRE E mediated the up regulation of GADD45 reporter by 1,25VD. The other putative VDREs are either nonfunctional or may act in a negative way. Overall, the data suggest that the regulation of GADD45 by 1,25VD is a complex process that may involve the interaction of the receptor bound to the VDRE and other transcription factors. The different VDREs may also function in a cell specific manner to mediate the regulation of GA DD45 expression by 1,25VD. GADD45 protein upregulation by 1,25VD is required for the hormone induced cell cycle arrest at the G2/M but not the G1/S checkpoint Our studies link GADD45 induction specifically to the inhibition by 1,25VD of cell cycle progression through the G2/M checkpoint. This linkage was established using cells in which GADD45 expression was compromised by the anti sense approach or genetic knock out. It is important to determine the effect of GADD45 anti sense on 1,25VD induc ed growth inhibition in the stably transfected cells, which would reveal
97 whether G2/M arrest is a major or minor factor in the overall growth inhibition. However, our analysis with MTT assay revealed little difference in the overall response to 1,25VD betw een anti sense GADD45 and control clones (data not shown). This may due to cell cycle arrest being shifted from G2/M to G1/S checkpoint in anti sense clones. Previous studies (Akutsu, 2001) have shown that GADD45 induction in squamous cell carcinoma by 1, 25VD is associated with an increased interaction with PCNA. Although we have not examined the association of GADD45 with PCNA, this seems unlikely in OVCAR3 cells. PCNA is required for DNA replication in S phase but our data show that G1/S arrest still occ urs in OVCAR3 cells expressing the anti sense cDNA of GADD45 (Fig. 13C and 13D). Instead of PCNA, our studies show a 1,25VD induced decrease in cdc2 activity, which is associated with a decrease in the level of cyclin B1 protein. The decrease was not obser ved in the stable clones expressing GADD45 anti sense cDNA (Fig. 15). The data suggest that GADD45 mediates the effect of 1,25VD on cdc2 activity. Our data concur with the proposed role of GADD45 in G2/M arrest induced by certain types of DNA damaging agen ts (Wang, 1999; Zhan, 1999). Zhan et al (1999) have shown that GADD45 inhibits the interaction between cyclin B and cdc2. Later, it was shown that GADD45 induced cell cycle arrest at G2/M is associated with an altered cellular distribution of cyclin B1 ( Jin, 2002), which seems to require a functional p53. Our conclusions, however, differ from the above studies since we detected a decrease in the level of cyclin B1 protein (Fig. 15). As mentioned earlier, OVCAR3 cells contain a mutant p53 (Yaginuma, 1992). Obviously, GADD45 exerts its effect on cdc2 activity and the G2/M transition in the OCa cells independently of p53. It remains to be determined whether the effect of 1,25VD on G2/M in OCa cells is exerted
98 through GADD45 alone or in combination with anothe r protein that functions similarly to p53. Besides its role in regulating G2/M transition, GADD45 plays an essential role in DNA repair and in the maintenance of genomic stability. MEFs derived from GADD45 null mice exhibit aneuploidy, chromosomal abnorma lities, gene amplification and centrosomal amplification (Hollander, 1999). GADD45 null mice display increased sensitivity to dimethylbenzanthracene induced carcinogenesis (Hollander, 2001). It is intriguing that dimethylbenzanthracene increased female ova rian tumors more efficiently in GADD45 null mice than wild type (Hollander, 2001). GADD45 induction by 1,25VD through VDR suggests that VDR may act in a p53 independent tumor suppressing pathway to affects ovarian genomic stability. Along this line, VDR ha s been shown to act as a bile acid sensor for the secondary bile acid lithocholic acid (LCA) (Makishima, 2002), a potential enteric carcinogen that induces DNA strand breaks, forms DNA adducts and inhibits DNA repair enzymes. It will be interesting to show whether ovarian carcinogens or DNA damaging processes (e.g. ovulation) activate VDR in ovarian epithelial cells and, if yes, whether the VD independent VDR activation functions in DNA repair. Down regulation of telomeras e by 1,25VD is mediated through de stabilization of hTERT mRNA Hormones regulate the expression of telomerase in hormone responsive cell systems. In some cases, induction or repression of telomerase has been considered a consequence of maturation and/or growth arrest, rather than a direct h ormonal effect. For example, androgens stimulate telomerase indirectly since the hTERT promoter construct
99 was not activated by androgen and transcription of the endogenous gene was not stimulated early enough in cultured cells to be considered a direct tar get of androgens. However, it has also been reported that inhibition of telomerase activity is an early event of the differentiation process in leukemia cells rather than its consequence (Albanell, 1996; Savoysky, 1996). Retinoids and VD negatively regula te telomerase that is associated with differentiation in leukemia cells. However, it is not yet clear at which level of gene expression this down regulation occurs. In addition, numerous studies showed that hormones directly regulate telomerase via differe nt mechanisms. For example, estrogens activates telomerase through the direct interaction of ligand activated ER with the ERE in the hTERT promoter. Progesterone regulates hTERT transcription via the MAP kinase cascade. There was a significant correlatio n of telomerase activity with hTERT mRNA expression but not with TP1 or hTR. hTERT is the major determinant of telomerase activity (Counter, 1998). In our study, the repression of telomerase activity by 1,25VD accompanied down regulation of hTERT mRNA as s hown by RT PCR and confirmed by Real time PCR. The present study demonstrates that 1,25VD down regulates telomerase activity mediated through repression of hTERT mRNA (Fig. 17 20). The stability of hTERT mRNA was decreased by 1,25VD (Fig. 20), demonstrati ng that the regulation was posttranscriptional. Our findings provide direct evidence that hTERT is a target gene for 1,25VD. Although EMSA revealed that the putative VDRE bound specifically to the VDR/RXR heterodimer, as shown by Ikeda et al 2003, it is worth noting that in their
100 study, this VDRE only mediated the down regulation of reporter activity by cotreatment of 1,25VD and 9 cis RA, but not 1,25VD alone. The limitation of this VDRE function may due to different cellular context. Transcriptional assa ys using luciferase reporter plasmids containing full length of hTERT promoter showed no down regulation by 1,25VD or by combination of 1,25VD and 9 cis RA in OCa cells. Furthermore, ChIP analysis showed that no detectable binding of VDR to this VDRE in v ivo All these data indicate that this putative VDRE is not functional in OVCAR3 cells. Real time PCR analysis revealed that 1,25VD decreased the stability of hTERT mRNA. Real time PCR showed that hTERT mRNA was degraded to 50% within ~10 hr by treatment of 1,25VD, while no detectable increase in the number of apoptotic cells within this timeframe (data not shown). The time course of destabilization of hTERT mRN A by 1,25VD precedes induction of apoptosis by 1,25VD, suggesting that decreases in hTERT mRNA are not a result of the induction of apoptosis in these cells. To our knowledge, this is the first study incident showing that the stability of hTERT mRNA was re gulated. This may open new aspects to study the regulation of hTERT, especially in the repression of telomerase. Sequence analysis of the hTERT promoter revealed binding sites for several transcription factors suggesting hTERT could be regulated by differ ent factors in different cellular contexts (Cong, 2002). The oncogene c myc has been shown to activate telomerase through two myc binding sites in hTERT promoter (Wu, 1999). Since 1,25VD have been shown to inhibit c myc (Saunders, 1993) and c myc binding s ites are present in full length hTERT reporter construct, this oncogene could mediate the repression of telomerase by 1,25VD. However, the persistent inhibition of hTERT mRNA in the
101 presence of cycloheximide and the lack of inhibition in reporter assays ru le out the indirect effect of 1,25VD mediated through c myc Over expression of hTERT allows the recovery of OCa cells after 1,25VD treatment and partially relieves 1,25VD induced apoptosis Treatment of 1,25VD in parental OCa cells causes massive apoptosi s leading to the loss of the entire population. In addition to the decrease of telomerase activity, apoptosis was also induced by 1,25VD and the recovery from 1,25VD treatment was slow in parental OVCAR3 cells. By contrast, the hTERT overexpressing cells m aintained telomerase activity at higher levels and telomere length at a size much longer than parental cells over time, 1,25VD induced apoptosis and growth inhibition were partially rescued, leading to cell recovery and growth (Fig 21 25). This indicates t hat a certain level of telomerase activity may play an important role in protecting cells from apoptosis in OCa cells. Our results provide the first direct evidence that telomerase may affect 1,25VD induced apoptosis in OCa cells. It is well known that th ere is strong correlation between telomere function and senescence ( Kolquist, 1998 ). Telomerase activation is shown as one of the three steps needed to overcome senescence and transform normal human epithelial cells to cancer cells (Hahn, 1999). Cells und ergoing senescence have a large and flat morphology, express acidic beta galactosidase ( b gal) and show a permanent G0/G1 arrest, whereas apoptotic cells show evidence of DNA fragmentation ( Biroccio, 2003 ). Many studies have linked telomerase inhibition to apoptosis. For example, hTERT expressing cells were more resistant to apoptosis induced by UV and g irradiation ( Gorbunova, 2002 ). Following ectopic expression of dominant negative TERT into transformed cells, growth
102 inhibition and apoptosis was induced ( Hahn, 1999; Zhang, 1999). In our model system, 1,25VD induced decrease in telomerase activity is correlated with 1,25VD induced apoptosis, not senescence. This is supported by our data that the senescence b gal, a marker for senescence, was not induced by 1,25VD treatment in OCa cells (data not shown). Instead, extensive cell death was observed. Our data strongly suggested that telomerase plays a critical role in cellular resistance to apoptosis. Little is known about the signaling pathways mediating VD i nduced apoptosis. Several factors involving mitochondrial and caspase changes were investigated. Bcl 2 is down regulated in VD induced apoptosis in breast and prostate cancer cells (Mathiasen, 1999; Blutt, 2000), pro apoptotic Bak is induced in colon canc er cells (Diaz, 2000). VD induced apoptosis is caspase dependent in prostate cancer cells, while in breast cancer cells, calpain may replace caspases as a key mediator (Mathiasen, 1999 and 2002). It is assumed that some anti apoptotic pathways are activat ed by telomerase or that the apoptotic pathway induced by 1,25VD is blocked by telomerase overexpression. As suggested in a recent report, o verexpression of wild type hTERT in HeLa cells increases their resistance to apoptosis, induced by the DNA damaging agent etoposide, and TERT suppresses apoptosis at a premitochondrial step by a mechanism requiring reverse transcriptase activity and 14 3 3 protein binding (Zhang, 2003). Overexpression of Bcl 2 and the caspase inhibitor zVAD fmk protected cells against apoptosis in the presence of telomerase inhibitors in pheochromocytoma cells (Fu, 1999). All these data suggest that the telomerase function is upstream of caspase activation and mitochondrial dysfunction. But different from those DNA damaging reagents ind uced apoptosis, 1,25VD induced apoptosis could mediate through different pathway. It will be
103 worthwhile to study how telomerase inserts anti apoptotic function to block 1,25VD induced apoptosis Transformed human cells enter crisis once TRF reach a length of ~4 kb. Telomeres that have been shortened to this degree may no longer protect chromosome ends, and in turn, may lead to the genomic instability and cell death (Counter, 1998). This is consisted with the fact that many cancers have mean telomere lengths well below normal and often close to the threshold required for cell survival. Tumor cells, such as OVCAR3 cells, have very short telomere and appear to require telomerase activity even for short term viability. In OVCAR3 cells, the short telomere length of 3 kb is close to the threshold level and was maintained by high telomerase activity. It is postulated that in these cells telomerase activity is required at nearly every division of cells replicating with critically short telomeres and inhibition of telomerase activity in these cells causes loss of viability through activation of apoptotic pathways (Zhang, 1999). It is known that telomerase inhibition may lead to a phenotypic lag in which cells would continue to divide until the point at which the t elomeres became critically short. This lag phase varies depending on the initial telomere length. In terms of telomerase as anti cancer target for VD, OCa cells, which have short telomeres, could have short lag phase and thus be efficiently inhibited by VD 1,25VD analogue EB1089 has potent anti tumor activity in vitro and in vivo and has strong implications to cancer treatment and prevention Since the pharmacological levels of 1,25VD (10 7 M) induce hypercalcemia, the goal of our research is to establish synthetic 1,25VD analogues, such as less calcemic EB1089, as chemotherapeutic treatment for OCa.
104 Our study is the first to demonstrate EB1089 inhibits growth of OCa cells and tumor xenografts in nude mice. Importantly, the tumor inhibition in vivo occurr ed without physical symptoms of hypercalcemia, and particularly promising is the minimal effect on serum calcium (Fig. 28). It demonstrates that EB1089 may be suitable for long term treatment and/or chemoprevention. EB1089 could also be useful for treatin g VD resistant cancers. A recent study demonstrated that down regulation of 24 hydroxylase enhaced 1,25VD levels and improved mitotic control of tumor cells (Cross, 2003). The rapid breakdown of 1,25VD was suspected to be the cause of the resistance of DU1 45, a prostate cancer cell line, to 1,25VD. In the presence of the 24 hydroxylase inhibitor, growth of DU145 cells was inhibited by 1,25VD and as much as in LNCaP cells (Zhao, 2001). EB1089 has 50 fold lower affinity than 1,25VD for 24 hydroxylase, and it is more slowly inactiviated by 24 hydroxylasion (Roy, 1995), supporting the potential effect of EB1089 in the treatment of VD resistant cancers. The pathogenesis of epithelial OCa is not completely understood, but it is believed that the process of recurre nt ovulation (incessant ovulation) causes genetic damage to ovarian epithelial cells and that sufficient genetic damage can lead to OCa in susceptible individuals ( Rodriguez, 2003) According to this model, reproductive and hormonal factors, including VD, retinoids, and non steroidal anti inflammatory drugs, may decrease OCa risk via their inhibitory effects on ovulation, leading to the biologic effects on the ovarian epithelium that are cancer preventive. Along this line, VDR expression is required for 1, 25VD mediated growth effects. We have shown the expression of VDR and its heterodimer partner, RXR, in human ovary and the response
105 of primary human ovarian cells to 1,25VD. These data indicate that the human ovary is a VD responsive organ and can be impac ted strongly by environmental VD, leading to decreased OCa risk. Most tumors have compromised p53 function and half of OCa are estimated to have lost p53 function, thus becoming refractory to drugs targeting p53 pathways. Examination of p53 independent response in p53 mutant OVCAR3 cells warrants closer attention. These tumors may however still be sensitive to synthetic VD since 1,25VD induces GADD45 and inhibit telomerase in a p53 independent manner. Better understanding of the molecular mechanism of VD is critical for determining the ultimate utility of 1,25VD analogues in the clinic. In conclusion, molecular studies show that 1,25VD inhibits OCa cell growth and is mediated through regulation of specific genes including GADD45 and telomerase. Our precl inical data show that 1,25VD analogue reduces OCa development in animals. Data from normal human ovarian tissues and cells demonstrate the presence of VDR and a response to 1,25VD. Studies from cellular, molecular and animal levels strongly suggest that VD represents a molecular target for chemotherapy and chemoprevention of OCa.
106 SUMMARY AND THE PERSPECTIVES OF FUTURE STUDY OCa is a deadly disease and its etiology is largely unknown. Lack of both reliable detection methods and early symptoms results in poor prognosis for patients with OCa. Relapse and resistance to current treatment necessitate the development of new therapeutic methods to fight this deadly disease. Our studies on VD action in OCa suggest that 1,25VD and its synthetic analogu es may be effective therapeutic treatments for OCa. GADD45 is identified as one of the primary target genes by 1,25VD in OCa cells. Based on the similarity to concensus sequences, several putative VDREs identified by EMSA are localized in GADD45 introns o r exons but not in the promoter region. Receptor response elements usually lie in the promoter region of regulated gene and all known VDRE have been identified in the promoter. Of particular interest to our study is VDRE in the 3 untranslated region of ex on 4 plays critical role in mediating the transcriptional regulation of GADD45. To our knowledge, this is the first study showing that a vitamin D enhancer element is localized in the exon region. Our study can not exclude that other VDREs act in a negativ e way and that different VDREs in the GADD45 genome may function in a cell specific manner to mediate the regulation of GADD45 expression by 1,25VD.
107 Since the specific VDRE is located in an exon and falls into the 3 untranslated region og GADD45, it prov ides an excellent model system to study how DNA response elements in the 3 end of the coding sequence may regulate the transcription of a target gene via the Pol II complex which bound to the promoter at the 5 end. Site specific transcription factors (Ag alioti, 2002), including nuclear hormone receptors (Louie, 2003), recruit components of the Pol II complex to the enhancer sequences and, after co activator mediated chromatin remodeling, the adjacent nucleosome slides downstream to initiate transcription. It remains to be seen whether DNA response elements at the 3 end of the coding sequence recruit Pol II components and, if so, how the modified nucleosome moves to the correct position to permit initiation of transcription since the nucleosome has to eith er slide upstream by a significant distance or jump across the coding region to initiate transcription. VD dependent apoptosis and cell cycle arrest could engage activation of the p38 MAPK pathway and induction of GADD45 (Sutter, 2003). Recent data showed that GADD45 is reqired for p38 activation. Disruption of GADD45 abrogates H ras induced cell cycle arrest and p38 activation (Bulavin, 2003). It is quite likely that upregulation of GADD45 by 1,25VD is required for activation of p38, that may mediate 1,25 VD induced growth inhibition. In support this hypothesis, it is known that most biological actions of VD are mediated through the nuclear VDR mediated expression of target genes. The study of VDRKO mice showed that nongenomic effects of VD in osteoblasts are abrogated in the absence of nuclear VDR (Erben, 2002) and suggest that some nongenomic responses require a functional nuclear VDR. Using GADD45 as a model
108 system, future work may decipher the cooperation of nongenomic and genomic effects of VD. Given that telomerase activity is regulated at multiple levels by different stimuli, including hormones, the mechanisms involved in telomerase regulation are far from established. Our data addressed a novel mechanism of down regulation of telomerase by 1,25VD. B etter understanding of the regulation of telomerase by 1,25VD will provide the basis for telomerase activity in VD targeted therapy. Although hTERT is regulated tightly at the promoter machinery and a putative VDRE in the hTERT promoter responds to a combi nation of 1,25VD and 9 cis RA in certain prostate cancer cells, it fails to respond to 1,25VD alone (Ikeda, 2003). It is not unusual that sequences match in vitro but do not provide a functional VDRE (Colnot, 2000; Gonzalez, 2002). We provide strong eviden ce that VD increase the degradation of hTERT mRNA by real time PCR analysis using 2 probes. Our study is the first to demonstrate down regulation of telomerase mediated by decreasing the hTERT mRNA stability. Given that upregulation of hTERT mRNA and telo merase activity induced by multiple oncogenetic factors in cancer cells, degradation of hTERT mRNA by 1,25VD could be an effective way to suppress the effect of multiple oncogenic pathways in OCa. Recent studies showed that double stranded RNA, including s mall interfering RNA (siRNA) and microRNA (miRNA) can induce the degradation of homologous RNAs in organisms as diverse as protozoa, animals, plants and fungi, resulting in posttranscriptional gene silencing (Lewis, 2003). MiRNA are endogenous ~22nt RNAs t hat arise from larger precursors transcribed from non protein coding genes. SiRNA arise
109 by cleavage of long, double stranded RNAs. Despite the differences in origin, miRNA ans siRNA are functionally interchangeable (Carrington, 2003; Nelson, 2003). 1,25VD induced degradation of hTERT mRNA may provide a model to study the posttranscriptional repression of hTERT mRNA by double stranded RNA. hTERT is not included in those ~400 targets of mammalian miRNA identified in the present database, perhaps due to the li mited sensitivity of current bioinformatic methods. The actual number of target genes regulated by each miRNA is likely to be substantially higher (Lewis, 2003). It remains to be seen if hTERT would be a target for miRNA; if so, it will be important to est ablish how 1,25VD mediates down regulation of hTERT mRNA through MiRNA. The ability of 1,25VD to induce cell cycle arrest and apoptosis without the involvement of p53 may prove useful in therapy. It is known that p53 stimulates the activities of p21/WAF1 gadd45 and bax genes to enhance their expression as a transcriptional factor resulting in cell cycle arrest, DNA repair and apoptosis (Bargonetti, 2002). p53 is mutated in 50% of human malignancies, including ovarian cancer and tumors with mutant p53 res ist conventional p53 target therapy. In our nodel system, 1,25VD stimulates GADD45 and down regulates telomerase in a p53 independent manner. 1,25VD therapy might compensate or substitute for part of the p53 function to induce cell cycle arrest and apoptos is. 1,25VD may prove valuable candidates for treating OCa, especially those OCa that have acquired resistance to other apoptosis inducing agents due to a mutation in p53. VD decreases the generation of single strand DNA induced by diethylnitrosamine (DEN), a mutagen that induces chromosome aberrations (Basak, 2000). DNA double
110 strand breaks are generated from mutagen induced DNA lesions in the S phase of the cell cycle. It is reasonable to assume that double strand breaks are repaired in the G2 phase by pos t replicational repair mechanism. Therfore, VD mediated suppression of double strand breaks could be mediated through GADD45, which has been shown to play role in DNA repair. GADD45 modified DNA accessibility on damaged chromatin (Carrier, 1999) and affect s chromatin remodeling of templates concurrent with DNA repair. Slower nucleotide excision repair was found in GADD45 deficient keratinocytes exposed to UV (Maeda, 2002). Based on these data GADD45 may participate in the coupling between chromatin assembly and DNA repair. Since telomerase maintains the telomere length and contributes to chromosome stability (Cech, 2004), it will be worthwhile to investigate the role of GADD45 and telomerase in VD protected chromosome stability and induced DNA repair. These two pathways will provide useful targets for DNA damaging chemotherapeutics against p53 defective OCa, which have decreased ability to repair chemotherapeutic damage. We were first to provide preclinical data on the 1,25VD analogue EB1089 in OCa. The goa l of future studies is to test EB1089 or other promising 1,25VD analogues in clinical trial of OCa. To increase the antiproliferative potency without increasing side effects, use of less calcemic analogues appears to be the most reasonable approach. Severa l promising new synthetic VD analogs are also under development, such as KH 1060, LG190119, deltanoids, 1 alpha hydroxyvitamin D5, vitamin D2, QW 1624F2 2, etc. EB1089 has been widely used in breast and colon cancer patients and stabilization of disease was observed in a phase I trial (Gulliford, 1998). In a phase II study of EB1089
111 in patients with inoperable hepatocellular carcinoma, strikingly, out of 33 patients, two had complete response, 12 stable diseases. Complete regression appeared after 6 and 2 4 months of treatment and lasted 29 to 36 months (Dalhoff, 2003). EB1089 was also well tolerated. Most patients tolerated a daily dose of 10 m g of EB1089. A Phase III trial in hepatoma are currently ongoing. These studies showed EB1089 response by a reduct ion of solid tumor size and provide the rationale that EB1089 can be used for treatment. The effectiveness of 1,25VD analogues in slowing the progression of prostate cancer, was shown in a study of an oral 1,25VD analogue (Rocaltrol) to treat early recurre nt prostate cancer. PSA doubling time was significantly prolonged by the treatment in all 7 cases (Zhao, 2001). Administration of 1,25VD analogues through dietary supplementation suggests that oral ingestion of VD based chemotherapy is an effective and fe asible approach. Oral administration and long term safety of 1,25VD analogues clearly has the advantage in terms of feasibility of cancer chemotherapy and chemopreventive agents (Welsh, 2003). Our study showed human OSE growth was inhibited by 1,25VD and that VDR, RXR are expressed in normal human ovary tissues, which support the suspected role of VD in OCa initiation. Our data suggest that VD may have a cancer preventative effect While n umerous issues remain to be addressed, it is essential to investiga te the anti tumor activity in the OCa prevention model after defining the downstream targets of VDR in the normal ovary. In summary, we link the specific molecular pathway to 1,25VDs biological function in this study. We provide the first molecular evi dence that G2/M arrest by 1,25VD in OCa cells is mediated through the induction of GADD45 via a novel exonic
112 enhancer. We are also first to show that 1,25VD induced apoptosis is mediated by destabilization of hTERT mRNA and a decrease in telomerase activit y. In addition, our study is the first to demonstrate that 1,25VD analogue EB1089 inhibits OCa xenograft in vivo With the ubiquitous expression of VDR and RXR in ovarian tissues and response of primary OSE to 1,25VD, our data strongly suggest further inv estigation of less calcemic synthetic 1,25VD analogues as chemopreventive and chemtherapeutic agents against OCa. Furthermore, half of OCa is estimated to lose p53 function, thus becoming refractory to drugs targeting p53 pathways. These tumors may howeve r still be sensitive to synthetic VD since 1,25VD induces GADD45 and inhibits telomerase in a p53 independent manner.
113 MATERIALS AND METHODS Materials pHG45 HC containing the 8 kb genomic sequence of human GADD45 (Hollander, 1993), pCMV45 containing the open reading frame of human GADD45 cDNA and pCMVAS45 containing human GADD45 cDNA in the anti sense orientation (Zhan, 1994), pCMVgal (Li, 2003), p91023B VDR (Baker, 1988; Hilliard, 1994), pCMX RXR b (Mangelsdorf, 1992), p23 containing rat 24 hydroxyla se promoter in pMAMMneoLuc (Arbour, 1998) and pBabhTERT (Vaziri, 1998) hTERT reporter construct pGL3 3328Luc (Kyo, 1999)have been described previously. pGL3 promoter, pGL3 basic and pGL3 control vectors were from Promega ( Madison, WI ). MEFs from wild type and GADD45 null mice have been described (Hollander, 1999). 1,25VD was from Calbiochem (La Jolla, CA). EB1089 was kindly provided by Dr. Binderup of Leo Pharmaceuticals Products (Ballerup, Denmark). Baculovirus expressed human VDR protein, human RXR b prot ein and anti RXR b antibody were from Affinity BioReagents Inc. (Golden, CO). Anti VDR antibody was from Chemicon International (Temecula, CA). Anti Flag M2 antibody and anti b actin antibody were from Sigma (St. Louis, MO). Anti GADD45 antibody (C 20), ant i cdc2 antibody and anti cyclin B1 antibody (D11) were from Santa Cruz Biotech (Santa Cruz, CA). All oligonucleotides were synthesized by Invitrogen (Carlsbad, CA). The sequence of primers used for the construction of
114 GADDLuc reporter by polymerase chain r eaction (PCR) is: 5 GGTGGTACGCGTCCCGAACTTCTCTTACCTACC 3 (forward) and 5 GGTGGTAGATCTACCCAAACTATGGCTGCACAC 3 (reverse). The sequence of the oligonucleotides in sense orientation for producing complementary double strand oligoes for EMSA and site mutagen esis is listed below: human OC VDRE 5 ttggtgactcaccGGGTGAacgGGGGCAtt 3; putative GADD45 VDRE A 5 ttgggcgtgcagGGGTCAtggGGGGTGacg 3; putative GADD45 VDRE B 5 taggtggGGGTCAggaGGGTGGctgcctttgt 3; putative GADD45 VDRE C 5 aactGTTTCActcAGGTCAgggtaa caagt 3; putative GADD45 VDRE D 5 cagcttgGGTTGCatgGGTTCAgactttgc 3; putative GADD45 VDRE E 5 gccaaggGGCTGAgtgAGTTCAactacatg 3; putative hTERT VDRE 5 cacccactggtaaggAGTTCAtggAGTTCAat 3; primer for the mutation of VDRE A 5 gtgcagGGGTCAtggGGG t T t acggggccgcggga 3; primer for the mutation of VDRE B 5 GGGTCAggaGGGT tt ctgcctttgtccgactagagtg 3; primer for the mutation of VDRE D 5 cagcttgGGTTGCatgGGTT tt gactttgcaatgtgtag 3; primer for the mutation of VDRE E 5 gccaaggGGCTGAgtgAGTT tt actacatgtt ctggg 3. Capital letters are used to indicate the hexameric core binding motifs in the VDRE primers and bold letters in lowercase are used to indicate the nucleotides in the mutagenesis primers that are different from the wild type sequence. The sequence of primers for chromatin immunoprecipitation (ChIP) assays is listed below: VDRE E region ( 2565/2767 ) CTGAACGGTGATGGCATCTG 3 (forward) and 5 CTGTTTCAACACAGCTTCCTTC 3 (reverse); promoter region 1 ( 1492/ 1241) 5 GTTGTCCATGGCTGACAACA 3 (forward) a nd 5 GCTCCCACATGCTTGCATTC 3 (reverse); promoter region 2 ( 505/ 310) 5
115 CACTTCTGAGGTAAACTTTGC 3 (forward) and 5 GAAGCAGGCTGCCAAGTGTT 3 (reverse). Colorimetric methylthiazole tetrazolium (MTT) assays and statistical analysis To measure cell growth, OVC AR3 cells were plated at 2 10 3 cells/well in 96 well plates and treated with VD or vehicle. MTT assays were performed as described (Li, 2001). OD 595 was read on a MRX microplate reader (DYNEX Technologies, Chantilly, VA). For cell growth and cell cycle a nalyses, statistical analysis was performed using the independent samples t test. P< 0.05 was considered to be statistically significant. Cell cycle and apoptosis analysis by flow cytometry To determine the cell cycle distribution, cells were harvested by trypsin digestion and fixed with 70% ethanol in PBS for 12 hours at 4 o C. Fixed cells were incubated overnight with 100 m g/ml RNase, stained with 50 m g/ml propidium iodide at 4 o C and subjected to cell cycle analysis on a FACScan (Becton Dickinson, Mountai n View, CA). To determine the apoptosis, cells were harvested by trypsin digestion and washed with PBS. Cell suspension in 1x assay buffer was added annexin V FITC and propidium iodide and incubated for 15min, then subjected to flow cytometry. Northern blo t analysis To determine the level of GADD45 mRNA, Northern blot was performed as described (Zhang, 2003). Briefly, OVCAR3 cells were incubated with 1,25VD or vehicle for the indicated times. Total cellular RNA was isolated by TRIzol (Invitrogen) method fol lowing manufactures instruction. Samples containing 20 m g RNA were run on a 1%
116 agarose gel in denaturing gel buffer (Ambion, Austin, TX) and transferred onto a nylon membrane. The membrane was pre hybridized at 65 o C for 4 h and hybridized with the GADD45 or GAPDH probe at 65 o C for overnight. Washes were performed in high stringency buffers. To prepare GADD45 probe, full length GADD45 cDNA was released from pCMV45 vector with Hind I I I /Xba I digestion, separated and recovered from agarose gel. GAPDH probe wa s from Ambion. The probes were labeled with 32 P using random primed DNA labeling kit (Ambion). Signal densities were analyzed with scion Image software (Scion Corp., Frederick, MD). Gel mobility shift assay (EMSA) EMSA was performed as described (Li, 2003 ; Gonzalez, 2002) with modifications. Briefly, double stranded oligonucleotides were end labeled with 32 P using a T4 polynucleotide kinase labeling system (Life Technologies, Rockville, MD). 1 m l of radiolabeled probe (roughly 50,000 cpm) was mixed with 1 9 m l DNA binding reaction mixture that contains 250 ng VDR, 250 ng RXR, 10 mM Tris Cl (pH7.9), 100 mM KCl, 0.1 mM EDTA, 15% glycerol, 100 m g/ml poly(dI:dC), 0.1 m g/ m l bovine serum albumin, 1 mM DTT and 10 7 M 1,25VD. The mixture was incubated at room tempe rature for 30 minutes. For competition experiments, VDR/RXR was pre incubated with 2 m g anti RXR b anti Flag M2 antibody or excess amount of cold probes on ice for 20 minutes before the EMSA reaction. The reaction mixture was resolved in a 5% non denaturin g polyarylamide gel and protein oligo complexes were revealed by autoradiography. Construction of luciferase reporter plasmids, deletion and site directed mutagenesis
117 To construct GADDLuc, GADD45 genomic DNA fragment from +366 to +2926 was amplified by PCR using primers described in the Material. The forward primer contains a MluI and the reverse primer a BglII site. The amplified PCR fragment was cloned into the Mlu I and Bgl I I sites of pGL3 promoter vector. Luc1 was generated by digesting the GADDLuc wit h Kpn I and religation. Luc2 was generated by digesting GADDLuc with Mlu I and EcoR I filling with Klenow fragments and religation. Luc3 was generated by sub cloning into Bgl I I site of pGL3 promoter vector a 440 bp DNA fragment released from GADDLuc with Bam H I and Bgl I I Luc4 was generated by digesting GADDLuc with Bgl I I and EcoR I filling with Klenow fragments and religation. Luc5 was generated by sub cloning into pGL3 promoter vector at Kpn I and Bgl I I sites a 777 bp fragment released from Luc2 with Kpn I a nd BamH I Site directed mutagenesis was performed as described (Lee, 2002) using QuikChange Site directed Mutagenesis kit ( Stratagene, La Jolla, CA). The sequence of all mutant constructs was verified by DNA sequencing. Transcriptional analysis For trans fection studies, OVCAR3 cells were plated in 15% FBS RPMI 1640 medium at 1 10 5 cells/well and HeLa cells in 10% FBS DMEM at 5 10 4 cells/well in 12 well plates. On the next day, OVCAR3 cells were transfected by Lipofectamine Plus and HeLa cells by Lipof ectamine following the protocol from Invitrogen. 4 h post transfection, cells were treated with 1,25VD or vehicle in fresh medium for 36 h. Cells
118 were harvested and luciferase and b galactosidase ( b gal) assays were performed as described (Lee, 2000 and 2002). Chromatin immunoprecipitation (ChIP) assays For ChIP assays, OVCAR3 cells were treated with EOH or 10 7 M VD for 60 min and cross linked with 1% formaldehyde at room temperature for 10 min. Then, the cells were incubated with 0.125 M glycine for 5 min, washed, scraped in ice cold PBS containing protease inhibitor cocktail (Roche) and lysed in buffer (pH 8.0) containing 5 mM PIPES, 85 mM KCl, 0.5 % NP 40 and protease inhibitor cocktail. Cell nuclei were re suspended in lysis buffer containing 50 mM Tris Cl (pH. 8.1), 10 mM EDTA, 1% SDS and protease inhibitor cocktail. Soluble chromatin was prepared by sonication and diluted in buffer containing 16.7 mM Tris Cl (pH 8.1), 0.01 % SDS, 1.1% Triton X 100, 1.2 mM EDTA, 167 mM NaCl and protease inhibitor c ocktail. The diluted chromatin solution was pre cleared with pre immune serum and protein G agarose (Santa Cruz) pre coated with sheared sperm DNA (Ambion). Immunoprecipitations were carried for overnight at 4 o C with rat anti VDR antibody or Rat IgG (Sigm a) followed by incubation with pre coated protein G agarose for 2 h at 4 o C. The beads were sequentially washed at room temperature for two times (10 min/wash) in each of the following buffers: the dilution buffer, TSE 500 wash buffer containing 20 mM Tris Cl (pH 8.1), 0.1% SDS, 1% Triton X 100, 2 mM EDTA, 500 mM NaCl) and LiCl/detergent wash buffer containing 100 mM Tris Cl (pH 8.1), 1% NP40, 1% deoxycholic acid, 500 mM LiCl). After the final wash in TE buffer containing 10 mM Tris Cl (pH 8.0) and 1 mM EDT A, the immunocomplexes were eluted from the beads with 50 mM NaHCO 3 and 1% SDS for two times. The immunocomplexes were pooled and heated at 65 o C overnight to reverse the cross
119 linking. DNA was extracted from the immunocomplexes using a QIAquick Spin Kit ( Qiagen, CA). 2 m l out of 30 m l DNA extract was used for PCR. Immunoblotting analysis Immunoblotting analysis of GADD45 protein was performed as previously described (Li, 2001) with modification. Briefly, cells were harvested in lysis buffer containing 50 mM Tris Cl (pH 7.5), 1% NP 40, 0.25% deoxycholic acid, 400 mM NaCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM Na 3 VO 4 1 mM NaF, and protease inhibitor cocktail. The protein concentration of the cell lysate was assayed using Bio Rad kit. Extracts containing 50 m g of protein were separated on a sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose membrane. GADD45 was detected using ECL Kit following the manufactures instruction (Amersham Pharmacia Biotech, Pis cataway, NJ). The immunoblotting analysis of cdc2, cyclinB1 and VDR was performed similarly as for GADD45 except that the cell extracts were prepared in buffer containing 20 mM Tris Cl (pH 7.5), 300 mM NaCl, 3 mM EDTA, 3 mM EGTA, 100 m M Na 3 VO 4 1% NP 40, a nd protease inhibitor cocktail (Zhang, 2003). Establishment of stable clones from OVCAR3 cells OVCAR3 cells were transfected with 10 m g pCMVAS45 or hTERT plasmid together with 0.5 m g pcDNA3 for the establishment of GADD45 anti sense, telomerase stable clon es. pcDNA3 alone transfected into OVCAR3 cells for the Vector OVCAR3 controls. For the establishment of stable clones with GADDLuc reporter, OVCAR3 cells were transfected with 10 m g GADDLuc plasmid and 0.5 m g pcDNA3. All stable clones
120 were obtained through selection with 100 m g/ml G418 for a period of about 4 weeks and isolated by cloning with glass cylinders. In vitro immunocomplex kinase assays In vitro immunocomplex kinase assays were performed as described (Lee, 2000) with minor modifications. In brie f, cells were washed with ice cold PBS and cellular extracts are prepared in buffer containing 20 mM Tris Cl (pH 7.5), 300 mM NaCl, 3 mM EDTA, 3 mM EGTA, 100 m M Na 3 VO 4 1% NP 40, protease inhibitor cocktail. Cellular extracts containing 200 m g protein were immunoprecipitated with anti cyclin B1 antibody. Kinase assays were performed at 30 o C in 20 m l reaction buffer containing 20 mM HEPES (pH7.5), 5 mM MgCl 2 2 mM EDTA, 2 mM EGTA, 1 mM dithiothreitol, 20 m M ATP, 3 m g histone H1, and 10 m Ci [ g 32 P] ATP. Reac tions were terminated by adding 2 SDS PAGE sample buffer, analyzed on a SDS PAGE and visualized by autoradiography. Telomerase activity assay Telomerase activity was measured with the telomerase PCR enzyme linked immunosorbent assay (ELISA) kit (Roch e) base on the telomeric repeat amplification protocol (TRAP) assay. Cells were suspended in lysis buffer (Roche), incubated on ice for 30 min, and then centrifuged at 14000 g for 20 min. Supernatants were used for the detection of telomerase or flash fr ozen and stored at 80 o C. According to the manufacturers instructions, 200 ng of protein extract were assayed for telomerase activity after 30 cycles of amplification by PCR. The resulting PCR product (5 ul) was quantified by ELISA. Telomerase activity w as expressed as absorbance values (OD) measured using a microtiter reader at 450 nm with a reference wavelength of 595 nm. All
121 assays were performed in duplicate and a dilution series of control telomerase extracts was always examined in parallel to give t itration curve for normalizing experimental variations. RT PCR analysis Analysis of the expression of hTERT and GAPDH were performed by reverse transcription PCR amplification as previously described (Nakamura, 1997). RNA was isolated with Trizol (invit rogen) according to the manufacturer. 1 g of total RNA was reverse transcribed using the RNA PCR kit version 2 (TaKaRa, Ohtsu, Japan) with oligo dT primers. To amplify the cDNA, l aliquots of the reverse transcribed cNNA (20ul) from 2 m g of RNA were subjected to PCR in 2 5 ul of 1 x buffer (10mM Tris Cl (pH8.3), 1.5 mM MgCl 2 50 mM KCl), containing 1 mM each of the dNTP; 1.25 U of Taq NDA polymerase (TaKaRa); and 0.2 uM of specific primers. Primer sequence was chosen to amplify a 145 bp region presenting in all transcripts of hTERT mRNA 1784 (5 cggaagagtgtctggagcaa 3) and 1910 (5 ggatgaagcggagtctgga 3) for 30 cycles (94 o C for 30s, 60 o C for 30s, 72 o C for 90s). As a positive control, GPDH mRNA was amplified in parallel using primers 5 ctcagacaccatggggaaggtga 3 and 5 atgatcttgaggctgttgtcata 3 for 16 cycles (94 o C for 30s, 55 o C for 30s, 72 o C for 45s). PCR products were electrophoresed in 15% polyacrylamide gel and stained with SYBR green. For semi quantification of hTERT mRNA expression, serially diluted cDNA re verse transcribed from 2 m g RNA (corresponding to 50 ng to 1 m g RNA) was subjected to RT PCR. Primers, probes for Real time PCR Primers and prodes for the hTERT gene were chosen with the assistance of the Primer Express (Perkin Elmer Applied Biosystems, Fo ster city, CA). To avoid
122 amplification of contaminating genomic DNA, either upstream or downstream primer was placed in a different exon. Forward primer of hTERT1 (1909F) and probe1 was placed in exon 4, whereas reverse primer (2017R) was spanning exon 4 a nd 5 junction. Forward primer of hTERT2 (3081F) was spanning exon 13 and 14 junction, probe2 and reverse primer (3162R) in exon 14. The nucleotide sequences are as follows: hTERT1: 1909F: 5 gtccagactccgcttcataa 3; 2017R: 5 gagacgctcggccctctt 3; FAM/TA MRAprobe1: 5 ttctggctcccacgacgtagtccatg 3; PCR product size: 109bp. hTERT2: 3081F: 5 cgtacaggtttcacgcatgtg 3; 3162R: 5 atgacgcgcaggaaaaatg 3; TAM/TAMRAprobe2: 5 agctcccatttcatcagcaagtttggaag 3. PCR product size: 82bp. Real time PCR RNA was prepared using the RNasy RNA isolation kit (Qiagen) and DNase digested on column using RNase free DNase set (Qiagen) and then reverse transcribed. An RNA pool was generated by mixing aliquots of RNA from cells treated with vehicle or 1,25VD for various times. Conc entrations of the pooled RNA ranging from 0 (buffer alone) to 50 ng/ m l were used in the PCR analysis to generate the standard curve for each gene. The Ct value was generated by the ABI PRISM 7700 SDS software version 1.7 and then exported to an Excel sprea dsheet where equations from the standard curve were generated. Using the Ct values, concentrations of the hTERT and GAPDH mRNAs were calculated from the equations. Each sample was analyzed at two different concentrations (50 ng/ m l and 2 ng/ m l) with only re sults in the most sensitive region of the standard curve presented. Samples at each concentration were analyzed in triplicates. hTERT levels were normalized to the corresponding input total RNA, base on quantitation of GAPDH. Telomere length analysis
123 Tel omere length was measured using a non radioactive chemiluminescent assay developed by Roche Diagnostics. In brief, genomic DNA was isolated using a Roche DNA isolation kit. 5 m g DNA was digested overnight with 20 U of Hinf I and Rsa I, fractionalized on a 0.8% agarose gel. Gels were denatured for 30 min and then neutratlized for 30 min. DNA was transfered to a nylon membrane, hybridized with digoxigenin 3 end tailing labeled (Roche) (TTAGGG) 4 oligonucleotides overnight at 50 o C in PerfectHyb solution ( Sigma). Membranes were washed twice in 2 x SSC, 0.1% SDS for 15 min at RT and twice in 0.1 x SSC, 0.1% SDS for 15 min at 50 o C. The probe was detected by the digoxigenin luminescence detection and processed by Southern blotting and chemiluminescent detecti on. The average telomere length can be determined by comparing the signals relative to a molecular weight standard. Immuno histochemical analysis To confirm the VDR and RXR expression in normal human tissue, paraffin embedded tissues were immunostained wit h anti VDR antibody and anti RXR antibody. The signal was detected with the avidin biotin complex (ABC) immunoperoxidase kit (Vectastatin Elite ABC, Vector Laboratories, CA). The cell nucleus was counterstained with hematoxylin. Positive (skin tissue) and negative controls (pre immune serum) were included in all immuno reactions. Nude mouse tumor studies The studies were performed as described (Blutt, 2000) with little modification. Briefly, OVCAR3 cells were trypsinized and resuspended in 50% Matrigel Matr ix (Becton Dickinson, San Jose, CA) at a concentration of 2x10 6 cells/100ul matrix. Female athymic nude mice, ~6 weeks of age, on a vitamin D deficient diet supplemented with
124 1% calcium, received injections s.c. with 2x10 6 cells on the dorsal surface. Afte r tumors grow to a volume of about 150mm 3 the mice were randomly divided to three groups treated with placebo or EB1089 at 0.3 or 1.0 m g/kg/day. EB1089 was delivered daily by gavage. The size of the tumor was determined by measuring the tumor diameters wit h a Vernier caliper every 5 days and using the formula tumor volume=dxDxD/2 (where d and D represent the shortest and the longest diameter, respectively).
125 REFERENCES 1. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T. 1 981. Differentiation of mouse myeloid leukemia cells induced by 1 alpha, 25 dihydroxyvitamin D3. Proc Natl Acad Sci U S A. 78:4990 4 2. Agalioti T, Chen G, Thanos D. 2002. Deciphering the transcriptional histone acetylation code for a human gene. Cell. 111:3 81 92 3. Ahonen MH, Zhuang YH, Aine R, Ylikomi T, Tuohimaa P. 2000. Androgen receptor and vitamin D receptor in human ovarian cancer: growth stimulation and inhibition by ligands. Int J Cancer. 86:40 6 4. Akhter J, Goerdel M, Morris DL. 1996. Vitamin D3 analo gue (EB 1089) inhibits in vitro cellular proliferation of human colon cancer cells. Br J Surg. 83:229 30 5. Akutsu N, Lin R, Bastien Y, Bestawros A, Enepekides DJ, Black MJ, White JH. 2001. Regulation of gene Expression by 1alpha,25 dihydroxyvitamin D3 and Its analog EB1089 under growth inhibitory conditions in squamous carcinoma Cells. Mol. Endocrinol. 15 1127 1139 6. Albanell J, Han W, Mellado B, Gunawardane R, Scher HI, Dmitrovsky E, Moore MA. 1996. Telomerase activity is repressed during differentiation o f maturation sensitive but not resistant human tumor cell lines. Cancer Res. 56:1503 8
126 7. Arbour NC, Ross, TK, Zierold C, Prahl JM, DeLuca HF. 1998. A highly sensitive method for large scale measurements of 1,25 dihydroxyvitamin D. Anal. Biochem. 255, 148 1 54 8. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Housman DE, Evans RM. 1987. Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science. 237:268 75 9. Bamberger C M Bamberger AM, De Castro M, Chrousos GP. 1995. Glucocorticoid receptor beta, a potential endogenous inhibitor of glucocorticoid action in humans. J. Clin. Invest. 95: 2435 2441 10. Bargonetti J, Manfredi JJ. 2002. Multiple roles of the tumor suppressor p53. Curr Opin Oncol. 14:86 91 11. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O'Malley BW. 1988. Cloning and expression of full length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci U S A. 85:3294 8 12. Basa k R, Saha BK, Chatterjee M. 2000. Inhibition of diethylnitrosamine induced rat liver chromosomal aberrations and DNA strand breaks by synergistic supplementation of vanadium and 1alpha,25 dihydroxyvitamin D(3). Biochim Biophys Acta. 1502:273 82 13. Baudet C, Chevalier G, Chassevent A, Canova C, Filmon R, Larra F, Brachet P, Wion D. 1996. 1,25 Dihydroxyvitamin D3 induces programmed cell death in a rat glioma cell line. J Neurosci Res. 46:540 50
127 14. Baudino TA, Kraichely DM, Jefcoat SC Jr, Winchester SK, Partridge NC, MacDonald PN. 1998. Isolation and characterization of a novel coactivator protein, NCoA 62, involved in vitamin D mediated transcription. J Biol Chem. 273:16434 41 15. Beato, M., Herrlich, P., and Schutz, G. 1995. Steroid hormone receptors: many actors i n search of a plot. Cell 83: 851 857 16. Belandia B, Parker MG. 2003. Nuclear receptors: a rendezvous for chromatin remodeling factors. Cell 114:277 80 17. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC. 1988. Immunocytochemical d etection of 1,25 dihydroxyvitamin D receptors in normal human tissues. J Clin Endocrinol Metab. 67:607 13 18. Bettoun DJ, Burris TP, Houck KA, Buck DW 2nd, Stayrook KR, Khalifa B, Lu J, Chin WW, Nagpal S. 2003. Retinoid X receptor is a nonsilent major contri butor to vitamin D receptor mediated transcriptional activation. Mol Endocrinol. 17:2320 8 19. Biroccio A, Amodei S, Antonelli A, Benassi B, Zupi G 2003. Inhibition of c Myc oncoprotein limits the growth of human melanoma cells by inducing cellular crisis. J Biol Chem. 278:35693 701 20. Blanquart C, Barbier O, Fruchart JC, Staels B, Glineur C. 2003. Peroxisome proliferator activated receptors: regulation of transcriptional activities and roles in inflammation. J Steroid Biochem Mol Biol. 85:267 73 21. Blumberg B, E vans RM. 1998. Orphan nuclear receptors -new ligands and new possibilities. Genes Dev. 12:3149 55
128 22. Blutt SE, Allegretto EA, Pike JW, Weigel NL. 1997. 1,25 dihydroxyvitamin D3 and 9 cis retinoic acid act synergistically to inhibit the growth of LNCaP prost ate cells and cause accumulation of cells in G1. Endocrinology 138, 1491 97 23. Blutt SE, McDonnell TJ, Polek TC, Weigel NL 2000. Calcitriol induced apoptosis in LNCaP cells is blocked by overexpression of Bcl 2. Endocrinology 141:5 9 24. Blutt SE, Polek TC, Stew art LV, Kattan MW, Weigel NL. 2000. A calcitriol analogue, EB1089, inhibits the growth of LNCaP tumors in nude mice. Cancer Res. 60:779 82 25. Bouillon R, Garmyn M, Verstuyf A, Segaert S, Casteels K, Mathieu C. 1995. Paracrine role for calcitriol in the immun e system and skin creates new therapeutic possibilities for vitamin D analogs. Eur J Endocrinol. 133:7 16 26. Braunstein I, Cohen Barak O, Shachaf C, Ravel Y, Yalon Hacohen M, Mills GB, Tzukerman M, Skorecki KL. 2001. Human telomerase reverse transcriptase pr omoter regulation in normal and malignant human ovarian epithelial cells. Cancer Res. 61:5529 36 27. Brown AJ, Dusso A, Slatopolsky E. 1999. Vitamin D. Am J Physiol. 277:157 75 28. Bulavin DV, Kovalsky O, Hollander MC, Fornace AJ Jr. 2003. Loss of oncogenic H ra s induced cell cycle arrest and p38 mitogen activated protein kinase activation by disruption of Gadd45a. Mol Cell Biol. 23:3859 71 29. Carrier F, Georgel PT, Pourquier P, Blake M, Kontny HU, Antinore MJ, Gariboldi M, Myers TG, Weinstein JN, Pommier Y, Fornace AJ Jr. 1999. Gadd45, a p53 responsive stress protein, modifies DNA accessibility on damaged chromatin. Mol Cell Biol. 19:1673 85
129 30. Carrington JC, Ambros V. 2003. Role of microRNAs in plant and animal development. Science. 301:336 8 31. Castillo AI, Jimenez L ara AM, Tolon RM, Aranda A. 1999. Synergistic activation of the prolactin promoter by vitamin D receptor and GHF 1: role of the coactivators, CREB binding protein and steroid hormone receptor coactivator 1 (SRC 1). Mol Endocrinol. 13:1141 54 32. Cech TR. 2004 Beginning to understand the end of the chromosome. Cell. 116:273 9 33. Chang C, Kokontis J, Liao S. 1988. Molecular cloning of human and rat complementary DNA encoding androgen receptors. Science 240: 324 26 34. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ. 20 01. Nuclear receptors and lipid physiology: opening the X files. Nature 294: 1866 1970 35. Chida K, Hashiba H, Fukushima M, Suda T, Kuroki T. 1985. Inhibition of tumor promotion in mouse skin by 1 alpha,25 dihydroxyvitamin D3. Cancer Res. 45, 5426 30 36. Cleme ns TL, Garrett KP, Zhou XY, Pike JW, Haussler MR, Dempster DW. 1988. Immunocytochemical localization of the 1,25 dihydroxyvitamin D3 receptor in target cells. Endocrinology. 122:1224 30 37. Cole TJ, Blendy JA, Monaghan AP, Krieglstein K, Schmid W, Aguzzi A, F antuzzi G, Hummler E, Unsicker K, and Schutz G. 1995. Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation. Genes Dev. 9: 1608 21
130 38. Colnot S, Ovejero C, Romagnolo B, Porte u A, Lacourte P, Thomasset M, Perret C. 2000. Transgenic analysis of the response of the rat calbindin D 9k gene to vitamin D. Endocrinology. 141:2301 8 39. Colston KW, Mackay AG, James SY, Binderup L, Chander S, Coombes RC. 1992. EB1089: a new vitamin D an alogue that inhibits the growth of breast cancer cells in vivo and in vitro. Biochem Pharmacol. 44:2273 80 40. Colston KW, James SY, Ofori Kuragu EA, Binderup L, Grant AG. 1997. Vitamin D receptors and anti proliferative effects of vitamin D derivatives in hum an pancreatic carcinoma cells in vivo and in vitro. Br J Cancer 76:1017 20 41. Cong YS, Wright WE, Shay JW. 2002. Human telomerase and its regulation. Microbiol Mol Biol Rev. 66:407 25 42. Conneely OM, Mulac Jericevic B, DeMayo F, Lydon JP, O'Malley BW. 2002. Re productive functions of progesterone receptors. Recent Prog Horm Res. 57:339 55 43. Counter CM, Hahn WC, Wei W, Caddle SD, Beijersbergen RL, Lansdorp PM, Sedivy JM, Weinberg RA. 1998. Dissociation among in vitro telomerase activity, telomere maintenance, and cellular immortalization. Proc Natl Acad Sci U S A. 95:14723 8 44. Counter CM, Meyerson M, Eaton EN, Ellisen LW, Caddle SD, Haber DA, Weinberg RA. 1998. Telomerase activity is restored in human cells by ectopic expression of hTERT (hEST2), the catalytic subun it of telomerase. Oncogene 16:1217 22 45. Couse, J.F., and Korach, K.S. 1998. Exploring the role of sex steroids through studies of receptor deficient mice. J. Mol. Med. 76: 497 511
131 46. Couse, J.F., and Korach, K.S. 1999. Estrogen receptor null mice: what have w e learned and where will they lead us. Endocrine Rev. 20: 358 417 47. Cross HS, Kallay E, Farhan H, Weiland T, Manhardt T. 2003. Regulation of extrarenal vitamin D metabolism as a tool for colon and prostate cancer prevention. Recent Results Cancer Res.164:41 3 25 48. Dalhoff K, Dancey J, Astrup L, Skovsgaard T, Hamberg KJ, Lofts FJ, Rosmorduc O, Erlinger S, Bach Hansen J, Steward WP, Skov T, Burcharth F, Evans TR. 2003. A phase II study of the vitamin D analogue Seocalcitol in patients with inoperable hepatocell ular carcinoma. Br J Cancer. 89:252 7 49. Devasa SS, et al 1999. Atlas of Cancer Mortality in the United States, 1950 94, p226 230. National Institutes of Health, National Cancer Institute (NIH Publication No. 99 4564), Maryland, USA 50. Diaz GD, Paraskeva C, Th omas MG, Binderup L, Hague A. 2000. Apoptosis is induced by the active metabolite of vitamin D3 and its analogue EB1089 in colorectal adenoma and carcinoma cells: possible implications for prevention and therapy. Cancer Res. 60:2304 12 51. Dodd RC, Cohen MS, Newman SL, Gray TK. 1983. Vitamin D metabolites change the phenotype of monoblastic U937 cells. Proc. Natl. Acad. Sci. U. S. A. 80:7538 41 52. Dokoh S, Donaldson CA, Marion SL, Pike JW, Haussler MR. 1983. The ovary: a target organ for 1,25 dihydroxyvitamin D 3 Endocrinology 112:200 6 53. Eisman JA, Barkla DH, Tutton PJ. 1987. Suppression of in vivo growth of human cancer solid tumor xenografts by 1,25 dihydroxyvitamin D3 Cancer Res. 47:21 5
132 54. Eisman JA, Koga M, Sutherland RL, Barkla DH, Tutton PJ. 1989. 1,25 Dihyd roxyvitamin D3 and the regulation of human cancer cell replication. Proc. Soc. Exp. Biol. Med. 191:221 6 55. Erben RG, Soegiarto DW, Weber K, Zeitz U, Lieberherr M, Gniadecki R, Moller G, Adamski J, Balling R. 2002. Deletion of deoxyribonucleic acid binding d omain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol. 16:1524 37 56. Fu W, Begley JG, Killen MW, Mattson MP. 1999. Anti apoptotic role of telomerase in pheochromocytoma cells. J Biol Chem. 274:7264 71 57. Fu W, K illen M, Culmsee C, Dhar S, Pandita TK, Mattson MP. 2000. The catalytic subunit of telomerase is expressed in developing brain neurons and serves a cell survival promoting function. J Mol Neurosci. 14:3 15 58. Giguere V. 1999. Orphan nuclear receptors: from g ene to function. Endocrine Rev. 20 : 689 725 59. Gill RK, Atkins LM, Hollis BW, Bell NH 1998. Mapping the domains of the interaction of the vitamin D receptor and steroid receptor coactivator 1. Mol Endocrinol. 12:57 65 60. Gniadecki R. 1996. Activation of Raf m itogen activated protein kinase signaling pathway by 1,25 dihydroxyvitamin D3 in normal human keratinocytes. J Invest Dermatol. 106:1212 7 61. Gniadecki R. 1998. Involvement of Src in the vitamin D signaling in human keratinocytes. Biochem Pharmacol. 55:499 5 03
133 62. Gonzalez EA, Disthabanchong S, Kowalewski R, Martin KJ. 2002. Mechanisms of the regulation of EGF receptor gene expression by calcitriol and parathyroid hormone in UMR 106 01 cells. Kidney Int. 61:1627 34 63. Gorbunova V, Seluanov A, Pereira Smith OM. 200 2. Expression of human telomerase (hTERT) does not prevent stress induced senescence in normal human fibroblasts but protects the cells from stress induced apoptosis and necrosis. J Biol Chem. 277:38540 9 64. Green S, Walter P, Kumar V, Krust A, Bornert JM, A rgos P, Chambon P. 1986. Human oestrogen receptor cDNA: sequence, expression and homology to v erb A. Nature 320: 134 9 65. Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC. 1998. A phase I study of the vitamin D analogue EB 1089 in patient s with advanced breast and colorectal cancer. Br J Cancer. 78:6 13 66. Guzey M, Kitada S, Reed JC. 2002. Apoptosis induction by 1alpha,25 dihydroxyvitamin D3 in prostate cancer. Mol Cancer Ther. 1:667 77 67. Guzey M, Sattler C, DeLuca HF. 1998. Combinational eff ects of vitamin D3 and retinoic acid (all trans and 9 cis) on proliferation, differentiation, and programmed cell death in two small cell lung carcinoma cell lines. Biochem Biophys Res Commun. 249:735 44 68. Hager G, Formanek M, Gedlicka C, Thurnher D, Knerer B, Kornfehl J. 2001. 1,25(OH)2 vitamin D3 induces elevated expression of the cell cycle regulating genes P21 and P27 in squamous carcinoma cell lines of the head and neck. Acta Otolaryngol. 121:103 9
134 69. Hahn WC, Counter CM, Lundberg AS, Beijersbergen RL, Br ooks MW, Weinberg RA. 1999. Creation of human tumour cells with defined genetic elements. Nature. 400:464 8 70. Haussler MR, Whitfield GK, Haussler CA, Hsieh JC, Thompson PD, Selznick SH, Dominguez CE, Jurutka PW. 1998. The nuclear vitamin D receptor: biolog ical and molecular regulatory properties revealed. J Bone Miner Res. 13:325 49 71. Hilliard GM 4th, Cook RG, Weigel NL, Pike JW. 1994. 1,25 dihydroxyvitamin D3 modulates phosphorylation of serine 205 in the human vitamin D receptor: site directed mutagenesis of this residue promotes alternative phosphorylation. Biochemistry. 33:4300 11 72. Hollander MC, Alamo I, Jackman J, Wang MG, McBride OW, Fornace AJ Jr. 1993. Analysis of the mammalian gadd45 gene and its response to DNA damage. J. Biol. Chem. 268:24385 93 73. H ollander MC, Sheikh MS, Bulavin DV, Lundgren K, Augeri Henmueller L, Shehee R, Molinaro TA, Kim KE, Tolosa E, Ashwell JD, Rosenberg MP, Zhan Q, Fernandez Salguero PM, Morgan WF, Deng CX, Fornace AJ Jr. 1999. Genomic instability in Gadd45a deficient mice. Nat Genet. 23:176 84 74. Hollander MC, Kovalsky O, Salvador JM, Kim KE, Patterson AD, Haines DC, Fornace AJ Jr. 2001. Dimethylbenzanthracene carcinogenesis in Gadd45a null mice is associated with decreased DNA repair and increased mutation frequency. Cancer R es. 61:2487 91
135 75. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Oro AE, Lebo R, Thompson EB, Rosenfeld MG, Evans RM. 1985. Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318: 635 41 76. Hsieh JC, Jurutka PW, Galliga n MA, Terpening CM, Haussler CA, Samuels DS, Shimizu Y, Shimizu N, Haussler MR. 1991. Human vitamin D receptor is selectively phosphorylated by protein kinase C on serine 51, a residue crucial to its trans activation function. Proc Natl Acad Sci U S A. 88 :9315 9 77. Hsieh T, Wu JM 1997. Induction of apoptosis and altered nuclear/cytoplasmic distribution of the androgen receptor and prostate specific antigen by 1alpha,25 dihydroxyvitamin D3 in androgen responsive LNCaP cells. Biochem Biophys Res Commun. 235:53 9 44 78. Ikeda N, Uemura H, Ishiguro H, Hori M, Hosaka M, Kyo S, Miyamoto K, Takeda E, Kubota Y. 2003. Combination treatment with 1alpha,25 dihydroxyvitamin D3 and 9 cis retinoic acid directly inhibits human telomerase reverse transcriptase transcription in p rostate cancer cells. Mol Cancer Ther. 2:739 46. 79. Issa LL, Leong GM, Eisman JA. 1998. Molecular mechanism of vitamin D receptor action. Inflamm Res. 47:451 75 80. James SY, Mackay AG, Binderup L, Colston KW. 1994. Effects of a new synthetic vitamin D analogu e, EB1089, on the oestrogen responsive growth of human breast cancer cells. J. Endocrinol. 141 :555 63
136 81. James SY, Mackay AG, Colston KW 1995. Vitamin D derivatives in combination with 9 cis retinoic acid promote active cell death in breast cancer cells. J M ol Endocrinol. 14:391 4 82. James SY, Mackay AG, Colston KW. 1996. Effects of 1,25 dihydroxyvitamin D3 and its analogues on induction of apoptosis in breast cancer cells. J Steroid Biochem Mol Biol. 58:395 401 83. Jin S, Tong T, Fan W, Fan F, Antinore MJ, Zhu X, Mazzacurati L, Li X, Petrik KL, Rajasekaran B, Wu M, and Zhan Q. 2002. GADD45 induced cell cycle G2 M arrest associates with altered subcellular distribution of cyclin B1 and is independent of p38 kinase activity. Oncogene 21 :8696 704 84. Jin S, Zhao H, Fan F, Blanck P, Fan W, Colchagie AB, Fornace AJ Jr, Zhan Q 2000. BRCA1 activation of the GADD45 promoter. Oncogene 19 : 4050 57 85. Johnson CS, Hershberger PA, Trump DL. 2002. Vitamin D related therapies in prostate cancer. Cancer Metastasis Rev. 21:147 58 86. Johns on JA, Grande JP, Roche PC, Kumar R. 1996. Immunohistochemical detection and distribution of the 1,25 dihydroxyvitamin D3 receptor in rat reproductive tissues. Histochem Cell Biol. 105 :7 15 87. Jurutka PW, Hsieh JC, Nakajima S, Haussler CA, Whitfield GK, Haus sler MR. 1996. Human vitamin D receptor phosphorylation by casein kinase II at Ser 208 potentiates transcriptional activation. Proc Natl Acad Sci U S A. 93:3519 24 88. Kastner P, Mark M, Chambon P. 1995. Nopnsteroid nuclear receptors: what are genetic studie s telling us about their role in real life? Cell 83 : 859 69
137 89. Kitano Y, Ikeda N, Okano M. 1991. Suppression of proliferation of human epidermal keratinocytes by 1,25 dihydroxyvitamin D3. Analysis of its effect on psoriatic lesion and of its mechanism using human keratinocytes in culture. Eur J Clin Invest. 21:53 8 90. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematsu Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S. 2003. The chromati n remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell.113:905 17 91. Kobayashi T, Hashimoto K, Yoshikawa K. 1998. Growth inhibition of human keratinocytes by a new vitamin D3 analogue in vitro. J Dermato l Sci. 16:158 64 92. Kolquist KA, Ellisen LW, Counter CM, Meyerson M, Tan LK, Weinberg RA, Haber DA, Gerald WL 1998. Expression of TERT in early premalignant lesions and a subset of cells in normal tissues. Nat Genet. 19:182 6 93. Koszewski NJ, Ashok S, Russell J 1999. Turning a negative into a positive: vitamin D receptor interactions with the avian parathyroid hormone response element. Mol Endocrinol. 13:455 65 94. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gutafsson JA. 1997. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors a and b Endocrinol. 138 : 863 870 95. Kumar R, Thompson EB. 1999. The structure of the nuclear hormone receptors. Steroids. 64:310 9
138 96. Kyo S, Kanaya T, Takakura M, Tanaka M, Yamashita A, Inoue H, Inoue M. 1999. Expression of human telomerase subunits in ovarian malignant, borderline and benign tumors. Int J Cancer. 80:804 9 97. Kyo S, Takakura M, Kanaya T, Zhuo W, Fujimoto K, Nishio Y, Orimo A, Inoue M. 1999. Estrogen activat es telomerase. Cancer Res. 59:5917 21 98. Le Menuet D, Zennaro MC, Viengchareun S, Lombes M. 2000. Transgenic mouse models to study human mineralocorticoid receptor function in vivo. Kidney Int. 57:1299 306 99. Lee H, Bai W. 2002. Regulation of estrogen recepto r nuclear export by ligand induced and p38 mediated receptor phosphorylation. Mol Cell Biol. 22:5835 45 100. Lee H, Jiang F, Wang Q, Nicosia SV, Yang J, Su B, Bai W. 2000. MEKK1 activation of human estrogen receptor alpha and stimulation of the agonistic activ ity of 4 hydroxytamoxifen in endometrial and ovarian cancer cells. Mol Endocrinol. 14:1882 96 101. Lefkowitz ES, Garland CF. 1994. Sunlight, vitamin D, and ovarian cancer mortality rates in US women. I nt. J. Epidemiol 23 :1133 36 102. Lewis BP, Shih IH, Jones Rho ades MW, Bartel DP, Burge CB. 2003. Prediction of mammalian microRNA targets. Cell. 115:787 98 103. Li P, Nicosia SV, Bai W. 2001. Antagonism between PTEN/MMAC1/TEP 1 and androgen receptor in growth and apoptosis of prostatic cancer cells. J Biol Chem. 276:20 444 50 104. Li P, Lee H, Guo S, Unterman TG, Jenster G, Bai W. 2003. AKT independent protection of prostate cancer cells from apoptosis mediated through
139 complex formation between the androgen receptor and FKHR. Mol Cell Biol. 23:104 18 105. Liu JP. 1999. Studies of the molecular mechanisms in the regulation of telomerase activity. FASEB J. 13:2091 104 106. Liu SM, Koszewski N, Lupez M, Malluche HH, Olivera A, Russell J. 1996. Characterization of a response element in the 5' flanking region of the avian (chicken) PTH g ene that mediates negative regulation of gene transcription by 1,25 dihydroxyvitamin D3 and binds the vitamin D3 receptor. Mol Endocrinol. 10:206 15 107. Louie MC, Yang HQ, Ma AH, Xu W, Zou JX, Kung HJ, Chen HW. 2003. Androgen induced recruitment of RNA polyme rase II to a nuclear receptor p160 coactivator complex. Proc Natl Acad Sci U S A. 100:2226 30 108. Lubahn DB, Joseph DR, Sullivan PM, Willard HF, French FS, Wilson EM. 1988. Cloning of human androgen receptor complementary DNA and localization to the X chromo some. Science 240:327 330 109. MacDonald PN, Baudino TA, Tokumaru H, Dowd DR, Zhang C. 2001. Vitamin D receptor and nuclear receptor coactivators: crucial interactions in vitamin D mediated transcription. Steroids. 66:171 6 110. Maeda T, Sim AB, Leedel DA, Chua PP, Chomey EG, Luong L, Tron VA. 2003. UV induces GADD45 in a p53 dependent and independent manner in human keratinocytes. J Cutan Med Surg. 7:119 23 111. Maenpaa PH, Vaisanen S, Jaaskelainen T, Ryhanen S, Rouvinen J, Duchier C, Mahonen A. 2001. Vitamin D(3) an alogs (MC 1288, KH 1060, EB 1089, GS 1558, and CB 1093): studies on their mechanism of action. Steroids. 66:223 5
140 112. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR, Mangelsdorf DJ 2002. Vitamin D receptor as an intestinal bile acid sensor. Science 296:1313 16 113. Mangelsdorf DJ, Borgmeyer U, Heyman RA, Zhou JY, Ong ES, Oro AE, Kakizuka A, Evans RM. 1992. Characterization of three RXR genes that mediate the action of 9 cis retinoic acid. Genes Dev. 6:329 44 114. Mangelsdorf DJ, Thummel C, B eato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, et al 1995. The nuclear receptor superfamily: the second decade. Cell. 83:835 9 115. Marcinkowska E, Wiedlocha A, Radzikowski C. 1997 1,25 Dihydroxyvitamin D3 induced activatio n and subsequent nuclear translocation of MAPK is upstream regulated by PKC in HL 60 cells. Biochem Biophys Res Commun. 241:419 26 116. Masutomi K, Yu EY, Khurts S, Ben Porath I, Currier JL, Metz GB, Brooks MW, Kaneko S, Murakami S, DeCaprio JA, Weinberg RA, St ewart SA, Hahn WC. 2003. Telomerase maintains telomere structure in normal human cells. Cell. 114:241 53 117. Mathiasen IS, Lademann U, Jaattela M. 1999. Apoptosis induced by vitamin D compounds in breast cancer cells is inhibited by Bcl 2 but does not involv e known caspases or p53. Cancer Res. 59:4848 56 118. Mathiasen IS, Sergeev IN, Bastholm L, Elling F, Norman AW, Jaattela M. 2002. Calcium and calpain as key mediators of apoptosis like death induced by vitamin D compounds in breast cancer cells. J Biol Chem. 2 77:30738 45
141 119. Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW. 1997. Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol. 11:1165 79 120. Morri son NA, Shine J, Fragonas JC, Verkest V, McMenemy ML, Eisman JA. 1989. 1,25 dihydroxyvitamin D responsive element and glucocorticoid repression in the osteocalcin gene. Science 246:1158 61 121. Muller M, Renkawitz R. 1991. The glucocorticoid receptor Biochim Biophys. Acta 1088:171 82 122. Nakamura TM, Morin GB, Chapman KB, Weinrich SL, Andrews WH, Lingner J, Harley CB, Cech TR. 1997. Telomerase catalytic subunit homologs from fission yeast and human. Science. 277:955 9 123. Nelson P, Kiriakidou M, Sharma A, Maniatak i E, Mourelatos Z. 2003. The microRNA world: small is mighty. Trends Biochem Sci. 28:534 40 124. Newbold RF. 2002. The significance of telomerase activation and cellular immortalization in human cancer. Mutagenesis. 17:539 50 125. Nishikawa J, Kitaura M, Matsumoto M, Imagawa M, Nishihara T. 1994. Difference and similarity of DNA sequence recognized by VDR homodimer and VDR/RXR heterodimer. Nucleic Acids Res. 22:2902 7 126. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT. 1990. Identification of a DNA sequen ce responsible for binding of the 1,25 dihydroxyvitamin D3 receptor and 1,25 dihydroxyvitamin D3 enhancement of mouse
142 secreted phosphoprotein 1 (SPP 1 or osteopontin) gene expression. Proc. Natl. Acad. Sci. U. S. A. 87:9995 9 127. Oda Y, Sihlbom C, Chalkley RJ, Huang L, Rachez C, Chang CP, Burlingame AL, Freedman LP, Bikle DD. 2003. Two distinct coactivators, DRIP/mediator and SRC/p160, are differentially involved in vitamin D receptor transactivation during keratinocyte differentiation. Mol Endocrinol. 17:2329 39 128. Park TW, Riethdorf S, Riethdorf L, Loning T, Janicke F. 1999. Differential telomerase activity, expression of the telomerase catalytic sub unit and telomerase RNA in ovarian tumors. Int J Cancer. 84:426 31 129. Peet DJ, Turley SD, Ma W, Janowski BA, Lobacca ro JM, Hammer RE, Mangelsdorf DJ. 1998. Cholesterol and bile acid metabolism are impaired in mice lacking the nuclear oxysterol receptor LXR alpha. Cell. 93:693 704 130. Pratt WB, Czar MJ, Stancato LF, Owens JK. 1993. The hsp56 immunophilin component of stero id receptor heterocomplexes: could this be the elusive nuclear localization signal binding protein? J Steroid Biochem Mol Biol. 46:269 79 131. Prudencio J, Akutsu N, Benlimame N, Wang T, Bastien Y, Lin R, Black MJ, Alaoui Jamali MA, White JH. 2001. Action of low calcemic 1alpha,25 dihydroxyvitamin D3 analogue EB1089 in head and neck squamous cell carcinoma. J Natl Cancer Inst. 93:745 53 132. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument Bromage H, Tempst P, Freedman LP. 1999. Ligand depe ndent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824 8
143 133. Ray DW, Davis J RE, White A, Clark AJL. 1996. Glucocorticoid receptor structure and function in glucocorticoid resistant small cell lung carcinoma cells. Cance r Res. 56: 3276 80 134. Ren JG, Xia HL, Tian YM, Just T, Cai GP, Dai YR. 2001. Expression of telomerase inhibits hydroxyl radical induced apoptosis in normal telomerase negative human lung fibroblasts. FEBS Lett. 488:133 8 135. Rivers C, Levy A, Hancock J, Lightman S, Norman M. 1999. Insertion of an amino acid in the DNA binding domain of the glucocorticoid receptor as a result of alternative splicing. J Clin Endocr Metab. 84:4283 6 136. Robinson Rechavi M, Escriva Garcia H, Laudet V. 2003. The nuclear receptor supe rfamily. J Cell Sci. 116:585 6 137. Rodriguez G. 2003 New insights regarding pharmacologic approaches for ovarian cancer prevention. Hematol Oncol Clin North Am. 17:1007 20 138. Rowland Goldsmith MA, Holmquist B, Henry HL. 1999. Genomic cloning, structure, and re gulatory elements of the 1 alpha, 25(OH)2D3 down regulated gene for cyclic AMP dependent protein kinase inhibitor. Biochim Biophys Acta. 1446:414 8 139. Roy S, Martel J, Tenenhouse HS 1995. Comparative effects of 1,25 dihydroxyvitamin D3 and EB 1089 on mouse r enal and intestinal 25 hydroxyvitamin D3 24 hydroxylase. J Bone Miner Res. 10:1951 9 140. Sambrook PN, Eisman JA 1994. Prediction of bone density from vitamin D receptor alleles. Nature 367:284 7
144 141. Saretzki G. 2003. Telomerase inhibition as cancer therapy. Canc er Lett. 194:209 19 142. Sato T, Takusagawa K, Asoo N, Konno K. 1982. Antitumor effect of 1 alpha hydroxyvitamin D3. Tohoku J. Exp. Med 138:445 6 143. Saunders DE, Christensen C, Lawrence WD, Malviya VK, Malone JM, Williams JR, Deppe G. 1992. Receptors for 1,25 dihydroxyvitamin D3 in gynecologic neoplasms. Gynecol Oncol 44:131 6 144. Saunders DE, Christensen C, Wappler NL, Schultz JF, Lawrence WD, Malviya VK, Malone JM, Deppe G. 1993. Inhibition of c myc in breast and ovarian carcinoma cells by 1,25 dihydroxyvitamin D3, retinoic acid and dexamethasone. Anticancer Drugs. 4:201 8 145. Saunders DE, Christensen C, Williams JR, Wappler NL, Lawrence WD, Malone JM, Malviya VK, Deppe G. 1995. Inhibition of breast and ovarian carcinoma cell growth by 1,25 dihydroxyvitamin D3 com bined with retinoic acid or dexamethasone. Anticancer Drugs. 6:562 9 146. Savoysky E, Yoshida K, Ohtomo T, Yamaguchi Y, Akamatsu K, Yamazaki T, Yoshida S, Tsuchiya M. 1996. Down regulation of telomerase activity is an early event in the differentiation of HL6 0 cells. Biochem Biophys Res Commun. 226:329 34 147. Scaglione Sewell BA, Bissonnette M, Skarosi S, Abraham C, Brasitus TA. 2000. A vitamin D3 analog induces a G1 phase arrest in CaCo 2 cells by inhibiting cdk2 and cdk6: roles of cyclin E, p21Waf1, and p27Kip 1. Endocrinology. 141:3931 9
145 148. Segaert S, Bouillon R. 1998. Vitamin D and regulation of gene expression. Curr Opin Clin Nutr Metab Care. 1:347 54 149. Simboli Campbell M, Narvaez CJ, Tenniswood M, Welsh J. 1996. 1,25 Dihydroxyvitamin D3 induces morphological an d biochemical markers of apoptosis in MCF 7 breast cancer cells. J. Steroid Biochem. Mol. Biol. 58:367 76 150. Smith ML, Chen IT, Zhan Q, Bae I, Chen CY, Gilmer TM, Kastan MB, O'Connor PM, Fornace AJ Jr. 1994. Interaction of the p53 regulated protein Gadd45 wi th proliferating cell nuclear antigen. Science 266 :1376 80 151. Sutter AP, Maaser K, Barthel B, Scherubl H. 2003. Ligands of the peripheral benzodiazepine receptor induce apoptosis and cell cycle arrest in oesophageal cancer cells: involvement of the p38MAPK s ignalling pathway. Br J Cancer. 89:564 72 152. Sutton AL, MacDonald PN. 2003. Vitamin D: more than a "bone a fide" hormone. Mol Endocrinol. 17:777 91 153. Toell A, Polly P, Carlberg C. 2000. All natural DR3 type vitamin D response elements show a similar function ality in vitro. Biochem J. 352:301 9 154. Vairapandi M, Azam N, Balliet AG, Hoffman B, Liebermann DA. 2000. Characterization of MyD118, Gadd45, and proliferating cell nuclear antigen (PCNA) interacting domains. PCNA impedes MyD118 AND Gadd45 mediated negative growth control. J Biol Chem. 275:16810 9 155. Vairapandi M., Balliet A.G., Fornace A.J., Hoffman Jr, B. and Liebermann D.A. 1996. The differentiation primary response gene MYD118 related to GADD45 encodes for a nuclear protein which interacts with PCNA and p 21 WAF1/CIP1 Oncogene 12 :2579 94
146 156. Vaziri H, Benchimol S. 1998. Reconstitution of telomerase activity in normal human cells leads to elongation of telomeres and extended replicative life span. Curr Biol. 8:279 82 157. Villa R, Folini M, Perego P, Supino R, Sett i E, Daidone MG, Zunino F, Zaffaroni N. 2000. Telomerase activity and telomere length in human ovarian cancer and melanoma cell lines: correlation with sensitivity to DNA damaging agents. Int J Oncol. 16:995 1002 158. Villena Heinsen C, Meyberg R, Axt Fliedner R, Reitnauer K, Reichrath J, Friedrich M. 2002. Immunohistochemical analysis of 1,25 dihydroxyvitamin D3 receptors, estrogen and progesterone receptors and Ki 67 in ovarian carcinoma. Anticancer Res. 22:2261 7 159. Wang XW, Zhan Q, Coursen JD, Khan MA, Kontny HU, Yu L, Hollander MC, O'Connor PM, Fornace AJ Jr, Harris CC. 1999. GADD45 induction of a G2/M cell cycle checkpoint. Proc. Natl. Acad. Sci. U. S. A. 96:3706 11 160. Waterhouse J et al 1976. Cancer Incidence in Five Continents Volume 2, International Agenc y for Research on Cancer Control (IRAC Scientific Publication No. 15), Lyon, France 161. Waterhouse J et al 1987. Cancer Incidence in Five Continents Volume 5, International Agency for Research on Cancer Control (IRAC Scientific Publication No. 88), Lyon, Fra nce 162. Webb AR, Kline L, Holick MF. 1988. Influence of season and latitude on the cutaneous synthesis of vitamin D3: exposure to winter sunlight in Boston and
147 Edmonton will not promote vitamin D3 synthesis in human skin. J Clin Endocrinol Metab. 67:373 8 163. We inberger C, Hollenberg SM, Rosenfeld MG, Evans RM. 1985. Domain structure of human glucocorticoid receptor and its relationship to the v erb A oncogene product. Nature. 318:670 2 164. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DJ, Evans RM. 1986. The c erb A gene encodes a thyroid hormone receptor. Nature 324:641 6 165. Welsh J. 1994. Induction of apoptosis in breast cancer cells in response to vitamin D and antiestrogens. Biochem Cell Biol. 72:537 45 166. Welsh J, Wietzke JA, Zinser GM, Byrne B, Smith K, Narvaez CJ. 2003. Vitamin D 3 receptor as a target for breast cancer prevention. J Nutr.133:2425S 2433S 167. White LK, Wright WE, Shay JW. 2001. Telomerase inhibitors. Trends Biotechnol. 19:114 20 168. Wu KJ, Grandori C, Amacker M, Simon Vermot N, Polack A, Lingner J, Dal la Favera R 1999. Direct activation of TERT transcription by c MYC. Nat Genet. 21:220 4 169. Wynford Thomas D. 1999. Cellular senescence and cancer. J Pathol. 187:100 11 170. Xie W, Radominska Pandya A, Shi Y, Simon CM, Nelson MC, Ong ES, Waxman DJ, Evans RM. 2001 An essential role for nuclear receptors SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci U S A. 98:3375 80 171. Xu D, Gruber A, Peterson C, Pisa P. 1996. Supression of telomerase activity in HL60 cells after treatment with differenti ating agents. Leukemia. 10:1354 7
148 172. Yaginuma Y, Westphal H. 1992. Abnormal structure and expression of the p53 gene in human ovarian carcinoma cell lines. Cancer Res. 52:4196 9 173. Yamamoto H, Shevde NK, Warrier A, Plum LA, DeLuca HF, Pike JW. 2003. 2 Methylen e 19 nor (20S) 1,25 dihydroxyvitamin D3 potently stimulates gene specific DNA binding of the vitamin D receptor in osteoblasts. J Biol. Chem. 278:31756 65 174. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiy ama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S. 1997. Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet. 16:391 6 175. Zhan Q, Antinore MJ, Wang XW, Carrier F, Smith ML Harris CC, Fornace AJ Jr. 1999. Association with Cdc2 and inhibition of Cdc2/Cyclin B1 kinase activity by the p53 regulated protein Gadd45. Oncogene 18:2892 900 176. Zhan Q, Lord KA, Alamo I Jr, Hollander MC, Carrier F, Ron D, Kohn KW, Hoffman B, Liebermann DA, Fornace AJ Jr. 1994. The gadd and MyD genes define a novel set of mammalian genes encoding acidic proteins that synergistically suppress cell growth. Mol. Cell Biol. 14:2361 71 177. Zhang P, Chan SL, Fu W, Mendoza M, Mattson MP. 2003. TERT suppresses apo ptotis at a premitochondrial step by a mechanism requiring reverse transcriptase activity and 14 3 3 protein binding ability. FASEB J. 17:767 9 178. Zhang X, Ma L, Enkemann SA, Pledger WJ. 2003. Role of Gadd45alpha in the density dependent G1 arrest induced by p27(Kip1). Oncogene. 22:4166 7
149 179. Zhang X, Mar V, Zhou W, Harrington L, Robinson MO. 1999. Telomere shortening and apoptosis in telomerase inhibited human tumor cells. Genes Dev. 13:2388 99 180. Zhao XY, Feldman D. 2001. The role of vitamin D in prostate cancer Steroids. 66:293 300 181. A unified nomenclature system for the nuclear receptor superfamily. 1999. Cell 97:161 3. Nuclear receptor nomenclature committee
About the Author Feng Jiang received her M.D. degree in Xian Medical University, China in 1994 and a M.S. in Pathology from Immunopathology Institute of Xian Medical University in 1997. She was a resident in the Department of Ophthal mology of the Second Affiliated Hospital of Xian Medical University from 1997 1999. She entered the Ph.D. program at Dr. Bais laboratory at the University of South Florida in 1999. She has several publications and participated in the 94 th and 95 th Annual Meetings of the American Association of Cancer Research in 2003 and 2004. She received an Outstanding graduate student award in 2001 and Superior presenters award at the Health Science Center Research Day in 2003 and 2004.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001469426
007 cr mnu|||uuuuu
008 040524s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000327
Molecular mechanism of vitamin D action and its implications in ovarian cancer prevention and therapy
h [electronic resource] /
by Feng Jiang.
[Tampa, Fla.] :
University of South Florida,
Thesis (Ph.D.)--University of South Florida, 2004.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 161 pages.
ABSTRACT: 1,25-dihydroxyvitamin D3 (1,25VD), the active form of vitamin D (VD), suppresses the growth of numerous human cancer cell lines by inhibiting cell cycle progression and inducing cell death. Genes that mediate each of these activities remain largely unidentified and there are no preclinical data for 1,25VD analogues in ovarian cancer (OCa). We hypothesize that 1,25VD and its analogues inhibit the development of OCa. In this study, we demonstrated, (a) 1,25VD causes cell cycle arrest at the G1/S and G2/M transition and induces apoptosis in OCa cells. (b) We also found that gadd45 is one of primary target genes for 1,25VD-mediated G2/M arrest. A direct repeat 3 (DR3) vitamin D response element (VDRE) is identified in the fourth exon of gadd45. This exonic VDRE forms a complex with the vitamin D receptor (VDR)/retinoid X receptor (RXR) heterodimer in vitro and mediates the induction of reporter activity by 1,25VD in vivo. VDR is recruited in a ligand-dependent manner to the exonic enhancer but not to the gadd45 promoter regions. In OCa cells expressing GADD45 anti-sense cDNA or GADD45-null mouse embryo fibroblasts, 1,25VD fails to induce G2/M arrest, suggesting that G2/M arrest induced by 1,25VD is mediated through GADD45. Further study showed that GADD45 mediates the effect of 1,25VD by decreasing cdc2 kinase activity. (c) hTERT, the catalytic subunit of telomerase, is identified as a primary target for 1,25VD. 1,25VD decreases telomerase activity and hTERT mRNA expression. The down-regulation of hTERT mRNA is due to decreased mRNA stability by 1,25VD, rather than decreased transcription of hTERT through VDRE. Clones stably transfected with hTERT showed higher telomerase activity and longer telomere length than parental cells. Moreover, hTERT clones resist 1,25VD-induced apoptosis and growth inhibition. In contrast to parental cells which do not recover from prolonged treatment with 1,25VD, hTERT clones re-grew rapidly after 1,25VD withdrawal. (d) We demonstrated that the 1,25VD analogue EB1089 inhibits OCa cells in vitro and OCa xenograft in vivo without inducing hypercalcemia. We also demonstrated precursors for epithelial OCa express VDR and human primary ovarian surface epithelial cells respond to 1,25VD. Taken together, these results strongly suggest that 1,25VD analogues may be effective in the chemoprevention and chemotherapy of OCa.
Adviser: Wenlong Bai
Vitamin D receptor.
x Pathology and Laboratory Medicine
t USF Electronic Theses and Dissertations.