This Article Abstract Full Text (PDF) All Versions of this Article: 26/5/662    most recent Final Manuscript Author Manuscript Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Nagpal, S. Articles by Rathnachalam, R. Search for Related Content PubMed PubMed Citation Articles by Nagpal, S. Articles by Rathnachalam, R. Pubmed/NCBI databases Compound via MeSH Substance via MeSH

Noncalcemic Actions of Vitamin D Receptor Ligands

Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana 46285

Correspondence: Address all correspondence and requests for reprints to: Sunil Nagpal, Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana 46285. E-mail: nagpal_sunil{at}lilly.com

	Abstract

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
1,25-Dihydroxyvitamin D₃ [1,25-(OH)₂D₃], the active metabolite of vitamin D₃, is known for the maintenance of mineral homeostasis and normal skeletal architecture. However, apart from these traditional calcium-related actions, 1,25-(OH)₂D₃ and its synthetic analogs are being increasingly recognized for their potent antiproliferative, prodifferentiative, and immunomodulatory activities. These actions of 1,25-(OH)₂D₃ are mediated through vitamin D receptor (VDR), which belongs to the superfamily of steroid/thyroid hormone nuclear receptors. Physiological and pharmacological actions of 1,25-(OH)₂D₃ in various systems, along with the detection of VDR in target cells, have indicated potential therapeutic applications of VDR ligands in inflammation (rheumatoid arthritis, psoriatic arthritis), dermatological indications (psoriasis, actinic keratosis, seborrheic dermatitis, photoaging), osteoporosis (postmenopausal and steroid-induced osteoporosis), cancers (prostate, colon, breast, myelodysplasia, leukemia, head and neck squamous cell carcinoma, and basal cell carcinoma), secondary hyperparathyroidism, and autoimmune diseases (systemic lupus erythematosus, type I diabetes, multiple sclerosis, and organ transplantation). As a result, VDR ligands have been developed for the treatment of psoriasis, osteoporosis, and secondary hyperparathyroidism. Furthermore, encouraging results have been obtained with VDR ligands in clinical trials of prostate cancer and hepatocellular carcinoma. This review deals with the molecular aspects of noncalcemic actions of vitamin D analogs that account for the efficacy of VDR ligands in the above-mentioned indications.

I. Introduction

II. VDR and the Regulation of Gene Expression

A. VDR and its functional unit

B. Regulation of gene expression

C. VDR cofactors

III. VDR Crystal Structure

IV. VDR Knockout Animals

V. Vitamin D Action in Autoimmune Disease Models

A. Multiple sclerosis (MS)

B. Rheumatoid arthritis (RA)

C. Inflammatory bowel diseases (IBDs)

D. Type I diabetes

E. Systemic lupus erythematosus (SLE)

F. Transplant rejection

VI. Vitamin D Action in Keratinocytes and Psoriasis

VII. Vitamin D Action on Prostate Cancer Cells

VIII. Vitamin D Action on Breast Cancer Cells

IX. Vitamin D Action on Colon Cancer Cells

X. Vitamin D Action on Leukemic Cells

XI. Vitamin D Action in Squamous Cell Carcinoma

XII. Vitamin D Action in Kaposi’s Sarcoma

XIII. Vitamin D Action on Bone and Osteoporosis

XIV. Vitamin D Action on Blood Pressure

XV. Conclusions

	I. Introduction

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
THE BIOLOGICAL ACTIONS of the hormonally active form of vitamin D₃, 1,25-dihydroxyvitamin D₃ [1,25-(OH)₂D₃] or calcitriol (Fig. 1), and its synthetic analogs are mediated by the nuclear vitamin D receptor (VDR). VDR is a ligand-dependent transcription factor belonging to the superfamily of steroid/thyroid hormone receptors (1) and has traditionally been associated with calcemic activities, namely, calcium and phosphorus homeostasis and maintenance of bone content. However, the observation that VDR is also present in cells other than those of the intestine, bone, kidney, and parathyroid gland led to the recognition of noncalcemic actions of VDR ligands. As a result, VDR is also known to be involved in cell proliferation, differentiation, and immunomodulation. For example, activated T and B lymphocytes, rheumatoid arthritis (RA) synoviocytes and macrophages, Kaposi’s sarcoma (KS), and prostate, breast, and colon cancer cells exhibit increased levels of VDR protein when compared with their normal counterparts. This activation or disease-specific up-regulation of VDR protein provides an opportunity to treat these conditions with VDR ligands. The expression of VDR in a variety of cell lines and primary cells, coupled with the increased evidence regarding the involvement of VDR in the processes of cell differentiation, inhibition of proliferation, and immunoregulation, has prompted testing of the therapeutic effect of VDR ligands in several human diseases (Table 1) as well as in various animal models of diseases. These efforts have led to the development of VDR ligands for the treatment of psoriasis (2, 3, 4), secondary hyperparathyroidism (5), and osteoporosis (6, 7). In addition, VDR ligands have shown some efficacy in limited open clinical trials for prostate cancer, myelodysplasia (a precancerous state), psoriatic arthritis, and RA. Examples of vitamin D analogs that have undergone clinical trials with positive outcome are shown in Table 1. VDR ligands have also shown efficacy in the prevention and treatment of inflammatory and autoimmune diseases in various animal models. The goal of this article is to review the progress in the field of noncalcemic actions of vitamin D and its analogs with a particular emphasis on the current and potential therapeutic applications of VDR ligands. To limit the references to a reasonable number, many recent up-to-date reviews are included, sometimes in place of relevant articles. We apologize to our colleagues when, due to lack of space, a recent review article is mentioned instead of the multiple original references.

View larger version (27K): [in this window] [in a new window]   FIG. 1. 1,25-(OH)2D3 and its synthetic analogs. Structures, common names, and chemical names of vitamin D analogs are presented.

	II. VDR and the Regulation of Gene Expression

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

B. Regulation of gene expression
VDR is a ligand-dependent transcription factor that can modulate the expression of vitamin D-responsive genes in three different ways (Fig. 2). It can positively regulate the expression of certain genes by binding to the VDREs present in their promoter regions (1, 8), or negatively regulate the expression of other genes by binding to negative VDREs (14, 15), or inhibit the expression of some genes by antagonizing the action of certain transcription factors, such as nuclear factor (NF)-AT and NF-B (16, 17, 18). Genes whose expression is induced by VDR ligands, and which are known to contain a VDRE in their promoter, are presented in Fig. 2 along with their known function. These genes include osteocalcin, osteopontin, receptor activator of NF-B ligand (RANKL), and carbonic anhydrase II, which are involved in extracellular bone matrix formation and bone remodeling (1, 19, 20, 21, 22, 23, 24). Other genes that contain a VDRE in their promoter region and show vitamin D-dependent up-regulation in their expression are the cell adhesion molecule ß3 integrin, tumor suppressor p21, calbindin-9k, 24-hydroxylase, human CYP3A4 and its rat and mouse CYP counterparts, involucrin, phospholipase C (PLC) 1, and IGF binding protein (IGFBP)-3 (1, 12, 25, 26, 27, 28, 29, 30, 31, 32). Genes that are down-regulated in response to 1,25-(OH)₂D₃ and its synthetic analogs are also listed in Fig. 2. The known hyperproliferative and inflammatory functions of these gene products indicate that many of the therapeutic effects of 1,25-(OH)₂D₃ and its analogs could result from their negative gene regulatory or transrepression activities. VDR ligands have been documented to inhibit the expression of cytokines, namely, IL-2 [T cell lines and peripheral blood mononuclear cells (PBMCs)] (33, 34), IL-12 (myelomonocytes) (35), TNF- (PBMCs) (34), interferon (IFN)- (PBMCs) (34), and granulocyte-macrophage colony-stimulating factor (GM-CSF; PBMCs) (36). Proliferation-associated genes that are repressed by VDR ligands include epidermal growth factor receptor (EGF-R) (keratinocytes) (37), c-myc (keratinocytes) (37), and K16 (psoriatic plaques) (38). PTH (parathyroid cells) (39) and PTHrP (osteoblasts and keratinocytes) (40, 41) that are involved in mineral homeostasis are also down-regulated by VDR ligands. Recently, rel B (NF-B component) showed vitamin D-dependent down-regulation in dendritic cells (DCs) (14). Negative regulation of PTH (42), PTHrP (43), and rel B (14) gene expression appears to occur through a DNA motif, called negative VDRE. However, the mechanism of VDR-dependent inhibition of IL-2 and GM-CSF expression appears to be more complex than the involvement of positive or negative VDREs. In the case of these cytokines, VDR first competes with NF-AT1 for binding to the composite NF-AT1- activator protein 1 (AP1) enhancer motif, and then it interacts with c-Jun. This apparent co-occupancy of the composite site by VDR-c-Jun leads to inhibition of activated IL-2 and GM-CSF expression (16, 17, 44). Both VDR monomers and VDR-RXR heterodimers are involved in inhibition of IL-2 and GM-CSF promoters.

View larger version (32K): [in this window] [in a new window]   FIG. 2. Regulation of gene expression by VDR ligands. A schematic representation of VDR-mediated regulation of gene expression is presented. All the genes with known VDREs are shown as positively (+) regulated genes. Genes that are negatively (–) regulated by vitamin D because of the presence of negative VDRE in their promoter regions as well as genes that are negatively regulated by mechanisms involving transcription factor antagonism (anti-NF-AT and anti-NF-B activities) are also shown. References are given in parentheses.

A complex of approximately 20 proteins called DRIP (VDR-interacting proteins)/SMCC (SRB- and MED-containing cofactor complex)/TRAP (thyroid hormone receptor-associated protein)/ARC (activator-recruited cofactor) that interacts with VDR, other nuclear receptors, and transcription factors has been described (55). The DRIP complex was found to be sufficient for in vitro ligand-dependent transcription by the RXR-VDR heterodimer (55). The RXR-VDR recruits the complete DRIP complex by ligand-mediated interaction with DRIP 205, a component of the complex. The current working model for the vitamin D-mediated transcription is thought to require ligand-dependent targeted recruitment of VDR-DRIP and VDR-SRC complexes to the VDRE in a sequential manner (56). The first step involves the recruitment of VDR-SRC or a VDR-histone acetyltransferase activity complex to a responsive promoter to facilitate the destabilization of the nucleosomal core. The unwound DNA then becomes a target for the VDR-DRIP complex, which contains mediator factors that are required for interaction with basal transcription factors and polymerase II. This two-step model is currently in vogue but may get more complex as additional DRIPs are discovered.

Three corepressors (Table 2) have been found to interact with VDR, and these are NcoR-1, NcoR-2, and Hairless (45). These corepressors recruit histone deacetylase activities that deacetylate the lysine residues present in the N-terminally located histone tails, resulting in chromatin compaction and silencing of genes. Ligand binding induces a receptor conformation that creates a hydrophobic cleft, thus rendering the nuclear receptor receptive to interaction with coactivators through their NR boxes (LXXLL motifs). A related motif present in corepressors NcoR-1 and silencing mediator of retinoid and thyroid receptor, called CoRNR box (I/LXXI/VI), was shown to be essential for the interaction of corepressors with the unliganded receptor (57, 58). Coactivators and corepressors provide "yin" and "yang" that are required for tighter control of transcription by physiological stimuli. The number of VDR-interacting cofactors indicates that even combinatorial and parallel receptor-interacting complexes may exist. Understandably, with the discovery of a plethora of receptor-interacting proteins, the transcription picture has become more complicated. But this scenario also provides an opportunity of increased understanding of tissue- and gene-selective transcription by natural and synthetic ligands.

	III. VDR Crystal Structure

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
Although structures for many nuclear receptors with or without ligands were available, obtaining a good model for 1,25-(OH)₂D₃ complexed VDR was challenging. The VDR LBD has an insertion domain from amino acid residues 165–215 that is poorly conserved between different species, with no obvious biological significance (59). Rochel et al. (60) engineered a VDR LBD lacking this loop and characterized this mutant protein to prove that it was able to bind the natural ligand and was also capable of transactivation. This mutant VDR-LBD was crystallized to solve the structure of the ligand-bound receptor. The crystal structure of the LBD of mutant VDR (60) resembles that of other nuclear receptors displaying three layers of 13- helices and a short ß-sheet of three strands (Fig. 3, A and B). The nomenclature of helices is based on the description of the structure of RXR LBD (61). Members of the nuclear receptor superfamily exhibit not only the same modular domain structure but also a moderately conserved LBD (60). The canonical folds of LBDs of all nuclear receptors consist of 10–13 -helices, two to five ß-sheets arranged in antiparallel orientations, and connecting loops of varying sizes (62). The relative position of N-terminal helix H1 is conserved among all nuclear receptors, and it provides intramolecular contacts for the stabilization of the global structure of LBD. In VDR, helices H1 and H3 are connected by two short helices, H2 and H3n, where H3n is predicted by secondary structure predictions in the place of a loop structure that is present in other nuclear receptors (60). The crystal structure of VDR LBD closely resembles that of retinoic acid receptor (RAR)-. However, a very conspicuous difference between VDR and RAR is that the loop connecting H1 and H3 (shown in green for RAR in Fig. 4) wraps around the three-stranded short ß-sheet in RAR (63), whereas it just passes by one end of ß-sheet in VDR (shown in pink in Fig. 4), thus leaving the sheet exposed. Note that the ß-sheets of RAR are shown in blue, and those of VDR are shown in red (Fig. 4). Concomitantly, the loop connecting the ß-strand-2 (ß2) and ß-strand-3 (ß3) in VDR depicted in red is folded out of the 1,25-(OH)₂D₃-binding pocket, whereas the corresponding loop in RAR (depicted in blue) is pointing toward the pocket, thus making the pocket smaller in RAR.

View larger version (47K): [in this window] [in a new window]   FIG. 3. Topology of VDR LBD crystal structures (x-ray coordinates; Protein Data Bank ID code 1DB1). A, Crystal structure of 1,25-(OH)2D3 bound to LBD of VDR is shown. The helices (H1-H12) are represented as cylinders (red, except helix-12 which is pink), ß-sheets are represented as arrows in yellow, and the ligand is shown in white. Positioning of the ligand in the ligand-binding cavity is clearly shown by making helices H3 and H2 translucent. B, Ribbon diagram of the VDR LBD in stereo. The 1,25-(OH)2D3 is represented as ball and stick embedded in translucent surface in pink color. The helices are numbered as H1, H3, H5, H11, and H12. Few of the residues close to the ligand are displayed in stick model identified with one letter amino acid code and the residue number in blue. The helix-7 (H7) is directly behind the ligand in this orientation and thus is not visible. However, its location can be seen in panel A, which is represented as cylinders. These figures were produced using the published crystal structure coordinates (60 ).

View larger version (50K): [in this window] [in a new window]   FIG. 4. Ribbon diagram of VDR and RAR LBDs. RAR (dark shaded) and VDR (light shaded) are optimally superimposed to show the important structural difference between them. The ß-sheet region (ß1, ß2, and ß3) for RAR is shown in blue, and the corresponding region for VDR is shown in red. The loop connecting the helix-1 and helix-3 (which includes helix-2 in the case of VDR) is colored pink in VDR, and the corresponding region is colored green for RAR. A total of 452 backbone atoms were used for superposition, and the RMS deviation was 1.03 Å. This figure was produced using the coordinates from the published crystal structures of VDR (x-ray coordinates; Protein Data Bank ID code 1DB1) and RAR (x-ray coordinates; Protein Data Bank ID code 2LBD) (60 63 ).

View larger version (33K): [in this window] [in a new window]   FIG. 5. Details of molecular interactions between 1,25-(OH)2D3 and VDR ligand binding cavity. LIGPLOT (267 ) was used to detail the molecular interactions between 1,25-(OH)2D3 and VDR ligand binding cavity. The interactions between amino acid residues lining the VDR ligand binding pocket and 1,25-(OH)2D3 are schematically depicted herein. 1,25-(OH)2D3 is shown in ball-and-stick in the middle. Residues that make potential hydrogen bonds to 1,25-(OH)2D3 are shown in ball-and-stick. Ligand bonds are violet, and bonds in the protein are reddish-yellow. Oxygen atoms, except in water molecules, are filled in red, nitrogen atoms in blue, and carbons in black. The water molecules (HOH) are shown in cyan. Red spikes denote the hydrophobic interactions between the ligand atoms and the protein residues. Hydrophobic contacts are discerned when the spikes from a ligand atom radiate out toward the protein residue (which is represented as arcs), and the spikes from that protein residue radiate out toward the same ligand atom. In other words, the red spikes from a ligand atom and a protein residue facing each other are indicative of the hydrophobic interaction between the two. Hydrogen bonds and hydrophobic contacts were calculated with HBPLUS (268 ).

	IV. VDR Knockout Animals

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
Gene knockout studies in mice and VDR mutations in humans have provided considerable insights into the physiological functions of vitamin D. Four groups have created VDR knockout animals (67, 68, 69, 70, 71). VDR null mice were phenotypically normal at birth but developed hypocalcemia, hyperparathyroidism, rickets/osteomalacia, and alopecia after weaning. Female null mutant mice also displayed uterine hypoplasia and impaired folliculogenesis. These mice died between 4 and 6 months of age. Similar symptoms occur in kindreds with VDR mutations in vitamin D-dependent rickets type II. Interestingly, feeding animals with a diet rich in calcium, phosphate, and lactose normalized all the symptoms in null mice, except for hair abnormalities. These observations suggest that increased intestinal calcium absorption is critical for 1,25-(OH)₂D₃ action on bone and calcium homeostasis. The calcium homeostasis defect could be explained by the reduced expression of duodenal epithelial calcium channels CAT1 and CAT2 in VDR null mice (71).

Vitamin D-deficient animals have also been shown to develop hypocalcemia, rickets, and hyperparathyroidism, but unlike vitamin D-dependent rickets type II patients, never developed alopecia. Thus, intact VDR is required for maintaining bone mineral homeostasis after birth as well as for normal hair development and hair follicle homeostasis. Hair-reconstitution studies in nude mice using VDR–/– keratinocytes showed normal hair follicle morphogenesis but a defective response to anagen initiation phase of the hair cycle (72). These studies also indicated that the defect lies in keratinocytes. In accordance with this notion, targeted expression of human VDR transgene to keratinocytes of VDR null mice prevented alopecia (73). Therefore, hair loss in VDR–/– mice appears to be due to keratinocytes defective in epithelial-mesenchymal interactions that are required for normal hair cycling.

Similar hair cycle defect with total alopecia was also observed in inactivating mutation in hairless (hr) gene and in mice with temporally controlled RXR mutations in the epidermis. Hairless gene product (Hr) acts as a corepressor of VDR, thyroid hormone receptor, and RAR-related orphan receptor , and associates with VDR in vitro and in vivo (74). It is tempting to hypothesize that VDR-Hr interaction may modulate hair cycling by controlling the expression of an inhibitor of hair cycle. Epidermis-specific temporal disruption of RXR in mouse showed alopecia and skin abnormalities (75). Like VDR ablation, progressive alopecia in epidermal targeted RXR–/– animals was attributed to defects in hair cycle. Because RXR is the heterodimer partner of other nuclear receptors (RAR, peroxisome proliferator activated receptor, liver X receptor, and thyroid hormone receptor) resident in skin, these results suggest that only the RXR-VDR signaling pathway is involved in hair cycling. Furthermore, the absence of keratinocyte differentiation abnormalities in the VDR null animals (76), but the presence of these defects in the temporally controlled RXR mutation in mouse epidermis points toward the redundancy of RXR partner nuclear receptor functions in mouse skin. VDR knockout mice were also found to have impaired insulin secretory capacity (77).

VDR/RXR compound null mutant mice have also been prepared and showed growth retardation, impaired bone formation, hypocalcemia, and alopecia, features typical of VDR null animals (78). The growth plate development in compound knockout animals was more severely impaired in comparison to VDR null animals. These findings indicate that both vitamin A and vitamin D signaling pathways are required for the normal development of growth plate chondrocytes.

	V. Vitamin D Action in Autoimmune Disease Models

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
In addition to its central role in calcium and bone metabolism, 1,25-(OH)₂D₃ has potent immunomodulatory effects on many immune cell types, including both innate and adaptive immune cells (79, 80, 81). Consistent with these effects, the VDR is widely expressed in most immune cell types such as antigen-presenting cells (APCs; monocyte/macrophage, DCs), natural killer cells (82, 83), T cells (84), and B cells (85). Furthermore, the intriguing effects of 1,25-(OH)₂D₃ have been demonstrated in several autoimmune disease models, namely, systemic lupus erythematosus (SLE) in lpr/lpr mice (86, 87), type I diabetes in nonobese diabetic (NOD) mice (88, 89, 90, 91, 92), collagen-induced arthritis (93, 94, 95), experimental allergic encephalomyelitis (EAE) (96, 97, 98, 99, 100, 101), and inflammatory bowel disease (IBD) (102).

A. Multiple sclerosis (MS)
MS is a chronic inflammatory autoimmune disease of the central nervous system (CNS) in which self-epitopes on myelinated nerve fibers are inappropriately recognized by adaptive immune cells of the host. The ensuing immune response recruits T cells and macrophages into the CNS, resulting in localized areas of inflammation and demyelination known as MS lesions. EAE is widely used as an animal model for human MS disease. Self-antigen reactive T helper I (Th1) cells have been demonstrated to play an essential role in both induction and effective phase of disease. Th1 cell differentiation is controlled by both antigen stimulation and cytokines, particularly IL-12 and IL-23, which subsequently induce synthesis of Th1-specific transcription factor, T-bet, and drives Th0 cells toward Th1 differentiation (103). Therefore, controlling Th1 development by inhibiting IL-12 production will benefit the Th1-mediated disease, such as MS. VDR ligands have been shown to inhibit IL-12 p70 production in freshly isolated human monocyte or PBMCs that are primed with IFN- and stimulated with lipopolysaccharide in a dose-dependent manner (104). Furthermore, mitogen-induced differentiation of neonatal CD4⁺ T cells into Th1 cells (IFN- secreting T cells) in vitro is dramatically inhibited by 1,25-(OH)₂D₃ and its analogs. Interestingly, the inhibition of Th1 development seems to be specific, because Th2 (IL-4 secreting T cells) cell differentiation is largely unaffected by this treatment (104).

When mice were immunized with self-antigen peptide, myelin oligodendrocyte glycoprotein (MOG35–55), and treated with 1,25-(OH)₂D₃, disease induction was inhibited as evidenced by the reduction of inflammatory infiltration and reduced demyelination of brain and spinal cord, along with decreased antigen-induced T cell proliferation during an in vitro T cell recall response. More importantly, it also inhibited Th1 development (104), suggesting that inhibiting CD4 Th1 effector function is one of the mechanisms by which 1,25-(OH)₂D₃ inhibits EAE. In contrast to CD4⁺ T cells, the protection of EAE from 1,25-(OH)₂D₃ treatment does not involve CD8⁺ T cells (97), although CD8⁺ T cells have the highest levels of VDR and have been implicated as both suppressors and effectors of the inflammation associated with EAE. In addition to effects on T cells, 1,25-(OH)₂D₃ has also been shown to decrease macrophage accumulation in the CNS, which can contribute to its protective effect during EAE (99). When EAE was induced in B10.PL mice after immunization with myelin basic protein, administration of 1,25-(OH)₂D₃ not only prevented the induction of the disease, but also ameliorated the disease when the treatment was administered at the appearance of the first disability symptoms. Interestingly, withdrawal of 1,25-(OH)₂D₃ resulted in the resumption of EAE. Deficiency of vitamin D in the food led to the increased EAE susceptibility in mice, suggesting that daily vitamin D in vivo prevents the occurrence of the autoimmune disease (105). Furthermore, immunization of VDR-deficient mice led to EAE, which was not suppressed by administration of 1,25-(OH)₂D₃, whereas EAE in wild-type animals was completely blocked by this treatment, suggesting that the VDR is necessary and directly involved in 1,25-(OH)₂D₃-mediated suppression of EAE.

Although the mechanism of EAE inhibition by 1,25-(OH)₂D₃ is not completely understood, many studies have clearly implicated that a major effect of VDR ligands is on T cell functions, mainly through inhibiting Th1 cell development and function while enhancing Th2 cell development (106). Administration of 1,25-(OH)₂D₃ in mice increases expression of IL-4 and TGF-ß1, which presumably play an important role in regulating T cell responses (100). Accordingly, VDR ligands were found to be less effective in reducing the progression of EAE in IL-4 null mice (102).

B. Rheumatoid arthritis (RA)
RA, one of the most common chronic inflammatory diseases, affects about 1% of the population and is characterized by articular infiltration of neutrophils, macrophages, T and B cells, and DCs, resulting in subsequent tissue damage (107). The collagen-induced arthritis model is the most commonly used arthritis model for human RA. Immunization of mice with type II collagen induces arthritis, which could be prevented by dietary supplementation or oral administration of 1,25-(OH)₂D₃ in both mouse and rat (93, 94, 95). More interestingly, administration of 1,25-(OH)₂D₃ or its analogs at the time of disease with early symptoms could prevent the progression to severe arthritis (92, 94). In addition, treated animals showed diminished serum levels of antibodies to rat collagen type II and reduced mitogen-induced proliferation of lymph node cells (94), suggesting that 1,25-(OH)₂D₃ suppresses RA by altering adaptive immunity.

Epidemiological studies have reported low serum levels of vitamin D and its metabolites in RA patients (108). In addition, an open-label clinical trial involving patients (Table 1) with psoriatic arthritis showed an improvement in disease symptoms after the oral administration of 1,25-(OH)₂D₃ (109). More significantly, in a 3-month open-label clinical trial in 19 RA patients being treated with standard disease-modifying antirheumatic drugs therapy for acute RA (Table 1), high-dose oral alfacalcidol (1 hydroxyvitamin D₃; Fig. 1) therapy showed a positive effect on disease activity in 89% of the patients, and only 11% of patients showed no improvement (110). This result provides strong evidence that VDR ligand can be used clinically for the treatment of RA.

1,25-(OH)₂D₃ has been detected in synovial fluid of arthritic joints, and the expression of VDR has also been reported in rheumatoid synovial tissue and at the site of cartilage erosion. Matrix metalloproteinases (MMPs) play an important role in the chondrolytic process of rheumatoid lesion. Animal studies have shown that the production of some MMPs may be up-regulated in rat chondrocytes by administration of 1,25-(OH)₂D₃ (111). Interestingly, treatment of rheumatoid synovial fibroblasts by 1,25-(OH)₂D₃ did not affect the expression of MMPs. However, when simultaneously stimulated with IL-1ß, the MMP production was reduced 50% compared with IL-1ß alone. Prostaglandins have a role in the immune system and inflammatory processes associated with RA. Interestingly, IL-1ß-stimulated prostaglandin E2 (PGE2) synthesis is also completely inhibited by 1,25-(OH)₂D₃ treatment (112). Therefore, VDR ligand can suppress both the IL-1ß-stimulated production of MMP and PGE2 in rheumatoid synovial fibroblasts, suggesting that VDR-mediated biological processes are important in controlling RA. However, the molecular mechanism by which VDR ligands inhibit IL-1ß-stimulated MMP and PGE2 production is not known.

C. Inflammatory bowel diseases (IBDs)
IBDs (ulcerative colitis and Crohn’s disease) are immune-mediated pathologies with unknown etiology that target the gastrointestinal tract. T cells, particularly CD4⁺ Th1 cells, which preferentially produce inflammatory cytokines (IL-2, TNF-, IFN-), have been shown to transfer Crohn’s-like symptoms to naive mice, and the production of Th1 cytokines is associated with IBD in humans (113, 114). Interestingly, IL-10-deficient mice develop colitis within 5–8 wk of life, and one third of these mice die after the development of severe anemia and weight loss (115). In this model, lack of vitamin D in the diet led IL-10 knockout mice to rapidly develop diarrhea, wasting disease, and mortality. In contrast, supplementation with 1,25-(OH)₂D₃ in diet significantly ameliorated the symptoms of IBD in IL-10 null mice. Similarly, treatment of animals with 1,25-(OH)₂D₃ blocked the progression and decreased the established IBD symptoms (115). 1,25-(OH)₂D₃ also inhibited the proliferation of rectal epithelial cells (116) and T cells (117) in active ulcerative colitis patients, suggesting the potential of VDR ligands in the treatment of IBDs.

D. Type I diabetes
Type I diabetes is an autoimmune disease characterized by the immune-mediated destruction of insulin-producing ß-cells of islets of Langerhans in the pancreas. Several cellular effector mechanisms leading to ß-cell destruction have been identified, including CD4⁺ and CD8⁺ T cells and macrophages (118). The NOD mouse, which spontaneously develops type I diabetes, is the most widely used animal model for type I diabetes (119). Based on the effect of VDR ligands on T cell and macrophage functions described earlier, it is reasonable to predict that type I diabetes would be modulated by VDR ligand. Epidemiological studies have shown an increase in the disease incidence when vitamin D deficiency was present in the first month of life in children. Moreover, in a recent study in NOD mice, vitamin D deficiency accelerated the onset of type I diabetes (120).

In fact, when 1,25-(OH)₂D₃ was administrated at 3 wk of age before the onset of insulitis, it effectively prevented the progression of diabetes in NOD mice. However, treatment was ineffective if administrated at 8 wk of age when insulitis was well established (121). In addition to the natural ligand, a vitamin D analog, KH 1060 (Fig. 1), has also been shown to prevent the onset of type I diabetes in NOD mice (91). Vitamin D analogs were also effective in the treatment of ongoing type I diabetes in the adult NOD mice by effectively blocking the disease course (92), raising the possibility of treatment of type I diabetes with VDR ligands before ß-cells are completely destroyed. Treatment of mice with a synthetic 1,25-(OH)₂D₃ analog also inhibited IL-12 production and blocked the infiltration of Th1 cells into the pancreas. It should be noted that IL-12 is a direct target of 1,25-(OH)₂D₃ (35). More interestingly, CD4⁺CD25⁺ Treg cells were also increased by this treatment in pancreatic lymph node (92). In fact, adaptive transfer of pancreatic lymph node-derived Treg cells into NOD mice completely prevented spontaneous diabetes. Interestingly, Treg cells accumulated preferentially in the pancreatic lymph nodes and islets, but not other lymph nodes or the spleen (92). In addition, 1,25-(OH)₂D₃ could induce an autoantigen-specific Th1 to Th2 immune shift in NOD mice immunized with glutamic acid decarboxylase 65 (89), suggesting that VDR ligands might work on multiple levels to modulate the immune system and prevent autoimmune diseases.

E. Systemic lupus erythematosus (SLE)
SLE is a systemic autoimmune disease. Patients with SLE produce autoantibodies to many tissue antigens including DNA, histones, red blood cells, platelets, and leukocytes, and as a result these individuals present with varying symptoms. The therapeutic potential of VDR ligands for the treatment of SLE was explored using an analog of 1,25-(OH)₂D₃, 22-oxa-1, 25-dihydroxyvitamin D₃ (Fig. 1). Symptoms of SLE were alleviated in MRL/lpr mice after treatment with 22-oxa-1, 25-dihydroxyvitamin D₃. In addition, at therapeutically effective doses of 22-oxa-1, 25-dihydroxyvitamin D₃, hypercalcemia was not observed in the autoimmune animals (122).

F. Transplant rejection
Organ and tissue transplantation are increasingly being performed worldwide, and these procedures many times lead to long-term survival. The commonly used strategy for preventing transplantation rejection involves the use of immunosuppressive agents such as corticosteroids and cyclosporin. 1,25-(OH)₂D₃ exerts a variety of immunomodulatory effects such as inhibition of T lymphocyte proliferation (123, 124, 125), down-regulation of cytokines IL-2 and IFN- (124), and DC maturation and survival (126). Based on these observations, 1,25-(OH)₂D₃ and its analogs have been tested either as single agents or in combination with other immunosuppressive agents such as cyclosporin in many experimental models, including heart (127, 128), liver (129), pancreatic islets (130, 131), and skin (132, 133). Interestingly, 1,25-(OH)₂D₃ treatment was shown to prolong allograft survival. For example, transplantation of rat islets under the kidney capsule in spontaneously diabetic NOD mice resulted in early graft failure in four of 10 recipients (131). Single treatment of NOD mice with KH1060 (Fig. 1), a vitamin D₃ analog, or cyclosporin did not result in statistically significant suppression of early graft failure. However, combination of both resulted in 100% early graft success (131). Similar results were also reported in chronic allograft rejection, where adventitial inflammation and internal hyperplasia in rat aortic allograft were inhibited by treatment with the VDR ligand (134). In addition to its direct immunosuppressive function in transplantation, 1,25-(OH)₂D₃ and its analogs also demonstrated important impact on the prevention of bone loss and improvement on bone quality after organ transplantation (135, 136, 137). The mechanism of vitamin D₃ in preventing transplant osteoporosis is probably due to its effect on hyperparathyroidism caused by the usage of corticosteroids or cyclosporin. Moreover, the immunomodulatory properties of 1,25-(OH)₂D₃ may enable the reduction of cyclosporin and/or corticosteroid dosage, which could potentially contribute to an improvement in posttransplantation-related bone loss.

In addition to its primary physiological role in regulating calcium homeostasis, vitamin D₃ has been demonstrated to have pleiotropic actions in the immune system. 1,25-(OH)₂D₃ was found to inhibit antigen-induced T cell proliferation (138) and cytokine production (139). APCs, in particular DCs, are also the primary targets for the immunosuppressive activity of 1,25-(OH)₂D₃. They play a central role in regulating immune response to self and foreign antigens. During the normal immune response, T cell response and specificity are conferred through the clonal restricted T cell receptor, which recognizes major histocompatibility complex (MHC) class I and class II molecule complexed with peptides. However, the potency of DC-mediated T cell activation also depends on their maturation status. Immature DCs express low levels of MHC class II costimulatory specific maturation markers, such as CD83, whereas mature DCs express high levels of these molecules in response to appropriate proinflammatory stimuli (140, 141). Immature DCs not only provide weak signals for activating T cells but also induce anergic and regulatory T cells (142). Mature DCs have considerably less antigen uptake and demonstrate increased antigen presentation and induction of costimulatory ligands on the surface. These cells also show enhanced production and secretion of immunomodulatory cytokines such as IL-12 (143). The maturation of DCs is regulated by many factors and cell-cell interactions. Among the factors that induce DC maturation are the components of pathological pattern recognition, which are recognized by cell surface Toll-like receptors and many cytokines including TNF-. T cells can induce DC maturation through their intimate interactions by activating TNF receptor family members expressed on DCs such as CD40. However, it is unknown whether the preservation of DCs in an immature state results from the absence of maturational stimuli or is also actively maintained in vivo.

VDR is widely expressed in most cell types in the immune system such as APCs, monocyte/macrophage, natural killer cells, and DCs (82, 83). 1,25-(OH)₂D₃ could affect DC function through differentiation, cytokine production, activation, maturation, as well as survival (126, 144). The most direct evidence of 1,25-(OH)₂D₃ effect on DCs comes from VDR knockout mouse studies, where DCs from VDR-deficient mice showed a significantly higher level of maturation markers such as class II MHC, CD40, CD80, and CD86 on cell surface (144). Under differentiation conditions in the presence of GM-CSF and IL-4 from purified human CD14⁺ monocytes, 1,25-(OH)₂D₃ completely inhibited DC differentiation with low-level expression of IL-12 and maturation markers (144, 145). Consequently, 1,25-(OH)₂D₃-treated DCs lead to impaired alloreactive T cell activation in vitro and in vivo (144, 146). Furthermore, 1,25-(OH)₂D₃ also affects maturing DCs, leading to decreased IL-12 secretion and enhanced IL-10 production, which subsequently affects T cell differentiation (126). Recent evidence also pointed out that 1,25-(OH)₂D₃ not only exhibits immunosuppressive effects on T cells and APCs, but also leads to the induction of tolerogenic DCs, which subsequently enhance CD4⁺CD25⁺ regulatory T cells and protect the allograft rejection (147). In conclusion, all the preclinical studies suggest that 1,25-(OH)₂D₃ and its analogs can be potentially used for the prevention of transplant rejection.

In summary, VDR ligands have pleiotropic effects on the immune system, including both innate and adaptive immunity. In particular, their effect on DCs is thought to induce tolerogenic DCs resulting in T cell anergy. Their effect on T cells can lead to the formation of regulatory T cells as well as accelerate Th2 cell development. All of this evidence suggests that VDR ligands have great potential for the development of therapies for multiple human autoimmune diseases. The challenge for the future is to develop safer oral VDR ligands without the hypercalcemic side effect.

	VI. Vitamin D Action on Keratinocytes and Psoriasis

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
Psoriasis, a recurrent inflammatory skin disorder, affects approximately 2% of the population. In addition, 5–10% of the patients develop psoriatic arthritis, with inflammation and swelling in the hands, feet, and large joints. Psoriasis is characterized by keratinocyte hyperproliferation, abnormal keratinocyte differentiation, and immune-cell infiltration into the epidermis and dermis. The most common form of psoriasis is plaque psoriasis or psoriasis vulgaris. At the molecular level, psoriasis lesions show a prominent loss of loricrin and filaggrin in the suprabasal layers of the epidermis and abnormal overexpression of other differentiation markers such as involucrin, transglutaminase I (TGase I), psoriasin, migration inhibitory factor related protein-8, and skin-derived antileukoproteinase. The expression of normal suprabasal keratins K1 and K10 is inhibited and replaced by the expression of the hyperproliferative keratins K6 and K16. Psoriatic epidermis also demonstrates an increased expression of IL-8 receptor, IL-6, IL-8, EGF-R, TGF, and amphiregulin. Lesions also show infiltration of IL-2, IFN-, and TNF--secreting CD8+ lymphocytes into epidermis and CD4+ lymphocytes into dermis, and these cytokines are thought to alter keratinocyte proliferation and differentiation.

The observations that keratinocytes and T cells express VDR and that 1,25-(OH)₂D₃ is a potent stimulator of keratinocyte differentiation provided a reasonable basis for the clinical use of VDR ligands for the treatment of psoriasis (82, 148). The first clinical evidence to support the use of vitamin D analogs was obtained fortuitously when a patient treated orally with 1-hydroxyvitamin D₃ (Fig. 1) for osteoporosis showed remarkable remission of psoriatic lesions (149). Subsequently, promising clinical results were obtained in studies using oral 1-hydroxyvitamin D₃, oral and topical 1,25-(OH)₂D₃ (calcitriol), and topical 1,24-(OH)₂D₃ (tacalcitol; Fig. 1). In these clinical trials, approximately 70–80% of the patients showed marked improvement, and complete clearance of the target lesions was observed in 20–25% of patients (150). A topical preparation of calcitriol/ (Silkis, 3 µg/g ointment) is being developed by Galderma Laboratories (Sophia-Antipolis, France) for the treatment of psoriasis. It has shown safety and efficacy at a calcitriol concentration of 3 µg/g ointment but resulted in increased risk of hypercalciuria at 15 µg/g concentration (151, 152). Therefore, it appears to show a therapeutic window of 5 between efficacy and side effect. Medicinal chemists have tried to develop 1,25-(OH)₂D₃ analogs with decreased hypercalcemia liability mostly by minor modifications of the secosteroidal backbone of vitamin D₃. Calcipotriol (calcipotriene or Dovonex, Leo Laboratories, Denmark; Fig. 1), a synthetic 1,25-(OH)₂D₃ analog used topically for the treatment of psoriasis, was chemically engineered to be metabolized quickly in systemic circulation, and as a result it is 100–200 times less calcemic than 1,25-(OH)₂D₃ (4). A comparison of hypercalcemic properties of calcitriol and calcipotriol is given in Table 3. A number of studies have confirmed the clinical efficacy of calcipotriol, and significant improvement has been observed in approximately 70% of the patients after 6–8 wk of topical therapy with twice daily application of the drug (4). The most common side effect of calcipotriol was cutaneous irritant reaction in approximately 20% of the patients (153). In comparative clinical trials, the efficacy of topical calcitriol was generally similar, and that of topical calcipotriol was slightly better than potent topical steroids (154, 155). Steroids result in early onset of action in psoriasis but cannot be used for longer periods, because of their skin thinning (telangiectasia) side effect, which results because of the alteration in skin collagen metabolism [decrease in type I and type III collagen, TIMP (tissue inhibitor of metalloproteinase) 1 and TIMP 2 message levels] by this class of drugs (156). Therefore, a combination of topical calcitriol or calcipotriol with potent topical steroids has been tried in patients. These clinical studies indicated more rapid onset of action, increased efficacy, and better tolerability of the combination regimens than the individual treatments (157, 158). As a result, a new formulation containing calcipotriol and betamethasone dipropionate (Daivobet, Leo Pharma AS, Ballerup, Denmark) is under consideration by drug regulatory authorities for the treatment of mild-to-moderate plaque-type psoriasis (159).

View larger version (20K): [in this window] [in a new window]   FIG. 6. Vitamin D-regulated genes in epithelial cancer cells. Genes that are positively or negatively regulated by 1,25-(OH)2D3 in prostate, colon, or breast cancer cells and keratinocytes/SCC are shown. The regulation of expression of these genes was confirmed by immunohistochemistry, Northern or Western blotting techniques.

APCs or DCs also play an important role in psoriasis and autoimmune diseases because they are involved in autoantigen presentation. It appears that APCs are one of the major targets of 1,25-(OH)₂D₃-mediated immunosuppressive action and VDR ligands prevent the differentiation, maturation, activation, and survival of DCs, leading to T cell hyporesponsiveness (126). 1,25-(OH)₂D₃ also increased the expression of IL-10 and decreased the expression of IL-12, two major cytokines that are involved in Th1-Th2 balance (147). It is believed that the development of more efficacious topical and oral VDR ligands, with improved side effect profiles, will further expand the treatment options for patients with psoriasis.

	VII. Vitamin D Action on Prostate Cancer Cells

Top Abstract I. Introduction II. VDR and the... III. VDR Crystal Structure IV. VDR Knockout Animals V. Vitamin D Action... VI. Vitamin D Action... VII. Vitamin D Action... VIII. Vitamin D Action... IX. Vitamin D Action... X. Vitamin D Action... XI. Vitamin D Action... XII. Vitamin D Action... XIII. Vitamin D Action... XIV. Vitamin D Action... XV. Conclusions References

 
Prostate cancer is the second leading malignancy after skin cancer, and it is also the second leading cause of cancer deaths among men in the United States (165). The discovery that the VDR is expressed in normal prostate, benign prostate hyperplasia (BPH), malignant prostate, and prostate cancer cell lines led to the recognition that BPH and prostate cancer could be potential targets for VDR ligands (166, 167, 168). Epidemiological studies have indicated an inverse relationship between mortality rates due to prostate cancer and UV light exposure. UV light is required for the synthesis of vitamin D in skin (169). In fact, one of the major risk factors for developing prostate cancer was low serum level of 25-hydroxyvitamin D (170). Several in vitro studies have demonstrated that 1,25-(OH)₂D₃ and its synthetic analogs inhibited the proliferation of prostate cancer cell lines (171) and primary epithelial cells from normal prostate, BPH, and prostate cancer (172, 173). VDR ligands also inhibited tumor cell