One area of cancer chemoprevention that has been intensively studied in recent years is biologic modifiers of cancer cells that are designed to retard proliferation ( 1 – 3 ), to induce differentiation of these cells to a quiescent, nondividing stage ( 4 ), and/or to promote cell death ( 5 – 8 ). In this issue of the Journal, Mehta et al. ( 9 ) report the effect of a novel vitamin D compound in a murine mammary gland chemoprevention model. The secosteroid hormone known as 1α,25-dihydroxyvitamin D 3 [1α,25(OH) 2 D 3 ] has been described as a key regulator of serum calcium. In the last two decades, however, it has also been found to have diverse biologic effects in normal and malignant tissues. These responses include the in vitro inhibition of proliferation and induction of differentiation of various cancer cells, such as those from the human hematopoietic system, breast, ovaries, colon, brain, and prostate ( 10 – 17 ). Initiation of these genomic responses is through a specific steroid hormone nuclear vitamin D 3 receptor (VDR) acting as a ligand-inducible transcription factor that binds the vitamin D 3 response element contained within the promoter/enhancer region of target genes ( 18 ).
Despite the intense research that has focused on 1α,25(OH) 2 D 3 since it was first characterized in 1971 ( 19 ), the exact mode of action by which it inhibits cancer cells remains largely unknown. In normal tissues not directly involved in calcium regulation, for example, the well-studied system of keratinocytes, exposure of these cells to 1α,25(OH) 2D3 increases the synthesis of transforming growth factor (TGF)-β1 and TGF-β2 ( 20 ), decreases expression of epidermal growth factor receptors ( 21 ), and leads to dephosphorylation of the retinoblastoma protein ( 22 ). In normal prostate cells, it exerts a differentiating effect in combination with testosterone ( 23 ). At pharmacologically active doses, 1α,25(OH) 2D3 can suppress the immune system ( 24 – 27 ) and can enhance monocyte-macrophage differentiation ( 28 ). Other specific, genomic effects observed in cancer cells exposed to 1α,25(OH) 2D3 include cell cycle arrest in G 1 . Many factors can lead to cell cycle arrest, but the cyclindependent kinase inhibitors known as p21 (waf 1) and p27kip 1 are pivotal to this process; the p21 (waf 1) gene contains a vitamin D 3 response element within its promotor region ( 29 ) and expression of the gene increased in response to 1α,25(OH) 2D3 . Also, expression of p27(kip 1) is markedly induced in certain cancer cell types (e.g., myeloid leukemia and prostate cancer) after their exposure to 1α,25(OH) 2D3 ( 17 , 30 – 32 ).
A major focus of chemoprevention research in the field of vitamin D and cancer has been to synthesize analogues of 1α,25(OH) 2D3 that have prominent antiproliferative effects against cancer cells without resulting in hypercalcemia when they are administered in vivo at pharmacologically active doses. This research has resulted in several analogues that have dramatic antiproliferative behavior, most noticeably analogues with double and triple bonds in the C/D ring and side chain ( 33 , 34 ), addition of three to six hexafluoride groups to the end of the side chain ( 17 , 28 , 35 ), or placement of the side chain in the 20-epi configuration ( 36 – 38 ). Initial clinical trials are under way; for instance, an ongoing phase I study in the U.K. is examining the effects of these analogues on breast cancer. Studies in vitro have shown that vitamin D3 analogues can inhibit the clonal proliferation of breast cancer cells at the 10-11-10-9 M range, with an associated increase in expression of bax and concurrent decrease in bcl-2 expression ( 37 ). Furthermore, potent hexafluoride analogues can reduce the breast cancer incidence and burden in N-nitroso-N-methylurea-treated rats ( 39 ). One area in which the therapeutic potential of vitamin D3 has been realized is in the treatment of psoriasis, where the topical application of potent analogues, including calcipotriene (Dovonex), controls the disease and does not significantly interfere with serum calcium levels ( 40 ).
The reason for the increased antiproliferative potency of the analogues is becoming clearer, as illustrated in Fig. 1 . Vitamin D 3 analogues usually bind less well to the D-binding protein in the blood and are, therefore, more readily available to enter the cells ( 41 ). Analogues may also extend the half-life of the VDR ( 42 ), or they may induce novel VDR conformations ( 43 ), which may either allow more efficient interactions with vitamin D 3 response elements and/or expand the array of vitamin D 3 response elements that can be activated. In addition, metabolic products of analogues may result in potent intermediates in vivo. For example, compared with the parental analogues, the 24-oxo metabolites have the same in vitro anticancer activities but have fewer effects on calcium levels in sera ( 44 , 45 ).
In vivo, VDR forms heterodimers with the retinoid X receptor ( 18 ), and the combination of 9-cis-retinoic acid and 1α,25(OH) 2D3 can synergistically increase expression of a reporter gene construct containing a vitamin D 3 response element within its promoter ( 46 ). Cooperation between these two receptor pathways has been the basis for combination therapy; we have previously demonstrated ( 47 , 48 ) synergistic inhibition of proliferation of human myeloid leukemia cells and MCF-7 breast cancer cells by a potent vitamin D 3 analogue in combination with 9-cis-retinoic acid. Thus, combinations of retinoids and vitamin D ligands may be an attractive prospect for control of deregulated cell growth. Also, various steroid hormone receptor enhancer proteins have been identified ( 49 , 50 ); by recruiting them in vivo, we may be able to accentuate further the positive therapeutic genomic effects of 1α,25(OH) 2D3 .
The study by Mehta et al. reported in this issue of the Journal presents an entirely novel class of vitamin D compounds (vitamin D 5 ). Utilizing a mammary gland lesion model to assess chemoprevention, the authors demonstrated preventive effects in vitro but no significant effect on serum calcium levels in vivo. Thus, the therapeutic index (ratio of its antiproliferative to its calcemic effects) for this compound is sufficiently high to warrant further investigations using other cancer cell types and model systems. The study by Mehta et al. and ongoing fundamental research into the effects of vitamin D compounds on cancer cells are elucidating the molecular effects and highlighting the therapeutic potential of these highly interesting compounds. Several analogues have already been identified that have significant inhibitory effects but that do not induce hypercalcemia; Mehta et al. add another compound to this list. Many of these compounds are potential candidates for clinical investigations.