Selective analogs of 1α,25-dihydroxyvitamin D3 for the study of specific functions of Vitamin D

New analogs of 1α,25-dihydroxyvitamin D₃ synthesized in our research group that show selective activity in vivo are presented along with supporting biological results. Compounds that act preferentially on intestine are 2-(3′-propylidene-19-nor-(20S or 20R))-1α,25-dihydroxyvitamin D₃ and 2-methylene-19-21-dinor-1α,25-dihydroxyvitamin D₃. Compounds that act anabolically to induce bone formation are 2-methylene-19-nor-(20S)-1α,25-dihydroxyvitamin D₃ (2MD), the 2α-methyl derivative, the 26,27-dimethyl derivative, and the 26-dimethylene derivative. Compounds that act preferentially on parathyroid glands are 2-methylene-19-nor-1α-hydroxy-homopregnacalciferol, the 20S-bishomo derivative and the 2-methylene-19,26,27-trinor-1α,25-dihydroxyvitamin D₃. These latter compounds do not elevate serum calcium until doses of the order of >300 μg/kg body weight are used, while parathyroid hormone levels are suppressed at much lower doses. Some of these novel analogs may ultimately be useful as new and safer therapeutic agents. Regardless of their clinical utility, they represent valuable research tools that can be used to study the specific functions of the Vitamin D hormone in vivo.     

Therapeutic potential of the 2-alkyl and 2-alkylidene-19-nor-(20S)-modified analogs of 1alpha,25-dihydroxyvitamin D3

Five analogs of 19-nor-1alpha,25-dihydroxyvitamin D(3) are described that show highly selective and potent activities. The 2-methylene-19-nor-(20S)-1alpha25-dihydroxyvitamin D(3) (2MD) and its 2alpha-methyl sister are selectively active on the osteoblast. 2MD is bone anabolic and causes bone formation in vivo and in vitro and is being developed as a therapy for bone loss diseases such as osteoporosis. 2-Methylene-19-nor (20S)-bishomo-1alpha-hydroxypregnacalciferol (2BMP) has no activity on calcium in vivo while totally suppressing circulating parathyroid hormone. Its homologs, i.e. 2-methylene-19-nor-1alpha-hydroxy-homopregnacalciferol (2MP) and 2-methylene-19-nor-1alpha-hydroxypregnacalciferol (2MPC) act similarly but are either less selective (2MP) or not as potent (2MPC). These abbreviated side chain analogs will be developed for diseases where a rise in serum calcium is not desired, as for example, cancer, renal osteodystrophy, psoriasis and autoimmune diseases.     

Differential effects of Vitamin D analogs on bone formation and resorption

Deficiency in Vitamin D and its metabolites leads to a failure in bone formation primarily caused by dysfunctional mineralization, suggesting that Vitamin D analogs might stimulate osteoblastic bone formation and mineralization. In this study, we compare the effect of selected Vitamin D analogs and active metabolite, 1alpha,25-dihydroxyvitamin D(3), 19-nor-1alpha, 25-dihydroxyvitamin D(2), and 1alpha-hydroxyvitamin D(2) or 1alpha,25-dihydroxyvitamin D(2) on bone formation and resorption. In a mouse calvariae bone primary organ culture system, all Vitamin D analogs and metabolite tested-stimulated collagen synthesis in a dose-dependent manner and 19-nor-1alpha, 25-dihydroxyvitamin D(2) was the most efficacious among three. 19-nor-1alpha, 25-dihydroxyvitamin D(2) and 1alpha,25-dihydroxyvitamin D(2) showed similar potencies and 1alpha,25-dihydroxyvitamin D(3) was less potent than others. Osteocalcin was also up-regulated in a dose-dependent manner, suggesting that the three Vitamin D analogs have the equal potencies on bone formation. 25-Hydroxyvitamin D-24-hydroxylase expression was induced in a dose-dependent manner and 19-nor-1alpha, 25-dihydroxyvitamin D(2) was less potent than other two compounds. In a mouse calvariae organ culture, all induced a net calcium release from calvariae in a dose-dependent manner, but the potency is in the order of 1alpha,25-dihydroxyvitamin D(2) congruent with1alpha,25-dihydroxyvitamin D(3)>19-nor-1alpha, 25-dihydroxyvitamin D(2). In a Vitamin D/calcium-restricted rat model, all caused an elevation in serum calcium in a dose-dependent manner. There is no significant difference between 1alpha,25-dihydroxyvitamin D(3) and 1alpha-hydroxyvitamin D(2) in potencies, but 19-nor-1alpha, 25-dihydroxyvitamin D(2) is at least 10-fold less potent than the other two compounds. Our results suggest that Vitamin D analogs have direct effects on bone resorption and formation, and 19-nor-1alpha, 25-dihydroxyvitamin D(2) may be more effective than 1alpha,25-dihydroxyvitamin D(3) and 1alpha-hydroxyvitamin D(2) on stimulating anabolic bone formation.     

Molecular structure of the rat vitamin D receptor ligand binding domain complexed with 2-carbon-substituted vitamin D3 hormone analogues and a LXXLL-containing coactivator peptide

We have determined the crystal structures of the ligand binding domain (LBD) of the rat vitamin D receptor in ternary complexes with a synthetic LXXLL-containing peptide and the following four ligands: 1alpha,25-dihydroxyvitamin D(3); 2-methylene-19-nor-(20S)-1alpha,25-dihydroxyvitamin D(3) (2MD); 1alpha-hydroxy-2-methylene-19-nor-(20S)-bishomopregnacalciferol (2MbisP), and 2alpha-methyl-19-nor-1alpha,25-dihydroxyvitamin D(3) (2AM20R). The conformation of the LBD is identical in each complex. Binding of the 2-carbon-modified analogues does not change the positions of the amino acids in the ligand binding site and has no effect on the interactions in the coactivator binding pocket. The CD ring of the superpotent analogue, 2MD, is tilted within the binding site relative to the other ligands in this study and to (20S)-1alpha,25-dihydroxyvitamin D(3) [Tocchini-Valentini et al. (2001) Proc. Natl. Acad. Sci. U.S.A. 98, 5491-5496]. The aliphatic side chain of 2MD follows a different path within the binding site; nevertheless, the 25-hydroxyl group at the end of the chain occupies the same position as that of the natural ligand, and the hydrogen bonds with histidines 301 and 393 are maintained. 2MbisP binds to the receptor despite the absence of the 25-hydroxyl group. A water molecule is observed between His 301 and His 393 in this structure, and it preserves the orientation of the histidines in the binding site. Although the alpha-chair conformer is highly favored in solution for the A ring of 2AM20R, the crystal structures demonstrate that this ring assumes the beta-chair conformation in all cases, and the 1alpha-hydroxyl group is equatorial. The peptide folds as a helix and is anchored through hydrogen bonds to a surface groove formed by helices 3, 4, and 12. Electrostatic and hydrophobic interactions between the peptide and the LBD stabilize the active receptor conformation. This stablization appears necessary for crystal growth.     

New analogs of 2-methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 with conformationally restricted side chains: evaluation of biological activity and structural determination of VDR-bound conformations

We have successfully prepared E- and Z- isomers of 17-20 dehydro analogs of 2-methylene-19-nor-(20S)-1alpha,25-dihydroxyvitamin D3 (2MD). Both isomers bind to the recombinant rat vitamin D receptor (VDR) with high affinity. The Z-isomer (Vit-III 17-20Z) displays activity in vivo and in vitro that is similar to 2MD. The in vitro activity of the E-isomer (Vit-III 17-20E) is comparable to the natural hormone, though in vivo this analog is significantly less calcemic. Crystal structures of the rat VDR ligand binding domain complexed with the analogs demonstrate that the Vit-III 17-20Z analog is oriented almost identically to 2MD, with only minor differences induced by the planar configuration around the C17-C20 double bond. The Vit-III 17-20E analog is oriented in a conformation distinct from both 2MD and the natural hormone. The structural comparisons suggest that the position of C21 in the ligand binding site may be an important determinant of biological activity.     

In vitro metabolism of 19-nor-1alpha, 25-(OH)2D2 in cultured cell lines: inducible synthesis of lipid- and water-soluble metabolites

The active vitamin D analog, 19-nor-1alpha,25-dihydroxyvitamin D2 (19-nor-1alpha,25-(OH)2D2), has a similar structure to the natural vitamin D hormone, 1a,25-dihydroxyvitamin D3 (1alpha,25-(OH)2D3), but lacks the C10-19 methylene group and possesses an ergosterol/ vitamin D2 rather than a cholesterol/vitamin D3 side chain. We have used this analog to investigate whether any of these structural features has any effect upon the type and rate of in vitro metabolism observed. Using a vitamin D-target cell, the human keratinocyte, HPK1A-ras, we observed formation of a number of metabolites, three of which were purified by extensive HPLC and conclusively identified by a combination of GC-MS and chemical derivatization as 19-nor-1alpha,24,25-(OH) 3D2, 19-nor-1alpha,24,25,26-(OH) 4D2, and 19-nor-1alpha,24,25,28-(OH)4,D2. The first metabolite is probably a product of the vitamin D-inducible cytochrome P450, P450cc24 (CYP24), while the latter two metabolites are likely to be further metabolic products of 19-nor-1alpha,24,25-(OH)3D2. These hydroxylated metabolites resemble those identified by other workers as products of the metabolism of 1alpha,25-(OH)2D2 in the perfused rat kidney. It therefore appears from the similar metabolic fate of 19-nor-1alpha,25-(OH)2D2 and 1alpha,25-(OH)2D2 that the lack of the C10-19 methylene group has little effect upon the nature of the lipid-soluble metabolic products and the rate of formation of these products seems to be comparable to that of products of 1alpha,25-(OH)2D3 in vitamin D-target cells. We also found extensive metabolism of 19-nor-1alpha,25(OH)2D2 to water-soluble metabolites in HPK1A-ras, metabolites which remain unidentified at this time. When we incubated 19-nor-1alpha,25-(OH)2D2 with the liver cell line HepG2, we obtained only 19-nor-1alpha,24,25-(OH)3D2. We conclude that 19-nor-1alpha,25-(OH)2D2 is efficiently metabolized by both vitamin D-target cells and liver cells.     

New highly calcemic 1 alpha,25-dihydroxy-19-norvitamin D(3) compounds with modified side chain: 26,27-dihomo- and 26,27-dimethylene analogs in 20S-series

New highly potent 2-substituted (20S)-1 alpha,25-dihydroxy-19-norvitamin D(3) analogs with elongated side chain were prepared by Wittig-Horner coupling of A-ring phosphine oxide with the corresponding protected (20S)-25-hydroxy Grundmann's ketones. Biologic evaluation in vitro and in vivo of the synthesized compounds was accomplished. All the synthesized vitamins possessing a 25-hydroxylated saturated side chain were slightly less active (3-5X) than 1 alpha,25-dihydroxyvitamin D(3) in binding to the porcine intestinal vitamin D receptor and significantly more potent (12-150X) in causing differentiation of HL-60 cells. In vivo, 2-methylene-26,27-dihomo and 2 alpha-methyl-26,27-dimethylene analogs were at least 10 times more active, and 2 alpha-methyl-26,27-dihomo compound at least 5 times more active than the vitamin D hormone both in stimulating intestinal calcium transport and bone calcium mobilization (serum calcium increase). It was also established that a 260 pmol dose of the corresponding 2 beta-methyl analogs had a similar effect on intestinal calcium transport and a much more pronounced effect on bone calcium mobilization as the same dose of 1 alpha,25-dihydroxyvitamin D(3).     

Methyl substitution of the 25-hydroxy group on 2-methylene-19-nor-1alpha,25-dihydroxyvitamin D3 (2MD) reduces potency but allows bone selectivity

The discovery of 2-methylene-19-nor-1alpha,25-dihydroxyvitamin D3 (2MD) as a bone selective and bone anabolic form of vitamin D has stimulated an investigation of structure/function of bone selectivity. Four new 2-substituted-19-norvitamin D analogs 3-6 have been developed to study the structure-activity relationship at C-25. As predicted, removing the 25-hydroxy group (compound 3) from the very potent analog 2MD and its 2-methyl derivatives (5 and 6) dramatically reduces in vitro activities, but biological potency is nearly fully restored in vivo likely due to in vivo 25-hydroxylation. The introduction of a methyl group at C-25 (compound 4) that blocks in vivo 25-hydroxylation reduces biological activity both in vitro and in vivo. However, analog 4 retains bone selectivity making it interesting as a possible therapeutic for bone loss diseases.     

Asymmetric catalytic ene-cyclization approach to 2-fluoro-19-nor-1,25-dihydroxyvitamin D(3) A-ring analog with significant transactivation activity

1alpha,25-Dihydroxyvitamin D(3) (1alpha,25(OH)2D3) has been shown to modulate not only proliferation and differentiation, but also apoptosis in malignant cells, indicating that it could be useful for the treatment of cancer and psoriasis. However, little information has been available on the binding conformation of the 1alpha,25(OH)2D3 molecule and its analogs with the vitamin D receptor (VDR). Therefore, we synthesized 2alpha-fluorinated A-ring analogs of 19-nor-1alpha,25(OH)2D3 in order to investigate the VDR-binding conformation of the A-rings on the basis of the (19)F NMR analysis. The 2alpha-fluoro-19-nor-1alpha,25-dihydroxyvitamin D3 A-ring analog thus synthesized via a asymmetric catalytic carbonyl-ene cyclization, shows significant activity in transactivation.     

2-methylene-19-nor-20(S)-1alpha-hydroxy-bishomopregnacalciferol [20(S)-2MbisP], an analog of vitamin D3 [1,25(OH)2D3], does not stimulate intestinal phosphate absorption at levels previously shown to suppress parathyroid hormone

Chronic kidney disease results in a reduction in 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) synthesis and an accumulation of phosphorus in the blood, leading to secondary hyperparathyroidism and renal osteodystrophy. Vitamin D analogs that retain the ability to suppress PTH but that are less calcemic and phosphatemic than the native hormone are preferred therapies for secondary hyperparathyroidism. However, even the most favored analog currently approved for the treatment of chronic kidney disease patients, i.e. 1,25-dihydroxy-19-nor-vitamin D2 (19-nor-D2, Zemplar), still retains some ability to stimulate intestinal absorption of calcium and phosphate. A recently described analog of vitamin D3, 2-methylene-19-nor-20(S)-1alpha-hydroxy-bishomopregnacalciferol [20(S)-2MbisP], suppresses PTH levels, but is unable to stimulate intestinal calcium absorption or bone resorption in rats. The present study shows that 20(S)-2MbisP is unable to stimulate intestinal phosphate absorption at levels known to suppress PTH secretion. Further, 19-nor-vitamin D2 under the same circumstances does stimulate phosphate absorption. Thus, 2MbisP has significant potential in the management of secondary hyperparathyroidism of renal failure.     

Production of 10-oxo-19-nor-25-hydroxyvitamin D3 by solubilized kidney mitochondria from chick and rat

Cholate-solubilized chick kidney mitochondria that 1-hydroxylated 25-hydroxyvitamin-D3 (25-OH-D3) upon reconstitution also produced 10-oxo-19-nor-25-OH-D3, which co-eluted with 1,25-dihydroxyvitamin D3 (1,25-(OH)2-D3) on normal phase high performance liquid chromatography (HPLC) with hexane:propanol-2 (9:1), the traditional chromatographic system for isolating 1,25-(OH)2-D3. The 10-oxo derivative was separated from 1,25-(OH)2-D3 by normal phase HPLC with dichloromethane:propanol-2 (19:1) or by reverse phase HPLC with methanol:water (4:1). Unlike 1,25-(OH)2-D3 production, formation of 10-oxo-19-nor-25-OH-D3 did not require a source of reducing equivalents and was blocked by the antioxidants, diphenyl-rho-phenylenediamine, and butylated hydroxytoluene, implicating a free radical or peroxidative synthetic mechanism. Rat kidney mitochondria solubilized with cholate or with cholate and Emulgen 911 produced 10-oxo-19-nor-25-OH-D3 but no detectable 1 alpha,25-(OH)2-D3. These results stress the importance of careful identification of vitamin D metabolites produced in vitro and suggest the use of alternate chromatographic conditions for isolating 1,25-(OH)2-D3 or inclusion of antioxidants in the assay of solubilized 1 alpha-hydroxylase to eliminate contamination of 1,25-dihydroxyvitamin D3 with 10-oxo-19-nor-25-OH-D3.     

Evaluation of C-2-substituted 19-nor-1alpha,25-dihydroxyvitamin D3 analogs as therapeutic agents for prostate cancer

1alpha,25-Dihydroxyvitamin D(3) (1alpha,25(OH)(2)D(3)) is known to inhibit the proliferation and invasiveness of prostate cancer cells. However, 1alpha,25(OH)(2)D(3) can cause hypercalcemia and is not suitable as a therapeutic agent. 19-Nor-vitamin D derivatives are known to be less calcemic when administered systemically. In order to develop more potent anti-cancer agents with less calcemic side effect, we therefore utilized (3)H-thymidine incorporation as an index for cell proliferation and examined the antiproliferative activities of nine C-2-substituted 19-nor-1alpha,25(OH)(2)D(3) analogs in the immortalized PZ-HPV-7 normal prostate cell line. Among the nine analogs we observed that the substitution with 2alpha- or 2beta-hydroxypropyl group produced two analogs having antiproliferative potency that is approximately 500- to 1000-fold higher than 1alpha,25(OH)(2)D(3). The (3)H-thymidine incorporation data were supported by the cell counting data after cells were treated with 1alpha,25(OH)(2)D(3), 19-nor-2alpha-(3-hydroxypropyl)-1alpha,25(OH)(2)D(3) or 19-nor-2beta-(3-hydroxypropyl)-1alpha,25(OH)(2)D(3) for 7 days. 19-Nor-2alpha-(3-hydroxypropyl)-1alpha,25(OH)(2)D(3) and 19-nor-2beta-(3-hydroxypropyl)-1alpha,25(OH)(2)D(3) were also shown to be about 10-fold more active than 1alpha,25(OH)(2)D(3) in cell invasion studies using prostate cancer cells. In conclusion, a substitution at the C-2 position of 19-nor-1alpha,25(OH)(2)D(3) molecule with a hydroxypropyl group greatly increased the antiproliferative and anti-invasion potencies. Thus, these two analogs could be developed to be effective therapeutic agents for treating early and late stages of prostate cancer.     

Differential effects of Vitamin D analogs on calcium transport

It is well known that the efficiency of intestinal active calcium transport is regulated by the Vitamin D receptor pathway and Vitamin D analogs seem to exhibit differential effects on intestinal active calcium transport. To investigate the molecular basis for the difference among Vitamin D analogs, we tested three Vitamin D analogs: 1,25-dihydroxyvitamin D(3), 19-nor-1,25-dihydroxyvitamin D(2), and 1alpha-hydroxyvitamin D(2) ex vivo and in vitro. In 5/6 nephrectomized rat intestinal active calcium transport, 19-nor-1,25-dihydroxyvitamin D(2) did not show a significant effects on intestinal active calcium transport at all the concentrations tested, while 1alpha-hydroxyvitamin D(2) at 0.33 and 0.67 microg/kg and 1,25-dihydroxyvitamin D(3) at 1microg/kg significantly stimulated calcium transport. In Caco-2 cells, 19-nor-1,25-dihydroxyvitamin D(2) did not show a significant effect on calcium transport, while 1,25-dihydroxyvitamin D(3) and 1,25-dihydroxyvitamin D(2) (the active form of 1alpha-hydroxyvitamin D(2)) stimulated calcium transport by 934 and 501% at 0.1microM, respectively. 1,25-Dihydroxyvitamin D(2) potently induced the expression of CALB3 and TRPV6 mRNA with an EC(50) of 0.3 and 1.0nM, whereas 19-nor-1,25-dihydroxyvitamin D(2) was 10-fold less potent than 1,25-dihydroxyvitamin D(2) in inducing CALB3 and TRPV6 mRNA. The three Vitamin D analogs had no significant effect on the expression of PMCA1 mRNA. These Vitamin D analogs did not change the expression of Vitamin D receptor (VDR) up to 10nM, but stimulated CYP24A1 expression in a dose-dependent manner with the potency in the order of 1,25-dihydroxyvitamin D(3)>1,25-dihydroxyvitamin D(2)=19-nor-1,25-dihydroxyvitamin D(2). These results suggest that the differential effect of Vitamin D analogs on stimulating intestinal and Caco-2 calcium transport may be in part due to its different effect on stimulating CALB3 and TRPV6 mRNA expression.     

On the specificity of vitamin D compounds binding to chick pig intestinal 1,25-dihydroxyvitamin D3 receptor

The binding activity of four vitamin D metabolites and/or analogs for the intestinal 1,25-dihydroxyvitamin D3 receptor was evaluated after incubation at 25 degrees C for 1 h or at 0-4 degrees C for 18 h. The incubation conditions, which had no effect on the binding of 1,25-dihydroxyvitamin D3, had a dramatic effect on the binding of 25-hydroxyvitamin D3 and 1 alpha-hydroxyvitamin D3 and a small but reproducible effect on 24,25-dihydroxyvitamin D3 binding to receptor. Affinities 10- to 20-fold higher were obtained for 25-hydroxyvitamin D3 and 1 alpha-hydroxyvitamin D3, and affinities 3-fold higher were obtained for 24,25-dihydroxyvitamin D3 at the 0-4 degrees C/18-h incubation. A comparison of intestinal receptor from chick and pig with nine vitamin D compounds showed no major differences between the two species. The relative affinity of the vitamin D analogs to compete with tritiated 1,25-dihydroxyvitamin D3 for the receptor in pig nuclear extract, expressed as ratios of the molar concentration required for 50% binding of the tritiated 1,25-dihydroxyvitamin D3 compared to nonradioactive 1,25-dihydroxyvitamin D3, are as follows: 1,25-dihydroxyvitamin D3 (1) = 1,25-dihydroxyvitamin D2 = 24-homo-1,25-dihydroxyvitamin D3 greater than 1,24,25-trihydroxyvitamin D3 (4) greater than 25-hydroxyvitamin D3 (21) = 10-oxo-19-nor-25-hydroxyvitamin D3 = 1 alpha-hydroxyvitamin D3 (37) greater than 24,25-dihydroxyvitamin D2 (257) much much greater than vitamin D3 (greater than 10(6)).