Altered Pharmacokinetics of 1,25-Dihydroxyvitamin D3 and 25-Hydroxyvitamin D3 in the Blood and Tissues of the 25-Hydroxyvitamin D-24-Hydroxy
http://www.100md.com
内分泌学杂志 2005年第2期
Department of Biochemistry, Queen’s University (S.M., V.B., G.J.), Kingston, Ontario, Canada K7L 3N6; Genetics Unit, Shriners Hospital for Children (A.A., R.S.-A.), Montréal, Québec, Canada H3G 1A6; Departments of Medicine, Surgery, and Human Genetics, McGill University (R.S.-A.), Montréal, Québec, Canada H3A 2T5; and Endocrine Unit, Massachusetts General Hospital, Harvard Medical School (Y.S., M.B.D.), Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Dr. Glenville Jones, Department of Biochemistry, Queen’s University, Kingston Ontario, Canada K7L 3N6. E-mail: gj1@post.queensu.ca.
Abstract
The 25-hydroxyvitamin D-24-hydroxylase (CYP24A1) plays an important role in regulating concentrations of both the precursor 25-hydroxyvitamin D3 [25(OH)D3] and the hormone 1,25-dihydroxyvitamin D3 [1,25(OH)2D3]. Previous studies suggest that Cyp24a1-null mice cannot clear exogenous 1,25(OH)2D3 efficiently. Here, we examined the metabolic clearance in Cyp24a1-null mice in vivo and in vitro using a physiological dose of [1?-3H]1,25(OH)2D3 or [26,27-methyl-3H]25(OH)D3. Cyp24a1-null mice showed difficulty in eliminating [1?-3H]1,25(OH)2D3 from the bloodstream and tissues over a 96-h time course, whereas heterozygotic mice eliminated the hormone within 6–12 h, although there was clearance of labeled hormone into water-soluble products involving liver in both genotypes. RT-PCR showed that Cyp24a1-null mice have decreased expression of 25-hydroxyvitamin D-1-hydroxylase that must play a role in their survival. After the administration of [26,27-methyl-3H]25(OH)D3, Cyp24a1-null mice showed higher [26,27-methyl-3H]25(OH)D3 levels and no [26,27-methyl-3H]24,25(OH)2D3 formation, whereas heterozygotic mice showed significant [26,27-methyl-3H]24,25(OH)2D3 production. Based upon in vitro experiments, keratinocytes from Cyp24a1-null mice fail to synthesize [1?-3H]calcitroic acid from [1?-3H]1,25(OH)2D3 or [26,27-methyl-3H]24,25(OH)2D3 from [26,27-methyl-3H]25(OH)D3 as do control mice, confirming the target cell catabolic role of CYP24A1 in these processes. Finally, the role of vitamin D receptor (VDR) in the vitamin D catabolic cascade was examined using VDR-null mice. Keratinocytes from VDR-null mice failed to metabolize [1?-3H]1,25(OH)2D3 confirming the importance of vitamin D-inducible, VDR-mediated, C24 oxidation pathway in target cells. These results suggest that the absence of CYP24A1 or VDR retards catabolism of 1,25(OH)2D3 and 25(OH)D3, reinforcing the physiological importance of CYP24A1 in vitamin D homeostasis.
Introduction
THE SEQUENCING OF the human genome has led to the realization that among a pool of 60 total cytochrome P450s (CYPs),1 there are three known and possibly other uncharacterized vitamin D-related CYPs dedicated to the metabolism of vitamin D. The key enzymes in vitamin D metabolism are the hepatic vitamin D-25-hydroxylase [CYP27A1 and possibly CYP2R1 (1)], renal 25-hydroxyvitamin D-1-hydroxylase (CYP27B1), and 25-hydroxyvitamin D-24-hydroxylase (CYP24A1). Biological activation of vitamin D occurs through the sequential hydroxylation catalyzed by CYP27A1 or CYP2R1 and CYP27B1, resulting in the formation of the active form of vitamin D3, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] (2, 3, 4). CYP24A1 is the enzyme acting on 1,25(OH)2D3 to initiate degradation of this potent metabolite and is tightly regulated. This enzyme is distributed in classical vitamin D target tissues, including kidney, intestine, bone, and skin. In target cells, 1,25(OH)2D3 has the ability to induce its own degradation by up-regulating CYP24A1 expression to attenuate 1,25(OH)2D3 action or prevent vitamin D toxicity (i.e. hypercalcemia) (5, 6, 7). CYP24A1 was discovered as the mitochondrial enzyme responsible for the hydroxylation at C-24 in the metabolism of 1,25(OH)2D3 and 25-hydroxyvitamin D3 [25(OH)D3] (8, 9, 10). C-24 oxidation represents the predominant catabolic pathway for 1,25(OH)2D3 in which sequential C-24 hydroxylation, C-24 ketonization, and C-23 hydroxylation, followed by oxidative cleavage to form tetranor-1,23(OH)2D3 and calcitroic acid, liberate an inactive, water-soluble product that is excreted in bile (5, 11). The secondary pathway involves C-26 hydroxylation, C-23 hydroxylation, and lactonization to form (23S,25R)-1,25(OH)2D3-26,23-lactone (12, 13). Both pathways have also been shown to participate in the metabolism of 25(OH)D3 (13). Studies of the recombinant CYP24A1 protein produced in Escherichia coli (14, 15), insect (16), and mammalian (17) cells have revealed that CYP24A1 exerts multicatalytic activity, facilitating sequential oxidation of C-23, C-24, and C-26 hydroxylation and side-chain cleavage.
The role of CYP24A1 has been established to be inactivation of both 1,25(OH)2D3 and 25(OH)D3 (5, 6, 13). Most evidence from kinetic analysis supports the concept that 1,25(OH)2D3, rather than 25(OH)D3, is the preferred substrate for CYP24A1 (18) and that this enzyme plays an important role in regulating the intracellular concentration as well as the half-life of 1,25(OH)2D3 in the body. Recently, this was also confirmed by the generation of a Cyp24a1 knockout mouse model (19). Initial reports suggest that Cyp24a1-null mice show prolonged elevations in plasma 1,25(OH)2D3 levels after exogenous 1,25(OH)2D3 administration. These findings are consistent with the poor viability of the Cyp24a1-null mice and its tendency to hypercalcemia and nephrocalcinosis during the neonatal period. Interestingly, Cyp24a1 mutant animals that survived past weaning showed lower baseline levels of circulating 1,25(OH)2D3 than wild-type controls. This fact raises questions about the physiological role of CYP24A1 in the vitamin D metabolism. It has been postulated that these animals may use an alternative pathway of 1,25(OH)2D3 catabolism to regulate circulating levels of the hormone.
The action of 1,25(OH)2D3 in target cells is mediated by the vitamin D receptor (VDR), which binds to vitamin D response elements in the promoter of genes. Two vitamin D response elements in the CYP24A1 promoter are responsible for the VDR-mediated up-regulation of CYP24A1 gene transcription (20, 21). In mice lacking the VDR gene, highly elevated plasma levels of 1,25(OH)2D3 were measured due to the impaired ligand-dependent activation of Cyp24a1 gene transcription and inadequate catabolic degradation of 1,25(OH)2D3 (22, 23). Hypocalcemia and a loss of VDR-mediated gene expression caused Cyp24a1 expression to be low in such animals. Thus, because of the inducible nature of vitamin D-target cell CYP24A1, the VDR-null mouse represents the equivalent of a partial Cyp24a1 knockout mouse model. Accordingly, a comparison of CYP24A1 enzyme activity in VDR-null mice and Cyp24a1-null mice would allow us to understand the roles of both VDR and CYP24A1 in vitamin D signaling and catabolic cascade.
The development of novel gene expression systems for CYPs has allowed examination of some long-standing questions in vitamin D metabolism. In the present study we focused on 1) elucidation of the overall importance of CYP24A1 in 1,25(OH)2D3 and 25(OH)D3 clearance, 2) direct comparison of the metabolism of 1,25(OH)2D3 and 25(OH)D3 in the Cyp24a1-null mouse in vivo and in vitro, and 3) elucidation of the role of VDR in 1,25(OH)2D3 metabolism using keratinocytes from the VDR-null mouse in vitro. The results confirm that ablation of either the Cyp24a1 or VDR gene profoundly retards metabolism of 1,25(OH)2D3 and 25(OH)D3 and that the surviving Cyp24a1-null mice owe their survival in part to down-regulation of Cyp27b1 expression.
Materials and Methods
Materials
1,25(OH)2D3, 25(OH)D3, 1,24,25(OH)3D3, tetranor-1,23(OH)2D3, 24,25(OH)2D3, and 1(OH)D3 were gifts from Drs. M. Calverley and A. M. Kissmeyer (Leo Pharmaceutical Products, Ballerup, Denmark). (23S,25R)-1,25(OH)2D3-26,23-lactone was a gift from Dr. S. Ishizuka (Teijin, Japan). [26,27-Methyl-3H]25(OH)D3 (30 Ci/mmol) was purchased from Amersham Biosciences (Little Chalfont, UK). [1?-3H]1,25(OH)2D3 (50 Ci/mmol) and calcitroic acid were synthesized by methods described previously (5). MEM, trypsin, and antibiotic/antimyotic were purchased from Invitrogen Life Technologies (Burlington, Canada). Fetal calf serum (FCS) was purchased from ICN (Costa Mesa, CA), and BSA was obtained from Roche (Laval, Canada). Chelex 100 was purchased from Bio-Rad Laboratories (Hercules, CA). Collagen (Vitrogen 100) was purchased from Cohesion Technologies, Inc. (Palo Alto, CA). Epidermal growth factor (EGF) and N'N-diphenyl-p-phenylenediamine (DPPD) were supplied by Sigma-Aldrich Corp. (St. Louis, MO). Liquid scintillation cocktail (Ready Flow III) for on-line radioactivity detection was purchased from Beckman Coulter, Inc. (Fullerton, CA). All organic solvents were of HPLC grade and were purchased from Caledon Laboratories Ltd. (Georgetown, Canada).
Animals
All studies performed were reviewed and approved by the institutional animal care and use committees. Animals were maintained in a virus- and parasite-free barrier facility and exposed to a 12-h light, 12-h dark cycle. Mice lacking Cyp24a1 were obtained as previously described (19). Genotyping for the Cyp24a1 mutation was performed by Southern blotting of BamHI digests using an EcoRI-PstI genomic probe downstream of the region of homology. The 12-kb wild-type allele and the 8.5-kb targeted allele were identified. Mice lacking VDR were obtained as previously described (22). VDR-null mice were genotyped by PCR analysis of tail DNA using primers specific for the neomycin resistance gene and the second zinc finger of the VDR.
In vivo metabolism study in Cyp24a1-null mice
For the 1,25(OH)2D3 metabolism study, 6-month-old Cyp24a–/– homozygotes and their Cyp24a+/– littermates received a single iv injection of [1?-3H]1,25(OH)2D3 (1 μCi; 8 ng) in 100 μl of a vehicle containing ethanol/mouse serum (1:9). At 3, 6, 24, 48, and 96 h postinjection, blood was taken by cardiac puncture, and liver and kidneys were collected. For the 25(OH)D3 metabolism study, 6-month-old Cyp24a1–/– homozygotes and their Cyp24a+/– littermates received a single iv injection of [26,27-methyl-3H]25(OH)D3 (5 μCi; 67 ng) in 100 μl of a vehicle containing ethanol/mouse serum (1:9). At 6, 24, and 48 h postinjection, blood was taken by cardiac puncture and liver, kidneys and intestine were collected. Plasma was obtained by centrifugation at 3000 rpm for 15 min. Liver, kidneys, and intestine were frozen in liquid N2 and stored at –20 C.
Isolation and preparation of primary keratinocytes
Primary keratinocytes were isolated from 2- to 3-d-old Cyp24a1-null mice and control littermates by a trypsin floating procedure as described previously (24). Briefly, the skin was dissected from the trunk of mice and floated on 0.25% trypsin for 16–20 h at 4 C. The epidermis was then peeled off the dermis, minced with scissors, and stirred in MEM with 4% (vol/vol) Chelex-treated FCS, EGF (10 ng/ml), 1% (vol/vol) antibiotic/antimycotic, and 0.05 mM CaCl2 (low calcium medium) for 1 h at 4 C. The resultant cell suspension was then filtered through a 70-μm pore size filter (Falcon, BD Biosciences, Franklin Lakes, NJ) to remove clumps of cells and debris. The keratinocytes were pelleted by centrifugation for 5 min at room temperature and plated in low calcium medium in collagen-coated, 60-mm dishes. Cells were maintained at 37 C in a humidified atmosphere of 5% CO2 in air. At 80–90% confluence, cells were reseeded at 1.8 x 105 cells/well in six-well plates in low calcium medium and grown to 80% confluence before addition of substrate.
In vitro metabolism study of 1,25(OH)2D3 and 25(OH)D3 in Cyp24a1-null mouse keratinocytes
Primary keratinocytes obtained from each mouse were treated with 10 nM 1,25(OH)2D3 (in ethanol) to induce the transcription of Cyp24a1. After 18 h, the cells were rinsed twice with PBS and incubated with 0.25 μCi [1?-3H]1,25(OH)2D3 (2.1 ng; 2.5 nM) in 2 ml low calcium MEM supplemented with 1% (wt/vol) BSA, 10 ng/ml EGF, 1% (vol/vol) antibiotic/antimycotic, and 100 μM DPPD for 24 h at 37 C. For the time-course study of 25(OH)D3, mouse primary keratinocytes that were not previously induced with 10 nM 1,25(OH)2D3 were incubated with 0.5 μCi [26,27-methyl-3H]25(OH)D3 (6.6 ng; 16.6 nM) in 2 ml low calcium MEM supplemented with 1% (wt/vol) BSA, 10 ng/ml EGF, 1% (vol/vol) antibiotic/antimycotic, and 100 μM DPPD for 0, 1, 3, 6, and 12 h at 37 C.
In vitro metabolism study in VDR-null mice keratinocytes
Mouse primary keratinocytes from VDR-null and wild-type mice were plated at 2 x 106 cells/plate in P-100 plates and grown to 80% confluence in low calcium MEM with 4% (vol/vol) Chelex-treated FCS, 10 ng/ml EGF, and 1% (vol/vol) antibiotic/antimycotic. Cells were incubated with 0.5 μCi [1?-3H]1,25(OH)2D3 (4.16 ng; 2.5 nM) including 1 μM 1,25(OH)2D3 or 10 μM 1,25(OH)2D3 in 4 ml low calcium MEM supplemented with 1% (wt/vol) BSA, 10 ng/ml EGF, 1% (vol/vol) antibiotic/antimycotic, and 100 μM DPPD for 24 h at 37 C.
RT-PCR analysis of CYP27B1 transcripts
The RT-PCR protocol used total RNA prepared with the TRIzol reagent (Invitrogen Life Technologies, Inc., Gaithersburg, MD) from the kidneys of 4-month-old Cyp24a1+/– and Cyp24a1–/– animals (25). The RT reaction contained oligo(deoxythymidine) and SuperScript II reverse transcriptase (Invitrogen Life Technologies, Inc.). PCR amplification used the following primers for Cyp27b1: 5'-CTGCGAGGAGGGGTAAGGTGTT-3' and 5'-GGAAACGGGGGAGGGGA-3'. Amplification was for 28 cycles. Control ?-actin amplification reactions used the primer pair 5'-GCTGCGTGTGGCCCCTAGG-3' and 5'-CAAGAAGGAAGGCTGGAAAAGAG-3'. Amplimers were detected by ethidium bromide staining of agarose electrophoresis gels.
Lipid extraction
Before the extraction, plasma was adjusted up to a volume of 1 ml with ice-cold buffered 0.9% saline, and tissues were homogenated in ice-cold buffered 0.9% saline (4 ml/g tissue). An internal standard of 2 μg 1(OH)D3 was added to blood and tissue homogenate samples to assess recovery during the extraction process. For the cell culture, incubations were terminated at designated time points with methanol. All samples were extracted using a modification of the method of Bligh and Dyer (26), in which chloroform was replaced by methylene chloride. The aqueous layer was reextracted with methylene chloride and 0.1% glacial acetic acid to isolate the water-soluble metabolites. After the extraction, aliquots of organic and aqueous fractions were subjected to liquid scintillation counter (LC1800, Beckman Coulter, Inc., Fullerton, CA). All samples were dried down under a stream of prepurified N2 at 37 C, resuspended, and stored in ethanol.
HPLC
Metabolites from in vivo or in vitro study were analyzed by either straight phase (organic-soluble metabolites) or reverse phase (water-soluble metabolites) HPLC using an Alliance 2695 Separations Module and a model 996 photodiode array detector (Waters Associates, Milford, MA). Radiolabeled organic- or water-soluble metabolites were detected using a RadioFlow Detector (LB 509, EG&G Berthold, Bad Wildbad, Germany). The organic phase extracts were dried down, redissolved, and separated on a Zorbax SIL column (3 μm; 62 x 80 mm; Agilent Technologies, Palo Alto, CA) with a hexane/isopropanol/methanol (HIM; 91%:7%:2%, vol/vol/vol) isocratic system at a flow rate of 1.0 ml/min. The water-soluble metabolites were separated on a Zorbax SB-C18 column (5 μm; 4.6 x 150 mm; Agilent Technologies) with a linear gradient solvent system, which started at 69%:30%:1% (vol/vol/vol) water/acetonitrile/10% acetic acid and was brought to 0%/99%/1% (vol/vol/vol) water/acetonitrile/10% acetic acid by 30 min at a flow rate of 1.0 ml/min. Metabolites were identified based on the characteristic vitamin D cis-triene chromophore (max = 265 nm; min = 228 nm; max/min = 1.75). Data analysis was performed with the aid of Millennium32 chromatographic software (Waters Associates, Milford, MA).
Statistical analysis
The significance of the difference between the means of the two groups was analyzed by t test, and differences were taken as statistically significant at P < 0.001 (***), P < 0.01 (**), or P < 0.05 (*).
Results
Metabolism of 1,25(OH)2D3 in Cyp24a1-null mice in vivo
Initial studies focused on the metabolic clearance and blood and tissue levels of [1?-3H]1,25(OH)2D3 and its circulating metabolites over a 96-h period in Cyp24a1–/– homozygotic mice and their Cyp24a1+/– littermates after a single iv injection of 1 μCi [1?-3H]1,25(OH)2D3. The results shown in Fig. 1 indicate that the marked differences in lipid-soluble radioactivity between Cyp24a1+/– control and Cyp24a1–/– mice were evident by as early as 6 h. Even at 6 h, organic fractions from blood, liver, and kidney tissues from Cyp24a1–/– mice showed approximately 2-fold higher [1?-3H]1,25(OH)2D3 levels than Cyp24a1+/– controls. By 96 h this difference became 20-fold for blood, 7-fold for liver, and 50-fold for kidney. [1?-3H]1,25(OH)2D3 was still identifiable in blood, liver, and kidney of Cyp24a1–/– mice at 96 h. The half-life (t1/2) of [1?-3H]1,25(OH)2D3 in Cyp24a1–/– mice was significantly longer than that in Cyp24a1+/– controls in blood (t1/2, 39 h vs. 4 h), liver (t1/2, 16 h vs. 5 h), and kidney (t1/2, 12 h vs. 5 h). The lipid-soluble 3H radioactivity remained mainly in the form of 1,25(OH)2D3, except in blood and liver of Cyp24a1+/– mice at early time points, where [1?-3H]1,25(OH)2D3-26,23-lactone was also detectable. [1?-3H]1,25(OH)2D3-26,23-lactone was completely absent from the blood of Cyp24a1–/– mice despite the fact that a measurable peak of [1?-3H]1,25(OH)2D3 was always greater in chromatograms of Cyp24a1–/– mice than in Cyp24a1+/– mice. As a result of these initial studies, subsequent experiments performed by our group emphasized the need to explore the water-soluble metabolites in more detail.
FIG. 1. Levels of [1?-3H]1,25(OH)2D3 in Cyp24a1–/–and Cyp24a1+/– mice after a single bolus dose. [1?-3H]1,25(OH)2D3 levels in blood, liver, and kidney in Cyp24a1-null mice were compared with heterozygotic littermates at 6, 24, 48, and 96 h after a single iv administration of 1 μCi [1?-3H]1,25(OH)2D3. Lipids from the extraction were separated on Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Radioactivity measured using radioisotope detector and identified as [1?-3H]1,25(OH)2D3 from its retention time is plotted. Each point represents the mean ± SE of three animals. Differences were statistically significant as indicated: *, P < 0.05; **, P < 0.01; ***, P < 0.001 (compared with Cyp24a1+/– mice by t test).
A second experiment, shown in Table 1, represents a time course over a 3- to 48-h period, where we observed a higher level of total radioactivity in the blood of Cyp24a1–/– animals by 24 h, a difference that became statistically significant by 48 h. The rapid clearance of [1?-3H]1,25(OH)2D3 from the blood of Cyp24a1+/– was also accompanied by the conversion of the lipid-soluble hormone into water-soluble products, such that over the 3- to 48-h period the percentage of recovered radioactivity found in the water-soluble fraction of blood rose dramatically from 45% to 74%. In contrast, the slower clearance of [1?-3H]1,25(OH)2D3 from the blood of the Cyp24a1–/– mice was accompanied by a modest increase in the same parameter (from 36% to 43% over 3–48 h), with the organic phase accounting for between 64% and 57% of the total radioactivity. In both genotypes, there was a rapid uptake of [1?-3H]1,25(OH)2D3 or its metabolites into liver and kidney by as early as 3 h. In the liver, not only were there more water-soluble products in both genotypes, but also the differences in total radioactivity were not as marked as observed in blood and kidney. Again in the liver of Cyp24a1+/– mice, there was a significant rise in the percentage of water-soluble radioactivity, increasing from 60% to 72% over the 3- to 48-h time period, a change that was absent in Cyp24a1–/– mice (58%, 53%, 64%, and 58% in the same 3- to 48-h period). In the kidney, arguably representing a typical vitamin D target tissue, the picture was even more pronounced, with significant differences in total radioactivity apparent by 6 h and a markedly lower percentage of the radioactivity residing in the organic fraction as unchanged [1?-3H]1,25(OH)2D3 in Cyp24a1+/– mice compared with Cyp24a1–/– mice. Indeed, in the Cyp24a1+/– mice, the water-soluble metabolites rose from 57% to 78% over the 3- to 48-h period, whereas in the Cyp24a1–/–– mice this parameter fairly stable at around 33%, or even fell.
TABLE 1. Time course of clearance of 1,25(OH)D2D3 in Cyp24a1–/– and Cyp24a1+/– mice after the administration of 1 μCi [1?-3H]1,25(OH)2D3
RT-PCR analysis of CYP27B1 transcripts in Cyp24a1-null mice
A previous report (19) found that the basal circulating levels of 1,25(OH)2D3 were approximately 3-fold lower in Cyp24a1–/– mice that survived past weaning than in wild-type controls. To examine whether surviving Cyp24a1–/– mice are capable of down-regulation of 1,25(OH)2D3 production, the expression of Cyp27b1 in Cyp24a1–/– mouse kidney was investigated. RT-PCR analysis using total RNA prepared from kidney tissue of 4-month-old Cyp24a1–/– and Cyp24a1+/– mice revealed detectable transcripts of Cyp27b1 mRNA in Cyp24a1+/–, but not Cyp24a1–/– mice (Fig. 2) despite the fact that there was equal expression of ?-actin mRNA. These observations suggest that the surviving Cyp24a1–/– mice may owe their survival in part to lower 1,25(OH)2D3 biosynthesis in the kidney.
FIG. 2. Cyp27b1 [25-hydroxyvitamin D-1-hydroxylase (1-OHase)] expression in the kidney from Cyp24a1–/– and Cyp24a1+/– mice. The expression of Cyp27b1 was analyzed by RT-PCR using total RNA prepared from kidney tissue of 4-month-old Cyp24a1–/– and Cyp24a1+/– mice. The conditions for the RT-PCR were described in Materials and Methods. The 451-bp Cyp27b1 amplimer and the 522-bp ?-actin amplimer are shown. Three animals were probed for each genotype; genotypes are identified below the figure. Note the dramatic reduction in Cyp27b1 expression in Cyp24a1–/– mice.
Metabolism of 1a,25(OH)2D3 in Cyp24a1-null mice keratinocytes in vitro
We next investigated the in vitro metabolism of [1?-3H]1,25(OH)2D3 using primary neonatal keratinocytes from Cyp24a1–/– and Cyp24a1+/– animals. When 0.25 μCi [1?-3H]1,25(OH)2D3 was incubated for 24 h after the induction of Cyp24a1 transcription, keratinocytes from Cyp24a1+/– mice demonstrated the complete breakdown of [1?-3H]1,25(OH)2D3 and the only metabolite identified in the organic fraction was [1?-3H]tetranor-1,23(OH)2D3 at a retention time of 17.73 min (Fig. 3A), whereas keratinocytes from Cyp24a1–/– mice failed to degrade substrate. The major metabolite observed was water soluble and ran as a discrete peak with a retention time of 11.13 min identical to that of the standard for calcitroic acid (Fig. 3B). In contrast, Cyp24a1–/– keratinocytes showed no metabolites generated through the C-24 oxidation pathway, and most of the radioactivity remained as [1?-3H]1,25(OH)2D3 in the organic fraction (Fig. 3A). Interestingly, despite the existence of high levels of [1?-3H]1,25(OH)2D3 in Cyp24a1–/– keratinocytes, neither C-23 nor C-26 oxidation was detected during the incubation. Furthermore, there was no evidence of alternative target cell pathways in Cyp24a1–/– keratinocytes that may substitute for C-24 oxidation and contribute to the catabolism of [1?-3H] 1,25(OH)2D3.
FIG. 3. HPLC profile of metabolites of [1?-3H]1,25(OH)2D3 generated by incubation of primary keratinocyte from Cyp24a1–/– and Cyp24a1+/– mice with 0.25 μCi [1?-3H] 1,25(OH)2D3 for 24 h after induction with 10 nM 1,25(OH)2D3. Lipid-soluble metabolites (A) were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) using a flow rate of 1.0 ml/min. Water-soluble metabolites (B) were separated on a Zorbax SCB18 (5 μm; 4.6 x 150 mm) column using a linear water-acetonitrile gradient system from 30–100% acetonitrile with 0.1% acetic acid over 30 min at a flow rate of 1.0 ml/min. Metabolites of [1?-3H]1,25(OH)2D3 were identified by flow radioactivity detection using synthetic standards. The experiment was performed using five to six animals from each group, and a representative chromatogram is shown.
Metabolism of 25(OH)D3 in Cyp24a1-null mice in vivo
To compare the metabolic clearance of 25(OH)D3 in Cyp24a1–/– and Cyp24a1+/– mice, we administered a 5-μCi dose of [26,27-methyl-3H]25(OH)D3 and followed the disappearance of [26,27-methyl-3H]25(OH)D3 and the production of C-24 oxidation products in blood and tissues. In comparison with the metabolic clearance measured for [1?-3H]1,25(OH)2D3, [26,27-methyl-3H]25(OH)D3 was clearly more slowly metabolized in both Cyp24a1–/– and Cyp24a1+/– mice. The obvious difference was observed in blood, where the pool of vitamin D-binding protein (DBP) presumably sequesters 25(OH)D3 (Table 2). Whereas the total radioactivity in the blood fell slightly from 30% to 20% of the dose in Cyp24a1+/– mice over the 48-h time course, this parameter dropped more slowly in the Cyp24a1–/– mice (39% at 48 h). This trend toward reduced rates of clearance in Cyp24a1–/– mice was more exaggerated when we focused on the amount of unchanged residual [26,27-methyl-3H]25(OH)D3 surviving in the blood and tissues. Figure 4 presents unchanged [26,27-methyl-3H]25(OH)D3 in the lipid-soluble fraction of blood and tissues in Cyp24a1+/– and Cyp24a1–/– mice after a bolus dose. By 48 h, differences in unchanged [26,27-methyl-3H]25(OH)D3 ratios were apparent in Cyp24a1+/– and Cyp24a1–/– mice (2.4-, 1.9-, 3.5- and 2.9-fold for blood, liver, kidney, and intestine, respectively).
TABLE 2. Time course of clearance and metabolism of 25(OH)D3 in Cyp24a1–/– and Cyp24a1+/– mice after the administration of 5 μCi [26,27-methyl-3H]25(OH)D3
FIG. 4. Levels of [26,27-methyl-3H]25(OH)D3 in Cyp24a1–/– and Cyp241a+/– mice after a single bolus dose. [26,27-Methyl-3H]25(OH)D3 levels in blood, liver, kidney, and intestine from Cyp24a1-null mice were compared with their heterozygotic littermates at 6, 24, and 48 h after a single iv administration with 5 μCi [26,27-methyl-3H]25(OH)D3. Lipids from the extraction were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Radioactivity measured using flow radioisotope detector and identified as [26,27-methyl-3H]25(OH)D3 from its retention time is plotted. Each data point is the mean ± SE of two animals. Differences were statistically significant as indicated: *, P < 0.05; **, P < 0.005 (compared with Cyp24a1+/– mice by the t test).
Although the lipid-soluble 3H remained mainly as [26,27-methyl-3H]25(OH)D3, Cyp24a1+/– mice showed the production of [26,27-methyl-3H]24,25(OH)2D3 and its subsequent metabolic product, [26,27-methyl-3H]24-oxo-25(OH)2D3, from 6 h postinjection onward, and the two metabolites were still accumulating 48 h after the injection in blood, liver, kidney, and intestine (Fig. 5 and Table 2). Unlike Cyp24a1+/– littermates, a total absence of [26,27-methyl-3H]24,25(OH)2D3 was observed in Cyp24a1–/– mice. Additional metabolites, such as [26,27-methyl-3H]1,25(OH)2D3 or [26,27-methyl-3H]1,24,25(OH)3D3, were not detected in either Cyp24a1+/– or Cyp24a1–/– mice at any time point in the blood and all tissues tested.
FIG. 5. HPLC profiles of blood and tissue samples extracted from Cyp24a1–/– and Cyp24a1+/– mice at 48 h after iv administration of [26,27-methyl-3H]25(OH)D3. Lipid extracts from blood, liver, kidney, and intestine (4.5 x 105 dpm of each) were chromatographed on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Metabolites of [26,27-methyl-3H]25(OH)D3 were identified by retention time with synthetic standards after detection using a flow radioisotope detector. The experiment was performed using two animals, and a typical chromatogram from each group is depicted.
Metabolism of 25(OH)D3 in Cyp24a1-null mice keratinocytes in vitro
For investigation of the kinetic metabolic profile of 25(OH)D3 in Cyp24a1-null mouse keratinocytes, cells were incubated with 0.5 μCi [26,27-methyl-3H]25(OH)D3 for various times. Despite the fact that the keratinocytes from Cyp24a1+/– mice were cultured with [26,27-methyl-3H]25(OH)D3 under DBP-free medium for up to 12 h, 80% of the radioactivity remained in the organic fraction. The only metabolite observed in the organic fraction of Cyp24a1+/– cells was [26,27-methyl-3H]24,25(OH)2D3, a metabolite that appeared as early as 3 h after the incubation and increased in the extracts up to 12 h (Fig. 6). Extracts from Cyp24a1–/– keratinocytes showed higher residual [26,27-methyl-3H]25(OH)D3 levels and no [26,27-methyl-3H] 24,25(OH)2D3 production over the full 12-h time course (Fig. 6). No peak corresponding to [26,27-methyl-3H]1,25(OH)2D3 or [26,27-methyl-3H]1,24,25(OH)3D3 was found at any time point in any keratinocyte culture tested.
FIG. 6. HPLC profiles of metabolites of [26,27-methyl-3H]25(OH)D3 generated in primary keratinocyte culture of Cyp24a1–/– and Cyp24a1+/– mice. Keratinocytes were incubated with 0.5 μCi [26,27-methyl-3H]25(OH)D3 for 0, 1, 3, 6, and 12 h. Lipid-soluble metabolites were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Metabolites of [26,27-methyl-3H]25(OH)D3 were identified by flow radioactivity detection using synthetic standards. The experiment was performed using three to seven animals in each group, and representative chromatograms are shown for 0, 6, and 12 h.
Metabolism of 1,25(OH)2D3 in VDR-null mice keratinocytes in vitro
Because of the inducible nature of vitamin D target cell CYP24A1, VDR-null mice represent a partial Cyp24a1 knockout model. To monitor CYP24A1 activity in vitamin D target cells by studying [1?-3H]1,25(OH)2D3 catabolism, primary keratinocytes from VDR–/– and VDR+/+ mice were isolated and incubated with [1?-3H]1,25(OH)2D3 (specific activity, 125 mCi/mmol) for 24 h. In the analysis of the lipid extracts between VDR+/+ and VDR–/– keratinocytes after the incubation, a significant difference was immediately observed in total radioactivity recovered in the organic fraction (36.9% vs. 80.9%) and in the aqueous fraction (63.1% vs. 19.1%; Fig. 7). When the production of lipid- and water-soluble metabolites was monitored by HPLC using both radioactive and UV detection, VDR+/+ keratinocytes had metabolized so efficiently that total products reached amounts equaling the concentration of remaining substrate by 24 h of incubation (Fig. 7). In contrast, the lipid extracts of keratinocytes derived from VDR–/– mice contained only [1?-3H]1,25(OH)2D3, confirming the lack of detectable metabolism of [1?-3H]1,25(OH)2D3 to lipid-soluble intermediates. In VDR+/+ keratinocytes, lipid-soluble products were identified as [1?-3H] 1,24,25(OH)3D3, [1?-3H]24-oxo-1,25(OH)2D3, [1?-3H]24-oxo-1,23,25(OH)3D3, and [1?-3H]tetranor-1,23(OH)2D3 by comigration with standards (Fig. 7). Previous studies have shown that the production of such metabolites is CYP24A1-catalyzed via a VDR-mediated process (5, 6, 10). The dramatic reduction in the C-24 oxidation products was not replaced by alternative pathways of metabolism in VDR–/– keratinocytes. Taken together, these results confirm the importance of vitamin D-inducible, VDR-mediated C-24 oxidation in 1,25(OH)2D3 target cell catabolism.
FIG. 7. HPLC profile of metabolites of [1?-3H]1,25(OH)2D3 generated in primary keratinocyte culture of VDR-null and wild-type mice. Keratinocytes were incubated with 0.5 μCi [1?-3H] 1,25(OH)2D3 (specific activity, 125 mCi/mmol) for 24 h. Lipid-soluble metabolites were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Metabolites of [1?-3H]1,25(OH)2D3 were identified by flow radioactivity detection using synthetic standards. The radioactivity recovered in the aqueous fraction, measured by liquid scintillation counter, is shown in the inset.
Discussion
This study clearly establishes that one important physiological role of CYP24A1 is the regulation of the catabolism of 1,25(OH)2D3. In Cyp24a1-null mice, altered pharmacokinetics of both 1,25(OH)2D3 and 25(OH)D3 in blood and tissues were observed; these mice showed a greatly reduced ability to clear [1?-3H]1,25(OH)2D3 from the body and convert it into water-soluble metabolites. The difficulty of Cyp24a1–/– mice to eliminate [1?-3H]1,25(OH)2D3 from the bloodstream was observed over the full 96-h experimental period and resulted in a difference of as much as 20- to 30-fold higher [1?-3H]1,25(OH)2D3 levels compared with Cyp24a1+/– littermates (Fig. 1). The tissue content of [1?-3H]1,25(OH)2D3 also remained high for at least 96 h in Cyp24a1–/– mice (Fig. 1). In the kidneys of Cyp24a –/– mice, the increased accumulation of administered [1?-3H]1,25(OH)2D3 without its metabolites indicates the impaired ability of the animals to maintain 1,25(OH)2D3 homeostasis in a vitamin D target tissue. These results support previous preliminary data (19) suggesting that Cyp24a1–/– mutants are severely impaired in their ability to eliminate 1,25(OH)2D3 from the bloodstream.
Not only do Cyp24a1-null mice lack 24-hydroxylated metabolites, but they also appear to lack 1,25(OH)2D3-26,23-lactone despite the fact that the concentration of the [1?-3H]1,25(OH)2D3 hormone was always higher in Cyp24a1–/– mice than in Cyp24a1+/– mice, and thus substrate was presumably not rate-limiting. This result reinforces the view that CYP24A1 is also responsible for hydroxylation of 1,25(OH)2D3 at C-23 and C-26. In contrast, the fact that half of the homologous mutant animals survived into adulthood despite the inactivation of the Cyp24a1 in mice that caused lethality in their littermates suggests the existence of an alternative pathway of inactivation of 1,25(OH)2D3 or some other form of compensation, such as altered rate of production of 1,25(OH)2D3. No detectable [1?-3H]3-epi-1,25(OH)2D3 (27) was found in mice in vivo or in keratinocytes in vitro even though HPLC conditions would have resolved it from the natural hormone. This suggests that 3-epimerization of 1,25(OH)2D3 is not the alternative pathway of inactivation responsible for eventual clearance of [1?-3H]1,25(OH)2D3 in Cyp24a1–/– mice. In contrast, a fraction of the radioactivity partitions into the aqueous layer during excretion at all time points, especially in the liver in both genotypes, and although the exact amount is difficult to quantify because the aqueous fraction always contains some unchanged [1?-3H]1,25(OH)2D3, this aqueous phase radioactivity most likely represents some other uncharacterized, water-soluble product. This is suggestive of an alternative undefined route of degradation, possibly involving glucuronidation or another route of conjugation (28, 29, 30). Indeed, because the relative contribution of the water-soluble fraction increases with time in the Cyp24a1+/– animals but not in the Cyp24a1–/– animals, the results support the hypothesis that a good portion of the clearance is CYP24A1 dependent.
However, one alternative hypothesis is that surviving animals down-regulate the synthesis of 1,25(OH)2D3. Indeed, we have previously reported that Cyp24a1–/– mice that survive past weaning (i.e. 50% of the mutant population) show baseline circulating levels of 1,25(OH)2D3 that are 2.8-fold lower than those in age-matched heterozygous littermates (19). This obviously allows these Cyp24a1–/– mutant mice to survive without significant impairment of mineral ion homeostasis, at least in the absence of a challenge to normocalcemia (such as pregnancy, for example) (19). It is noteworthy that Cyp24a1 mutant mice showed significantly increased expression levels of vitamin D-dependent genes involves in VDR in the kidney at the basal point compared with heterozygote controls, and the expression levels in these mice did not respond to 1,25(OH)2D3 administration (19). Iida et al. (31) suggested that regulation of VDR expression would allow reciprocal control of CYP27B1 and CYP24A1 activity in renal proximal tubules. We postulated that it might be possible that up-regulated VDR in the kidney of Cyp24a1–/– mice acts on Cyp27b1 expression to suppress the production of 1,25(OH)2D3, resulting in lower circulating levels of 1,25(OH)2D3. The RT-PCR data presented here suggest that surviving Cyp24a1+/– animals show reduced expression of Cyp27b1 mRNA compared with their Cyp24a1+/– littermates at 4 months of age. Accordingly, these results confirm that pathways of 1,25(OH)2D3 elimination are through the inducible expression of CYP24A1 in vivo and possibly an as yet to be defined basal clearance pathway, but that Cyp24a1+/– animals also adapt to vitamin D intoxication by down-regulation of 1,25(OH)2D3 production.
After the administration of [26,27-methyl-3H]25(OH)D3, higher residual [26,27-methyl-3H]25(OH)D3 levels and no [26,27-methyl-3H]24,25(OH)2D3 formation were observed in blood and tissues (liver, kidney, and intestine) of Cyp24a1–/– mice over 48 h. In contrast to our observations with [1?-3H]1,25(OH)2D3, metabolic clearance of [26,27-methyl-3H]25(OH)D3 was slower in both Cyp24a1–/– and Cyp24a1+/– mice. Administered [26,27-methyl-3H]25(OH)D3 stayed mostly in the blood over the full 48-h time course, with limited uptake by tissues in either genotype. These data are consistent with the longer half-life reported for 25(OH)D3 and reflect the fact that the clearance of 25(OH)D3 is greatly affected by its sequestration by the plasma carrier protein, DBP, that is synthesized in the liver. Although there was a slightly higher level of tissue total radioactivity in the tissues of Cyp24a1–/– mice compared with Cyp24a1+/– mice, there was a marked increase in residual [26,27-methyl-3H]25(OH)D3 levels in the blood and tissues (liver, kidney, and intestine) of Cyp24a1–/– mice compared with Cyp24a1+/– littermates, reflecting the reduced rate of metabolism. Indeed, chromatography showed the complete absence of 24-hydroxylated metabolites in the blood and tissues of Cyp24a1–/– mice. These results indicate the critical importance of CYP24A1 in the catabolism of 25(OH)D3.
Another way to investigate the actions of CYP24A1 in the natural setting using Cyp24a1–/– mice is to perform in vitro studies to avoid the complexity of the in vivo setting. Keratinocytes were chosen as target cells to examine the physiological role of CYP24A1 in the catabolic cascade of vitamin D in vitro. When primary neonatal keratinocytes were prepared from Cyp24a1–/– and Cyp24a1+/– mice, the findings were essentially the same as those of the in vivo study, demonstrating in particular the lack of calcitroic acid production and 24,25(OH)2D3 formation in keratinocytes from Cyp24a1–/– mice. Previous studies demonstrated that skin is capable of activating vitamin D3 and 25(OH)D3 through sequential hydroxylations, including 1-hydroxylation, and the resulting 1,25(OH)2D3 plays a role in epidermal homeostasis in normal and diseased skin (32, 33, 34). Interestingly, in our hands these Cyp24a1–/– keratinocytes failed to produce detectable amounts of 1,25(OH)2D3 production even though they were cultured under low calcium conditions, and the masking action of CYP24A1 was absent. The lack of 1,25(OH)2D3 production was possibly due to the presence of FCS and its associated vitamin D metabolites in the culture medium. The fact that none of the metabolites, including C-23/C-26 oxidation products, was detected from the incubation with keratinocytes from Cyp24a1–/– mice leads to the conclusion that CYP24A1 plays multiple roles in 1,25(OH)2D3 and 25(OH)D3 catabolism. To date, our results suggest that CYP24A1 is the main enzyme required for the catabolism of 1,25(OH)2D3 and 25(OH)D3, although it might be prudent to not completely exclude alternative pathways until we have fully defined the nature of water-soluble biliary metabolites in the Cyp24a1–/– mice.
The results from the keratinocyte study of 1,25(OH)2D3 in VDR–/– mice indicate that Cyp24a1 expression is mediated through VDR. When primary keratinocytes were incubated with a physiological concentration of 1,25(OH)2D3, VDR+/+ mice demonstrated a series of metabolites of 1,25(OH)2D3 through the C-24/C-23 oxidation pathway. However, keratinocytes from VDR–/– mice showed a complete block of target cell metabolism of 1,25(OH)2D3 (as shown in Fig. 7). The difference illustrates the highly inducible nature of target cell Cyp24a1. Even when substrate was increased to pharmacological concentrations, little metabolism was observed in cells from VDR–/– mice compared with VDR+/+ mice (data not shown). One published in vivo study using VDR-null mice (35) demonstrated that the degradation of 1,25(OH)2D3 in VDR-null mice was substantially reduced compared with that in wild-type control mice. Interestingly, unlike the in vitro study, 1,25,26(OH)3D3 was found in plasma, kidney, and liver of VDR-null mice (35). This suggests that an enzyme that catalyzes the 26-hydroxylation of 1,25(OH)2D3, that would be independent of VDR stimulation, may exist in vivo. Recently, several liver cytochrome P450 isoforms, including CYP2R1, CYP2J3, CYP3A4, as well as CYP27A1, have been shown to possess vitamin D side-chain hydroxylation activity, and one wonders if this activity can be directed at the C26 position as well as the C25 position, as is currently claimed (1, 36, 37).
In conclusion, the Cyp24a1-null mouse is a valuable model, with potential to establish the pharmacological as well as the physiological role of CYP24A1. The availability of the Cyp24a1-null mouse also opens up new approaches to determine the degree of involvement of CYP24A1 in vitamin D analog metabolism.
References
Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1: a microsomal vitamin D 25-hydroxylase. J Biol Chem 278:38084–38093
Jones G, Strugnell SA, Deluca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231
Masuda S, Jones G 2003 Vitamin D analogs: drug design based on proteins involved in vitamin D signal transduction. Curr Drug Targets Immune Endocr Metab Disord 3:43–66
Omdahl JL, Morris HA, May BK 2002 Hydroxylase enzymes of the vitamin D pathway: expression, function, and regulation. Annu Rev Nutr 22:139–166
Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180
Lohnes D, Jones G 1987 Side chain metabolism of vitamin D3 in osteosarcoma cell line UMR-106. Characterization of products. J Biol Chem 262:14394–14401
Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762
Ohyama Y, Okuda K 1991 Isolation and characterization of a cytochrome P-450 from rat kidney mitochondria that catalyzes the 24-hydroxylation of 25-hydroxyvitamin D3. J Biol Chem 266:8690–8695
Chen KS, Prahl JM, DeLuca HF 1993 Isolation and expression of human 1,25-dihydroxyvitamin D3 24-hydroxylase cDNA. Proc Natl Acad Sci USA 90:4543–4547
Itoh S, Yoshimura T, Iemura O, Yamada E, Tsujikawa K, Kohama Y, Mimura T 1995 Molecular cloning of 25-hydroxyvitamin D3 24-hydroxylase (Cyp-24) from mouse kidney: its inducibility by vitamin D3. Biochim Biophys Acta 1264:26–28
Esvelt RP, Schnoes HK, Deluca HF 1979 Isolation and characterization of 1-hydroxy-23-carboxy-tetranorvitamin D: a major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983
Ishizuka S, Norman AW 1987 Metabolic pathways from 1,25-dihydroxyvitamin D3 to 1,25-dihydroxyvitamin D3-26,23-lactone. Stereo-retained and stereo-selective lactonization. J Biol Chem 262:7165–7170
Siu-Caldera ML, Zou L, Ehrlich MG, Schwartz ER, Ishizuka S, Reddy GS 1995 Human osteoblasts in culture metabolize both 1,25-dihydroxyvitamin D3 and its precursor 25-hydroxyvitamin D3 into their respective lactones. Endocrinology 136:4195–4203
Akiyoshi-Shibata M, Sakaki T, Ohyama Y, Noshiro M, Okuda K, Yabusaki Y 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. A study with the mature enzyme expressed in Escherichia coli. Eur J Biochem 224:335–343
Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant escherichia coli cells producing rat mitochondrial CYP24A1. CYP24A1 can convert 1,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48
Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472
Masuda S, Gao M, Zhang A, Kaufmann M, Jones G 2003 Importance of cytochrome P450-mediated metabolism in the mechanism of action of vitamin D analogs. Recent Results Cancer Res 164:189–202
Sakaki T, Sawada N, Komai K, Shiozawa S, Yamada S, Yamamoto K, Ohyama Y, Inouye K 2000 Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24A1. Eur J Biochem 267:6158–6165
St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxy-vitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666
Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D-responsive element in the 5'-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550
Chen KS, DeLuca HF 1995 Cloning of the human 1,25-dihydroxyvitamin D3-24-hydroxylase gene promoter and identification of two vitamin D-responsive elements. Biochim Biophys Acta 1263:1–9
Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835
Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama 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 16:391–396
Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141
Sakai Y, Demay MB 2000 Evaluation of keratinocyte proliferation and differentiation in vitamin D receptor knockout mice. Endocrinology 141:2043–2049
Bligh EG, Dyer WG 1957 A rapid method of total lipid extraction and purification. Can J Biochem 37:911–917
Siu-Caldera ML, Sekimoto H, Weiskopf A, Vouros P, Muralidharan KR, Okamura WH, Bishop J, Norman AW, Uskokovic MR, Schuster I, Reddy GS 1999 Production of 1,25-dihydroxy-3-epi-vitamin D3 in two rat osteosarcoma cell lines (UMR 106 and ROS 17/2.8): existence of the C-3 epimerization pathway in ROS 17/2.8 cells in which the C-24 oxidation pathway is not expressed. Bone 24:457–463
Londowski JM, Kost SB, Meier W, Labler L, Kumar R 1986 Biological activity of the C-1, C-3, C-25, ?-D-glucopyranosides of 1,25-dihydroxyvitamin D3. J Pharmacol Exp Ther 237:837–840
Ayton B, Martin F 1987 Conjugated forms of [3H]1, 25-dihydroxyvitamin in rat bile. J Steroid Biochem 26:667–677
Shimada K, Nakatani I, Saito K, Mitamura K 1996 Separation and characterization of mono-glucuronides of vitamin D3 and 25-hydroxyvitamin D3 in rat bile by high-performance liquid chromatography. Biol Pharm Bull 19:491–494
Iida K, Shinki T, Yamaguchi A, DeLuca HF, Kurokawa K, Suda T 1995 A possible role of vitamin D receptors in regulating vitamin D activation in the kidney. Proc Natl Acad Sci USA 92:6112–6116
Ray S, Ray R, Holick MF 1995 Metabolism of 3H-1,25-dihydroxyvitamin D3 in cultured human keratinocytes. J Cell Biochem 59:117–122
Lehmann B, Genehr T, Knuschke P, Pietzsch J, Meurer M 2001 UVB-induced conversion of 7-dehydrocholesterol to 1,25-dihydroxyvitamin D3 in an in vitro human skin equivalent model. J Invest Dermatol 117:1179–1185
Xie Z, Munson SJ, Huang N, Portale AA, Miller WL, Bikle DD 2002 The mechanism of 1,25-dihydroxyvitamin D3 autoregulation in keratinocytes. J Biol Chem 277:36987–36990
Endres B, Kato S, DeLuca HF 2000 Metabolism of 1,25-dihydroxyvitamin D3 in vitamin D receptor-ablated mice in vivo. Biochemistry 39:2123–2129
Yamasaki T, Izumi S, Ide H, Ohyama Y 2004 Identification of a novel rat microsomal vitamin D3 25-hydroxylase. J Biol Chem 279:22848–22856
Gupta RP, Hollis BW, Patel SB, Patrick KS, Bell NH 2004 CYP3A4 is a human microsomal vitamin D 25-hydroxylase. J Bone Miner Res 19:680–688
Jones G, Byford V, Sakai Y, Demay MB 2000 Dramatic reduction of 1,25-(OH)2D3 catabolism in keratinocytes from VDR-null mice. J Bone Miner Res 15(Suppl 1):S181
Jones G, Byford V, Arabian A, St-Arnaud R 2000 Altered pharmacokinetics of 1,25-(OH)2D3 in blood and tissues of the CYP24-null mouse. J Bone Miner Res 15(Suppl 1):S199(Sonoko Masuda, Valarie By)
Address all correspondence and requests for reprints to: Dr. Glenville Jones, Department of Biochemistry, Queen’s University, Kingston Ontario, Canada K7L 3N6. E-mail: gj1@post.queensu.ca.
Abstract
The 25-hydroxyvitamin D-24-hydroxylase (CYP24A1) plays an important role in regulating concentrations of both the precursor 25-hydroxyvitamin D3 [25(OH)D3] and the hormone 1,25-dihydroxyvitamin D3 [1,25(OH)2D3]. Previous studies suggest that Cyp24a1-null mice cannot clear exogenous 1,25(OH)2D3 efficiently. Here, we examined the metabolic clearance in Cyp24a1-null mice in vivo and in vitro using a physiological dose of [1?-3H]1,25(OH)2D3 or [26,27-methyl-3H]25(OH)D3. Cyp24a1-null mice showed difficulty in eliminating [1?-3H]1,25(OH)2D3 from the bloodstream and tissues over a 96-h time course, whereas heterozygotic mice eliminated the hormone within 6–12 h, although there was clearance of labeled hormone into water-soluble products involving liver in both genotypes. RT-PCR showed that Cyp24a1-null mice have decreased expression of 25-hydroxyvitamin D-1-hydroxylase that must play a role in their survival. After the administration of [26,27-methyl-3H]25(OH)D3, Cyp24a1-null mice showed higher [26,27-methyl-3H]25(OH)D3 levels and no [26,27-methyl-3H]24,25(OH)2D3 formation, whereas heterozygotic mice showed significant [26,27-methyl-3H]24,25(OH)2D3 production. Based upon in vitro experiments, keratinocytes from Cyp24a1-null mice fail to synthesize [1?-3H]calcitroic acid from [1?-3H]1,25(OH)2D3 or [26,27-methyl-3H]24,25(OH)2D3 from [26,27-methyl-3H]25(OH)D3 as do control mice, confirming the target cell catabolic role of CYP24A1 in these processes. Finally, the role of vitamin D receptor (VDR) in the vitamin D catabolic cascade was examined using VDR-null mice. Keratinocytes from VDR-null mice failed to metabolize [1?-3H]1,25(OH)2D3 confirming the importance of vitamin D-inducible, VDR-mediated, C24 oxidation pathway in target cells. These results suggest that the absence of CYP24A1 or VDR retards catabolism of 1,25(OH)2D3 and 25(OH)D3, reinforcing the physiological importance of CYP24A1 in vitamin D homeostasis.
Introduction
THE SEQUENCING OF the human genome has led to the realization that among a pool of 60 total cytochrome P450s (CYPs),1 there are three known and possibly other uncharacterized vitamin D-related CYPs dedicated to the metabolism of vitamin D. The key enzymes in vitamin D metabolism are the hepatic vitamin D-25-hydroxylase [CYP27A1 and possibly CYP2R1 (1)], renal 25-hydroxyvitamin D-1-hydroxylase (CYP27B1), and 25-hydroxyvitamin D-24-hydroxylase (CYP24A1). Biological activation of vitamin D occurs through the sequential hydroxylation catalyzed by CYP27A1 or CYP2R1 and CYP27B1, resulting in the formation of the active form of vitamin D3, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] (2, 3, 4). CYP24A1 is the enzyme acting on 1,25(OH)2D3 to initiate degradation of this potent metabolite and is tightly regulated. This enzyme is distributed in classical vitamin D target tissues, including kidney, intestine, bone, and skin. In target cells, 1,25(OH)2D3 has the ability to induce its own degradation by up-regulating CYP24A1 expression to attenuate 1,25(OH)2D3 action or prevent vitamin D toxicity (i.e. hypercalcemia) (5, 6, 7). CYP24A1 was discovered as the mitochondrial enzyme responsible for the hydroxylation at C-24 in the metabolism of 1,25(OH)2D3 and 25-hydroxyvitamin D3 [25(OH)D3] (8, 9, 10). C-24 oxidation represents the predominant catabolic pathway for 1,25(OH)2D3 in which sequential C-24 hydroxylation, C-24 ketonization, and C-23 hydroxylation, followed by oxidative cleavage to form tetranor-1,23(OH)2D3 and calcitroic acid, liberate an inactive, water-soluble product that is excreted in bile (5, 11). The secondary pathway involves C-26 hydroxylation, C-23 hydroxylation, and lactonization to form (23S,25R)-1,25(OH)2D3-26,23-lactone (12, 13). Both pathways have also been shown to participate in the metabolism of 25(OH)D3 (13). Studies of the recombinant CYP24A1 protein produced in Escherichia coli (14, 15), insect (16), and mammalian (17) cells have revealed that CYP24A1 exerts multicatalytic activity, facilitating sequential oxidation of C-23, C-24, and C-26 hydroxylation and side-chain cleavage.
The role of CYP24A1 has been established to be inactivation of both 1,25(OH)2D3 and 25(OH)D3 (5, 6, 13). Most evidence from kinetic analysis supports the concept that 1,25(OH)2D3, rather than 25(OH)D3, is the preferred substrate for CYP24A1 (18) and that this enzyme plays an important role in regulating the intracellular concentration as well as the half-life of 1,25(OH)2D3 in the body. Recently, this was also confirmed by the generation of a Cyp24a1 knockout mouse model (19). Initial reports suggest that Cyp24a1-null mice show prolonged elevations in plasma 1,25(OH)2D3 levels after exogenous 1,25(OH)2D3 administration. These findings are consistent with the poor viability of the Cyp24a1-null mice and its tendency to hypercalcemia and nephrocalcinosis during the neonatal period. Interestingly, Cyp24a1 mutant animals that survived past weaning showed lower baseline levels of circulating 1,25(OH)2D3 than wild-type controls. This fact raises questions about the physiological role of CYP24A1 in the vitamin D metabolism. It has been postulated that these animals may use an alternative pathway of 1,25(OH)2D3 catabolism to regulate circulating levels of the hormone.
The action of 1,25(OH)2D3 in target cells is mediated by the vitamin D receptor (VDR), which binds to vitamin D response elements in the promoter of genes. Two vitamin D response elements in the CYP24A1 promoter are responsible for the VDR-mediated up-regulation of CYP24A1 gene transcription (20, 21). In mice lacking the VDR gene, highly elevated plasma levels of 1,25(OH)2D3 were measured due to the impaired ligand-dependent activation of Cyp24a1 gene transcription and inadequate catabolic degradation of 1,25(OH)2D3 (22, 23). Hypocalcemia and a loss of VDR-mediated gene expression caused Cyp24a1 expression to be low in such animals. Thus, because of the inducible nature of vitamin D-target cell CYP24A1, the VDR-null mouse represents the equivalent of a partial Cyp24a1 knockout mouse model. Accordingly, a comparison of CYP24A1 enzyme activity in VDR-null mice and Cyp24a1-null mice would allow us to understand the roles of both VDR and CYP24A1 in vitamin D signaling and catabolic cascade.
The development of novel gene expression systems for CYPs has allowed examination of some long-standing questions in vitamin D metabolism. In the present study we focused on 1) elucidation of the overall importance of CYP24A1 in 1,25(OH)2D3 and 25(OH)D3 clearance, 2) direct comparison of the metabolism of 1,25(OH)2D3 and 25(OH)D3 in the Cyp24a1-null mouse in vivo and in vitro, and 3) elucidation of the role of VDR in 1,25(OH)2D3 metabolism using keratinocytes from the VDR-null mouse in vitro. The results confirm that ablation of either the Cyp24a1 or VDR gene profoundly retards metabolism of 1,25(OH)2D3 and 25(OH)D3 and that the surviving Cyp24a1-null mice owe their survival in part to down-regulation of Cyp27b1 expression.
Materials and Methods
Materials
1,25(OH)2D3, 25(OH)D3, 1,24,25(OH)3D3, tetranor-1,23(OH)2D3, 24,25(OH)2D3, and 1(OH)D3 were gifts from Drs. M. Calverley and A. M. Kissmeyer (Leo Pharmaceutical Products, Ballerup, Denmark). (23S,25R)-1,25(OH)2D3-26,23-lactone was a gift from Dr. S. Ishizuka (Teijin, Japan). [26,27-Methyl-3H]25(OH)D3 (30 Ci/mmol) was purchased from Amersham Biosciences (Little Chalfont, UK). [1?-3H]1,25(OH)2D3 (50 Ci/mmol) and calcitroic acid were synthesized by methods described previously (5). MEM, trypsin, and antibiotic/antimyotic were purchased from Invitrogen Life Technologies (Burlington, Canada). Fetal calf serum (FCS) was purchased from ICN (Costa Mesa, CA), and BSA was obtained from Roche (Laval, Canada). Chelex 100 was purchased from Bio-Rad Laboratories (Hercules, CA). Collagen (Vitrogen 100) was purchased from Cohesion Technologies, Inc. (Palo Alto, CA). Epidermal growth factor (EGF) and N'N-diphenyl-p-phenylenediamine (DPPD) were supplied by Sigma-Aldrich Corp. (St. Louis, MO). Liquid scintillation cocktail (Ready Flow III) for on-line radioactivity detection was purchased from Beckman Coulter, Inc. (Fullerton, CA). All organic solvents were of HPLC grade and were purchased from Caledon Laboratories Ltd. (Georgetown, Canada).
Animals
All studies performed were reviewed and approved by the institutional animal care and use committees. Animals were maintained in a virus- and parasite-free barrier facility and exposed to a 12-h light, 12-h dark cycle. Mice lacking Cyp24a1 were obtained as previously described (19). Genotyping for the Cyp24a1 mutation was performed by Southern blotting of BamHI digests using an EcoRI-PstI genomic probe downstream of the region of homology. The 12-kb wild-type allele and the 8.5-kb targeted allele were identified. Mice lacking VDR were obtained as previously described (22). VDR-null mice were genotyped by PCR analysis of tail DNA using primers specific for the neomycin resistance gene and the second zinc finger of the VDR.
In vivo metabolism study in Cyp24a1-null mice
For the 1,25(OH)2D3 metabolism study, 6-month-old Cyp24a–/– homozygotes and their Cyp24a+/– littermates received a single iv injection of [1?-3H]1,25(OH)2D3 (1 μCi; 8 ng) in 100 μl of a vehicle containing ethanol/mouse serum (1:9). At 3, 6, 24, 48, and 96 h postinjection, blood was taken by cardiac puncture, and liver and kidneys were collected. For the 25(OH)D3 metabolism study, 6-month-old Cyp24a1–/– homozygotes and their Cyp24a+/– littermates received a single iv injection of [26,27-methyl-3H]25(OH)D3 (5 μCi; 67 ng) in 100 μl of a vehicle containing ethanol/mouse serum (1:9). At 6, 24, and 48 h postinjection, blood was taken by cardiac puncture and liver, kidneys and intestine were collected. Plasma was obtained by centrifugation at 3000 rpm for 15 min. Liver, kidneys, and intestine were frozen in liquid N2 and stored at –20 C.
Isolation and preparation of primary keratinocytes
Primary keratinocytes were isolated from 2- to 3-d-old Cyp24a1-null mice and control littermates by a trypsin floating procedure as described previously (24). Briefly, the skin was dissected from the trunk of mice and floated on 0.25% trypsin for 16–20 h at 4 C. The epidermis was then peeled off the dermis, minced with scissors, and stirred in MEM with 4% (vol/vol) Chelex-treated FCS, EGF (10 ng/ml), 1% (vol/vol) antibiotic/antimycotic, and 0.05 mM CaCl2 (low calcium medium) for 1 h at 4 C. The resultant cell suspension was then filtered through a 70-μm pore size filter (Falcon, BD Biosciences, Franklin Lakes, NJ) to remove clumps of cells and debris. The keratinocytes were pelleted by centrifugation for 5 min at room temperature and plated in low calcium medium in collagen-coated, 60-mm dishes. Cells were maintained at 37 C in a humidified atmosphere of 5% CO2 in air. At 80–90% confluence, cells were reseeded at 1.8 x 105 cells/well in six-well plates in low calcium medium and grown to 80% confluence before addition of substrate.
In vitro metabolism study of 1,25(OH)2D3 and 25(OH)D3 in Cyp24a1-null mouse keratinocytes
Primary keratinocytes obtained from each mouse were treated with 10 nM 1,25(OH)2D3 (in ethanol) to induce the transcription of Cyp24a1. After 18 h, the cells were rinsed twice with PBS and incubated with 0.25 μCi [1?-3H]1,25(OH)2D3 (2.1 ng; 2.5 nM) in 2 ml low calcium MEM supplemented with 1% (wt/vol) BSA, 10 ng/ml EGF, 1% (vol/vol) antibiotic/antimycotic, and 100 μM DPPD for 24 h at 37 C. For the time-course study of 25(OH)D3, mouse primary keratinocytes that were not previously induced with 10 nM 1,25(OH)2D3 were incubated with 0.5 μCi [26,27-methyl-3H]25(OH)D3 (6.6 ng; 16.6 nM) in 2 ml low calcium MEM supplemented with 1% (wt/vol) BSA, 10 ng/ml EGF, 1% (vol/vol) antibiotic/antimycotic, and 100 μM DPPD for 0, 1, 3, 6, and 12 h at 37 C.
In vitro metabolism study in VDR-null mice keratinocytes
Mouse primary keratinocytes from VDR-null and wild-type mice were plated at 2 x 106 cells/plate in P-100 plates and grown to 80% confluence in low calcium MEM with 4% (vol/vol) Chelex-treated FCS, 10 ng/ml EGF, and 1% (vol/vol) antibiotic/antimycotic. Cells were incubated with 0.5 μCi [1?-3H]1,25(OH)2D3 (4.16 ng; 2.5 nM) including 1 μM 1,25(OH)2D3 or 10 μM 1,25(OH)2D3 in 4 ml low calcium MEM supplemented with 1% (wt/vol) BSA, 10 ng/ml EGF, 1% (vol/vol) antibiotic/antimycotic, and 100 μM DPPD for 24 h at 37 C.
RT-PCR analysis of CYP27B1 transcripts
The RT-PCR protocol used total RNA prepared with the TRIzol reagent (Invitrogen Life Technologies, Inc., Gaithersburg, MD) from the kidneys of 4-month-old Cyp24a1+/– and Cyp24a1–/– animals (25). The RT reaction contained oligo(deoxythymidine) and SuperScript II reverse transcriptase (Invitrogen Life Technologies, Inc.). PCR amplification used the following primers for Cyp27b1: 5'-CTGCGAGGAGGGGTAAGGTGTT-3' and 5'-GGAAACGGGGGAGGGGA-3'. Amplification was for 28 cycles. Control ?-actin amplification reactions used the primer pair 5'-GCTGCGTGTGGCCCCTAGG-3' and 5'-CAAGAAGGAAGGCTGGAAAAGAG-3'. Amplimers were detected by ethidium bromide staining of agarose electrophoresis gels.
Lipid extraction
Before the extraction, plasma was adjusted up to a volume of 1 ml with ice-cold buffered 0.9% saline, and tissues were homogenated in ice-cold buffered 0.9% saline (4 ml/g tissue). An internal standard of 2 μg 1(OH)D3 was added to blood and tissue homogenate samples to assess recovery during the extraction process. For the cell culture, incubations were terminated at designated time points with methanol. All samples were extracted using a modification of the method of Bligh and Dyer (26), in which chloroform was replaced by methylene chloride. The aqueous layer was reextracted with methylene chloride and 0.1% glacial acetic acid to isolate the water-soluble metabolites. After the extraction, aliquots of organic and aqueous fractions were subjected to liquid scintillation counter (LC1800, Beckman Coulter, Inc., Fullerton, CA). All samples were dried down under a stream of prepurified N2 at 37 C, resuspended, and stored in ethanol.
HPLC
Metabolites from in vivo or in vitro study were analyzed by either straight phase (organic-soluble metabolites) or reverse phase (water-soluble metabolites) HPLC using an Alliance 2695 Separations Module and a model 996 photodiode array detector (Waters Associates, Milford, MA). Radiolabeled organic- or water-soluble metabolites were detected using a RadioFlow Detector (LB 509, EG&G Berthold, Bad Wildbad, Germany). The organic phase extracts were dried down, redissolved, and separated on a Zorbax SIL column (3 μm; 62 x 80 mm; Agilent Technologies, Palo Alto, CA) with a hexane/isopropanol/methanol (HIM; 91%:7%:2%, vol/vol/vol) isocratic system at a flow rate of 1.0 ml/min. The water-soluble metabolites were separated on a Zorbax SB-C18 column (5 μm; 4.6 x 150 mm; Agilent Technologies) with a linear gradient solvent system, which started at 69%:30%:1% (vol/vol/vol) water/acetonitrile/10% acetic acid and was brought to 0%/99%/1% (vol/vol/vol) water/acetonitrile/10% acetic acid by 30 min at a flow rate of 1.0 ml/min. Metabolites were identified based on the characteristic vitamin D cis-triene chromophore (max = 265 nm; min = 228 nm; max/min = 1.75). Data analysis was performed with the aid of Millennium32 chromatographic software (Waters Associates, Milford, MA).
Statistical analysis
The significance of the difference between the means of the two groups was analyzed by t test, and differences were taken as statistically significant at P < 0.001 (***), P < 0.01 (**), or P < 0.05 (*).
Results
Metabolism of 1,25(OH)2D3 in Cyp24a1-null mice in vivo
Initial studies focused on the metabolic clearance and blood and tissue levels of [1?-3H]1,25(OH)2D3 and its circulating metabolites over a 96-h period in Cyp24a1–/– homozygotic mice and their Cyp24a1+/– littermates after a single iv injection of 1 μCi [1?-3H]1,25(OH)2D3. The results shown in Fig. 1 indicate that the marked differences in lipid-soluble radioactivity between Cyp24a1+/– control and Cyp24a1–/– mice were evident by as early as 6 h. Even at 6 h, organic fractions from blood, liver, and kidney tissues from Cyp24a1–/– mice showed approximately 2-fold higher [1?-3H]1,25(OH)2D3 levels than Cyp24a1+/– controls. By 96 h this difference became 20-fold for blood, 7-fold for liver, and 50-fold for kidney. [1?-3H]1,25(OH)2D3 was still identifiable in blood, liver, and kidney of Cyp24a1–/– mice at 96 h. The half-life (t1/2) of [1?-3H]1,25(OH)2D3 in Cyp24a1–/– mice was significantly longer than that in Cyp24a1+/– controls in blood (t1/2, 39 h vs. 4 h), liver (t1/2, 16 h vs. 5 h), and kidney (t1/2, 12 h vs. 5 h). The lipid-soluble 3H radioactivity remained mainly in the form of 1,25(OH)2D3, except in blood and liver of Cyp24a1+/– mice at early time points, where [1?-3H]1,25(OH)2D3-26,23-lactone was also detectable. [1?-3H]1,25(OH)2D3-26,23-lactone was completely absent from the blood of Cyp24a1–/– mice despite the fact that a measurable peak of [1?-3H]1,25(OH)2D3 was always greater in chromatograms of Cyp24a1–/– mice than in Cyp24a1+/– mice. As a result of these initial studies, subsequent experiments performed by our group emphasized the need to explore the water-soluble metabolites in more detail.
FIG. 1. Levels of [1?-3H]1,25(OH)2D3 in Cyp24a1–/–and Cyp24a1+/– mice after a single bolus dose. [1?-3H]1,25(OH)2D3 levels in blood, liver, and kidney in Cyp24a1-null mice were compared with heterozygotic littermates at 6, 24, 48, and 96 h after a single iv administration of 1 μCi [1?-3H]1,25(OH)2D3. Lipids from the extraction were separated on Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Radioactivity measured using radioisotope detector and identified as [1?-3H]1,25(OH)2D3 from its retention time is plotted. Each point represents the mean ± SE of three animals. Differences were statistically significant as indicated: *, P < 0.05; **, P < 0.01; ***, P < 0.001 (compared with Cyp24a1+/– mice by t test).
A second experiment, shown in Table 1, represents a time course over a 3- to 48-h period, where we observed a higher level of total radioactivity in the blood of Cyp24a1–/– animals by 24 h, a difference that became statistically significant by 48 h. The rapid clearance of [1?-3H]1,25(OH)2D3 from the blood of Cyp24a1+/– was also accompanied by the conversion of the lipid-soluble hormone into water-soluble products, such that over the 3- to 48-h period the percentage of recovered radioactivity found in the water-soluble fraction of blood rose dramatically from 45% to 74%. In contrast, the slower clearance of [1?-3H]1,25(OH)2D3 from the blood of the Cyp24a1–/– mice was accompanied by a modest increase in the same parameter (from 36% to 43% over 3–48 h), with the organic phase accounting for between 64% and 57% of the total radioactivity. In both genotypes, there was a rapid uptake of [1?-3H]1,25(OH)2D3 or its metabolites into liver and kidney by as early as 3 h. In the liver, not only were there more water-soluble products in both genotypes, but also the differences in total radioactivity were not as marked as observed in blood and kidney. Again in the liver of Cyp24a1+/– mice, there was a significant rise in the percentage of water-soluble radioactivity, increasing from 60% to 72% over the 3- to 48-h time period, a change that was absent in Cyp24a1–/– mice (58%, 53%, 64%, and 58% in the same 3- to 48-h period). In the kidney, arguably representing a typical vitamin D target tissue, the picture was even more pronounced, with significant differences in total radioactivity apparent by 6 h and a markedly lower percentage of the radioactivity residing in the organic fraction as unchanged [1?-3H]1,25(OH)2D3 in Cyp24a1+/– mice compared with Cyp24a1–/– mice. Indeed, in the Cyp24a1+/– mice, the water-soluble metabolites rose from 57% to 78% over the 3- to 48-h period, whereas in the Cyp24a1–/–– mice this parameter fairly stable at around 33%, or even fell.
TABLE 1. Time course of clearance of 1,25(OH)D2D3 in Cyp24a1–/– and Cyp24a1+/– mice after the administration of 1 μCi [1?-3H]1,25(OH)2D3
RT-PCR analysis of CYP27B1 transcripts in Cyp24a1-null mice
A previous report (19) found that the basal circulating levels of 1,25(OH)2D3 were approximately 3-fold lower in Cyp24a1–/– mice that survived past weaning than in wild-type controls. To examine whether surviving Cyp24a1–/– mice are capable of down-regulation of 1,25(OH)2D3 production, the expression of Cyp27b1 in Cyp24a1–/– mouse kidney was investigated. RT-PCR analysis using total RNA prepared from kidney tissue of 4-month-old Cyp24a1–/– and Cyp24a1+/– mice revealed detectable transcripts of Cyp27b1 mRNA in Cyp24a1+/–, but not Cyp24a1–/– mice (Fig. 2) despite the fact that there was equal expression of ?-actin mRNA. These observations suggest that the surviving Cyp24a1–/– mice may owe their survival in part to lower 1,25(OH)2D3 biosynthesis in the kidney.
FIG. 2. Cyp27b1 [25-hydroxyvitamin D-1-hydroxylase (1-OHase)] expression in the kidney from Cyp24a1–/– and Cyp24a1+/– mice. The expression of Cyp27b1 was analyzed by RT-PCR using total RNA prepared from kidney tissue of 4-month-old Cyp24a1–/– and Cyp24a1+/– mice. The conditions for the RT-PCR were described in Materials and Methods. The 451-bp Cyp27b1 amplimer and the 522-bp ?-actin amplimer are shown. Three animals were probed for each genotype; genotypes are identified below the figure. Note the dramatic reduction in Cyp27b1 expression in Cyp24a1–/– mice.
Metabolism of 1a,25(OH)2D3 in Cyp24a1-null mice keratinocytes in vitro
We next investigated the in vitro metabolism of [1?-3H]1,25(OH)2D3 using primary neonatal keratinocytes from Cyp24a1–/– and Cyp24a1+/– animals. When 0.25 μCi [1?-3H]1,25(OH)2D3 was incubated for 24 h after the induction of Cyp24a1 transcription, keratinocytes from Cyp24a1+/– mice demonstrated the complete breakdown of [1?-3H]1,25(OH)2D3 and the only metabolite identified in the organic fraction was [1?-3H]tetranor-1,23(OH)2D3 at a retention time of 17.73 min (Fig. 3A), whereas keratinocytes from Cyp24a1–/– mice failed to degrade substrate. The major metabolite observed was water soluble and ran as a discrete peak with a retention time of 11.13 min identical to that of the standard for calcitroic acid (Fig. 3B). In contrast, Cyp24a1–/– keratinocytes showed no metabolites generated through the C-24 oxidation pathway, and most of the radioactivity remained as [1?-3H]1,25(OH)2D3 in the organic fraction (Fig. 3A). Interestingly, despite the existence of high levels of [1?-3H]1,25(OH)2D3 in Cyp24a1–/– keratinocytes, neither C-23 nor C-26 oxidation was detected during the incubation. Furthermore, there was no evidence of alternative target cell pathways in Cyp24a1–/– keratinocytes that may substitute for C-24 oxidation and contribute to the catabolism of [1?-3H] 1,25(OH)2D3.
FIG. 3. HPLC profile of metabolites of [1?-3H]1,25(OH)2D3 generated by incubation of primary keratinocyte from Cyp24a1–/– and Cyp24a1+/– mice with 0.25 μCi [1?-3H] 1,25(OH)2D3 for 24 h after induction with 10 nM 1,25(OH)2D3. Lipid-soluble metabolites (A) were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) using a flow rate of 1.0 ml/min. Water-soluble metabolites (B) were separated on a Zorbax SCB18 (5 μm; 4.6 x 150 mm) column using a linear water-acetonitrile gradient system from 30–100% acetonitrile with 0.1% acetic acid over 30 min at a flow rate of 1.0 ml/min. Metabolites of [1?-3H]1,25(OH)2D3 were identified by flow radioactivity detection using synthetic standards. The experiment was performed using five to six animals from each group, and a representative chromatogram is shown.
Metabolism of 25(OH)D3 in Cyp24a1-null mice in vivo
To compare the metabolic clearance of 25(OH)D3 in Cyp24a1–/– and Cyp24a1+/– mice, we administered a 5-μCi dose of [26,27-methyl-3H]25(OH)D3 and followed the disappearance of [26,27-methyl-3H]25(OH)D3 and the production of C-24 oxidation products in blood and tissues. In comparison with the metabolic clearance measured for [1?-3H]1,25(OH)2D3, [26,27-methyl-3H]25(OH)D3 was clearly more slowly metabolized in both Cyp24a1–/– and Cyp24a1+/– mice. The obvious difference was observed in blood, where the pool of vitamin D-binding protein (DBP) presumably sequesters 25(OH)D3 (Table 2). Whereas the total radioactivity in the blood fell slightly from 30% to 20% of the dose in Cyp24a1+/– mice over the 48-h time course, this parameter dropped more slowly in the Cyp24a1–/– mice (39% at 48 h). This trend toward reduced rates of clearance in Cyp24a1–/– mice was more exaggerated when we focused on the amount of unchanged residual [26,27-methyl-3H]25(OH)D3 surviving in the blood and tissues. Figure 4 presents unchanged [26,27-methyl-3H]25(OH)D3 in the lipid-soluble fraction of blood and tissues in Cyp24a1+/– and Cyp24a1–/– mice after a bolus dose. By 48 h, differences in unchanged [26,27-methyl-3H]25(OH)D3 ratios were apparent in Cyp24a1+/– and Cyp24a1–/– mice (2.4-, 1.9-, 3.5- and 2.9-fold for blood, liver, kidney, and intestine, respectively).
TABLE 2. Time course of clearance and metabolism of 25(OH)D3 in Cyp24a1–/– and Cyp24a1+/– mice after the administration of 5 μCi [26,27-methyl-3H]25(OH)D3
FIG. 4. Levels of [26,27-methyl-3H]25(OH)D3 in Cyp24a1–/– and Cyp241a+/– mice after a single bolus dose. [26,27-Methyl-3H]25(OH)D3 levels in blood, liver, kidney, and intestine from Cyp24a1-null mice were compared with their heterozygotic littermates at 6, 24, and 48 h after a single iv administration with 5 μCi [26,27-methyl-3H]25(OH)D3. Lipids from the extraction were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Radioactivity measured using flow radioisotope detector and identified as [26,27-methyl-3H]25(OH)D3 from its retention time is plotted. Each data point is the mean ± SE of two animals. Differences were statistically significant as indicated: *, P < 0.05; **, P < 0.005 (compared with Cyp24a1+/– mice by the t test).
Although the lipid-soluble 3H remained mainly as [26,27-methyl-3H]25(OH)D3, Cyp24a1+/– mice showed the production of [26,27-methyl-3H]24,25(OH)2D3 and its subsequent metabolic product, [26,27-methyl-3H]24-oxo-25(OH)2D3, from 6 h postinjection onward, and the two metabolites were still accumulating 48 h after the injection in blood, liver, kidney, and intestine (Fig. 5 and Table 2). Unlike Cyp24a1+/– littermates, a total absence of [26,27-methyl-3H]24,25(OH)2D3 was observed in Cyp24a1–/– mice. Additional metabolites, such as [26,27-methyl-3H]1,25(OH)2D3 or [26,27-methyl-3H]1,24,25(OH)3D3, were not detected in either Cyp24a1+/– or Cyp24a1–/– mice at any time point in the blood and all tissues tested.
FIG. 5. HPLC profiles of blood and tissue samples extracted from Cyp24a1–/– and Cyp24a1+/– mice at 48 h after iv administration of [26,27-methyl-3H]25(OH)D3. Lipid extracts from blood, liver, kidney, and intestine (4.5 x 105 dpm of each) were chromatographed on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Metabolites of [26,27-methyl-3H]25(OH)D3 were identified by retention time with synthetic standards after detection using a flow radioisotope detector. The experiment was performed using two animals, and a typical chromatogram from each group is depicted.
Metabolism of 25(OH)D3 in Cyp24a1-null mice keratinocytes in vitro
For investigation of the kinetic metabolic profile of 25(OH)D3 in Cyp24a1-null mouse keratinocytes, cells were incubated with 0.5 μCi [26,27-methyl-3H]25(OH)D3 for various times. Despite the fact that the keratinocytes from Cyp24a1+/– mice were cultured with [26,27-methyl-3H]25(OH)D3 under DBP-free medium for up to 12 h, 80% of the radioactivity remained in the organic fraction. The only metabolite observed in the organic fraction of Cyp24a1+/– cells was [26,27-methyl-3H]24,25(OH)2D3, a metabolite that appeared as early as 3 h after the incubation and increased in the extracts up to 12 h (Fig. 6). Extracts from Cyp24a1–/– keratinocytes showed higher residual [26,27-methyl-3H]25(OH)D3 levels and no [26,27-methyl-3H] 24,25(OH)2D3 production over the full 12-h time course (Fig. 6). No peak corresponding to [26,27-methyl-3H]1,25(OH)2D3 or [26,27-methyl-3H]1,24,25(OH)3D3 was found at any time point in any keratinocyte culture tested.
FIG. 6. HPLC profiles of metabolites of [26,27-methyl-3H]25(OH)D3 generated in primary keratinocyte culture of Cyp24a1–/– and Cyp24a1+/– mice. Keratinocytes were incubated with 0.5 μCi [26,27-methyl-3H]25(OH)D3 for 0, 1, 3, 6, and 12 h. Lipid-soluble metabolites were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Metabolites of [26,27-methyl-3H]25(OH)D3 were identified by flow radioactivity detection using synthetic standards. The experiment was performed using three to seven animals in each group, and representative chromatograms are shown for 0, 6, and 12 h.
Metabolism of 1,25(OH)2D3 in VDR-null mice keratinocytes in vitro
Because of the inducible nature of vitamin D target cell CYP24A1, VDR-null mice represent a partial Cyp24a1 knockout model. To monitor CYP24A1 activity in vitamin D target cells by studying [1?-3H]1,25(OH)2D3 catabolism, primary keratinocytes from VDR–/– and VDR+/+ mice were isolated and incubated with [1?-3H]1,25(OH)2D3 (specific activity, 125 mCi/mmol) for 24 h. In the analysis of the lipid extracts between VDR+/+ and VDR–/– keratinocytes after the incubation, a significant difference was immediately observed in total radioactivity recovered in the organic fraction (36.9% vs. 80.9%) and in the aqueous fraction (63.1% vs. 19.1%; Fig. 7). When the production of lipid- and water-soluble metabolites was monitored by HPLC using both radioactive and UV detection, VDR+/+ keratinocytes had metabolized so efficiently that total products reached amounts equaling the concentration of remaining substrate by 24 h of incubation (Fig. 7). In contrast, the lipid extracts of keratinocytes derived from VDR–/– mice contained only [1?-3H]1,25(OH)2D3, confirming the lack of detectable metabolism of [1?-3H]1,25(OH)2D3 to lipid-soluble intermediates. In VDR+/+ keratinocytes, lipid-soluble products were identified as [1?-3H] 1,24,25(OH)3D3, [1?-3H]24-oxo-1,25(OH)2D3, [1?-3H]24-oxo-1,23,25(OH)3D3, and [1?-3H]tetranor-1,23(OH)2D3 by comigration with standards (Fig. 7). Previous studies have shown that the production of such metabolites is CYP24A1-catalyzed via a VDR-mediated process (5, 6, 10). The dramatic reduction in the C-24 oxidation products was not replaced by alternative pathways of metabolism in VDR–/– keratinocytes. Taken together, these results confirm the importance of vitamin D-inducible, VDR-mediated C-24 oxidation in 1,25(OH)2D3 target cell catabolism.
FIG. 7. HPLC profile of metabolites of [1?-3H]1,25(OH)2D3 generated in primary keratinocyte culture of VDR-null and wild-type mice. Keratinocytes were incubated with 0.5 μCi [1?-3H] 1,25(OH)2D3 (specific activity, 125 mCi/mmol) for 24 h. Lipid-soluble metabolites were separated on a Zorbax SIL (3 μm; 62 x 80 mm) column using the solvent HIM (91:7:2) at a flow rate of 1.0 ml/min. Metabolites of [1?-3H]1,25(OH)2D3 were identified by flow radioactivity detection using synthetic standards. The radioactivity recovered in the aqueous fraction, measured by liquid scintillation counter, is shown in the inset.
Discussion
This study clearly establishes that one important physiological role of CYP24A1 is the regulation of the catabolism of 1,25(OH)2D3. In Cyp24a1-null mice, altered pharmacokinetics of both 1,25(OH)2D3 and 25(OH)D3 in blood and tissues were observed; these mice showed a greatly reduced ability to clear [1?-3H]1,25(OH)2D3 from the body and convert it into water-soluble metabolites. The difficulty of Cyp24a1–/– mice to eliminate [1?-3H]1,25(OH)2D3 from the bloodstream was observed over the full 96-h experimental period and resulted in a difference of as much as 20- to 30-fold higher [1?-3H]1,25(OH)2D3 levels compared with Cyp24a1+/– littermates (Fig. 1). The tissue content of [1?-3H]1,25(OH)2D3 also remained high for at least 96 h in Cyp24a1–/– mice (Fig. 1). In the kidneys of Cyp24a –/– mice, the increased accumulation of administered [1?-3H]1,25(OH)2D3 without its metabolites indicates the impaired ability of the animals to maintain 1,25(OH)2D3 homeostasis in a vitamin D target tissue. These results support previous preliminary data (19) suggesting that Cyp24a1–/– mutants are severely impaired in their ability to eliminate 1,25(OH)2D3 from the bloodstream.
Not only do Cyp24a1-null mice lack 24-hydroxylated metabolites, but they also appear to lack 1,25(OH)2D3-26,23-lactone despite the fact that the concentration of the [1?-3H]1,25(OH)2D3 hormone was always higher in Cyp24a1–/– mice than in Cyp24a1+/– mice, and thus substrate was presumably not rate-limiting. This result reinforces the view that CYP24A1 is also responsible for hydroxylation of 1,25(OH)2D3 at C-23 and C-26. In contrast, the fact that half of the homologous mutant animals survived into adulthood despite the inactivation of the Cyp24a1 in mice that caused lethality in their littermates suggests the existence of an alternative pathway of inactivation of 1,25(OH)2D3 or some other form of compensation, such as altered rate of production of 1,25(OH)2D3. No detectable [1?-3H]3-epi-1,25(OH)2D3 (27) was found in mice in vivo or in keratinocytes in vitro even though HPLC conditions would have resolved it from the natural hormone. This suggests that 3-epimerization of 1,25(OH)2D3 is not the alternative pathway of inactivation responsible for eventual clearance of [1?-3H]1,25(OH)2D3 in Cyp24a1–/– mice. In contrast, a fraction of the radioactivity partitions into the aqueous layer during excretion at all time points, especially in the liver in both genotypes, and although the exact amount is difficult to quantify because the aqueous fraction always contains some unchanged [1?-3H]1,25(OH)2D3, this aqueous phase radioactivity most likely represents some other uncharacterized, water-soluble product. This is suggestive of an alternative undefined route of degradation, possibly involving glucuronidation or another route of conjugation (28, 29, 30). Indeed, because the relative contribution of the water-soluble fraction increases with time in the Cyp24a1+/– animals but not in the Cyp24a1–/– animals, the results support the hypothesis that a good portion of the clearance is CYP24A1 dependent.
However, one alternative hypothesis is that surviving animals down-regulate the synthesis of 1,25(OH)2D3. Indeed, we have previously reported that Cyp24a1–/– mice that survive past weaning (i.e. 50% of the mutant population) show baseline circulating levels of 1,25(OH)2D3 that are 2.8-fold lower than those in age-matched heterozygous littermates (19). This obviously allows these Cyp24a1–/– mutant mice to survive without significant impairment of mineral ion homeostasis, at least in the absence of a challenge to normocalcemia (such as pregnancy, for example) (19). It is noteworthy that Cyp24a1 mutant mice showed significantly increased expression levels of vitamin D-dependent genes involves in VDR in the kidney at the basal point compared with heterozygote controls, and the expression levels in these mice did not respond to 1,25(OH)2D3 administration (19). Iida et al. (31) suggested that regulation of VDR expression would allow reciprocal control of CYP27B1 and CYP24A1 activity in renal proximal tubules. We postulated that it might be possible that up-regulated VDR in the kidney of Cyp24a1–/– mice acts on Cyp27b1 expression to suppress the production of 1,25(OH)2D3, resulting in lower circulating levels of 1,25(OH)2D3. The RT-PCR data presented here suggest that surviving Cyp24a1+/– animals show reduced expression of Cyp27b1 mRNA compared with their Cyp24a1+/– littermates at 4 months of age. Accordingly, these results confirm that pathways of 1,25(OH)2D3 elimination are through the inducible expression of CYP24A1 in vivo and possibly an as yet to be defined basal clearance pathway, but that Cyp24a1+/– animals also adapt to vitamin D intoxication by down-regulation of 1,25(OH)2D3 production.
After the administration of [26,27-methyl-3H]25(OH)D3, higher residual [26,27-methyl-3H]25(OH)D3 levels and no [26,27-methyl-3H]24,25(OH)2D3 formation were observed in blood and tissues (liver, kidney, and intestine) of Cyp24a1–/– mice over 48 h. In contrast to our observations with [1?-3H]1,25(OH)2D3, metabolic clearance of [26,27-methyl-3H]25(OH)D3 was slower in both Cyp24a1–/– and Cyp24a1+/– mice. Administered [26,27-methyl-3H]25(OH)D3 stayed mostly in the blood over the full 48-h time course, with limited uptake by tissues in either genotype. These data are consistent with the longer half-life reported for 25(OH)D3 and reflect the fact that the clearance of 25(OH)D3 is greatly affected by its sequestration by the plasma carrier protein, DBP, that is synthesized in the liver. Although there was a slightly higher level of tissue total radioactivity in the tissues of Cyp24a1–/– mice compared with Cyp24a1+/– mice, there was a marked increase in residual [26,27-methyl-3H]25(OH)D3 levels in the blood and tissues (liver, kidney, and intestine) of Cyp24a1–/– mice compared with Cyp24a1+/– littermates, reflecting the reduced rate of metabolism. Indeed, chromatography showed the complete absence of 24-hydroxylated metabolites in the blood and tissues of Cyp24a1–/– mice. These results indicate the critical importance of CYP24A1 in the catabolism of 25(OH)D3.
Another way to investigate the actions of CYP24A1 in the natural setting using Cyp24a1–/– mice is to perform in vitro studies to avoid the complexity of the in vivo setting. Keratinocytes were chosen as target cells to examine the physiological role of CYP24A1 in the catabolic cascade of vitamin D in vitro. When primary neonatal keratinocytes were prepared from Cyp24a1–/– and Cyp24a1+/– mice, the findings were essentially the same as those of the in vivo study, demonstrating in particular the lack of calcitroic acid production and 24,25(OH)2D3 formation in keratinocytes from Cyp24a1–/– mice. Previous studies demonstrated that skin is capable of activating vitamin D3 and 25(OH)D3 through sequential hydroxylations, including 1-hydroxylation, and the resulting 1,25(OH)2D3 plays a role in epidermal homeostasis in normal and diseased skin (32, 33, 34). Interestingly, in our hands these Cyp24a1–/– keratinocytes failed to produce detectable amounts of 1,25(OH)2D3 production even though they were cultured under low calcium conditions, and the masking action of CYP24A1 was absent. The lack of 1,25(OH)2D3 production was possibly due to the presence of FCS and its associated vitamin D metabolites in the culture medium. The fact that none of the metabolites, including C-23/C-26 oxidation products, was detected from the incubation with keratinocytes from Cyp24a1–/– mice leads to the conclusion that CYP24A1 plays multiple roles in 1,25(OH)2D3 and 25(OH)D3 catabolism. To date, our results suggest that CYP24A1 is the main enzyme required for the catabolism of 1,25(OH)2D3 and 25(OH)D3, although it might be prudent to not completely exclude alternative pathways until we have fully defined the nature of water-soluble biliary metabolites in the Cyp24a1–/– mice.
The results from the keratinocyte study of 1,25(OH)2D3 in VDR–/– mice indicate that Cyp24a1 expression is mediated through VDR. When primary keratinocytes were incubated with a physiological concentration of 1,25(OH)2D3, VDR+/+ mice demonstrated a series of metabolites of 1,25(OH)2D3 through the C-24/C-23 oxidation pathway. However, keratinocytes from VDR–/– mice showed a complete block of target cell metabolism of 1,25(OH)2D3 (as shown in Fig. 7). The difference illustrates the highly inducible nature of target cell Cyp24a1. Even when substrate was increased to pharmacological concentrations, little metabolism was observed in cells from VDR–/– mice compared with VDR+/+ mice (data not shown). One published in vivo study using VDR-null mice (35) demonstrated that the degradation of 1,25(OH)2D3 in VDR-null mice was substantially reduced compared with that in wild-type control mice. Interestingly, unlike the in vitro study, 1,25,26(OH)3D3 was found in plasma, kidney, and liver of VDR-null mice (35). This suggests that an enzyme that catalyzes the 26-hydroxylation of 1,25(OH)2D3, that would be independent of VDR stimulation, may exist in vivo. Recently, several liver cytochrome P450 isoforms, including CYP2R1, CYP2J3, CYP3A4, as well as CYP27A1, have been shown to possess vitamin D side-chain hydroxylation activity, and one wonders if this activity can be directed at the C26 position as well as the C25 position, as is currently claimed (1, 36, 37).
In conclusion, the Cyp24a1-null mouse is a valuable model, with potential to establish the pharmacological as well as the physiological role of CYP24A1. The availability of the Cyp24a1-null mouse also opens up new approaches to determine the degree of involvement of CYP24A1 in vitamin D analog metabolism.
References
Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1: a microsomal vitamin D 25-hydroxylase. J Biol Chem 278:38084–38093
Jones G, Strugnell SA, Deluca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231
Masuda S, Jones G 2003 Vitamin D analogs: drug design based on proteins involved in vitamin D signal transduction. Curr Drug Targets Immune Endocr Metab Disord 3:43–66
Omdahl JL, Morris HA, May BK 2002 Hydroxylase enzymes of the vitamin D pathway: expression, function, and regulation. Annu Rev Nutr 22:139–166
Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180
Lohnes D, Jones G 1987 Side chain metabolism of vitamin D3 in osteosarcoma cell line UMR-106. Characterization of products. J Biol Chem 262:14394–14401
Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762
Ohyama Y, Okuda K 1991 Isolation and characterization of a cytochrome P-450 from rat kidney mitochondria that catalyzes the 24-hydroxylation of 25-hydroxyvitamin D3. J Biol Chem 266:8690–8695
Chen KS, Prahl JM, DeLuca HF 1993 Isolation and expression of human 1,25-dihydroxyvitamin D3 24-hydroxylase cDNA. Proc Natl Acad Sci USA 90:4543–4547
Itoh S, Yoshimura T, Iemura O, Yamada E, Tsujikawa K, Kohama Y, Mimura T 1995 Molecular cloning of 25-hydroxyvitamin D3 24-hydroxylase (Cyp-24) from mouse kidney: its inducibility by vitamin D3. Biochim Biophys Acta 1264:26–28
Esvelt RP, Schnoes HK, Deluca HF 1979 Isolation and characterization of 1-hydroxy-23-carboxy-tetranorvitamin D: a major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983
Ishizuka S, Norman AW 1987 Metabolic pathways from 1,25-dihydroxyvitamin D3 to 1,25-dihydroxyvitamin D3-26,23-lactone. Stereo-retained and stereo-selective lactonization. J Biol Chem 262:7165–7170
Siu-Caldera ML, Zou L, Ehrlich MG, Schwartz ER, Ishizuka S, Reddy GS 1995 Human osteoblasts in culture metabolize both 1,25-dihydroxyvitamin D3 and its precursor 25-hydroxyvitamin D3 into their respective lactones. Endocrinology 136:4195–4203
Akiyoshi-Shibata M, Sakaki T, Ohyama Y, Noshiro M, Okuda K, Yabusaki Y 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. A study with the mature enzyme expressed in Escherichia coli. Eur J Biochem 224:335–343
Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant escherichia coli cells producing rat mitochondrial CYP24A1. CYP24A1 can convert 1,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48
Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472
Masuda S, Gao M, Zhang A, Kaufmann M, Jones G 2003 Importance of cytochrome P450-mediated metabolism in the mechanism of action of vitamin D analogs. Recent Results Cancer Res 164:189–202
Sakaki T, Sawada N, Komai K, Shiozawa S, Yamada S, Yamamoto K, Ohyama Y, Inouye K 2000 Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24A1. Eur J Biochem 267:6158–6165
St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxy-vitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666
Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D-responsive element in the 5'-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550
Chen KS, DeLuca HF 1995 Cloning of the human 1,25-dihydroxyvitamin D3-24-hydroxylase gene promoter and identification of two vitamin D-responsive elements. Biochim Biophys Acta 1263:1–9
Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835
Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama 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 16:391–396
Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141
Sakai Y, Demay MB 2000 Evaluation of keratinocyte proliferation and differentiation in vitamin D receptor knockout mice. Endocrinology 141:2043–2049
Bligh EG, Dyer WG 1957 A rapid method of total lipid extraction and purification. Can J Biochem 37:911–917
Siu-Caldera ML, Sekimoto H, Weiskopf A, Vouros P, Muralidharan KR, Okamura WH, Bishop J, Norman AW, Uskokovic MR, Schuster I, Reddy GS 1999 Production of 1,25-dihydroxy-3-epi-vitamin D3 in two rat osteosarcoma cell lines (UMR 106 and ROS 17/2.8): existence of the C-3 epimerization pathway in ROS 17/2.8 cells in which the C-24 oxidation pathway is not expressed. Bone 24:457–463
Londowski JM, Kost SB, Meier W, Labler L, Kumar R 1986 Biological activity of the C-1, C-3, C-25, ?-D-glucopyranosides of 1,25-dihydroxyvitamin D3. J Pharmacol Exp Ther 237:837–840
Ayton B, Martin F 1987 Conjugated forms of [3H]1, 25-dihydroxyvitamin in rat bile. J Steroid Biochem 26:667–677
Shimada K, Nakatani I, Saito K, Mitamura K 1996 Separation and characterization of mono-glucuronides of vitamin D3 and 25-hydroxyvitamin D3 in rat bile by high-performance liquid chromatography. Biol Pharm Bull 19:491–494
Iida K, Shinki T, Yamaguchi A, DeLuca HF, Kurokawa K, Suda T 1995 A possible role of vitamin D receptors in regulating vitamin D activation in the kidney. Proc Natl Acad Sci USA 92:6112–6116
Ray S, Ray R, Holick MF 1995 Metabolism of 3H-1,25-dihydroxyvitamin D3 in cultured human keratinocytes. J Cell Biochem 59:117–122
Lehmann B, Genehr T, Knuschke P, Pietzsch J, Meurer M 2001 UVB-induced conversion of 7-dehydrocholesterol to 1,25-dihydroxyvitamin D3 in an in vitro human skin equivalent model. J Invest Dermatol 117:1179–1185
Xie Z, Munson SJ, Huang N, Portale AA, Miller WL, Bikle DD 2002 The mechanism of 1,25-dihydroxyvitamin D3 autoregulation in keratinocytes. J Biol Chem 277:36987–36990
Endres B, Kato S, DeLuca HF 2000 Metabolism of 1,25-dihydroxyvitamin D3 in vitamin D receptor-ablated mice in vivo. Biochemistry 39:2123–2129
Yamasaki T, Izumi S, Ide H, Ohyama Y 2004 Identification of a novel rat microsomal vitamin D3 25-hydroxylase. J Biol Chem 279:22848–22856
Gupta RP, Hollis BW, Patel SB, Patrick KS, Bell NH 2004 CYP3A4 is a human microsomal vitamin D 25-hydroxylase. J Bone Miner Res 19:680–688
Jones G, Byford V, Sakai Y, Demay MB 2000 Dramatic reduction of 1,25-(OH)2D3 catabolism in keratinocytes from VDR-null mice. J Bone Miner Res 15(Suppl 1):S181
Jones G, Byford V, Arabian A, St-Arnaud R 2000 Altered pharmacokinetics of 1,25-(OH)2D3 in blood and tissues of the CYP24-null mouse. J Bone Miner Res 15(Suppl 1):S199(Sonoko Masuda, Valarie By)