Vitamin D receptor displays DNA binding and transactivation as a heterodimer with the retinoid X receptor, but not with the thyroid hormone receptor.

The vitamin D receptor (VDR) is a transcription factor believed to function as a heterodimer with the retinoid X receptor (RXR). However, it was reported [Schräder et al., 1994] that, on putative vitamin D response elements (VDREs) within the rat 9k and mouse 28k calcium binding protein genes (rCaBP 9k and mCaBP 28k), VDR and thyroid hormone receptor (TR) form heterodimers that transactivate in response to both 1,25-dihydroxyvitamin D(3) (1,25(OH)(2)D(3)) and triiodothyronine (T(3)). We, therefore, examined associations of these receptors on the putative rCaBP 9k and mCaBP 28k VDREs, as well as on established VDREs from the rat osteocalcin (rOC) and mouse osteopontin (mOP) genes, plus the thyroid hormone response element (TRE) from the rat myosin heavy chain (rMHC) gene. In gel mobility shift assays, we found no evidence for VDR-TR heterodimer interaction with any tested element. Further, employing these hormone response elements linked to reporter genes in transfected cells, VDR and TR mediated responses to their cognate ligands only from the rOC/mOP and rMHC elements, respectively, while the CaBP elements were unresponsive to any combination of ligand(s). Utilizing the rOC and mOP VDREs, two distinct repressive actions of TR on VDR-mediated signaling were demonstrated: a T(3)-independent action, presumably via direct TR-RXR competition for DNA binding, and a T(3)-dependent repression, likely by diversion of limiting RXR from VDR-RXR toward the formation of TR-RXR heterodimers. The relative importance of these two mechanisms differed in a response element-specific manner. These results may provide a partial explanation for the observed association between hyperthyroidism and bone demineralization/osteoporosis.

Vitamin D receptors from patients with resistance to 1,25-dihydroxyvitamin D3: point mutations confer reduced transactivation in response to ligand and impaired interaction with the retinoid X receptor heterodimeric partner.

Hereditary hypocalcemic vitamin D-resistant rickets is attributable to defects in the nuclear receptor for 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3]. Two novel point mutations (I314S and R391C) identified in the hormone-binding domain of the human vitamin D receptor (VDR) from patients with hereditary hypocalcemic vitamin D-resistant rickets confer the receptor with sharply reduced 1,25-(OH)2D3-dependent transactivation. These natural mutations, especially R391C, also lead to a second specific consequence, namely impaired heterodimeric interaction with retinoid X receptor (RXR). While the transactivation ability of the I314S mutant can be largely restored by providing excess 1,25-(OH)2D3, R391C activity is more effectively restored with exogenous RXR. These observations are reflected also in the clinical course of each patient: the patient bearing the I314S mutation showed a nearly complete cure with pharmacological doses of a vitamin D derivative, whereas the patient bearing R391C responded only partially to such therapy. Further tests with patient fibroblasts and transfected cells show that the activity of the I314S VDR mutant is augmented somewhat by added RXR, while transactivation by the R391C mutant is best corrected by RXR in the presence of excess hormone. Thus, the effects of hormone vs. RXR in bolstering these mutant VDRs, such that they mediate efficient transactivation, are not entirely separable. The unique properties of these genetically altered receptors establish a new subclass of natural human VDR mutants that illustrate, in vivo, the importance of both 1,25-(OH)2D3 binding and heterodimerization with RXR in VDR action.

Purified human vitamin D receptor overexpressed in E. coli and baculovirus systems does not bind 1,25-dihydroxyvitamin D3 hormone efficiently unless supplemented with a rat liver nuclear extract.

We report here that highly purified human vitamin D receptor (hVDR) derived from E. coli or baculovirus expression systems does not exhibit saturable, high affinity 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) ligand binding when these preparations alone are analyzed. Inclusion of rat liver nuclear extract, which does not itself contain detectable 1,25(OH)2D3 binding activity, is required to endow hVDR isolated from bacterial or insect cells with the property of high affinity hormone binding (Kd = 0.13-0.22 nM). This observation should facilitate the valid assay of 1,25(OH)2D3 binding activity and kinetics in samples of overexpressed hVDR. Moreover, since rat liver nuclear extract contains retinoid X receptors and possibly other auxiliary factors capable of forming heterodimers with hVDR that in turn associate with vitamin D responsive elements, we hypothesize that like DNA binding, 1,25(OH)2D3 binding to hVDR requires the cooperation of a co-receptor or some uncharacterized receptor activating/stabilizing factor.

1,25-Dihydroxyvitamin D3 does not up-regulate vitamin D receptor messenger ribonucleic acid levels in hypophosphatemic mice.

The effect of 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) administration on duodenal vitamin D receptor (VDR) mRNA levels in hypophosphatemic (Hyp) mice, a murine homologue of human X-linked hypophosphatemic rickets, was examined. Basal levels of VDR mRNA in Hyp mice were similar to those of normal littermates and, in normal mice, VDR mRNA levels were up-regulated 1.8-2.7-fold after injection of 1 microgram/kg 1,25(OH)2D3. In contrast, no significant change in VDR mRNA was observed in Hyp mice treated with 1,25(OH)2D3. To determine the effect of phosphate repletion on VDR mRNA levels, high-phosphate diet was fed to Hyp mice. Although plasma phosphorus concentration was restored to normal, up-regulation of VDR mRNA was not recovered with phosphate supplementation. These results indicate that the vitamin D-resistance in Hyp mice is not caused by hypophosphatemia, per se, and may result from a fundamental molecular defect in vitamin D action at the intestine which could be related to ineffective up-regulation of VDR mRNA by 1,25(OH)2D3.

Stable expression of the calbindin-D28K complementary DNA interferes with the apoptotic pathway in lymphocytes.

The WEHI7.2 thymoma cell line undergoes apoptotic cell death when exposed to glucocorticoids and agents that increase intracellular cAMP. Several lines of evidence indicate that calcium may play an important role in events culminating in lymphocyte apoptosis. In these studies, calbindin-D28K was stably overexpressed in WEHI7.2 cells to determine if increasing the Ca(2+)-binding capacity of the cell interferes with the apoptotic pathway. Indeed, stable expression of calbindin-D28K decreased the apoptotic effects of dexamethasone and forskolin, and the level of resistance to these agents correlated with the relative amount of calbindin expressed in each line. Overexpression of calbindin also increased cell survival in the presence of the calcium ionophore A23187. The stably expressed calcium-binding protein appeared to exert its protective effect subsequent to transcriptional activation, since glucocorticoid- and cAMP-induced gene expression were not affected. These data support the proposal that calcium fluxes are involved in apoptosis and suggest that high level expression of proteins that buffer calcium fluxes can effectively suppress death in apoptosis-susceptible cells.

In vitro transcription and translation of the human 1,25-dihydroxyvitamin D3 receptor cDNA.

A fragment of the complementary deoxyribonucleic acid to the human 1,25-dihydroxyvitamin D3 receptor protein containing essentially the entire open reading frame was transcribed and translated in vitro. The resulting protein was then demonstrated to exhibit the physical and functional features, i.e. molecular weight, immunoreactivity, 1,25-dihydroxyvitamin D3 binding, and DNA-cellulose binding, of the native human receptor from the T47D cell line. This validates the authenticity of the cDNA in a cell free system and provides a biochemical means of generating this rare and labile macromolecule to use in heretofore difficult structure/function studies.

Estrogen binding, receptor mRNA, and biologic response in osteoblast-like osteosarcoma cells.

High specific activity estradiol labeled with iodine-125 was used to detect approximately 200 saturable, high-affinity (dissociation constant approximately equal to 1.0 nM) nuclear binding sites in rat (ROS 17/2.8) and human (HOS TE85) clonal osteoblast-like osteosarcoma cells. Of the steroids tested, only testosterone exhibited significant cross-reactivity with estrogen binding. RNA blot analysis with a complementary DNA probe to the human estrogen receptor revealed putative receptor transcripts of 6 to 6.2 kilobases in both rat and human osteosarcoma cells. Type I procollagen and transforming growth factor-beta messenger RNA levels were enhanced in cultured human osteoblast-like cells treated with 1 nM estradiol. Thus, estrogen can act directly on osteoblasts by a receptor-mediated mechanism and thereby modulate the extracellular matrix and other proteins involved in the maintenance of skeletal mineralization and remodeling.

Cloning and expression of full-length cDNA encoding human vitamin D receptor.

Complementary DNA clones encoding the human vitamin D receptor have been isolated from human intestine and T47D cell cDNA libraries. The nucleotide sequence of the 4605-base pair (bp) cDNA includes a noncoding leader sequence of 115 bp, a 1281-bp open reading frame, and 3209 bp of 3' noncoding sequence. Two polyadenylylation signals, AATAAA, are present 25 and 70 bp upstream of the poly(A) tail, respectively. RNA blot hybridization indicates a single mRNA species of approximately equal to 4600 bp. Transfection of the cloned sequences into COS-1 cells results in the production of a single receptor species indistinguishable from the native receptor. Sequence comparisons demonstrate that the vitamin D receptor belongs to the steroid-receptor gene family and is closest in size and sequence to another member of this family, the thyroid hormone receptor.

Molecular cloning of complementary DNA encoding the avian receptor for vitamin D.

Vitamin D3 receptors are intracellular proteins that mediate the nuclear action of the active metabolite 1,25-dihydroxyvitamin D3 [1,25(OH)2D3]. Two receptor-specific monoclonal antibodies were used to recover the complementary DNA (cDNA) of this regulatory protein from a chicken intestinal lambda gt11 cDNA expression library. The amino acid sequences that were deduced from this cDNA revealed a highly conserved cysteine-rich region that displayed homology with a domain characteristic of other steroid receptors and with the gag-erbA oncogene product of avian erythroblastosis virus. RNA selected via hybridization with this DNA sequence directed the cell-free synthesis of immunoprecipitable vitamin D3 receptor. Northern blot analysis of polyadenylated RNA with these cDNA probes revealed two vitamin D receptor messenger RNAs (mRNAs) of 2.6 and 3.2 kilobases in receptor-containing chicken tissues and a major cross-hybridizing receptor mRNA species of 4.2 kilobases in mouse 3T6 fibroblasts. The 4.2-kilobase species was substantially increased by prior exposure of 3T6 cells to 1,25(OH)2D3. This cDNA represents perhaps the rarest mRNA cloned to date in eukaryotes, as well as the first receptor sequence described for an authentic vitamin.

Postpartum resolution of hypocalcemia in a lactating hypoparathyroid patient.

A 24 year old woman with post-surgical hypoparathyroidism was studied during pregnancy and lactation. During pregnancy the patient required less vitamin D therapy for control of her hypoparathyroidism and, while lactating, maintained a normal serum calcium without any supplemental vitamin D. The serum parathyroid hormone concentration and plasma 1,25 (OH)2 vitamin D concentration were undetectable and low normal respectively at a time when the serum calcium concentration was normal and the patient was not on vitamin D therapy. Urinary calcium excretion was low during this period and may explain the normalization of the serum calcium. The mechanism by which the improvement in calcium metabolism occurred is unknown, but may be secondary to a direct effect of prolactin on calcium homeostasis.

Phosphorus administration in patients with profound hypophosphatemia.

The influence of severe hypophosphatemia (less than or equal to 1.0 mg/dl) on vitamin D metabolism was prospectively determined in 11 patients before and after intravenous phosphorus administration. Evidence of liver dysfunction was present in ten patients. The mean (+/- SE) plasma 25 hydroxycholecalciferol [25(OH)D] was significantly decreased before phosphorus therapy when compared to control subjects (9.4 +/- 1.3 vs. 17.8 +/- 1.3 ng/ml, P less than 0.001). With phosphorus administration, serum phosphorus increased from 0.59 +/- 0.07 to 2.58 +/- 0.09 mg/dl while 1,25 dihydroxycholecalciferol [1,25(OH)2D] decreased from 34.6 +/- 4.3 to 14.3 +/- 2.9 pg/ml (P less than 0.001). Plasma 25(OH)D, plasma immunoreactive PTH (both amino and carboxyterminal) and serum calcium did not change after phosphorus administration, suggesting that phosphorus alone was responsible for the change in plasma 1,25(OH)2D concentration. An inverse correlation was found between serum phosphorus and plasma 1,25(OH)2D (r = -0.62, P less than 0.005). In addition, a direct correlation was observed between plasma 25(OH)D and 1,25(OH)2D both before (r = 0.66, P less than 0.005) and after (r = 0.74, P less than 0.005) phosphorus administration. Thus, the decrease in 1,25(OH)2D levels with phosphorus therapy suggests a role of serum phosphate in the regulation of this sterol, and hypophosphatemia or phosphorus depletion may change the relationship of substrate [25(OH)D] to product [1,25(OH)2D].

Association of 1,25-dihydroxyvitamin D3 with cultured 3T6 mouse fibroblasts. Cellular uptake and receptor-mediated migration to the nucleus.

The uptake of 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) by intact cells was investigated using the cultured embryonic 3T6 mouse fibroblast as a model. Suspended cells, incubated for 60-90 min in serum-containing culture medium supplemented with 1,25-(OH)2D3 (2 nM), maximally accumulate hormone which becomes bound to a typical vitamin D 3.3 S receptor protein. Incubation of cells with varying concentrations of 1,25-(OH)2D3 reveals the presence of 21,000 receptor molecules/3T6 cell, with an apparent uptake constant of 6-8 X 10(-10) M at 37 degrees C. This value contrasts with the equilibrium dissociation constant (Kd) for 1,25-(OH)2D3 binding of 6 X 10(-11) M as determined at 2 degrees C in disrupted cell cytosol. The distribution of unoccupied (R0) receptors is predominantly (greater than 85%) cytosolic in the hormone-deprived state (1,25-(OH)2D3 less than 0.05 nM), whereas exposure to 1,25-(OH)2D3 (2 nM) leads to almost complete nuclear localization of the occupied receptor at both 2 and 37 degrees C. This phenomenon was similarly supported through reconstitution of receptor and purified 3T6 nuclei in vitro in which binding also occurs at 2 degrees C. The majority (65%) of intact cell-formed receptor-nuclear complexes can be solubilized by micrococcal nuclease treatment, suggesting the participation of DNA in the acceptor binding site for the 1,25-(OH)2D3 receptor. Consistent with these data, DNA-binding of receptor also occurred in vitro at 2 degrees C and was a characteristic of both occupied (Rs) and unoccupied receptors. However, elution of the latter occurred at reduced ionic strength, implying that the hormone does physically alter the receptor protein. This binding was also sensitive to prior ethidium bromide saturation of DNA-cellulose, but not phosphocellulose. Although the biologic effects of the 1,25-(OH)2D3 hormone in 3T6 fibroblasts are as yet unknown, the present findings support previous work with 1,25-(OH)2D3 receptors and suggest that this cell represents a good model for the study of nuclear events associated with the molecular action of 1,25-(OH)2D3.

Biosynthesis, purification and receptor binding properties of high specific radioactivity 1 alpha, 24(R),25-trihydroxy-[26,27-methyl-3H]-vitamin D3.

A procedure for the biosynthesis and purification of 1 alpha, 24(R),25-trihydroxy[26,27-methyl-3H]-vitamin D3 (1,24,25-(OH)3[3H]D3) is reported. A kidney homogenate from chicks receiving a high calcium diet (3%) and oral supplements of 1 alpha, 25-dihydroxyvitamin D3 (1,25-(OH)2D3) was used for C-24-hydroxylation of 1 alpha, 25-dihydroxy[26,27-methyl-3H]-vitamin D3 (1,25-(OH)2[3H]D3), in vitro. Extraction and purification of the homogenate lipid fraction by Sephadex LH-20 and high performance liquid chromatography yielded radiochemically pure 1,24-25-(OH)3[3H]D3 with a specific radioactivity equivalent to the initial substrate (166 Ci/mmol). The authenticity of the generated metabolite was assessed by co-migration with synthetic 1,24,25-(OH)3D3 on high performance liquid chromatography and by equimolar competition with authentic radioinert 1,24,25-(OH)3D3 for binding to a purified receptor protein from rat kidney. Binding studies indicate the trihydroxylated metabolite competes 40-50% as effectively as 1,25-(OH)2D3 for hormone binding sites. Further analysis of 1,24,25-(OH)3D3-receptor interaction reveals a high-affinity, saturable binding with an apparent K4 of 2.2 x 10(-9) M. These studies demonstrate that although slightly less active than 1,25-(OH)2D3, 1,24,25-(OH)3D3 is capable of hormone-like interactions, in vitro. The availability of this high specific radioactivity sterol should allow for clarification of its potential physiologic significance.

Effect of metabolic acidosis in intestinal absorption of calcium and phosphorus.

To investigate the effect of metabolic acidosis on intestinal calcium (Ca) and phosphorus (P) absorption and vitamin D metabolism, metabolic balance studies and in vitro gut sac uptake of 45Ca and [32P]phosphate were performed in rats maintained on low-Ca and moderately low-P diet and fed NH4Cl for 3 or 9 days and pair-fed controls. Plasma 1,25(OH)2D concentration was measured in the rats fed NH4Cl for 9 days and their controls. Net Ca and P absorption was 87-92% in the acidotic rats and did not differ from control. Moreover, gut sac uptakes of 45Ca and [32P]phosphate were not different from control. Plasma 1,25(OH)2D was higher in the ammonium chloride-fed rats than in controls (213 +/- 44 vs. 110 +/- 12 pg/ml), and serum P was lower in the acidotic animals (4.6 +/- 0.7 vs. 7.6 +/- 0.3 mg/dl). These data indicate that metabolic acidosis does not depress the augmented intestinal absorption of calcium and phosphorus noted during their dietary deprivation nor reduce the plasma level of 1,25(OH)2D.

Effect of the menstrual cycle on calcium-regulating hormones in the normal young woman.

To examine the effect of estrogens on circulating levels of calcium-regulating hormones, total serum calcium, ionized calcium, inorganic phosphate, immunoreactive parathyroid hormone, calcitonin, 1,25-dihydroxyvitamin D, and estradiol (E2) levels were assayed on menstrual cycle days 3 and 13 in 12 healthy young women. Despite a 3-fold increase in serum E2 levels by midcycle, no differences in the other factors were observed. Furthermore, iv infusion of calcium (3 mg/kg over a 10-min period) produced similar changes in the serum levels of immunoreactive parathyroid hormone and calcitonin on days 3 and 13 in 8 women. Our findings suggest that in normal menstruating women, endogenous increases in E2 neither directly nor indirectly stimulate 1,25-dihydroxyvitamin D production, nor do they affect circulating levels of the hormones known to influence calcium homeostasis in man.

Dynamic changes in circulating 1,25-dihydroxyvitamin D during reproduction in rats.

The concentrations of 1,25-dihydroxyvitamin D [1,25-(OH)2D], calcium, and phosphorus were measured in the serum of rats during pregnancy and at various stages of lactation. The concentration of 1,25-(OH)2D hormone increased almost two-fold during pregnancy and the latter part of lactation, but decreased to control levels or very low values immediately after birth and weaning, respectively. Furthermore, the concentration of 1,25-(OH)2D was inversely correlated with the concentration of calcium, suggesting that circulating 1,25-(OH)2D fluctuates in concert with calcium demands during the reproductive cycle. Parathyroidectomy in lactating rats caused a 70 percent inhibition of the normally observed 1,25-(OH)2D increase, indicating that parathyroid hormone, in response to changes in serum calcium, is a primary modulator of 1,25-(OH)2D during lactation.

Calcium and phosphorus deficiency in rats: effects on PTH and 1,25-dihydroxyvitamin D3.

Weanling male Holtzman rats were fed calcium.deficient, phosphorus-deficient, or control diets for 8 wk. Parathyroid hormone (PTH) was measured by radioimmunoassay, and 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) by a competitive binding assay. Rats fed the calcium-deficient diet (0.01% calcium, 0.6% phosphorus) became mildly hypocalcemic after 6 days. Serum calcium levels reached 5.5 +/- 0.4 mg/dl (mean +/- SD) in 5 wk (control 10.3 +/- 0.4 mg/dl). PTH increased from 285 +/- 112 to 3658 +/- 428 pg/ml within 6 wk. Maximum serum levels of 1,25(OH)2D3 (111.8 +/- 17.3 vs. control 11.4 +/- 3.8 ng/dl) were reached at 3 wk and thereafter declined to 44.6 +/- 14.0 ng/dl. In rats fed the phosphorus-deficient diet (0.6% calcium, 0.04% phosphorus), serum phosphorus fell within 24 h from 9.1 +/- 0.6 to 3.2 +/- 0.1 mg/dl, recovered to 5.6 +/- 0.4 mg/dl for 2-3 days, and then declined again. Serum calcium reached a maximum of 14.4 +/- 0.4 mg/dl at day 2 (control 10.8 +/- 0.5 mg/dl) and then slowly declined. PTH decreased within 24 h from 243 +/- 59 to 36 +/- 0 pg/ml in phosphorus-depleted rats. Serum levels of 1,25(OH)2D3 increased within 24 h and remained elevated after 6 wk of phosphorus deprivation (61.2 +/- 11.7 ng/dl vs. control 18.3 +/- 0.4 ng/dl).

Lack of effect of hypomagnesemia on elevated plasma 1,25-dihydroxycholecalciferol in acutely phosphate-deprived rats.

The effect of hypomagnesemia occurring in phosphate deprivation on elevated levels of plasma 1,25-dihydroxycholecalciferol (1,25(OH)2D3) was evaluated in rats. Phosphate deprivation was associated with an elevated plasma 1,25(OH)2D3, hypophosphatemia, and hypomagnesemia. With oral magnesium supplement during phosphate deprivation, plasma magnesium did not fall, but plasma inorganic phosphate fell and 1,25(OH)2D3 rose to comparable levels seen in phosphate-deprived rats lacking magnesium supplement. These results indicate that a rise in plasma 1,25(OH)2D3 during phosphate deprivation is not due to concomitant hypomagnesemia and support the postulate that phosphate deprivation per se is responsible for the elevated plasma 1,25(OH)2D3.