DNA bending is induced by binding of vitamin D receptor-retinoid X receptor heterodimers to vitamin D response elements.

The ability of vitamin D receptor-retinoid X receptor (VDR-RXR) heterodimers to induce a DNA bend upon binding to various vitamin D response elements (VDRE) has been investigated by circular permutation and phasing analysis. Recombinant rat VDR expressed in the baculovirus system and purified recombinant human RXR beta have been used. The VDREs were from 1,25-dihydroxyvitamin D3 (1,25-[OH]2D3) enhanced genes (rat osteocalcin, rOC; mouse osteopontin, mOP, and rat 1,25-dihydroxyvitamin D3-24-hydroxylase, r24-OHase), and a 1,25-(OH)2D3 repressed gene (human parathyroid hormone, hPTH). As shown by circular permutation analysis, VDR-RXR induced a distortion in DNA fragments containing various VDREs. Calculated distortion angles were similar in magnitude (57 degrees, 56 degrees, 61 degrees, and 59 degrees, respectively for rOC, mOP, r24-Ohase, and hPTH). The distortions took place with or without a 1,25-(OH)2D3 ligand. The centers of the apparent bend were found in the vicinity of the midpoint of all VDREs, except for rOC VDRE which was found 4 bp upstream. Phasing analysis was performed with DNA fragments containing mOP VDRE and revealed that VDR-RXR heterodimers induced a directed bend of 26 degrees, not influenced by the presence of hormone. In this study we report that similar to other members of the steroid and thyroid nuclear receptor superfamily, VDR-RXR heterodimers induce DNA bending.

Identification of a transcription factor that binds to the promoter region of the human parathyroid hormone gene.

A putative transcription factor binds a site adjacent to the negative vitamin D responsive element (VDRE) in the promoter region of the human parathyroid hormone gene. Deletion and mutation analysis reveal the binding site for this factor overlaps with the proximal repeat element of the VDRE. It includes additional nucleotides at the 3' end of the VDRE. This site has the sequence TTTGAACCTATAGTTGAGAT and a core sequence TGAACCTAT needed for binding of the factor. Experiments with specific anti-vitamin D receptor (VDR) antibodies demonstrate that VDR is not found in the factor/DNA complex. However, removing the VDR from the nuclear extract by immunoprecipitation eliminated the binding complex, and the addition of recombinant VDR to the depleted extract did not restore the factor's ability to bind to the DNA, suggesting that the factor and VDR are closely associated. Transfection experiments with various reporter constructs indicate that the factor is required for the high transcriptional activity of the human PTH gene. This high activity is significantly suppressed by 1,25-dihydroxyvitamin D3. This factor seems to be expressed in several cell types including rat osteoblasts and pituitary. Additionally, some human cancer cell lines express a high level of this factor.

Analysis of binding of the 1,25-dihydroxyvitamin D3 receptor to positive and negative vitamin D response elements.

The binding of the 1,25-dihydroxyvitamin D3 receptor to the vitamin D response elements (VDREs) in the rat osteocalcin (OSC-DRE), mouse osteopontin (MOP-DRE), rat calbindin D-9k (CaBP-DRE), and human parathyroid hormone genes (PTH-DRE) was studied. Binding of VDR to the three positive VDREs is cooperative. The degree of cooperativity is highest with the calbindin VDRE compared with either the OSC-DRE or the MOP-DRE. This cooperativity is largely absent in the case of the negative element, the PTH-DRE. The VDR binds in order of decreasing affinity to PTH-DRE > OSC-DRE = MOP-DRE > CaBP-DRE. Thus, the greatest affinity is associated with the lowest degree of cooperativity. Further study has revealed that the PTH-VDRE actually consists of two repeat elements like all other VDREs and is not a single six-base sequence. A nuclear factor has also been found that binds downstream from the GGTTCA element in the PTH promoter. The binding site of this factor overlaps the PTH-DRE nucleotide sequence.

Recent advances in the molecular biology of vitamin D action.

Following the cloning and deletion analysis of the vitamin D receptor, most recent advances have been in the isolation and characterization of the DNA response elements found in the promoter region of target genes of vitamin D. Vitamin D, like the thyroid and retinoid hormones, binds to repeat sequences, but the repeats are separated by three nonspecified bases. The action of the VDR requires the presence of the RXR proteins and evidently other proteins that are involved in regulating transcriptions. A possible role of phosphorylation of the ligand binding domain of the VDR in transcription has also appeared. Very likely, the molecular events involved in vitamin D stimulation or suppression of a target gene will include its interaction with a number of transcription factors, both in the regulation of transcription and in the actual machinery involved in the transcription process through polymerase II. Although likely, it is not entirely clear whether the genomic action of vitamin D can account for all of its biological activities. Nongenomic actions of the vitamin D hormone have been reported, but convincing evidence that this is of biological importance in vivo is lacking. Advances in our understanding of the vitamin D mechanism of action can clearly be expected from physical studies of cloned and expressed vitamin D receptor and its subdomains, elucidation of the transcription factors in vitamin D-modulated transcription of target genes, elucidation of the role of phosphorylation in the transcription process, and the identification of important genes that are regulated in the specific target tissues responsive to vitamin D. This will definitely remain as a very active field of investigation well into the future.

Transcriptional control of the osteocalcin gene by 1,25-dihydroxyvitamin D-2 and its 24-epimer in rat osteosarcoma cells.

The effects of two vitamin D analogs, 1,25-dihydroxyvitamin D-2 and 24-epi-1,25-dihydroxyvitamin D-2, were examined on osteocalcin gene expression in the rat osteosarcoma cell line ROS 17/28. Our results indicate that these analogs are more transcriptionally active than 1,25-dihydroxyvitamin D-3, particularly the 24-epimer. Assessment of reporter gene chloramphenicol acetyltransferase (CAT) activity, using the vitamin D responsive element (VDRE) derived from the human osteocalcin gene promoter. revealed that both analogs stimulated CAT activity 5- to 10-fold. 1,25-Dihydroxyvitamin D-2 was slightly more active than 1,25-dihydroxyvitamin D-3, while the 24-epimer was twice as effective. 1,25-Dihydroxyvitamin D-3 also stimulated osteocalcin mRNA accumulation by 2-fold over vehicle-treated cells, 1,25-dihydroxyvitamin D-2 by 2.5-fold, and 24-epi-1,25-dihydroxyvitamin D-2 by 4-fold. Electrophoretic mobility shift assays using the osteocalcin vitamin D responsive element revealed no increase in DNA binding with either analog when compared to 1,25-(OH)2D3. Examination of CAT activity using the rat 24-hydroxylase VDRE indicated no significant difference in transcription with these compounds, suggesting that the vitamin D-2 analogs preferentially activate osteocalcin gene expression.

Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter.

The interaction between the two vitamin D response elements (DRE) located at -154 to -134 base pairs (bp) and -262 to -238 bp from the transcription initiation site has been studied using reporter gene assays and binding assays by electrophoretic gel shift measurements. 3 half-sites separated by 3 bp were found necessary for transactivation by the -154 to -125 DRE, while 2 half-sites separated by 3 bp were needed for the DRE at -262 to -238 to function. However, the two DREs together provided maximal activity. The 93-bp fragment separating the two DREs was not required and could be deleted. The most effective binding by receptor was found with the two complete DREs (dissociation constant (Kd) = 13.7 pM), although each DRE bound to the receptor and nuclear accessory factor with about 5 nM Kd. The two DREs (a total of 5 half-sites) apparently account for most if not all of the transactivation of the rat 24-hydroxylase by 1,25-dihydroxyvitamin D3. This system represents the most powerful of the DREs reported to date.

Identification of a vitamin D-response element in the rat calcidiol (25-hydroxyvitamin D3) 24-hydroxylase gene.

The calcidiol (25-hydroxyvitamin D3) 24-hydroxylase is one of the key enzymes in the metabolism of vitamin D. This enzyme acts on both calcidiol and calcitriol (1,25-dihydroxyvitamin D3) to initiate degradation of these potent vitamin D metabolites and is tightly regulated. Calcitriol itself induces this enzyme and acts at the transcriptional level. Transcriptional regulation of genes by calcitriol has been shown to occur via the vitamin D-receptor binding to a vitamin D-response element located upstream of the transcription start site. We now report a vitamin D-response element located between nt -262 and nt -238 of the rat calcidiol 24-hydroxylase gene. This sequence binds the calcitriol receptor and confers vitamin D-dependent transactivation of transcription to its own, as well as heterologous, promoter.

Vitamin D-influenced gene expression via a ligand-independent, receptor-DNA complex intermediate.

A lingering question regarding the regulation of target gene expression by 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3] has been the delineation of vitamin D receptor (VDR)-DNA binding and transactivation. This report confirms that initial VDR-DNA interaction occurs in a ligand-independent fashion. An electrophoretic mobility-shift analysis demonstrated that VDR, derived from extracts of the small intestines of vitamin D-deficient rats, is capable of binding a vitamin D response element (DRE). Additional mobility-shift studies using either porcine-derived VDR or recombinant rat VDR from insect cells revealed DRE-binding capability in the absence of 1,25-(OH)2D3. The reactions were performed in various salt environments, with the maximum of porcine VDR-DRE and rat VDR-DRE binding detected at 100 mM and 150 mM KCl, respectively. The addition of 1,25-(OH)2D3 to an identical set of reaction mixtures resulted in increased DRE binding with greater affinities exhibited by both VDR types. These two phenomena were confirmed upon examination of an elution profile of VDR bound to DRE-linked Sepharose. When a linear KCl gradient was used for elution without the addition of 1,25-(OH)2D3, the peak of VDR was 205 mM KCl; the presence of exogenous hormone shifted the maximum VDR elution to a position corresponding to 265 mM KCl. Based on these data and previous reports on VDR-mediated transactivation, we propose a model for 1,25-(OH)2D3-influenced target gene expression.

Phosphorylation is involved in transcriptional activation by the 1,25-dihydroxyvitamin D3 receptor.

The 1,25-dihydroxyvitamin D3 receptor becomes phosphorylated upon treatment with 1,25-dihydroxyvitamin D3. We have investigated the role of phosphorylation in the transcriptional activity induced by 1,25-dihydroxyvitamin D3 through its receptor. An active 1,25-dihydroxyvitamin D3-dependent transcription system was reconstituted in CV-1 cells by co-transfection of plasmids containing the rat 1,25-(OH)2D3 receptor DNA and a functional vitamin D response element (DRE) in a reporter gene construct. Treatment of these transiently transfected CV-1 cells with modulators of protein kinase A (8-Br-cAMP, PKIA and H-9) and phosphatases (Okadaic acid) resulted in mimicking or abolishing the transcriptional activity of 1,25-dihydroxyvitamin D3 in a receptor-dependent fashion. These modulators directly altered 1,25-dihydroxyvitamin D3 receptor phosphorylation. Therefore, the present results strongly suggest that phosphorylation plays a central role in the transcriptional activity of the 1,25-dihydroxyvitamin D3 receptor.

Identification of a 1,25-dihydroxyvitamin D3-response element in the 5'-flanking region of the rat calbindin D-9k gene.

The rat calbindin D-9k gene is transcriptionally regulated by 1,25-dihydroxyvitamin D3 in the intestine. We have examined the 5'-flanking region of this gene and identified a 1,25-dihydroxyvitamin D3-responsive element (DRE) between nucleotides -489 and -445. This element confers 1,25-dihydroxyvitamin D3 responsiveness through its native promoter and the heterologous thymidine kinase promoter, and it contains the sequence GGGTGTCGGAAGCCC, which is homologous to the other previously identified DREs. Incubation of this element with the 1,25-dihydroxyvitamin D3 receptor produced a specific protein-DNA complex, which shifted to a higher molecular weight form upon the addition of a monoclonal antibody specific to the 1,25-dihydroxyvitamin D3 receptor. Therefore, the 5'-flanking region of the rat calbindin D-9k gene contains a DRE that mediates the enhanced expression of this gene by 1,25-dihydroxyvitamin D3 in the intestine.

Methylation of the vitamin D-dependent CaBP gene (calbindin 9 Kd) does not mediate tissue or vitamin D regulation.

The vitamin D-dependent intestinal calcium binding protein gene is predominantly expressed in the intestine. In this report we have examined the possibility that methylation of the gene might play a role in its tissue-specific expression employing genomic Southern analysis. None of the Hpa II and Hha I sites examined by the indicated probes in and around the gene were found to be methylated in the intestine, kidney and liver. No change in the methylation of these sites was detected in response to 1,25-dihydroxy-vitamin D3 administration to vitamin D-deficient rats under conditions which stimulate the expression of the gene. These results indicate that the rat intestinal calcium binding protein gene is not methylated in these tissues, at the indicated sites and, therefore, methylation seems not to be involved in the regulation of this gene's expression.

Molecular cloning of the cDNA and chromosomal gene for vitamin D-dependent calcium-binding protein of rat intestine.

A cDNA encoding the vitamin D-dependent rat intestinal calcium-binding protein has been isolated by screening a rat intestinal cDNA library. The cDNA is 406 nucleotides long and appears to contain all the sequences of the mRNA. The cDNA includes the entire protein coding region. It consists of 237 nucleotides coding for 79 amino acids, including the starting methionine, flanked by 62 and 107 noncoding nucleotides at the 5' and 3' ends, respectively. Using the cloned cDNA, we have isolated a genomic clone from a rat liver genomic library. Restriction mapping and Southern analysis using synthetic oligonucleotides localized the gene to a 4.0-kilobase-pair HindIII fragment.

Mechanism of copper transport from plasma to hepatocytes.

The effects of plasma components on the kinetics of copper transport by rat hepatocytes were examined in an attempt to determine how copper is mobilized from plasma for uptake by the liver. Specific protein-facilitated transport was indicated by saturation kinetics, competition by related substrates, and similar kinetic parameters for uptake and efflux. For copper uptake, Km = 11 +/- 0.6 microM and Vmax = 2.7 +/- 0.6 nmol Cu/(min X mg protein). Zinc is a competitive inhibitor of copper uptake, and copper competes for zinc uptake. Copper efflux from preloaded cells is biphasic. The kinetic parameters for the initial rapid phase are similar to the parameters for uptake. Copper transport by hepatocytes is strictly passive. A variety of metabolic inhibitors have no effect on uptake and initial rates are solely dependent on extracellular-intracellular concentration gradients. Albumin markedly inhibits copper uptake by a substrate removal mechanism, and histidine facilitates albumin-inhibited copper uptake. The active species that delivers copper to hepatocytes under conditions of excess albumin and excess histidine is the His2Cu complex. Experiments with [3H]His2 64Cu showed that the transported species is free ionic copper. The kinetic parameters of copper transport by hepatocytes isolated from the brindled mouse model of Menkes' disease are normal. However, these cells show a decreased capacity to accumulate copper on prolonged incubation. An intracellular metabolic defect seems to be involved.

Copper efflux kinetics from rat hepatocytes.

The kinetics of copper efflux from rat hepatocytes were determined to further characterize the hepatic Cu(II) transport system. Efflux was biphasic. Net efflux was rapid for 1-5 min, and 35-45% of preloaded copper was lost by 40 min. Efflux was negligible after 40 min. The retained percentage was independent of the preloading concentration, but the total amount of intracellular copper that was available for efflux gradually decreased as the duration of the preloading period increased. Unlabeled extracellular Cu(II) displaced 64Cu from intracellular pools, but exchange with intracellular Cu was not required for uptake. Zinc also displaced copper but less effectively than copper. This implies some specificity in intracellular binding components. No transstimulation of uptake or efflux was detected. Copper efflux was strictly passive. Extracellular Cu(II) decreased the rate of efflux and preloading inhibited Cu(II) uptake. Metabolic inhibitors had no effect on the rate or total amount of 64Cu efflux, and significant efflux occurred at 4 degrees C. Moreover, when a steady state was attained at the end of a preloading period, the effective intracellular concentration of free copper approximated the Cu(II) concentration in the extracellular medium. By use of this estimate for the intracellular concentration of free copper, efflux kinetic parameters were obtained from initial (1-min) rate data (Km = 5.5 +/- 0.8 microM; Vmax = 1.1 +/- 0.1 nmol X min-1 X mg prot-1). These parameters are similar to those obtained for Cu(II) uptake, and the saturation kinetics observed are consistent with specific facilitated efflux by this copper transport system.(ABSTRACT TRUNCATED AT 250 WORDS)

Mobilization of copper(II) from plasma components and mechanisms of hepatic copper transport.

The effects of plasma Cu(II) ligands on the kinetics of Cu(II) transport by rat liver parenchymal cells were determined to examine how Cu(II) is mobilized from plasma and transported into liver cells. Albumin markedly inhibited Cu(II) uptake at Cu(II)-to-albumin molar ratios of 3:1 or less. Kinetic analyses showed that albumin inhibits Cu(II) uptake by reducing the concentration of free Cu(II) in solution. Under conditions of excess albumin to Cu(II), histidine facilitated albumin-inhibited uptake of Cu(II). Threonine, glutamine, and most other amino acids were without effect. Moreover, the facilitation effect of a low-molecular-weight plasma fraction (less than or equal to 5,000) was largely accounted for by its histidine concentration. The tripeptide Gly-His-Lys also inhibited Cu(II) uptake into hepatocytes by the same mechanism as albumin. The inhibitory effects of albumin and Gly-His-Lys were additive with or without histidine. The active species in the Cu(II), albumin, and histamine mixtures was shown to be the His2Cu(II) complex. Vmax for this complex was identical to the Vmax for free Cu(II), but the Km was slightly higher [15 microM vs. 11 microM for free Cu(II)]. Concurrent determinations of [3H]-histidine and 64Cu(II) uptake showed that histidine was not transported with Cu(II) from His X Cu(II) or His2Cu(II) complexes. The data are consistent with histidine mobilizing Cu(II) from albumin by competing for Cu(II), interaction of the His2Cu(II) complex with the putative hepatic copper transport protein, and transport of copper as free ionic copper.

Kinetics of Cu(II) transport and accumulation by hepatocytes from copper-deficient mice and the brindled mouse model of Menkes disease.

The kinetics of 64Cu(II) uptake, efflux and net accumulation were examined with hepatocytes isolated from hemizygous and heterozygous brindled mice to determine the basis for the low hepatic copper levels in these mice. Initial rate data (0.5 min) for 64Cu(II) uptake showed no significant differences between the normal and mutant hepatocytes for the Km and Vmax parameters for copper uptake. The rate and total efflux of 64Cu from preloaded cells were also normal for both the hemizygous and heterozygous brindled hepatocytes. Normal and mutant hepatocytes exhibited an overshoot in net 64Cu(II) uptake when cells were incubated continuously with 64Cu(II) for 6 h. Maximal intracellular levels were attained at 1.5 to 2.5 h before a lower steady state was achieved by approximately 3 h. The mutant hepatocytes were also normal with respect to the time-delayed efflux process responsible for the overshoot in terms of 64Cu lost/min/mg of protein. However, both the hemizygous and heterozygous hepatocytes showed approximately 25% less than normal maximal copper accumulation. This accounted for the approximately 25% lower copper levels at steady state and also equaled the Cu deficiency in the resting whole livers. The impaired Cu accumulation characteristic is not a general property of Cu-deficient hepatocytes because nutritionally Cu-deficient hepatocytes exhibited higher than normal net 64Cu accumulation. The effect was Cu-specific since 65Zn(II) accumulation was normal with the mutant hepatocytes. The results are consistent with expression of the primary defect at the cellular level in the liver in the mouse mutants, and by inference, Menkes disease.

Copper transport kinetics by isolated rat hepatocytes.

Uptake and efflux of 64Cu were examined to determine whether hepatic parenchymal cells exhibit the kinetic criteria of a specific transport system for copper and related trace metals. Saturation kinetics were clearly indicated by both v versus [Cu] and 1/v versus 1/[Cu] plots (Km = 11 +/- 0.6 microM and Vmax = 2.7 nmol Cu X min-1 X mg prot-1). Identical results were obtained by cold-copper analyses, and contributions from simple diffusion or nonspecific binding were not detected. Virtually all of the accumulated 64Cu was intracellular by 0.5 min (the initial velocity period), with approximately 40% in the cytosolic fraction. Several related trace metals inhibited 64Cu uptake, but Ni(II) at a 10:1 molar excess did not. Zn(II) acted as a simple competitive inhibitor of 64Cu uptake (Ki = 16 microM). Efflux from preloaded cells was biphasic, with an initial rapid phase of approximately 5 min. Approximately 35% of preloaded 64Cu was transported out of the cells by 40 min, and little efflux occurred thereafter. Thus, hepatocytes exhibit saturation kinetics, competition by related substrates, and countertransport criteria of specific facilitated transport. A wide variety of metabolic inhibitors have no effect on 64Cu uptake under the same conditions that inhibit the active transport of bile acids. Specific inhibitor tests for electrogenic coupling were also negative. Because the identical kinetic parameters were obtained for free 64Cu and the 1:1 64Cu-histidine complex, it is inferred that copper is probably transported as the free ion. Cells incubated with greater than or equal to 10 microM 64Cu showed a net loss of copper after 40- to 60-min incubation, which may involve specific hepatic mechanisms in copper homeostasis.