Parathyroid hormone decreases renal vitamin D receptor expression in vivo

Abstract

The vitamin D receptor (VDR) is a nuclear transcription factor responsible for mediating the biological activities of 1,25-dihydroxyvitamin D₃ [1,25(OH)₂D₃]. Renal and parathyroid gland VDR content is an important factor in calcium homeostasis, vitamin D metabolism, and the treatment of secondary hyperparathyroidism and renal osteodystrophy. In these tissues, VDR expression is highly regulated by the calcium and vitamin D status. Although 1,25(OH)₂D₃ up-regulates VDR expression, hypocalcemia and vitamin D deficiency result in drastically reduced expression of the receptor. The generation of 25-hydroxyvitamin D₃-1α-hydroxylase-null mice, which are incapable of endogenously producing 1,25(OH)₂D₃, has allowed us to investigate the influence of parathyroid hormone (PTH) on VDR expression independent of PTH-mediated increases in 1,25(OH)₂D₃. Administration of human PTH (1-34) (110 μg/kg per day) for 48 h reduced renal VDR levels from 515 to 435 fmol/mg protein (15%, P < 0.03) in wild-type mice. In the 25-hydroxyvitamin D₃-1α-hydroxylase-null mice, PTH administration strongly reduced renal VDR levels, from 555 to 394 fmol/mg protein (29%, P < 0.001). These results demonstrate that PTH is a potent down-regulator of VDR expression in vivo.

Parathyroid hormone (PTH) and 1,25-dihydroxyvitamin D₃ [1,25(OH)₂D₃], the vitamin D hormone, are responsible for maintaining serum calcium values at sufficient levels for many physiological events, including proper neuromuscular function and bone mineralization. Their coordinated activities depend on several metabolic and functional interactions. In conditions of low serum calcium, the calcium-sensing receptor of the parathyroid gland stimulates PTH secretion (1). PTH increases circulating levels of 1,25(OH)₂D₃ by increasing expression of the renal 25-hydroxyvitamin D₃-1α-hydroxylase (1α-hydroxylase) gene (2, 3), which encodes the enzyme responsible for the production of the vitamin D hormone. 1,25(OH)₂D₃ exerts the biological actions of vitamin D by binding to the vitamin D receptor (VDR), a steroid/thyroid hormone nuclear receptor, and regulating the transcription of target genes. Through this mechanism, 1,25(OH)₂D₃ stimulates the intestinal absorption of calcium while 1,25(OH)₂D₃ and PTH act in concert to increase bone resorption and renal reabsorption of calcium (4). As a form of negative feedback regulation, 1,25(OH)₂D₃ decreases PTH expression by negatively regulating the PTH gene (5).

The biological response to 1,25(OH)₂D₃ is directly related to the VDR content of target tissues (6, 7). Thus, the regulation of receptor expression is a critical determinant of hormone activity. The VDR has been shown to be developmentally regulated, expressed in a tissue-specific manner, and regulated by a variety of physiological factors and hormones (8). It is well established that 1,25(OH)₂D₃ can stimulate receptor expression in the kidney and parathyroid gland while having minimal influence on intestinal VDR expression (9-13). Several studies also have noted a positive correlation between calcium and VDR expression. Hypocalcemia induced by dietary calcium restriction dramatically reduces renal VDR levels and prevents 1,25(OH)₂D₃ from increasing VDR expression (14-16). Reduced dietary calcium also has been associated with diminished VDR mRNA in avian and rat parathyroid gland (11, 12). If left untreated, renal failure with subsequent vitamin D deficiency and hypocalcemia will lead to a decrease in parathyroid gland VDR content and the development of vitamin D resistance (17, 18). The molecular trigger responsible for the hypocalcemia-mediated decline in renal and parathyroid gland VDR content is hereto-fore unknown.

PTH expression is inversely related to serum calcium values (19). Because hypocalcemic animals have high circulating levels of PTH and reduced renal and parathyroid gland VDR expression, we hypothesized that PTH may down-regulate VDR expression. Previous investigations into PTH-mediated regulation of VDR expression yielded mixed results. A PTH-mediated increase of VDR and VDR mRNA was reported in osteoblast-like UMR-106 cells (20, 21), whereas a PTH-mediated decrease in VDR and its transcript was reported in ROS 17/2.8 osteoblast cells (22). In vivo, osmotic PTH administration to rats nearly doubled renal VDR levels (23). However, these in vivo results were complicated by the PTH-mediated activation of the 1α-hydroxylase gene and the increase in circulating levels of 1,25(OH)₂D₃. Recently, our laboratory and others have generated 1α-hydroxylase-null (1α-hydroxylase^-/-) mice that are incapable of endogenously producing 1,25(OH)₂D₃ (24, 25). We have used these null mice to directly assess the effect of PTH on VDR expression, and from these studies, we have uncovered an additional functional interaction between PTH and the vitamin D endocrine system that is of both physiological and clinical significance.

Materials and Methods

Chemicals. 1,25(OH)₂D₃ was purchased from Tetrionics (Madison, WI). Human PTH (1-34) was obtained from Bachem.

Mice. Null mice were generated in-house (J.L.V., J.M.P., C. Kimmel-Jehan, M. Mendelsohn, E. W. Danielson, K.D.H., and H.F.D., unpublished data) and maintained on a C57BL/6 background. The majority of the 1α-hydroxylase gene coding region is obliterated in these mice, and no 1α-hydroxylase transcript is detectable by real-time RT-PCR of total kidney RNA. The mice grow, develop, and reproduce normally when maintained on a balanced diet supplemented with 1,25(OH)₂D₃ but quickly become rachitic when fed a diet devoid of 1,25(OH)₂D₃ and phosphorus. Administration of 1,25(OH)₂D₃ effectively reverses the rachitic condition; its biological precursor, 25-hydroxyvitamin D₃, is ineffective as a cure. For this study, wild-type (1α-hydroxylase^+/+) and 1α-hydroxylase^-/- offspring of 1α-hydroxylase heterozygotes were weaned onto a diet containing fat-soluble vitamins A, E, and K, with 0.47% calcium and 4 ng of 1,25(OH)₂D₃ per mouse per day (26). At 7 weeks of age, mice were transferred onto a 20% lactose/2.0% calcium/1.25% phosphorus diet supplemented with 1 ng of 1,25(OH)₂D₃ per mouse per day (27). One week later, mice were implanted with Alzet microosmotic pumps delivering either human PTH (1-34) (110 μg/kg per day) or a 97.9% saline/2.0% heat-inactivated serum/0.1% 1 N HCl vehicle. Mice were euthanized by using CO₂ 48 h later. Experimental protocols were reviewed and approved by the Research Animal Resources Center (University of Wisconsin).

Serum Analysis. After CO₂ euthanasia, blood was collected and centrifuged to obtain serum. For serum calcium analysis, serum was diluted 1:40 in 0.1% LaCl₃, and serum calcium was measured by using an atomic absorption spectrometer (model 3110, Perkin-Elmer). Serum phosphorus was quantified with a colorimetric assay by using malachite green (28).

Kidney Homogenate Preparation. Kidney whole-cell extract was prepared by using a modified method of Pierce et al. (29) and Sandgren and DeLuca (14). All steps were performed on ice or at 4°C. Kidneys were minced with a razor blade and washed twice with Tris·HCl EDTA DTT (TED)Na₁₅₀ containing a panel of protease inhibitors. The buffer was decanted and replaced with 1 vol (vol/vol) of TED plus inhibitors. After homogenization with a Tissue-Tearor (BioSpec Products, Bartlesville, OK), 1 vol of TEDK₆₀₀Mg₂₀ was added, and homogenization was repeated. Samples were centrifuged at 20,000 × g for 1 h. Supernatant was divided into aliquots, frozen under liquid nitrogen, and stored at -80°C until analysis. The buffers contained 50 mM Tris·HCl (pH 7.4), 1.5 mM EDTA, and 5 mM DTT. Either 150 mM NaCl or 600 mM KCl/20 mM MgCl₂ was added where appropriate. The panel of protease inhibitors (Sigma) consisted of 150 μM aprotinin, 130 μM bestatin, 10 μM leupeptin, 1 μM pepstatin A, and 1 mM phenylmethylsulfonyl fluoride.

Analysis of Kidney Homogenate. VDR content was determined by using an ELISA developed in this laboratory (30). The protein concentration of the homogenates was determined by the method of Bradford (31), using BSA as a standard.

RNA Isolation. Total RNA was isolated from mouse kidney with Tri Reagent (Molecular Research Center, Cincinnati) according to the manufacturer's protocol.

Quantitative RT-PCR. RNA was DNase-treated (RQ-1 RNase-Free DNase, Promega) and then reverse-transcribed by using a first-strand synthesis system for RT-PCR (SuperScript, Invitrogen). Real-time PCR was performed in a LightCycler (Roche Diagnostics) according to the manufacturer's recommendations. SYBR green dye (Roche Applied Science) was used for quantification of double-stranded DNA after every cycle. The following primer sequences were used: β-actin, (forward) 5′-TGGAATCCTGTGGCATCCATGAAAC-3′ and (reverse) 5′-TAAAACGCAGCTCAGTAACAGTCCG-3′; 1α-hydroxylase, (forward) 5′-CCGCGGGCTATGCTGGAAC-3′ and (reverse) 5′-CTCTGGGCAAAGGCAAACATCTGA-3′; and VDR, (forward) 5′-CTCCTCGATGCCCACCACAAGACCTACG-3′ and (reverse) 5′-GTGGGGCAGCATGGAGAGCGGAGACAG-3′. All amplicons were 200-500 bp in length and were sequenced to confirm specificity of amplification. Throughout real-time PCR analysis, product identities were confirmed by melting curve analysis. The quantification of each gene was relative to a standard curve generated from a serially diluted sample. The relative amount of each experimental gene was then normalized to the abundance of the housekeeping gene β-actin. These values were standardized such that a maximum value of 1.0 was assigned to the group with the highest gene expression.

Statistical Analysis. The two-tailed Student t test was used to quantify statistical differences between experimental groups. Results are expressed as the mean ± SE.

Results

PTH Increases Serum Calcium and Decreases Serum Phosphorus. To investigate the effect of PTH on VDR expression, 1α-hydroxylase^+/+ and 1α-hydroxylase^-/- mice were implanted with an osmotic pump that delivered human PTH (110 μg/kg per day) for 48 h. Mice were maintained on a high-calcium diet to minimize circulating levels of endogenous PTH. Osmotic PTH administration dramatically increased the serum calcium levels in both the wild-type and null mice (Fig. 1A). This increase in serum calcium can be attributed to a PTH-stimulated increase in bone catabolism (32). The ability of PTH to stimulate renal phosphorus excretion resulted in significantly reduced serum phosphorus levels (Fig. 1B). Combined, the serum calcium and phosphorus data illustrate the biological activity of the administered PTH.

Fig. 1.

Serum calcium and phosphorus. 1α-Hydroxylase^+/+ and 1α-hydroxylase^-/- mice were maintained on a 20% lactose diet containing 2.0% calcium, 1.25% phosphorus, and 1 ng of 1,25(OH)₂D₃ per mouse per day. Either PTH (110 μg/kg per day) or vehicle was administered by microosmotic pump for 48 h. At death, blood was collected and serum was isolated. Assays were performed as described in Materials and Methods. (A) Serum calcium. (B) Serum phosphorus. All values are reported as the mean ± SEM (n = 10). ^*, significance at P < 0.005 vs. vehicle-treated mice; †, significance at P < 0.03.

PTH Dramatically Up-Regulates 1α-Hydroxylase mRNA in Wild-Type Mice. The ability of PTH to stimulate 1,25(OH)₂D₃ production was demonstrated by Garabedian et al. in 1972 (2). Since then, the mechanism by which PTH activates 1α-hydroxylase gene expression to stimulate production of the vitamin D hormone has been delineated (3, 33). To monitor the bioactivity of the administered PTH on transcriptional regulation and verify the absence of the 1α-hydroxylase gene product in the null mice, we quantified renal 1α-hydroxylase mRNA levels. As shown in Fig. 2, the 1α-hydroxylase mRNA was expressed at a basal level in the 1α-hydroxylase^+/+ mice. PTH administration to these mice caused a 25-fold increase in 1α-hydroxylase transcript levels. As expected, the 1α-hydroxylase transcript was not detectable in either the vehicle or PTH-treated 1α-hydroxylase^-/- mice.

Fig. 2.

1α-Hydroxylase transcript levels. 1α-Hydroxylase^+/+ and 1α-hydroxylase^-/- mice were maintained on a 20% lactose diet containing 2.0% calcium, 1.25% phosphorus, and 1 ng of 1,25(OH)₂D₃ per mouse per day. Either PTH (110 μg/kg per day) or vehicle was administered by microosmotic pump for 48 h. One kidney per mouse was harvested into Tri Reagent, and total RNA was prepared. After reverse transcription, 1α-hydroxylase mRNA was quantified by using real-time PCR, and standardized to β-actin mRNA. (Top) Values are charted as relative mRNA with a maximal value of 1. Values are reported as the mean ± SEM (n = 10). ^*, significance at P < 0.001 vs. vehicle-treated mice. N.D., not detectable. (Middle and Bottom) A representative sample from each group was electrophoresed through an ethidium bromide-stained gel and was UV-illuminated. M.W., pGEM (Promega) molecular weight markers. Lane 1, +/+, vehicle; lane 2, +/+ PTH; lane 3, -/- vehicle; lane 4, -/- PTH.

PTH Reduces Renal VDR Levels in Vivo. After 48 h of continuous PTH administration, renal VDR levels were quantified with an ELISA assay. PTH caused a modest (15%) yet significant (P < 0.03) drop in the renal VDR levels of the wild-type mice (Fig. 3). In the 1α-hydroxylase^-/- mice, PTH caused a dramatic 29% decrease in VDR expression (P < 0.001). In wild-type animals, PTH administration increases circulating levels of 1,25(OH)₂D₃ by activating expression of the anabolic 1α-hydroxylase enzyme. 1,25(OH)₂D₃ is a known up-regulator of renal VDR expression (9, 10, 15, 16). The greater PTH-mediated suppression of VDR expression in the 1α-hydroxylase^-/- mice can be explained by the inability of PTH to stimulate 1,25(OH)₂ D₃ production in these animals. Overall, these results clearly demonstrate that PTH is a potent down-regulator of VDR expression in vivo.

Fig. 3.

Suppression of renal VDR by PTH. 1α-Hydroxylase^+/+ and 1α-hydroxylase^-/- mice were maintained on a 20% lactose diet containing 2.0% calcium, 1.25% phosphorus, and 1 ng of 1,25(OH)₂D₃ per mouse per day. PTH (110 μg/kg per day) or vehicle was administered by microosmotic pump for 48 h. One kidney per mouse was harvested into isotonic buffer, whole-cell extract was prepared, and VDR content was determined by ELISA and standardized to total protein, as described in Materials and Methods. Each group contained 10 mice. The values represent the mean ± SEM. ^*, significance at P < 0.03 vs. vehicle-treated mice; †, significance at P < 0.001.

Renal VDR Transcript Levels Are Modestly Decreased by PTH. As an initial investigation into the mechanism by which PTH decreases renal VDR levels, real-time PCR was performed to quantify the renal VDR transcript levels. Although the differences were not statistically significant, PTH administration resulted in a trend of reduced VDR mRNA levels both in wild-type and null mice (Fig. 4). The data suggest that the suppression of VDR by PTH may be signaled through repression of VDR transcript expression, but other mechanisms cannot be excluded.

Fig. 4.

VDR transcript levels. 1α-Hydroxylase^+/+ and 1α-hydroxylase^-/- mice were maintained on a 20% lactose diet containing 2.0% calcium, 1.25% phosphorus, and 1 ng of 1,25(OH)₂D₃ per mouse per day. Either PTH (110 μg/kg per day) or vehicle was administered by microosmotic pump for 48 h. One kidney per mouse was harvested into Tri Reagent, and total RNA was prepared. After reverse transcription, VDR mRNA was quantified by using real-time PCR and was standardized to β-actin mRNA. The mean ± SEM (n = 10) is expressed as relative mRNA with a maximal value of 1.

Discussion

Hypocalcemia as a result of dietary calcium restriction or vitamin D deficiency results in a reduction of renal and parathyroid gland VDR (14, 17). Because serum calcium levels are inversely proportional to serum PTH, we hypothesized that PTH may suppress VDR expression in vivo. Previous investigations into the PTH regulation of receptor content were complicated by the PTH-mediated activation of the 1α-hydroxylase gene and subsequent increase in circulating levels of 1,25(OH)₂D₃, a known up-regulator of VDR expression. The development of transgenic mice with targeted ablation of the 1α-hydroxylase gene has allowed us to separate the effects of PTH and 1,25(OH)₂D₃ and investigate the in vivo effect of PTH administration on VDR content. PTH administration by osmotic pump for 48 h resulted in the anticipated increase in serum calcium and decrease in serum phosphorus (Fig. 1). PTH treatment drastically up-regulated the 1α-hydroxylase mRNA levels of the wild-type mice, whereas these transcripts were not detectable in the 1α-hydroxylase^-/- mice (Fig. 2). As shown in Fig. 3, administration of PTH to wild-type mice resulted in a significant reduction of renal VDR expression (15%, P < 0.03). A more striking decline of ≈30% (P < 0.001) was observed for the 1α-hydroxylase^-/- mice. In rats, PTH administration has been reported to increase renal VDR levels 1.8-fold, a change that was attributed to increased circulating levels of 1,25(OH)₂D₃ (23). A coinfusion experiment in which PTH partially blocked a 1,25(OH)₂D₃-mediated increase in renal VDR was described in the same report. This result suggested that PTH may have suppressive actions on VDR expression. The use of 1α-hydroxylase^-/- mice has allowed us to separate the effects of PTH and 1,25(OH)₂D₃, and clearly demonstrate that PTH suppresses VDR expression in vivo.

The hypocalcemia and PTH-mediated decrease in VDR expression is of physiological and medical significance. The kidney is an important tissue for the regulation of the metabolism of 1,25(OH)₂D₃, because both the anabolic 1α-hydroxylase and the catabolic 25-hydroxyvitamin D₃-24-hydroxylase (24-hydroxylase) are expressed in the renal proximal tubular cells. The PTH-mediated down-regulation of renal VDR may be a protective measure to block 1,25(OH)₂D₃-mediated suppression of the 1α-hydroxylase and induction of the 24-hydroxylase, resulting in a net increase in serum 1,25(OH)₂D₃ levels that would be beneficial to a hypocalcemic animal (34).

In advanced renal failure, treatment with calcitriol or vitamin D analogs is used to prevent secondary hyperparathyroidism and renal osteodystrophy (35). Without early intervention, hypocalcemia and parathyroid hyperplasia preface a reduction in parathyroid gland VDR content and vitamin D resistance (18). The mechanism established in this report, PTH-mediated suppression of VDR expression, is likely responsible for this reduction of parathyroid gland VDR content. Our findings emphasize the importance of early treatment with vitamin D compounds in renal failure patients to prevent vitamin D resistance.

Although parathyroid gland and renal receptor content is diminished in hypocalcemic conditions, studies on rats and mice have concluded that intestinal VDR expression is not significantly affected (10, 13). This constitutive VDR expression may reflect important noncalcemic actions of intestinal VDR, such as the receptor-mediated bile acid detoxification recently suggested by Makishima et al. (36). The tissue specificity of hypocalcemia-mediated suppression of VDR expression may be related to PTH responsiveness; the lack of PTH receptor in intestinal cells may allow them to escape regulation by PTH and constitutively express VDR. Notably, Meyer et al. (37) reported that a low-calcium diet repressed intestinal VDR mRNA expression in chickens, suggesting that a factor other than PTH may be repressing VDR expression in this model.

In addition to its classical role in regulating calcium homeostasis, 1,25(OH)₂D₃ also inhibits cellular proliferation, stimulates cell differentiation, and influences the immune response (38). The immunosuppressive properties of 1,25(OH)₂D₃ are being investigated for therapeutic intervention in several autoimmune diseases. Interestingly, dietary calcium restriction drastically reduced the effectiveness of 1,25(OH)₂D₃ in preventing disease onset in experimental autoimmune encephalomyelitis (39) and in reducing diabetes incidence in the nonobese diabetic mouse (J. B. Zella and H.F.D., unpublished data). The influence of PTH and calcium status on lymphocyte VDR expression merits investigation.

The mechanism by which PTH down-regulates VDR expression is not clear. Although we observed a trend toward reduced VDR mRNA expression in PTH-treated mice, the reductions were not statistically significant (Fig. 4). Previous investigations into the calcium and 1,25(OH)₂D₃-mediated regulation of renal receptor content unveiled dramatic transcriptional regulation, because the VDR mRNA was expressed at a basal level in the absence of either calcium or 1,25(OH)₂D₃ (13, 16). The identification of PTH as the trigger responsible for hypocalcemia-mediated suppression of renal VDR enables focused research into the general mechanism, signaling molecules, and pathways responsible for PTH-mediated down-regulation of VDR content. Because of the significance of receptor regulation, further experiments must be performed to elucidate this mechanism and to identify the cells and tissues responsive to PTH regulation.