INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

The reabsorption of inorganic phosphate (P_i) in the renal proximal tubule plays a key role in overall P_i homeostasis (1, 2). 1,25-Dihydroxyvitamin D₃ (1,25-(OH)₂D₃)¹ regulates P_i homeostasis in the bone, intestine, and kidney (2). However, the effect of 1,25-(OH)₂D₃ on the reabsorption of P_i in the kidney remains unclear. Contradictory results showing an increase or decrease in P_i excretion in response to 1,25-(OH)₂D₃ have been reported (2). The results may be due to differences in the mode of action (genomic action versus non-genomic action) of 1,25-(OH)₂D₃, and/or to differences in the experimental conditions including the time of exposure, dose of 1,25-(OH)₂D₃, and previous status of vitamin D and parathyroid hormone (PTH) (2).

A study using a micropuncture technique, in situ microperfusion, isolated perfused tubules, and primary cell cultures revealed an axial heterogeneity in proximal tubular P_i transport (2). The extent of Na⁺-P_i co-transport is greater in the proximal convoluted tubules (PCTs) than in the proximal straight tubules (PSTs) (3-8). Kinetic studies have shown that the greater P_i transport in the PCT is attributable to a higher V_max of Na⁺-P_i co-transport (6-8). Several Na⁺-P_i co-transporters have been isolated from the kidney cortex of various species (9-11). They have been classified into two different types on the basis of their predicted amino acid sequences: type I, which includes NaPi-1 (rabbit), NPT-1 (human), Npt-1 (mouse), and RNaPi-1 (rat); and type II, which includes NaPi-2/3 (rat, human), NaPi-4 (OK cell), NaPi-6 (rabbit), and NaPi-7 (mouse) (9-11). Both types of Na⁺-P_i co-transporters are localized in PCTs and PSTs. With the use of polyclonal antibodies and cDNA probes, the regulation of the expression of the rat renal type I and type II transporters by several physiological factors has been studied (9-11). The type II transporter was found to be regulated mainly by dietary P_i and PTH. In addition, the regulation of the type II transporter by dietary P_i and PTH was shown to differ between PSTs and PCTs (12, 13). In contrast, insulin and glucose can affect the expression of the rat type I transporter (14).

To clarify the action of 1,25-(OH)₂D₃ on renal P_i transport, we have now examined the regulation by 1,25-(OH)₂D₃ of the expression of the type II phosphate transporter NaPi-2 at the mRNA and protein levels in the rat kidney cortex. We also characterized the promoter of the human type II phosphate transporter NaPi-3 gene with regard to transcriptional regulation by the vitamin D receptor (VDR), because we and another group demonstrated that the structures of the type II Na-P_i transporters (rat NaPi-2, human NaPi-3 and mouse NaPi-7) gene are highly conserved (15, 16). In addition, 1,25-(OH)₂D₃ is known to affect renal Na-P_i co-transport activity in these three species (2).

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Vitamin D-deficient Animals-- Male Wistar rats (age, 3 weeks; body weight, ~40 g) purchased from Japan SLC (Shizuoka, Japan) and fed a vitamin D-free diet (Diet 11) ad libitum for 6 weeks and subsequently a vitamin D-free and calcium-free diet (Diet 11-Ca) for 1 week (17). The rats with low plasma calcium and undetectable levels of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D obtained thus were subjected to the experiments. For repletion, the deficient-rats were intravenously injected with 1,25-(OH)₂D₃ (6.25 µg (15 nmol, 2500 IU) per kilogram of body weight) dissolved in ethanol-propylene glycol (1:4, v/v) (18).

Preparation of Brush-border Membrane Vesicles (BBMVs) and Transport Measurements-- Kidneys were sliced horizontally in 3-mm sections. The outer 3-mm portion of the cortex (superficial cortex) and the inner cortex (juxtamedullary cortex), including the outer-most portion of red medulla were used for the preparation of BBMVs (19). The enzyme activity profiles of these cortex preparations indicate that they correspond to BBMVs of PCTs and PSTs, respectively. The BBMVs were prepared by the Ca²⁺ precipitation method as described previously (20), and their purity was assessed by the measurement of the leucine aminopeptidase, Na⁺- and K⁺-dependent ATPase, and cytochrome-c-oxidase activities (21). The uptake of [³²P]P_i was measured by a rapid filtration technique (21) with transport solution (100 mM NaCl, 100 mM mannitol, 20 mM Hepes-Tris (pH 7.5), and 0.01 to 10 mM KH₂³²PO₄ (9000 Ci/mmol; NEN Life Science Products, Boston, MA)).

Northern Blot Analysis-- Total RNA was isolated from the kidney tissue by IsoGen RNA extraction regent (Nippon Gene, Tokyo). Total RNA was separated by electrophoresis with a 1.2% agarose gel containing 2.2 M formaldehyde. The resolved RNA was transferred to a Hybond-N⁺ membrane (Amersham, Buckinghamshire, United Kingdom). The hybridization and washing and the analysis of the data were carried out essentially as described previously (22). Rat type I transporter rNaPi-1 (nucleotides 58 to +356, relative to the translation start site (23)) and type II transporter NaPi-2 (nucleotides +543 to +1639, relative to the translation start site (24)) cDNA probes were prepared by polymerase chain reaction with kidney total cDNA and specific oligonucleotide primers. The probes were labeled with [-³²P]dCTP (110 TBq/mmol; ICN, Costa Mesa, CA) with the use of a Megaprime labeling system (Amersham).

Immunoblot Analysis-- For the immunoblot analysis, BBMVs were prepared as described above and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The separated proteins were transferred electrophoretically to a Hybond ECL nitrocellulose membrane (Amersham). The membrane was treated with non-fat dried milk and 1:1,000 diluted anti-NaPi-2 antibody prepared previously (24). The membrane was also treated with horseradish peroxidase-conjugated anti-rabbit IgG as the secondary antibody. The signal was detected by an enhanced chemiluminescence (ECL) system (Amersham).

Immunohistochemistry-- Rats were anesthetized with pentobarbital and transcardially perfused with saline followed by 4% (w/v) paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4). Both kidneys were removed, immersed in fixative for 15 h, and processed for the preparation of cryostat sections. After microwave irradiation (for 10 min in 10 mM citrate buffer, pH 6.0) and treatment with hydrogen peroxide, the sections were exposed overnight at 4 °C to the NaPi-2-specific antibodies (1:2,000 dilution). Visualization was achieved by incubation with Cy3-labeled goat anti-rabbit IgG (Chemicon, Temecula, CA) for 2 h at 37 °C. The sections were observed with a confocal laser scanning microscope (TCS-4D, Leica, Bensheim, Germany).

Cell Culture-- COS-7 cells (RIKEN Cell Bank, Saitama, Japan, (25)) were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Rat osteosarcoma ROS-17/2.8 cells (RIKEN Cell Bank, (26)) were maintained in Ham's F-12 medium supplemented with 10% fetal bovine serum. Both cell lines were cultured at 37 °C under a humidified atmosphere containing 5% CO₂.

Plasmid Construction-- Two luciferase reporter vectors (p3P2400 and p3P1260) containing the 5'-flanking region of the human NaPi-3 gene were described in a previous study (15). The luciferase reporter plasmids pTKDRCF, pTKDRCR, pTKDRCmt2, and pTKDRCmt3 containing the DR-C sequence of the NaPi-3 gene promoter in the forward and reverse directions, respectively, were constructed by placing synthetic double-stranded DNA (annealed oligonucleotides with XhoI cleavage sites (5'-TCGAGATCAGGGGCAGCAAGGGCAGAAATG-3' and 5'-TCGACATTTCTGCCCTTGCTGCCCCTGATC-3') and with the mutation bases indicated in Table I for pTKDRCmt2 and pTKDRCmt3) corresponding to nucleotides 1982 to 1957 (relative to the transcription start site) of NaPi-3 upstream of the minimum promoter region of the herpes simplex virus-thymidine kinase gene (kindly provided by H. Kondo) (27) in the pGL3 vector (Promega, Madison, WI). Two additional reporter plasmids (pTKDRCmt2 and pTKDRCmt3) were also constructed by the above method using synthetic oligonucleotides containing mutations as described in Table I, respectively. A p3P1850TK was constructed with KpnI-SacI-digested polymerase chain reaction-amplified DNA fragment with the following two primers: 5'-CGGGATCCAGGCTGGTCTCGAACTCC-3' (corresponding to nucleotides 2121 to 2102, with an added KpnI clevage site), and 5'-ATTGCTCCAGGAGCTC-3' (corresponding to nucleotides 1859 to 1843, contains the SacI cleavage site), the herpes simplex virus-thymidine kinase minimum promoter, and pGL-3 vector. A human VDR expression vector (28) was constructed by subcloning an EcoRI fragment containing the full-length human VDR cDNA into pcDL-SR-296 (29). The -galactosidase expression vector pCMV- (CLONTECH, Palo Alto, CA) was used as an internal control. Each plasmid was purified with a plasmid purification kit (QIAGEN, Hilden, Germany).

Transfection of COS-7 and ROS-17/2.8 Cells-- COS-7 cells (1.5 × 10⁵) were transferred to a 35-mm plastic dish and transfected with 0.5 µg of luciferase reporter vector, 0.05 µg of human VDR expression vector, and 0.5 µg of pCMV- with the use of TransIT-LT1 lipofection reagent (Pan Vera Corp., Madison, WI). ROS-17/2.8 cells (2.0 × 10⁵), also in 35-mm dishes, were transfected with 0.5 µg of luciferase reporter vector and 0.5 µg of pCMV-, again with the use of TransIT-LT1. After transfection, the cells were incubated under standard conditions for 24 h and then exposed to 1,25-(OH)₂D₃ for 15 h. The cells were then harvested in cell lysis buffer supplied with a luciferase assay kit (Pica-gene; Toyo Ink, Tokyo), and the lysates were assayed for luciferase activity, -galactosidase activity, and protein concentration (30).

Preparation of Nuclear Extracts from COS-7 Cells-- Nuclear extracts from COS-7 cells were prepared as described by Arakawa et al. (30), using a slight modification of the method established by Dignam et al. (31). COS-7 cells were transfected with the human VDR expression vector by the DEAE-dextran method (32).

In Vitro Synthesis of VDR and Retinoid X Receptor-- The human VDR expression vector pSG5/hVDR (28) and the murine RXR expression vector pSG5/m RXR (kindly provided by P. Chambon) were used to synthesize the encoded protein in vitro protein synthesis system (Single Tube Protein System (STP) 2, Novagen, Madison, WI). The 50-µl reaction mixture contained 0.5 µg of expression vector, STP System 2 Transcription Mix, and STP System 2 Translation Mix, and 25 µM methionine.

Electrophoretic Mobility Shift Assay (EMSA)-- Eleven double-stranded oligonucleotides corresponding to direct repeat-like sequences in the 5'-flanking region of the NaPi-3 gene (15), human osteocalcin gene (33), and rat 25-hydroxyvitamin D-24-hydroxylase gene (34), as well as various mutant sequences of the DR-C region of the NaPi-3 gene were synthesized (Table I). The oligonucleotides were purified electrophoretically on a 15% polyacrylamide gel and labeled by T4 polynucleotide kinase with [-³²P]ATP (167 TBq/mmol; ICN). The binding reaction was performed for 30 min at room temperature in a final volume of 20 µl containing 1 µg of poly(dI-dC) (Pharmacia, Uppsala, Sweden), 20 mM Hepes-KOH (pH 7.9), 100 mM KCl, 0.1 mM EDTA, 10% (v/v) glycerol, 5 µg of nuclear extract or 2 µl of in vitro synthesized protein pretreated with 1,25-(OH)₂D₃, and 25 fmol of probe (1 × 10⁵ cpm). The reaction mixture was then subjected to electrophoresis on a 6% polyacrylamide gel with 1 × TAE (40 mM Tris-HCl, 40 mM acetic acid, 1 mM EDTA) as electrode buffer at a constant current of 30 mA for 2 h. The gel was dried and analyzed with a bio-imaging analyzer (BAS-1500, Fuji-film, Tokyo).

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Na⁺-dependent P_i Uptake-- BBMVs were prepared from the superficial and juxtamedullary cortex of rat kidneys. The P_i uptake remained linear for up to 3 min in both types of vesicles (data not shown). The initial rate of P_i uptake was greater in the BBMVs from the superficial cortex than in those from the juxtamedullary cortex (785 ± 164 and 554 ± 121 pmol/mg of protein per min, respectively (means ± S.E., n = 6)) of normal rats (Fig. 1A). In the vitamin D-deficient animals, the initial rate of P_i uptake in BBMVs from the juxtamedullary cortex was substantially decreased (215 ± 49 pmol/mg of protein/min) whereas that in the vesicles from the superficial cortex was slightly increased (987 ± 121 pmol/mg of protein/min). Forty-eight hours after the injection of 1,25-(OH)₂D₃ into vitamin D-deficient rats, the initial rate of P_i uptake was ~160% (897 ± 143 pmol/mg of protein/min) and ~50% (398 ± 115 pmol/mg of protein/min) of the values of normal animals for BBMVs from the juxtamedullary and superficial cortex, respectively (Fig. 1B).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effects of vitamin D deficiency on Na+-Pi co-transport activity. A, Na+/Pi co-transport activity in BBMVs from the superficial (SC) and juxtamedullary (JM) cortexes of vitamin D-deficient and normal rats. B, effect of 1,25-(OH)2D3 on Na+-Pi co-transport activity in vitamin D-deficient rats. Vitamin D-deficient rats were injected intravenously with 1,25-(OH)2D3 (6.25 µg/kg) and killed at various times thereafter.

Northern Blot Analysis-- The amounts of NaPi-2 mRNA (~2.4 kilobase) and protein (~90-110 kDa) did not differ between the superficial and juxtamedullary cortexes of the normal rats (data not shown). The amounts of NaPi-2 mRNA and protein in the juxtamedullary cortex were markedly decreased in the vitamin D-deficient animals (Fig. 2, A and B). In contrast, the type I transporter rNaPi-1 mRNA and protein levels were not significantly different between the normal and vitamin D-deficient animals. In addition, the levels of neutral basic amino acid transporter mRNA and protein were not changed.

View larger version (67K): [in this window] [in a new window]   Fig. 2.   Messenger RNA and protein levels in the superficial and juxtamedullary cortexes of vitamin D-deficient rat kidney. A, Northern blot analysis of total RNA (20 µg) from the superficial (SC) and juxtamedullary (JM) cortexes of vitamin D-deficient and normal animals with 32P-labeled cDNA rNaPi-1 (type I), NaPi-2 (type II), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probes. B, Western blot analysis of BBMVs from vitamin D-deficient and normal animals, using specific antibodies. Cont, normal rats; Vit D, vitamin D-deficient rats; NBAT, neutral basic amino acid transporter.

View larger version (75K): [in this window] [in a new window]   Fig. 3.   Time course of mRNA and protein levels in the superficial (SC) and juxtamedullary (JM) cortexes of vitamin D-deficient rats after the administration of 1,25-(OH)2D3. Vitamin D-deficient rats were injected intravenously with 1,25-(OH)2D3 (6.25 µg/kg) and killed at various times thereafter. The amounts of mRNA (A) and protein (B) were then determined for the superficial and juxtamedullary cortexes as in Fig. 2. Data are representative of three separate experiments.

Immunohistochemistry-- In the normal animals, the localization of NaPi-2 immunoreactivity in the apical membrane of tubular cells was shown in the juxtamedullary and superficial cortexes (Fig. 4, A and E). In the vitamin D-deficient animals, the intensity of NaPi-2 immunoreactivity was decreased (Fig. 4, B and F), to a greater extent in the juxtamedullary cortex than in the superficial cortex. The administration of 1,25-(OH)₂D₃ to vitamin D-deficient rats resulted in the gradual but slight diminishment of NaPi-2 immunoreactivity from the superficial cortex (Fig. 4, C and D) and a gradual increase in the amount of NaPi-2 in the juxtamedullary region (Fig. 4, G and H).

View larger version (65K): [in this window] [in a new window]   Fig. 4.   Immunohistochemical analysis of the effect of the administration of 1,25-(OH)2D3 on the expression of NaPi-2 immunoreactivity in the kidney of vitamin D-deficient rats. A-D, NaPi-2 immunoreactivity in the superficial region of the kidney cortex of a normal rat (A) and vitamin D-deficient rats (B-D) 0, 12, and 48 h, respectively, after the administration of 1,25-(OH)2D3. E-H, NaPi-2 immunoreactivity in the juxtamedullary nephrons of a normal rat (E) and vitamin D-deficient rats (F-H) 0, 12, and 48 h, respectively, after the administration of 1,25-(OH)2D3.

Identification of a Functional Vitamin D-responsive Element (VDRE) in the 5'-Flanking Region of the NaPi-3 Gene-- To further characterize the effect of 1,25-(OH)₂D₃ on the expression of a type II co-transporter gene, we performed a functional analysis of the human NaPi-3 gene promoter in COS-7 cells expressing the human VDR. 1,25-(OH)₂D₃ was previously shown to stimulate the transcriptional activity of this promoter in COS-7 cells (15). The luciferase activity of COS-7 cells expressing VDR and transfected with the luciferase reporter vector p3P2400, which contains 2462 base pairs (nucleotides 2409 to +53) of the NaPi-3 gene and all three direct repeat-like motifs (DR-A, DR-B, and DR-C) present in the promoter, increased markedly up on the exposure of the cells to 1,25-(OH)₂D₃. In contrast, the luciferase activity in the cells transfected with p3P1260, which contains 1312 base pairs (nucleotides 1259 to +53) of the NaPi-3 gene but lacks DR-C, was not affected by 1,25-(OH)₂D₃ (Fig. 5). In addition, the luciferase activity of the cells transfected with p3P1850TK, which lacks 1854 to +53, but contains the minimum promoter of herpes simplex virus-thymidine kinase, was increased by the exposure to 1,25-(OH)₂D₃. The human type I co-transporter NPT-1 gene promoter (nucleotides 1414 to +109) did not respond to 1,25-(OH)₂D₃ (data not shown).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Deletion analysis of the human NaPi-3 gene promoter. COS-7 cells were transfected with 0.5 µg of a NaPi-3 reporter plasmid (p3P2400, p3P1260, p3P1850TK, pTK, pTKDRCF, pTKDRCR, pTKDRCmt2, and pTKDRCmt3), 0.05 µg of VDR expression vector, and 0.5 µg of pCMV-. The construction of each plasmid is described under "Experimental Procedures." Cells were incubated in control medium for 24 h and then in the presence (closed columns) or absence (open columns) of 50 nM 1,25-(OH)2D3 for 15 h. The cells were harvested and assayed for luciferase activity, which was corrected for differences in transfection efficiency by normalization with -galactosidase activity. Data are expressed as fold induction by 1,25-(OH)2D3 and are mean ± S.E. from triplicate determinations. Similar results were obtained in two additional experiments. *, p < 0.02.

Binding of VDR-RXR Heterodimer to the VDRE of the NaPi-3 Gene-- The VDRE of the NaPi-3 gene was investigated further by an EMSA with various oligonucleotides as probes and competitors (Table I). The EMSA demonstrated the formation of a complex between an oligonucleotide containing the VDRE of the rat 25-hydroxyvitamin D-24-hydroxylase gene promoter and the VDR-RXR heterodimer (Fig. 6A). The formation of this complex was inhibited in the presence of DR-C but not in the presence of DR-A or DR-B oligonucleotides. An oligonucleotide containing DR-C formed a complex with the VDR-RXR heterodimer but not with either VDR or RXR alone (Fig. 6B).

View larger version (54K): [in this window] [in a new window]   Fig. 6.   EMSA of the interaction of the VDRE of the NaPi-3 gene promoter and VDR and RXR. A, EMSA with an oligonucleotide containing the VDRE of the rat 25-hydroxyvitamin D-24-hydroxylase gene as the probe and in vitro synthesized VDR and RXR. Oligonucleotides containing DR-A, DR-B, or DR-C of the NaPi-3 gene promoter were added as competitors at a 50-fold molar excess relative to the probe. B, EMSA with the DR-C oligonucleotide as the probe, performed in the presence of 4 µl of reticulocyte lysate (VDR(), RXR()), 2 µl of reticulocyte lysate and 2 µl of in vitro-synthesized RXR (VDR(), RXR(+)), 2 µl of reticulocyte lysate and 2 µl of in vitro synthesized VDR (VDR(+), RXR()), or 2 µl of in vitro synthesized RXR and 2 µl of in vitro VDR (VDR(+), RXR(+)). In the absence of VDR and RXR, a major band was detected in the EMSAs. This results from the nonspecific binding of DR-C to endogenous proteins in rabbit reticulocyte lysate.

EMSA Revealed the Formation of a Prominent DNA-Protein complex-- The DNA-protein complex could be observed in EMSA using the DR-C oligonucleotide as a probe and the nuclear extract of COS-7 cells expressing human VDR (Fig. 7). The formation of this complex was inhibited in the presence of either an oligonucleotide containing the VDRE of the human osteocalcin gene promoter (nucleotides 501 to 483) (33), or a monoclonal antibody (9A7) to chick VDR which could inhibit the VDR-DNA complex formation described previously (35). In addition, we compared the binding affinity of the VDREs of the human osteocalcin and NaPi-3 genes. The binding affinity of the NaPi-3 VDRE was slightly but not significantly lower than that of the osteocalcin VDRE (data not shown). In addition, neither competition or inhibition were observed by AP-1 oligonucleotide and c-Fos antibodies.

View larger version (89K): [in this window] [in a new window]   Fig. 7.   EMSA analysis with oligonucleotide DR-C and nuclear extract of COS-7 cells expressing human VDR. EMSAs were performed with 32P-labeled oligonucleotide DR-C as the probe and the nuclear extract of COS-7 cells expressing human VDR. An oligonucleotide containing the VDRE of the human osteocalcin gene (hOC-VDRE) was incubated with nuclear extract for 30 min at 20 °C before the addition of probe, where indicated. A monoclonal antibody (9A7) to chick VDR (anti-VDR Ab) and a monoclonal antibody to c-Fos (anti-c-Fos; purchased from Santa Cruz Biotech Inc., Santa Cruz, CA) as controls were incubated with nuclear extract for 2 h at 4 °C before the addition of the probe and poly(dI-dC), where indicated. An AP-1 consensus oligonucleotide (5'-CGCTTGATGACTCAGCCGGAA-3', Santa Cruz Biotech Inc.) was used as the control for the competition analysis.

Mutation Analysis of the DR-C Sequence of the NaPi-3 Gene Promoter-- To confirm that the DR-C sequence 5'-GGGGCAGCAAGGGCA-3' is the target sequence of the RXR-VDR heterodimer, we performed EMSAs with oligonucleotides containing specific mutations of this sequence as competitors of DR-C. The mt1 oligonucleotide, in which CA in the 5'-flanking region of the candidate VDRE sequence was changed to AC, showed binding activity similar to that of the wild-type DR-C (Fig. 8, Table I). The mutation of AG in the 5' half-site to GT (mt2) inhibited the ability to interact with VDR-RXR. The mutation of the first and second (GG to TC; mt3) or second and third (GG to TC; mt4) nucleotides of the 5' half-site of the candidate VDRE, or of the third nucleotide of the 3-nucleotide spacer and the first nucleotide of the 3' half-site (AA to TT; mt6) abolished the binding activity. The binding activity results of mt2 and mt3 were consistent with the transcriptional activities shown in Fig. 5. Finally, the mutation of the first and second nucleotides (GC to TA) in the 3-nucleotide spacer (mt5) had no effect on the binding activity. Thus, the VDR-RXR heterodimer recognized the sequence 5'-GGGGCAGCAAGGGCA-3' in the promoter of the NaPi-3 gene.

View larger version (85K): [in this window] [in a new window]   Fig. 8.   EMSA analysis with mutant DR-C oligonucleotides. EMSAs were performed with 32P-labeled oligonucleotide DR-C as the probe, the nuclear extract of COS-7 cells expressing human VDR, and various mutant DR-C oligonucleotides at the indicated concentrations as competitors. The sequences of the oligonucleotides are shown in Table I.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

The present study showed that the administration of 1,25-(OH)₂D₃ increased the Na⁺-dependent P_i uptake in the juxtamedullary cortex of vitamin D-deficient rats. The results of our transport studies performed with BBMVs isolated from the superficial and juxtamedullary cortexes, are consistent with previous studies using micropuncture, in situ microperfusion, isolated perfused tubules, and primary cell cultures that have demonstrated axial heterogeneity for proximal tubular Na⁺-P_i co-transport, and with the finding that Na⁺-P_i co-transport activity is greater in the PCTs than in the PSTs (3-8). The enzyme activity profiles of brush-border membrane from the superficial and juxtamedullary cortexes indicate that they correspond to those of PCTs and PSTs, respectively (36-38). In this context, type II Na⁺-P_i co-transporters in the luminal membrane appeared more abundant in PCTs than in PSTs. The up-regulation of Na⁺-P_i co-transporters in response to 1,25-(OH)₂D₃ was observed in the juxtamedullary (but not superficial) cortex, suggesting that 1,25-(OH)₂D₃ may increase type II Na⁺-P_i co-transporter expression and activity in PST cells.

These observations suggest the functional difference of the VDR between PCTs and PSTs. The VDR is present not only in classical target tissues of 1,25-(OH)₂D₃ but also in many other tissues (39). The effects of 1,25-(OH)₂D₃ in proximal and distal tubules are not uniform; for example, 1,25-(OH)₂D₃ reduces the expression of 25-hydroxyvitamin D-1-hydroxylase in proximal tubules and increases the expression of the Ca²⁺-binding protein in distal tubules (40, 41). In addition, the VDR is down-regulated in PCT cells when the renal production of 1,25-(OH)₂D₃ is stimulated. The regulation of VDR expression may underlie the reciprocal control of 25-hydroxyvitamin D-24-hydroxylase and 25-hydroxyvitamin D-1-hydroxylase activities in PCTs (42). Our present data indicate that 1,25-(OH)₂D₃ may differently regulate type II Na⁺-P_i co-transporters differently in PCTs and PSTs.

However, immunohistochemical studies showed that 1,25-(OH)₂D₃ increased the amount of immunoreactivity with anti-NaPi-2 antibody in the juxtamedullary cortex but not the superficial cortex despite the presence of immunoreactive PCT in both cortexes (Fig. 3). Thus, the action of 1,25-(OH)₂D₃ may be different between juxtamedullary PCTs and superficial PCTs as well as between PCTs and PSTs. In addition, a recent report showed the heterogeneity of vitamin D actions on Na⁺,K⁺-ATPase activity, 25-hydroxyvitamin D-24-hydroxylase activity, and 25-hydroxyvitamin D-1-hydroxylase activity in superficial and juxtamedullary PCTs (43); that is, these enzyme activities in juxtamedullary PCTs are more responsive to vitamin D than are those in superficial PCTs. However, these observations may be due, at least in part, to the effects of PTH, because PTH suppresses Na⁺/P_i co-transport activity not only by mainly an enhancement of the endocytosis of transporter protein from the plasma membrane but also by a reduction of the mRNA levels of type II Na⁺-P_i co-transporter (13). We therefore estimated the alteration of the serum PTH level after the vitamin D administration to rats. The serum PTH level was markedly high during the vitamin D deficiency state (about 580 ± 123 pg/dl), whereas the level was slightly decreased at 12 h after the administration of 1,25-(OH)₂D₃ (490 ± 89 pg/dl) and was normalized after 48 h (40 ± 11 pg/dl). The level even at 12 h was still much higher than the normal level. It is interesting that 1,25-(OH)₂D₃ up-regulated the type II Na⁺-P_i co-transporter in the juxtamedullary PCT in the presence of a high level of PTH. In addition, the plasma phosphate levels were not significantly different between the vitamin D deficiency state and at 12 h after the 1,25-(OH)₂D₃ administration. In light of these results, this up-regulation may be the consequence of a direct action of 1,25-(OH)₂D₃ rather than due to the changes of PTH or P_i levels.

To further study the mechanism of up-regulation by 1,25-(OH)₂D₃, the human NaPi-3 gene VDRE was identified. The relatively small number of natural VDREs that have been characterized indicates that these elements consist of two imperfect direct repeats of the nucleotide sequence GGGTGA separated by three nucleotides (44). The genes for osteocalcin, osteopontin, and 25-hydroxyvitamin D-24-hydroxylase have provided the most information concerning transcriptional activation by 1,25-(OH)₂D₃. The expression of the 25-hydroxyvitamin D-24-hyroxylase gene is controlled by two independent VDREs (nucleotides 259 to 245 and 151 to 137) (45). The transcriptional response to 1,25-(OH)₂D₃ of the 25-hydroxyvitamin D-24-hydroxylase gene promoter in COS-7 cells was markedly greater than those of the osteocalcin and NaPi-3 gene promoters. The human osteocalcin VDRE and NaPi-3 VDRE revealed similar affinity to VDR-RXR heterodimer. The calbindin D9k gene promoter is not transcriptionally responsive to 1,25-(OH)₂D₃, suggesting that the large increase in calbindin mRNA induced by 1,25-(OH)₂D₃ may be mediated primarily by post-transcriptional mechanisms, as reported previously (46).

Early studies indicated that a nuclear accessory factor is required for the VDR to bind to DNA (47). Highly purified VDR derived from baculovirus or yeast expression systems and in vitro synthesized VDR were unable to interact directly with VDREs, suggesting that the VDR is unable to form natural homodimers (48, 49). RXR is a candidate for this nuclear accessory factor. Whereas in vitro synthesized VDR formed a complex with the osteocalcin VDRE that was not enhanced by the addition of RXR (50), the VDR formed a complex with the NaPi-3 VDRE only in the presence of RXR.

The VDRE of the NaPi-3 gene is located ~2 kilobases upstream of the transcription initiation site, which makes it the most distant from the transcription start site among the known VDREs. This may be due to a unique property of the regulation of the NaPi-3 gene by 1,25-(OH)₂D₃. The administration of 1,25-(OH)₂D₃ to vitamin D-deficient rats resulted in a decrease of NaPi-2 mRNA in the superficial cortex, suggesting the possible existence of a negative VDRE in the gene promoter. The promoter of the human PTH gene contains a sequence that mediates transcriptional repression in response to 1,25-(OH)₂D₃ (51). Unlike other VDREs, only a single-copy motif (AGGTTCA) is apparent in this promoter sequence (nucleotides 125 to 101) in the human PTH gene. This sequence mediated transcriptional repression in response to 1,25-(OH)₂D₃ in GH4C1 cells but not in ROS-17/2.8 cells, suggesting the requirement of a cell-specific factor in addition to the VDR for the 1,25-(OH)₂D₃-induced inhibition of transcription (52). An identical motif (AGGTTCA, nucleotides 93 to 86, relative to the transcription start site) is present in a similar position in the human NaPi-3 gene (15). In the present study, the up-regulation mechanism of the renal type II Na⁺-P_i co-transporter by 1,25-(OH)₂D₃ in vitamin D-deficient rats was partly elucidated by the identification of a novel VDRE in the human NaPi-3 gene promoter. However, further studies necessary to clarify the mechanism of the down-regulation of type II Na⁺-P_i co-transporter by 1,25-(OH)₂D₃. In this context, the position of the VDRE in the NaPi-3 gene may be located distant from this negative VDRE. Transcriptional repression of the human NaPi-3 gene promoter by 1,25-(OH)₂D₃ was not observed in this study.