Regulation of Type II Renal Na+-dependent Inorganic Phosphate Transporters by 1,25-Dihydroxyvitamin D3

Vitamin D is an important regulator of phosphate homeostasis. The effects of vitamin D on the expression of renal Na+-dependent inorganic phosphate (Pi) transporters (types I and II) were investigated. In vitamin D-deficient rats, the amounts of type II Na+-dependent Pi transporter (NaPi-2) protein and mRNA were decreased in the juxtamedullary kidney cortex, but not in the superficial cortex, compared with control rats. The administration of 1,25-dihydroxyvitamin D3(1,25-(OH)2D3) to vitamin D-deficient rats increased the initial rate of Pi uptake as well as the amounts of NaPi-2 mRNA and protein in the juxtamedullary cortex. The transcriptional activity of a luciferase reporter plasmid containing the promoter region of the human type II Na+-dependent Pi transporter NaPi-3 gene was increased markedly by 1,25-(OH)2D3 in COS-7 cells expressing the human vitamin D receptor. A deletion and mutation analysis of the NaPi-3 gene promoter identified the vitamin D-responsive element as the sequence 5′-GGGGCAGCAAGGGCA-3′ nucleotides −1977 to −1963 relative to the transcription start site. This element bound a heterodimer of the vitamin D receptor and retinoid X receptor, and it enhanced the basal transcriptional activity of the promoter of the herpes simplex virus thymidine kinase gene in an orientation-independent manner. Thus, one mechanism by which vitamin D regulates Pi homeostasis is through the modulation of the expression of type II Na+-dependent Pi transporter genes in the juxtamedullary kidney cortex. Vitamin D is an important regulator of phosphate homeostasis. The effects of vitamin D on the expression of renal Na+-dependent inorganic phosphate (Pi) transporters (types I and II) were investigated. In vitamin D-deficient rats, the amounts of type II Na+-dependent Pi transporter (NaPi-2) protein and mRNA were decreased in the juxtamedullary kidney cortex, but not in the superficial cortex, compared with control rats. The administration of 1,25-dihydroxyvitamin D3(1,25-(OH)2D3) to vitamin D-deficient rats increased the initial rate of Pi uptake as well as the amounts of NaPi-2 mRNA and protein in the juxtamedullary cortex. The transcriptional activity of a luciferase reporter plasmid containing the promoter region of the human type II Na+-dependent Pi transporter NaPi-3 gene was increased markedly by 1,25-(OH)2D3 in COS-7 cells expressing the human vitamin D receptor. A deletion and mutation analysis of the NaPi-3 gene promoter identified the vitamin D-responsive element as the sequence 5′-GGGGCAGCAAGGGCA-3′ nucleotides −1977 to −1963 relative to the transcription start site. This element bound a heterodimer of the vitamin D receptor and retinoid X receptor, and it enhanced the basal transcriptional activity of the promoter of the herpes simplex virus thymidine kinase gene in an orientation-independent manner. Thus, one mechanism by which vitamin D regulates Pi homeostasis is through the modulation of the expression of type II Na+-dependent Pi transporter genes in the juxtamedullary kidney cortex. The reabsorption of inorganic phosphate (Pi) in the renal proximal tubule plays a key role in overall Pihomeostasis (1Murer H. Biber J. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2481-2509Google Scholar, 2Berndt T.J. Knox F.G. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2511-2531Google Scholar). 1,25-Dihydroxyvitamin D3(1,25-(OH)2D3) 1The abbreviations used are: 1,25-(OH)2D3, 1,25-dihydroxyvitamin D3; Pi, inorganic phosphate; Na+-Pi co-transport, Na+-dependent Pi transport; rNaPi-1, rat type I Na+-dependent Pi transporter; NaPi-2, rat type II Na+-dependent Pi transporter; NaPi-3, human type II Na+-dependent Pi transporter; PTH, parathyroid hormone; VDR, vitamin D receptor; PCT, proximal convoluted tubule; PST, proximal straight tubule; BBMV, brush-border membrane vesicle; RXR, retinoid X receptor; EMSA, electrophoretic mobility shift assay; VDRE, vitamin D-responsive element. 1The abbreviations used are: 1,25-(OH)2D3, 1,25-dihydroxyvitamin D3; Pi, inorganic phosphate; Na+-Pi co-transport, Na+-dependent Pi transport; rNaPi-1, rat type I Na+-dependent Pi transporter; NaPi-2, rat type II Na+-dependent Pi transporter; NaPi-3, human type II Na+-dependent Pi transporter; PTH, parathyroid hormone; VDR, vitamin D receptor; PCT, proximal convoluted tubule; PST, proximal straight tubule; BBMV, brush-border membrane vesicle; RXR, retinoid X receptor; EMSA, electrophoretic mobility shift assay; VDRE, vitamin D-responsive element.regulates Pi homeostasis in the bone, intestine, and kidney (2Berndt T.J. Knox F.G. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2511-2531Google Scholar). However, the effect of 1,25-(OH)2D3 on the reabsorption of Pi in the kidney remains unclear. Contradictory results showing an increase or decrease in Piexcretion in response to 1,25-(OH)2D3 have been reported (2Berndt T.J. Knox F.G. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2511-2531Google Scholar). 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) (2Berndt T.J. Knox F.G. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2511-2531Google Scholar). A study using a micropuncture technique, in situmicroperfusion, isolated perfused tubules, and primary cell cultures revealed an axial heterogeneity in proximal tubular Pitransport (2Berndt T.J. Knox F.G. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2511-2531Google Scholar). The extent of Na+-Pico-transport is greater in the proximal convoluted tubules (PCTs) than in the proximal straight tubules (PSTs) (3McKeown J.W. Brazy P.C. Dennis V.W. Am. J. Physiol. 1979; 237: F312-F318PubMed Google Scholar, 4Suzuki M. Capparelli A. Jo O.D. Kawaguchi Y. Miyahara T. Kidney Int. 1989; 34: 268-272Abstract Full Text PDF Scopus (15) Google Scholar, 5Ullrich K.J. Rumrich G. Kloss S. Pflügers Arch. 1977; 372: 269-274Crossref PubMed Scopus (34) Google Scholar, 6Turner S.T. Dousa T.P. Kidney Int. 1985; 27: 879-885Abstract Full Text PDF PubMed Scopus (22) Google Scholar, 7Levi M Am. J. Physiol. 1990; 258: F1616-F1624PubMed Google Scholar, 8Quamme G.A. Am. J. Physiol. 1990; 258: F356-F363PubMed Google Scholar). Kinetic studies have shown that the greater Pi transport in the PCT is attributable to a higher V max of Na+-Pi co-transport (6Turner S.T. Dousa T.P. Kidney Int. 1985; 27: 879-885Abstract Full Text PDF PubMed Scopus (22) Google Scholar, 7Levi M Am. J. Physiol. 1990; 258: F1616-F1624PubMed Google Scholar, 8Quamme G.A. Am. J. Physiol. 1990; 258: F356-F363PubMed Google Scholar). Several Na+-Pi co-transporters have been isolated from the kidney cortex of various species (9Biber J. Custer M. Magagnin S. Hayes G. Werner A. Lötscher M. Kaissling B. Murer H. Kidney Int. 1996; 49: 981-985Abstract Full Text PDF PubMed Scopus (95) Google Scholar, 10Tenenhouse H.S. J. Bone Miner. Res. 1997; 12: 159-164Crossref PubMed Scopus (54) Google Scholar, 11Murer H. Biber J. Pflügers Arch. 1997; 433: 379-389Crossref PubMed Scopus (90) Google Scholar). 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) (9Biber J. Custer M. Magagnin S. Hayes G. Werner A. Lötscher M. Kaissling B. Murer H. Kidney Int. 1996; 49: 981-985Abstract Full Text PDF PubMed Scopus (95) Google Scholar, 10Tenenhouse H.S. J. Bone Miner. Res. 1997; 12: 159-164Crossref PubMed Scopus (54) Google Scholar, 11Murer H. Biber J. Pflügers Arch. 1997; 433: 379-389Crossref PubMed Scopus (90) Google Scholar). Both types of Na+-Pi 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 (9Biber J. Custer M. Magagnin S. Hayes G. Werner A. Lötscher M. Kaissling B. Murer H. Kidney Int. 1996; 49: 981-985Abstract Full Text PDF PubMed Scopus (95) Google Scholar, 10Tenenhouse H.S. J. Bone Miner. Res. 1997; 12: 159-164Crossref PubMed Scopus (54) Google Scholar, 11Murer H. Biber J. Pflügers Arch. 1997; 433: 379-389Crossref PubMed Scopus (90) Google Scholar). The type II transporter was found to be regulated mainly by dietary Pi and PTH. In addition, the regulation of the type II transporter by dietary Pi and PTH was shown to differ between PSTs and PCTs (12Levi M. Arar M. Kaissling B. Murer H. Biber J. Pflügers Arch. 1994; 426: 5-11Crossref PubMed Scopus (17) Google Scholar, 13Kempson S.A. Lötscher M. Kaissling B. Biber J. Murer H. Levi M. Am. J. Physiol. 1995; 268: F784-F791PubMed Google Scholar). In contrast, insulin and glucose can affect the expression of the rat type I transporter (14Li H. Ren P. Onwochei M. Ruch R.J. Xie Z. Am. J. Physiol. 1996; 271: E1021-E1028Crossref PubMed Google Scholar). To clarify the action of 1,25-(OH)2D3 on renal Pi 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-Pitransporters (rat NaPi-2, human NaPi-3 and mouse NaPi-7) gene are highly conserved (15Taketani Y. Miyamoto K. Tanaka K. Katai K. Chikamori M. Tatsumi S. Segawa H. Yamamoto H. Morita K. Takeda E. Biochem. J. 1997; 324: 927-934Crossref PubMed Scopus (45) Google Scholar, 16Hartmann C.M. Hewson A.S. Kos C.H. Hilfiker H. Soumounou Y. Murer H. Tenenhouse H.S. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 7409-7414Crossref PubMed Scopus (49) Google Scholar). In addition, 1,25-(OH)2D3 is known to affect renal Na-Pi co-transport activity in these three species (2Berndt T.J. Knox F.G. Seldin D.W. Giebisch G. The Kidney: Physiology and Pathophysiology. 2nd Ed. Ravan Press, NY1992: 2511-2531Google Scholar). 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 (17Kimura T. Okano T. Tsugawa N. Okamura Y. Kobayashi T. J. Bone Miner. Met. 1994; 12: S7-S11Crossref Scopus (3) Google Scholar). 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) (18Reeve L.E. Jorgensen N.A. DeLuca H.F. J. Nutr. 1982; 36: 122-126Google Scholar). 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 (19Yusufi A.N.K. Murayama N. Gapstur S.M. Szczepanska-Konkel M. Dousa T.P. Biochim. Biophys. Acta. 1994; 1191: 117-132Crossref PubMed Scopus (30) Google Scholar). 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 (20Minami H. Kim J.R. Tada K. Takahashi F. Miyamoto K. Nakabou Y. Sakai K. Hagihira H. Gastroenterology. 1993; 105: 692-697Abstract Full Text PDF PubMed Scopus (23) Google Scholar), and their purity was assessed by the measurement of the leucine aminopeptidase, Na+- and K+-dependent ATPase, and cytochrome-c-oxidase activities (21Nakagawa N. Arab N. Ghishan F.K. J. Biol. Chem. 1991; 266: 13616-13620Abstract Full Text PDF PubMed Google Scholar). The uptake of [32P]Pi was measured by a rapid filtration technique (21Nakagawa N. Arab N. Ghishan F.K. J. Biol. Chem. 1991; 266: 13616-13620Abstract Full Text PDF PubMed Google Scholar) 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)). 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 (22Takahashi Y. Taketani Y. Endo T. Yamamoto S. Kumegawa M. Biochim. Biophys. Acta. 1994; 1212: 217-224Crossref PubMed Scopus (65) Google Scholar). Rat type I transporter rNaPi-1 (nucleotides −58 to +356, relative to the translation start site (23Li H. Xie Z. Cell. Mol. Biol. Res. 1995; 41: 451-460PubMed Google Scholar)) and type II transporter NaPi-2 (nucleotides +543 to +1639, relative to the translation start site (24Magagnin S. Werner A. Markovich D. Sorribas V. Stange G. Biber J. Murer H. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5979-5983Crossref PubMed Scopus (327) Google Scholar)) cDNA probes were prepared by polymerase chain reaction with kidney total cDNA and specific oligonucleotide primers. The probes were labeled with [α-32P]dCTP (110 TBq/mmol; ICN, Costa Mesa, CA) with the use of a Megaprime labeling system (Amersham). 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 (24Magagnin S. Werner A. Markovich D. Sorribas V. Stange G. Biber J. Murer H. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5979-5983Crossref PubMed Scopus (327) Google Scholar). 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). 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). COS-7 cells (RIKEN Cell Bank, Saitama, Japan, (25Gluzman Y. Cell. 1981; 23: 175-182Abstract Full Text PDF PubMed Scopus (1449) Google Scholar)) 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, (26Majeska R. Rodan S. Rodan G. Endocrinology. 1980; 107: 1494-1503Crossref PubMed Scopus (417) Google Scholar)) 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. Two luciferase reporter vectors (p3P2400 and p3P1260) containing the 5′-flanking region of the human NaPi-3 gene were described in a previous study (15Taketani Y. Miyamoto K. Tanaka K. Katai K. Chikamori M. Tatsumi S. Segawa H. Yamamoto H. Morita K. Takeda E. Biochem. J. 1997; 324: 927-934Crossref PubMed Scopus (45) Google Scholar). 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) (27Hayashi S. Goto K. Okada T.S. Kondoh H. Genes Dev. 1987; 1: 818-828Crossref PubMed Scopus (70) Google Scholar) 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 p3PΔ1850TK was constructed withKpnI-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 (28Arai H. Miyamoto K. Taketani Y. Yamamoto H. Iemori Y. Morita K. Tonai T. Nishisho T. Mori S. Takeda E. J. Bone Miner. Res. 1997; 12: 915-921Crossref PubMed Scopus (552) Google Scholar) was constructed by subcloning anEcoRI fragment containing the full-length human VDR cDNA into pcDL-SRα-296 (29Takebe Y. Seiki M. Fujisawa J. Hoy P. Yokota K. Arai K. Yoshida M. Arai N. Mol. Cell. Biol. 1988; 8: 466-472Crossref PubMed Google Scholar). The β-galactosidase expression vector pCMV-β (CLONTECH, Palo Alto, CA) was used as an internal control. Each plasmid was purified with a plasmid purification kit (QIAGEN, Hilden, Germany).Table ISequences of the oligonucleotide used for the gel mobility shift assayNameSequencePositionDR-A5′-TTAATTGGGGGACAGAGGGAGGGCAG-3′−89 to −64DR-B5′-GGAACGAGGGGACCCTGGGAACAAGG-3′−251 to −226DR-C5′-GATCAGGGGCAGCAAGGGCAGAAATG-3′−1982 to −1957r24-VDRE5′-AGCTAGAGCGCACCCGCTGAACCCTGGGCT-3′−264 to −238hOC-VDRE5′-AGCTTCCGGGTGAACGGGGGCATTA-3′−501 to −483mt15′-ATACGGGGCAGCAAGGGCAGAAATG-3′−1981 to −1957mt25′-ATCGTGGGCAGCAAGGGCAGAAATG-3′−1981 to −1957mt35′-ATCATCGGCAGCAAGGGCAGAAATG-3′−1981 to −1957mt45′-ATCAGTCGCAGCAAGGGCAGAAATG-3′−1981 to −1957mt55′-ATCAGGGGCATAAAGGGCAGAAATG-3′−1981 to −1957mt65′-ATCAGGGGCAGCTTGGGCAGAAATG-3′−1981 to −1957Mutation bases are indicated by double underlining. Open table in a new tab Mutation bases are indicated by double underlining. 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-β 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-β, 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, β-galactosidase activity, and protein concentration (30Arakawa T. Nakamura M. Yoshimoto T. Yamamoto S. FEBS Lett. 1995; 363: 105-110Crossref PubMed Scopus (39) Google Scholar). Nuclear extracts from COS-7 cells were prepared as described by Arakawaet al. (30Arakawa T. Nakamura M. Yoshimoto T. Yamamoto S. FEBS Lett. 1995; 363: 105-110Crossref PubMed Scopus (39) Google Scholar), using a slight modification of the method established by Dignam et al. (31Dignam J.D. Martin P.L. Shastry B.S. Roeder R.G. Methods Enzymol. 1983; 101: 582-598Crossref PubMed Scopus (744) Google Scholar). COS-7 cells were transfected with the human VDR expression vector by the DEAE-dextran method (32Maruyama K. Wang H-M. Miyajima A. Takebe Y. Arai N. Karam J.D. Chao L. Warr G.W. Methods in Nucleic Acids Research. CRC Press, Boca Raton, FL1991: 283-306Google Scholar). The human VDR expression vector pSG5/hVDR (28Arai H. Miyamoto K. Taketani Y. Yamamoto H. Iemori Y. Morita K. Tonai T. Nishisho T. Mori S. Takeda E. J. Bone Miner. Res. 1997; 12: 915-921Crossref PubMed Scopus (552) Google Scholar) and the murine RXRα expression vector pSG5/m RXRα (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. Eleven double-stranded oligonucleotides corresponding to direct repeat-like sequences in the 5′-flanking region of the NaPi-3 gene (15Taketani Y. Miyamoto K. Tanaka K. Katai K. Chikamori M. Tatsumi S. Segawa H. Yamamoto H. Morita K. Takeda E. Biochem. J. 1997; 324: 927-934Crossref PubMed Scopus (45) Google Scholar), human osteocalcin gene (33Kerner S.A. Scott R.A. Pike J.W. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4455-4459Crossref PubMed Scopus (336) Google Scholar), and rat 25-hydroxyvitamin D-24-hydroxylase gene (34Zierold C. Darwish H.M. DeLuca H.F. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 900-902Crossref PubMed Scopus (142) Google Scholar), 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 [γ-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 vitrosynthesized 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). BBMVs were prepared from the superficial and juxtamedullary cortex of rat kidneys. The Pi uptake remained linear for up to 3 min in both types of vesicles (data not shown). The initial rate of Pi 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. 1 A). In the vitamin D-deficient animals, the initial rate of Pi 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 Pi 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.1 B). 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, Aand 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. Moreover, the amounts of NaPi-2 mRNA and protein in the juxtamedullary cortex were 2.5- and 3.1-fold of the value for normal rats in the vitamin D-deficient rats 12 h after the administration of 1,25-(OH)2D3 (Fig.3, A and B), respectively. In contrast, the amounts of NaPi-2 mRNA and protein in the superficial cortex showed only small changes in response to vitamin D deprivation and slightly decreased after the 1,25-(OH)2D3 administration. 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 andH). 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 (15Taketani Y. Miyamoto K. Tanaka K. Katai K. Chikamori M. Tatsumi S. Segawa H. Yamamoto H. Morita K. Takeda E. Biochem. J. 1997; 324: 927-934Crossref PubMed Scopus (45) Google Scholar). 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 p3PΔ1850TK, 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). To determine whether the DR-C sequence possesses functional VDRE activity, we constructed luciferase reporter plasmids that contain DR-C in the forward (pTKDRCF) or reverse (pTKDRCR) direction upstream of the minimal promoter of the thymidine kinase gene of herpes simplex virus. COS-7 cells, which express both VDR and retinoid X receptor (RXR), were transfected with each of these plasmids separately, and the inducibility of luciferase activity by 1,25-(OH)2D3 was examined. 1,25-(OH)2D3 stimulated luciferase activity to a similar extent in the COS-7 cells transfected with either plasmid (Fig. 5). To further test whether the DR-C sequence 5′-GGGGCAGCAAGGGCA-3′ is the target sequence of the candidate VDRE, we determined the luciferase activity of the cells transfected with pTKDRCmt2 and pTKDRCmt3, which contain the oligonucleotide mutated in the VDRE half-site. The mutation of AG in the 5′ half-site to GT (pTKDRCmt2) or the first and second (GG to TC; mt3) of the 5′ half-site of the VDRE completely inhibited the ability to respond to 1,25-(OH)2D3. Similar results were obtained by the transfection of ROS-17/2.8 cells endogenously expressing VDR (data not shown). The VDRE of the NaPi-3 gene was investigated further by an EMSA with various oligonucleotides as probes and competitors (TableI). 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. 6 A). 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. 6 B). 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) (33Kerner S.A. Scott R.A. Pike J.W. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 4455-4459Crossref PubMed Scopus (336) Google Scholar), or a monoclonal antibody (9A7γ) to chick VDR which could inhibit the VDR-DNA complex formation described previously (35Pike J.W. Sleator N.M. Haussler M.R. J. Biol. Chem. 1987; 262: 1305-1311Abstract Full Text PDF PubMed Google Scholar). 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 n