The human paracellin-1 gene (hPCLN-1): renal epithelial cell-specific expression and regulation
http://www.100md.com
《美国生理学杂志》
Laboratory of Developmental Nephrology, Faculty of Medicine, Technion-Israel Institute of Technology, and Pediatric Nephrology Unit, Department of Nephrology, Rambam Medical Center, Haifa, Israel
ABSTRACT
Tubular reabsorption of Mg2+ is mediated by the tight junction protein paracellin-1, which is encoded by the gene PCLN-1 (CLDN16) and exclusively expressed in the kidney. Tubular Mg2+ reclamation is modulated by many hormones and factors. The aim of this study was to define regulatory elements essential for renal tubular cell-specific expression of human PCLN-1 (hPCLN-1) and to explore the effect of Mg2+ transport modulators on the paracellin-1 gene promoter. Endogenous paracellin-1 mRNA and protein were detected in renal cell lines opossom kidney (OK), HEK293, and MDCT, but not in the fibroblast cell line NIH3T3. A 7.5-kb hPCLN-1 5'-flanking DNA sequence along with seven 5'-deletion products were cloned into luciferase reporter vectors and transiently transfected into the renal and nonrenal cells. The highest levels of luciferase activity resulted from transfection of a 5'-flanking 2.5-kb fragment (pJ2M). This activity was maximal in OK cells, was orientation dependent, and was absent in NIH3T3 cells. Mg2+ deprivation significantly increased pJ2M-driven activity in transfected OK cells, whereas Mg2+ load decreased it compared with conditions of normal Mg2+. Deletion analysis along with electrophoretic mobility-shift assay demonstrated that OK cells contain nuclear proteins, which bind a 70-bp region between –1633 and –1703 of major functional significance. Deleting this 70-bp segment, which contains a single peroxisome proliferator-response element (PPRE), or mutating the PPRE, caused a 60% reduction in luciferase activity. Stimulating the 70-bp sequence with 1,25(OH)2 vitamin D decreased luciferase activity by 52%. This effect of 1,25(OH)2 vitamin D was abolished in the absence of PPRE or in the presence of mutated PPRE. We conclude that the PPRE within this 70-bp DNA region may play a key role in the cell-specific and regulatory activity of the hPCLN-1 promoter. Ambient Mg2+ concentration and 1,25(OH)2 vitamin D may modulate paracellular, paracellin-1-mediated, Mg2+ transport at the transcriptional level. 1,25(OH)2 vitamin D exerts its activity on the hPCLN-1 promoter likely via the PPRE site.
magnesium; renal tubule; transcription; promoter; gene expression; gene regulation; peroxisome proliferator response element; 1,25(OH)2 vitamin D
MAGNESIUM IS THE MOST ABUNDANT divalent cation in the intracellular fluid. It plays a critical role in a wide variety of metabolic and cellular processes, including cellular energy storage, DNA/RNA processing, ion transport, membrane stabilization, and nerve conduction (33). Abnormalities in Mg2+ homeostasis are relatively common in clinical practice and may lead to neuromuscular disturbances, central nervous system manifestations, and cardiovascular dysfunction (16, 30). In mammals, the kidney is the principal organ responsible for Mg2+ balance (16, 30). Normally, >95% of the filtered Mg2+ is reabsorbed by the renal tubule. Ten to fifteen percent of the filtered Mg2+ is reabsorbed in the proximal tubule and 10% in the distal tubule. The major site of Mg2+ reabsorption is the thick ascending limb of the loop of Henle (TAL), where 60–70% of the filtered load is reclaimed (16, 30). Mg2+ transport in this tubule segment occurs primarily through paracellular conductance driven by the lumen positive electrical potential (30). While renal Mg2+ handling has been thoroughly investigated at the tubular and cellular levels (8, 16, 30), the molecular mechanisms of tubular Mg2+ reabsorption are poorly understood.
Recently, Simon et al. (38) using positional cloning, have identified a human gene, hPCLN-1 (also known as CLDN16, NCBI accession no. NM-006580), mutations in which cause familial hypomagnesemia-hypercalciuria syndrome. hPCLN-1 consists of five exons and resides on chromosome 3q27. The gene encodes a protein, paracellin-1, which is composed of 305 amino acids (38). Northern blot analysis of human tissues has shown that the 3.5-kb PCLN-1 mRNA transcript is expressed exclusively in the kidney (38). RT-PCR analysis of mRNA from nephron segments of the rabbit (38) and rat (42) has demonstrated that PCLN-1 is expressed in the TAL and the distal convoluted tubule (DCT). The paracellin-1 protein, which is located in the paracellular tight junctions of the TAL and DCT, is a member of the claudin family of tight junction proteins (27) and appears to mediate resorption of both Mg2+ and Ca2+ (38).
In the kidney, the specialized reabsorptive and/or secretory function of each tubule segment depends upon its structural arrangement and upon the specific pattern of gene expression in each tubular cell type. The promoters of several transporter and channel genes including aquaporin (28), the Na+-phosphate cotransporter (36), the Na+-K+-Cl– cotransporter (39), and chloride channels (40) as well as the promoters of nephrin (46) and cadherin (13, 43) genes have been cloned and shown to direct kidney-specific expression in vitro and/or in transgenic mice. Several transcription factors including myc-associated zinc finger proteins and Krüppel-like factor (41), hepatocyte nuclear factor-3 (39), and hepatocyte nuclear factor-1 (2), were found to be involved in kidney-specific expression of the ClC-K1 chloride channel, thiazide-sensitive Na-Cl cotransporter, and cadherin genes, respectively. However, very little is known about the regulatory elements responsible for cell-specific expression of transporter and channel genes in the kidney. As yet, the promoter region of the human paracellin-1 gene has not been characterized, and the molecular mechanisms for renal epithelial-specific activity of this gene have not been investigated.
Many factors are known to modulate Mg2+ reabsorption in various nephron segments. These factors include hormones such as insulin, 1,25(OH)2 vitamin D, aldosterone, and parathyroid hormone as well as nonhormonal factors such as Mg2+ restriction and load and acid-base changes (8, 30). Most of these factors influence both transcellular magnesium transport in the DCT as well as paracellular transport of this cation in the TAL (8, 30). However, very little is known about the molecular mechanisms of this modulation and whether it occurs at the protein or DNA/RNA level. The magnesium restriction (30)- and 1,25(OH)2 vitamin D (32)-induced increase in Mg2+ uptake by renal MDCT cells was diminished by pretreatment of cells with actinomycin D, suggesting that this stimulation occurs through transcriptional activation. Nevertheless, the molecular mechanisms whereby hormones and other factors modulate Mg2+ transport across the paracellular pathway in the renal tubule are unknown.
The purpose of this study was to define cis-acting promoter regulatory elements and to examine trans-acting factors essential for renal tubular epithelial-specific expression of hPCLN-1. We also explored the effect of modulators of Mg2+ transport on the hPCLN-1 gene promoter. We show that a 70-bp 5'-flanking region of the paracellin-1 gene may determine renal epithelial-specific expression of this gene. We demonstrate an increase in hPCLN-1 promoter activity in response to Mg2+ depletion and a decrease in response to Mg2+ load. In addition, we show that 1,25(OH)2 vitamin D may modulate Mg2+ transport at the transcriptional level, probably via the peroxisome proliferator response element (PPRE) contained within the 70-bp region.
MATERIALS AND METHODS
Opossum kidney (OK) cells (provided by Dr. J Green, Technion, Haifa, Israel), human embryonic kidney (HEK293) cells (provided by Dr. K. Skorecki, Technion, Haifa, Israel), mouse distal convoluted tubule (MDCT) cells (provided by Dr. P. Friedman, University of Pittsburgh, Pittsburgh, PA), and mouse embryonic fibroblast (NIH3T3) cells were grown and maintained in DMEM/F-12 supplemented with 10% fetal calf serum, 2 mM glutamine, 50 IU/ml penicillin, and 50 μg/ml streptomycin at 37°C in a humidified atmosphere of 95% air-5% CO2.
RT-PCR followed by Southern blot analysis. Total RNA isolated with Tri-reagent (MRB, Cincinnati, OH) from the cell lines mentioned above was reverse-transcribed using OmniscriptRT (Qiagen, Hilden, Germany) with random hexamers (Promega, Madison, WI). PCR was performed using HotStarTaq DNA polymerase (Qiagen) with two sets of primers (Table 1). The first (F1,R1) complementary to exon 1 of hPCLN-1 and the second (F2,R2) complementary to a region between exons 3 and 5 of PCLN-1 (spanning two introns) highly homologous among human, rat, and mouse (mPCLN-1). The resultant DNA was separated on 1% agarose gels and transferred to nylon membranes (Osmonics, Minnetonka, MN). The membranes were probed with PCLN-1 cDNA probes generated by PCR using primers F1,R1 for hPCLN-1 and primers F3,R2 (Table 1) from exon 5 for mPCLN-1, with human genomic DNA as a template. A similar procedure was carried out with actin primers (F4,R4) as control. Probes were 32P labeled by random priming (Biological Industries, Beit Ha’Emek, Israel) using [-32P]dCTP (DuPont-New England Nuclear, Boston, MA).
View this table:
Western blot analysis. Protein extracts from HEK293, OK, and NIH3T3 cells were prepared using standard protocols (35). Proteins were separated on 10% SDS polyacrylamide gels, transferred to nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany), and probed with anti-human paracellin-1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) using chemiluminescense (Biological Industries). To verify antibody specificity, a peptide competition assay was carried out with 100-fold excess of paracellin-derived peptide.
Cloning of hPCLN-1 5'-flanking DNA. Using the published sequence of human paracellin-1 cDNA (NCBI accession no. NM-006580), we localized the hPCLN-1 gene on chromosome 3, drafts of which have recently been deposited in NCBI (accession no. NT-00962). A DNA fragment of 7.5 kb in the 5'-flanking sequence region, to but not including the PCLN-1 translation start site (Fig. 1), was synthesized using PCR with human genomic DNA as a template and primers F5/R5 (Table 1). To minimize the possibility of PCR-generated mutations, the High-Fidelity Long-Range PCR system (Roche, Mannheim, Germany) was used. The fragment was TA cloned into pCR-XL-TOPO vector (Invitrogen, Carlsbad, CA) and sequence verified. Computer analysis (Wisconsin Package version 8.0, Genetic Computer Group) of the 5'-flanking region of hPCLN-1 from the gene’s translation start site was carried out, and putative transcription factor binding sites were identified.
Generation of promoter/reporter constructs. The 7.5-kb 5'-flanking sequence described above was subcloned into pGL3-basic vector (Promega) upstream of a luciferase reporter gene (and named pJ12) (Fig. 1). In addition, a set of seven deletion fragments decreasing in size from the 5'-end of the 7,514-bp genomic fragment were produced and cloned into pGL3-basic vector. Five inserts were prepared using PCR-based strategies with genomic DNA as template with primers F6, F7, F8, F9, F10, and R5 (Table 1). The additional two were prepared by removing specific 5'-segments from cloned hPCLN-1 sequences in pGL3-basic, using restriction enzymes. All eight constructs were sequenced to verify the orientation and integrity of the inserts. As illustrated in Fig. 1, the 5'-ends of the deletion fragments were positioned at –4687 (pJ2L), –3986 (pJ2/3.9), –3317 (pJ2/3.3), –2554 (pJ2M), –1982 (pJ2/1.9), –1458 (pJ2.1/4), and –733 (pJ2.7), respectively. All fragments extended to the ATG site, but did not include it. All eight constructs were cloned into the vector in sense orientation, and the 2.5-kb fragment in pJ2M was also cloned in antisense orientation.
A reporter plasmid containing a 586-bp nested deletion (from position –1772 to –1186) within this 2.5-kb DNA region (termed pJ2Mdel) was generated using PCR with pJ2M as template and primers F12/R12 (Table 1) complementary to the gap ends. Following DpnI treatment, the PCR-generated pGL3-basic construct was ligated, and the sequence was verified. Empty pGL3 vector containing no insert was used as a negative control. pGL3-control containing the SV40 enhancer/promoter was used as a positive control. Promoter activity was estimated from the ratio of hPCLN-1-driven luciferase activity in pGL3-basic vector to promoterless pGL3-basic vector.
For reporter gene activity driven by distal hPCLN-1 promoter fragments, the DNA region from position –1738 to –1493 as well as a series of five deletion fragments 5'-truncated in increments of 35 bp were PCR generated with human genomic DNA and primers F13, F14, F15, F16, F17, F18, with R13. All six fragments were cloned into the pGL3-promoter vector, which carries an SV40 minimal promoter upstream to the luciferase reporter gene. In these experiments, promoter activity was determined from the ratio of hPCLN-1 promoter-driven luciferase activity in the pGL3-promoter vector to that in the empty pGL3-promoter vector.
Point mutations in the 2.5-kb insert-containing plasmid, pJ2M, and the 210-bp insert-containing plasmid, pJ210 (see RESULTS), were generated using PCR with the mutated primers F19 and R14 (Table 1).
In all transfection experiments, plasmid pCH110 (Pharmacia, Uppsala, Sweden), containing the LacZ gene driven by the CMV promoter, was used to normalize for transfection efficiency. DNA for transfections was purified using Nucleobond AX (Macherey Nagel, Duren, Germany).
Transient transfections and reporter gene assays. OK, HEK293, and MDCT cells were plated (5 x 104/dish) in 24-well dishes in serum-containing medium. Cotransfections were performed using Fugene 6 (1.2 μl/well, Roche) with 0.3 μg reporter plasmid and 0.3 μg pCH110. NIH3T3 cells (4 x 105/dish) were seeded in 6-well plates and incubated overnight at 37°C. Cells were transfected 24 h later using Polyfect (10 μl/well, Qiagen) with 0.75 μg reporter plasmid and 0.75 μg pCH110. All plates were incubated at 37°C for 48 h. For enzymatic assays, cells were washed with PBS (150 mM NaCl, 15 mM sodium phosphate, pH 7.3) and lysed by incubating in 200 μl/well M-Per (Pierce, Cheshire, UK) for 5 min at 37°C. Lysed cells were centrifuged, and the supernatant was aliquoted (50 μl/well) into 96-well plates. Fifty microliters of Luciferase Assay Reagent (Promega) were automatically added, and the light intensity of the reaction was immediately read in a luminometer (Lucy, Anthos, Austria) for a period of 10 s. Luciferase activity was normalized to -galactosidase activity, which was measured in identical cell lysates. One hundred sixty microliters of ONPG substrate (Sigma, St. Louis, MO) were added to 30 μl of cell lysate in each well of a 96-well plate. The reaction mixture was incubated for 30 min at 37°C, or until yellow color developed. -Galactosidase measurements were performed by a luminometer with a 405-nm filter. Measurements of luciferase and -galactosidase were performed in duplicate.
In some experiments, 48 h after transfection, the medium was replaced with fresh medium containing 0 (low), 0.7 (normal), or 1.5 mM (high) Mg2+ (Biological Industries). In other experiments, the fresh medium contained 1,25(OH)2 vitamin D (from a stock solution of 10–4 M in ethanol, Sigma) at a final concentration of 5 x 10–7 M. Control experiments were carried out with 0.5% ethanol. Cells were exposed to experimental media for 24 h.
EMSA. Nuclear extracts were prepared from cells using the method of Dignam (9). Briefly, confluent cells were grown on 100-mm plates, washed in 3 ml of PBS (150 mM NaCl, 15 mM sodium phosphate, pH 7.3) supplemented with protease inhibitor mix (Complete, Roche), scraped, and pelleted. The pellet was resuspended in ice-cold lysis buffer (in mM: 10 HEPES, pH 7.9, 1.5 MgCl2, 420 NaCl, 0.2 EDTA, and 1 DTT as well as protease inhibitor mix) and incubated at 4°C for 20 min. Following centrifugation, the supernatant, containing the nuclear extract, was diluted 1:2 with (in mM) 20 HEPES (pH 7.9), 100 KCl, 0.2 EDTA, and 1 DTT as well as 20% glycerol and protease inhibitors. Protein concentration was measured using Bradford reagent (Sigma) at 595 nm with bovine serum albumin as the standard.
Double-stranded oligonucleotides corresponding to promoter sequences of interest were end-labeled with [-32P]ATP (DuPont-New England Nuclear) using T4 polynucleotide kinase (NEB, Beverly, MA). Binding reactions contained (in mM) 10 Tris·HCl (pH 8.0), 250 KCl, 0.5 EDTA, and 0.2 DTT as well as 0.1% Triton-X 100, 12.5% glycerol (vol/vol), 1 μg poly-dIdC-labeled probe (5 x 104 counts/min), and 50 μg of nuclear extracts. In some reactions, a 50-fold molar excess of unlabeled double-stranded oligonucleotide was added for specific competition. Following a 30-min incubation period on ice, complexes were resolved on 4% nondenaturing polyacrylamide gels in 1x TBE buffer. The gels were dried and autoradiographed.
RESULTS
Cloning the 5'-flanking region of hPCLN-1. A DNA fragment stretching over 7.5 kb in the 5'-flanking region of hPCLN-1, reaching 9 bp from the gene’s translation start site, was isolated. Computer analysis of this fragment disclosed a variety of putative transcription factor binding sites, among them recognition sites for several transcription factors known to be tissue restricted and expressed in the kidney. These included hepatocyte nuclear factor 5 (HNF-5), GATA factors, and a peroxisone proliferator-activated receptor (PPAR) binding site. The hPCLN-1 promoter also contained binding sites for transcription factors involved in signal transduction, such as activator protein (AP)-1 and AP-3 as well as a putative TATA box.
The 5'-flanking regions of PCLN-1 in mouse and rat genomic DNA were located using NCBI-BLAST with the respective cDNA sequences. Comparison of the 5'-flanking region of hPCLN-1 to that in mouse (accession no. NW-000107, region: 23187252 ... 23287252) and rat (accession no. AC106700) revealed very little overall sequence identity within 2.4 kb 5' to the translation site. However, the 51-bp region from position –392 to –341, relative to the hPCLN-1 translation start site, was 79 and 81% identical to the mouse and rat PCLN-1 5'-flanking region sequences, respectively.
Comparison of the hPCLN-1 5'-flanking region to that of a variety of genes encoding various ion channels and transporters, known to be specifically expressed in the kidney, such as the voltage-gated Cl– channels ClC-K1 and ClC-K2, aquaporin-2, the K+ channel ROMK, Tamn-Horsfall protein, and the thiazide-sensitive NaCl cotransporter, did not reveal areas of sequence homology.
Kidney cell-specific expression of hPCLN-1. RT-PCR followed by Southern blot analysis with nested probes (see MATERIALS AND METHODS) detected endogenous PCLN-1 mRNA in HEK293 and OK cells using an hPCLN-1 probe and in HEK293 and MDCT cells using an mPCLN-1 probe (Fig. 2A). Paracellin-1 mRNA was not detected in NIH3T3 cells.
Western blot analysis with paracellin-1-specific antibody detected paracellin-1 protein expression in HEK293 and OK but not in NIH3T3 cells (Fig. 2B). Addition of 100-fold molar excess of paracellin-1 peptide to the antibody, 24 h before incubation with the blot, competed out the paracellin-1 band (data not shown).
Kidney cell-specific activity of the hPCLN-1 promoter. To verify that the 5'-flanking region of PCLN-1 contained a functional cell-specific promoter, reporter gene assays were conducted. The 7.5-kb hPCLN-1 5' flanking region, as well as three 5'-deletion products, were cloned upstream to a luciferase reporter gene in promoterless pGL3-basic vectors in sense orientation.
The resulting plasmids (Fig. 1), designated pJ12 (7,514-bp insert), pJ2L (4,687-bp insert), pJ2M (2,554-bp insert) and pJ2.7 (733-bp insert), were transfected into the PCLN-1-expressing cell lines HEK293, OK, and MDCT. Reporter gene activities were compared with those in transfected NIH3T3 cells, which do not express PCLN-1. As shown in Fig. 3, all three PCLN-1-expressing renal cell lines displayed a similar pattern of luciferase activity when transfected with pJ12, pJ2L, pJ2M, and pJ2.7. Luciferase activity was maximally induced by pJ2M (2.5 kb) and gradually diminished, as the insert size was either increased or decreased. Specifically, the lowest luciferase activity resulted from transfection of the 7.5-kb insert-containing plasmid (pJ12). Truncation of the 5'-flanking region from –7,514 to –4,687 bp (pJ2L) did not significantly increase luciferase activity. A further decrease in size from –4,687 to –2554 bp (pJ2M) caused a significant increase in luciferase activity in all three renal cell lines. When the 7.5-kb fragment in the luciferase vector was further shortened to 0.733 kb (pJ2.7), luciferase activity in all four renal cell lines was markedly reduced. Transfection of each of the four constructs into the control cell line NIH3T3 showed negligible induction of luciferase activity.
Taken together, these results indicated that the 2.5-kb hPCLN-1 promoter fragment cloned in pJ2M contained an active promoter. The lack of stimulation in NIH3T3 cells suggested that the activity of the hPCLN-1 promoter was kidney cell specific. The results also suggested that kidney-specific expression of the hPCLN-1 promoter was due, at least in part, to tissue-specific transcriptional regulation.
Although the induction pattern was similar in all three PCLN-1-expressing renal cell lines, its magnitude was very different between cells. The highest levels of induction appeared in OK cells and the lowest in MDCT cells. HEK293 cells displayed intermediate levels of induction. Based on these findings, OK cells were selected as experimental cells in our next set of experiments.
Deletion analysis of the hPCLN-1 promoter. To further explore the 5'-flanking region of hPCLN-1, a second set of deletion constructs was tested (Fig. 1), two between pJ2L and pJ2M (designated pJ2/3.9 and pJ2/3.3) and two between pJ2M and pJ2.7 (designated pJ2/1.9 and pJ/1.4). All eight constructs were transiently transfected into OK and NIH3T3 cells. As evident from Fig. 4, the highest level of luciferase activity resulted from transfection of the 2.5-kb insert-containing plasmid (pJ2M). Increasing the size of this fragment by 763 bp (pJ2/3.3) caused a 50% reduction in luciferase activity, which remained unchanged when the 2.5-kb fragment was lengthened by 1,432 bp (pJ2/3.9). When the 2.5-kb fragment was lengthened by 2,133 (pJ2L) or 4,960 bp (pJ12), luciferase activity gradually diminished by 60 and 90%, respectively, compared with pJ2M. Decreasing the size of the 2.5-kb fragment by 572 (pJ2/1.9) and 1,096 bp (pJ2/1.4) brought about a 30 and 77% reduction in luciferase activity, respectively. A further truncation of the insert from 1,458 to 733 bp left luciferase activity at 23% that of pJ2M. No significant differences in luciferase activity between the various fragments tested were demonstrated in control NIH3T3 cells (data not shown).
To further explore the 2.5-kb 5'-flanking region of hPCLN-1, which displayed the highest promoter activity, an orientation study was performed in OK cells. As shown in Fig. 5, when the 2.5-kb promoter fragment in pGL3 vector was reversed (pJ235M), luciferase activity was 80% lower than activity induced by the same sequence in sense orientation (pJ2M).
Taken together, these results suggested that a positive regulatory element, most likely located between positions –1982 and –1458, and a negative regulatory element likely positioned between positions –3317 and –2554, are involved in hPCLN-1 promoter activity in renal cells.
Binding of nuclear proteins to the hPCLN-1 promoter. To investigate the location and nature of the presumed positive regulatory element between positions –1982 and –1458, binding of nuclear proteins to this region was examined. The 525-bp DNA fragment was divided into 15 sequential double-stranded oligonucleotides, which were 5' end-labeled with 32P and incubated with nuclear proteins. DNA-protein complexes were resolved from unbound DNA by nondenaturing gel electrophoresis. Binding patterns were compared between nuclear extracts from OK cells and those from control NIH3T3 cells. As shown in Fig. 6, incubation of seven sequential DNA fragments located in the proximal part of the 525-bp sequence (between positions –1770 and –1491) with nuclear extracts from OK cells produced several retarded bands ( in Fig. 6) that were absent when the DNA was incubated with nuclear extracts from NIH3T3 cells. Binding was specific, since addition of a 50-fold molar excess of unlabeled oligonucleotide abolished DNA-protein complexes. DNA binding was absent in lanes without the nuclear extract. When oligonucleotides originating from the distal part of the 500-bp region (from position –1982 to –1771) were incubated with nuclear proteins from OK and NIH3T3 cells, the protein-DNA binding pattern was similar in the two cell lines (data not shown).
Taken together, these results suggested that OK cells, the renal cell line OK, but not nonrenal NIH3T3 cells, contain nuclear proteins that bind specifically to the hPCLN-1 promoter in the 280-bp region between positions –1770 and –1491.
Deletion analysis of the DNA region containing the proposed positive regulatory element of hPCLN-1. To investigate the functional importance of the 280-bp hPCLN-1 promoter region implicated in transcriptional activation and nuclear protein binding, a deletion of the region between positions –1772 and –1186 within the pJ2M reporter vector was created. When the resulting plasmid, designated pJ2Mdel, was transfected into OK cells, luciferase activity was reduced by 60% relative to pJ2M (Fig. 7A).
To further define the positive regulatory element within the 280-bp promoter sequence of functional importance, deletion analysis of this segment was carried out. The hPCLN-1 promoter sequence from positon –1738 to –1493 was 5'-truncated in increments of 35 bp. These progressively shorter DNA fragments were cloned upstream to a luciferase reporter gene in SV40 minimal promoter-containing vectors, pGL3-promoter, and transfected into OK cells. Figure 7B illustrates that luciferase activity was highest with the construct containing the hPCLN-1 promoter sequence from position –1703 to –1493. Truncation of the promoter to position –1668 caused little change in reporter activity. However, when the promoter was truncated to position –1633, reporter activity diminished by 50%. Two more sequential 35-bp deletions caused only minor reductions in luciferase activity.
Taken together, these experiments suggest that the 70-bp DNA sequence between positions –1703 and –1633, which binds nuclear proteins from renal cells, contains a positive regulatory element involved in transcriptional activation. Computer analysis of this segment, revealed a single PPRE at position –1655, which is known to bind the transcription factor PPAR (11, 15, 20).
Most PPREs described so far consist of a direct repeat of two hexamer half-sites separated by several nucleotides (14). The first half-site is highly conserved (AGGTCA) whereas the second is not. The sequence of the PPRE identified in the hPCLN-1 promoter contains one half-site, which is a perfect consensus sequence, but no clearly recognizable second half-site. This sequence, however, has been shown to enable transcription factor binding and activation (3).
Several in vivo (37, 44) and in vitro (7, 30) studies have demonstrated that Mg2+ restriction increases Mg2+ transport in the renal tubule whereas Mg2+ load decreases it (22, 31). To explore the molecular mechanisms of this modulation, we examined the effect of changes in ambient Mg2+ concentration on the activity of the hPCLN-1 promoter. For this purpose, OK cells were transiently transfected with the 2.5-kb hPCLN-1 promoter fragment (pJ2M) driving maximal luciferase activity. Following transfection, the cells were exposed for 24 h to media containing 0 (low), 0.7 (normal), or 1.5 mM (high) Mg2+. As demonstrated in Fig. 8, Mg2+ restriction caused a 52% increase, and Mg2+ load a 72% decrease, in luciferase activity in OK cells compared with conditions of normal Mg2+ concentration. These findings suggest that ambient Mg2+ concentration may modulate Mg2+ transport at the transcriptional level.
Effect of 1,25(OH)2 vitamin D on the PPRE-containing DNA region in the hPCLN-1 promoter. Since the vitamin D receptor (VDR) and PPAR both belong to the nuclear receptor superfamily of transcription factors and thus share several biological characteristics (6, 15, 26), we examined the effect of 1,25(OH)2 vitamin D on the hPCLN-1 promoter sequence containing the PPRE half-site. We first examined the action of 1,25(OH)2 vitamin D on the 2.5-kb hPCLN-1 promoter fragment (pJ2M). For this purpose, OK cells, transiently transfected with this fragment were exposed to 5 x 10–7 M 1,25(OH)2 vitamin D for 16–20 h, which caused promoter activity to decrease by 56% (Fig. 9A). Experimental conditions were selected based on preliminary experiments with varying 1,25(OH)2 vitamin D concentrations (10–6 to 10–7M) (data not shown).
We next examined whether the effect of 1,25(OH)2 vitamin D on the promoter was PPRE dependent. For this purpose, OK cells transfected with pGL3-promoter vector containing a 210-bp hPCLN-1 promoter fragment (pJ210) corresponding to the hPCLN-1 promoter sequence between positions –1703 and –1493 harboring the PPRE site, or a 70-bp 5' deletion product (pJ140) of this same promoter sequence, between positions –1633 and –1493, lacking the PPRE, were exposed to 1,25(OH)2 vitamin D under similar conditions. 1,25(OH)2 vitamin D treatment reduced the activity of the PPRE-containing promoter fragment by 52% compared with control non-1,25(OH)2 vitamin D-treated cells (Fig. 9B). This effect of 1,25(OH)2 vitamin D was markedly diminished in cells transfected with the PPRE-lacking fragment. Moreover, the actual difference in promoter activity between the PPRE-containing and PPRE-lacking fragments decreased fivefold following 1,25(OH)2 vitamin D treatment (Fig. 9B).
To further establish the role of PPRE in hPCLN-1 promoter activity, three bases of the PPRE half-site were point-mutated in both pJ2M and pJ210. OK cells transfected with these constructs were exposed to 1,25(OH)2 vitamin D as above. As shown in Fig. 10A, the decrease in pJ2M activity following 1,25(OH)2 vitamin D treatment was greatly diminished in cells transfected with the mutated promoter. When a similar experiment was carried out with mutated pJ210 (Fig. 10B), the 1,25(OH)2 vitamin D-induced reduction in luciferase activity seen in cells transfected with wild-type, PPRE-containing pJ210 was completely abolished once the PPRE site was mutated.
These experiments suggest that 1,25(OH)2 vitamin D modulates hPCLN-1 activity by a mechanism that appears to involve the PPRE half-site in the gene promoter.
DISCUSSION
In this study, we describe the transcriptional analysis of the promoter of the human paracellin-1 gene. We demonstrate that a 70-bp region between positions –1633 and –1703 may play a key role in the activity of the hPCLN-1 promoter (Figs. 6 and 7). Furthermore, we provide evidence that an interplay between a positive regulatory element located within this 70-bp region, and a more distally located negative regulatory element on the 5'-flanking region of the paracellin-1 gene, may determine renal cell-specific expression of this gene (Fig. 4). In addition, we demonstrate that ambient Mg2+ concentration affects hPCLN-1 promoter activity (Fig. 8). Finally, we show that 1,25(OH)2 vitamin D modulates hPCLN-1 promoter activity via this 70-bp, PPRE-containing, hPCLN-1 promoter fragment (Figs. 9 and 10), thereby suggesting that paracellin-1 gene expression is subject to PPAR/PPRE-mediated transcriptional regulation by this hormone.
Reabsorption of solutes across the tubular epithelial layer depends on two separate routes: a transcellular pathway and a paracellular pathway (1, 47). The paracellular pathway is regulated by tight junctions, which form a barrier to the diffusion of solutes across epithelial cells and function as a boundary between the apical and basolateral membranes maintaining epithelial cell polarity (21). Paracellin-1, which appears to regulate the paracellular transport of Mg2+ in the TAL, is the first tight junction protein reported to be involved in ion resorption (38). Paracellin-1, or claudin 16, (38) belongs to the claudin family of proteins that participate in the formation of tight junction strands in various tissues (27) and are thought to have a major role in regulating the magnitude and nature of paracellular permeability (1). The disease mutations found in paracellin-1 in familial hypomagnesemia-hypercalciuria syndrome result in an increased resistance to the electrical seal of tight junctions, thereby decreasing epithelial ionic permeability and selectively impeding magnesium and calcium reabsorption (38, 47). Recently, it has been reported that deletion of the paracellin-1 gene is responsible for renal tubular dysplasia in cattle (29). This finding suggests that paracellin-1, in addition to its function in ion resorption, may play an important role in the normal development and organization of the renal tubule.
hPCLN-1 mRNA was found to be expressed exclusively in the kidney (38, 42). Although cis-acting regulatory elements involved in regulating gene transcription may be dispersed throughout the gene locus, often the proximal promoter region contains elements sufficient for high levels of tissue-specific gene transcription. These elements may function as binding sites for tissue-restricted proteins that arbitrate transcriptional activation in expressing cells. In an effort to identify cis-acting elements involved in regulating kidney-specific expression of hPCLN-1, we focused our study on the 5'-flanking region of this gene. We isolated a 7.5-kb genomic sequence corresponding to the 5'-flanking region of hPCLN-1 and, using reporter gene assays, showed that the proximal 2.5-kb region contained cis-acting, positive and negative, regulatory elements that play a role in renal epithelial-specific expression of hPCLN-1. This promoter displayed high activity in PCLN-1-expressing renal cell lines but not in the fibroblast cell line NIH3T3. In the rabbit (38) and rat (42) kidney, PCLN-1 expression has been shown to be restricted to the TAL and the DCT. Several renal cell lines were used in this study. These included HEK293 and MDCT cells, as well as OK cells of proximal tubular origin. These renal cells were shown to express paracellin-1 mRNA (Fig. 2A) and protein (Fig. 2B) and to display hPCLN-1 promoter-driven luciferase activity (Fig. 3) as opposed to mouse embryonic fibroblast cells (NIH3T3), which demonstrated no expression/activity, indicating that the PCLN-1 promoter was kidney specific.
In our study, OK cells displayed the highest level of hPCLN-1 promoter-driven reporter gene activity (Fig. 3). Hence, these cells served as the recipient cells in most of our transfection experiments. It is not entirely clear why OK cells of proximal tubular origin express paracellin-1, which was found only in the TAL and the DCT of the rabbit and the rat. Several considerations could provide an explanation for this finding. First, the nephron segment-specific expression of the paracellin-1 gene has been examined in the kidney of these two rodents only. It is possible that the expression of the paracellin-1 gene in other animals and species, including humans, extends to the tight junction of more proximal nephron segments. Second, although several properties of OK cells are consistent with a proximal tubular site of origin, this cell line was originally derived from the whole kidney and may possess characteristics of more distal regions of the nephron (10, 19). Finally, in the immature rat, Mg2+ reabsorption occurs predominantly via the paracellular pathway in the proximal tubule rather than in the loop of Henle (23, 30). It is possible that the OK cell line used in our study contains cells with characteristics of the immature proximal tubule, including expression and cell-specific activity of genes, such as PCLN-1, participating in Mg2+ reabsorption in the immature proximal tubule tight junction. Taken together, we believe that the expression of paracellin-1 mRNA and protein in OK cells, coupled with the marked increase in reporter gene activity driven by the hPCLN-1 promoter in this cell line, has made OK cells suitable as experimental cells in our study.
Deletion analysis of reporter gene constructs containing hPCLN-1 promoter DNA suggested that hPCLN-1 transcription may be regulated by a proximal positive regulatory element positioned between –1982 and –1458 and a more distally located negative regulatory element between positions –3317 and –2554 (Fig. 4). EMSA studies of the presumed positive regulatory element-bearing region indicated that OK cells contain nuclear proteins that bind specifically to this functionally important region of the hPCLN-1 promoter (Fig. 6). NIH3T3 nuclear proteins did not bind to this DNA region. These proteins are potentially involved in tissue-specific expression of hPCLN-1. When this protein-binding DNA region was deleted from the 2.5-kb PCLN-1 promoter, a 60% reduction in promoter activity occurred. Detailed analysis revealed a 70-bp sequence within this DNA region, which seems to be responsible, at least in part, for modulating kidney-specific hPCLN-1 transcription (Fig. 7). Computer analysis of this segment disclosed a single PPAR binding site (PPRE).
Tubular Mg2+ reclamation is modulated by a variety of hormonal and nonhormonal factors (8, 16, 30). Mg2+ restriction and Mg2+ load influence both transcellular Mg2+ transport in the DCT and paracellular Mg2+ flux in the TAL (7, 22, 31, 37, 44). Micropuncture experiments in the rat nephron have demonstrated that a low-Mg2+ diet leads to urinary retention of Mg2+ due to increased Mg2+ reabsorption in the loop of Henle (37, 44). Culturing MDCT cells in Mg2+-free medium increased their Mg2+ transport rate (7, 30). Pretreatment of MDCT cells with actinomycin D, an inhibitor of transcription, resulted in a significant decrease in this adaptive response, suggesting that the adaptive regulation of Mg2+ depletion may involve gene transcription (30). As opposed to the effect of Mg2+ deprivation, Mg2+ infusion (22) and acute elevation of Mg2+ concentration at the contraluminal membrane of the TAL (31) in rats inhibited Mg2+ resorption in this tubule segment. The molecular mechanisms underlying the adaptive response of tubular Mg2+ transport to changes in Mg2+ levels have not been investigated. In this study, we show that ambient Mg2+ concentration affects hPCLN-1 promoter activity in OK cells. Specifically, Mg2+ restriction increases hPCLN-1 promoter activity, whereas Mg2+ load reduces the activity of this promoter. These findings are in concert with the modulatory effects of Mg2+ restriction and load on Mg2+ transport observed at the tubular and cellular levels (7, 22, 31, 37, 44) and demonstrate, for the first time, that changes in Mg2+ availability may influence Mg2+ transport at the transcriptional level. The exact molecular mechanisms whereby this Mg2+ level-induced effect is achieved remain unknown. The possible involvement of the cell membrane-bound Ca2+/Mg2+-sensing receptor, or the potential role of a hypothetical Mg2+ response element residing on the hPCLN-1 promoter in the Mg2+-induced effect on transcription of this gene, remains to be explored.
Paracellular Mg2+ transport in the TAL is known to be regulated by several hormones. These include parathyroid hormone, arginine vasopressin, aldosterone, insulin, and 1,25(OH)2 vitamin D, among others (30). Using microperfusion studies of isolated mouse TAL segments, it has been shown that most of these hormonal responses are mediated by changing the transepithelial voltage or by altering the permeability of the paracellular pathway (24, 30, 45). However, the exact molecular mechanisms of this hormone-induced effect on Mg2+ transport have not been investigated. 1,25(OH)2 vitamin D has been shown to increase Mg2+ uptake by MDCT cells (32). The effect of this hormone on paracellular Mg2+ transport has not been explored. Several hormones, including 1,25(OH)2 vitamin D (6), exert many of the biological actions by receptor-mediated effects on gene transcription. In this study, we show that 1,25(OH)2 vitamin D decreases hPCLN-1 promoter-driven luciferase activity (Figs. 9 and 10). OK cells are known to harbor vitamin D receptors (18) and, as discussed above, appear to express the paracellin-1 gene as well as possess the machinery necessary for its activity. Hence, this cell line may serve as an excellent model with which to explore the effect of this hormone on the paracellin-1 gene. Taken together, our findings provide the first direct evidence that paracellular, paracellin-1-mediated Mg2+ transport may be regulated at the transcriptional level by 1,25(OH)2 vitamin D. However, the physiological significance of the transcription-inhibiting effect of 1,25(OH)2 vitamin D shown in our study and its role in overall magnesium transport in the renal tubule remain to be clarified.
The effect of 1,25(OH)2 vitamin D on the hPCLN-1 promoter was demonstrated in our study to involve the PPRE half-site located between positions –1633 and –1703 within this promoter (Figs. 9 and 10). The PPARs comprise a group of transcription factors that belong to the nuclear receptor superfamily to which the VDR, the thyroid hormone receptor, and the all-trans-retinoic acid receptor also belong (11, 15, 20). PPARs are ligand-regulated transcription factors that control gene expression by binding to specific response elements (PPREs) within promoters, having formed heterodimers with the retinoid X receptor (11, 15, 20). Three PPAR isoforms, PPAR, PPAR, and PPAR, have been identified (11, 15) and are expressed in several tissues including the kidney (11, 48). PPARs participate in a variety of biological processes common to various cell types, including lipid metabolism, glucose homeostasis, inflammation, cell proliferation and differentiation, apoptosis, and early development (4, 11). However, despite their abundant expression in various segments of the renal tubule (11, 48), very little is known about specific tubular transport processes that are controlled by the PPARs. Noteworthy is a recent study demonstrating enhanced renal tubular cell surface expression of the epithelial sodium channel in response to PPAR activation (12).
In our study, we focused on 1,25(OH)2 vitamin D, known to modulate Mg2+ transport on one hand (30, 32) and to interact with the PPAR system on the other. The DNA binding site of the VDR is 46% homologous to the DNA binding site of PPAR, and both receptors are known to heterodimerize with the retinoid X receptor before binding to their specific DNA motifs (6, 15, 26). However, recent studies suggest that there is considerable flexibility in the binding sites recognized by the VDR, including the existence of some single half-sites (17). Furthermore, the outcome of the interaction between the VDR and its binding site may be an increase (5) or a decrease (25) in gene transcription. Of note is a recent study demonstrating that the VDR represses transcriptional activity of PPAR in COS1 cells (34). In our study, we provide evidence for PPRE-dependent transcriptional regulation of hPCLN-1 by 1,25(OH)2 vitamin D, and we show here, for the first time, that the PPAR/PPRE axis may play a specific role in the hormonal regulation of Mg2+ transport in the renal tubule.
In conclusion, an interplay between a positive regulatory element and a more distally located negative regulatory element on the 5'-flanking region of the paracellin-1 gene may determine renal epithelial-specific activity of this gene. The 70-bp, PPRE-containing, DNA region between positions –1633 and –1703 may play a key role in the activity of the PCLN-1 promoter. Ambient Mg2+ concentration and 1,25(OH)2 vitamin D may modulate paracellular, paracellin-1-driven, Mg2+ transport at the transcriptional level. 1,25(OH)2 vitamin D exerts its activity on the hPCLN-1 promoter likely via the PPRE site.
Future studies utilizing transgenic animals may verify whether the activity of this 70-bp DNA fragment, examined in the cell culture model, demonstrates tubular epithelial specifity in the intact organism, where the transgene is exposed to a more physiologically relevant environment (i.e., hormones, proteins, etc.). Once established, it is possible that these transgenic animals will serve as an excellent model with which to study the transcriptional effects of a variety of physiological as well as pathophysiological factors on the paracellin-1 gene, and thus on tubular Mg2+ reabsorption.
GRANTS
This work was supported by the Ruth and Allen Ziegler Fund for Pediatric Research and the Kronovet Fund for Medical Research, Technion-Israel Institute of Technology.
ACKNOWLEDGMENTS
We thank Dr. Karl Skorecki, Dr. Maty Tzukerman, and Dr. Sara Selig for helpful suggestions and Ayal Kelmachter and Adva Hermoni for technical assistance. We thank Judi Fichman and Hagar Shafrir for expert secretarial assistance.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
E. Efrati and Julia Arsentiev-Rozenfeld contributred equally to this work.
REFERENCES
Anderson JM. Molecular structure of tight junctions and their role in epithelial transport. News Physiol Sci 16: 126–130, 2001.
Bai Y, Pontoglio M, Hiesberger T, Sinclair AM, and Igarashi P. Regulation of kidney-specific Ksp-cadherin gene promoter by hepatocyte nuclear factor-1. Am J Physiol Renal Physiol 283: F839–F851, 2002.
Barrero MJ, Camarrero N, Marrero PF, and Haro D. Control of human carnitine palmitoyltransferase II gene transcription by peroxisome proliferator-activated receptor through a partially conserved peroxisome proliferator-responsive element. Biochem J 369: 721–729, 2003.
Berger J and Moller DE. The mechanisms of action of PPARs. Annu Rev Med 53: 409–435, 2002.
Breen EC, Van Wijen AJ, Lian JB, Stein GS, and Stein JL. In vivo occupancy of vitamin D responsive element in the osteocalcin gene supports vitamin D-dependent transcriptional upregulation in intact cells. Proc Natl Acad Sci USA 91: 12902–12906, 1994.
Brown AJ, Dusso A, and Slatopolsky E. Vitamin D. Am J Physiol Renal Physiol 277: F157–F175, 1999.
Dai LJ, Raymond L, Friedman PA, and Quamme GA. Mechanisms of amiloride stimulation of Mg2+ uptake in immortalized mouse distal convoluted tubule cells. Am J Physiol Renal Physiol 272: F249–F256, 1997.
Dai LJ, Ritchie G, Kerstan D, Kang HS, Cole DE, and Quamme GA. Magnesium transport in the renal distal convoluted tubule. Physiol Rev 81: 51–84, 2001.
Dignam JD, Lebovitz RM, and Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res 11: 1475–1489, 1983.
Gomes P, Xu J, Serrao P, Doria S, Jose PA, and Soares-da-Silva P. Expression and function of sodium transporters in two opossum kidney cell clonal sublines. Am J Physiol Renal Physiol 283: F73–F85, 2002.
Guan Y and Breyer MD. Peroxisome proliferator-activated receptors (PPARs): novel therapeutic targets in renal disease. Kidney Int 60: 14–30, 2001.
Hong g Lockhart A, Davis B, Rahmoune H, Baker S, Ye L, Thompson P, Shou Y, O’Shaughnessy K, Ronco P, and Brown J. PPAR activation enhances cell surface ENac via up-regulation of SGK1 in human collecting duct cells. FASEB J 17: 1966–1968, 2003.
Igarashi P, Cooduvalli S, Shashikant R, Thomson RB, Whyte DA, Liu-Chen S, Ruddle FH, and Aronson PS. Ksp-cadherin gene promoter. II. Kidney-specific activity in transgenic mice. Am J Physiol Renal Physiol 277: F599–F610, 1999.
Ijpenberg A, Jeannin E, Wahli W, and Desvergne B. Polarity and specific sequence requirements of peroxisome proliferator-activated receptor (PPAR)/retinoid X receptor heterodimer binding to DNA. A functional analysis of the malic enzyme gene PPAR response element. J Biol Chem 272: 20108–20117, 1997.
Issemann I and Green S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature 347: 645–650, 1990.
Kelepouris E and Agus ZS. Hypomagnesemia: renal magnesium handling. Semin Nephrol 18: 58–73, 1998.
Koszewski NJ, Kiessling S, and Malluche H. Isolation of genomic DNA sequences that bind vitamin D receptor complexes. Biochem Biophys Res Commun 283: 188–194, 2001.
Koszewski NJ and Rowan A. Heterodimer requirement for gene regulation by vitamin D in variant OK cells. Steroids 68: 307–314, 2003.
Koyama H, Goodpasture C, Miller MM, Teplitz RL, and Riggs AD. Establishment and characterization of a cell line from the American opossum (Didelphys virginiana). In Vitro 14: 239–246, 1978.
Laudet V, Hanni C, Coll J, Catzeflis F, and Stehelin D. Evolution of the nuclear receptor gene superfamily. EMBO J 11: 1003–1013, 1992.
Le Grimellec C, Friedlander G, el Yandouzi EH, Zlatkine P, and Giocondi MC. Membrane fluidity and transport properties in epithelia. Kidney Int 42: 825–836, 1992.
Le Grimellec C, Roinel N, and Morel F. Simultaneous Mg, Ca, P, K, Na and Cl analysis in rat tubular fluid. II. During acute Mg plasma loading. Pflügers Arch 340: 197–210, 1973.
Lelievre-Pegorier M, Merlet-Benichou C, Roinel N, and de Rouffignac C. Developmental pattern of water and electrolyte transport in rat superficial nephrons. Am J Physiol Renal Fluid Electrolyte Physiol 245: F15–F21, 1983.
Mandon B, Siga E, Chabardes D, Firsov D, Roinel N, and De Rouffignac C. Insulin stimulates Na+, Cl–, Ca2+, and Mg2+ transports in TAL of mouse nephron: cross-potentiation with AVP. Am J Physiol Renal Fluid Electrolyte Physiol 265: F361–F369, 1993.
Meszaros JG, Karin NJ, Akanbi K, and Farach-Carson MC. Down-regulation of L-type Ca2+ channel transcript levels by 1,25-dihydroxyvitamin-D3. J Biol Chem 271: 32981–32985, 1996.
Misra P, Owuor ED, Li W, Yu S, Qi C, Meyer K, Zhu YJ, Rao MS, Kong AN, and Reddy JK. Phosphorylation of transcriptional coactivator peroxisome proliferator-activated receptor (PPAR)-binding protein (PBP). Stimulation of transcriptional regulation by mitogen-activated protein kinase. J Biol Chem 277: 48745–48754, 2002.
Morita K, Furuse M, Fujimoto K, and Tsukita SH. Claudin multigene family encoding four-transmembrane domain protein components of tight junction strands. Proc Natl Acad Sci USA 96: 511–516, 1999.
Nelson RD, Stricklett P, Gustafson C, Stevens A, Ausiello D, Brown D, and Kohan DE. Expression of an AQP2 Cre recombinase transgene in kidney and male reproductive system of transgenic mice. Am J Physiol Cell Physiol 275: C216–C226, 1998.
Ohba Y, Kitagawa H, Kitoh K, Sasaki Y, Takami M, Shinkai Y, and Kunieda T. A deletion of the paracellin-1 gene is responsible for renal tubular dysplasia in cattle. Genomics 68: 229–236, 2000.
Quamme GA. Renal magnesium handling: new insights in understanding old problems. Kideny Int 52: 1180–1195, 1997.
Quamme GA and Dirks JH. Intraluminal and contraluminal magnesium on magnesium and calcium transfer in the rat nephron. Am J Physiol Renal Fluid Electrolyte Physiol 238: F187–F198, 1980.
Ritchie G, Kerstan D, Dai LJ, Kang HS, Canaff L, Hendy GN, and Quamme GA. 1,25(OH)2D3 stimulates Mg2+ uptake into MDCT cells: modulation by extracellular Ca2+ and Mg2+. Am J Physiol Renal Physiol 280: F868–F878, 2001.
Romani A and Scarpa A. Regulation of cell magnesium. Arch Biochem Biophys 298: 1–12, 1992.
Sakuma T, Miyamoto T, Jiang W, Kakizawa T, Nishio S, Suzuki S, Takeda T, Oiwa A, and Hashizume K. Inhibition of peroxisome proliferator-activated receptor signaling by vitamin D receptor. Biochem Biophys Res Commun 312: 513–519, 2003.
Sambrook J, Fritsh EF, and Maniatis T. Molecular Cloning: A Laboratory Manual (2nd ed.). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989.
Shachaf C, Skorecki KL, and Tzukerman M. Role of AP2 consensus sites in regulation of rat Npt2 (sodium-phosphate cotransporter) promoter. Am J Physiol Renal Physiol 278: F406–F416, 2000.
Shafik IM and Quamme GA. Early adaptation of renal magnesium reabsorption in response to magnesium restriction. Am J Physiol Renal Fluid Electrolyte Physiol 257: F974–F977, 1989.
Simon DB, Lu Y, Chaote KA, Velazquez H, Al-Sabban E, Praga M, Casari G, Bettinelli A, Colussi G, Rodriguez-Soriano J, McCredie D, Milford D, Sanjad S, and Lifton RP. Paracellin-1, a renal tight junction protein required for paracellular Mg2+ resorption. Science 285: 103–106, 1999.
Taniyama Y, Sato K, Sugawara A, Urono A, Ikeda Y, Kudo M, Ito S, and Takeuchi K. Renal tubule-specific transcription and chromosomal localization of rat thiazide-sensitive Na-Cl cotransporter gene. J Biol Chem 276: 26260–26268, 2001.
Uchida S, Rai T, Yatsushige H, Matsumura Y, Kawasaki M, Sasaki S, and Marumo F. Isolation and characterization of kidney cell-specific ClC-K1 chloride channel gene promoter. Am J Physiol Renal Physiol 274: F602–F610, 1998.
Uchida S, Tanaka Y, Ito H, Saitoh-Ohara F, Inazawa J, Yokoyama KK, Sasaki S, Marumo F. Transcriptional regulation of the CLC-K1 promoter by myc-associated zinc finger protein and kidney-enriched Kruppel-like factor, a novel zinc finger repressor. Mol Cell Biol 20: 7319–7331, 2000.
Weber S, Schlingmann KP, Peters M, Nejsum LN, Nielsen S, Engel H, Grzeschik KH, Seyberth HW, Grone HJ, Nusing R, and Konrad M. Primary gene structure and expression studies of rodent paracellin-1. J Am Soc Nephrol 12: 2664–2667, 2001.
Whyte DA, Li C, Thomson RB, Nix SL, Zanjani R, Karp SL, Aronson PS, and Igarashi P. Ksp-cadherin gene promoter. I. Characterization and renal epithelial cell-specific activity. Am J Physiol Renal Physiol 277: F587–F598, 1999.
Wittner M, Jounier S, Deschenes G, de Rouffignac C, and Di Stefano A. Cellular adaptation of the mouse cortical thick ascending limb of Henle’s loop (CTAL) to dietary magnesium restriction: enhanced transepithelial Mg2+ and Ca2+ transport. Pflügers Arch 439: 765–771, 2000.
Wittner M, Mandon B, Roinel N, de Rouffignac C, and Di Stefano A. Hormonal stimulation of Ca2+ and Mg2+ transport in the cortical thick ascending limb of Henle’s loop of the mouse: evidence for a change in the paracellular pathway permeability. Pflügers Arch 423: 387–396, 1993.
Wong MA, Cui S, and Quaggin SE. Identification and characterization of a glomerular-specific promoter from the human nephrin gene. Am J Physiol Renal Physiol 279: F1027–F1032, 2000.
Wong V and Goodenough DA. Paracellular channels (Abstract). Science 285: 52, 1999.
Yang T, Michele DE, Park J, Smart AM, Lin Z, Brosius 3rd FC, Schnermann JB, and Briggs JP. Expression of peroxisomal proliferator-activated receptors and retinoid X receptors in the kidney. Am J Physiol Renal Physiol 277: F966–F973, 1999.(Edna Efrati , Julia Arsen)
ABSTRACT
Tubular reabsorption of Mg2+ is mediated by the tight junction protein paracellin-1, which is encoded by the gene PCLN-1 (CLDN16) and exclusively expressed in the kidney. Tubular Mg2+ reclamation is modulated by many hormones and factors. The aim of this study was to define regulatory elements essential for renal tubular cell-specific expression of human PCLN-1 (hPCLN-1) and to explore the effect of Mg2+ transport modulators on the paracellin-1 gene promoter. Endogenous paracellin-1 mRNA and protein were detected in renal cell lines opossom kidney (OK), HEK293, and MDCT, but not in the fibroblast cell line NIH3T3. A 7.5-kb hPCLN-1 5'-flanking DNA sequence along with seven 5'-deletion products were cloned into luciferase reporter vectors and transiently transfected into the renal and nonrenal cells. The highest levels of luciferase activity resulted from transfection of a 5'-flanking 2.5-kb fragment (pJ2M). This activity was maximal in OK cells, was orientation dependent, and was absent in NIH3T3 cells. Mg2+ deprivation significantly increased pJ2M-driven activity in transfected OK cells, whereas Mg2+ load decreased it compared with conditions of normal Mg2+. Deletion analysis along with electrophoretic mobility-shift assay demonstrated that OK cells contain nuclear proteins, which bind a 70-bp region between –1633 and –1703 of major functional significance. Deleting this 70-bp segment, which contains a single peroxisome proliferator-response element (PPRE), or mutating the PPRE, caused a 60% reduction in luciferase activity. Stimulating the 70-bp sequence with 1,25(OH)2 vitamin D decreased luciferase activity by 52%. This effect of 1,25(OH)2 vitamin D was abolished in the absence of PPRE or in the presence of mutated PPRE. We conclude that the PPRE within this 70-bp DNA region may play a key role in the cell-specific and regulatory activity of the hPCLN-1 promoter. Ambient Mg2+ concentration and 1,25(OH)2 vitamin D may modulate paracellular, paracellin-1-mediated, Mg2+ transport at the transcriptional level. 1,25(OH)2 vitamin D exerts its activity on the hPCLN-1 promoter likely via the PPRE site.
magnesium; renal tubule; transcription; promoter; gene expression; gene regulation; peroxisome proliferator response element; 1,25(OH)2 vitamin D
MAGNESIUM IS THE MOST ABUNDANT divalent cation in the intracellular fluid. It plays a critical role in a wide variety of metabolic and cellular processes, including cellular energy storage, DNA/RNA processing, ion transport, membrane stabilization, and nerve conduction (33). Abnormalities in Mg2+ homeostasis are relatively common in clinical practice and may lead to neuromuscular disturbances, central nervous system manifestations, and cardiovascular dysfunction (16, 30). In mammals, the kidney is the principal organ responsible for Mg2+ balance (16, 30). Normally, >95% of the filtered Mg2+ is reabsorbed by the renal tubule. Ten to fifteen percent of the filtered Mg2+ is reabsorbed in the proximal tubule and 10% in the distal tubule. The major site of Mg2+ reabsorption is the thick ascending limb of the loop of Henle (TAL), where 60–70% of the filtered load is reclaimed (16, 30). Mg2+ transport in this tubule segment occurs primarily through paracellular conductance driven by the lumen positive electrical potential (30). While renal Mg2+ handling has been thoroughly investigated at the tubular and cellular levels (8, 16, 30), the molecular mechanisms of tubular Mg2+ reabsorption are poorly understood.
Recently, Simon et al. (38) using positional cloning, have identified a human gene, hPCLN-1 (also known as CLDN16, NCBI accession no. NM-006580), mutations in which cause familial hypomagnesemia-hypercalciuria syndrome. hPCLN-1 consists of five exons and resides on chromosome 3q27. The gene encodes a protein, paracellin-1, which is composed of 305 amino acids (38). Northern blot analysis of human tissues has shown that the 3.5-kb PCLN-1 mRNA transcript is expressed exclusively in the kidney (38). RT-PCR analysis of mRNA from nephron segments of the rabbit (38) and rat (42) has demonstrated that PCLN-1 is expressed in the TAL and the distal convoluted tubule (DCT). The paracellin-1 protein, which is located in the paracellular tight junctions of the TAL and DCT, is a member of the claudin family of tight junction proteins (27) and appears to mediate resorption of both Mg2+ and Ca2+ (38).
In the kidney, the specialized reabsorptive and/or secretory function of each tubule segment depends upon its structural arrangement and upon the specific pattern of gene expression in each tubular cell type. The promoters of several transporter and channel genes including aquaporin (28), the Na+-phosphate cotransporter (36), the Na+-K+-Cl– cotransporter (39), and chloride channels (40) as well as the promoters of nephrin (46) and cadherin (13, 43) genes have been cloned and shown to direct kidney-specific expression in vitro and/or in transgenic mice. Several transcription factors including myc-associated zinc finger proteins and Krüppel-like factor (41), hepatocyte nuclear factor-3 (39), and hepatocyte nuclear factor-1 (2), were found to be involved in kidney-specific expression of the ClC-K1 chloride channel, thiazide-sensitive Na-Cl cotransporter, and cadherin genes, respectively. However, very little is known about the regulatory elements responsible for cell-specific expression of transporter and channel genes in the kidney. As yet, the promoter region of the human paracellin-1 gene has not been characterized, and the molecular mechanisms for renal epithelial-specific activity of this gene have not been investigated.
Many factors are known to modulate Mg2+ reabsorption in various nephron segments. These factors include hormones such as insulin, 1,25(OH)2 vitamin D, aldosterone, and parathyroid hormone as well as nonhormonal factors such as Mg2+ restriction and load and acid-base changes (8, 30). Most of these factors influence both transcellular magnesium transport in the DCT as well as paracellular transport of this cation in the TAL (8, 30). However, very little is known about the molecular mechanisms of this modulation and whether it occurs at the protein or DNA/RNA level. The magnesium restriction (30)- and 1,25(OH)2 vitamin D (32)-induced increase in Mg2+ uptake by renal MDCT cells was diminished by pretreatment of cells with actinomycin D, suggesting that this stimulation occurs through transcriptional activation. Nevertheless, the molecular mechanisms whereby hormones and other factors modulate Mg2+ transport across the paracellular pathway in the renal tubule are unknown.
The purpose of this study was to define cis-acting promoter regulatory elements and to examine trans-acting factors essential for renal tubular epithelial-specific expression of hPCLN-1. We also explored the effect of modulators of Mg2+ transport on the hPCLN-1 gene promoter. We show that a 70-bp 5'-flanking region of the paracellin-1 gene may determine renal epithelial-specific expression of this gene. We demonstrate an increase in hPCLN-1 promoter activity in response to Mg2+ depletion and a decrease in response to Mg2+ load. In addition, we show that 1,25(OH)2 vitamin D may modulate Mg2+ transport at the transcriptional level, probably via the peroxisome proliferator response element (PPRE) contained within the 70-bp region.
MATERIALS AND METHODS
Opossum kidney (OK) cells (provided by Dr. J Green, Technion, Haifa, Israel), human embryonic kidney (HEK293) cells (provided by Dr. K. Skorecki, Technion, Haifa, Israel), mouse distal convoluted tubule (MDCT) cells (provided by Dr. P. Friedman, University of Pittsburgh, Pittsburgh, PA), and mouse embryonic fibroblast (NIH3T3) cells were grown and maintained in DMEM/F-12 supplemented with 10% fetal calf serum, 2 mM glutamine, 50 IU/ml penicillin, and 50 μg/ml streptomycin at 37°C in a humidified atmosphere of 95% air-5% CO2.
RT-PCR followed by Southern blot analysis. Total RNA isolated with Tri-reagent (MRB, Cincinnati, OH) from the cell lines mentioned above was reverse-transcribed using OmniscriptRT (Qiagen, Hilden, Germany) with random hexamers (Promega, Madison, WI). PCR was performed using HotStarTaq DNA polymerase (Qiagen) with two sets of primers (Table 1). The first (F1,R1) complementary to exon 1 of hPCLN-1 and the second (F2,R2) complementary to a region between exons 3 and 5 of PCLN-1 (spanning two introns) highly homologous among human, rat, and mouse (mPCLN-1). The resultant DNA was separated on 1% agarose gels and transferred to nylon membranes (Osmonics, Minnetonka, MN). The membranes were probed with PCLN-1 cDNA probes generated by PCR using primers F1,R1 for hPCLN-1 and primers F3,R2 (Table 1) from exon 5 for mPCLN-1, with human genomic DNA as a template. A similar procedure was carried out with actin primers (F4,R4) as control. Probes were 32P labeled by random priming (Biological Industries, Beit Ha’Emek, Israel) using [-32P]dCTP (DuPont-New England Nuclear, Boston, MA).
View this table:
Western blot analysis. Protein extracts from HEK293, OK, and NIH3T3 cells were prepared using standard protocols (35). Proteins were separated on 10% SDS polyacrylamide gels, transferred to nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany), and probed with anti-human paracellin-1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) using chemiluminescense (Biological Industries). To verify antibody specificity, a peptide competition assay was carried out with 100-fold excess of paracellin-derived peptide.
Cloning of hPCLN-1 5'-flanking DNA. Using the published sequence of human paracellin-1 cDNA (NCBI accession no. NM-006580), we localized the hPCLN-1 gene on chromosome 3, drafts of which have recently been deposited in NCBI (accession no. NT-00962). A DNA fragment of 7.5 kb in the 5'-flanking sequence region, to but not including the PCLN-1 translation start site (Fig. 1), was synthesized using PCR with human genomic DNA as a template and primers F5/R5 (Table 1). To minimize the possibility of PCR-generated mutations, the High-Fidelity Long-Range PCR system (Roche, Mannheim, Germany) was used. The fragment was TA cloned into pCR-XL-TOPO vector (Invitrogen, Carlsbad, CA) and sequence verified. Computer analysis (Wisconsin Package version 8.0, Genetic Computer Group) of the 5'-flanking region of hPCLN-1 from the gene’s translation start site was carried out, and putative transcription factor binding sites were identified.
Generation of promoter/reporter constructs. The 7.5-kb 5'-flanking sequence described above was subcloned into pGL3-basic vector (Promega) upstream of a luciferase reporter gene (and named pJ12) (Fig. 1). In addition, a set of seven deletion fragments decreasing in size from the 5'-end of the 7,514-bp genomic fragment were produced and cloned into pGL3-basic vector. Five inserts were prepared using PCR-based strategies with genomic DNA as template with primers F6, F7, F8, F9, F10, and R5 (Table 1). The additional two were prepared by removing specific 5'-segments from cloned hPCLN-1 sequences in pGL3-basic, using restriction enzymes. All eight constructs were sequenced to verify the orientation and integrity of the inserts. As illustrated in Fig. 1, the 5'-ends of the deletion fragments were positioned at –4687 (pJ2L), –3986 (pJ2/3.9), –3317 (pJ2/3.3), –2554 (pJ2M), –1982 (pJ2/1.9), –1458 (pJ2.1/4), and –733 (pJ2.7), respectively. All fragments extended to the ATG site, but did not include it. All eight constructs were cloned into the vector in sense orientation, and the 2.5-kb fragment in pJ2M was also cloned in antisense orientation.
A reporter plasmid containing a 586-bp nested deletion (from position –1772 to –1186) within this 2.5-kb DNA region (termed pJ2Mdel) was generated using PCR with pJ2M as template and primers F12/R12 (Table 1) complementary to the gap ends. Following DpnI treatment, the PCR-generated pGL3-basic construct was ligated, and the sequence was verified. Empty pGL3 vector containing no insert was used as a negative control. pGL3-control containing the SV40 enhancer/promoter was used as a positive control. Promoter activity was estimated from the ratio of hPCLN-1-driven luciferase activity in pGL3-basic vector to promoterless pGL3-basic vector.
For reporter gene activity driven by distal hPCLN-1 promoter fragments, the DNA region from position –1738 to –1493 as well as a series of five deletion fragments 5'-truncated in increments of 35 bp were PCR generated with human genomic DNA and primers F13, F14, F15, F16, F17, F18, with R13. All six fragments were cloned into the pGL3-promoter vector, which carries an SV40 minimal promoter upstream to the luciferase reporter gene. In these experiments, promoter activity was determined from the ratio of hPCLN-1 promoter-driven luciferase activity in the pGL3-promoter vector to that in the empty pGL3-promoter vector.
Point mutations in the 2.5-kb insert-containing plasmid, pJ2M, and the 210-bp insert-containing plasmid, pJ210 (see RESULTS), were generated using PCR with the mutated primers F19 and R14 (Table 1).
In all transfection experiments, plasmid pCH110 (Pharmacia, Uppsala, Sweden), containing the LacZ gene driven by the CMV promoter, was used to normalize for transfection efficiency. DNA for transfections was purified using Nucleobond AX (Macherey Nagel, Duren, Germany).
Transient transfections and reporter gene assays. OK, HEK293, and MDCT cells were plated (5 x 104/dish) in 24-well dishes in serum-containing medium. Cotransfections were performed using Fugene 6 (1.2 μl/well, Roche) with 0.3 μg reporter plasmid and 0.3 μg pCH110. NIH3T3 cells (4 x 105/dish) were seeded in 6-well plates and incubated overnight at 37°C. Cells were transfected 24 h later using Polyfect (10 μl/well, Qiagen) with 0.75 μg reporter plasmid and 0.75 μg pCH110. All plates were incubated at 37°C for 48 h. For enzymatic assays, cells were washed with PBS (150 mM NaCl, 15 mM sodium phosphate, pH 7.3) and lysed by incubating in 200 μl/well M-Per (Pierce, Cheshire, UK) for 5 min at 37°C. Lysed cells were centrifuged, and the supernatant was aliquoted (50 μl/well) into 96-well plates. Fifty microliters of Luciferase Assay Reagent (Promega) were automatically added, and the light intensity of the reaction was immediately read in a luminometer (Lucy, Anthos, Austria) for a period of 10 s. Luciferase activity was normalized to -galactosidase activity, which was measured in identical cell lysates. One hundred sixty microliters of ONPG substrate (Sigma, St. Louis, MO) were added to 30 μl of cell lysate in each well of a 96-well plate. The reaction mixture was incubated for 30 min at 37°C, or until yellow color developed. -Galactosidase measurements were performed by a luminometer with a 405-nm filter. Measurements of luciferase and -galactosidase were performed in duplicate.
In some experiments, 48 h after transfection, the medium was replaced with fresh medium containing 0 (low), 0.7 (normal), or 1.5 mM (high) Mg2+ (Biological Industries). In other experiments, the fresh medium contained 1,25(OH)2 vitamin D (from a stock solution of 10–4 M in ethanol, Sigma) at a final concentration of 5 x 10–7 M. Control experiments were carried out with 0.5% ethanol. Cells were exposed to experimental media for 24 h.
EMSA. Nuclear extracts were prepared from cells using the method of Dignam (9). Briefly, confluent cells were grown on 100-mm plates, washed in 3 ml of PBS (150 mM NaCl, 15 mM sodium phosphate, pH 7.3) supplemented with protease inhibitor mix (Complete, Roche), scraped, and pelleted. The pellet was resuspended in ice-cold lysis buffer (in mM: 10 HEPES, pH 7.9, 1.5 MgCl2, 420 NaCl, 0.2 EDTA, and 1 DTT as well as protease inhibitor mix) and incubated at 4°C for 20 min. Following centrifugation, the supernatant, containing the nuclear extract, was diluted 1:2 with (in mM) 20 HEPES (pH 7.9), 100 KCl, 0.2 EDTA, and 1 DTT as well as 20% glycerol and protease inhibitors. Protein concentration was measured using Bradford reagent (Sigma) at 595 nm with bovine serum albumin as the standard.
Double-stranded oligonucleotides corresponding to promoter sequences of interest were end-labeled with [-32P]ATP (DuPont-New England Nuclear) using T4 polynucleotide kinase (NEB, Beverly, MA). Binding reactions contained (in mM) 10 Tris·HCl (pH 8.0), 250 KCl, 0.5 EDTA, and 0.2 DTT as well as 0.1% Triton-X 100, 12.5% glycerol (vol/vol), 1 μg poly-dIdC-labeled probe (5 x 104 counts/min), and 50 μg of nuclear extracts. In some reactions, a 50-fold molar excess of unlabeled double-stranded oligonucleotide was added for specific competition. Following a 30-min incubation period on ice, complexes were resolved on 4% nondenaturing polyacrylamide gels in 1x TBE buffer. The gels were dried and autoradiographed.
RESULTS
Cloning the 5'-flanking region of hPCLN-1. A DNA fragment stretching over 7.5 kb in the 5'-flanking region of hPCLN-1, reaching 9 bp from the gene’s translation start site, was isolated. Computer analysis of this fragment disclosed a variety of putative transcription factor binding sites, among them recognition sites for several transcription factors known to be tissue restricted and expressed in the kidney. These included hepatocyte nuclear factor 5 (HNF-5), GATA factors, and a peroxisone proliferator-activated receptor (PPAR) binding site. The hPCLN-1 promoter also contained binding sites for transcription factors involved in signal transduction, such as activator protein (AP)-1 and AP-3 as well as a putative TATA box.
The 5'-flanking regions of PCLN-1 in mouse and rat genomic DNA were located using NCBI-BLAST with the respective cDNA sequences. Comparison of the 5'-flanking region of hPCLN-1 to that in mouse (accession no. NW-000107, region: 23187252 ... 23287252) and rat (accession no. AC106700) revealed very little overall sequence identity within 2.4 kb 5' to the translation site. However, the 51-bp region from position –392 to –341, relative to the hPCLN-1 translation start site, was 79 and 81% identical to the mouse and rat PCLN-1 5'-flanking region sequences, respectively.
Comparison of the hPCLN-1 5'-flanking region to that of a variety of genes encoding various ion channels and transporters, known to be specifically expressed in the kidney, such as the voltage-gated Cl– channels ClC-K1 and ClC-K2, aquaporin-2, the K+ channel ROMK, Tamn-Horsfall protein, and the thiazide-sensitive NaCl cotransporter, did not reveal areas of sequence homology.
Kidney cell-specific expression of hPCLN-1. RT-PCR followed by Southern blot analysis with nested probes (see MATERIALS AND METHODS) detected endogenous PCLN-1 mRNA in HEK293 and OK cells using an hPCLN-1 probe and in HEK293 and MDCT cells using an mPCLN-1 probe (Fig. 2A). Paracellin-1 mRNA was not detected in NIH3T3 cells.
Western blot analysis with paracellin-1-specific antibody detected paracellin-1 protein expression in HEK293 and OK but not in NIH3T3 cells (Fig. 2B). Addition of 100-fold molar excess of paracellin-1 peptide to the antibody, 24 h before incubation with the blot, competed out the paracellin-1 band (data not shown).
Kidney cell-specific activity of the hPCLN-1 promoter. To verify that the 5'-flanking region of PCLN-1 contained a functional cell-specific promoter, reporter gene assays were conducted. The 7.5-kb hPCLN-1 5' flanking region, as well as three 5'-deletion products, were cloned upstream to a luciferase reporter gene in promoterless pGL3-basic vectors in sense orientation.
The resulting plasmids (Fig. 1), designated pJ12 (7,514-bp insert), pJ2L (4,687-bp insert), pJ2M (2,554-bp insert) and pJ2.7 (733-bp insert), were transfected into the PCLN-1-expressing cell lines HEK293, OK, and MDCT. Reporter gene activities were compared with those in transfected NIH3T3 cells, which do not express PCLN-1. As shown in Fig. 3, all three PCLN-1-expressing renal cell lines displayed a similar pattern of luciferase activity when transfected with pJ12, pJ2L, pJ2M, and pJ2.7. Luciferase activity was maximally induced by pJ2M (2.5 kb) and gradually diminished, as the insert size was either increased or decreased. Specifically, the lowest luciferase activity resulted from transfection of the 7.5-kb insert-containing plasmid (pJ12). Truncation of the 5'-flanking region from –7,514 to –4,687 bp (pJ2L) did not significantly increase luciferase activity. A further decrease in size from –4,687 to –2554 bp (pJ2M) caused a significant increase in luciferase activity in all three renal cell lines. When the 7.5-kb fragment in the luciferase vector was further shortened to 0.733 kb (pJ2.7), luciferase activity in all four renal cell lines was markedly reduced. Transfection of each of the four constructs into the control cell line NIH3T3 showed negligible induction of luciferase activity.
Taken together, these results indicated that the 2.5-kb hPCLN-1 promoter fragment cloned in pJ2M contained an active promoter. The lack of stimulation in NIH3T3 cells suggested that the activity of the hPCLN-1 promoter was kidney cell specific. The results also suggested that kidney-specific expression of the hPCLN-1 promoter was due, at least in part, to tissue-specific transcriptional regulation.
Although the induction pattern was similar in all three PCLN-1-expressing renal cell lines, its magnitude was very different between cells. The highest levels of induction appeared in OK cells and the lowest in MDCT cells. HEK293 cells displayed intermediate levels of induction. Based on these findings, OK cells were selected as experimental cells in our next set of experiments.
Deletion analysis of the hPCLN-1 promoter. To further explore the 5'-flanking region of hPCLN-1, a second set of deletion constructs was tested (Fig. 1), two between pJ2L and pJ2M (designated pJ2/3.9 and pJ2/3.3) and two between pJ2M and pJ2.7 (designated pJ2/1.9 and pJ/1.4). All eight constructs were transiently transfected into OK and NIH3T3 cells. As evident from Fig. 4, the highest level of luciferase activity resulted from transfection of the 2.5-kb insert-containing plasmid (pJ2M). Increasing the size of this fragment by 763 bp (pJ2/3.3) caused a 50% reduction in luciferase activity, which remained unchanged when the 2.5-kb fragment was lengthened by 1,432 bp (pJ2/3.9). When the 2.5-kb fragment was lengthened by 2,133 (pJ2L) or 4,960 bp (pJ12), luciferase activity gradually diminished by 60 and 90%, respectively, compared with pJ2M. Decreasing the size of the 2.5-kb fragment by 572 (pJ2/1.9) and 1,096 bp (pJ2/1.4) brought about a 30 and 77% reduction in luciferase activity, respectively. A further truncation of the insert from 1,458 to 733 bp left luciferase activity at 23% that of pJ2M. No significant differences in luciferase activity between the various fragments tested were demonstrated in control NIH3T3 cells (data not shown).
To further explore the 2.5-kb 5'-flanking region of hPCLN-1, which displayed the highest promoter activity, an orientation study was performed in OK cells. As shown in Fig. 5, when the 2.5-kb promoter fragment in pGL3 vector was reversed (pJ235M), luciferase activity was 80% lower than activity induced by the same sequence in sense orientation (pJ2M).
Taken together, these results suggested that a positive regulatory element, most likely located between positions –1982 and –1458, and a negative regulatory element likely positioned between positions –3317 and –2554, are involved in hPCLN-1 promoter activity in renal cells.
Binding of nuclear proteins to the hPCLN-1 promoter. To investigate the location and nature of the presumed positive regulatory element between positions –1982 and –1458, binding of nuclear proteins to this region was examined. The 525-bp DNA fragment was divided into 15 sequential double-stranded oligonucleotides, which were 5' end-labeled with 32P and incubated with nuclear proteins. DNA-protein complexes were resolved from unbound DNA by nondenaturing gel electrophoresis. Binding patterns were compared between nuclear extracts from OK cells and those from control NIH3T3 cells. As shown in Fig. 6, incubation of seven sequential DNA fragments located in the proximal part of the 525-bp sequence (between positions –1770 and –1491) with nuclear extracts from OK cells produced several retarded bands ( in Fig. 6) that were absent when the DNA was incubated with nuclear extracts from NIH3T3 cells. Binding was specific, since addition of a 50-fold molar excess of unlabeled oligonucleotide abolished DNA-protein complexes. DNA binding was absent in lanes without the nuclear extract. When oligonucleotides originating from the distal part of the 500-bp region (from position –1982 to –1771) were incubated with nuclear proteins from OK and NIH3T3 cells, the protein-DNA binding pattern was similar in the two cell lines (data not shown).
Taken together, these results suggested that OK cells, the renal cell line OK, but not nonrenal NIH3T3 cells, contain nuclear proteins that bind specifically to the hPCLN-1 promoter in the 280-bp region between positions –1770 and –1491.
Deletion analysis of the DNA region containing the proposed positive regulatory element of hPCLN-1. To investigate the functional importance of the 280-bp hPCLN-1 promoter region implicated in transcriptional activation and nuclear protein binding, a deletion of the region between positions –1772 and –1186 within the pJ2M reporter vector was created. When the resulting plasmid, designated pJ2Mdel, was transfected into OK cells, luciferase activity was reduced by 60% relative to pJ2M (Fig. 7A).
To further define the positive regulatory element within the 280-bp promoter sequence of functional importance, deletion analysis of this segment was carried out. The hPCLN-1 promoter sequence from positon –1738 to –1493 was 5'-truncated in increments of 35 bp. These progressively shorter DNA fragments were cloned upstream to a luciferase reporter gene in SV40 minimal promoter-containing vectors, pGL3-promoter, and transfected into OK cells. Figure 7B illustrates that luciferase activity was highest with the construct containing the hPCLN-1 promoter sequence from position –1703 to –1493. Truncation of the promoter to position –1668 caused little change in reporter activity. However, when the promoter was truncated to position –1633, reporter activity diminished by 50%. Two more sequential 35-bp deletions caused only minor reductions in luciferase activity.
Taken together, these experiments suggest that the 70-bp DNA sequence between positions –1703 and –1633, which binds nuclear proteins from renal cells, contains a positive regulatory element involved in transcriptional activation. Computer analysis of this segment, revealed a single PPRE at position –1655, which is known to bind the transcription factor PPAR (11, 15, 20).
Most PPREs described so far consist of a direct repeat of two hexamer half-sites separated by several nucleotides (14). The first half-site is highly conserved (AGGTCA) whereas the second is not. The sequence of the PPRE identified in the hPCLN-1 promoter contains one half-site, which is a perfect consensus sequence, but no clearly recognizable second half-site. This sequence, however, has been shown to enable transcription factor binding and activation (3).
Several in vivo (37, 44) and in vitro (7, 30) studies have demonstrated that Mg2+ restriction increases Mg2+ transport in the renal tubule whereas Mg2+ load decreases it (22, 31). To explore the molecular mechanisms of this modulation, we examined the effect of changes in ambient Mg2+ concentration on the activity of the hPCLN-1 promoter. For this purpose, OK cells were transiently transfected with the 2.5-kb hPCLN-1 promoter fragment (pJ2M) driving maximal luciferase activity. Following transfection, the cells were exposed for 24 h to media containing 0 (low), 0.7 (normal), or 1.5 mM (high) Mg2+. As demonstrated in Fig. 8, Mg2+ restriction caused a 52% increase, and Mg2+ load a 72% decrease, in luciferase activity in OK cells compared with conditions of normal Mg2+ concentration. These findings suggest that ambient Mg2+ concentration may modulate Mg2+ transport at the transcriptional level.
Effect of 1,25(OH)2 vitamin D on the PPRE-containing DNA region in the hPCLN-1 promoter. Since the vitamin D receptor (VDR) and PPAR both belong to the nuclear receptor superfamily of transcription factors and thus share several biological characteristics (6, 15, 26), we examined the effect of 1,25(OH)2 vitamin D on the hPCLN-1 promoter sequence containing the PPRE half-site. We first examined the action of 1,25(OH)2 vitamin D on the 2.5-kb hPCLN-1 promoter fragment (pJ2M). For this purpose, OK cells, transiently transfected with this fragment were exposed to 5 x 10–7 M 1,25(OH)2 vitamin D for 16–20 h, which caused promoter activity to decrease by 56% (Fig. 9A). Experimental conditions were selected based on preliminary experiments with varying 1,25(OH)2 vitamin D concentrations (10–6 to 10–7M) (data not shown).
We next examined whether the effect of 1,25(OH)2 vitamin D on the promoter was PPRE dependent. For this purpose, OK cells transfected with pGL3-promoter vector containing a 210-bp hPCLN-1 promoter fragment (pJ210) corresponding to the hPCLN-1 promoter sequence between positions –1703 and –1493 harboring the PPRE site, or a 70-bp 5' deletion product (pJ140) of this same promoter sequence, between positions –1633 and –1493, lacking the PPRE, were exposed to 1,25(OH)2 vitamin D under similar conditions. 1,25(OH)2 vitamin D treatment reduced the activity of the PPRE-containing promoter fragment by 52% compared with control non-1,25(OH)2 vitamin D-treated cells (Fig. 9B). This effect of 1,25(OH)2 vitamin D was markedly diminished in cells transfected with the PPRE-lacking fragment. Moreover, the actual difference in promoter activity between the PPRE-containing and PPRE-lacking fragments decreased fivefold following 1,25(OH)2 vitamin D treatment (Fig. 9B).
To further establish the role of PPRE in hPCLN-1 promoter activity, three bases of the PPRE half-site were point-mutated in both pJ2M and pJ210. OK cells transfected with these constructs were exposed to 1,25(OH)2 vitamin D as above. As shown in Fig. 10A, the decrease in pJ2M activity following 1,25(OH)2 vitamin D treatment was greatly diminished in cells transfected with the mutated promoter. When a similar experiment was carried out with mutated pJ210 (Fig. 10B), the 1,25(OH)2 vitamin D-induced reduction in luciferase activity seen in cells transfected with wild-type, PPRE-containing pJ210 was completely abolished once the PPRE site was mutated.
These experiments suggest that 1,25(OH)2 vitamin D modulates hPCLN-1 activity by a mechanism that appears to involve the PPRE half-site in the gene promoter.
DISCUSSION
In this study, we describe the transcriptional analysis of the promoter of the human paracellin-1 gene. We demonstrate that a 70-bp region between positions –1633 and –1703 may play a key role in the activity of the hPCLN-1 promoter (Figs. 6 and 7). Furthermore, we provide evidence that an interplay between a positive regulatory element located within this 70-bp region, and a more distally located negative regulatory element on the 5'-flanking region of the paracellin-1 gene, may determine renal cell-specific expression of this gene (Fig. 4). In addition, we demonstrate that ambient Mg2+ concentration affects hPCLN-1 promoter activity (Fig. 8). Finally, we show that 1,25(OH)2 vitamin D modulates hPCLN-1 promoter activity via this 70-bp, PPRE-containing, hPCLN-1 promoter fragment (Figs. 9 and 10), thereby suggesting that paracellin-1 gene expression is subject to PPAR/PPRE-mediated transcriptional regulation by this hormone.
Reabsorption of solutes across the tubular epithelial layer depends on two separate routes: a transcellular pathway and a paracellular pathway (1, 47). The paracellular pathway is regulated by tight junctions, which form a barrier to the diffusion of solutes across epithelial cells and function as a boundary between the apical and basolateral membranes maintaining epithelial cell polarity (21). Paracellin-1, which appears to regulate the paracellular transport of Mg2+ in the TAL, is the first tight junction protein reported to be involved in ion resorption (38). Paracellin-1, or claudin 16, (38) belongs to the claudin family of proteins that participate in the formation of tight junction strands in various tissues (27) and are thought to have a major role in regulating the magnitude and nature of paracellular permeability (1). The disease mutations found in paracellin-1 in familial hypomagnesemia-hypercalciuria syndrome result in an increased resistance to the electrical seal of tight junctions, thereby decreasing epithelial ionic permeability and selectively impeding magnesium and calcium reabsorption (38, 47). Recently, it has been reported that deletion of the paracellin-1 gene is responsible for renal tubular dysplasia in cattle (29). This finding suggests that paracellin-1, in addition to its function in ion resorption, may play an important role in the normal development and organization of the renal tubule.
hPCLN-1 mRNA was found to be expressed exclusively in the kidney (38, 42). Although cis-acting regulatory elements involved in regulating gene transcription may be dispersed throughout the gene locus, often the proximal promoter region contains elements sufficient for high levels of tissue-specific gene transcription. These elements may function as binding sites for tissue-restricted proteins that arbitrate transcriptional activation in expressing cells. In an effort to identify cis-acting elements involved in regulating kidney-specific expression of hPCLN-1, we focused our study on the 5'-flanking region of this gene. We isolated a 7.5-kb genomic sequence corresponding to the 5'-flanking region of hPCLN-1 and, using reporter gene assays, showed that the proximal 2.5-kb region contained cis-acting, positive and negative, regulatory elements that play a role in renal epithelial-specific expression of hPCLN-1. This promoter displayed high activity in PCLN-1-expressing renal cell lines but not in the fibroblast cell line NIH3T3. In the rabbit (38) and rat (42) kidney, PCLN-1 expression has been shown to be restricted to the TAL and the DCT. Several renal cell lines were used in this study. These included HEK293 and MDCT cells, as well as OK cells of proximal tubular origin. These renal cells were shown to express paracellin-1 mRNA (Fig. 2A) and protein (Fig. 2B) and to display hPCLN-1 promoter-driven luciferase activity (Fig. 3) as opposed to mouse embryonic fibroblast cells (NIH3T3), which demonstrated no expression/activity, indicating that the PCLN-1 promoter was kidney specific.
In our study, OK cells displayed the highest level of hPCLN-1 promoter-driven reporter gene activity (Fig. 3). Hence, these cells served as the recipient cells in most of our transfection experiments. It is not entirely clear why OK cells of proximal tubular origin express paracellin-1, which was found only in the TAL and the DCT of the rabbit and the rat. Several considerations could provide an explanation for this finding. First, the nephron segment-specific expression of the paracellin-1 gene has been examined in the kidney of these two rodents only. It is possible that the expression of the paracellin-1 gene in other animals and species, including humans, extends to the tight junction of more proximal nephron segments. Second, although several properties of OK cells are consistent with a proximal tubular site of origin, this cell line was originally derived from the whole kidney and may possess characteristics of more distal regions of the nephron (10, 19). Finally, in the immature rat, Mg2+ reabsorption occurs predominantly via the paracellular pathway in the proximal tubule rather than in the loop of Henle (23, 30). It is possible that the OK cell line used in our study contains cells with characteristics of the immature proximal tubule, including expression and cell-specific activity of genes, such as PCLN-1, participating in Mg2+ reabsorption in the immature proximal tubule tight junction. Taken together, we believe that the expression of paracellin-1 mRNA and protein in OK cells, coupled with the marked increase in reporter gene activity driven by the hPCLN-1 promoter in this cell line, has made OK cells suitable as experimental cells in our study.
Deletion analysis of reporter gene constructs containing hPCLN-1 promoter DNA suggested that hPCLN-1 transcription may be regulated by a proximal positive regulatory element positioned between –1982 and –1458 and a more distally located negative regulatory element between positions –3317 and –2554 (Fig. 4). EMSA studies of the presumed positive regulatory element-bearing region indicated that OK cells contain nuclear proteins that bind specifically to this functionally important region of the hPCLN-1 promoter (Fig. 6). NIH3T3 nuclear proteins did not bind to this DNA region. These proteins are potentially involved in tissue-specific expression of hPCLN-1. When this protein-binding DNA region was deleted from the 2.5-kb PCLN-1 promoter, a 60% reduction in promoter activity occurred. Detailed analysis revealed a 70-bp sequence within this DNA region, which seems to be responsible, at least in part, for modulating kidney-specific hPCLN-1 transcription (Fig. 7). Computer analysis of this segment disclosed a single PPAR binding site (PPRE).
Tubular Mg2+ reclamation is modulated by a variety of hormonal and nonhormonal factors (8, 16, 30). Mg2+ restriction and Mg2+ load influence both transcellular Mg2+ transport in the DCT and paracellular Mg2+ flux in the TAL (7, 22, 31, 37, 44). Micropuncture experiments in the rat nephron have demonstrated that a low-Mg2+ diet leads to urinary retention of Mg2+ due to increased Mg2+ reabsorption in the loop of Henle (37, 44). Culturing MDCT cells in Mg2+-free medium increased their Mg2+ transport rate (7, 30). Pretreatment of MDCT cells with actinomycin D, an inhibitor of transcription, resulted in a significant decrease in this adaptive response, suggesting that the adaptive regulation of Mg2+ depletion may involve gene transcription (30). As opposed to the effect of Mg2+ deprivation, Mg2+ infusion (22) and acute elevation of Mg2+ concentration at the contraluminal membrane of the TAL (31) in rats inhibited Mg2+ resorption in this tubule segment. The molecular mechanisms underlying the adaptive response of tubular Mg2+ transport to changes in Mg2+ levels have not been investigated. In this study, we show that ambient Mg2+ concentration affects hPCLN-1 promoter activity in OK cells. Specifically, Mg2+ restriction increases hPCLN-1 promoter activity, whereas Mg2+ load reduces the activity of this promoter. These findings are in concert with the modulatory effects of Mg2+ restriction and load on Mg2+ transport observed at the tubular and cellular levels (7, 22, 31, 37, 44) and demonstrate, for the first time, that changes in Mg2+ availability may influence Mg2+ transport at the transcriptional level. The exact molecular mechanisms whereby this Mg2+ level-induced effect is achieved remain unknown. The possible involvement of the cell membrane-bound Ca2+/Mg2+-sensing receptor, or the potential role of a hypothetical Mg2+ response element residing on the hPCLN-1 promoter in the Mg2+-induced effect on transcription of this gene, remains to be explored.
Paracellular Mg2+ transport in the TAL is known to be regulated by several hormones. These include parathyroid hormone, arginine vasopressin, aldosterone, insulin, and 1,25(OH)2 vitamin D, among others (30). Using microperfusion studies of isolated mouse TAL segments, it has been shown that most of these hormonal responses are mediated by changing the transepithelial voltage or by altering the permeability of the paracellular pathway (24, 30, 45). However, the exact molecular mechanisms of this hormone-induced effect on Mg2+ transport have not been investigated. 1,25(OH)2 vitamin D has been shown to increase Mg2+ uptake by MDCT cells (32). The effect of this hormone on paracellular Mg2+ transport has not been explored. Several hormones, including 1,25(OH)2 vitamin D (6), exert many of the biological actions by receptor-mediated effects on gene transcription. In this study, we show that 1,25(OH)2 vitamin D decreases hPCLN-1 promoter-driven luciferase activity (Figs. 9 and 10). OK cells are known to harbor vitamin D receptors (18) and, as discussed above, appear to express the paracellin-1 gene as well as possess the machinery necessary for its activity. Hence, this cell line may serve as an excellent model with which to explore the effect of this hormone on the paracellin-1 gene. Taken together, our findings provide the first direct evidence that paracellular, paracellin-1-mediated Mg2+ transport may be regulated at the transcriptional level by 1,25(OH)2 vitamin D. However, the physiological significance of the transcription-inhibiting effect of 1,25(OH)2 vitamin D shown in our study and its role in overall magnesium transport in the renal tubule remain to be clarified.
The effect of 1,25(OH)2 vitamin D on the hPCLN-1 promoter was demonstrated in our study to involve the PPRE half-site located between positions –1633 and –1703 within this promoter (Figs. 9 and 10). The PPARs comprise a group of transcription factors that belong to the nuclear receptor superfamily to which the VDR, the thyroid hormone receptor, and the all-trans-retinoic acid receptor also belong (11, 15, 20). PPARs are ligand-regulated transcription factors that control gene expression by binding to specific response elements (PPREs) within promoters, having formed heterodimers with the retinoid X receptor (11, 15, 20). Three PPAR isoforms, PPAR, PPAR, and PPAR, have been identified (11, 15) and are expressed in several tissues including the kidney (11, 48). PPARs participate in a variety of biological processes common to various cell types, including lipid metabolism, glucose homeostasis, inflammation, cell proliferation and differentiation, apoptosis, and early development (4, 11). However, despite their abundant expression in various segments of the renal tubule (11, 48), very little is known about specific tubular transport processes that are controlled by the PPARs. Noteworthy is a recent study demonstrating enhanced renal tubular cell surface expression of the epithelial sodium channel in response to PPAR activation (12).
In our study, we focused on 1,25(OH)2 vitamin D, known to modulate Mg2+ transport on one hand (30, 32) and to interact with the PPAR system on the other. The DNA binding site of the VDR is 46% homologous to the DNA binding site of PPAR, and both receptors are known to heterodimerize with the retinoid X receptor before binding to their specific DNA motifs (6, 15, 26). However, recent studies suggest that there is considerable flexibility in the binding sites recognized by the VDR, including the existence of some single half-sites (17). Furthermore, the outcome of the interaction between the VDR and its binding site may be an increase (5) or a decrease (25) in gene transcription. Of note is a recent study demonstrating that the VDR represses transcriptional activity of PPAR in COS1 cells (34). In our study, we provide evidence for PPRE-dependent transcriptional regulation of hPCLN-1 by 1,25(OH)2 vitamin D, and we show here, for the first time, that the PPAR/PPRE axis may play a specific role in the hormonal regulation of Mg2+ transport in the renal tubule.
In conclusion, an interplay between a positive regulatory element and a more distally located negative regulatory element on the 5'-flanking region of the paracellin-1 gene may determine renal epithelial-specific activity of this gene. The 70-bp, PPRE-containing, DNA region between positions –1633 and –1703 may play a key role in the activity of the PCLN-1 promoter. Ambient Mg2+ concentration and 1,25(OH)2 vitamin D may modulate paracellular, paracellin-1-driven, Mg2+ transport at the transcriptional level. 1,25(OH)2 vitamin D exerts its activity on the hPCLN-1 promoter likely via the PPRE site.
Future studies utilizing transgenic animals may verify whether the activity of this 70-bp DNA fragment, examined in the cell culture model, demonstrates tubular epithelial specifity in the intact organism, where the transgene is exposed to a more physiologically relevant environment (i.e., hormones, proteins, etc.). Once established, it is possible that these transgenic animals will serve as an excellent model with which to study the transcriptional effects of a variety of physiological as well as pathophysiological factors on the paracellin-1 gene, and thus on tubular Mg2+ reabsorption.
GRANTS
This work was supported by the Ruth and Allen Ziegler Fund for Pediatric Research and the Kronovet Fund for Medical Research, Technion-Israel Institute of Technology.
ACKNOWLEDGMENTS
We thank Dr. Karl Skorecki, Dr. Maty Tzukerman, and Dr. Sara Selig for helpful suggestions and Ayal Kelmachter and Adva Hermoni for technical assistance. We thank Judi Fichman and Hagar Shafrir for expert secretarial assistance.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
E. Efrati and Julia Arsentiev-Rozenfeld contributred equally to this work.
REFERENCES
Anderson JM. Molecular structure of tight junctions and their role in epithelial transport. News Physiol Sci 16: 126–130, 2001.
Bai Y, Pontoglio M, Hiesberger T, Sinclair AM, and Igarashi P. Regulation of kidney-specific Ksp-cadherin gene promoter by hepatocyte nuclear factor-1. Am J Physiol Renal Physiol 283: F839–F851, 2002.
Barrero MJ, Camarrero N, Marrero PF, and Haro D. Control of human carnitine palmitoyltransferase II gene transcription by peroxisome proliferator-activated receptor through a partially conserved peroxisome proliferator-responsive element. Biochem J 369: 721–729, 2003.
Berger J and Moller DE. The mechanisms of action of PPARs. Annu Rev Med 53: 409–435, 2002.
Breen EC, Van Wijen AJ, Lian JB, Stein GS, and Stein JL. In vivo occupancy of vitamin D responsive element in the osteocalcin gene supports vitamin D-dependent transcriptional upregulation in intact cells. Proc Natl Acad Sci USA 91: 12902–12906, 1994.
Brown AJ, Dusso A, and Slatopolsky E. Vitamin D. Am J Physiol Renal Physiol 277: F157–F175, 1999.
Dai LJ, Raymond L, Friedman PA, and Quamme GA. Mechanisms of amiloride stimulation of Mg2+ uptake in immortalized mouse distal convoluted tubule cells. Am J Physiol Renal Physiol 272: F249–F256, 1997.
Dai LJ, Ritchie G, Kerstan D, Kang HS, Cole DE, and Quamme GA. Magnesium transport in the renal distal convoluted tubule. Physiol Rev 81: 51–84, 2001.
Dignam JD, Lebovitz RM, and Roeder RG. Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res 11: 1475–1489, 1983.
Gomes P, Xu J, Serrao P, Doria S, Jose PA, and Soares-da-Silva P. Expression and function of sodium transporters in two opossum kidney cell clonal sublines. Am J Physiol Renal Physiol 283: F73–F85, 2002.
Guan Y and Breyer MD. Peroxisome proliferator-activated receptors (PPARs): novel therapeutic targets in renal disease. Kidney Int 60: 14–30, 2001.
Hong g Lockhart A, Davis B, Rahmoune H, Baker S, Ye L, Thompson P, Shou Y, O’Shaughnessy K, Ronco P, and Brown J. PPAR activation enhances cell surface ENac via up-regulation of SGK1 in human collecting duct cells. FASEB J 17: 1966–1968, 2003.
Igarashi P, Cooduvalli S, Shashikant R, Thomson RB, Whyte DA, Liu-Chen S, Ruddle FH, and Aronson PS. Ksp-cadherin gene promoter. II. Kidney-specific activity in transgenic mice. Am J Physiol Renal Physiol 277: F599–F610, 1999.
Ijpenberg A, Jeannin E, Wahli W, and Desvergne B. Polarity and specific sequence requirements of peroxisome proliferator-activated receptor (PPAR)/retinoid X receptor heterodimer binding to DNA. A functional analysis of the malic enzyme gene PPAR response element. J Biol Chem 272: 20108–20117, 1997.
Issemann I and Green S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature 347: 645–650, 1990.
Kelepouris E and Agus ZS. Hypomagnesemia: renal magnesium handling. Semin Nephrol 18: 58–73, 1998.
Koszewski NJ, Kiessling S, and Malluche H. Isolation of genomic DNA sequences that bind vitamin D receptor complexes. Biochem Biophys Res Commun 283: 188–194, 2001.
Koszewski NJ and Rowan A. Heterodimer requirement for gene regulation by vitamin D in variant OK cells. Steroids 68: 307–314, 2003.
Koyama H, Goodpasture C, Miller MM, Teplitz RL, and Riggs AD. Establishment and characterization of a cell line from the American opossum (Didelphys virginiana). In Vitro 14: 239–246, 1978.
Laudet V, Hanni C, Coll J, Catzeflis F, and Stehelin D. Evolution of the nuclear receptor gene superfamily. EMBO J 11: 1003–1013, 1992.
Le Grimellec C, Friedlander G, el Yandouzi EH, Zlatkine P, and Giocondi MC. Membrane fluidity and transport properties in epithelia. Kidney Int 42: 825–836, 1992.
Le Grimellec C, Roinel N, and Morel F. Simultaneous Mg, Ca, P, K, Na and Cl analysis in rat tubular fluid. II. During acute Mg plasma loading. Pflügers Arch 340: 197–210, 1973.
Lelievre-Pegorier M, Merlet-Benichou C, Roinel N, and de Rouffignac C. Developmental pattern of water and electrolyte transport in rat superficial nephrons. Am J Physiol Renal Fluid Electrolyte Physiol 245: F15–F21, 1983.
Mandon B, Siga E, Chabardes D, Firsov D, Roinel N, and De Rouffignac C. Insulin stimulates Na+, Cl–, Ca2+, and Mg2+ transports in TAL of mouse nephron: cross-potentiation with AVP. Am J Physiol Renal Fluid Electrolyte Physiol 265: F361–F369, 1993.
Meszaros JG, Karin NJ, Akanbi K, and Farach-Carson MC. Down-regulation of L-type Ca2+ channel transcript levels by 1,25-dihydroxyvitamin-D3. J Biol Chem 271: 32981–32985, 1996.
Misra P, Owuor ED, Li W, Yu S, Qi C, Meyer K, Zhu YJ, Rao MS, Kong AN, and Reddy JK. Phosphorylation of transcriptional coactivator peroxisome proliferator-activated receptor (PPAR)-binding protein (PBP). Stimulation of transcriptional regulation by mitogen-activated protein kinase. J Biol Chem 277: 48745–48754, 2002.
Morita K, Furuse M, Fujimoto K, and Tsukita SH. Claudin multigene family encoding four-transmembrane domain protein components of tight junction strands. Proc Natl Acad Sci USA 96: 511–516, 1999.
Nelson RD, Stricklett P, Gustafson C, Stevens A, Ausiello D, Brown D, and Kohan DE. Expression of an AQP2 Cre recombinase transgene in kidney and male reproductive system of transgenic mice. Am J Physiol Cell Physiol 275: C216–C226, 1998.
Ohba Y, Kitagawa H, Kitoh K, Sasaki Y, Takami M, Shinkai Y, and Kunieda T. A deletion of the paracellin-1 gene is responsible for renal tubular dysplasia in cattle. Genomics 68: 229–236, 2000.
Quamme GA. Renal magnesium handling: new insights in understanding old problems. Kideny Int 52: 1180–1195, 1997.
Quamme GA and Dirks JH. Intraluminal and contraluminal magnesium on magnesium and calcium transfer in the rat nephron. Am J Physiol Renal Fluid Electrolyte Physiol 238: F187–F198, 1980.
Ritchie G, Kerstan D, Dai LJ, Kang HS, Canaff L, Hendy GN, and Quamme GA. 1,25(OH)2D3 stimulates Mg2+ uptake into MDCT cells: modulation by extracellular Ca2+ and Mg2+. Am J Physiol Renal Physiol 280: F868–F878, 2001.
Romani A and Scarpa A. Regulation of cell magnesium. Arch Biochem Biophys 298: 1–12, 1992.
Sakuma T, Miyamoto T, Jiang W, Kakizawa T, Nishio S, Suzuki S, Takeda T, Oiwa A, and Hashizume K. Inhibition of peroxisome proliferator-activated receptor signaling by vitamin D receptor. Biochem Biophys Res Commun 312: 513–519, 2003.
Sambrook J, Fritsh EF, and Maniatis T. Molecular Cloning: A Laboratory Manual (2nd ed.). Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989.
Shachaf C, Skorecki KL, and Tzukerman M. Role of AP2 consensus sites in regulation of rat Npt2 (sodium-phosphate cotransporter) promoter. Am J Physiol Renal Physiol 278: F406–F416, 2000.
Shafik IM and Quamme GA. Early adaptation of renal magnesium reabsorption in response to magnesium restriction. Am J Physiol Renal Fluid Electrolyte Physiol 257: F974–F977, 1989.
Simon DB, Lu Y, Chaote KA, Velazquez H, Al-Sabban E, Praga M, Casari G, Bettinelli A, Colussi G, Rodriguez-Soriano J, McCredie D, Milford D, Sanjad S, and Lifton RP. Paracellin-1, a renal tight junction protein required for paracellular Mg2+ resorption. Science 285: 103–106, 1999.
Taniyama Y, Sato K, Sugawara A, Urono A, Ikeda Y, Kudo M, Ito S, and Takeuchi K. Renal tubule-specific transcription and chromosomal localization of rat thiazide-sensitive Na-Cl cotransporter gene. J Biol Chem 276: 26260–26268, 2001.
Uchida S, Rai T, Yatsushige H, Matsumura Y, Kawasaki M, Sasaki S, and Marumo F. Isolation and characterization of kidney cell-specific ClC-K1 chloride channel gene promoter. Am J Physiol Renal Physiol 274: F602–F610, 1998.
Uchida S, Tanaka Y, Ito H, Saitoh-Ohara F, Inazawa J, Yokoyama KK, Sasaki S, Marumo F. Transcriptional regulation of the CLC-K1 promoter by myc-associated zinc finger protein and kidney-enriched Kruppel-like factor, a novel zinc finger repressor. Mol Cell Biol 20: 7319–7331, 2000.
Weber S, Schlingmann KP, Peters M, Nejsum LN, Nielsen S, Engel H, Grzeschik KH, Seyberth HW, Grone HJ, Nusing R, and Konrad M. Primary gene structure and expression studies of rodent paracellin-1. J Am Soc Nephrol 12: 2664–2667, 2001.
Whyte DA, Li C, Thomson RB, Nix SL, Zanjani R, Karp SL, Aronson PS, and Igarashi P. Ksp-cadherin gene promoter. I. Characterization and renal epithelial cell-specific activity. Am J Physiol Renal Physiol 277: F587–F598, 1999.
Wittner M, Jounier S, Deschenes G, de Rouffignac C, and Di Stefano A. Cellular adaptation of the mouse cortical thick ascending limb of Henle’s loop (CTAL) to dietary magnesium restriction: enhanced transepithelial Mg2+ and Ca2+ transport. Pflügers Arch 439: 765–771, 2000.
Wittner M, Mandon B, Roinel N, de Rouffignac C, and Di Stefano A. Hormonal stimulation of Ca2+ and Mg2+ transport in the cortical thick ascending limb of Henle’s loop of the mouse: evidence for a change in the paracellular pathway permeability. Pflügers Arch 423: 387–396, 1993.
Wong MA, Cui S, and Quaggin SE. Identification and characterization of a glomerular-specific promoter from the human nephrin gene. Am J Physiol Renal Physiol 279: F1027–F1032, 2000.
Wong V and Goodenough DA. Paracellular channels (Abstract). Science 285: 52, 1999.
Yang T, Michele DE, Park J, Smart AM, Lin Z, Brosius 3rd FC, Schnermann JB, and Briggs JP. Expression of peroxisomal proliferator-activated receptors and retinoid X receptors in the kidney. Am J Physiol Renal Physiol 277: F966–F973, 1999.(Edna Efrati , Julia Arsen)