INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

PHOSPHATE (P_i) reabsorption in the mammalian proximal tubule is altered under a variety of physiological and pathophysiological conditions (for review, see Refs. 3, 6, 22). Brush-border membrane Na-P_i cotransport was found to be the rate-limiting and pathophysiologically altered transport step. Two different brush-border membrane Na-P_i cotransporters have been identified by expression cloning techniques. Both, the type I and type II transporters were found to be exclusively located in the proximal tubule (for review, see Refs. 13, 19-21).

The type II Na-P_i cotransporter was identified from different animal species (9, 11, 16, 27, 29, 30, 32), including the opossum kidney-derived OK cell line (NaPi-4, Ref. 27), the only renal cell model expressing this transporter in significant amounts and frequently used to dissect the cellular and regulatory mechanisms involved in regulation of proximal tubular brush-border membrane Na-P_i cotransport (for review, see Ref. 22). The type II Na-P_i cotransporter is altered under a variety of physiological and pathophysiological conditions (for reviews, see Refs. 13, 19-21). Such regulatory phenomena were also seen in OK cells, for example, for parathyroid hormone (PTH; 5, 17, 23a), for alterations in the medium phosphate content (4, 18, 25, 26), for glucocorticoids (31), for thyroid hormone [triiodothyronine (T₃); see Ref. 28], and for epidermal growth hormone (1).

The gene structure for the human and murine type II Na-P_i cotransporter gene (NPT2 and m-Npt2) has recently been determined (8). In the present study, we report the isolation and characterization of the 5'-flanking region of the OK cell type II Na-P_i cotransporter (ok-Npt2). We demonstrate that the ok-Npt2 promoter is active in the homologous OK cell system and hardly regulated by PTH or changes in the phosphate concentration in the medium.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. Restriction enzymes were obtained from Biofinex (Switzerland); DNA modifying enzymes, pGL3 luciferase reporter gene plasmids, and luciferase reporter gene assay were from Promega (Madison, WI); pBluescript SK⁺ plasmid was acquired from Stratagene (La Jolla, CA); pCMV-LacZ was kindly provided by Dr. S. Rusconi, University of Zurich, Switzerland; pUC19 Nco I was obtained from Dr. H. Rhyner, Children's Hospital of Zurich, Switzerland; [-³²P]dCTP, [-³⁵S]dATP, and [-³²P]ATP were purchased from NEN Research Products (Du Pont Canada) and oligonucleotides were synthesized by Microsynth.

Gene isolation and mapping. Genomic DNA from cultured OK cells (3B/2 clone) was extracted and dialyzed according to current protocols (2). A genomic Southern blot was hybridized with a ³²P-labeled 5' end probe (oligolabeling kit, Pharmacia) isolated from OK cell NaPi-4 cDNA (27) in pSPORT1 (560-bp Mlu I-Sac I, see Fig. 1) overnight at 42°C in 50% formamide solution supplemented with 5% dextran sulfate (2). The filter was washed twice with 2× SSC (1× SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7.0)-0.1% sodium dodecyl sulfate (SDS) at room temperature and twice with 0.5× SSC-0.1% SDS at 55°C and exposed on Kodak films with intensifying screen. A 12- to 15-kb BamH I genomic fragment could be detected on the genomic Southern blot. For construction of a library, OK cell genomic DNA was digested with BamH I and size fractionated on a 10-40% sucrose gradient. A fraction of 12-20 kb was ligated into DASH II-BamH I (Stratagene), and 10⁶ plaque-forming units were screened with the probe mentioned above. A genomic clone of ~14 kb (OKGL41.1, see Fig. 1) was isolated. A 6,000-bp EcoR I-BamH I fragment was subcloned into pBluescript SK⁺ (Stratagene). The DNA was purified in small quantities by spin columns (Qiagen).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Genomic organization of OK cell Npt2 (ok-Npt2). A genomic BamH I clone of 14 kb (OKGL41.1) was isolated from a OK cell DASH II library. A 6,000-bp EcoR I-BamH I subfragment was cloned into pBluescript SK+ and sequenced. It contained exons 1 to 3 (solid boxes), introns 1 to 3 (open boxes), and 4.7 kb of 5'-flanking region. A consensus binding site for TFIID (TATA box) at 30 and a "GCAAT" element at 124 were identified. The 5'-flanking region as well as intron 3 of the OK cell Npt2 gene exhibit long GGAA repeats of unknown function (microsatellites; gray shaded boxes). Top: the 560-bp Mlu I-Sac I cDNA probe used for screening the DASH II library; the polylinker is depicted as solid black, the 5' untranslated region is shaded in gray, and the coding region is open. Bottom: different constructs used for the analysis of promoter function; fragments of different size ranging from 4700 to 67 and including 54 bp of exon 1 have been cloned in front of the luciferase reporter gene of pGL3 basic (Promega) to estimate the corresponding transcriptional activity.

The DNA was sequenced on both strands by the dideoxynucleotide chain termination method (Pharmacia Kit) and analyzed using the University of Wisconsin Genetics Computer Group Software, version 8.1.

Cell culture. The opossum kidney proximal tubule cell line (3B/2) and the human HeLa-S3 cell line were grown in Dulbecco's modified Eagle's medium-Ham's F12 (DMEM-F12) supplemented with 10% fetal calf serum (FCS) and 2 mM L-glutamine. The human macrophage cell line U937, the Madin-Darby canine kidney (MDCK) cell line, and the mouse kidney cortex cell line MCT (10) were cultivated in DMEM supplemented with 10% FCS. All cell cultures were kept at 37°C in a humidified environment (5% CO₂, 95% air).

ok-Npt2 promoter-luciferase constructs: electroporation and enzyme assays. A 327-bp genomic fragment was generated by polymerase chain reaction (PCR) on the 6,000-bp EcoR I-BamH I subfragment of clone OKGL41.1 with an antisense primer localized within exon 1 (NaPi4-31/50R, see Fig. 2) and a sense primer starting at position 327 on the ok-Npt2 gene. The fragment was digested with Apa I (located in the 5'-flanking region of ok-Npt2), subcloned into pBluescript SK⁺ cut with Apa I and Sma I (pB327), and sequenced. To generate the luciferase reporter gene construct pGL327, the insert was excised from pB327 by BamH I and Kpn I (both located within the polylinker of pBluescript SK⁺) and subcloned into pGL3 basic vector digested by Kpn I and Bgl II. Genomic DNA fragments of 140 to 208 bp were generated by PCR on pB327 (mentioned above) with a pBluescript SK⁺ primer (5' CCATGATTACGCCAAGCTC 3') and different ok-Npt2-specific sense primers. The construct pGL208 extends from position 154, pGL206 from 152, pGL198 from 144, pGL194 from 140, pGL173 from 119, pGL168 from 114, pGL140 from 86 and pGL121 from 67, and the whole set of PCR generated fragments end within exon 1 at position +54. The amplified DNA was digested with BamH I (located in the polylinker of pBluescript SK⁺), subcloned into the pGL3 basic vector digested with Bgl II and Sma I, and sequenced on both strands.

View larger version (84K): [in this window] [in a new window]   Fig. 2.   Sequence of the 5'-flanking region of the OK cell Npt2 gene (ok-Npt2). This 3,017-bp sequence is part of the 6,000-bp EcoR I-BamH I fragment (see Fig. 1), which has been sent to the GenBank/EMBL Data Bank (6 kb of sequence; accession no. AJ003021). The first nucleotide corresponding to the OK cell NaPi-4 cDNA is depicted as +16, and the main transcription start site is at +1. Exons 1-3 are underlined, and coding sequence is in uppercase letters. Translation start site (double underlined) is located within the second exon. A possible TFIID binding site and the conserved GCAAT element are boxed in bold. A potential AP1 site at 304 is underlined. The two microsatellite GGAA repeats of unknown function are in italic lowercase. The 3' end primer (NaPi4-31/50R) used for generation of the OK cell promoter constructs is marked in the gray shaded box.

The luciferase reporter gene constructs pGL1700 and pGL4700 were generated by linking a 1,400-bp Sac I-Apa I or a 4,400-bp EcoR I-Apa I subfragment of clone OKGL 41.1 to the 327-bp Apa I-BamH I fragment mentioned above.

For standardization of the promoter activity in different cell lines, the pGL3 promoter (pGL3-SV40, enhancerless simian virus promoter) and the pGL3 basic vector (Promega) were used. As internal standard, a -galactosidase-expressing vector pCMV-LacZ was cotransfected. Transient cell transfections were performed by electroporation. Approximately 10⁷ cells in DMEM without FCS were electroporated at 960 µF and 250 V with 5 µg (OK cells), 10 µg (MCT, MDCK and U937), or 20 µg (HeLa-S3) of pGL3 constructs and equal amounts of pCMV-LacZ and plated on 3.5-cm diameter dishes. The cells were harvested 48 h after transfection in 100 µl of 1× lysis buffer (Promega). The -galactosidase reaction was performed in 0.1 M NaPO₄, pH 7.5, 4 mg/ml o-nitrophenyl--galactopyranoside, and 1 mM MgCl₂ for 30 min to 4 h at 37°C, and the optical density was measured at 420 nm. Luciferase was assayed with a luciferase kit (Promega) and measured in a luminometer (Lumac Biocounter M1500).

For studies related to the effect of PTH and phosphate concentration, the electroporated cells were plated into 24-well plates in triplicate and kept for 4 h in DMEM-F12 supplemented with 10% FCS to improve vitality. Subsequently cultured cells were switched for 24 h to the corresponding medium (22 mM , 5% CO₂, and pH 7.4) without FCS. For studies with altering phosphate concentration, cells were kept in phosphate-free DMEM (Sigma) supplemented with 0.1, 0.9, and 1.5 mM phosphate. To determine the hormonal effect, the cells were either incubated in 10⁸ M PTH or in its vehicle. The cells were used for uptake studies and then harvested in the reporter gene lysis buffer for determination of luciferase and -galactosidase activity. Protein concentration and P_i uptake was performed as described (23a).

Rapid amplification of 5' cDNA ends (5'-RACE). For 5'-RACE OK cell total RNA (10 µg) isolated with Trizol (GIBCO-BRL) was retrotranscribed with 20 U Moloney murine leukemia virus reverse transcriptase (MLV-RT, Promega) using a NaPi-4 cDNA-specific (OK cell; Ref. 27) primer (nt 386-405; antisense). The product was cleaned by the GeneClean II kit (BIO 101). The 5' end of the cDNA was extended by a polynucleotide transferase (30 U, GIBCO-BRL) in presence of 0.4 mM dATP. PCR was performed with a second NaPi-4 cDNA-specific primer (nt 276-291; antisense), a T₁₇ adapter primer, and an adapter primer (7). The PCR product was digested with Sal I (located in the adapter primer) and Nco I (NaPi-4 cDNA) and subcloned into pUC19 Nco and sequenced. By this method we obtained four different extension products (see RESULTS).

Primer extension. A quantity of 10 µg of OK cell total RNA (see above) was retrotranscribed with 20 U MLV-RT (Promega) in presence of 80,000 cpm of a ³²P-labeled 5' NaPi-4 cDNA-specific (OK cell; Ref. 27) primer (nt 45-65 or nt 50-70, both antisense). Labeling of the oligonucleotide was performed with 8 U polynucleotide kinase (Promega) in presence of 50 µCi [-³²P]ATP. The extension product was ethanol precipitated and loaded on a 6% sequencing gel in presence of a sequencing ladder performed with the primers mentioned above on genomic DNA (pGL327).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Structure of the OK cell type II Na/P_i-cotransporter (Npt2) 5'-flanking region. The organization of the ok-Npt2 promoter region was determined by screening a genomic OK cell DASH BamH I library with a 560-bp Mlu I-Sac I fragment representing the very 5' end of the OK cell type II Na-P_i cotransporter cDNA (see Fig. 1). A single clone OKGL41.1 of ~14 kb was obtained, and a 6,000-bp EcoR I-BamH I subfragment was subcloned into plasmid pBluescript SK and fully sequenced (see Figs. 1 and 2). It contained in its 3' region exon 1 corresponding to nt 12-64, exon 2 corresponding to nt 65-177, and exon 3 corresponding to nt 178-333 of the OK cell NaPi-4 cDNA (27). A comparison of the intron/exon boundaries is presented in Table 1. In ok-Npt2, exon 2 contains only 1 base of 5'-untranslated region, in contrast to the m-Npt2 and h-NPT2 where the translation initiation site is also on exon 2 but further downstream (8). A high homology in the organization between mouse, human, and opossum NPT/Npt2 genes is suggested by the identity of encoded amino acids and coding phase usage at the intron/exon boundaries (Table 1; introns 2 and 3).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Sequence comparison of the 5'-flanking regions of the OK cell (ok-), human (h-), and the murine (m-) Npt2. Comparison of the first 170 bp of the OK cell, the human, and the murine promoter including the first exon reveal sequence homology (*) around a GCAAT element and a conserved TFIID binding site (both boxed and underlined). Transcription initiation sites of the ok- and m-NPT2 gene determined by both rapid amplification of 5' cDNA ends (5'-RACE) and primer extension are underlined. Exon 1 of the OK cell, human, and murine type II Na-Pi cotransporter gene is depicted in uppercase letters; +1, the major transcription initiation site in ok-Npt2; and +16, the start of the cDNA previously identified (27).

View larger version (67K): [in this window] [in a new window]   Fig. 4.   Determination of the transcription initiation site by primer extension (arrowheads) and by 5'-RACE (asterisks), for comparison. Sequence corresponding to the OK cell NaPi-4 cDNA (starting at +16) is given in bold, potential transcription initiation sites as obtained by primer extension are marked by arrows, and those obtained from sequencing 5'-RACE products are indicated by asterisks. For further details see MATERIALS AND METHODS.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   ok-Npt2 promoter activity in the OK cell line 3B/2. Fragments of different size ranging from 4700 to 67 and all including 54 bp of exon 1 (see Fig. 1) were subcloned into the luciferase reporter vector pGL3 to estimate the corresponding transcriptional activity. The fragments missing the GCAAT element, pGL121 to pGL173, were similar to the promoterless pGL3 construct in activity measurements. The larger fragments including the conserved GCAAT region pGL194 to pGL4700 had activities similar to the enhancerless pGL3-SV40 promoter construct. At least 3 independent experiments were done in triplicate.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Analysis of the OK cell Npt2 promoter in different cell lines. The promoterless luciferase constructs pGL3, the OK cell promoter construct pGL327 (see Fig. 1), and the enhancerless SV40 promoter construct pGL3-SV40 were electroporated in different cell lines (A-E) in presence of the pCMV-LacZ vector for standardization. Luciferase and galactosidase activities were measured 48 h posttransfection. High activity of the OK cell promoter could be observed in the homologous system of OK cells (A) and in human HeLa cells (B). In the human macrophage cell line U937 (C), in the murine kidney cell line MCT (D), and the canine renal distal tubular cell line MDCK (E), the OK cell promoter had only a low activity. For each cell line data from 3 independent experiments done in triplicate were used.

8

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Effect of different "stimuli" known to influence the activity of the type II Na-Pi cotransporter. OK cells (3B/2) were electroporated in presence of diverse pGL3 constructs. For each transfection, cells were equally plated in 24-well plates and after 4 h switched to fetal calf serum-free medium treated with 108 M parathyroid hormone (PTH) in DMEM-F12 at 5% CO2 (A) or with 0.1, 0.9 or 1.5 mM phosphate DMEM medium (Sigma) at 5% CO2 (B) for 24 h. In several experiments the cytomegalovirus (CMV) promoter activity of the vector used for standardization (pCMV-LacZ) was two- to threefold increased in presence of PTH. Therefore, we did not include the -galactosidase assay for standardization in A.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Opossum kidney cells (OK cells) are the only renal cell line expressing the type II Na-P_i cotransporter (NaPi-4) in significant amounts. Furthermore, in this cell line the transporter is regulated like in the proximal tubule (for review see Refs. 13, 19-22). Therefore, OK cells are ideally suited in homologous transfection studies to characterize structure and function relationships of the type II Na-P_i cotransport promoter.

We report the characterization of the 5'-flanking region of the ok-Npt2 gene including the promoter region and exons 1 to 3. The first two exons are separated by a relatively small intron of 443 bp. Within the 5'-flanking region and adjacent to exon 3, two extended GGAA repeats (microsatellites) of unknown function are present. The promoter/enhancer region of the ok-Npt2 gene contains a typical TATA box and a conserved GCAAT element, which seem to enhance promoter activity. These motifs are also found at corresponding sites within the published human and murine type II Na-P_i cotransporter genes (Fig. 3). In all three species the translation start site is within the second exon. For both the murine and the OK cell Npt2 genes, the conserved TATA box is 31 bp away from the major transcription initiation site (see Fig. 3). In the ok-Npt2 gene an AP1 site is present at position 304; two AP1 sites have also been identified in the human NPT2 gene but are located further upstream (29).

In contrast to OK cells, no promoter activity could be observed in the human macrophage cell line U397, in the Madin Darby canine kidney MDCK cell line of distal tubular origin, and in the murine kidney cortex cell line MCT, and we could prove by RT-PCR and Northern blotting that in the human, murine, and dog cell lines an intrinsic type II Na-P_i cotransporter is absent (data not shown). These results indicate that an element might be present within the 327-bp fragment derived from the ok-Npt2 gene that promotes transcriptional activity in OK cells. The negative RT-PCR result with the murine cell line MCT, a SV40 transformed cell line derived from microdissected mice proximal tubules (10), indicates a lack of cell differentiation. In Northern hybridizations, we could also not observe an expression of the type II Na-P_i cotransporter in "proximal tubular" LLC-PK cells (17; data not shown). It remains unknown why only OK cells have retained the capacity to express the type II Na-P_i cotransporter in physiological relevant amounts and with the basic and regulatory characteristics of the type II Na-P_i cotransporter in its proximal tubular environment.

Dietary phosphorus and PTH are important regulators of renal phosphate reabsorption. Dietary phosphate restriction for several days leads in rats to an increased brush-border membrane type II Na-P_i cotransporter expression on protein and mRNA level that cannot be prevented by actinomycin D (14, 15). Interestingly, the early effects (2 h) are independent of changes in mRNA content. A similar phenomenon can also be observed in OK cells. In our studies P_i deprivation led to an increase in apical Na-P_i cotransport activity that could not be prevented by actinomycin D (4). Similar to the rat studies, P_i deprivation of OK cells (clone 3B/2) led at early time points to transport alterations not paralleled by alteration in mRNA content (18). Under chronic conditions (24 h), P_i deprivation of OK cells led to an increased apical expression of Na-P_i cotransport activity that was not always paralleled by an increase in mRNA content (Ref. 18; M. Custer and J. Biber, unpublished observations). These data suggest that P_i deprivation (in the diet or medium) leads to an altered Na-P_i cotransport activity that can occur independent of an increased transcription rate. The data obtained from the reporter gene studies with the ok-Npt2 promoter in the "responsive" OK cell environment are in direct support of the interpretations mentioned above.

For PTH we could recently show that inhibition of rat proximal tubular Na-P_i cotransport is related to a withdrawal of the protein from the apical surface; also, the mRNA level was found to be reduced after PTH administration (2 h) to parathyroidectomized rats (12). Similar results were obtained in the OK cell system, where PTH treatment leads in short-term experiments to a rapid internalization of the Na-P_i cotransporter protein from the apical membrane (23). The present reporter gene studies were not able to detect a change in type II Na-P_i cotransporter promoter activity after prolonged PTH treatment (24 h).

We have also performed reporter gene studies with additional hormones known to regulate the type II Na-P_i cotransporter in OK cells. Experiments were performed for 24 h with dexamethasone (10⁶ M, see Ref. 31), T₃ (10⁷ M, see Ref. 28), with insulin-like growth factor I (10⁷ M), and with epidermal growth factor (10⁸ M, see Ref. 1). For all these conditions, we could not document reproducible and significant changes in transcriptional activity (data not shown). In contrast to our previous studies (~50% increase, see Ref. 28), we were unable to observe a T₃-induced increase in specific mRNA content (data not shown). Although we have no satisfactory explanation for this apparent discrepancy, we must conclude that T₃ effects on promoter activity seem to be rather small and that present experimental conditions (e.g., cell passage number and source of FCS in culture medium) might have prevented detection of small alterations in promoter activity and specific mRNA-content.

The present results indicate that the ok-Npt2 promoter is "constitutively" active and contains elements required for cell-specific expression. We were unable to document large changes in promoter activity for different modulators of the type II Na-P_i cotransporter in renal cells (proximal tubule and OK cells), indicating that the protein abundance might be regulated predominantly at the posttranscriptional level.