Main

The mammalian kidney is an important arbiter of extracellular phosphate(Pi) homeostasis. Renal reabsorption of filtered Pi occurs predominantly in the proximal tubule where concentrative Na+-gradient dependent Pi transport across the brush-border membrane is the rate limiting step in the overall Pi reabsorptive process and the major site for its regulation(1, 2). Recently, homologous cDNAs encoding high affinity, renal-specific, brush-border membrane Na+-Pi cotransporters (NPT2 type, Genome Data Base symbol SLC17A2) were cloned from several mammalian species(3–6), thereby providing useful probes to study the regulation of renal NPT2 gene expression and determine whether NPT2 is a candidate gene for Mendelian disorders of renal Pi reabsorption(7, 8).

We demonstrated that renal NPT2 gene expression is significantly reduced in X-linked Hyp mice, a murine homologue of human X-linked hypophosphatemia(9). We also showed that the NPT2 gene maps to human chromosome 5q35(10) and mouse chromosome 13B (X. Zhang, H. S. Tenenhouse, A. S. Hewson, H. Murer, and P. Eydoux, manuscript submitted for publication) by FISH, using the corresponding NPT2 cDNAs as probes. These findings indicate that mutations in NPT2 are not responsible for the homologous human and murine X-linked hypophosphatemias and suggest that the gene at the X-linked hypophosphatemia (Hyp) locus is involved in the regulation of renal NPT2 gene expression. A candidate gene for X-linked hypophosphatemia has recently been identified(11).

To determine whether NPT2 is a candidate gene for autosomal disorders of renal Pi wasting(7, 8), polymorphic markers mapping close to the NPT2 gene on chromosome 5q35(10) are required for linkage analysis. However, the application of this approach has been complicated by a second report which localized the NPT2 gene to human chromosome 5q13(12). The present study was undertaken to resolve this discrepancy. We used NPT2-specific primers to PCR amplify genomic DNA from a panel of human-hamster somatic cell hybrids which either contain or lack DNA in regions q13 and q35 of chromosome 5(13, 14) and a panel of chromosome 5 radiation hybrids (J. D. McPherson, unpublished data). In addition, we identified polymorphic microsatellite markers that flank the NPT2 locus on human chromosome 5 using the radiation hybrid panel and confirmed the localization of the NPT2 gene to chromosome 5q35 by FISH using an NPT2 genomic clone as probe.

METHODS

Deletion panel mapping. NPT2-specific oligonucleotide primers(forward 1405: 5′-CCGCTCACACTGGGTTCCAA-3′; reverse 1667: 5′-AGTGAGGGCAGCAGCAGGAA-3′) were used to amplify DNA isolated from somatic cell hybrids obtained from a chromosome 5 deletion panel(13, 14). The NPT2 primers are numbered according to their position in the NPT2 cDNA [NaPi-3 in(3)]. The PCR amplification conditions were 1.5 mM MgCl2, 0.4 μM of each primer, 200 μM of each dNTP and 1 unit of Taq polymerase(Boehringer Mannheim) with the supplied buffer. The PCR cycling parameters were 94 °C, 3 min (94 °C, 30 s/64 °C, 5 s/72 °C, 30 s) × 33; 72 °C, 7 min with an MJ Research PTC200 Thermocycler. PCR products and DNA Molecular Weight Marker VI (Boehringer Mannheim, Indianapolis, IN) were separated on 6% acrylamide gels.

Radiation hybrid mapping. NPT2-specific oligonucleotide primers(forward 618: 5′-GGAGGTGAGCTCTGCCATC-3′; reverse 764: 5′-CAGTTAAAGCAGTCATGCACC-3′) were used to amplify DNA isolated from 180 chromosome 5-specific radiation hybrids (J. D. McPherson, unpublished data). PCR conditions were essentially the same as above with the following cycling parameters: 94 °C, 1 min (94 °C, 30 s/60 °C, 30 s/72°C, 30 s) × 35; 72 °C, 10 min. PCR products and DNA Molecular Weight Markers (Boehringer Mannheim) were separated on 3% agarose gels and the hybrids scored for the presence or absence of the NPT2 locus. The position of NPT2 relative to previously ordered chromosome 5 markers was determined using the rh2pt and RHMAXLIK programs with the SAVMAX value of the latter set to 8(15).

FISH. A human PAC clone (203K6) containing the NPT2 locus was isolated from an arrayed library (generously provided by Pieter de Jong) using the NPT2-specific primers (forward 618 and reverse 764) described above. DNA from the PAC 203K6 clone was direct labeled with fluorescein-dUTP (Prime-It Fluor, Stratagene, La Jolla, CA) as recommended by the supplier and hybridized to human metaphase chromosomes, and images were collected as previously described(16). Over 30 metaphase spreads were examined.

RESULTS

Figure 1 depicts the results of PCR amplification of DNA isolated from a human chromosome 5 natural deletion panel, selected from human/hamster somatic cell hybrids(13, 14). An NPT2 PCR product of the expected size (962 bp) was generated with DNA from all cell lines containing the distal portion of the long arm of chromosome 5(HHW105, HHW1064, HHW1421, HHW1499, and HHW1600) but not with DNA from HHW1138, a cell line lacking the region of interest (Fig. 1). These findings localize NPT2 to 5q33.2-qter. Moreover, the demonstration that a PCR product was obtained with DNA from HHW1064, a cell line lacking the 5q13 region (Fig. 1), indicates that NPT2 does not map to this locus as reported by Ghishan et al.(12). No PCR product was detected in negative controls run in the absence of DNA or in the presence of hamster (UCW56) DNA (Fig. 1). A PCR product of the expected size was obtained with human (46XY) DNA as template (Fig. 1).

Figure 1
figure 1

Regional mapping of NPT2 by PCR amplification of DNA from a panel of human-hamster somatic cell hybrids. Solid lines below the lanes indicate the portion of human chromosome 5 present in the indicated hybrids. The PCR conditions were 1.5 mM MgCl2, 0.4 μM of each primer(forward 1405: 5′-CCGCTCACACTGGGTTCCAA-3′; reverse 1667: 5′-AGTGAGGGCAGCAGCAGGAA-3′), 200 μM of each dNTP and 1 unit of Taq polymerase (Boehringer Mannheim). The PCR cycling parameters were 94 °C, 3 min; (94 °C, 30 s/64 °C, 5 s/72 °C, 30 s)× 33; 72 °C, 7 min with an MJ Research PTC200 thermocycler. PCR products were separated on 6% acrylamide gels. Visible fragments of 2176 and 1766 bp (doublet), 1230, 1033, 653, and 517 bp were obtained with DNA Molecular Weight Marker VI (Boehringer Mannheim).

Full size image

Radiation hybrid mapping was performed by PCR analysis using NPT2-specific primers and DNA isolated from 180 chromosome 5-specific radiation hybrids. The hybrids were scored for the presence or absence of the NPT2 locus by the presence or absence of a 269 bp NPT2 PCR product. The position of NPT2 was determined relative to 248 markers previously ordered on human chromosome 5 by radiation hybrid mapping (J. D. McPherson, unpublished data). The NPT2 locus was localized between D5S498 and D5S469 using the RHMAXLIK program(15) at a log likelyhood > 10E-3. The calculated distance between NPT2 and D5S469 was 32.2 centiRays (cR) and between NPT2 and D5S498 was 65.9 cR, with associated LOD scores of 16.02 and 8.08, respectively. Calibration of the chromosome 5 radiation map, using both the chromosome size and sequence-tagged sites generated from the end clones of an 850-kb YAC, yielded an approximate value of 40 kb/cR, in agreement with that of a previous study(17). Thus, NPT2 is estimated to be 2.6 Mb distal to D5S498 and 1.3 Mb proximal to D5S469. The known localization of the dinucleotide repeat markers D5S498 and D5S469 to distal 5q is consistent with the data from the chromosome 5 deletion mapping panel shown in Figure 1.

The localization of NPT2 to human chromosome 5q35 was also confirmed by hybridization of human metaphase chromosomes with fluorescein-labeled, denatured genomic DNA from a human PAC clone (203K6) containing the NPT2 locus(data not shown). The genomic probe hybridized only to the q35 region of chromosome 5. Of all chromosomes examined with hybridization to 5q35, greater than 90% showed a signal on both chromatids. No signal was observed on any other chromosome.

DISCUSSION

We demonstrate by three independent methods that the gene encoding the high affinity, renal-specific brush-border membrane Na+-Pi cotransporter (NPT2, gene symbol SLC172A) maps to human chromosome 5q35. The methods include PCR amplification of genomic DNA from a panel of human-hamster somatic cell hybrids that either lack or contain DNA in regions 5q13 and 5q35(13, 14) and a chromosome 5 radiation hybrid panel(J. D. McPherson, unpublished data), using NPT2-specific primers. In addition, FISH was performed with an NPT2 genomic clone as probe. The present findings are consistent with our previous mapping data in which a human NPT2 cDNA was used as a probe for FISH(10). We suggest that the localization of NPT2 to chromosome 5q13(12) was likely the result of nonspecific hybridization (only nine cells of 50 examined gave a positive signal)(12). Moreover, improper chromosomal assignment may have occurred since hybridization was not performed directly on banded chromosomes(12).

Using the radiation hybrid panel, we were able to confirm the localization of NPT2, to position NPT2 relative to other chromosome 5 loci, and to identify flanking polymorphic microsatellite markers D5S498 and D5S469. These data provide a basis for linkage analysis to determine whether the Na+-Pi cotransporter is a candidate gene for several autosomal disorders of renal phosphate wasting(7, 8).

The NPT2-specific primers used in the present study were designed from the cDNA sequence(3). Each of the NPT2-specific primer pairs spans an intron, generating a PCR product that is larger than that expected from the cDNA sequence. The size of NPT2 introns was derived by cloning and characterization of the human NPT2 gene(18). Primer pairs “forward 618/reverse 764” and “forward 1405/reverse 1667” amplify introns 6 and 12 which are 122 and 700 bp, respectively(18). Because NPT2 is renal-specific and not expressed in white blood cells or skin fibroblasts, a knowledge of the intron/exon boundaries of the NPT2 gene is necessary for mutation analysis of the NPT2 coding sequence in genomic DNA samples. This information is relevant to patients with autosomal disorders of renal Pi wasting once linkage to polymorphic markers flanking NPT2 on human chromosome 5q35 is established.