Hereditary Hypophosphatemic Rickets with Hypercalciuria Is ... - JASN

J Am Soc Nephrol 12: 507–514, 2001

Hereditary Hypophosphatemic Rickets with Hypercalciuria Is Not Caused by Mutations in the Na/Pi Cotransporter NPT2 Gene ANDREW O. JONES,*㛳 JORDANA TZENOVA,* DANIELLE FRAPPIER,㛳 M. JOYCE CRUMLEY,㛳 NICOLE M. ROSLIN,㛳 CLAUDINE H. KOS,* MARTIN TIEDER,¶ CRAIG B. LANGMAN,‡‡ WILLEM PROESMANS,§§ THOMAS O. CARPENTER,㛳㛳 ALLAN RICE,㛳㛳 DONALD ANDERSON,¶¶ KENNETH MORGAN,†‡㛳 T. MARY FUJIWARA,†‡㛳 and HARRIET S. TENENHOUSE*†§㛳 Departments of *Biology, †Human Genetics, ‡Medicine, and §Pediatrics, McGill University, Montreal, Canada; 㛳McGill University Health Centre Research Institute, Montreal, Canada; ¶Department of Nephrology, Assaf Harofeh Medical Center, Beer Yaakov, Israel; ‡‡Department of Pediatric Nephrology, Northwestern University, Chicago, Illinois; §§Department of Pediatrics, University Ziekenhuis Gasthuisberg, Leuven, Belgium; 㛳㛳Department of Pediatrics, Yale University School of Medicine, New Haven, Connecticut; and ¶¶ Department of Pediatric Endocrinology, John Hunter Children’s Hospital, Newcastle, Australia.

Abstract. Hereditary hypophosphatemic rickets with hypercalciuria (HHRH), a renal phosphate (Pi) wasting disease first described in an extended Bedouin kindred, is characterized by hypophosphatemia, elevated serum 1,25-dihydroxyvitamin D levels, hypercalciuria, rickets, and osteomalacia. Correction of all abnormalities, except for renal Pi wasting, can be achieved by oral Pi supplementation. These findings and the demonstration that mice that are homozygous for the disrupted Na/Pi cotransporter gene Npt2 exhibit many of the biochemical features of HHRH suggested that mutations in the human orthologue NPT2 might be responsible for HHRH. The NPT2 gene

in affected individuals from the Bedouin kindred and four small families was screened for mutations to test this hypothesis. No putative disease-causing mutation was found. Two single nucleotide polymorphisms (SNP), a silent substitution in exon 7 and a nucleotide substitution in intron 4, were identified, and neither consistently segregated with HHRH in the Bedouin kindred. Linkage analysis indicated that the two NPT2 intragenic SNP as well as five microsatellite markers in the NPT2 gene region were not linked to HHRH in the Bedouin kindred. Therefore, this is evidence to exclude NPT2 as a candidate gene for HHRH in the families that were studied.

Hereditary hypophosphatemic rickets with hypercalciuria (HHRH; OMIM 241530) is a disorder of phosphate (Pi) homeostasis characterized by decreased renal Pi reabsorption, hypophosphatemia, an elevated serum concentration of 1,25dihydroxyvitamin D (1,25(OH)2D), hypercalciuria, and a decreased circulating concentration of parathyroid hormone (PTH) (1). These biochemical abnormalities are associated with bone pain, muscle weakness, growth retardation, and evidence of rickets and osteomalacia (1,2). The high serum 1,25(OH)2D concentration and hypercalciuria distinguish HHRH from Mendelian renal Pi wasting disorders, such as X-linked hypophosphatemia (XLH; OMIM 307800) and autosomal dominant hypophosphatemic rickets (ADHR; OMIM 193100) (3,4). Neither XLH nor ADHR is associated with

elevated circulating levels of the vitamin D hormone, despite hypophosphatemia, which serves as a stimulus for increased renal synthesis and decreased renal catabolism of 1,25(OH)2D (3,4). HHRH was first described in a large inbred Bedouin kindred (1), and only a few sporadic cases have been subsequently reported (5– 8). In an extensive study of the Bedouin kindred, 21 members were found to be clinically healthy but displayed “idiopathic” hypercalciuria (IH) and biochemical measurements that were intermediate between normal and affected members of the kindred (2). On the basis of the phenotypic findings in individuals with HHRH and IH, it was suggested that individuals with HHRH and IH are homozygous and heterozygous, respectively, for the same mutant allele (2). Treatment of HHRH is accomplished by oral Pi supplementation. It increases growth rate, corrects the rickets and osteomalacia, and normalizes all biochemical abnormalities, with the exception of renal Pi wasting (1). On the basis of these findings, it was suggested that HHRH is a primary disorder of renal Pi reabsorption. The kidney plays a key role in the maintenance of Pi

Received April 11, 2000. Accepted August 28, 2000. Correspondence to Dr. Harriet S. Tenenhouse, Montreal Children’s Hospital, 2300 Tupper Street, Montreal, Quebec H3H 1P3, Canada. Phone: 514-9344400, ext. 2342; Fax: 514-934-4331; E-mail: [email protected] 1046-6673/1203-0507 Journal of the American Society of Nephrology Copyright © 2001 by the American Society of Nephrology

508

Journal of the American Society of Nephrology

homeostasis (9). Of the three classes of Na/Pi cotransporters identified in mammalian kidney, the type II Na/Pi cotransporter, Npt2, is the most abundant at the mRNA level (10). The Npt2 protein was localized exclusively to the brush border membrane (BBM) of proximal tubule cells (11), where the majority of filtered Pi is reabsorbed, and is a target for regulation by PTH and dietary Pi, major regulators of renal Pi handling (12). The human orthologue, NPT2 (gene SLC34A1 in the Human Gene Nomenclature Database; http://www.gene.ucl.ac.uk/nomenclature) was mapped to chromosome region 5q35 by fluorescence in situ hybridization (13). The NPT2 gene spans 16 kb and consists of 13 exons (14); the initiation codon starts at nucleotide 82 in the second exon (15). The 2.5-kb coding region is predicted to code for a 635 amino acid protein that has eight transmembrane domains (15). Recent studies in mice in which the Npt2 gene was disrupted by targeted mutagenesis have underscored the importance of Npt2 in the maintenance of Pi homeostasis (16). Mice that are homozygous for the disrupted Npt2 gene (Npt2⫺/⫺) exhibit increased urinary excretion of Pi, decreased BBM Na/Pi cotransport, hypophosphatemia, and an elevated serum concentration of 1,25(OH)2D, with attendant hypercalcemia, hypercalciuria, and reduced circulating PTH levels (16). The biochemical features of Npt2⫺/⫺ mice resemble those of patients with HHRH, and on the basis of these findings, we hypothesized that NPT2 is a candidate gene for HHRH. The present study was undertaken to test this hypothesis in the Bedouin kindred in which the disease was first characterized (1,2), as well as in four small families from different countries.

Materials and Methods Pedigrees The principal study population was the extended Bedouin kindred characterized by Tieder et al. (1,2). DNA was available from 63 members of the family; 10 were classified as HHRH and 28 as IH (Figure 1). In addition, four small families were studied (Figure 2): family A from France (of Turkish origin), family B from Belgium, family C from the United States (of Scottish-Irish origin), and family D from Australia. Phenotypic data on families A to D are presented in Table 1. In addition, DNA from 18 Chinese, 19 French Canadians, 22 Greeks, and 20 Italians were used to assess the variation of NPT2 allele frequencies. The studies were approved by institutional review committees, and all participants gave informed consent.

DNA Sequencing Intronic primers 10 to 50 bp from the intron/exon boundaries were designed to amplify the 13 NPT2 exons (Table 2). Exons were amplified as single fragments or amplified together with other exons. PCR was done in a volume of 50 ␮l with 100 ng of genomic DNA, 2.5 U of Platinum Taq (BRL Life Technologies, Burlington, Ontario, Canada), 0.4 ␮M primer, and 1 ␮M dNTP. PCR conditions were as follows: 30 s at 94°C, 30 s at the annealing temperature of the specific primer pair (Table 2), and 60 s at 72°C, for 35 cycles, with a final extension at 72°C for 10 min. PCR products were sequenced either using the T7 Sequenase Kit with [␣-33P]dNTP (Amersham Pharmacia Biotech, Baie d’Urfe, Quebec, Canada) or on an ABI 373A sequencer (Laval University Sequencing Service, Quebec, Quebec, Canada; or

J Am Soc Nephrol 12: 507–514, 2001

McGill University Sheldon Biotechnology Center, Montreal, Quebec, Canada).

Screening of NPT2 Polymorphisms Assays based on amplification-created or naturally occurring restriction sites were developed to screen for the single nucleotide polymorphisms (SNP) identified by DNA sequencing. The T-to-C substitution in intron 4 creates an HgaI site when amplified with primer I4R (5'-GATGTGGAGGTGCTGGAGCTCT-3') and a primer (5'-GGTTCCTGAAGGGCCTTGGACgA-3') designed with a mismatch (denoted by lower-case g). The PCR products were digested overnight with 4 U of HgaI at 37°C. Exon 7 was amplified using the forward primer for exons 7 to 8 (Table 2) and 7R (5'-AGGAGGCTGGGTCCCTTGTTAC-3') at an annealing temperature of 61°C. The T-to-C substitution in exon 7 abolishes an NlaIII site, and overnight digestion of the PCR product was performed with 5 U of NlaIII at 37°C. The digests were run on 10% acrylamide minigels and stained with ethidium bromide; genotypes were assigned on the basis of the digestion pattern. For the intron 4 SNP, the T allele has one fragment of 158 bp; the C allele has two fragments of 131 and 27 bp. For the exon 7 SNP, the T allele has four fragments of 105, 87, 62, and 36 bp; the C allele has three fragments of 192, 62, and 36 bp. An allelespecific hybridization assay was also developed to screen for the exon 7 SNP (details available from the authors).

Genotyping Microsatellite Markers DNA samples from the Bedouin kindred were genotyped for markers that flank NPT2 in chromosome region 5q35 (D5S2108, D5S2058, D5S498, D5S469, and D5S408) by PCR amplification using the protocol for [␥-32P] end-labeled primers (17) with minor modifications. Primer sequences were obtained from the Genome Database (http://gdbwww.gdb.org). Marker order and sex-averaged distances were obtained from the Marshfield Medical Research Foundation web site (http://research.marshfieldclinic.org/genetics). NPT2 was placed between D5S498 and D5S469 on the basis of radiation hybrid mapping (http://chrom5.hsis.uci.edu). Thus, our marker order and approximate intermarker distances are D5S2108—1.33 cM—D5S2058— 0.98 cM—D5S498—2.29 cM—NPT2—2.29 cM—D5S469— 6.26 cM—D5S408. We manually constructed haplotypes by attempting to minimize the number of recombinants and assuming that there was no mutation of marker alleles.

Statistical Analyses Evidence for linkage between a locus for HHRH and chromosome 5 markers was analyzed in the Bedouin kindred using a genetic model of an autosomal recessive trait with either complete penetrance or 75% penetrance, and a disease allele frequency of 0.025. Although HHRH is a rare disease, we assumed that it may be more common in the Bedouin population than in the general population and chose a relatively high carrier prevalence of 5%. Two-point parametric linkage analysis was done using the MLINK program of the FASTLINK package (version 4.1P) (18,19) of the LINKAGE programs (version 5.2) (20). Marker allele frequencies were obtained using the program NEWPREP (Whitehead Institute for Biomedical Research, Boston, MA). Multipoint nonparametric linkage analysis was performed with SimWalk2, version 2.60 (21). This program uses Markov chain Monte Carlo methods to assess the degree of allele sharing that is identical by descent among affected members of a pedigree. The analysis was run using 50,000 simulations of the pedigree. We report P values for statistic A, which measures the number of different paths that lead to marker alleles that appear among affected individuals. Statistic A was

J Am Soc Nephrol 12: 507–514, 2001

Exclusion of NPT2 as a Candidate Gene for HHRH

509

Figure 1. Partial pedigree of the Bedouin kindred showing genealogical links among individuals with hereditary hypophosphatemic rickets with hypercalciuria (HHRH). The two individuals with HHRH selected for DNA sequencing are marked with an asterisk. Double marriage bars indicate consanguinity. Haplotypes for the NPT2 single nucleotide polymorphisms (SNP) and flanking microsatellite markers are shown below each individual. The assignment of haplotypes was based on all 63 family members with genotype data, which included siblings and cousins (not shown in pedigree) of individuals with HHRH. Parentheses indicate haplotypes that were inferred.

510


J Am Soc Nephrol 12: 507–514, 2001

Figure 2. Four small families (A, B, C, and D) with HHRH. In family B, individual II-3 was considered to have HHRH, although there was no evidence of rickets (see Table 1). Individuals with HHRH who were selected for DNA sequencing are marked with an asterisk. The variability of the intermediate phenotype is indicated by filled quadrants. Haplotypes of the NPT2 SNP are shown below each individual.

Table 1. Clinical data for families A to Da Family A b

Age at measurement (yr) Serum Pi (mM) TmP/GFR (mM) Urine Ca (mg/kg per d) Urine Ca/creatinine Serum 1,25(OH)2D (pM) Histomorphometry Bone mineral density X-ray

Family B

Family C b

V-1

II-1

II-2

II-3

I-1

I-2

II-1

II-2b

8 0.6 (1.3–1.85) 0.48 (1.82–2.36) NA 1 (⬍0.5) 288 (72–144) NA NA

20 0.9 (1.0–1.5) 0.68 (1.15–2.44) 6.5 (⬍4.4) NA 178 (48–156) NA Below normal

18 1.22 (1.0–1.5) 1.12 (1.15–2.44) 2 (⬍4.4) NA 91 (48–156) NA NA

11 1.1 (1.16–1.87) 0.93 (1.16–1.29) 7 (⬍4.4) NA 340 (36–168) NA NA

48 0.42 (1.0–1.5) 0.4 (1.15–2.44) NA 0.32 (⬍0.6) 173 (36–168) NA NA

45 1.03 (1.0–1.5) 1.2 (1.15–2.44) NA 0.6 (⬍0.6) 164 (36–168) NA NA

15 1.0 (1.0–1.5) 1.44 (1.15–2.44) NA 0.25 (⬍0.6) 161 (36–168) NA NA

Rickets

NA

NA

Osteopenia

NA

NA

NA

14 0.77 (1.0–1.5) 0.8 (1.15–2.44) 5.6 (⬍4.0) 0.94 (⬍0.6) 234 (48–156) NA Well below normal Osteopenia, metaphyseal and epiphyseal fraying

a Pi, phosphate; TmP/GFR, tubular maximum for Pi reabsorption per GFR; HHRH, hereditary hypophosphatemic rickets with hypercalciuria; NA, data not available. Measurements were obtained at the time of diagnosis by the investigating physician. Normal ranges for age-matched control subjects are given in parentheses. TmP/GFR was determined by nomogram (36). Urine Ca/creatinine values are based on 24-h urine collections. Bone mineral density was determined by dual-energy X-ray absorptiometry of the lumbar spine (L2 to L4). Data for the individual with HHRH in family A is from reference 37; data were unavailable for other family members. b Individuals with a diagnosis of HHRH.

J Am Soc Nephrol 12: 507–514, 2001


chosen because it tends to have the most power for detecting linkage to a recessively inherited trait.

Results Phenotypic Characteristics of Individuals with HHRH Individuals with HHRH in the extended Bedouin kindred (Figure 1) were previously characterized by hypophosphatemia, decreased tubular maximum for Pi reabsorption per GFR (TmP/GFR), elevated serum 1,25(OH)2D levels, hypercalciuria, rickets, and osteomalacia (1,2,22). Individuals with HHRH in families A to D (Figure 2) exhibited hypophosphatemia and decreased TmP/GFR when compared with agematched control subjects (Table 1). In addition, they exhibited hypercalciuria and elevated serum concentrations of 1,25(OH)2D. Skeletal abnormalities were evident in all cases. The intermediate phenotype in families A to D was more variable and differed somewhat from the IH phenotype characterized by Tieder et al. (1,2). Phenotypic variation was observed both within and between families and is shown in Table 1 and Figure 2. Figure 1 shows a partial pedigree of the Bedouin kindred with 12 individuals having a diagnosis of HHRH. The members of the Bedouin kindred are related in many different ways; there are eight known inbreeding loops. The relationships among the 10 nuclear families with a child affected with HHRH range from parent– offspring to third cousins or third cousins once-removed in two different ways. Not all pairs of nuclear families are known to be related to each other, and there is no single recent ancestor common to all of the parents based on the available genealogical information. The parents of

511

three sibships are consanguineous; two sets of parents are related as first cousins and one set as second cousins onceremoved. Consanguinity is also evident in family A, in which the parents of the affected individual are related as second cousins (Figure 2).

Mutation Analysis Six individuals with HHRH were selected for DNA sequence analysis: two from the Bedouin kindred and one each from families A to D (see Figures 1 and 2). Sequencing of the NPT2 exons in these six individuals did not reveal any putative disease-causing mutation. In addition, no differences were found in 120 bp of the promoter region among the individuals with HHRH, control subjects, and published sequences (14,23); available intronic sequence agreed with that reported previously (23). Two SNP were identified in intron 4 and exon 7. Sequencing revealed both a T and a C in intron 4, 54 bp from the 3' end of exon 4 (nucleotide 78 in GenBank accession AH007379). The common name for this SNP is designated as IVS4⫹54T⬎C, and the systematic name is c.469⫹54T⬎C. The second SNP was a silent substitution in codon 258 (CAT or CAC) in exon 7 (T or C at nucleotide 774 where, for numbering purposes, nucleotide 1 is the A in the initiation codon; nucleotide 855 in GenBank reference sequence NM_003052). The common name for this SNP is designated as H258H, and the systematic name is c.774T⬎C. Single-strand conformation analysis was also used to search for mutations in the Bedouin kindred and in families B and D. No conformational changes were observed other than the band shift caused by H258H (data not shown).

Table 1.—Continued Family D I-2

I-3

II-1

II-2

II-3

II-4

II-5b

II-6

48 0.89 (1.0–1.5) NA

37 0.97 (1.0–1.5) NA

19 0.90 (1.0–1.5) NA

17 0.80 (1.0–1.5) NA

14 0.75 (1.16–1.87) NA

11 0.75 (1.16–1.87) NA

16 0.96 (1.0–1.5) NA

NA 0.5 (⬍0.6) 96 (38–162) Increased bone turnover NA

NA 1.0 (⬍0.6) 79 (38–162) Increased bone formation Marginally below normal Mild genu valgum

NA 0.66 (⬍0.6) 68 (38–162) NA

NA 0.62 (⬍0.6) 94 (38–162) NA

NA 0.32 (⬍0.6) 344 (38–162) Osteomalacia

NA 0.24 (⬍0.6) 68 (38–162) NA

NA

NA

Slightly below normal

NA

NA

NA

NA 0.85 (⬍0.6) 240 (38–162) Mild increase in bone formation Slightly below normal NA

10 0.77 (1.16–1.87) 0.76 (1.16–1.29) NA 0.8 (⬍0.6) 566 (45–175) Osteomalacia

Abnormal trabecular pattern, widening of metaphyses, and rickets

NA

NA

Markedly below normal Mild rachitic changes

512


J Am Soc Nephrol 12: 507–514, 2001

Table 2. Primers for DNA sequence analysis Exon(s)

PCR Length (bp)

TAa

Forward Primer (5⬘–3⬘)

Reverse Primer (5⬘–3⬘)

1 2 3 4 5 2–5 2–6 6–8 7–8 9–10 11–13

290 357 332 254 214 957 2300 718 468 588 1900

58°C 58°C 66°C 65°C 61°C 60°C 62°C 60°C 60°C 59°C 62°C

GAGGAATCTCAGGGGTGAGGTT ATGGGGGTTTGTGCAGGTGA AAGTTGACTGCTCCCACACCCT GCAAGAGGCGGCAGCAGTCCCTG CTGAAGGGCCTTGGACAATG ATGGGGGTTTGTGCAGGTGA ATGGGGGTTTGTGCAGGTGA GGAGTTGTGGTGGTGCAGGA GAGCTGACTGCCTTGCAATGTG CCTCAGGCTGTGTTTTCCTA GCACTAGGGATGCGGTTTCC

ATCAAGCTTATCGATACC AGGTATGCTCAGCAAGGGCCACA CAGGGACTGCTGCCGCCTCTTGC CATTGTCCAAGGCCCTTCAGGA GCTCTTCCCCACCCTGGT GCTCTTCCCCACCCTGGT AGCCCCTCATACCCCTCAGC GAGTCTGAAGAGGGAGTCCT GAGTCTGAAGAGGGAGTCCT CTGTGCCCTGCCCAAGTCAA CGTGCAGAGGACCTGAAATC

a

Annealing temperature.

DNA samples from all of the available family members were typed for the two NPT2 SNP. The data in Figure 1 show that they do not consistently segregate with HHRH in the Bedouin kindred. Among those with HHRH, there were two genotypic combinations of IVS4⫹54T⬎C and H258H: CT/CT (n ⫽ 7) and TT/CC (n ⫽ 3). Segregation of the SNP in families A to D does not provide any evidence for or against linkage to HHRH (Figure 2). Both NPT2 SNP were polymorphic in an ethnically diverse sample of 79 individuals. The allele frequencies did not differ significantly among the Chinese, French Canadians, Greeks, and Italians (Fisher’s exact test, P ⫽ 0.08 for IVS4⫹54T⬎C and P ⫽ 0.13 for H258H). The predominant allele for IVS4⫹54T⬎C was C (allele frequency ⫽ 0.78; SD ⫽ 0.03) and for H258H was T (allele frequency ⫽ 0.71; SD ⫽ 0.04). Thirty-four of the 79 individuals were heterozygous for one or both SNP.

Linkage Analysis Visual inspection of the genotypes in the Bedouin kindred indicated that two of the individuals with HHRH (604 and 675), who are children of first cousins, and five others with HHRH (527, 529, 655, 583, and 589) were not homozygous for the NPT2 SNP (Figure 1). Furthermore, none of the three patients who were homozygous for the NPT2 SNP (603, 664, and 666) were homozygous for the haplotype consisting of the flanking microsatellite markers. Notably, the two affected siblings who are children of first cousins (603 and 604) share only one haplotype, and the other pair of affected siblings shares no haplotype (655 and 664). In families B, C, and D, none of the individuals with HHRH were homozygous for both NPT2 SNP (Figure 2). In family A, the HHRH patient is homozygous for the NPT2 SNP but is not homozygous for four of the five microsatellite markers (data not shown). Thus, within each of these kindreds, we infer that individuals with HHRH are not homozygous for an undetected mutation in NPT2 that was inherited from a recent common ancestor. Parametric linkage analysis in the Bedouin kindred, under

the assumption that HHRH is a completely penetrant autosomal recessive trait, indicated that there were recombinants between a locus for HHRH and the NPT2 SNP as well as for four of the five microsatellite markers. Linkage to NPT2 within 1% recombination could be excluded (lod scores of ⫺2.4 and ⫺2.6 for the intron 4 and exon 7 SNP, respectively). Linkage could be excluded within 5% recombination with D5S2108 and D5S2058, within 2% recombination with D5S498, and within 13% recombination with D5S408 (lod scores ⬍ ⫺2). Only two alleles were observed for D5S469, and this marker was not informative in most matings. Reducing the penetrance to 75% gave essentially the same results; lod scores of ⫺3.9 and ⫺4.1 were obtained for linkage of HHRH to the intron 4 and exon 7 SNP, respectively, at 0% recombination. Because a precise genetic model is not known for HHRH, nonparametric analysis was also performed to test for linkage. The NPT2 SNP are not sufficiently informative for two-point nonparametric linkage analysis; therefore, multipoint analysis was done with SimWalk2. The program was run four times using 50,000 simulations of the pedigree for each run. The average empirical P values for statistic A of the four runs was greater than 0.18 (SD ranged from 0.02 to 0.08) at every marker (D5S2108, D5S2058, D5S498, D5S469, D5S408, and the two NPT2 SNP). In particular, the average P value was 0.23 for both NPT2 SNP (P values ranged from 0.12 to 0.27 over the four runs; SD was 0.07 for both SNP). Clustering among affected individuals of alleles descending from the founders is expected when there is linkage. The multipoint nonparametric analysis of the Bedouin kindred indicates that there is no significant excess marker allele sharing among pedigree members affected with HHRH and provides further evidence against linkage of NPT2 to a locus for HHRH. Linkage analysis was not done on families A to D as the available data do not have sufficient power to test for linkage.

Discussion The present study was undertaken to determine whether mutations in the NPT2 gene are responsible for HHRH. We

J Am Soc Nephrol 12: 507–514, 2001

found no disease-causing mutation in the NPT2 gene in six individuals with HHRH from five unrelated families. Two SNP were identified: one in exon 7 that is a silent substitution in codon 258 (His) and one in intron 4. The results of linkage analysis in the Bedouin kindred using these SNP and five microsatellite markers that flank NPT2 provide evidence to exclude NPT2 as a candidate gene for HHRH under the assumption of autosomal recessive inheritance with complete or incomplete (75%) penetrance. In addition, there was no evidence for linkage, based on increased sharing of NPT2 alleles inherited identical by descent. Haplotype analysis provided further evidence that individuals with HHRH are not homozygous for an unidentified NPT2 mutation that was inherited from a recent common ancestor. Exclusion of NPT2 as a candidate gene for HHRH in the families studied is consistent with the observed differences in the skeletal phenotype in individuals with HHRH and in mice that are homozygous for the disrupted Npt2 gene. Individuals with HHRH have radiologic and histomorphometric evidence of rickets and osteomalacia (1,22). In contrast, Npt2⫺/⫺ mice are characterized by a unique age-dependent bone phenotype. At weaning, Npt2⫺/⫺ mice exhibit poorly developed metaphyseal trabeculae and retarded secondary ossification in the epiphysis when compared with wild-type littermates (16). However, with increasing age, the knockout mice exhibit a reversal and eventual correction of the skeletal abnormalities observed at weaning (16). An age-related improvement of the bone abnormalities is not evident in individuals with HHRH, and treatment with oral Pi supplementation is necessary to heal the rickets and osteomalacia (1). It has been suggested that HHRH and IH could be the result of mutations in two different genes (2). The strongest support against this hypothesis is that when individuals with HHRH are treated with oral Pi, both the hypophosphatemia and the hypercalciuria are corrected. However, if HHRH is an autosomal recessive disease and individuals with IH are heterozygous for a mutant allele, then IH must be an incompletely penetrant trait because not all obligate heterozygotes have IH. In families A to D, however, the variability in clinical presentation suggests that mutations in different genes may be involved. Possible candidate genes for HHRH include genes that encode other renal Pi transporters or their regulators. The type I Na/Pi cotransporter accounts for approximately 13% of known Na/Pi cotransporter mRNA in mouse kidney (10) and, as is the case for Npt2, the immunoreactive protein was localized to the BBM of proximal tubule cells (24). The human orthologue NPT1 (also called SLC17A1) was mapped to chromosome region 6p22 (25). Analysis of allele sharing among individuals with HHRH using three microsatellite markers (D6S2238, D6S2237, and D6S2233) in a 500-kb region that includes NPT1 failed to provide evidence of linkage to HHRH in the Bedouin kindred under the assumption that HHRH is a completely penetrant autosomal recessive disease (26). The type III Na/Pi cotransporters that have a broad tissue distribution (27) account for less than 1% of known Na/Pi cotransporter mRNA in mouse kidney (10). They have been postulated to serve as housekeeping Na/Pi cotransporters, and,


513

thus, their involvement in HHRH is unlikely. Other likely candidate genes for HHRH are the orthologues of diphor-1 (28) and Pius (29), which map to human chromosome region 1q21 (30) and 3p21 (31), respectively. These proteins modify Pi transport, and both have been shown to stimulate Na/Pi cotransport in oocytes co-injected with Npt2 (28,29). The reabsorption of solutes and ions in the proximal tubule can also occur through tight junction-regulated paracellular pathways. The gene for paracellin-1 was recently identified by positional cloning, and mutations were found in patients with hypomagnesemia and magnesium wasting (32). These findings, as well as functional studies, implicated paracellin-1 in the paracellular transport of Mg2⫹ (32). Although tight junctions are more selective for cations than anions, further studies are necessary to determine whether a similar mechanism exists for renal Pi reabsorption. BBM Na/Pi cotransport is extremely sensitive to changes in pH, and this effect is in large part explained by a competition of protons with Na ions for interaction with the cotransporter protein (33). Accordingly, genes involved in the regulation of luminal pH are likely to be important in modifying renal Na/Pi cotransport and may have a role to play in the pathophysiology of HHRH. XLH, which is phenotypically similar to HHRH, with the exception of inappropriately regulated serum 1,25(OH)2D concentration, is caused by mutations in the PHEX gene, which has significant similarity to members of the M13 endopeptidase family (34). Although the endogenous substrate(s) for PHEX has not yet been identified, it is postulated to be a paracrine/endocrine factor involved in the regulation of skeletal mineralization and Pi homeostasis (34). Thus, mutations in genes involved in the PHEX pathway are possible candidates for HHRH. The two SNP that were identified in NPT2 may be useful for association studies of multifactorial diseases such as nephrolithiasis and osteoporosis in which NPT2 may play a role. Given the recent study suggesting that SNP can perturb mRNA secondary structure (35), the silent substitution in exon 7 may have an impact on NPT2 gene expression, which in turn could affect Pi homeostasis. In summary, we identified two SNP in the NPT2 gene and demonstrated that these sequence variants as well as microsatellite markers that flank the NPT2 gene are not linked to HHRH in the families that we studied. Further studies, using candidate gene or genome-wide scan approaches, are necessary to determine the genetic factor(s) responsible for HHRH.

Acknowledgments We are grateful to all of the individuals who participated in our study. We thank Carl Brewer for assistance with computer programs. This work was supported by grants from The Medical Research Council of Canada (GR-13297 to HST) and the Networks of Excellence Program—the Canadian Genetic Diseases Network and the Mathematics of Information Technology and Complex Systems. TMF was supported by a gift to McGill University from Alcan Aluminum Limite´e.

References 1. Tieder M, Modai D, Samuel R, Arie R, Halabe A, Bab I, Gabizon D, Lieberman UA: Hereditary hypophosphatemic rickets with hypercalciuria. N Engl J Med 312: 611– 617, 1985

514


2. Tieder M, Modai D, Shaked U, Samuel R, Arie R, Halabe A, Maor J, Weissgarten J, Averbukh Z, Cohen N, Edelstein S, Lieberman UA: “Idiopathic” hypercalciuria and hereditary hypophosphatemic rickets. N Engl J Med 316: 125–129, 1987 3. Rasmussen H, Tenenhouse HS: Mendelian hypophosphatemias. In: The Metabolic and Molecular Basis of Inherited Disease, 7th Ed., edited by Scriver CR, Beaudet AL, Sly WS, Valle D, New York, McGraw Hill, 1995, pp 3717–3745 4. Tenenhouse HS, Econs MJ: Mendelian hypophosphatemias. In: The Metabolic and Molecular Basis of Inherited Disease, 8th Ed., edited by Scriver CR, Beaudet AL, Sly WS, Valle D, New York, McGraw Hill, 2000, pp 5039 –5067 5. Nishiyama S, Inoue F, Matsuda I: A single case of hypophosphatemic rickets with hypercalciuria. J Pediatr Gastroenterol Nutr 5: 826 – 829, 1986 6. Proesmans WC, Fabrey G, Marchal GJ, Gillis PL, Bouillon R: Autosomal dominant hypophosphatemia with elevated serum 1,25 dihydroxyvitamin D and hypercalciuria. Pediatr Nephrol 1: 479 – 484, 1987 7. Chen C, Carpenter T, Steg N, Baron R, Anast C: Hypercalciuric hypophosphatemic rickets, mineral balance, bone histomorphometry, and therapeutic implications of hypercalciuria. Pediatrics 84: 276 –280, 1989 8. Tieder M, Arie R, Bab I, Maor J, Liberman UA: A new kindred with hereditary hypophosphatemic rickets with hypercalciuria: Implications for correct diagnosis and treatment. Nephron 62: 176 –181, 1992 9. Tenenhouse HS: Cellular and molecular mechanisms of renal phosphate transport. J Bone Miner Res 12: 159 –164, 1997 10. Tenenhouse HS, Roy S, Martel J, Gauthier C: Differential expression, abundance and regulation of Na⫹-phosphate cotransporter genes in murine kidney. Am J Physiol 44: F527–F534, 1998 11. Custer M, Lo¨tscher M, Biber J, Murer H, Kaissling B: Expression of Na-Pi cotransport in rat kidney: Localization by RT-PCR and immunohistochemistry. Am J Physiol 266: F767–F774, 1994 12. Levi M, Kempson SA, Lotscher M, Biber J, Murer H: Molecular regulation of renal phosphate transport. J Membr Biol 154: 1–9, 1996 13. Kos CH, Tihy F, Econs MJ, Murer H, Lemieux N, Tenenhouse HS: Localization of a renal sodium phosphate cotransporter gene to human chromosome 5q35. Genomics 19: 176 –177, 1994 14. Hartmann CM, Hewson AS, Kos CH, Hilfiker H, Soumoumou Y, Murer H, Tenenhouse HS: Structure of murine and human renal type II Na⫹-phosphate cotransporter genes (Npt2 and NPT2). Proc Natl Acad Sci USA 93: 7409 –7414, 1996 15. Magagnin S, Werner A, Markovich D, Sorribas V, Stange G, Biber J, Murer H: Expression cloning of human and rat renal cortex Na/Pi cotransport. Proc Natl Acad Sci USA 90: 5979 – 5983, 1993 16. Beck L, Karaplis AC, Amizuka N, Hewson AS, Ozawa H, Tenenhouse HS: Targeted inactivation of Npt2 in mice leads to severe renal phosphate wasting, hypercalciuria and skeletal abnormalities. Proc Natl Acad Sci USA 95: 5372–5377, 1998 17. Hudson TJ, Clark CD, Gschwend M, Justice-Higgins E: PCR methods of genotyping. In: Current Protocols in Human Genetics, edited by Dracopoli NC, Haines JL, Korf BR, Moir DT, Morton CC, Seidman CE, Seidman JG, Smith DR, New York, John Wiley & Sons, 1997, pp 2.5.1–2.5.23 18. Cottingham RWJ, Idury RM, Schaffer AA: Faster sequential genetic linkage computations. Am J Hum Genet 53: 252–263, 1993 19. Becker A, Geiger D, Schaffer AA: Automatic selection of loop breakers for genetic linkage analysis. Hum Hered 48: 49 – 60, 1998

J Am Soc Nephrol 12: 507–514, 2001

20. Lathrop GM, Lalouel JM, Julier C, Ott J: Strategies for multilocus linkage analysis in humans. Proc Natl Acad Sci USA 81: 3443–3446, 1984 21. Sobel E Lange K: Descent graphs in pedigree analysis: Applications to haplotyping, location scores, and marker-sharing statistics. Am J Hum Genet 58: 1323–1337, 1996 22. Gazit D, Tieder M, Liberman UA, Passi-Even L, Bab IA: Osteomalacia in hereditary hypophosphatemic rickets with hypercalciuria: A correlative clinical-histomorphometric study. J Clin Endocrinol Metab 72: 229 –235, 1991 23. Taketani Y, Miyamoto K, Tanaka K, Katai K, Chikamori M, Tatsumi S, Segawa H, Yamamoto H, Morita K, Takeda E: Gene structure and functional analysis of the human Na⫹/phosphate co-transporter. Biochem J 324: 927–934, 1997 24. Biber J, Custer M, Werner A, Kaissling B, Murer H: Localization of NaPi-1, a Na/Pi cotransporter, in rabbit kidney proximal tubules. II. Localization by immunohistochemistry. Pflu¨gers Arch 424: 210 –215, 1993 25. Kos CH, Tihy F, Murer H, Lemieux N, Tenenhouse HS: Comparative mapping of Na⫹-phosphate cotransporter genes, NPT1 and NPT2, in human and rabbit. Cytogenet Cell Genet 75: 22–24, 1996 26. Kos CH: The role of the renal sodium-dependent phosphate cotransporter genes, NPT1 and NPT2, in inherited hypophosphatemias, unpublished doctoral dissertation, Montreal, Quebec, Canada, McGill University, 1998 27. Kavanaugh MP, Miller DG, Zhang W, Law W, Kozak SL, Kabat D, Miller AD: Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-phosphate symporters. Proc Natl Acad Sci USA 91: 7071–7075, 1994 28. Custer M, Spindler B, Verrey F, Murer H, Biber J: Identification of a new gene product (diphor-1) regulated by dietary phosphate. Am J Physiol 273: F801–F806, 1997 29. Norbis F, Boll M, Stange G, Markovich D, Verrey F, Biber J, Murer H: Identification of a cDNA/protein leading to an increased Pi-uptake in Xenopus laevis oocytes. J Membr Biol 156: 19 –24, 1997 30. White KE, Biber J, Murer H, Econs MJ: A PDZ domain-containing protein with homology to Diphor-1 maps to human chromosome 1q21. Ann Hum Genet 62: 287–290, 1998 31. White KE, Econs MJ: Localization of PIUS, a stimulator of cellular phosphate uptake, to human chromosome 3p21.3. Somat Cell Mol Genet 24: 71–74, 1998 32. Simon DB, Lu Y, Choate KA, Velazquez H, Al-Sabban E, Praga M, Casari G, Bettinelli A, Colussi G, Rodriguez-Soriano J, McCredie D, Milford D, Sanjad S, Lifton RP: Paracellin-1, a renal tight junction protein required for paracellular Mg2⫹ resorption. Science 285: 103–106, 1999 33. Amstutz M, Mohrmann M, Gmaj P, Murer H: Effect of pH on phosphate transport in rat renal brush border membrane vesicles. Am J Physiol 248: F705–F710, 1985 34. Tenenhouse HS: X-linked hypophosphatemia: A homologous disorder in humans and mice. Nephrol Dial Transplant 14: 333–341, 1999 35. Shen LX, Basilion JP, Stanton VP Jr: Single-nucleotide polymorphisms can cause different structural folds of mRNA. Proc Natl Acad Sci USA 96: 7871–7876, 1999 36. Walton RJ, Bijvoet OL: Nomogram for the derivation of renal threshold phosphate concentration. Lancet 2: 309 –310, 1975 37. Sermet-Gaudelus I, Garabedian M, Lenoir G, Souberbielle JC, Lemenager F, Niaudet P, Tieder M, Dechaux M: A new kindred of hereditary hypophosphatemic rickets with hypercalciuria (HHRH) [Abstract]. J Am Soc Nephrol 9: 570A, 2000