Digenic Heterozygous Mutations in SLC34A3 and SLC34A1 Cause Dominant Hypophosphatemic Rickets with Hypercalciuria

Abstract

Context

Hypophosphatemia and metabolic bone disease are associated with hereditary hypophosphatemic rickets with hypercalciuria (HHRH) due to biallelic mutations of SLC34A3 encoding the NPT2C sodium-phosphate cotransporter and nephrolithiasis/osteoporosis, hypophosphatemic 1 (NPHLOP1) due to monoallelic mutations in SLC34A1 encoding the NPT2A sodium-phosphate cotransporter.

Objective

To identify a genetic cause of apparent dominant transmission of HHRH.

Design and Setting

Retrospective and prospective analysis of clinical and molecular characteristics of patients studied in 2 academic medical centers.

Methods

We recruited 4 affected and 3 unaffected members of a 4-generation family in which the proband presented with apparent HHRH. We performed clinical examinations, biochemical and radiological analyses, and molecular studies of genomic DNA.

Results

The proband and her affected sister and mother carried pathogenic heterozygous mutations in 2 related genes, SLC34A1 (exon 13, c.1535G>A; p.R512H) and SLC34A3 (exon 13, c.1561dupC; L521Pfs*72). The proband and her affected sister inherited both gene mutations from their mother, while their clinically less affected brother, father, and paternal grandmother carried only the SLC34A3 mutation. Renal phosphate-wasting exhibited both a gene dosage–effect and an age-dependent attenuation of severity.

Conclusions

We describe a kindred with autosomal dominant hypophosphatemic rickets in which whole exome analysis identified digenic heterozygous mutations in SLC34A1 and SLC34A3. Subjects with both mutations were more severely affected than subjects carrying only one mutation. These findings highlight the challenges of assigning causality to plausible genetic variants in the next generation sequencing era.

Phosphorus homeostasis represents an equipoise between dietary phosphate absorption, influx and efflux of phosphorus from bone and intracellular stores, and reabsorption of filtered phosphate in the renal proximal tubule (1). The renal reabsorption of phosphate is controlled by the integrated actions of parathyroid hormone (PTH), 1,25-dihydroxyvitamin D (1,25(OH)₂D), and phosphotonins such as fibroblast growth factor 23 (FGF23) on 2 members of the SLC34 solute carrier family, the renal sodium-coupled inorganic phosphate cotransporters NPT2A and NPT2C (2, 3). These 2 proteins constitute the main Na(+)-driven pathways for apical entry of inorganic phosphate (Pi) across renal epithelium, and are predominantly expressed in the proximal tubules of the kidney. Recently, PiT-2 (SLC20A2), a member of solute carrier family 20, has also been shown to mediate phosphate reabsorption in the renal proximal tubule, however (4).

NPT2A and NPT2C share physical and functional similarities, but differ on their ability and need to interact with the scaffolding protein Na/H exchange regulatory factor 1 (NHERF1). Expression of NPT2A requires interaction of its C-terminal region with NHERF1, which in turn interacts with cytoskeletal proteins (5). The importance of NHERF1 in phosphate homeostasis is supported by studies showing that genetic deficiency of NHERF1 in mice (6) and humans (7) impairs NPT2A expression, leading to renal phosphate loss and hypophosphatemia. The expression and action of NPT2A and NPT2C are decreased by PTH and FGF23 and, at least for NPT2A, increased by 1,25(OH)₂D (8, 9).

Appreciation of the key roles that these transporters play in mammalian phosphorus homeostasis has come from the development of knockout mice for each gene and the identification of naturally occurring mutations that can lead to phosphate-handling dysfunction in humans. These studies have shown that loss-of-function mutations in these 2 genes lead to overlapping hypophosphatemic syndromes that are associated with hypercalciuria and metabolic bone disease. Hereditary hypophosphatemic rickets with hypercalciuria (HHRH; OMIM # 241530) is an uncommon metabolic bone disorder caused by biallelic loss-of-function mutations in SLC34A3 encoding NPT2C (10), with an estimated prevalence of 1:250 000 (11, 12). The loss of NPT2C transporters leads to renal phosphate wasting and hypophosphatemia with appropriately suppressed serum levels of FGF23 that result in elevated serum concentrations of 1,25(OH)₂D, with consequent low PTH and hypercalciuria (10). Heterozygous mutations in NPT2C cause idiopathic hypercalciuria and low bone density. By contrast, biallelic mutations in NPT2A cause idiopathic infantile hypercalcemia, whereas heterozygous mutations in SLC34A1 that reduce expression or function of NPT2A co-transporters result in a milder phenotype consisting of hypophosphatemia, renal calcification, and osteoporosis (NPHLOP1, OMIM #612286) (13).

Here we describe an unusual family in which HHRH exhibited an apparent autosomal dominant pattern of transmission due to heterozygous mutations in both SLC34A3 and SLC34A1, providing genetic evidence of the cooperative roles of NPT2A and NPT2C in regulating systemic phosphate homeostasis and implicating a novel digenic mechanism for this disorder.

Materials and Methods

Human subjects

This study was approved by the institutional review board of the Children’s Hospital of Philadelphia (CHOP). Written informed consent, and assent as appropriate, was obtained from all participants.

We recruited the proband and all available family members over 4 generations. We collected retrospective and prospective clinical, radiological, biochemical, and molecular data. The diagnosis of rickets was based on radiological criteria in younger subjects or, in older subjects, the presence of current skeletal abnormalities (e.g., genu varus or valgus deformity) with a history of bone pain and deformity during youth. Not all subjects were able to undergo all analyses.

Genetic analyses

DNA from the proband was subjected to targeted genetic analysis using a commercial NGS gene panel (Connective Tissue Gene Tests, Abnormal Mineralization Disorders Panel) that evaluated fifteen genes associated with abnormal mineralization including: ALPL, ANKH, CASR, CLCN5, CYP27B1, DMP1, ENPP1, FAH, FGF23, OCRL, PHEX, SLC34A1, SLC34A3, SLC9A3R1, and VDR. We used Sanger sequencing (14) to confirm all mutations and to genotype all other participating members of the family (primers and conditions for PCR are available upon request). In addition, the proband (III-7), her brother (III-2), and her parents (II-5 and II-4), underwent whole exome sequencing (WES) to evaluate potential consanguinity and to identify other possible defects that could explain the disproportionate short stature.

Briefly, all the raw reads were aligned to the reference human genome using the Burrows-Wheeler Aligner (BWA-Mem) (15) and single-nucleotide variants (SNVs) and small insertions/deletions (INDELs) were captured using the Genome Analysis Tool Kit (GATK) (16). The kinship coefficient was calculated between every 2 samples via KING (17) to confirm reported relationships and identify potential consanguinity. Haplotype analysis was performed using Beagle 4.0 (18) with 1000 Genomes Project phase 3 reference panel and incorporating pedigree information (i.e., the familial relationships between samples) for a 2 Mb region flanking the SLC34A3 mutation (9(GRCh37):g.140,130,625). To confirm the imputed haplotypes, we used PCR to amplify several fragments from genomic DNA that corresponded to the SLC34A3 mutation and rs28434439 (primers 5′CCACTTCTTCTTCAACCTGGC3′ and 5′GACAGGGTCTCATTCTCCGT3′) and subcloned fragments into pGEM-T vector (Promega Corp, Madison, WI). We then transformed DH5α Escherichia coli with the plasmids, extracted DNA from colonies, and sequenced DNA from 40 unique clones for each subject analyzed using standard Sanger techniques.

Biochemical analyses

All biochemical analyses were performed using standard clinical assays. Serum concentrations of 25-hydroxyvitamin D (25(OH)D) were measured by LC/MS-MS (CHOP) and serum concentrations of 1,25(OH)₂D were measured using the DiaSorin LIAISON XL quantitative chemiluminescent immunoassay (ARUP Laboratories). Serum intact PTH levels were determined by an immunoassay performed on a Siemens IMMULITE 2000 and FGF23 levels were measured by Mayo Clinic Laboratories using an immunoassay that detects c-terminal fragments and intact FGF23. Tubular maximum phosphate reabsorption per GFR (TmP/GFR) was calculated using the Walton and Bijvoet nomogram (19).

Results

Clinical phenotype

The proband (Fig. 1, III-7) was referred to CHOP at age 11 years 11 months for evaluation of rickets, short stature, and hypercalciuria with bilateral renal stone disease. She had been born with a normal length but had developed short stature (< 1st percentile) by age 5 years, and she received a 1-year course of growth hormone with no improvement in height velocity. She was subsequently diagnosed with hypophosphatemic rickets at age 8 years and initially treated with calcitriol plus sodium phosphate. At presentation to CHOP her height was 132.6 cm (Z = −2.46), weight: 43.2 kg (57 percentile, Z = 0.18), BMI was 24.57 kg/m² (94 percentile). Her armspan was 131 cm and her upper segment was 68 cm with a lower segment of 64 cm (US/LS 1.06, normal less than 0.92). Menarche was at age 11 years, and at age 14 years her height Z-score was −3.92. She has an older sister (III-6, see Fig. 1 and Table 1) and an older brother (III-5) with similar histories of rickets, renal stones, and short stature. All 3 siblings have visibly disproportionate short stature with apparently normal-sized trunks. Another older brother (III-2) has short stature but no skeletal deformity. Their mother (II-5) has markedly bowed legs and was short, and had a history of bone pain as a child. Their father (II-4) has a history of renal stones and short stature, but no skeletal deformities to suggest a history of rickets. Various other relatives were also short. Renal ultrasounds or other imaging studies were not performed on all tested subjects. There were no histories of dental abscesses, hearing impairments, or fragility fractures in affected members of the family.

Figure 1.

Four-generation family pedigree.

The phenotype and genotype is indicated for each individual, SLC34A3 exon 13, c.1561dupC; p.L521Pfs*72, SLC34A1 exon 13, c.1535 G>A; p.R512H. WT denotes wild-type allele; NT, not tested. Subject II-1 was still-born (SB) at 38 weeks while subject II-2 died at 3 years of uncertain cause; neither subject was tested. The arrow indicates the proband, circles denote females and squares denote males.

Table 1.

Molecular, Clinical, and Biochemical Characteristics of Subjects

	III-7	III-6	III-2	II-4	II-5	I-2	Reference range
SLC34A1 Mutation	Present	Present	Absent	Absent	Present	Absent
SLC34A3 Mutation1	Present (M)	Present (M)	Present (M)	Present	Present	Present
SLC34A3 haplotype2	AB	AB	AB	AB	BC
Age (yrs)/Gender	14/F	20/F	31/M	48/M	52/F	68/F
Height (Z)	−3.92	−4.18	−2.73	−3.12	−2.88	-2.85
Serum Phosphorus, mg/dL3	2.1 (3.6–6.0)	2.9 (3.1–5.5)	2.3	3.3	3.5	3.2	2.7–4.4
Serum Calcium, mg/dL	9.6	9.3	10	10.3	9.7	10.1	8.5–10.5
Alk phos, U/L4	178 (60–280)	886	117	86	90	108	40–150
25(OH)D, ng/mL	14.9	8	19.6	15.1	11.1	21.6	20–80
1,25(OH)2D, pg/mL	198	198	171	94.3	106	68.3	36–96
PTH, pg/mL	5.8	23	12.3	<3.0	24.7	43.4	15–65
FGF23, RU/mL	57	62	<50	<50	57	99	< 230 age 3 mos- 17 yrs; <180 over 17 years
TmP/GFR, mg/dL5	1.75 (2.9–6.5)	2.9 (2.9–6.5)	1.98 (2.5–3.5)	2.63 (2.2–3.4)	3.13 (2.2–3.6)	2.87 (2–3.4)
uCa/Cr, mg/mg	0.28	0.17	0.15	0.12	0.16	0.14	<0.22

¹ M refers to inheritance of mutant SCL34A3 allele from mother (II-5).

² Based on haplotype analysis (please refer to Fig. 3), both parents (II-5 and II-4) have same mutant slc34a3 allele (haplotype b) but different wild-type alleles (A or C); father (II-4) is AB and mother (II-5) is bc.

³ Given in parentheses are age- and sex-dependent reference values ( (46).

⁴ Given in parentheses are age- and sex-dependent reference values (47).

⁵ TmP/GFR = maximum tubular phosphate reabsorption per glomerular filtration rate; in parentheses are age- and sex-dependent reference values (48).

Genetic analyses

We identified heterozygous mutations in 2 different genes in the proband (Fig. 2): a nucleotide duplication in exon 13 of SLC34A3 (c.1561dupC; p.Leu521Profs*72) that results in premature termination 72 amino acids downstream and is predicted to be pathogenic, and a transition in exon 13 of SLC34A1 (c.1535G>A; p.Arg512His) that results in replacement of a highly conserved arginine (52/53 species, to nematode; GenBank accession KRZ90293.1) by histidine. Review of publicly available databases indicates that the SLC34A3 mutation (c.1561dupC; p.Leu521Profs*72) (https://www.ncbi.nlm.nih.gov/clinvar/RCV000522409.1/) is estimated to be present in 0.01% of the population, but is not present as a homozygous mutation (20). In addition, all tested subjects were homozygous for the SLC34A3 major (T) allele at c.1538 (p.Val513), except subject IV-1, who was heterozygous (p.Glu513Val) (Fig. 1). This is considered a benign polymorphism.

Figure 2.

Sequencing electropherograms.

Sequence electropherograms demonstrating a missense mutation in SLC34A1 (exon 13, c.1535G>A; p.R512H) and a frameshift mutation in SLC34A3 (exon 13, Leu521Profs*72). Homozygous wild-type sequences are shown above, and heterozygous mutations are shown below, and indicated with arrows.

In silico analysis of the SLC34A1 p.Arg512His mutation reveals this substitution to be disease causing, probably damaging, and deleterious by MutationTaster, PolyPhen2, and SIFT algorithms, respectively. Variant c.1535G>A (p.Arg512His) has not been linked to disease previously and has a very low allele frequency (8.49e^-5; single nucleotide variant: 5-176824902-G-A [RCh37]) in the gnomAD database, and is not found in the homozygous state (21). A change in the same codon that results in a p.Arg512Cys substitution had been previously reported as a SLC34A1 mutation associated with autosomal recessive nephrolithiasis/nephrocalcinosis (22, 23). Functional characterization of the p.Arg512Cys mutation in Xenopus laevis oocytes had shown very little expression of the mutant NPT2A protein, which failed to reach the plasma membrane (23). The mutation showed significantly reduced phosphate transport in oocytes, and there was no evidence of dominant suppressor effect when the mutant protein was co-expressed with wild-type NPT2A (23).

The proband’s sister (III-6) and mother (II-5) both carried the same 2 heterozygous mutations in SLC34A1 and SLC34A3, while her brother (III-2), father (II-4) and paternal grandmother (I-2) carried only the SLC34A3 mutation. WES of the proband (III-7), her brother (III-2), and her parents (II-5 and II-4) further confirmed these genotypes, and indicated that the mother and father are not related based on kinship calculations. Haplotype analysis indicated that both parents share a haplotype around the mutant SLC34A3 gene as determined using the markers shown in Fig. 3. Because the SLC34A3 mutation is relatively rare, it is most likely that the shared haplotype carries the mutation. To confirm this hypothesis, we selected the closest informative SNP marker (rs28434439) and amplified the surrounding region including the SLC34A3 mutation. The amplified DNA fragments were subsequently subcloned and individually sequenced to obtain clearer haplotype information.

Figure 3.

Haplotype imputation by using polymorphism markers flanking the SLC34A3 mutation. Note that rs28434439 (in bold) is the closest informative SNP to the site of the SLC34A3 mutation and it was used to design the subcloning assay. Based on the results shown in Supplemental Fig. 1 (24), both parents of the proband carried the SLC34A3 (c.1561dupC) mutation on the same haplotype (blue). However, the father and mother carried distinct wild-type alleles, shown in red and black, respectively, that allow us to assign parental origin of the mutant allele to their children.

As shown in Supplemental Figure 1 (24), we observed that the SLC34A3 mutant allele and rs28434439 alternative allele are present in DNA fragments from different cloned colonies in the proband and her father, confirming our hypothesis that the shared haplotype harbors the SLC34A3 mutation. Haplotype sequencing of the mother’s DNA also confirmed our hypothesis (see Supplemental data (24)). We therefore were able to use simple PCR and Sanger sequencing around rs28434439 to distinguish the 2 parental wild-type SLC34A3 alleles, which allowed us to determine the parental origin for the SLC34A3 mutant allele that was inherited by members of generation III (24). The proband and her sister (III-6) inherited both the SLC34A1 and the SLC34A3 mutations from their mother (II-5), while the proband’s brother (III-2) inherited only a single SLC34A3 mutant allele from their mother (II-5). The proband’s father (II-4) and paternal grandmother (I-2) carried only the common SLC34A3 mutant allele.

Additionally, WES demonstrated a heterozygous missense mutation (NM_006031.5:c.G16C; p.Glu6Gln) in exon 1 of the PCNT gene encoding pericentrin in the proband, her brother (III-2), father (II-5) and grandmother (I-2), but not in other tested relatives (data not shown). This mutation is present in some but not all PCNT transcripts, and is determined to be a variant (rs761540486) of uncertain significance in ClinVar. It is reported as an uncommon variant allele in gnomAD (frequency 1.23e^-5), with no homozygotes identified. Biallelic loss-of-function mutations in PCNT cause microcephalic osteodysplastic primordial dwarfism type II (OMIM # 605925), which is characterized by intrauterine and postnatal growth restriction with severe short stature, microcephaly, and bone dysplasia (25). To date, all patients with primordial dwarfism type II have been either homozygous (26) or compound heterozygous (27) for PCNT mutations, and short stature has not been reported among heterozygous carriers. In our kindred, several family members were heterozygous for the PCNT mutation, but this mutation did not segregate with height, as both the proband’s sister (III-6) and mother (II-5) have short stature and do not carry the PCNT mutation. We did not find any other pathogenic variants.

Biochemical evaluation

Biochemical evaluation of the kindred demonstrated no obvious differences in biochemical parameters based on SLC34A3 and SLC34A1 genotype (see Table 1). Overall, serum levels of phosphorus ranged from low to low-normal for age, and affected subjects had reduced renal phosphate reabsorption as determined by TMP/GFR, which appeared to improve with age. There was also evidence of mild hypercalciuria in some affected subjects, which may have been attenuated in several subjects by co-existing vitamin D insufficiency. Serum levels of calcium ranged from normal to high-normal; alkaline phosphatase activity was elevated within the context of clinically active rickets, and was otherwise normal. The serum concentrations of 25(OH)D ranged from low to insufficient, while serum concentrations of 1,25(OH)₂D were mildly elevated. The serum levels of intact PTH ranged from suppressed to normal, while serum levels of C-terminal FGF23 were low relative to the reduced serum phosphorus concentrations.

Discussion

The identification of pathogenic alleles in SLC34A1 and SLC34A3 has informed our understanding of the relative roles of NPT2A and NPT2C in phosphate transport in the renal proximal tubule. Here, we describe an unusual family in which heterozygous mutations in SLC34A1 and SLC34A3 caused significantly more severe disease when both mutations are present in the same individual. Each mutation by itself appears to cause a less severe phenotype. These observations suggest a gene dosage effect of mutant alleles of 2 related but distinct genes, SLC34A1 and SLC34A3, providing evidence for a digenic form of HHRH. The 2 parents in this extended kindred carried the same SLC34A3 mutation, and limited haplotype analysis strongly suggested that the 2 alleles were identical. Because the kinship calculation between both parents argues against consanguinity, we presume that the presence of these identical alleles in the 2 unrelated subjects represents identity by descent of a common founder mutation that occurred many generations ago. Therefore, the marriage of these 2 carrier parents likely reflects the effect of assortative mating (28). Perhaps even more remarkable, the proband and her affected siblings inherited both heterozygous mutations, located on separate chromosomes, from a single parent, their mother. The transmission of genetic disease in this family shows a very unusual manifestation of dominant disease transmission due to digenic inheritance, wherein 2 loci are necessary to express, or extremely modify, the severity of a phenotype.

Compared to monogenic disease inheritance, digenic inheritance does not follow the logical precepts of Mendelian segregation, and is therefore underdiagnosed due to the difficulty in verifying true digenic effects. Until recently, another barrier to diagnosing digenic disorders had been the practice of sequencing single candidate genes. However, the availability of next-generation sequencing panels that allow cost-effective analysis of multiple genes that are associated with a clinical phenotype has increased awareness of digenic inheritance in genetically heterogeneous disorders. The DIDA (Digenic Disease Database) is a curated database that currently contains 258 digenic combinations causative of 54 diseases (29), and includes endocrine and renal disorders such as Alport syndrome (30), hypogonadotropic hypogonadism (31), Gitelman syndrome (32), and pituitary stalk interruption syndrome (33).

Classical HHRH is an autosomal recessive metabolic disorder of childhood, in which loss of NPT2C-dependent phosphate transport in the renal proximal tubule leads to hypophosphatemic rickets and/or osteomalacia. Although symptomatic rickets is the typical presenting feature of HHRH, recent studies indicate that some patients may experience either a later onset or later discovery of biochemical and/or clinical abnormalities during adulthood (14, 34). Reduced bone mass, skeletal deformity, or hypercalciuria with nephrolithiasis may be important and often unrecognized features in these patients. Patients who carry heterozygous SLC34A3 mutations often have an attenuated phenotype (35, 36) that includes mild hypophosphatemia and/or elevations in 1,25(OH)₂D (34), as well as hypercalciuria and renal calcifications (11, 37). Moreover, the age of onset for nephrolithiasis may be earlier in patients who have SLC34A1 mutations compared to carriers of SLC34A3 mutations (22, 23). Hence, it is likely that recurrent renal stones and short stature in the father (II-4) are due to the heterozygous SLC34A3 mutation, whereas his similarly affected heterozygous son (III-2) (Table 1) has short stature and hypophosphatemia but has not yet manifested clinical nephrolithiasis. Unfortunately, we were unable to perform a renal ultrasound to evaluate him, or several other family members such as the mother (II-5), for subclinical renal calcification. This is a limitation of our study.

NPT2A, encoded by SLC34A1, also plays an important role in renal tubular phosphate reabsorption and mineral metabolism. Biallelic mutations in SLC34A1 are a common cause of idiopathic infantile hypercalcemia (13), a more severe disorder than HHRH in very young patients. Remarkably, patients with idiopathic infantile hypercalcemia experience a gradual amelioration of hypercalcemia as they mature, which suggests that, as in the mouse, NPT2A may play a reduced role in phosphate homeostasis in older individuals. By contrast, the clinical implications of heterozygous variants in SLC34A1 are less clear (23, 38, 39). Nephrolithiasis/osteoporosis hypophosphatemic 1 (NPHLOP1) is associated with monoallelic SLC34A1 variants (40), and a growing number of reports show that heterozygous SLC34A1 mutations are present in some adult patients with recurrent renal stones (23, 41). Similarly, NPT2A also appears to be very important in mice, and constitutes some 70% to 80% of cotransport activity (42). Homozygous Slc34a1^−/− knockout mice demonstrate increased urinary phosphate levels, hypophosphatemia, elevated serum concentrations of 1,25(OH)₂D, renal calcification, and hypercalcemia (38, 43, 44). Young mice also have skeletal abnormalities that include poorly developed trabecular bones and retarded secondary ossification (44). Heterozygous mice are healthy but manifest a mild biochemical phenotype with normal serum levels of phosphorus, but evidence of phosphaturia and elevated serum concentrations of 1,25(OH)₂D (44).

The patients we describe in this report demonstrate phenotypes that reflect a gene-dose effect of mutant alleles of SLC34A and SLC34C. Being heterozygous for a mutation in both SLC34A1 and SLC34A3 resulted in hypophosphatemic rickets and renal stones, whereas carrying just the SLC34A3 mutation resulted in only renal stones and hypophosphatemia. This effect is similar to that observed in mice with homozygous ablation of both Slc34a1 and Slc34a3 (i.e., double knockout), which results in severe hypophosphatemia, hypercalciuria, and rickets that is more pronounced than in mice that are homozygous for ablation of either single gene (45). Moreover, based on our analysis of this family and a review of the literature, inheriting 2 heterozygous mutations in SLC34A1 and SLC34A3 results in a phenotype that is more severe than that caused by only 1 heterozygous mutation, but less severe than biallelic mutations in either gene. Our findings support the premise that NPT2A and NPT2C have synergistic and nonredundant effects on renal phosphate transport and play independent roles in the regulation of phosphate homeostasis.

Finally, we note that there appeared to be an attenuation of renal phosphate wasting with advancing age in affected members of this family. This may be more reflective of the decreasing importance of NPT2A for phosphate homeostasis with age, as both humans (13) and mice (44, 45) that are homozygous for SLC34A1 mutations show amelioration of the phenotype as they get older.

In summary, this is the first case of digenic inheritance of heterozygous mutations in SLC34A3 and SLC34A1 resulting in HHRH. The unusual phenotypes and pattern of inheritance in this family emphasize the importance of considering polygenic mutations as a basis of disease, and advocate for utilization of diagnostic multigene panels that allow interrogation of diverse proteins related to a specific physiological pathway or pathogenic phenotype.

Glossary

Abbreviations

1,25(OH)2D

1,25-dihydroxyvitamin D

25(OH)D

25-hydroxyvitamin D

CHOP

Children’s Hospital of Philadelphia

FGF23

fibroblast growth factor 23

HHRH

hereditary hypophosphatemic rickets with hypercalciuria

NHERF1

Na/H exchange regulatory factor 1

PTH

parathyroid hormone

TmP/GFR

tubular maximum tubular phosphate reabsorption per glomerular filtration rate

WES

whole exome sequencing

Additional Information

Disclosure Summary: The authors have nothing to disclose. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Data Availability: All data generated or analyzed during this study are included in this published article or in the data repositories listed in References.