Role of sodium‐dependent Pi transporter/Npt2c on Pi homeostasis in klotho knockout mice different properties between juvenile and adult stages

Abstract SLC34A3/NPT2c/NaPi‐2c/Npt2c is a growth‐related NaPi cotransporter that mediates the uptake of renal sodium‐dependent phosphate (Pi). Mutation of human NPT2c causes hereditary hypophosphatemic rickets with hypercalciuria. Mice with Npt2c knockout, however, exhibit normal Pi metabolism. To investigate the role of Npt2c in Pi homeostasis, we generated α‐klotho−/−/Npt2c−/− (KL2cDKO) mice and analyzed Pi homeostasis. α‐Klotho−/− (KLKO) mice exhibit hyperphosphatemia and markedly increased kidney Npt2c protein levels. Genetic disruption of Npt2c extended the lifespan of KLKO mice similar to that of α‐Klotho−/−/Npt2a−/− mice. Adult KL2cDKO mice had hyperphosphatemia, but analysis of Pi metabolism revealed significantly decreased intestinal and renal Pi (re)absorption compared with KLKO mice. The 1,25‐dihydroxy vitamin D3 concentration was not reduced in KL2cDKO mice compared with that in KLKO mice. The KL2cDKO mice had less severe soft tissue and vascular calcification compared with KLKO mice. Juvenile KL2cDKO mice had significantly reduced plasma Pi levels, but Pi metabolism was not changed. In Npt2cKO mice, plasma Pi levels began to decrease around the age of 15 days and significant hypophosphatemia developed within 21 days. The findings of the present study suggest that Npt2c contributes to regulating plasma Pi levels in the juvenile stage and affects Pi retention in the soft and vascular tissues in KLKO mice.

In the present study, we investigated the role of Npt2c in KLKO mice. Our findings revealed that although Npt2c does not affect the plasma Pi concentration, it has the same effects as Npt2a to suppress calcification and increase the lifespan in KLKO mice.

| Metabolic cages to collect urine and fecal samples
The mice were individually placed in metabolic cages at 10:00 a.m. for quantitative urine and fecal collection for 24 hr with free access to food and water. Fecal samples were ashed according to a modified protocol (Ikuta et al., 2019;Ikuta et al., 2018). The fecal samples were collected and placed in beakers and allowed to dry at 110°C for no more than 24 hr. The samples were then ashed at 250°C for 3 hr and at 550°C for 24 hr in a muffle furnace. The samples were cooled, weighed, and digested in HCl with heat, and the sample volume was standardized to 5 ml.

| RNA extraction and cDNA synthesis
Total RNA was extracted from mouse tissues using ISOGEN (Wako) according to the manufacturer's instructions. After treatment with DNase (Invitrogen), cDNA was synthesized with or without the Moloney murine leukemia virus, reverse transcriptase (Invitrogen), and oligo(dT)12-18 primer. Specific primers for NaPi transporters, Cyp27A1, Cyp24A1, IL-6, IL-1β, and GAPDH were used for the PCR reactions. The PCR primer sequences are shown in Table  2. To check for genomic DNA contamination, a reverse transcriptase negative control experiment was performed (data not shown).

| Protein sample purification
Brush border membrane vesicles (BBMVs) were prepared from kidney and intestine using the Ca 2+ precipitation method, and used for immunoblotting and Pi transport analyses, as described previously (Furutani et al., 2013;Ikuta et al., 2018;Schlingmann et al., 2016).

| Transport assay
Transport of 32 P into BBMVs was measured by the rapid filtration technique as described previously (Ikuta et al., 2018;Segawa et al., 2004). The transport rate of Pi into the kidney BBMVs was determined at 30, 60 and 120 s at 25°C with an inward gradient of 100 mM NaCl or 100 mM KCl and 0.1 mM KH 2 PO 4 (pH 7.5). The Na + -dependent Pi transport activity (activity rate in the presence of Na + in the absence of Na + ) is shown. All measurements were performed in triplicate.
Intestinal absorption was assessed on the basis of the 32 P blood level after gavage of a test solution using a previously described protocol with modifications (Matsuo et al., 2005;Van Cromphaut et al., 2001). The test solution (pH7.4) contained 128 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgSO 4 , and 4 mM KH 2 PO 4 (80 μCi/ml). For the study, 5 μl of the test solution per gram body weight was administered by gavage. Blood samples were obtained at the indicated time-points and analyzed by liquid scintillation counting.

| Immunoblotting
Protein samples were heated at 95°C for 5 min in sample buffer in the presence of 2-mercaptoethanol and subjected to 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The separated proteins were transferred by electrophoresis to Immobilon-P polyvinylidene difluoride (Millipore) and treated with diluted antibodies. Signals were detected using Immobilon Western (Millipore).

| Histologic analysis
Mouse tissues were fixed with 4% paraformaldehyde overnight at 4°C and embedded in paraffin. Serial sections (5 µm thick) of several tissues were mounted on MAS-coated slides (Matsunami Glass IND, Ltd.). The sections were treated for hematoxylin and von Kossa staining prior to light microscopic observations. The von Kossa staining for mineral deposits was performed by applying 5% silver nitrate to the sections and exposing them to bright light for 30 min (Schlingmann et al., 2016). The sections were slightly counterstained with hematoxylin.

| Bone analysis
The femurs of all groups were fixed with 4% paraformaldehyde overnight at 4°C, decalcified for 4 weeks with 10% EDTA, and then, embedded into paraffin for immunohistochemical examinations. For von Kossa staining, tibiae were immersed in a mixture containing 2% paraformaldehyde and 2.5% glutaraldehyde diluted in a 0.067 M cacodylate buffer (pH 7.4) and post-fixed with 1% osmium tetraoxide in a 0.067 M cacodylate buffer for 4 hr at 4°C. After post-fixation, the tibiae were embedded in epoxy resin (Epon 812, Taab, Berkshire, UK) and sliced at a 500 nm thickness using by an ultramicrotome. Immunohistochemical analyses of mouse bone sections were performed as described previously. Briefly, the sections were immersed into 0.3% H 2 O 2 in methanol for 30 min to block endogenous peroxidase. To reduce nonspecific binding, 1% bovine serum albumin (BSA; Serologicals Proteins Inc.) in PBS (1% BSA-PBS) was applied to the sections for 20 min. The sections were then incubated with rabbit polyclonal antisera against tissue-nonspecific alkaline phosphatase (ALP) (Oda et al., 1999). We rapidly purified soluble forms of glycosylphosphatidylinositol-anchored proteins using human tissue-nonspecific ALP, or rabbit polyclonal anti-dentin matrix protein-1 (DMP-1; Takara Bio) with 1% BSA-PBS at room temperature for 2 hr as previously described, with some modifications (Oda et al., 1999). The thus treated sections were incubated with HRP-conjugated anti-rabbit F I G U R E 1 Physiologic analysis, renal phosphate transport, and transporter expression in KLKO mice. (a) Serum fibroblast growth factor (FGF)23, (b) plasma 1,25(OH) 2 D 3 , (c) plasma parathyroid hormone (PTH), (d) blood ionized Ca, (e) plasma Pi, (f) urinary Ca, and (g) urinary Pi excretion. Male mice at 8-weeks of age (n = 5-20) were used. Values are mean ± SE. *p < .05, **p < .01. Metabolic cages were used for 24-hr collection of urine from mice. (h) Renal Na + -dependent and (i) -independent Pi transport activity in renal BBMVs isolated from the kidneys of 8-week-old WT and KLKO mice. Values are mean ± SE. *p < .05. (j) Western blot analysis of renal BBMVs isolated from the kidneys of 8-weekold WT and KLKO mice (n = 3-5). Each lane was loaded with 20 μg of BBMVs. Actin was used as an internal control. Relative intensity of Npt2a and Npt2c expression in WT mice was defined as 1.0. Values are mean relative intensity ± SE. **p < .01 versus WT mice. (k) Immunofluorescence staining of DAPI (blue), villin (red), and Npt2a or Npt2c (green) in kidney sections of 8-week-old WT and KLKO mice. Sections were prepared from kidneys embedded in OCT compound and frozen. Scale bar; 100 μm antibody (Chemicon International, Temecula) for l hr, and thereafter, the immunoreactivities were visualized by using diaminobenzidine tetrahydrochloride as a chromogen.
For double detection of ALP and tartrate-resistant acid phosphatase (TRAP), the sections immunostained for ALP were incubated with a mixture of 8 mg of naphthol AS-BI

| Statistical analysis
Data are expressed as means ± SE. Differences among multiple groups were analyzed by ANOVA. The significance of differences between the two experimental groups was established by ANOVA followed by Student's t-test. A p value of less than .05 was considered significant.

| Physiologic analysis, renal phosphate transport, and transporter expression in KLKO mice
The phenotypes of the KLKO mice were confirmed at 8 weeks of age ( Figure 1). Compared with Klotho +/+ (WT) mice, KLKO mice exhibited high plasma concentrations of FGF23 and 1,25(OH) 2 D 3 , hyperphosphatemia, and hypercalciuria, but not ionized Ca or PTH, as described previously ( Figure  1a-f) (Ohnishi et al., 2009a). Urinary Pi excretion was significantly lower in KLKO mice than in WT mice ( Figure 1g). Renal Na + -dependent Pi transport activities were significantly higher in KLKO mice than in WT mice ( Figure  1h). Na + -independent Pi transport activities did not differ between WT and KLKO mice (Figure 1i). Npt2a protein expression levels did not differ significantly between WT and KLKO mice (Figure 1j). In contrast, Npt2c protein expression levels were significantly higher in KLKO mice than in WT mice (Figure 1j). Npt2a and Npt2c protein expression was confirmed by immunofluorescence staining (Figure  1k). Npt2a and Npt2c mRNA levels were not significantly different between WT and KLKO mice ( Figure S1).

KL2cDKO mice
Feeding of a low Pi diet can rescue the kl/kl mouse phenotype (Morishita et al., 2001; Segawa et al., 2007). In the present study, we confirmed the effects of a low Pi diet on KLKO mice ( Figure S2). KLKO mice were fed a low Pi diet from 8 weeks of age. A low Pi diet increased the body weight and extended the lifespan of KLKO mice, similar to that of kl/kl mice ( Figure S2a). After 8 days, a low Pi diet significantly decreased plasma Pi levels in KLKO mice compared with baseline levels before starting the test diet ( Figure S2b).
Plasma FGF23, 1,25(OH) 2 D 3 , and Pi levels, and urinary Pi excretion levels of the mice at 8 weeks of age are shown in Figure 2c-f. The phenotypes of the KL2aDKO mice were consistent with the previous description (Ohnishi et al., 2009a). Plasma FGF23 levels were also high in KL2cDKO mice, similar to KLKO and KL2aDKO mice (Figure 2c). KLKO, KL2aDKO, Npt2aKO, and KL2cDKO mice had higher levels of 1,25(OH) 2 D 3 than WT mice (Figure 2d). Real-time PCR of renal 1αOHase and 24OHase mRNA was performed ( Figure  S3). The 1αOHase mRNA levels were significantly higher in KLKO, KL2aDKO, Npt2aKO, and KL2cDKO mice than in WT mice ( Figure S3). In contrast, 24OHase mRNA levels were lower in KLKO, KL2aDKO, Npt2aKO, KL2cDKO, and Npt2cKO mice than in WT mice ( Figure S3b). KL2aDKO mice had significantly lower plasma Pi levels than KLKO mice, as described previously (Figure 2e)  In the present study, we measured urinary Pi excretion using metabolic cages. KL2aDKO mice had much higher urinary Pi excretion levels compared with the other groups of mice (Figure 2f). Urinary Pi excretion levels were significantly higher in KL2cDKO mice than in KLKO mice ( Figure  2f). Hypercalcemia and hypercalciuria were also more severe F I G U R E 2 Characteristics of klotho/Npt2c double-KO (KL2cDKO) mice at 8 weeks of age.  Figure S3c and d). KL2cDKO mice did not exhibit hypercalcemia, but had higher levels of urinary Ca excretion compared with WT mice (Figure S3c and d). Urinary Ca excretion levels were not significantly different between KLKO and KL2cDKO mice ( Figure S3c and d).

| Npt2c plays an important role in KLKO mice during the juvenile period
Because Npt2c is highly expressed during the juvenile period, we previously reported that Npt2c is a growth-related Pi transporter (Ohkido et al., 2003;Segawa et al., 2002). KLKO mice  at 5 weeks of age showed hyperphosphatemia and hypercalciuria, but not hyperphosphaturia or hypercalcemia (Figure 4a and b, and Figure S4a and b). The Npt2a protein expression levels were significantly lower in 5-week-old KLKO mice compared with WT mice (Figure 4c). In contrast, the Npt2c protein expression levels were significantly higher in 5-week-old KLKO  (Figure 4c). Furthermore, Npt2a and Npt2c protein expression in 5-week-old KLKO mice was confirmed by immunofluorescence staining (Figure 4d). Npt2a mRNA levels were significantly lower in KLKO mice than in WT mice at 5 weeks of age ( Figure S4c). Npt2c mRNA levels, however, were not different between WT and KLKO mice at 5 weeks of age ( Figure S4c). Disruption of Npt2c in KLKO mice at 5 weeks of age significantly decreased the plasma Pi levels compared with KLKO mice, but plasma Pi levels were significantly higher in KL2cDKO mice than in WT and Npt2cKO mice ( Figure  4e). Urinary Pi excretion did not differ significantly among groups (Figure 4f). Furthermore, KL2aDKO and Npt2aKO mice exhibited hypercalcemia and hypercalciuria compared with the other mouse groups (Figure S4d and e). KL2cDKO mice had slight, but significant hypercalcemia compared with WT and KLKO mice ( Figure S4d). Furthermore, KL2cDKO mice had hypercalciuria compared with WT mice, but were not different from KLKO mice ( Figure S4e). KL2aDKO and KL2cDKO mice at 5 weeks of age had higher 1,25(OH) 2 D 3 levels than WT mice (Figure 4g). Furthermore, the 1,25(OH)2D3 levels were significantly suppressed in KL2aDKO mice, but not in KL2cDKO mice compared with KLKO mice (Figure 4g).
Real-time PCR of renal 1αOHase and 24OHase mRNA was performed ( Figure S4f and g). The 1αOHase mRNA levels tended to be higher in KLKO, KL2aDKO, Npt2aKO, and KL2cDKO mice than in WT mice ( Figure S4f). In contrast, 24OHase mRNA levels tended to be lower in KLKO, KL2aDKO, Npt2aKO, KL2cDKO, and Npt2cKO mice than in WT mice ( Figure S4g). Real-time PCR on inflammation markers, interleukin (IL)-6 and IL-1β, are shown in Figure 4h and i. Disruption of Npt2c in KLKO mice significantly decreased the IL-6 and IL-1β mRNA levels to those in KLKO mice (Figure 4f and g).

| Bone histochemical analysis
Bone histochemical analysis was performed in 5-week-old mice ( Figure 5). The bone volume of the femoral metaphyses in Npt2aKO and KL2aDKO mice seemed to be markedly reduced, while those in KLKO, Npt2cKO, and KL2cDKO mice appeared to be similar or slightly decreased compared with that in WT mice ( Figure 5A-L).
Von Kossa staining to evaluate the metaphyseal mineralization in Npt2aKO and KL2aDKO mice revealed markedly reduced metaphyseal trabeculae in both groups ( Figure 5O, P, c, d). Npt2aKO mice exhibited the broad areas of unmineralized bone matrix in the metaphyses, while KL2aDKO mice showed only slightly unmineralized bone matrix in the corresponding areas, indicating that bone mineralization of KL2aDKO mice seemed to have recovered to some extent by Npt2a deficiency ( Figure 5O, P, c, d). Unlike Npt2aKO mice, mineralization in metaphyseal trabeculae was similar between Npt2cKO mice and WT mice ( Figure 5M, R, a, f). Despite a similar metaphyseal bone volume as in WT, KLKO and KL2cDKO mice exhibited very large of unmineralized bone matrix compared with the WT mice ( Figure 5M, N, Q, a, b, e).
Histologically assessment by double immunostaining for ALP/TRAP revealed similar numbers of ALP-positive osteoblasts and TRAP-reactive osteoclasts in all the groups (Figure 5g-l). Interestingly, an intense accumulation of DMP-1 immunoreactivity was observed in the osteocyte lacunae of KLKO and KL2cDKO mice (Figure 5n and q). In contrast, Npt2aKO, KL2aDKO, and Npt2cKO mice showed a distribution pattern of DMP-1 immunoreactivity in the osteocytic lacunar-canalicular system that was similar to that of WT mice (Figure 5m, o, p, r).

KL2cDKO mice
Intestinal Pi absorption in KLKO mice has not been studied in detail. KLKO mice may have increased intestinal Pi absorption due to the high plasma vitamin D levels. The high vitamin D levels may also contribute to hyperphosphatemia in KLKO mice. Fecal Pi excretion was measured in juvenile (5-weekold) and adult KLKO (8-week-old) mice using metabolic cages (Figure 6a-d). In the juvenile period, KLKO, KL2cDKO, and Npt2cKO mice had higher food intake (mg/g body weight) than WT mice, but there was no difference in fecal Pi excretion (Figure 6a and b). In adults, food intake (mg/g body weight) did not differ between groups, but fecal Pi excretion was significantly lower in KLKO mice compared with WT mice (Figure  6c and d). Disruption of Npt2c in 8-week-old KLKO mice, however, recovered the fecal Pi excretion level compared with 8-week-old KLKO mice (Figure 6d). Furthermore, to confirm the intestinal Pi absorption, 32 P was administered orally and the Pi absorption rate was measured at 8 weeks of age (Figure 6e). The Pi absorption rates ( 32 P transfer from intestine to the blood) were significantly increased in KLKO mice compared with WT mice (Figure 6e). Disruption of Npt2c in 8-week-old KLKO mice significantly decreased the Pi absorption rate compared with KLKO mice (Figure 6e).
Next, we investigated the reasons for the suppression of fecal Pi excretion in KL2cDKO mice compared with KLKO mice. Npt2c is mainly expressed in the kidney (Ohkido et al., 2003;Segawa et al., 2002). In the present study, we evaluated Npt2c in the mouse intestine. Npt2c mRNA expression was detected in the proximal and distal mouse intestine ( Figure  6f). Furthermore, intestinal Npt2c mRNA expression levels were significantly increased in 8-week-old KLKO mice, but not in 5-week-old KLKO mice, compared with WT mice (Figure 6g).
Intestinal Pi transporter mRNA expression is shown in Figure 7. Intestinal PiT1 and PiT2 mRNA levels were significantly increased in KLKO mice compared with WT mice (Figure 7a and b). Intestinal Npt2b mRNA levels were not different in KLKO mice compared with WT mice ( Figure  7c). Disruption of Npt2c in KLKO mice suppressed the induction of PiT1 and PiT2 mRNA (Figure 7a and b).

| Role of Npt2c in controlling blood Pi levels in juvenile mice
Finally, we confirmed renal Npt2a and Npt2c protein expression around the juvenile period ( Figure S5a and b). As described previously, Npt2a protein expression gradually increased with growth ( Figure S5a and b). In contrast, Npt2c protein expression rapidly increased at 28 days of age (28D; Figure S5a and b). PTH and FGF23 levels were measured in mice at 15, 21, 28, and 60 days ( Figure S5c and d). Plasma PTH levels in mice at 21, 28, and 60 days were significantly lower than those at 15 days ( Figure S5c). In contrast, FGF23 levels in mice were lower at 21 and 28 days than at 15 and 60 days ( Figure S5d). Next, we measured plasma Pi levels in juvenile Npt2cKO mice (Figure 8). In juvenile Npt2cKO mice, plasma Pi levels began to decrease around 15 days of age and showed significant hypophosphatemia at 21 days ( Figure 8).
The lifespan of both KL2aDKO and KL2cDKO mice was significantly extended compared with that of KLKO mice. In KL2aDKO mice, despite the increased plasma 1,25(OH) 2 D 3 levels, we predicted that the increased renal Metabolic cages were used for measurement of 24-hr food intake (mg/g body weight [BW]), and collection of feces from mice (n = 5-9). Values are mean ± SE. # p < .05, #′ p < .01 versus WT mice, a p < .05 versus KLKO mice. (e) Intestinal phosphate absorption assays in 8-week-old WT, KLKO, and KL2cDKO mice. Change in the blood Pi at 4 and 60 min after administration of the 32 P test solution. Relative rate of 32 P absorption in WT mice was defined as 100%. Values are mean ± SE, #′ p < .01, a' p < .01, n = 3-4. (f) Npt2c mRNA in the kidney, proximal and distal intestine of WT mice. (g) Real-time PCR analysis for Npt2c mRNA in the proximal and distal intestine of male WT and KLKO mice (n = 5-9). GAPDH was used as an internal control. Values are mean ± SE. **p < .01 versus WT mice soft-tissue and vascular calcification, even in the presence of extremely high serum Ca and 1,25(OH) 2 D 3 levels compared with KLKO mice. Survival curves indicated that the lifespan of KL2cDKO mice was extended similarly to that of KL2aDKO mice compared with KLKO mice. The plasma Pi levels in adult KL2cDKO mice (8-week-old) did not decrease, however, and plasma 1,25(OH) 2 D 3 levels also remained high. Renal Pi excretion was significantly increased in KL2cDKO mice compared with KLKO mice. In addition, adult (8-weekold) KLKO mice had significantly decreased fecal Pi excretion, and increased the levels of intestinal PiT1, PiT2, and Npt2c mRNA compared with WT mice. In vivo 32 P oral administration, to KL2cDKO mice significantly suppressed intestinal Pi absorption compared with that in the KLKO mice. Thus, in KL2cDKO mice, although there was no change in the plasma Pi concentration, the body Pi load was reduced by increased intestinal and renal Pi excretion compared with that in KLKO mice. Such changes are expected to result in the suppression of calcification (kidney and aorta).
On the other hand, plasma Pi levels were reduced in 5-week-old KL2cDKO mice compared with KLKO mice. Renal Npt2c is specifically increased during the juvenile F I G U R E 7 Intestinal Pi transporter in 8-week-old KL2cDKO mice. (a) PiT1, (b) PiT2, and (c) Npt2b mRNA levels in the proximal and distal intestine by real-time PCR analysis. Male mice at 8 weeks of age (n = 5-9) were used. GAPDH was used as an internal control.  period, suggesting that Npt2c function is involved in renal Pi reabsorption. Urinary Pi excretion levels, however, were not altered in 5-week-old KL2cDKO mice compared with KLKO mice. Furthermore, previous and current data revealed no change in Pi excretion in 5-week-old Npt2cKO mice. We predicted that organs other than the kidney that express Npt2c contribute to the decreased plasma Pi concentration. Deletion of Npt2c in KLKO mice did not have a clear recovery effect on bone phenotype. In addition, bone abnormalities in KLKO mice may have been caused by factors such as increased FGF23 due to klotho deficiency rather than suppression of plasma Pi levels, as described previously (Hikone et al., 2017).
Previous analyses of Npt2cKO mice (4 and 8 weeks of age) showed no clear abnormalities in Pi metabolism (Segawa, Onitsuka, Kuwahata, et al., 2009). Therefore, the role of Npt2c in the pathogenesis of HHRH due to Npt2c abnormalities is unclear in mice. We examined plasma Pi levels in Npt2cKO mice before weaning. Npt2cKO mice exhibited a significant decrease in plasma Pi levels at 21 days of age. We expect that Npt2c maintains postnatal plasma Pi levels. During the time when Npt2c is considered important, we evaluated the FGF23 and PTH concentrations, and found that FGF23 concentrations were significantly decreased. Changes in the plasma FGF23 levels may also play an important role in the induction of postnatal Npt2c protein.
The difference in the role of NaPi-2c in humans and mice is not well understood. On the basis of our findings, mouse NaPi-2c is necessary for increasing plasma Pi levels during the growth phase. In the mature period, it contributes to the absorption/excretion of Pi in the intestine/ kidney rather than maintenance of the plasma Pi concentration, suggesting that it is involved in the retention of Pi in the body. On the other hand, human NaPi-2c, like mouse NaPi-2a, is expected to be an important molecule for maintaining plasma Pi levels. Although patients with HHRH typically present in childhood with rickets and/or nephrolithiasis, patients occasionally present as adults with low bone density (Dasgupta et al., 2014;Dhir, Li, Hakonarson, & Levine, 2017). NaPi-2c may have different roles in the body depending on age.
Finally, Npt2c may contribute to maintaining plasma Pi levels for growth during the juvenile period and may also be important for Pi retention in adults by controlling intestinal and renal Pi absorption. Figure 9 summarizes the KLKO, KL2aDKO, and KL2cDKO mouse phenotypes. Analysis of mouse NaPi-2c may be useful toward understanding the role of NaPi-2c in phenotypically different HHRH patients.