SEARCH

SEARCH BY CITATION

Keywords:

  • fibroblast growth factor 23;
  • X-linked hypophosphatemia;
  • phosphatonin;
  • neutralizing antibody;
  • novel therapy

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

X-linked hypophosphatemia (XLH), characterized by renal phosphate wasting, is the most common cause of vitamin D-resistant rickets. It has been postulated that some phosphaturic factor plays a causative role in XLH and its murine homolog, the Hyp mouse. Fibroblast growth factor 23 (FGF23) is a physiological phosphaturic factor; its circulatory level is known to be high in most patients with XLH and Hyp mice, suggesting its pathophysiological role in this disease. To test this hypothesis, we treated Hyp mice with anti-FGF23 antibodies to inhibit endogenous FGF23 action. A single injection of the antibodies corrected the hypophosphatemia and inappropriately normal serum 1,25-dihydroxyvitamin D. These effects were accompanied by increased expressions of type IIa sodium-phosphate cotransporter and 25-hydroxyvitamin-D-1α-hydroxylase and a suppressed expression of 24-hydroxylase in the kidney. Repeated injections during the growth period ameliorated the rachitic bone phenotypes typically observed in Hyp mice, such as impaired longitudinal elongation, defective mineralization, and abnormal cartilage development. Thus, these results indicate that excess actions of FGF23 underlie hypophosphatemic rickets in Hyp mice and suggest a novel therapeutic potential of the FGF23 antibodies for XLH.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

X-linked hypophosphatemic rickets (XLH) is the most common form of inherited hypophosphatemic rickets and is characterized by hypophosphatemia caused by renal phosphate wasting.(1) The responsible gene for XLH is PHEX (phosphate-regulating gene with homologies to endopeptidases on the X-chromosome), which encodes a 749 amino acid type I membrane protein.(2) Various mutations in the PHEX gene, which are believed to impair the function of the PHEX protein, have been identified in XLH patients(2–4); however, the molecular mechanism by which a functional loss of this gene product leads to hypophosphatemic rickets is still uncertain. XLH has been hypothesized to be caused by putative humoral factor(s), termed phosphatonin. This idea has arisen substantially from the characterization of the Hyp mouse, a murine homolog of XLH, with a large deletion in the 3′ region of Phex gene.(5,6) In this mouse, it has been suggested that some circulatory factor(s) derived from extrarenal tissue is responsible for hypophosphatemic rickets.(7,8)

Fibroblast growth factor 23 (FGF23) was identified as a causative factor for autosomal dominant hypophosphatemic rickets/osteomalacia (ADHR) and tumor-induced osteomalacia (TIO).(9,10) Excess action of FGF23 has been shown to cause hypophosphatemic rickets/osteomalacia in studies using transgenic mice or recombinant protein,(11–15) whereas loss of its action, as occurs in Fgf23-knockout mice or patients with tumoral calcinosis, results in hyperphosphatemia, elevated 1,25-dihydroxyvitamin D (1,25D) levels, and abnormal ossification.(16–18) These observations point to an indispensable role of this molecule in the physiological regulation of phosphate and vitamin D metabolism. In addition to ADHR and TIO, it is evident that FGF23 is circulating at higher levels in XLH patients(19,20) and Hyp mice.(18,21) In Hyp mice, the serum concentration of FGF23 is 10- to 20-fold higher than that in normal mice, and its mRNA expression has been shown to be upregulated in osteocytes.(22,23) Furthermore, a recent study showed that the hypophosphatemic character of Hyp mice was reversed by crossing Hyp with Fgf23KO mice.(18) Given the evidence and its biological action, excess amount of FGF23 in XLH/Hyp is considered to play the substantial and causative role ascribed to “phosphatonin.”

In this study, we addressed this hypothetical causative role of FGF23 in Hyp mice by blocking its endogenous activity with neutralizing antibodies. We previously developed anti-FGF23 mouse monoclonal antibodies that are capable of specifically inhibiting FGF23 action in vitro and in vivo and showed that administration of antibodies into normal mice transiently caused hyperphosphatemia and high 1,25D levels.(24) By using these powerful tools, we also aimed to evaluate their therapeutic effects in Hyp mice. As a result, this study shows the potential utility of this intervention as a novel therapeutic maneuver for hypophosphatemic rickets/osteomalacia.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Antibodies

FN1 and FC1, which are mouse monoclonal antibodies (IgG1) with neutralizing activity to FGF23, were affinity-purified from the conditioned media of the hybridoma cells using Protein G Sepharose (GE Healthcare) and reconstituted in PBS. As an isotype-matched control antibody, we used anti-human thrombopoietin (TPO) mouse monoclonal antibody (IgG1) without cross-reactivity to murine TPO.(25) FN1 structurally recognizes the N-terminal portion of FGF23, whereas FC1 recognizes single region comprised of several residues in the C-terminal domain of FGF23. FN1 and FC1 can block FGF23′s ability to associate with its FGF receptor and Klotho, a co-receptor for FGF23/FGFRs, respectively.(24) When these antibodies are mixed, they showed significantly synergistic and longer-lasting effect than that by each antibody alone.(24) Therefore, we used a 1:1 mixture of FN1 and FC1 in every study. (Hereafter we refer to this mixture as FGF23Ab.)

Mice

Female Hyp/X mice were originally purchased from Jackson Laboratory (Bar Harbor, ME, USA) and bred with male C57BL/6J mice to obtain male Hyp/Y and normal littermates. All mice were fed with standard rodent chow CE-2 (Crea) containing 1.0% Ca and 1.1% P and tap water ad libitum. All animal studies were reviewed and approved by the institutional Animal Care and Use Committee at the Discovery Research Laboratories, Kirin Pharma.

Single dose study

FGF23Ab was administered subcutaneously into male Hyp mice (6–8 wk of age) and normal littermates (WT) at 4 (2 mg/kg of FN1 and 2 mg/kg of FC1 each) or 16 mg/kg (8 mg/kg of FN1 and 8 mg/kg of FC1 each) as a rapid bolus injection. As a control group, 16 mg/kg of anti-TPO antibody (control IgG1) was injected. To monitor the time course of changes in serum phosphate and calcium, blood samples were sequentially collected from the orbital cavity at 0, 1, 3, 5, 7, 11, and 14 days after injection. To evaluate fractional excretion of phosphate (FEPi), urine samples were collected from 48 to 72 h after the injection by rearing mice individually in metabolic cages (Sugiyamagen). To analyze renal expressions of type IIa sodium-phosphate cotransporter (NaPi2a) and vitamin D-metabolizing enzymes, mice were killed at 12 h after injection to obtain the kidneys. To monitor the time course of changes in 1,25D, mice were killed at 1, 4, 7, 10, and 14 days after dosing to collect whole blood from the heart.

Repeated dose study

FGF23Ab (4 or 16 mg/kg) or control IgG1 (16 mg/kg) was subcutaneously administered into male Hyp or normal littermates at 4 wk of age (day 0), followed by an additional four injections on days 7, 14, 21, and 28. Body weight and tail length were recorded weekly. Mice were killed on day 31 to collect blood samples (from the heart), tibial and femoral bones, and kidneys. To analyze the bone turnover rate, intraperitoneal injection (30 mg/kg) of tetracycline (Sigma-Aldrich) and subcutaneous injection (30 mg/kg) of calcein (Wako) were performed at 7 and 2 days before death, respectively.

Serum and urine parameters

Serum samples were prepared in MicroTainer (Becton, Dickinson and Company). Serum/urine concentrations of phosphate, calcium, and blood urea nitrogen, and alkaline phosphatase (ALP) were determined by test-Wako kits (Wako). Creatinine levels were measured by creatinine kits (Kainos Laboratory). PTH and 1,25D levels were determined by Rat Intact PTH RIA Kit (Immutopics) and 1,25D RIA Kit (TFB), respectively.

Renal expression of NaPi2a and vitamin D-metabolizing enzymes

Proximal tubular brush border membrane vesicle (BBMV) was prepared from each kidney with a previously reported method.(14) A 20-μg fraction of BBMV was subjected to Western blot using anti-NaPi2a polyclonal rat antibody raised by immunizing with the C-terminal peptide of mouse NaPi2a.(14) Total RNA was prepared from the frozen kidney using QIAGEN RNeasy (QIAGEN) and reverse-transcribed using the SuperScript III First-Strand Synthesis System (Invitrogen). Quantitative analysis of Cyp27b1 and Cyp24 mRNA was performed on a real-time PCR system (7900HT [Applied Biosystems] with QuantiTest SYBR Green PCR kit [QIAGEN]) and results were corrected by that of an internal control, Rps18. The primer-pairs were as follows-Cyp27b1: 5′-AGCTCCACTTGGCTCTTTGC-3′ and 5′-GCTAGGTACCAGGATGCCAAGAT-3′; Cyp24: 5′-ATTTGCATCTCTTCCCTTCGG-3′ and 5′-GCTCAGGTAGCACTTCAAAATGG-3′; Rps18: 5′-TTCTGGCCAACGGTCTAGACAAC-3′ and 5′-CCAGTGGTCTTGGTGTGCTGA-3′.

Immunohistochemistry

Each kidney was fixed in 10% formalin and embedded in paraffin to make 3-μm sections. Anti-NaPi2a polyclonal rat antibody was used for the detection in immunohistochemistry, and the signals were visualized by a Vectastain ABC-AP Standard Kit (Vector Laboratories) and an Alkaline Phosphatase Substrate Kit I (Vector Laboratories).

Histological analyses of bone

Soft X-ray radiograms were obtained by μFX-1000 (Fuji). The femurs were dried at 100°C for 48 h and ashed at 600°C for 24 h to determine BMC. The tibias were fixed in 70% ethanol, stained with Villanueva bone stain, embedded in methyl methacrylate, sectioned at 5 μm, and stained with Goldner stain. Using the same undecalcified sections, histomorphometric analysis was performed on secondary spongiosa, starting at 1.5 mm from the end of primary spongiosa, and at 0.15 mm from the endosteal surface. This area was split into 0.3 × 0.3-mm square fields so that the largest number of fields containing secondary spongiosa could be obtained within the defined area. Each parameter was measured in each square field, and results were averaged. Individual animal data were obtained from 8 to 23 fields.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Single injection of FGF23Ab in Hyp mice

A single subcutaneous injection of FGF23Ab (a 1:1 mixture of two anti-FGF23 antibodies; see Materials and Methods section) succeeded in increasing serum phosphate levels in Hyp mice in a dose-dependent manner (4 and 16 mg/kg; Fig. 1A). In contrast, the FGF23Ab treatment did not result in a significant change in serum calcium levels, with the exception of day 7, where a slight and transient increase was seen in the higher dose group (Fig. 1A). The FGF23Ab treatments (4 and 16 mg/kg) increased serum phosphate levels in Hyp mice by 24 h after injection and normalized them within 3 days. Serum phosphate levels increased similarly in the two dose groups, but peak levels in the higher-dose (16 mg/kg) group were higher than those of wildtype mice treated with a control antibody (control Ab, at 16 mg/kg).

thumbnail image

Figure Figure 1. The improvement of hypophosphatemia after a single injection of FGF23Ab. (A) Time course of changes in serum phosphate and calcium levels. All results represent means ± SE (n = 6). Statistical analysis was conducted for the results at each time point by Dunnett's method. ap < 0.05 and bp < 0.01 vs. WT control. cp < 0.05 and dp < 0.01 vs. Hyp control. (B) Renal fractional excretion of phosphate for 24 h. Urine samples were collected from 48 to 72 h after injection. Results represent means ± SE (n = 6). Statistical significance was evaluated by Bonferroni's method. (C) Renal NaPi2a protein expression at 12 h after the injection (n = 3). As a control, isotype-matched IgG (16 mg/kg) was injected into normal littermates (WT, control) or Hyp mice (Hyp, control). The numbers in parentheses refer to the FGF23Ab doses given (mg/kg).

Download figure to PowerPoint

The hypophosphatemia in Hyp mice is well known to result from renal phosphate wasting caused by the reduction of NaPi2a protein in the brush border membrane of renal proximal tubules.(26) We therefore examined the effects of FGF23Ab on urinary phosphate excretion and renal NaPi2a expression. Because the increases in serum phosphate levels were clearly observed within the 48- to 72-h postinjection period, we collected urine samples during this period. The FGF23Ab treatment completely normalized the excess renal phosphate excretion seen in the Hyp mice (Fig. 1B). This correction of urinary phosphate excretion was accompanied by a recovery of NaPi2a protein expression (Fig. 1C). These effects were sustained for at least 3 days after injection (data not shown).

A single injection of FGF23Ab also resulted in a significant increase in serum 1,25D levels. The highest levels of 1,25D were observed at 24 h after the injections in both the lower- and higher-dose groups (Fig. 2A). In the higher-dose group, the elevated levels of 1,25D were maintained for 7 days. These changes in serum 1,25D levels were caused by marked changes in the renal expressions of two key vitamin D-metabolizing enzymes: 25-hydroxyvitamin-D-1α-hydroxylase (1αOHase) and 24-hydroxylase (24OHase). The FGF23Ab treatment significantly increased 1αOHase (Cyp27b1) mRNA expression (48- and 82-fold of WT for the low- and high-dose groups, respectively), whereas it restored 24OHase (Cyp24) mRNA expression to normal levels (Fig. 2B).

thumbnail image

Figure Figure 2. Changes of vitamin D metabolism by the single injection of FGF23Ab. (A) Time course of changes in serum concentrations of 1,25D. Each datum represents mean ± SE (n = 6). ap < 0.01 vs. WT control and Hyp control (Dunnett). (B) Quantitative real-time PCR analysis for renal Cyp27b1 (1αOHase) and Cyp24 (24OHase) mRNA levels at 24 h after the injection (n = 3). The results were corrected for Rps18 (ribosomal protein S18) mRNA level and represent the ratio to that of WT group (mean ± SD). Statistical significance was evaluated by Bonferroni's method. ap < 0.05 vs. WT control. cp < 0.05 and dp < 0.01 vs. Hyp control.

Download figure to PowerPoint

Repeated injections of FGF23Ab in juvenile Hyp mice

Next, the therapeutic potential benefit of the FGF23Ab was examined during the growth period, particularly focusing on the skeletal development in the Hyp mice. We started to treat male Hyp mice at 4 wk of age with FGF23Ab (4 or 16 mg/kg) or control Ab (16 mg/kg). The treatments were repeated weekly for 1 mo (total of five injections). As shown in Fig. 1A, the lower dose (4 mg/kg) of FGF23Ab resulted in a transient normalization of serum phosphate levels, whereas a higher dose (16 mg/kg) of FGF23Ab enabled serum phosphate to be sustained at normal or high levels for almost 1 wk. Therefore, weekly injections of 4 and 16 mg/kg of FGF23Ab were expected to result in low-normal and normal-high levels of serum phosphate throughout the experiments, respectively, as long as the magnitude of changes remained constant with each repeated treatment.

Biochemical analyses were performed on sera collected on day 31 (i.e., 31 days after the first injection; 3 days after the final injection) and on urine samples collected from day 30 to day 31. The data are summarized in Table 1. Serum phosphate levels in the 4 and 16 mg/kg FGF23Ab-treated Hyp mice on day 31 were normal and high, respectively ( Table. 1). A dose-dependent decrease in the fractional excretion of phosphate was observed in the FGF23Ab-treated Hyp mice, which was accompanied by the evident recovery of NaPi2a protein expression in the kidney (Fig. 3). The Hyp mice treated with FGF23Ab also showed significantly elevated serum 1,25D levels compared with those of the control group. These data showed that the neutralizing effect of FGF23Ab was not attenuated over the course of the extended weekly injection period.

Table Table 1.. Blood and Urinary Parameters on Day 31
Thumbnail image of
thumbnail image

Figure Figure 3. Immunohistochemistry of renal NaPi2a protein in the kidneys on day 31 after repeated injections. Original magnification, × 200.

Download figure to PowerPoint

Serum and urinary excretion of calcium were elevated in a dose-dependent manner. However, there was no evidence of nephrocalcinosis, as judged by von Kossa staining or ash content analysis of kidneys (data not shown). In addition, neither blood urea nitrogen (BUN) nor serum creatinine levels were affected by the treatments. Although the treatments significantly increased serum phosphate levels in the higher dose group during the experimental period, serum PTH levels were depressed relatively to controls (Table 1). To reassess these effects on PTH levels, we conducted a similar study using a different colony of mice. The treatments were started at 4 wk of age, and injections were repeated weekly for 6 wk (n = 10 each). Consequently, the PTH levels were again decreased in both the 4- and 16-mg/kg groups (WT control, 20.0 ± 2.3 pg/ml; Hyp control, 41.3 ± 3.3 pg/ml; Hyp FGF23Ab [4 mg/kg], 11.0 ± 1.4 pg/ml [p < 0.01 vs. Hyp control]; Hyp FGF23Ab [16 mg/kg], 13.1 ± 0.8 pg/ml [p < 0.01 vs. Hyp control]), confirming the data that the FGF23Ab treatment suppressed serum PTH levels.

It is significant that abnormally increased serum total ALP activity in Hyp mice was corrected by the treatment (Table 1). Because this parameter did not change with a single injection of FGF23Ab (data not shown), this result is likely to represent effects of the repeated injections and suggests an improvement in bone metabolism (as discussed later).

Skeletal changes by repeated injections of FGF23Ab in juvenile Hyp mice

The FGF23Ab treatment during the growth period markedly improved the growth retardation of Hyp mice (Fig. 4). Relative to untreated controls, the treated Hyp mice exhibited an accelerated weight gain, and the final body weight was close to that of the normal littermates at the end of the study. A short tail, which is another typical characteristic of the Hyp mouse, was also ameliorated by the treatment, whereas its elongation ceased at 5 wk of age in the untreated Hyp control (Fig. 4). Because this improvement in tail length suggested a general effect on longitudinal bone elongation, we analyzed the long bones of the legs (Fig. 5). The FGF23Ab treatments indeed improved the elongation of femoral and tibial bones (femur: WT control, 14.7 ± 0.1 mm; Hyp control, 10.4 ± 0.3 mm; Hyp FGF23Ab [4 mg/kg], 12.1 ± 0.2 mm [p < 0.01 vs. Hyp control]; Hyp FGF23Ab [16 mg/kg], 12.4 ± 0.1 mm, [p < 0.01 vs. Hyp control]). In addition, the treatment alleviated the enlarged and distorted epiphyses and diaphyses typically observed in Hyp bones (Fig. 5A) and corrected the BMC (Fig. 5B).

thumbnail image

Figure Figure 4. Time courses of changes in body weight (left) and tail length (right) during repeated injections of FGF23Ab. All results represent means ± SE (n = 5–7). Statistical analysis was conducted for the results at each time point. ap < 0.05 and bp < 0.01 vs. WT control. cp < 0.05 and dp < 0.01 vs. Hyp control (Dunnett). All data of tail length of both treated and control Hyp were statistically different from those of WT control (p < 0.01, Dunnett).

Download figure to PowerPoint

thumbnail image

Figure Figure 5. Ameliorative effects of repeated injections of FGF23Ab on long bones. (A) Representative roentgenograms of femurs and tibias isolated on day 31. (B) Femoral ash weight (% of total dry tissue weight) on day 31. The result represents mean ± SE (n = 5–7). ap < 0.05 and bp < 0.01 vs. WT control. cp < 0.01 vs. Hyp control (Dunnett).

Download figure to PowerPoint

Histological changes in bone of juvenile Hyp mice treated with FGF23Ab

Consistent with the improvement in longitudinal bone length, the FGF23Ab treatment markedly improved the columnar structure of proximal metaphyseal growth plates (Fig. 6). In the control Hyp mice, the columnar structure of chondrocytes in growth plates was severely disorganized, and the number of hypertrophic chondrocytes was markedly increased, which resulted in the abnormal thickening of the growth plate structure (Fig. 6A). In contrast, these abnormalities were drastically improved with treatment. Improvement was observed substantially in the hypertrophic zone. Indeed, the number of hypertrophic chondrocytes of treated mice was almost same as that of normal control mice (WT control, 2.5 ± 0.3 cells; Hyp control, 8.3 ± 1.4 cells; Hyp FGF23Ab [4 mg/kg], 3.2 ± 0.4 cells [p < 0.01 vs. Hyp control]; Hyp FGF23Ab [16 mg/kg], 2.5 ± 0.6 cells [p < 0.01 vs. Hyp control]). This reduction in the number of hypertrophic chondrocytes was dose dependent and resulted in a decrease in the total thickness of the growth plate (Fig. 6B). In addition, the primary spongiosa in the higher-dose group was substituted by calcified bone, whereas this area in untreated Hyp mice was occupied by unmineralized osteoid (Fig. 6A).

thumbnail image

Figure Figure 6. Histological analysis of tibial metaphyseal growth plate on day 31. (A) Undecalcified sections visualized by Villanueva-Goldner stain. Mineralized bone and unmineralized osteoid are colored in green and red, respectively. (B) The quantitative measurement of total growth plate thickness. All results represent means ± SE (n = 5–7). ap < 0.05 and bp < 0.01 vs. WT control. cp < 0.01 vs. Hyp control (Dunnett).

Download figure to PowerPoint

Tibial trabecular or cortical bones were analyzed histologically. In Hyp mice, bones were poorly mineralized and occupied with abundant osteoid. The repeated injections of FGF23Ab during the growth period, however, drastically reduced the osteoid and increased the mineralized bone in both trabecular and cortical bones (Fig. 7).

thumbnail image

Figure Figure 7. Histological analysis of tibial proximal metaphysis on day 31. Undecalcified sections visualized by Villanueva-Goldner stain (green, mineralized bone; red, unmineralized osteoid). (A-D) Lower magnification (×100) of tibial proximal metaphysis. Trabeculae squared with dashed line are enlarged (E-H). (I-L) Metaphyseal cortices (magnification, ×400). Counter stain, hematoxylin.

Download figure to PowerPoint

To quantitatively analyze the changes in the treated bone, a histomorphometric measurement was performed on secondary cancellous bones at the proximal tibial metaphysis (Table 2). The FGF23Ab treatment increased trabecular number and decreased its separation. In addition, the FGF23Ab treatment largely decreased the quantity of osteoid and increased the amount of mineralized bone in the assessed area. To determine kinetic parameters of mineralization, the bones of all animals were double-labeled before death. The bones of the untreated Hyp mice showed no evidence of labeling by either marker, indicating their extremely low mineral incorporation activity. In contrast, the bones of the FGF23Ab-treated Hyp mice showed clear labeling by the first label injected (100% of the bones analyzed), as well as by the second label injected (40% and 85% for the 4- and 16-mg/kg treated groups, respectively), indicating that the mineralization process had resumed after treatment. Mineral apposition rate and bone formation rate in the higher dose group were nearly 70% of those rates seen in the normal littermates. It is noteworthy that osteoblast and osteoclast surfaces in the FGF23Ab-treated Hyp mice were higher than those in normal mice, which might have contributed to the rapid recovery in bone formation and remodeling. Thus, repeated administration of FGF23Ab both restored the structure of the growth plate and the mineralization process and triggered the resumption of bone turnover.

Table Table 2.. Histomorphometry of Secondary Cancellous Bones on day 31
Thumbnail image of

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

The inhibition of excess activity of FGF23 in Hyp mice by FGF23 antibodies resulted in dramatic improvement of most phenotypes. The FGF23Ab treatment improved the defect in the abnormal renal phosphate reabsorption, so that serum phosphate levels were elevated in Hyp mice. The inappropriate 1,25D levels, which occurred despite the hypophosphatemic condition and are considered to be one of the typical features of this disease, were also overcome by FGF23Ab treatment. The successful disappearance of Hyp characteristics by the FGF23Ab directly indicates that FGF23 plays a causative role in hypophosphatemia and inappropriate vitamin D metabolism in XLH/Hyp, confirming the accumulating evidence that FGF23 is the long-sought phosphatonin.(18,27)

Our study further indicated that direct blockade of FGF23 action can be an effective therapy for XLH. The anti-hypophosphatemic effects of the antibodies seem considerably more robust and long-lasting than those attained by oral phosphate supplementation, an approach currently used in XLH patients,(1) or previously reported drugs, such as indomethacin(28) and dipyridamole.(29) In addition, the repeated injections of FGF23Ab tended to lower the serum PTH levels. It is known that XLH patients often exhibit secondary hyperparathyroidism, caused, at least in part, by continuous oral phosphate supplementation, and excess PTH, in turn, could also contribute to the hypophosphatemia. In this regard, the decreased PTH could have contributed to the increases in serum phosphate levels to some degree in the treated Hyp mice. The overall efficacy and duration of the anti-hypophosphatemic effect observed with the FGF23Ab may thus be attributable to a combined mechanism involving a direct block of FGF23 activity, an indirect suppression of PTH secretion and stimulation of 1,25D production. In addition, a therapeutic use of FGF23Ab could potentially offer advantages over the use of oral phosphate supplementation, in as much as the risks of secondary hyperparathyroidism and phosphate-induced diarrhea would likely be reduced.(30)

Repeated FGF23Ab treatments during growth ameliorated the defects in skeletal development seen in the Hyp mice and resulted in improved body weight gain. This study was performed in mice between 4 and 8 wk of age. This experimental period was long enough to test the hypothesis that by suppressing FGF23 activity during growth, we could reverse at least some of the Hyp characteristics. Indeed, there were clear dose-dependent effects on several key histomorphometric parameters, such that pharmaceutical benefits of the FGF23Ab treatment on skeletal development were largely proved by the current 1-mo experiment. However, the low-dose group still exhibited some defects in bone turnover and mineralization., These defects could likely be improved further by initiating the treatment at an earlier age and extending for a longer duration. Our study also showed an apparently excess mineralization in metaphyseal primary cancellous bone in the treated Hyp mice. We speculate that the enhanced mineralization probably was derived from a simultaneous mineralization of a large amount of osteoid and hypertrophic chondrocytes in the growth plate that could have been remodeled afterward, as indicated by the increased bone resorption. In this regard, a study of longer-term treatment with antibodies could provide more information.

On the other hand, our study also suggests potential risks of FGF23 blockade when considering therapeutic uses. In general, excess amount of 1,25D can cause hypercalcemia, hypercalciuria, and ectopic calcification, such that the tolerable dose of 1,25D for XLH treatment is limited. Treatment with the FGF23 antibodies resulted in a transient but significant elevation of serum 1,25D because of its strong stimulatory effect on 1αOHase mRNA expression. In this study, however, it is of note that the effect of either single or repeated injections of FGF23Ab on serum calcium levels was not striking, and nephrocalcinosis was not observed. This could be a unique characteristic of FGF23Ab compared with the use of 1,25D or its analog. Nevertheless, given the dose-dependent increase in urinary calcium excretion observed with repeated FGF23Ab treatment, a more complete inhibition of FGF23 action could potentially result in hypercalcemia or ectopic mineralization, as seen in the Fgf23KO mouse and patients with tumoral calcinosis.(16–18) This point would be more important in longer-term treatment; therefore, it must be carefully addressed when considering possible clinical uses.

Several mechanisms of action of the FGF23Ab treatment were not fully addressed in our study. First, the pathophysiological impact of excess FGF23 activity on the secretion of PTH in Hyp mice remains unclear. A recent study of compound mutant mice generated by cross-breeding the PTH-null and Hyp mice suggests that the increased PTH secretion in Hyp mice helps to prevent the hypocalcemia caused by the FGF23-mediated decrease in serum level of 1,25D.(31) In addition, the continuous action of recombinant FGF23 in mice has been shown to cause secondary hyperparathyroidism.(11,12) These observations imply that FGF23 indirectly increases PTH secretion by decreasing serum 1,25D levels. Our observations of suppressed serum PTH levels with FGF23Ab treatment may thus be attributed to direct effect of an increase in serum 1,25D level. On the other hand, recent studies showed that FGF23 can suppress PTH secretion in vivo and in vitro.(32,33) These observations are not easily reconciled with our current results. Further studies are thus required to elucidate the relationship between FGF23 and PTH secretion, particularly under hypophosphatemic conditions.

The mechanism of the drastic amelioration of Hyp bone by FGF23 antibodies also remains to be clarified. Treatment with FGF23Ab corrected not only the defect of bone mineralization in Hyp mice but also the defects in longitudinal bone growth accompanied by normalized columnar structures and mineralization of chondrocytes, indicating corrective effects on chondrocytes differentiation and subsequent processes of endochondral ossification. Furthermore, Hyp mice treated with FGF23 antibodies exhibited increased osteoblast numbers, as well as enhanced bone resorption, suggesting a high rate of bone turnover. Given recent reports suggesting direct actions of FGF23 on bone,(34,35) it is possible that the skeletal effects that we observed were in part caused by a loss of FGF23 action within the bone compartment. On the other hand, expression of Klotho has been shown to be a key determinant for the tissue-specific action of circulatory FGF23,(36,37) and expression of this molecule in bone has not been reported. It is thus unlikely that FGF23 activates a Klotho-dependent signaling pathway in bone. Rather it is more conceivable that the dramatically improved phosphate metabolism and enhanced 1,25D action per se are responsible for the observed effects on bone. The efficacy of concomitant oral supplementation of phosphate and 1,25D (or its analogs) on bone has been shown in both patients and Hyp mice,(1,38,39) so that it is widely accepted as the most established treatment for XLH patients despite several unmet issues previously described. The next subject of interest should be to discriminate between indirect effects of corrections in phosphate and elevated 1,25D levels and potential local actions of FGF23 in bone by analyzing the effect of FGF23 antibodies in cell or organ culture or in Hyp mice fed with a vitamin D-deficient diet.

In summary, this study provides new insights into the pathological role of FGF23 in Hyp mice and possibly XLH patients; these indicate that FGF23 blockade through neutralizing antibodies offers a therapeutic modality for XLH patients. In addition, the FGF23 antibodies may open new therapeutic potential for other FGF23-related disorders of renal phosphate wasting and bone mineralization defects, such as ADHR, TIO, and autosomal recessive hypophosphatemic rickets/osteomalacia.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

The authors thank K. Ono, R. Hino, N. Kobayashi, S. Miyata, S. Wakita, and M. Abe for excellent technical assistance and Drs. T. Sakai and M. Paskewitz for fruitful discussion. This work was partly supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan and from the Ministry of Health, Labor and Welfare of Japan.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Alon US 2006 Hypophosphatemic vitamin D-resistant rickets. In: FavusMJ (ed.) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 6th ed. The American Society for Bone and Mineral Research, Washington, DC, USA, 342345.
  • 2
    The Hyp Consortium 1995 A gene (PEX) with homologies to endopeptidases is mutated in patients with X-linked hypophosphatemic rickets. Nat Genet 11 : 130136.
  • 3
    Holm IA, Nelson AE, Robinson BG, Mason RS, Marsh DJ, Cowell CT, Carpenter TO 2001 Mutational analysis and genotype-phenotype correlation of the PHEX gene in X-linked hypophosphatemic rickets. J Clin Endocrinol Metab 86 : 38893899.
  • 4
    Sabbagh Y, Jones AO, Tenenhouse HS 2000 PHEXdb, a locus-specific database for mutations causing X-linked hypophosphatemia. Hum Mutat 16 : 16.
  • 5
    Beck L, Soumounou Y, Martel J, Krishnamurthy G, Gauthier C, Goodyer CG, Tenenhouse HS 1997 Pex/PEX tissue distribution and evidence for a deletion in the 3′ region of the Pex gene in X-linked hypophosphatemic mice. J Clin Invest 99 : 12001209.
  • 6
    Strom TM, Francis F, Lorenz B, Boddrich A, Econs MJ, Lehrach H, Meitinger T 1997 Pex gene deletions in Gy and Hyp mice provide mouse models for X-linked hypophosphatemia. Hum Mol Genet 6 : 165171.
  • 7
    Meyer RA Jr, Meyer MH, Gray RW 1989 Parabiosis suggests a humoral factor is involved in X-linked hypophosphatemia in mice. J Bone Miner Res 4 : 493500.
  • 8
    Nesbitt T, Coffman TM, Griffiths R, Drezner MK 1992 Crosstransplantation of kidneys in normal and Hyp mice. Evidence that the Hyp mouse phenotype is unrelated to an intrinsic renal defect. J Clin Invest 89 : 14531459.
  • 9
    The ADHR Consortium 2000 Autosomal dominant hypophosphataemic rickets is associated with mutations in FGF23. Nat Genet 26 : 345348.
  • 10
    Shimada T, Mizutani S, Muto T, Yoneya T, Hino R, Takeda S, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T 2001 Cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia. Proc Natl Acad Sci USA 98 : 65006505.
  • 11
    Bai X, Miao D, Li J, Goltzman D, Karaplis AC 2004 Transgenic mice overexpressing human fibroblast growth factor 23 (R176Q) delineate a putative role for parathyroid hormone in renal phosphate wasting disorders. Endocrinology 145 : 52695279.
  • 12
    Bai XY, Miao D, Goltzman D, Karaplis AC 2003 The autosomal dominant hypophosphatemic rickets R176Q mutation in fibroblast growth factor 23 resists proteolytic cleavage and enhances in vivo biological potency. J Biol Chem 278 : 98439849.
  • 13
    Larsson T, Marsell R, Schipani E, Ohlsson C, Ljunggren O, Tenenhouse HS, Juppner H, Jonsson KB 2004 Transgenic mice expressing fibroblast growth factor 23 under the control of the alpha1(I) collagen promoter exhibit growth retardation, osteomalacia, and disturbed phosphate homeostasis. Endocrinology 145 : 30873094.
  • 14
    Shimada T, Hasegawa H, Yamazaki Y, Muto T, Hino R, Takeuchi Y, Fujita T, Nakahara K, Fukumoto S, Yamashita T 2004 FGF-23 is a potent regulator of vitamin D metabolism and phosphate homeostasis. J Bone Miner Res 19 : 429435.
  • 15
    Shimada T, Muto T, Urakawa I, Yoneya T, Yamazaki Y, Okawa K, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T 2002 Mutant FGF-23 responsible for autosomal dominant hypophosphatemic rickets is resistant to proteolytic cleavage and causes hypophosphatemia in vivo. Endocrinology 143 : 31793182.
  • 16
    Araya K, Fukumoto S, Backenroth R, Takeuchi Y, Nakayama K, Ito N, Yoshii N, Yamazaki Y, Yamashita T, Silver J, Igarashi T, Fujita T 2005 A novel mutation in fibroblast growth factor 23 gene as a cause of tumoral calcinosis. J Clin Endocrinol Metab 90 : 55235527.
  • 17
    Shimada T, Kakitani M, Yamazaki Y, Hasegawa H, Takeuchi Y, Fujita T, Fukumoto S, Tomizuka K, Yamashita T 2004 Targeted ablation of Fgf23 demonstrates an essential physiological role of FGF23 in phosphate and vitamin D metabolism. J Clin Invest 113 : 561568.
  • 18
    Sitara D, Razzaque MS, Hesse M, Yoganathan S, Taguchi T, Erben RG, Juppner H, Lanske B 2004 Homozygous ablation of fibroblast growth factor-23 results in hyperphosphatemia and impaired skeletogenesis, and reverses hypophosphatemia in Phex-deficient mice. Matrix Biol 23 : 421432.
  • 19
    Jonsson KB, Zahradnik R, Larsson T, White KE, Sugimoto T, Imanishi Y, Yamamoto T, Hampson G, Koshiyama H, Ljunggren O, Oba K, Yang IM, Miyauchi A, Econs MJ, Lavigne J, Juppner H 2003 Fibroblast growth factor 23 in oncogenic osteomalacia and X-linked hypophosphatemia. N Engl J Med 348 : 16561663.
  • 20
    Yamazaki Y, Okazaki R, Shibata M, Hasegawa Y, Satoh K, Tajima T, Takeuchi Y, Fujita T, Nakahara K, Yamashita T, Fukumoto S 2002 Increased circulatory level of biologically active full-length FGF-23 in patients with hypophosphatemic rickets/osteomalacia. J Clin Endocrinol Metab 87 : 49574960.
  • 21
    Liu S, Zhou J, Tang W, Jiang X, Rowe DW, Quarles LD 2006 Pathogenic role of Fgf23 in Hyp mice. Am J Physiol Endocrinol Metab 291 : E38E49.
  • 22
    Liu S, Guo R, Simpson LG, Xiao ZS, Burnham CE, Quarles LD 2003 Regulation of fibroblastic growth factor 23 expression but not degradation by PHEX. J Biol Chem 278 : 3741937426.
  • 23
    Liu S, Tang W, Zhou J, Vierthaler L, Quarles LD 2007 Distinct roles for intrinsic osteocyte abnormalities and systemic factors in regulation of FGF23 and bone mineralization in Hyp mice. Am J Physiol Endocrinol Metab 293 : E1636E1644.
  • 24
    Yamazaki Y, Tamada T, Kasai N, Urakawa I, Aono Y, Hasegawa H, Fujita T, Kuroki R, Yamashita T, Fukumoto S, Shimada T 2008 Anti-FGF23 neutralizing antibodies demonstrate the physiological role and structural features of FGF23. J Bone Miner Res 23 : 15091518.
  • 25
    Tahara T, Kuwaki T, Matsumoto A, Morita H, Watarai H, Inagaki Y, Ohashi H, Ogami K, Miyazaki H, Kato T 1998 Neutralization of biological activity and inhibition of receptor binding by antibodies against human thrombopoietin. Stem Cells 16 : 5460.
  • 26
    Tenenhouse HS, Beck L 1996 Renal Na(+)-phosphate cotransporter gene expression in X-linked Hyp and Gy mice. Kidney Int 49 : 10271032.
  • 27
    Yuan B, Takaiwa M, Clemens TL, Feng JQ, Kumar R, Rowe PS, Xie Y, Drezner MK 2008 Aberrant Phex function in osteoblasts and osteocytes alone underlies murine X-linked hypophosphatemia. J Clin Invest 118 : 722734.
  • 28
    Baum M, Loleh S, Saini N, Seikaly M, Dwarakanath V, Quigley R 2003 Correction of proximal tubule phosphate transport defect in Hyp mice in vivo and in vitro with indomethacin. Proc Natl Acad Sci USA 100 : 1109811103.
  • 29
    Prie D, Blanchet FB, Essig M, Jourdain JP, Friedlander G 1998 Dipyridamole decreases renal phosphate leak and augments serum phosphorus in patients with low renal phosphate threshold. J Am Soc Nephrol 9 : 12641269.
  • 30
    Tenenhouse HS, Martel J, Rubin J, Harvey N 1994 Effect of phosphate supplementation on the expression of the mutant phenotype in murine X-linked hypophosphatemic rickets. Bone 15 : 677683.
  • 31
    Bai X, Miao D, Goltzman D, Karaplis AC 2007 Early lethality in Hyp mice with targeted deletion of Pth gene. Endocrinology 148 : 49744983.
  • 32
    Ben-Dov IZ, Galitzer H, Lavi-Moshayoff V, Goetz R, Kuro OM, Mohammadi M, Sirkis R, Naveh-Many T, Silver J 2007 The parathyroid is a target organ for FGF23 in rats. J Clin Invest 117 : 40034008.
  • 33
    Krajisnik T, Bjorklund P, Marsell R, Ljunggren O, Akerstrom G, Jonsson KB, Westin G, Larsson TE 2007 Fibroblast growth factor-23 regulates parathyroid hormone and 1alpha-hydroxylase expression in cultured bovine parathyroid cells. J Endocrinol 195 : 125131.
  • 34
    Sitara D, Kim S, Razzaque MS, Bergwitz C, Taguchi T, Schuler C, Erben RG, Lanske B 2008 Genetic evidence of serum phosphate-independent functions of FGF-23 on bone. PLoS Genet 4 : e1000154.
  • 35
    Wang H, Yoshiko Y, Yamamoto R, Minamizaki T, Kozai K, Tanne K, Aubin JE, Maeda N 2008 Overexpression of fibroblast growth factor 23 suppresses osteoblast differentiation and matrix mineralization in vitro. J Bone Miner Res 23 : 939948.
  • 36
    Kurosu H, Ogawa Y, Miyoshi M, Yamamoto M, Nandi A, Rosenblatt KP, Baum MG, Schiavi S, Hu MC, Moe OW, Kuro-o M 2006 Regulation of fibroblast growth factor-23 signaling by klotho. J Biol Chem 281 : 61206123.
  • 37
    Urakawa I, Yamazaki Y, Shimada T, Iijima K, Hasegawa H, Okawa K, Fujita T, Fukumoto S, Yamashita T 2006 Klotho converts canonical FGF receptor into a specific receptor for FGF23. Nature 444 : 770774.
  • 38
    Marie PJ, Travers R, Glorieux FH 1982 Bone response to phosphate and vitamin D metabolites in the hypophosphatemic male mouse. Calcif Tissue Int 34 : 158164.
  • 39
    Ono T, Tanaka H, Yamate T, Nagai Y, Nakamura T, Seino Y 1996 24R,25-dihydroxyvitamin D3 promotes bone formation without causing excessive resorption in hypophosphatemic mice. Endocrinology 137 : 26332637.