Journal Facts

Journal
BoneKEy Knowledge Environment
ISSN
1533-4368
Volume
Year
2001

Article Facts

Title
Hereditary hypophosphataemic rickets: Role for a fibroblast growth factor, FGF23
DOI
10.1138/2001027
Author

Online Date
05/09/2001

Article Links

BoneKEy-Osteovision | Commentary

Hereditary hypophosphataemic rickets: Role for a fibroblast growth factor, FGF23


DOI:10.1138/2001027

The disorders of hypophosphatemic rickets represent the commonest causes of hereditary rickets and can be divided into two main groups according to the predominant metabolic abnormality (). In the first group, hypophosphatemia is the result of a renal tubular defect, which may consist of either a single (isolated) transport defect in phosphate handling, as occurs in the X-linked and autosomal dominant forms of hypophosphatemic rickets (XLH and ADHR, respectively), or multiple transport defects in the handling of phosphate, amino acids, glucose, bicarbonate and potassium as occurs in the Fanconi syndromes of Lowe (oculocerebro-renal syndrome) and Dent's disease. In the second group, vitamin D metabolism is abnormal, either because of a defect in the 1α-hydroxylase enzyme or because of defects in the 1,25-dihydroxy vitamin D3 receptor (VDR) leading to end organ resistance. The application of molecular genetic approaches has helped to elucidate some of the mechanisms underlying these disorders of hereditary hypophosphatemic rickets. Thus, XLH has been shown to be due to inactivating mutations of the phosphate-regulating gene with homologies to endopeptidases on the X chromosome, PHEX (); Lowe's syndrome is caused by mutations that result in a deficiency of lipid phosphatase that likely controls cellular levels of the metabolite, phosphatidyl inositol 4, 5 bisphosphate; Dent's disease results from loss of function mutations of a member of the voltage-gated chloride channel family, CLC-5 (); vitamin D-dependent rickets (VDDR) type I results from a deficiency of the renal 1α-hydroxylase enzyme, which is a cytochrome P450 enzyme that forms part of the superfamily of heme-containing proteins that are bound to the membranes of microsomes and mitochondria and serve as oxidation-reduction components of the mixed-function oxidase system; and VDDR type II is caused by mutations involving the VDR, which is closely related to the thyroid hormone receptors and represents another member of the transacting transcriptional factors that include the family of steroid hormone receptors. Studies by Econs and colleagues () have now identified the molecular basis of ADHR and elucidated a role for a member of the fibroblast growth factor (FGF) family. Additional studies by Econs and colleagues () and Shimada et al. () link ADHR with the non-hereditary disorder of oncogenic hypophosphatemic osteomalacia (OHO). These studies and their implications for phosphate homeostasis will be reviewed.

ADHR is characterised by low serum phosphate concentrations, bone pain, rickets that result in deformities of the legs, osteomalacia and dental caries. ADHR and XLH thus have marked clinical similarities but differ in their modes of inheritance. Genetic linkage studies mapped the ADHR locus to chromosome 12p13.3 () and defined a 1.5 Mb critical region that contained 12 genes. Mutational analyses of 6 of these 12 genes revealed the occurrence of missense mutations involving a new member of the fibroblast growth factor (FGF) family, FGF23, in 4 unrelated ADHR families. The FGF23 gene consists of 3 exons that span 10 Kb of genomic sequence, and encode a predicted 251 amino acid protein that has 25% to 35% homology with the FGF family members; FGF23 is most closely related to FGF15, FGF19 and FGF21. The gene contains a 146bp 5′ UTR and a 710bp 3′ UTR with a predicted poly(A) signal located 831bp downstream of the Stop codon. The FGF23 gene is normally transcribed at low levels only, such that it can only be detected by RT-PCR, and not by Northern blot analysis in human heart, liver, thyroid/parathyroid, small intestine, testis, skeletal muscle and foetal chondrocytes. However, the FGF23 gene is transcribed at high levels, which were detectable by Northern blot analysis, in tumorous cells such as the chronic myelogenous leukaemia cell line K562 and in those from OHO tumours (). Western blot analysis has revealed the presence of a 32kDa band, consistent with a post-translationally modified 231 amino acid protein, which lacked a putative N-terminal signal peptide encompassing codons 1-24. Three missense mutations of FGF23 have been identified in 4 unrelated ADHR families, and these cluster at codons 176 and 179. Two unrelated ADHR have an identical mutation involving codon 176, in which the normal positively charged arginine residue is replaced by a polar but uncharged glutamine residue (Arg 176 Gln). The other two mutations involve codon 179, and in one ADHR family the normal arginine residue is replaced by a non-polar tryptophan (Trp) residue (Arg 179 Trp) and in the other ADHR family, it is replaced by a glutamine residue (Arg 179 Gln). The clustering of these ADHR missense mutations that alter arginine residues has lead to the speculation that they may cause a gain of function. Interestingly, mutational analysis of FGF23, in 18 patients, who had hypophosphatemic rickets but did not have PHEX mutations, revealed no abnormalities, suggesting a role(s) for other genes in these hereditary disorders of hypophosphatemic rickets.

However, investigations of the non hereditary disorder of OHO has revealed a role for FGF23, together with some intriguing possibilities of a phosphate homeostatic pathway involving FGF23, PHEX and a yet to be identified phosphaturic factor, referred to as phosphatonin. OHO is a rare disorder characterised by hypophosphatemia, hyperphosphaturia, a low circulating 1,25- dihydroxy-vitamin D3 concentration and osteomalacia, and occurs in previously unaffected individuals. Thus there are similarities between the tumour-induced form of hypophosphatemia, XLH and ADHR. In OHO, the clinical and biochemical abnormalities resolve rapidly after the removal of the tumour, whereas in XLH and ADHR these abnormalities are lifelong. However, the similarities between OHO, ADHR and XLH suggest that they may involve the same phosphate-regulating pathway, and it is important to note that OHO tumours do express PHEX, which is mutated in XLH. Econs and colleagues, therefore explored the possibility that FGF23, which is mutated in ADHR, may also be expressed in OHO tumours and furthermore that FGF23 may be a secreted protein. Indeed, OHO tumours were found by Northern blot analysis to contain high levels of FGF23 mRNA and Western blot analysis confirmed the presence of the 32kDa FGF23 protein. Moreover, transient transfection of FGF23 constructs in OK-E, COS-7 and HEK293 cells revealed efficient synthesis and secretion of FGF23 protein by all 3 cell lines. Thus, OHO tumours abundantly express FGF23, which is a secreted protein, and these findings suggest that FGF23 may be the phosphaturic factor “phosphatonin”. Shimada and colleagues cloned FGF23 from a hemangiopericytoma that caused OHO (). When they expressed the protein, both an approximately 32 kDa form and a short form were isolated; the truncated protein had Ser 180 at its amino terminus, identifying a presumptive cleavage site at Arg 179/Ser 180. Remembering that mutations of Arg 176 and Arg 179 are associated with ADHR, one could postulate a unifying hypothesis to explain the aetiology of the hypophosphatemia in these 3 disorders, based upon a simple enzyme (PHEX) - substrate (FGF23, phosphatonin) interaction () that results in excess phosphatonin either because of impaired degradation, (in XLH and ADHR) or increased production in OHO. For example, in XLH the inactivating PHEX mutations fail to degrade phosphatonin (or FGF23) while in ADHR the mutant FGF23 protein is not cleared by PHEX because the presumptive cleavage site at Arg 179 has been abolished, with the net result being an excess of phosphatonin (FGF23). In contrast, in OHO there is overproduction of FGF23 (ie phosphatonin) such that this exceeds the capacity of the normal PHEX activity, with a the net result again being an excess of phosphatonin. In support of a direct role of FGF23 in OHO, Shimada and colleagues report that direct injection of recombinant FGF23 produced mild phosphaturia, and growth in nude mice of CHO cells that produced FGF23 caused severe hypophosphatemia and rickets ().

Attractive as this hypothesis may be, one still needs to prove that FGF23 is a substrate for PHEX and that FGF23 has a phosphaturic action independent of parathyroid hormone (PTH). In the latter regard, Shimada and colleagues reported that FGF23 did not have direct effects on phosphate reabsorption in the renal tubule cell line OK (), suggesting either that further processing of FGF23 is required for manifestation of its phosphaturic effects, or that FGF23 induces a final phosphaturic factor. Furthermore, this hypothesis does not embody roles for matrix extracellular phosphoglycoprotein (MEPE), which has been proposed as another candidate for the phosphaturic factor in OHO (), or for the human homologues of staniocalcin 1 and staniocalcin 2, which stimulate and inhibit renal phosphate reabsorption, respectively. Much still remains to be elucidated in the pathways regulating phosphate homeostasis, and the identification of roles for FGF23 in the etiology of the hypophosphatemia in both ADHR and OHO represent significant important steps.

Creative Commons License This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.