The role of DMP1 in FGF23-induced hypophosphatemia

Fibroblast growth factor 23 (FGF23) is a phosphaturic hormone produced by bone. Hypophosphatemic rickets diseases, such as X-linked hypophosphatemia (XLH) and autosomal recessive hypophosphatemic rickets (ARHR), are associated with FGF23 excess, impaired skeletal growth and osteomalacia leading to debilitating bone pain and fractures. Therapeutic trials testing the prevention of FGF23 phosphaturic effect using FGF23 blocking antibody in XLH have shown great promise to improve serum Pi levels and bone mineralization. Further studies are needed to determine whether blocking FGF23 will be efficacious in the long term and in other diseases associated with FGF23 excess, including ARHR. In humans and mice, XLH and ARHR type I are respectively caused by inactivating mutations of phosphate regulating gene with homologies to endopeptidase X-linked (PHEX) and dentin matrix protein (DMP1) that result in nearly identical elevation in FGF23 production by osteocytes, and overlapping bone and mineral phenotypes. Previous studies conducted in Phex mutant (Hyp) mice and Dmp1-null (Dmp1KO) mice suggested that PHEX is a functional partner of DMP1 explaining the convergence of both mutant phenotypes. PHEX and DMP1 are mainly expressed in mineralized tissues, and minimal expression was reported in soft tissues, including kidney. In preliminary data for this project, we show that osteocyte-specific deletion of Fgf23 (Fgf23cKO) in Dmp1KO mice results in a full correction of serum Pi levels, but in incomplete corrections of FGF23 levels, bone growth and mineralization, suggesting that additional Pi-independent mechanisms are involved in the pathogenesis of ARHR. Intriguingly, hypophosphatemia resurged in older Dmp1KO/Fgf23cKO compound mutant mice despite lower FGF23 levels. In additional preliminary data, induction of FGF23 levels by a 2% Pi diet leads to increased Pi excretion by the kidney in wild-type (WT) mice but not in mice with Dmp1 genetic overexpression (Dmp1TG) (vs. 0.2% Pi diet), suggesting a role for DMP1 in blocking FGF23 effects in the kidney. In final preliminary data, kidney transplantation performed from Dmp1TG donor into Hyp recipient mice resulted in reduced phosphaturia despite further elevations of FGF23 (vs. isogenic Hyp transplant), suggesting that kidney expressed PHEX and DMP1 repress FGF23-induced phosphaturia. The goal of this project is to investigate the pathogenesis of hypophosphatemia induced by DMP1 deficiency. In Aim 1, we will define the direct contribution of FGF23 excess and hypophosphatemia to impaired bone mineralization in mice with ARHR at different ages. In Aim 2, we will establish the role of DMP1 in the kidney using dietary phosphate challenges in mice with genetic overexpression of DMP1, FGF23 or both. Finally, in Aim 3, we will investigate the contribution of kidney DMP1 deficiency to the pathophysiology of ARHR and XLH by performing kidney transplants in mice with ARHR and XLH. Ultimately, this project will support the development of therapeutic strategies that will improve outcomes in diseases associated with FGF23 excess.