FGF23 Is Not Required to Regulate Fetal Phosphorus Metabolism but Exerts Effects Within 12 Hours After Birth

Abstract

Loss of fibroblast growth factor-23 (FGF23) causes hyperphosphatemia, extraskeletal calcifications, and early mortality; excess FGF23 causes hypophosphatemia with rickets or osteomalacia. However, FGF23 may not be important during fetal development. FGF23 deficiency (Fgf23 null) and FGF23 excess (Phex null) did not alter fetal phosphorus or skeletal parameters. In this study, we further tested our hypothesis that FGF23 is not essential for fetal phosphorus regulation but becomes important after birth. Although coreceptor Klotho null adults have extremely high FGF23 concentrations, intact FGF23 was normal in Klotho null fetuses, as were fetal phosphorus and skeletal parameters and placental and renal expression of FGF23 target genes. Pth/Fgf23 double mutants had the same elevation in serum phosphorus as Pth null fetuses, as compared with normal serum phosphorus in Fgf23 nulls. We examined the postnatal time courses of Fgf23 null, Klotho null, and Phex null mice. Fgf23 nulls and Klotho nulls were normal at birth, but developed hyperphosphatemia, increased renal expression of NaPi2a and NaPi2c, and reduced renal phosphorus excretion between 5 and 7 days after birth. Parathyroid hormone remained normal. In contrast, excess FGF23 exerted effects in Phex null males within 12 hours after birth, with the development of hypophosphatemia, reduced renal expression of NaPi2a and NaPi2c, and increased renal phosphorus excretion. In conclusion, although FGF23 is present in the fetal circulation at levels that may equal adult values, and there is robust expression of FGF23 target genes in placenta and fetal kidneys, FGF23 itself is not an important regulator of fetal phosphorous metabolism.

Figure 1.

Loss of Klotho does not disturb fetal phosphorus, PTH, calcitriol, or placental phosphorus transport. In contrast to the very high intact FGF23 concentrations that have been found in postnatal (young and adult) Klotho null mice, (A) Klotho null fetuses had normal serum intact FGF23. Klotho null fetuses showed no disturbances in (B) serum phosphorus and (C) calcium, (D) amniotic fluid phosphorus and (E) calcium, (F) serum PTH, or (G) serum calcitriol. (H) Placental transport of ³²P at 5 minutes was normal when corrected for ⁵¹Cr-EDTA diffusion or when ³²P was analyzed alone (not shown). The numbers of observations are indicated in parentheses. mM, mmol/L.

Figure 2.

Skeletal parameters are normal in Klotho null fetuses. (A) Klotho null tibias showed normal endochondral development with no alteration in the lengths or cellular morphology of the cartilaginous or boney compartments, and a normal distribution of mineral (black stain due to von Kossa). Fetuses were reduced to ash and the mineral content of that ash was then assayed. On ED 18.5, Klotho null fetuses had normal (B) ash weight as well as ash content of (C) phosphorus, (D) calcium, and magnesium (not shown). The numbers of observations are indicated in parentheses.

Figure 3.

Expression of Cyp24a1 among kidneys of the 3 null genotypes vs corresponding WT. Cyp24a1 expression was normal in Fgf23 null and Klotho null fetal kidneys, but significantly increased in Phex null male fetal kidneys. To compare all 6 different genotypes, all values were normalized to the WT placentas from the Fgf23 colony. The numbers of observations are indicated in parentheses.

Figure 4.

Serum phosphorus and skeletal mineral content in Fgf23/Pth double mutant fetuses. (A) Loss of PTH caused the expected increase in serum phosphorus, whereas superimposed loss of FGF23 in double-mutants (DKO) had no effect. Loss of PTH reduced skeletal (C) calcium but not (B) phosphorus or (D) magnesium content, whereas loss of FGF23 did not affect any of these parameters. The numbers of observations are indicated in parentheses. mM, mmol/L.

Figure 5.

Postnatal changes in serum phosphorus, renal expression of sodium-phosphate cotransporters, and urine phosphorus excretion among Fgf23, Klotho, and Phex mutant colonies. (A) After an interval of normal serum phosphorus, Fgf23 null mice became hyperphosphatemic by day 7 after birth (left panel), and on that day renal expression of NaPi2a and NaPi2c (middle panel) were significantly increased, whereas renal phosphorus excretion (right panel) was modestly but significantly reduced. (B) Klotho null mice developed hyperphosphatemia by day 5 after birth (left panel), and this coincided with increased renal expression of NaPi2a and NaPi2c (middle panel), and a trend for reduced renal phosphorus excretion (right panel). In contrast to the delayed effects that loss of FGF23 and Klotho have on postnatal phosphorus metabolism, (C) the effects of high levels of FGF23 in Phex null males are shown. Reduced serum phosphorus developed within 12 hours after birth (left panel), accompanied by reduced renal expression of NaPi2a and NaPi2c (middle panel), and increased renal phosphorus excretion (right panel). In all 3 mutant models, there was no change in renal NaPi2b expression at any time point (not shown). The numbers of observations are indicated in parentheses. mM, mmol/L.

Figure 6.

Postnatal changes in serum PTH in Fgf23 and Klotho mutant colonies. With (A) Fgf23 nulls and (B) Klotho nulls, the slow emergence of hyperphosphatemia was not accompanied by any compensatory increase in PTH. The numbers of observations are indicated in parentheses. mM, mmol/L.