Identification of the type II Na(+)-Pi cotransporter (Npt2) in the osteoclast and the skeletal phenotype of Npt2-/- mice.

We previously reported that a type II sodium phosphate (Na(+)-Pi) cotransporter (Npt2) protein is expressed in osteoclasts and that Pi limitation decreases osteoclast-mediated bone resorption in vitro. We also demonstrated that mice homozygous for the disrupted Npt2 gene (Npt2-/-) exhibit a unique age-dependent bone phenotype that is associated with significant hypophosphatemia. In the present study, we sought to identify the Npt2 cDNA in mouse osteoclasts and characterize the impact of Npt2 gene ablation on osteoclast function and bone histomorphometry. We demonstrate that the osteoclast Npt2 cDNA sequence is identical to that of the proximal renal tubule and, thus, not an isoform or splice variant thereof. Histomorphometric analysis revealed that, at 25 days of age, Npt2-/- mice exhibited a reduction in osteoclast number and eroded perimeters, relative to wild-type mice. Moreover, although the number of metaphyseal trabeculae was reduced in 25-day-old Npt2-/- mice, trabecular bone volume was normal due to increased trabecular width. At 115 days of age, the decrease in osteoclast index persisted in Npt2-/- mice relative to wild-type littermates. However, mineralizing and osteoblast surfaces and bone formation rates were increased, and, although trabecular number was still reduced, trabecular bone volume was higher than that of wild-type mice. These data demonstrate a link between osteoclast activity and trabecular development in young Npt2-/- mice, and suggest that an age-related adaptation to Npt2 deficiency is apparent in osteoclast and osteoblast function and bone formation.

Disease-causing missense mutations in the PHEX gene interfere with membrane targeting of the recombinant protein.

PHEX is homologous to the M13 zinc metallopeptidases, a class of type II membrane glycoproteins. Although more than 140 mutations in the PHEX gene have been identified in patients with X-linked hypophosphatemia (XLH), the most prevalent form of inherited rickets, the molecular consequences of disease-causing PHEX mutations have not yet been investigated. We examined the effect of PHEX missense mutations on cellular trafficking of the recombinant protein. Four mutant PHEX cDNAs were generated by PCR mutagenesis: C85R, G579R and S711R, identified in XLH patients, and E581V, previously engineered in neutral endopeptidase 24.11, where it abolished catalytic activity but not plasma membrane targeting. Wild-type and mutant PHEX cDNAs were transfected in HEK(293) cells and PHEX protein expression was characterized. In contrast to the wild-type and E581V PHEX proteins, the C85R, G579R and S711R mutants were completely sensitive to endoglycosidase H digestion, indicating that they were not fully glycosylated. Sequestration of the disease-causing mutant proteins in the endoplasmic reticulum (ER) and plasma membrane localization of wild-type and E581V PHEX proteins was demonstrated by immunofluorescence and cell surface biotinylation. Of the three mutant PHEX proteins, the S711R was the least stable and the only one that could be rescued from the ER to the plasma membrane in cells grown at 26 degrees C. The chemical chaperone glycerol failed to correct defective targeting of all three mutant proteins. Our data provide a mechanism for loss of PHEX function in XLH patients expressing the C85R, G579R and S711R mutations.

Effect of phosphate supplementation on the expression of the mutant phenotype in murine X-linked hypophosphatemic rickets.

The X-linked Hyp mouse, a murine homologue of X-linked hypophosphatemia in humans, is characterized by rachitic bone disease, hypophosphatemia, impaired renal brush-border membrane Na(+)-phosphate cotransport and abnormal regulation of renal vitamin D metabolism. We demonstrated that short-term phosphate supplementation decreases renal 1,25-dihydroxyvitamin D3 (1,25-(OH)2D) catabolism and increases serum 1,25-(OH)2D levels in Hyp mice (Tenenhouse & Jones 1990). In the present study, we compared several other parameters in normal and Hyp mice fed control (1%) and high (1.6%) phosphate diets for 4 days. Phosphate supplementation significantly raised serum phosphate levels and decreased renal brush-border membrane Na(+)-phosphate but not Na(+)-glucose, cotransport in both genotypes (67% of control diet, p < 0.05). However, under both dietary conditions, the phosphate/glucose transport ratio was significantly reduced in Hyp mice (58% of normal littermates, p < 0.05). Renal PTH-stimulated cAMP accumulation, which was significantly blunted in Hyp mice compared to normal mice under control dietary conditions (p < 0.05), was not altered by phosphate supplementation in either genotype. Serum alkaline phosphatase activity was significantly higher than normal in Hyp mice on the control diet and was further increased in mutants but not in normals fed the high phosphate diet (p < 0.05). Measurements of serum bilirubin and electrophoresis of serum alkaline phosphatase suggested that the elevation in serum alkaline phosphatase activity in phosphate-supplemented Hyp mice represents the bone-derived isozyme.(ABSTRACT TRUNCATED AT 250 WORDS)

Abnormal regulation of renal vitamin D catabolism by dietary phosphate in murine X-linked hypophosphatemic rickets.

Hyp mice exhibit increased renal catabolism of vitamin D metabolites by the C-24 oxidation pathway (1988. J. Clin. Invest. 81:461-465). To examine the regulatory influence of dietary phosphate on the renal vitamin D catabolic pathway in Hyp mice, we measured C-24 oxidation of 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) in renal mitochondria isolated from Hyp mice and normal littermates fed diets containing 0.03% (low-Pi), 1% (control-Pi), and 1.6% (high-Pi) phosphate. In normal mice the low-Pi diet led to a rise in serum 1,25(OH)2D (22.2 +/- 1.8 to 48.1 +/- 6.8 pg/ml, P less than 0.05) and no change in C-24 oxidation products (0.053 +/- 0.006 to 0.066 +/- 0.008 pmol/mg protein per min) when compared with the control diet. In Hyp mice the low-Pi diet elicited a fall in serum 1,25(OH)2D (21.9 +/- 1.2 to 8.0 +/- 0.2 pg/ml, P less than 0.05) and a dramatic increase in C-24 oxidation products (0.120 +/- 0.017 to 0.526 +/- 0.053 pmol/mg protein per min, P less than 0.05) when compared with the control diet. The high-Pi diet did not significantly alter serum levels of 1,25(OH)2D or C-24 oxidation products in normal mice. Hyp mice on the high-Pi diet experienced a rise in serum 1,25(OH)2D (21.9 +/- 1.2 to 40.4 +/- 7.3, P less than 0.05) and a fall in C-24 oxidation products (0.120 +/- 0.017 to 0.043 +/- 0.007 pmol/mg protein per min, P less than 0.05). The present results demonstrate that the defect in C-24 oxidation of 1,25(OH)2D3 in Hyp mice is exacerbated by phosphate depletion and corrected by phosphate supplementation. The data suggest that the disorder in vitamin D metabolism in the mutant strain is secondary to the perturbation in phosphate homeostasis.

The Gy mutation: another cause of X-linked hypophosphatemia in mouse.

An X-linked dominant mutation (gyro, gene symbol Gy) in the laboratory mouse causes hypophosphatemia, rickets/osteomalacia, circling behavior, inner ear abnormalities, and sterility in males and a milder phenotype in females. Gy maps closely (crossover value 0.4-0.8%) to another X-linked gene (Hyp) that also causes hypophosphatemia in the mouse. Gy and Hyp genes have similar quantitative expression in serum phosphorus values, renal excretion of phosphate, and impairment of Na+/phosphate cotransport by renal brush-border membrane vesicles. These findings indicate that independent translation products of two X-linked genes serve phosphate transport in mouse kidney and thereby control phosphate content of extracellular fluid. The Gy translation product, unlike the Hyp product, is also expressed in the inner ear. These findings have implications for our understanding of the human counterpart known as "X-linked hypophosphatemia."

Metabolism of 25-hydroxyvitamin D3 in renal slices from the X-linked hypophosphatemic (Hyp) mouse: abnormal response to fall in serum calcium.

The effect of the X-linked Hyp mutation on 25-hydroxyvitamin D3 (25-OH-D3) metabolism in mouse renal cortical slices was investigated. Vitamin D replete normal mice and Hyp littermates fed the control diet synthesized primarily 24,25-dihydroxyvitamin D3 (24,25-(OH)2D3); only minimal synthesis of 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) was detected in both genotypes and 1,25-(OH)2D3 formation was not significantly greater in Hyp mice relative to normal littermates, despite hypophosphatemia and hypocalcemia in the mutants. Calcium-deficient diet fed to normal mice reduced serum calcium (p less than 0.01), increased renal 25-hydroxyvitamin D3-1-hydroxylase (1-OHase) activity (p less than 0.05), and decreased 25-hydroxyvitamin D3-24-hydroxylase (24-OHase) activity (p less than 0.05). In contrast, Hyp littermates on the calcium-deficient diet had decreased serum calcium (p less than 0.01), without significant changes in the renal metabolism of 25-OH-D3. Both normal and Hyp mice responded to the vitamin D-deficient diet with a fall in serum calcium (p less than 0.01), significantly increased renal 1-OHase, and significantly decreased renal 24-OHase activities. In Hyp mice, the fall in serum calcium on the vitamin D-deficient diet was significantly greater than that observed on the calcium-deficient diet. Therefore the ability of Hyp mice to increase renal 1-OHase activity when fed the vitamin D-deficient diet and their failure to do so on the calcium-deficient diet may be related to the resulting degree of hypocalcemia. The results suggest that although Hyp mice can respond to a disturbance of calcium homeostasis, the in vivo signal for the stimulation of renal 1-OHase activity may be set at a different threshold in the Hyp mouse; i.e. a lower serum calcium concentration is necessary for Hyp mice to initiate increased synthesis of 1,25(-OH)2D3.

Orthophosphate transport in the erythrocyte of normal subjects and of patients with X-linked hypophosphatemia.

We have examined the mechanism of TCA-soluble orthophosphate (Pi) transfer across the membrane of mature human erythrocytes in normal subjects and in patients with X-linked hypophosphatemia (X-LH). The studies were carried out largely at pH 7.4 and 37 degrees C, in partial stimulation of conditions in vivo. (a) At physiological concentrations (1-2 mM) Pi enters the intact normal erythrocyte down its chemical gradient and under no conditions could we identify a steady-state trans-membrane gradient for Pi greater than 0.6. Calculations of the phosphate anion distribution ratio using the Nernst equation yield theoretical values that closely approximate observed values. (b) Glycolytic inhibitors have little effect on total entry of 32Pi inti erythrocytes but they do affect the intracellular distribution of Pi. In the presence of iodoacetamide, label accumulates almost exclusively in the orthophosphate pool and less than 1% enters the organic phosphate pool. (c) Specific activity measurements in unblocked cells indicate that Pi anion equilibrates first with its intracellular Pi pool. These initial findings imply that neither group translocation, nor energy coupling, influence Pi permeation into the human erythrocytes. (d) The relationship between 32P entry and extracellular Pi concentration is parabolic in the presence of chloride, and linear in the presence of sulfate. The kinetics of concentration dependent entrance cannot be examined and saturability of Pi entry cannot be identified under these conditions. (e) The competitive inhibitor arsenate partially inhibits the initial rate and steady-state flux of orthophosphate in erythrocytes treated with iodoacetamide to inhibit glycolysis. However, a significant portion of Pi transport escapes arsenate inhibition. (f) Activation energies for Pi entry, in nonglycolizing erythrocytes are much higher than those required by simple diffusion in an aqueous system. (g) Neither the inward or outward movement of Pi is modulated by trans-phosphate. These latter findings suggest that transport of phosphate across the human erythrocyte is compatible with slow facilitated diffusion with symmetry for influex and efflux. The transmembrane chemical distribution ratio, and the equilibrium flux of Pi were not different from normal in the X-LH erythrocyte. Nor did the extracellular Pi concentration, arsenate, or temperature affect Pi entry differently in the two types of cells. We dedjce that different gene products serve the diffusional type of Pi transport in the erythrocyte membrane and the saturable component of transepithelial absorption in the gut and kidney. Only the latter is affected by the X-LH mutation. The former is apparently present not only in erythrocytes but also in epithelial tissue, where it can serve the absorption of pharmacologic amounts of Pi in the therapeutic repair of the depleted phosphate pools in X-LH.