Renal phosphate transport and vitamin D metabolism in X-linked hypophosphatemic Gy mice: responses to phosphate deprivation

A novel Mixture Model Method for identification of differentially expressed genes from DNA microarray data

Background

The main goal in analyzing microarray data is to determine the genes that are differentially expressed across two types of tissue samples or samples obtained under two experimental conditions. Mixture model method (MMM hereafter) is a nonparametric statistical method often used for microarray processing applications, but is known to over-fit the data if the number of replicates is small. In addition, the results of the MMM may not be repeatable when dealing with a small number of replicates. In this paper, we propose a new version of MMM to ensure the repeatability of the results in different runs, and reduce the sensitivity of the results on the parameters.

Results

The proposed technique is applied to the two different data sets: Leukaemia data set and a data set that examines the effects of low phosphate diet on regular and Hyp mice. In each study, the proposed algorithm successfully selects genes closely related to the disease state that are verified by biological information.

Conclusion

The results indicate 100% repeatability in all runs, and exhibit very little sensitivity on the choice of parameters. In addition, the evaluation of the applied method on the Leukaemia data set shows 12% improvement compared to the MMM in detecting the biologically-identified 50 expressed genes by Thomas et al. The results witness to the successful performance of the proposed algorithm in quantitative pathogenesis of diseases and comparative evaluation of treatment methods.

The genomic response of the mouse kidney to low-phosphate diet is altered in X-linked hypophosphatemia

The mechanism for the renal adaptation to low-phosphate diets is not well understood. Whether the Hyp mutation of the Phex gene blocks this adaptation is also not clear. To gain further insight into this, 5-wk-old normal and Hyp mice were fed a control (1.0% P) or low-phosphate diet (0.03% P) for 3-5 days. Renal RNA was hybridized to Affymetrix U74Av2 microarrays (5 arrays/group). Of the 5,719 detectable genes on each array, 290 responded significantly (P < 0.01) to low-phosphate diet in normal mice. This was reduced significantly (P < 0.001) to 7 in the Hyp mice. This suggested that the adaptations of the normal kidney to a low-phosphate environment were blocked by the Hyp mutation. The Npt2 phosphate transporter, vitamin D 1alpha- and 24-hydroxylases, and calbindins D9K and D28K responded in the expected fashion. Genes with significant (P < 0.05) diet-by-genotype interaction were analyzed by GenMAPP and MAPPFinder. This revealed a cluster of differentially expressed genes associated with microtubule-based processes. Most alpha- and beta-tubulins and most kinesins had responses to low-phosphate diet in normal mice which were abolished or reversed in Hyp mice. In summary, renal adaptation to low-phosphate diet involved changes in the mRNA expression of specific genes. Disruption of these responses in Hyp mice may contribute to their abnormal phosphate homeostasis.

Renal expression of the sodium/phosphate cotransporter gene, Npt2, is not required for regulation of renal 1 alpha-hydroxylase by phosphate

Several reports have suggested that the regulation of renal 1,25-dihydroxyvitamin D [1,25-(OH)(2)D] synthesis by extracellular phosphate (Pi) is dependent on normal transepithelial Pi transport by the renal tubule. Mice homozygous for the disrupted Na/Pi cotransporter gene Npt2 (Npt2(-/-)) exhibit renal Pi wasting, an approximately 85% decrease in renal brush border membrane Na/Pi cotransport, hypophosphatemia, and an increase in serum 1,25-(OH)(2)D concentration. We undertook 1) to determine the mechanism for the increased circulating levels of 1,25-(OH)(2)D in Npt2(-/-) mice and 2) to establish whether renal 1alpha-hydroxylase was appropriately regulated by dietary Pi in the absence of Npt2 gene expression. On a control diet, the 2.5-fold increase in the serum 1,25-(OH)(2)D concentration in Npt2(-/-) mice, relative to that in Npt2(+/+) littermates, is associated with a corresponding increase in renal mitochondrial 25-hydroxyvitamin D-1 alpha-hydroxylase (1 alpha-hydroxylase) activity and messenger RNA (mRNA) abundance. A low Pi diet elicits an increase in serum 1,25-(OH)(2)D concentration, renal 1alpha-hydroxylase activity, and mRNA abundance in Npt2(+/+) and Npt2(-/-) mice to similar levels in both mouse strains. A high Pi diet has no effect on serum 1,25-(OH)(2)D concentration, renal 1 alpha-hydroxylase activity, or mRNA abundance in Npt2(+/+) mice, but normalizes these parameters in Npt2(-/-) mice. In addition, renal 24-hydroxylase mRNA abundance is significantly reduced in Npt2(-/-) mice compared with that in Npt2(+/+) mice under all dietary conditions. In summary, we demonstrate that 1) increased renal synthesis of 1,25-(OH)(2)D is responsible for the increased serum 1,25-(OH)(2)D concentration in Npt2(-/-) mice; and 2) renal 1alpha-hydroxylase gene expression is appropriately regulated by dietary manipulation of serum Pi in both Npt2(+/+) and Npt2(-/-) mice. Thus, intact renal Na/Pi cotransport is not required for the regulation of renal 1alpha-hydroxylase by Pi.

Positional cloning of the PEX gene: new insights into the pathophysiology of X-linked hypophosphatemic rickets

X-linked hypophosphatemic rickets (HYP) is the most common form of hereditary renal phosphate wasting. The hallmarks of this disease are isolated renal phosphate wasting with inappropriately normal calcitriol concentrations and a mineralization defect in bone. Studies in the Hyp mouse, one of the murine models of the human disease, suggest that there is an approximately 50% decrease in both message and protein of NPT-2, the predominant sodium-phosphate cotransporter in the proximal tubule. However, human NPT-2 maps to chromosome 5q35, indicating that it is not the disease gene. Positional cloning studies have led to the identification of a gene, PEX, which is responsible for the disorder. Further studies have led to identification of the murine Pex gene, which is mutated in the murine models of the disorder. These studies, in concert with other studies, have led to improved understanding of the pathophysiology of HYP and a new appreciation for the complexity of normal phosphate homeostasis.

Regulation of renal vitamin D-24-hydroxylase by phosphate: effects of hypophysectomy, growth hormone and insulin-like growth factor I

Dietary phosphate (Pi) regulates the production of 1,25-dihydroxyvitamin D3. This control is blunted in hypophysectomized (HPX) rats but can be restored by growth hormone (GH) or IGF-I. The regulation of the vitamin D catabolism by Pi, GH and IGF-I is less clear. In the present study, we found that the activity and transcript levels of the catabolic enzyme, vitamin D-24-hydroxylase (24-OHase) were decreased 3- and 5-fold, respectively, during Pi restriction in normal rats, but this effect is greatly reduced in HPX rats. Examination of the serum chemistries revealed that HPX rats on the high Pi diet had lower serum Pi levels than normal rats on this diet, presumably due to the known defective reabsorption of Pi by HPX rats. Treatment of HPX rats, adapted to a 0.6% P diet, with GH (150 micrograms) or IGF-I (80 micrograms) suppressed 24-OHase mRNA levels by 88% and 64%, respectively, by 20 hours and these effects were preceded by decreases in serum Pi. Our findings show that the 24-OHase is regulated by dietary Pi and this control is modulated by hypophysectomy, GH and IGF-I.

New perspectives on the biology and treatment of X-linked hypophosphatemic rickets

This article updates the practicing pediatrician's knowledge of the hypophophatemic disorders that may occur in children. The classic X-linked disorder is emphasized. Details of clinical manifestations, the wide spectrum of disease severity, and complications of the disorder in adults are reviewed. Recent research, new genetic findings, and speculations regarding pathophysiology are discussed. A strategy for approaching medical treatment of X-linked hypophosphatemic rickets is provided, together with complications of treatment and treatment after cessation of growth.

Renal Na(+)-phosphate cotransporter gene expression in X-linked Hyp and Gy mice

The X-linked Hyp and Gy mutations are murine homologues of X-linked hypophosphatemia (XLH), a dominant disorder of phosphate (Pi) homeostasis characterized by growth retardation, rickets, hypophosphatemia and decreased renal tubular maximum for Pi reabsorption relative to glomerular filtration rate (Tmp/GFR). In Hyp and Gy mice, the decrease in Tmp/GFR is associated with a reduction in renal brush-border membrane (BBM) Na(+)-Pi cotransport that can be ascribed to a decrease in renal-specific, Na(+)-Pi cotransporter (NPT2) mRNA and protein abundance. Although renal NPT2 gene expression is reduced in Hyp and Gy mice, the NPT2 gene does not map to the X chromosome. These findings exclude NPT2 as a candidate gene for murine and human X-linked hypophosphatemias and suggest that genes at the Hyp, Gy and XLH (HYP) loci are involved in regulation of NPT2 gene expression. Both Hyp and Gy mice respond to low Pi diet with an increase in BBM Na(+)-Pi cotransport, NPT2 mRNA and protein. The increase in NPT2 protein in Pi-depleted mice far exceeds the increase in NPT2 mRNA, suggesting that translational or post-translational mechanisms are involved in the adaptive process. NPT2 protein is localized to the apical surface of the proximal tubule, where immunostaining in both normal and Hyp mice is increased in response to low Pi diet. Pi-deprived Hyp and Gy mice fail to show an increase in Tmp/GFR, indicating that adaptation at the BBM is not sufficient for the overall increase in Tmp/GFR in response to low Pi diet.

Renal expression of Na+-phosphate cotransporter mRNA and protein: effect of the Gy mutation and low phosphate diet

The X-linked Gy mutation is closely linked, but not allelic, to Hyp and is characterized by rickets, hypophosphatemia, decreased renal tubular maximum for phosphate (Pi) reabsorption (TmP) and a specific reduction in renal brush-border membrane (BBM) Na+-Pi cotransport. Gy mice, like their normal littermates, respond to a low-Pi diet with an increase in BBM Na+-Pi cotransport, but fail to show an adaptive increase in Tmp. Using an antibody raised against the NH2 terminal peptide of the rat renal-specific Na+-Pi cotransporter (NaPi-2) and a NaPi-2 cDNA probe, we examined the effect of the Gy mutation and low-Pi diet (0.03% Pi) on NaPi-2 protein and mRNA abundance. The reduction in BBM Na+-Pi cotransport in Gy mice (51 +/- 5% of normal, P < 0.05) was associated with a decrease in NaPi-2 protein (46 +/- 12% of normal, P < 0.05) and mRNA abundance (76 +/- 5%, P < 0.05). The low-Pi diet elicited a two- to three-fold increase in Na+-Pi cotransport in both normal and Gy mice that was accompanied by a large increase in NaPi-2 protein (10.2-fold in normal and 16.9-fold in Gy mice) and a modest increase in NaPi-2 mRNA (1.3-fold in both mouse strains, P < 0.05). The present data demonstrate that (1) the renal defect in BBM Pi transport in Gy mice can be ascribed to a deficit in NaPi-2 protein and mRNA abundance, (2) both normal and Gy mice respond to low Pi with an adaptive increase in NaPi-2 protein that exceeds the increase in Na+-Pi cotransport activity and NaPi-2 mRNA, (3) the adaptive increase in NaPi-2 protein and mRNA are not sufficient for the overall increase in TmP following Pi restriction.

Femoral abnormalities and vitamin D metabolism in X-linked hypophosphatemic (Hyp and Gy) mice

X-linked hypophosphatemia is a genetic bone disease in humans and mice. Two closely linked mutations in mice, Hyp and Gy, cause low plasma phosphate and a rachitic and osteomalacic bone disease. Because of the controversy as to whether Gy is a good model for X-linked hypophosphatemia, the phenotypic severity of these two mutations was compared in both sexes and on two genetic backgrounds. The depression in plasma levels of phosphate was similar in all 10-week-old mutant mice. Male Hyp mice and heterozygous female Hyp mice were affected with similar severity in terms of reduced tail growth, shortened femora, reduced femoral mineral content, and abnormal mineral composition of the femoral matrix. In contrast, male Gy mice did not survive on the C57BL/6J background and were more severely affected than female Gy mice on the B6C3H background. The hybrid B6C3H background ameliorated the bone disease compared with the inbred C57BL/6J background for both mutant strains. There was no evidence of change in the plasma levels of 1,25-dihydroxyvitamin D, duodenal level of vitamin D-dependent calcium-binding protein, or urinary level of calcium in these adult mutant mice. In summary, Gy mice have a sexual dimorphism not present in Hyp mice. These two genes may indicate the presence of multiple gene loci in the human disease, with multiple proteins involved in the pathophysiology of the bone disease.

Phosphate transport by osteoblasts from X-linked hypophosphatemic mice

Hypophosphatemic vitamin D-resistant rickets is characterized by impaired renal reabsorption of Pi. The underlying mechanism of this abnormality remains unknown. Because the osteoblast is likely a target for the HYP mutation, we investigated the Pi transport activity in osteoblasts isolated from the murine homologue for the human disease, the Hyp mouse. Kinetic analysis of sodium-dependent Pi uptake in quiescent normal and Hyp osteoblasts indicated no significant differences in apparent maximal capacity (Vmax) and apparent affinity (Km) of the carrier for Pi. In rapidly growing cells, higher levels of Pi uptake were observed in mutants cells associated with a 1.4- to 1.7-fold increase in Vmax and no change in Km for Pi. This increase in Pi uptake seemed related to changes in the sodium electrochemical gradient inasmuch as a similar increase was observed in alanine uptake. The adaptive response of sodium-dependent Pi transport to Pi deprivation was not altered in mutant cells relative to normal cells. To test whether the expression of a Pi transport defect was dependent on a humoral factor for its expression, we evaluated the activity of the serum from Hyp mice on Pi transport in osteoblasts from both genotypes. No difference in activity was observed between sera from normal and mutant mice. In summary, cultured osteoblasts derived from Hyp mice did not express impaired sodium-dependent Pi transport when compared with cells from normal mice.

Intestinal malabsorption of 45calcium in young Gy mice, a second model for X-linked hypophosphatemia

X-linked hypophosphatemia, a common metabolic bone disease in humans and mice (the Hyp and Gy mutations), is characterized by decreased plasma phosphate, decreased renal tubular reabsorption of phosphate, rickets, and osteomalacia. The question of whether intestinal malabsorption of calcium contributes to the bone disease is controversial. Intestinal absorption of 45Ca was studied in three different mouse colonies: Gy on B6C3H background, Hyp on B6C3H background, and Hyp on C57BL/6J background, all at 4 weeks of age. The duodenum was isolated by sutures, and 45Ca in a 150 mM NaCl and 2 mM CaCl2 solution at pH 7.2 was injected into the lumen. Absorption was measured by the amount of 45Ca remaining in the lumen and by the plasma isotope level. The Gy and Hyp mice of both sexes significantly malabsorbed 45Ca at 4 weeks of age compared to normal littermates. Following the 4 week study, intestinal absorption was measured at 2, 7-8, and 12 weeks of age in normal and Gy mice on the B6C3H background. At 2 and 7-8 weeks of age, the Gy males significantly malabsorbed 45Ca compared to their normal littermates. Serum 1,25-dihydroxyvitamin D was not significantly altered in Gy males at 4 weeks of age. This suggests the possibility of resistance of the intestine to stimulation. Malabsorption of calcium in young Gy and Hyp mice may exacerbate the low mineralization in their rachitic bone disease.

Renal adaptation to phosphate deprivation: lessons from the X-linked Hyp mouse

The X-linked Hyp mutation, a murine homologue of X-linked hypophosphatemia in humans, is characterized by renal defects in phosphate reabsorption and vitamin D metabolism. In addition, the renal adaptive response to phosphate deprivation in mutant Hyp mice differs from that of normal littermates. While Hyp mice fed a low phosphate diet retain the capacity to exhibit a significant increase in renal brush-border membrane sodium-phosphate cotransport in vitro, the mutants fail to show an adaptive increase in maximal tubular reabsorption of phosphate per volume of glomerular filtrate (TmP/GFR) in vivo. Moreover, unlike their normal counterparts, Hyp mice respond to phosphate restriction with a fall in the serum concentration of 1,25-dihydroxyvitamin D [1,25(OH)2D] that can be ascribed to increased renal 1,25(OH)2D catabolism. The dissociation between the adaptive brush-border membrane phosphate transport response and the TmP/GFR and vitamin D responses observed in Hyp mice is also apparent in X-linked Gy mice and hypophysectomized rats. Based on these findings and the notion that transport across the brush-border membrane reflects proximal tubular function, we suggest that the adaptive TmP/GFR response requires the participation of 1,25(OH)2D or a related metabolite and that a more distal segment of the nephron is the likely target for the 1,25(OH)2D-dependent increase in overall tubular phosphate conservation.