Liver X receptor-activating ligands modulate renal and intestinal sodium–phosphate transporters

Abstract

Cholesterol is pumped out of the cells in different tissues, including the vasculature, intestine, liver, and kidney, by the ATP-binding cassette transporters. Ligands that activate the liver X receptor (LXR) modulate this efflux. Here we determined the effects of LXR agonists on the regulation of phosphate transporters. Phosphate homeostasis is regulated by the coordinated action of the intestinal and renal sodium–phosphate (NaPi) transporters, and the loss of this regulation causes hyperphosphatemia. Mice treated with DMHCA or TO901317, two LXR agonists that prevent atherosclerosis in ApoE or LDLR knockout mice, significantly decreased the activity of intestinal and kidney proximal tubular brush border membrane sodium gradient-dependent phosphate uptake, decreased serum phosphate, and increased urine phosphate excretion. The effects of DMHCA were due to a significant decrease in the abundance of the intestinal and renal NaPi transport proteins. The same effect was also found in opossum kidney cells in culture after treatment with either agonist. There was increased nuclear expression of the endogenous LXR receptor, a reduction in NaPi4 protein abundance (the main type II NaPi transporter in the opossum cells), and a reduction in NaPi co-transport activity. Thus, LXR agonists modulate intestinal and renal NaPi transporters and, in turn, serum phosphate levels.

Increase in serum inorganic phosphate (Pi) concentration (hyperphosphatemia) is associated with endothelial dysfunction¹ and increased incidence of cardiovascular disease,² including accelerated atherosclerosis,³ vascular stiffness,⁴ and vascular calcification.^2,5–7 We have recently found that hyperphosphatemia in vivo and increase in extracellular Pi in vascular smooth muscle cells grown in cell culture induce lipid accumulation and vascular calcification, further emphasizing a role for Pi in vascular disease.⁸ Serum Pi concentration is determined by coordinated activity of the renal and intestinal sodium-gradient-dependent Pi (Na-Pi) transporters.^9–11 In the renal proximal tubule, at least three different phosphate transporters are expressed in the brush border membrane: type II NaPi-2a and NaPi-2c, and type III PiT-2.^12,13 Interestingly, both type III NaPi transporters, PiT-1 and PiT-2, are expressed in mouse ileum; however, a third type II NaPi transporter (NaPi-2b) is considered to be the main transporter that mediates phosphate absorption in the gut.^14,15 Novel pathways that can inhibit renal and intestinal Na-Pi transporters and prevent hyperphosphatemia, especially in the presence of chronic kidney disease, are likely to have important effects in the inhibition of hyperphosphatemia-mediated cardiovascular disease.

The nuclear receptors are involved in the regulation of essential metabolic functions, including glucose and lipid metabolism, reverse cholesterol transport, and inflammation.^16–18 All of these factors have an important role in the development of cardiovascular disease. Activation of liver X receptor (LXR), a nuclear receptor, has been shown to prevent the development of atherosclerosis in ApoE-knock-out¹⁹ and low-density lipoprotein receptor-knockout²⁰ mice.^21,22 In addition, LXR activation reduces the expression of several genes, iNOS, COX2, MMP9, IL-1β, and IL-6, which are mediators of inflammation and atherosclerosis.^23,24 LXR is present in two different isoforms. LXRα (NR1H3) is mostly expressed in liver, intestine, kidney, spleen, macrophages, and adipose tissue. The second isoform LXRβ (NR1H2) is ubiquitously expressed.²⁵ LXRs belong to a family of the type II nuclear receptors, which form heterodimers with the retinoid X receptor and, on ligand binding, stimulate the expression of target genes.^26,27 Recently, the oxidized cholesterol derivatives (oxysterols) have been identified as their natural ligands for LXR.²⁸ Oxysterols are formed in amounts proportional to the cholesterol content in the cell; therefore, LXRs operate as cholesterol sensors, which protect from cholesterol overload by inhibiting intestinal cholesterol absorption.^19,29 LXRs stimulate cholesterol efflux from cells via the activation of adenosine-triphosphate-binding cassette (ABC) transporters for the subsequent transport of cholesterol to the liver, conversion to bile acids, and biliary excretion.^27,30,31 However, some synthetic non-steroidal LXR agonists (TO901317 and GW3965) have shown to induce lipogenesis mainly through the activation of sterol-regulatory-element-binding protein 1c, a master regulator of lipids. In contrast to this, our group²² and others³² have demonstrated that a new steroidal LXR ligand, N,N-dimethyl-3β-hydroxy-cholenamide (DMHCA), activates the ABC transporters that mediate reverse cholesterol transport but does not activate lipogenesis.

In addition to the liver and the intestine, LXRα and LXRβ are also highly expressed in the kidney.^33,34 Although LXR agonists have been shown to increase the activity of the intestinal and renal ABC cholesterol transporters ABCA1 and ABCG1,^35,36 their potential effect in the modulation of intestinal and renal Na-Pi transporters have not been studied.

In this study, we document a novel role for the LXR-activating ligands, DMHCA and TO901317, in the inhibition of the major renal and intestinal Na-Pi transporters, resulting in a decrease of serum phosphate levels. This study along with our previous findings of reduction of atherosclerosis, suggests that LXR-activating ligands, such as DMHCA, capable of inducing reverse cholesterol transport without the lipogenic effects might be a promising therapeutic agent in the prevention of hyperphosphatemia and its cardiovascular consequences.

RESULTS

The effects of the LXR agonists on renal and intestinal gene regulation

In our initial studies, we determined the effects of both T0901317 and DMHCA on potential LXR targets in the kidney and the intestine.²² We found that both T0901317 and DMHCA increased ABCA1 and ABCG1 mRNA abundance in the kidney and the ileum (Figure 1). As previously shown in hepatocytes and macrophages,³² T0901317 also increased sterol-regulatory-element-binding protein 1c, FAS, and SCD-1 mRNA abundance in the kidney and the ileum; however, the effects of DMHCA on sterol-regulatory-element-binding protein 1c, FAS, and SCD-1 were minimal (Figure 1). These results were also in agreement with the previous data, showing the activation of LXR target genes in the kidney when the mice were treated with TO901317.³⁷ Upregulation of ABCA1 and stearoyl-CoA desaturase 1 protein in kidney and ileum was also confirmed by western blotting and immunofluorescence microscopy (Figure 1c–e). No significant effects were observed in the expression of carbohydrate-responsive-element-binding protein and liver pyruvate kinase.

Figure 1. Effect of the LXR agonist DMHCA and TO901317 on the abundance of LXR target genes in mouse kidney and ileum.

(a, b) LXR target gene mRNA abundance in kidney and ileum was analyzed by real-time quantitative PCR. TO901317 induced significant increases in the mRNA abundance of ABCA1 and ABCG1 as well as SREBP1c, FAS, and SCD1. Increases in the mRNA abundance of these genes with DMHCA were lower than activation by TO901317, especially activation of lipogenic genes, such as SREBP1c, FAS, and SCD1. DMHCA or TO901317 did not activate ChREBP and LPK, neither in kidney nor in ileum. (c) ABCA1 and SCD1 protein expression in mouse kidney was analyzed by western blotting. TO901317 induced a more significant upregulation of both of these proteins in kidney, which was also confirmed by immunofluorescence. (d, e) The expressions of ABCA1 and SCD1 were also increased in mouse ileum. Protein increase of SCD1 was not observed with DMHCA. Values represent means ± s.e.m., at least n = 6 mice per group. ChREBP, carbohydrate-responsive-element-binding protein; DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; LPK, liver pyruvate kinase; LXR, liver X receptor; NS, nonsignificant; SCD1, stearoyl-CoA desaturase 1; SREBP1c, sterol-regulatory-element-binding protein 1c.

Treatment with DMHCA or TO901317 causes decreases in Na⁺-dependent phosphate uptake in kidney and ileum brush border membrane (BBM)

In kidney BBM, sodium-dependent Pi uptake was reduced by 20% and 15% in the DMHCA- and TO901317-treated mice, respectively. Treatment with DMHCA had even more marked effects in the ileum where sodium-dependent Pi uptake was reduced by 56% and by 51% when mice were treated with TO901317 (Figure 2). In all cases, sodium-gradient-independent Pi transport was measured by using choline chloride rather than sodium chloride. An average of 5% of total uptake in the kidney BBM and an average of 14% of the total uptake in the ileum BBM was due to Na-independent Pi uptake, and this Na-independent component of total uptake was modified by neither DMHCA nor TO901317. Therefore, both drugs are inhibiting the transport of Pi in both epithelia.

Figure 2. Treatment with DMHCA or TO901317 reduces Na⁺-dependent phosphate uptake in mouse ileum and kidney BBM.

(a) Sodium-dependent ³²P uptake was reduced in ileum BBM in DMHCA- or TO901317-treated mice. (b) Sodium-dependent ³²P was also reduced in kidney BBM in DMHCA- or TO901317-treated mice. Small panels in the upper right show sodium-independent uptake, measured in presence of Cl-choline, compared with the total phosphate uptake (NaCl). These values represent the average of two different experiments with at least n = 10 per group. BBM, brush border membrane; DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; NaPi, sodium–phosphate; Pi, inorganic phosphate.

The DMHCA- or TO901317-induced decrease in renal and intestinal NaPi transport activity was paralleled by an increase in urinary Pi excretion and a small but significant decrease in serum Pi concentration (Figure 3). Changes in the urinary glucose or protein excretion were not detectable after treatment; however, a small significant decrease in the urine pH was detected in the treated mice (Supplementary Figure A online).

Figure 3. Treatment with DMHCA or TO901317 decreases blood phosphate concentration, increases urine phosphate excretion in mouse, and does not change serum calcium concentration.

(a) Treatment with TO901317 caused a 20% decrease in serum phosphate concentration, with a smaller reduction of 14% after treatment with DMHCA. (b) Approximately a 30% increase in urine phosphate excretion with either compound. (c) No changes in the serum calcium concentration were observed with either compound. At least n = 10 mice per group. DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; NS, nonsignificant; Pi, inorganic phosphate.

Treatment with DMHCA or TO901317 decreases serum Pi and increases serum FGF23

To determine the serum levels of Pi, Ca, FGF23, and parathyroid hormone, blood was collected when animals were killed. Treatment with TO901317 caused a 20% decrease in serum phosphate concentration, with a smaller reduction of 14% after treatment with DMHCA (Figure 3a). This is associated with an increase of ~30% in the urine phosphate excretion with either compound (Figure 3b). No significant changes in the serum calcium concentration were observed (Figure 3c). Additionally treatment with TO901317 caused a 64% increase in serum FGF23 concentration, with a smaller increase of 46% after treatment with DMHCA (Figure 4a). No significant changes in the serum parathyroid hormone levels were observed after treatment with either compound (Figure 4b).

Figure 4. Effects of DMHCA and TO901317 on mouse serum FGF23 and serum PTH.

(a) Treatment with TO901317 caused a 64% increase in serum FGF23 concentration, with a smaller increase of 46% after treatment with DMHCA. (b) Changes on the serum PTH levels were determined to be not significant for both compounds. At least n = 10 mice per group. DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; FGF23, fibroblast growth factor 23; NS, nonsignificant; PTH, parathyroid hormone.

Treatment with DMHCA or TO901317 causes decreases in NaPi cotransporter protein and mRNA abundance in kidney and ileum

To determine the mechanism of the LXR-agonist-mediated decrease in renal BBM NaPi cotransport activity, we determined the abundance of BBM NaPi cotransporters by western blotting. We found that treatment with DMHCA or TO901317 caused significant decreases in the protein abundance of all of the three renal transporters, namely, NaPi-2a, NaPi-2c, and Pit-2 (Figure 5a). The effect of these compounds on the renal NaPi transporter abundance was independent of alterations in the protein abundance of the PDZ-domain-interacting proteins, namely, NHERF-1 and PDZK-1 (Figure 5c). These proteins are well known for the regulation of the renal phosphate transporters. In addition, we found that DMHCA did not alter renal Na-glucose SGLT-2 protein levels, whereas TO901317 caused a decrease of SGLT-2 protein abundance (Supplementary Figure C online).

Figure 5. Effects of DMHCA or TO901317 on renal BBM NaPi transporter protein abundance and NaPi transporter mRNA abundance.

(a) Mouse kidney brush border membrane (BBM) vesicles were isolated after treatment with DMHCA or TO901317. A significant decrease in protein abundance was observed in all three NaPi transporters: 57% for NaPi-2a, 40% for NaPi-2c, and 30% for PiT-2. (b) NaPi transporter mRNA abundance was measured by quantitative PCR and normalized against cyclophilin A. Decreases in NaPi-2a and NaPi-2c mRNA levels were observed with either DMHCA or TO901317. (c) NHERF1 and PDZK1 protein levels were not affected by these LXR agonists. At least n = 6–8 mice per group. BBM, brush border membrane; DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; NHERF1, Na/H exchange regulatory factor-1; NaPi, sodium-phosphate; NS, nonsignificant.

Measurements of the mRNA abundance of these transporters in parallel samples of the kidney cortex by real-time quantitative PCR indicate that these ligands decrease the mRNA abundance of NaPi-2a and NaPi-2c but not Pit-2 (Figure 5b).

Treatment with DMHCA or TO901317 also caused a significant decrease in ileum BBM NaPi-2b protein abundance, with no significant effects on Pit-1 protein abundance (Figure 6a). The effect of these ligands on the intestinal NaPi-2b transporter abundance was independent of the alterations in the protein abundance of the PDZ-domain-interacting proteins, namely, NHERF-1 and PDZK-1 (Figure 6c). In addition, we found no effects of these LXR agonists on the intestinal Na-glucose SGLT1 transporter protein levels (Supplementary Figure B online).

Figure 6. Effects of DMHCA or TO901317 on intestinal BBM NaPi transporter protein abundance and NaPi transporter mRNA abundance.

(a) Mouse ileum brush border membrane (BBM) vesicles were isolated after treatment with DMHCA or TO901317. There was a significant 61% decrease in the protein abundance of the major intestinal NaPi transporter NaPi-2b with TO901317 treatment, and a 52% after treatment with DMHCA, while there were no significant changes in type III NaPi transporter PiT1 protein expression. (b) NaPi2b mRNA abundance was reduced by 50% after treatment with DMHCA, and by 75% with TO901317. (c) NHERF1 and PDZK1 protein levels were not significantly affected by DMHCA or TO901317. At least n = 6 mice per group. BBM, brush border membrane; DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; NHERF1, Na/H exchange regulatory factor-1; NaPi, sodium-phosphate; NS, nonsignificant.

The effect of DMHCA or TO901317 on NaPi-2b protein abundance was associated with a parallel decrease in NaPi-2b mRNA; 75% after treatment with TO901317 and 50% with DMHCA (Figure 6b).

DMHCA or TO901317 causes decreases in NaPi cotransport activity and NaPi-4 protein abundance in opossum kidney (OK) cells in culture

To determine whether DMHCA or TO901317 has direct modulatory effects on NaPi cotransport activity, independent of systemic metabolic and hormonal factors, we studied their effects in OK cells. OK cells are a well-established model of the renal proximal tubule, which expresses the endogenous type IIa NaPi cotransporter, also known as NaPi-4. Treatment of OK cells with DMHCA induced translocation of LXR to the nucleus (Figure 7a). This effect was also observed with TO901317 compound (data not shown). Both LXR agonists caused a dose-dependent decrease in OK cell NaPi cotransport activity (Figure 7b), measured by whole-confluent cells ³²P uptake. Western blot of apical membranes isolated from OK cells and immunofluorescence studies indicate that these agonists caused parallel decreases in the apical membrane NaPi-4 protein abundance (Figure 7c, d and Supplementary Figure D online).

Figure 7. Treatment of opossum kidney (OK) cells with LXR agonists DMHCA or TO901317 induces activation of endogenous LXR, decrease in the phosphate uptake in a dose-dependent manner, and decrease in the expression of the endogenous NaPi transporter (NaPi4).

(a) Activation and translocation of the endogenous LXR nuclear receptor is shown by immunofluorescence after incubation of the OK cells with DMHCA. Notice the increased red signal inside the nucleus in the treated cells. (b) Correlation between concentration of the LXR agonist and reduction of the whole cells ³²P uptake was observed for both compounds. (c) Reduced expression of the endogenous NaPi4 phosphate transporter is observed by immunofluorescence in the apical membrane of OK cells. Notice the decrease in the red signal, and (d) this is confirmed by western blot in isolated OK cell BBM. At least n = 3 per group. BBM, brush border membrane; DMHCA, N,N-dimethyl-3β-hydroxy-cholenamide; NaPi, sodium-phosphate; NS, nonsignificant; OK, opossum kidney; Pi, inorganic phosphate; SCD1, stearoyl-CoA desaturase 1.

DISCUSSION

Hyperphosphatemia is a major risk factor for cardiovascular disease. Any interventions that decrease serum Pi concentration and possibly prevent the cardiovascular consequences of hyperphosphatemia are welcomed. In this study, we show that TO901317 and DMHCA, two LXR-activating ligands, which have been previously described and shown to prevent atherosclerosis in ApoE-knockout mice,²² also causes a significant decrease in serum Pi concentration by inhibiting the activity of the renal and intestinal NaPi transporters.

The effects of TO901317 or DMHCA in decreasing renal proximal tubular BBM NaPi cotransport activity are reflected by an increase in the urinary Pi excretion and significant decreases in the abundance of NaPi-2a, NaPi-2c, and Pit-2. Although the decreases in NaPi-2a and NaPi-2c protein abundance may be mediated by the transcriptional mechanisms, the decrease in Pit-2 protein abundance seems to be independent of Pit-2 transcriptional regulation. These results are associated with a significant increase of the FGF23 serum levels after treatment with LXR agonist. FGF23 is a well-known phosphaturic hormone³⁸ capable of downregulating the expression of the renal NaPi transporters. In addition, it is also important to mention that the protein levels of the Na-glucose transporters SGLT2 in kidney BBM are also reduced after treatment with TO901317, with no significant effects with DMHCA.

It is now quite well established that NaPi–PDZ-type (PSD-95, discs-large, and ZO-1) protein interactions are important for the regulation of NaPi-2a and NaPi-2c protein expression in the proximal tubular apical BBM.^39–41 Our studies indicate that the effects of these LXR agonists on NaPi-2a, NaPi-2c, and NaPi-2b protein abundance are independent of alterations on BBM expression of the PDZ proteins, NHERF-1 or PDZK-1. However, LXR-induced modifications in NHERF-1 or PDZK-1–NaPi interactions cannot be ruled out.

DMHCA and TO901317 also have marked effects in inhibiting intestinal BBM NaPi cotransport activity. This inhibition occurs via a major decrease in the BBM protein abundance of the major intestinal NaPi transporter NaPi-2b. There is also a parallel decrease in NaPi-2b mRNA abundance, which indicates that DMHCA and TO901317 may downregulate NaPi-2b via transcriptional mechanisms.

To determine whether LXR activation has also direct effects in modulating NaPi cotransport activity, we have performed parallel studies in OK cells, a well-established model of the renal proximal tubule.^39,42 We found that both of the LXR agonists cause a dose-dependent decrease in OK cells NaPi cotransport activity by decreasing the apical BBM abundance of the OK cell type II NaPi cotransporter NaPi-4 protein. Additional studies indicate that DMHCA as well as TO901317 induce increased nuclear expression of LXR protein, further supporting that these drugs are LXR-activating ligands; this is correlated with the upregulation of LXR target genes, including ABCA1 (data not shown), after treatment of the OK cells with this agonist.

Our study demonstrates that LXR has a novel role in inhibiting renal and intestinal NaPi transporters and decreasing serum Pi concentration through multiple mechanisms, including phosphatonin modulation and direct control of Pi transporter abundance in epithelia. This effect of LXR, along with its well-established effects in mediating reverse cholesterol transport, inhibiting inflammatory cytokines, and preventing atherosclerosis, establishes it as a major target for the prevention of hyperphosphatemia and the associated cardiovascular complications.

MATERIALS AND METHODS

Animals and diets

Male C57Bl/6 mice were purchased from Jackson Laboratory (Bar Harbor, ME) and maintained in a clean environment on a regular 12-h light–12-h dark cycle. Before the initiation of the corresponding diets, mice were kept on a standard laboratory chow diet (Harland Teklad 2019 chow diet) with 0.9% of Ca and 0.7% of Pi. Male C57Bl/6 mice were fed chow diet containing (a) no ligands, or (b) DMHCA (80 mg/kg body weight/day), or (c) T0901317 (35 mg/kg body weight/day) for 15 days. Diets were supplemented with the respective LXR ligand at a level sufficient to provide the appropriate mg/kg food dose on consumption of a 5 g diet by a 25 g mouse/day. Body weight and food intake were monitored regularly. We studied n = 24 mice in each treatment group: 12 mice for renal and intestinal BBM isolation and 12 mice for renal and intestinal RNA isolation. On the 14th day of the treatment, the mice were placed in metabolic balance cages for urine collection. Animal experiments were approved by the Institutional Animal Care and Research Advisory Committee of the University of Colorado at Denver.

Cell culture

OK proximal tubule cells were grown in DMEM-F-12 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 50 U/ml penicillin, and 50 µg/ml streptomycin. For experimental work, cells were seeded on porous membrane inserts (Corning, Lowell, MA). After confluency, cells were placed in DMEM-F-12 supplemented with 0.2% fetal bovine serum and penicillin/streptomycin to get them quiescent for 24–48 h before treatment. Cells were treated with different concentration of LXR agonist, either DMHCA or TO901317 (stocks were resuspended in dimethyl sulfoxide). Working solutions of these agonists were prepared in DMEM-F-12 supplemented with 0.2% fetal bovine serum and penicillin/streptomycin. Cells were treated with 1:1000 dilution of dimethyl sulfoxide (control) or DMHCA or TO901317 for 24 h.

Materials and antibodies

All chemicals were obtained from Sigma (Saint Louis, MI), except when noted. A polyclonal rabbit anti-NaPi-IIa antibody was generated by Affinity Bio Reagents (Golden, CO) and used at 1:5,000 for western blotting.¹³ A rabbit anti-NaPi-IIc antibody was custom-made by Davids Biotechnologie (Regensburg, Germany), as previously described,¹³ and was used at 1:1000 for western blotting. The polyclonal rabbit anti-NaPi-2b, anti-PiT1, and anti-PiT2 antibodies were also custom generated by Davids Biotechnologie (Regensburg, Germany) as described before,¹⁴ and was used at 1:1,000 dilution for western blotting. The goat anti-LXRα/β was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The rabbit anti-ABCA1 was purchased from Novus Biologicals (Littleton, CO), and the rabbit anti-stearoyl-CoA desaturase 1 was purchased from Cell Signaling Technology (Danvers, MA). The rabbit anti-Na/H exchange regulatory factor-1 antibody was purchased from Sigma. The rabbit anti-PDZK1 was a kindly gift from Dr David Silver (Columbia University).

BBM vesicle isolation

Mice were anesthetized via an intraperitoneal injection of 50 mg/kg pentobarbital sodium (Pentothal, Abbott Laboratories, Abbott Park, IL). After clamping of the renal vessels, blood was drawn for biochemical analysis, the kidneys and the ileum were removed, and the ileum mucosa was scraped for BBM isolation.

Kidney slices from two mice were combined in 7.5 ml isolation buffer consisting of 15 mmol/l Tris · HCl (pH 7.4), 300 mmol/l mannitol, 5 mmol/l ethylene glycol tetraacetic acid, and 1 Roche Complete inhibitor tablet per 250 ml buffer. The kidney slices were homogenized using a Potter-Elvejham homogenizer with 8–10 rapid strokes and transferred to a chilled capable tube. Kidney residues remaining on the homogenizer were rinsed off with 10 ml water that was then added to the kidney homogenate. BBM was prepared by a double Mg²⁺ precipitation. For the first Mg²⁺ precipitation, MgCl₂ was added to the homogenate (final concentration of 15 mmol/l), and the homogenate was shaken every 5 min on ice for 20 min before centrifugation at 2500 g for 15 min. The supernatant was subjected to a second Mg²⁺ precipitation; and from the resulting supernatant, the BBM was recovered by centrifugation at 38,000 g for 40 min. The BBM was resuspended, and its protein content quantified.

Ileum BBM was similarly isolated by double Mg²⁺ precipitation as describe above. BBM of the OK cells was isolated by Mg²⁺ precipitation. Briefly, OKP cells were grown to confluence in 100 mm dishes. At 24 h before the experiment, the cells were placed in DMEM medium containing 0.2% fetal bovine serum to synchronize them. Cells were incubated with control media (1:1000 dimethyl sulfoxide) or DMHCA or TO901317 for 24 h. After treatment, the cells were washed in ice cold PBS and scraped into isolation buffer (15 mmol/l Tris (pH7.4), 300 mmol/l mannitol, 5 mmol/l ethylene glycol tetraacetic acid, and one Mini-Complete tablet (Roche)) on ice. The cells were then homogenized by aspirating 30 times through a 23-gauge needle. MgCl2 was added to a final concentration of 15 mmol/l, and the homogenate was shaken on ice for 20 min. The homogenate was centrifuged at 2500 × g at 4 C for 15 min. The supernatant was removed and spun at 60,000 × g for 40 min. The final pellet was resuspended in isolation buffer.

Pi transport assays

Phosphate transport from kidney or ileum was measured by rapid filtration of radioactive ³²Pi uptake in freshly isolated BBM vesicles.⁴³ The BBM and the uptake solution were incubated for 10 s at 25°C for kidney and 30 s at 37°C for ileum. Phosphate transport in OK cells was measured by radioactive. ³²Pi uptake in treated confluent cells for 6 min at 25°C, as described.^14,40,44

Urine and blood analysis

Blood samples were collected in heparin-containing tubes during sacrifice. The 24 h and spot urine was collected in animals treated for 2 weeks. Plasma obtained after centrifugation and urine samples were analyzed for phosphate (Pi) concentrations by using the commercial kit Stanbio Liqui-UV (Stanbio; Boerne, TX). Creatinine concentration in urine was determined using QuantiChrom Creatinine Assay (BioAssay Systems; Hayward, CA). FGF-23 (C-Term) and intact parathyroid hormone were determined with specific ELISA kits from Immunotopics (San Clemente, CA). n = 10–12 animals per group was used in these assays.

Western blotting

BBM proteins (20 or 30 µg) were separated by 10% SDS-polyacrylamide gel electrophoresis, and transferred onto nitrocellulose membranes. Membranes were blocked with 5% milk in PBS Tween 20 before incubation with primary antibodies diluted in PBTS overnight at 4°C. After washes with phosphate buffered saline and Tween 20, membranes were incubated with Licor-conjugated (LI-COR, Lincoln, NE) donkey secondary antibodies diluted 1:5,000 for 1 h. Membranes were scanned using Licor system. Densitometry data are presented as average ± s.d.

RNA extraction and real-time quantitative PCR

Total RNA was isolated from kidney cortex and ileum using the Qiagen RNeasy Mini Kit, and complementary DNA was synthesized using reverse transcription reagents from Bio-Rad (Hercules, CA). The mRNA level was quantified using a Bio-Rad iCyCler real-time PCR machine. Cyclophilin A was used as an internal control, and the amount of RNA was calculated by the comparative threshold cycle method as recommended by the manufacturer. All of the data were calculated from duplicate reactions of three different experiments.

Confocal microscopy

Cells were washed with PBS before blocking for 30 min with 5% goat serum and permeabilized with 0.1% saponin in PBS. Cells were then incubated overnight at 4°C with primary antibodies. After washing with saponin solution, sections were incubated with secondary fluorescent goat antibodies for 1 h. After washing three times, the cells were mounted in Vectashield (Vector Labs, Burlingame, CA). Confocal images were acquired on a Zeiss 510 NLO-META LSM laser scanning confocal microscope (Carl Zeiss, Thornwood, NY), and the Olympus Fluoview 1000 confocal microscope (Olympus, Center Valley, PA).

Statistical analysis

Data are expressed as means ± s.d., *P<0.05, **P<0.005, and ***P<0.001. Data were analyzed for statistical significance by unpaired Student’s t-test or one-way analysis of variance.