Main

The neonatal proximal tubule has a lower rate of volume absorption than that of the adult segment(1–3). Transport rates for bicarbonate, glucose, and amino acids are all substantially less in the immature proximal tubule than that in the adult(2–5). However, there is substantial evidence that there is a maturational decrease in proximal tubule phosphate transport(6–11). The maximal rate of phosphate transport per volume of GFR in neonatal rats is higher than that of adult rats(7–9). In the isolated perfused guinea pig kidney there is a higher rate of phosphate transport per volume of GFR in neonates than in that of adults indicating an inherently higher rate of transport by the neonatal kidney(10). BBMVs from neonatal guinea pigs have a higher Vmax than adults consistent with a higher rate of neonatal proximal tubule phosphate transport(11).

The serum glucocorticoid concentration increases during the perinatal period(12–14). This rise in glucocorticoids may play an important role in the maturational increase in proximal tubule volume, glucose, and bicarbonate transport(1, 2, 15–19). We have recently provided evidence that glucocorticoids may play a role in the maturational decrease in proximal tubule phosphate transport(16). Although we found that administration of glucocorticoids to neonatal rabbits resulted in a decrease in BBMV sodium-dependent phosphate transport, there was no change of the Na-Pi cotransporter density as assessed by phosphate-protectable Na-dependent equilibrium binding of phosphonoformic acid(16).

Two families of the Na-Pi cotransport have recently been cloned from kidney cortex(20–22), which have been designed NaPi-1(20) and NaPi-6(22) in the rabbit. NaPi-1 and NaPi-6 belong to two different families of Na-Pi cotransporters(22, 23); NaPi-6 unlike NaPi-1, is regulated by changes in dietary phosphate intake and has functional properties consistent with the BBM Na-Pi cotransporter in the rabbit(22). The purpose of the present study was to examine the maturation of NaPi-6 mRNA and protein abundance and examine whether glucocorticoids modulate the abundance of NaPi-1 and NaPi-6 mRNA and protein abundance in neonate renal cortex.

METHOD

Animals. New Zealand white rabbit pregnant does were housed at our institution for 1 wk before their expected date of delivery. Neonatal rabbits were cared for by their mothers. At 3-5 d of life they received either daily s.c. injections of dexamethasone (10 μg/100 g of body weight) or vehicle for 4 d, including a dose 2 h before they were killed. These doses of dexamethasone were pharmacologic and resulted in higher concentration of glucocorticoids than that observed at this age. Dexamethasone was dissolved in 150 mmol/L NaCl and 2 mmol/L K2HPO4 (pH 7.5). In experiments designed to examine the maturation of the NaPi-6 mRNA and protein abundance, 4-wk-old rabbits were weaned and placed on rabbit chow for 1 wk before study. These young rabbits were compared with 4-5 kg adult rabbits.

BBMV isolation. Control and dexamethasone-treated neonatal rabbits were killed, and the kidneys were rapidly removed and placed in an ice-cold isolation buffer containing (in millimoles/L) 300 mannitol, 15 HEPES, 5 EGTA, titrated to pH 7.5 with Tris. The cortex was homogenized with a Teflon glass Potter Eljevhem homogenizer. BBMV were then isolated by differential centrifugation and magnesium precipitation as previously described(16, 24, 25). The final BBMV fraction was resuspended in buffer containing (in millimoles/L) 300 mannitol, 16 HEPES, titrated to pH 7.5 with Tris. The protein concentration was measured using the Lowry method with crystalline BSA as the standard(26). To minimize potential day to day variation in BBMV isolation procedure, we isolated BBM from the kidneys of control and dexamethasone-treated neonatal rabbits simultaneously each day. Brush border enzyme activity measurements for alkaline phosphatase and γ-glutamyltransferase were performed as previously described in our laboratory(16, 24, 25).

BBMV transport. Sodium gradient-dependent phosphate transport was measured in freshly isolated BBMVs by radiotracer uptake followed by rapid millipore filtration as previously described(16, 24, 25). Ten microliters of BBMV preloaded in an intravesicular buffer of (in millimoles/L) 300 mannitol, 16 HEPES, pH 7.5 with Tris, were vortex mixed at 25°C with 40 μL of extravesicular uptake buffer of (in millimoles/L) 150 NaCl, 16 HEPES, pH adjusted to 7.5 with Tris, and 25 μmol/L K2H32PO4. Uptake after 10 s (representing the initial linear rate) was terminated by the addition of an ice-cold stop solution. All uptake measurements were performed in triplicate, and the uptake was calculated on the basis of specific activity determined in each experiment and expressed as picomoles/10 s/mg of protein. Sodium-independent 32P uptake was studied in experiments where sodium was replaced by choline and was found to be negligible.

RNA isolation and Northern blotting. Renal cortex from control and dexamethasone-treated neonates were homogenized in RNazol [1:1, phenol-RNazol stock (4 M guanidinium thiocyanate, 25 mmol/L disodium-citrate, pH 7.0), 0.5% Sarcosyl] containing 3.6 mmol/L β-mercaptoethanol. RNA was extracted using 3 M NaOAc (pH 4.0) and chloroform, purified using isopropanol precipitation, and washed twice with 75% ethanol(27). Poly(A)+ RNA was purified using oligo(dT) column chromatography. Five micrograms of poly(A)+ RNA were fractionated by agarose-formaldehyde gel electrophoresis and transferred to a nylon filter (GeneScreen Plus; Dupont NEN, Boston, MA). The filter was prehybridized at 42°C for 4 h with 5× SSC, 5 × Denhardt's solution (Ficoll, BSA, and polyvinylpyrrolidone, each at 1 mg/mL), 0.5% SDS, and 0.5 mg/mL of sheared salmon sperm DNA, then hybridized to double-stranded uniformly32 P-labeled cDNA probes (>106 counts/min/mL) in the above hybridization solution at 42°C for 16 h. Full-length cDNA probes for NaPi-6(22), NaPi-1(20), and GAPDH(28) were labeled by a random hexamer method (oligo labeling kit; Pharmacia Biotech Inc., Piscataway, NJ) using [α-32P]dCTP. Blots were washed three times with 2 × SSC/0.1% SDS at room temperature for 5 min and twice with 0.1% SSC/1% SDS at 55°C for 40 min. Message abundance was quantitated using autoradiography and densitometry.

SDS-PAGE and immunoblotting. BBMs were denatured and separated on a 7.5% polyacrylamide gel as previously described(17, 29). The proteins were transferred to polyvinylidiene difluoride membrane overnight at 140 mA at 4°C. The blot was blocked with fresh Blotto (5% nonfat milk, 0.1% Tween 20, and PBS, pH 7.4) for 4 h, and then a primary antibody to NaPi-1(30) or NaPi-6(21, 22, 31) was added at a 1:4000 dilution and incubated for 8-12 h at 4°C. The blot was washed with Blotto, and then the secondary horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin was added for 1 h at 1/10,000 dilution. The blot was washed with Blotto, and enhanced chemiluminescence was used to detect bound antibody (DuPont NEN). The Na-Pi protein abundance was quantitated using densitometry.

Statistical analysis:. All data are expressed as means ± SEM. An unpaired t test was used to compare results between control and dexamethasone-treated neonatal rabbits.

RESULTS

Previous studies have shown that the rate of renal phosphate transport in young animals is greater than that in adults(6–11). To examine if there was a maturational change in renal NaPi-6 mRNA and protein abundance, we compared five weaned young animals eating rabbit chow for 1 wk with five adults. As shown in Figure 1, NaPi-6 protein abundance was 2-fold higher in young compared with adult animals (100 ± 16 versus 47 ± 7 arbitrary densitometric units, p< 0.05). There was no change in NaPi-6 mRNA abundance.

Figure 1
figure 1

Comparison of BBMV (50 μg) NaPi-6 protein abundance (top bands) and renal cortical NaPi-6 mRNA (20 μg) (middle bands) from young weaned rabbits and adult rabbits.

Full size image

The effect of glucocorticoids on Na-Pi uptake in BBMV from neonates is shown in Figure 2. There was a significant reduction in sodium-dependent phosphate transport, confirming our previous results(16). As in our previous study, there was over a 9-fold enrichment of the BBMV preparation for γ-glutamyltranspeptidase and alkaline phosphatase in the control and dexamethasone group. As previously found, the sodium-independent rate of Pi uptake was insignificant.

Figure 2
figure 2

Effect of dexamethasone on Na-Pi transport in neonatal renal BBMV. Dexamethasone-treated neonates received s.c. daily injections of (10 μg/100 g of body weight) for 4 d starting at 3-5 d of life. Controls received vehicle. *p < 0.05 vs control.

Full size image

The effect of dexamethasone on NaPi-6, NaPi-1, and GAPDH mRNA abundance is shown in Figure 3 and Table 1. There were eight experiments in each group, and NaPi-1 and NaPi-6 mRNA abundance was normalized for GAPDH mRNA abundance in the corresponding lane. As can be seen, dexamethasone had no significant effect on NaPi-1 or NaPi-6 mRNA abundance.

Figure 3
figure 3

Effect of dexamethasone on neonatal NaPi-6, NaPi-1 and GAPDH mRNA abundance from renal cortex. Each lane had 5μg of poly(A)+ RNA. There was no effect of glucocorticoids on NaPi-6 and NaPi-1 mRNA abundance. Con, control;Dex, dexamethasone.

Full size image
Table 1 Effect of dexamethasone on NaPi-6 and NaPi-1 mRNA and protein abundance
Full size table

The effect of glucocorticoids on NaPi-6 and NaPi-1 protein abundance was determined in 12 experiments using Western blot analysis. As is shown in Figure 4 and Table 1, there was no effect of dexamethasone on NaPi-1 protein abundance; however, there was over a 3-fold reduction in NaPi-6 protein abundance in the dexamethasone group (p < 0.001). Thus dexamethasone selectively reduces the NaPi-6 and not NaPi-1.

Figure 4
figure 4

Immunoblot of NaPi-6 and NaPi-1 protein abundance. Samples were 100 μg of renal cortical BBMV from control and dexamethasone-treated rabbits. There was a significant reduction in NaPi-6, but not NaPi-1, protein abundance. Con, control; Dex, dexamethasone.

Full size image

DISCUSSION

Previous studies have shown a higher rate of phosphate transport in neonates compared with adult animals(6–11). This is due to a higher intrinsic rate of neonatal phosphate transport(10, 11). In the present study we found a higher BBMV NaPi-6 protein abundance in young rabbits to account, at least in part, for these maturational changes. There was no change in NaPi-6 mRNA abundance. These findings are similar to maturational changes in Na-Pi mRNA and protein abundance recently described in the rat(32).

There is increasing evidence that the perinatal increase in glucocorticoid concentration plays an important role in proximal tubule maturation. Administration of glucocorticoids to neonates, or to the late gestation fetus, has been shown to increase sodium-dependent glucose and proline transport(16), Na-K-ATPase activity(16, 18, 19, 33, 34), apical membrane Na+/H+ antiporter activity(2, 34), and basolateral membrane Na(HCO3)3 activity(2). The effect of glucocorticoids on the Na+/H+ antiporter is isoform-specific in that NHE-3 mRNA and protein abundance increase with glucocorticoids, but there is no effect on NHE-1(17).

We have recently examined the effect of glucocorticoids on neonatal Na-Pi cotransport activity using the same protocol as that in this study(16). Glucocorticoid-treated neonates had a lower rate of Na-Pi cotransport activity that was due to a decrease in Vmax with no change in the Km for phosphate. Pi-protectable Na-phosphonoformic acid binding was not affected by glucocorticoids. However, in that study we pointed out that phosphonoformic acid has been shown to bind to other anion transporters and was thus only a crude approximation of phosphate transporter density(35). The present study using specific antibodies to NaPi-6 shows that glucocorticoids do decrease Na-Pi transporter density.

In our previous study examining the effect of glucocorticoids on Na-Pi cotransport, we found that glucocorticoids altered lipid composition and membrane fluidity(16). BBMV from glucocorticoid-treated rabbits had a significant decrease in sphingomyelin and increase in phosphatidylcholine and phosphatidylinositol content. BBMs from glucocorticoid-treated animals had a decrease in BBM fluidity assayed using fluorescence anisotropy of 1,6-diphenyl-1,3,5-hexatriene and trimethylammonium-1,6-diphenyl-1,3,5-hexatriene. A decrease in membrane fluidity has been shown to decrease BBM sodium-dependent phosphate transport(24). The relative contributions of the alteration in membrane fluidity and NaPi-6 to the decrease in phosphate transport in dexamethasone-treated animals is unclear, but it is likely from the current data that the decrease in NaPi-6 transporter abundance plays a greater role.

Two Na-Pi transporters have been cloned from rabbit kidney cortex(20, 22). NaPi-1 is localized to the apical membrane of the proximal tubule(30), but its abundance is not affected by maneuvers that are known to regulate Na-Pi cotransporter activity (such as dietary phosphate intake)(22). NaPi-1 has been designated to belong to a type I Na-Pi cotransporter system that has only 20% homology at the amino acid level to the type II systems(36). NaPi-1 likely is an anion channel(37). The type II Na-Pi cotransports are also localized to the apical membrane of the proximal tubules(30) and have similar characteristics when induced in Xenopus oocytes to mammalian BBM Na-Pi cotransporter(21, 38). NaPi-6, the rabbit type II Na-Pi cotransporter, has recently been shown to be up-regulated by low dietary phosphate intake(22).

The effect of dexamethasone has recently been examined in the adult rat(39). Dexamethasone treatment resulted in a decrease in BBMV sodium-dependent phosphate uptake to a comparable degree as found in the present study. As we have previously shown, the decrease was due to a lower Vmax with no change in the Km for phosphate. Dexamethasone-treated rats had a decrease in mRNA and protein abundance for NaPi-2, the rat type II transporter. Thus, the decrease in BBMV Na-Pi cotransport activity was paralleled by a decrease in transporter abundance using a specific antibody.

In the present study we reexamined the mechanism for the reduction in Na-Pi cotransporter activity in neonatal rabbits. Dexamethasone-treated rabbits had a 3-fold reduction in NaPi-6 protein abundance, but no effect on NaPi-6 mRNA abundance. These results differ from the effect of dexamethasone in adult rats where there was a 2.5-fold reduction in both NaPi-2 mRNA and protein abundance(39). The specific effect of dexamethasone NaPi-6 protein abundance could be due to an effect of glucocorticoids on posttranscriptional regulation. Dr. Adrian Spitzer's laboratory has provided evidence for an isoform of the Na-Pi cotransporter which is present during growth(32). This isoform is recognized by the antibody used in this study, and it is possible that glucocorticoids specifically decrease the abundance of this growth-related isoform. However, we did not detect an additional band on our Northern blots, even under low stringency conditions. Thus we have no evidence for this growth-related isoform in rabbits.

In summary, the present study suggests the maturational decrease in phosphate transport is, in part, mediated by a decrease in BBMV NaPi-6 abundance. In addition, the present study reexamined the mechanisms for the glucocorticoid mediated reduction in Na-Pi cotransport in neonatal rabbits. Unlike our previous study which used Na-dependent equilibrium binding of phosphonoformic acid as an assay for Na-Pi protein abundance, the present study demonstrates a reduction in NaPi-6 protein abundance. There was no effect of glucocorticoids on NaPi-6 mRNA abundance or NaPi-1 mRNA and protein abundance. These data are consistent with glucocorticoids playing a role in the postnatal decrease in proximal tubule phosphate transport.