The sodium-dependent renal phosphate transporter (Npt2, Na-Pi IIa) is the major regulated phosphate transporter in the renal proximal convoluted tubule. Npt2 associates with a number of PDZ-containing proteins including Na+H+ exchanger regulatory factor-1 (NHERF-1). To determine whether NHERF-1 is involved in the acute regulation of phosphate transport, wild-type and NHERF-1 (–/–) mice were stabilized on a high-phosphate diet and then acutely changed to a low-phosphate diet. At 24 h after the change to a low-phosphate diet, there was a significant decrease in the urinary excretion of phosphate in both groups but the urinary excretion of phosphate in NHERF-1 (–/–) mice was significantly higher than in wild-type animals (1,097 ± 356 vs. 255 ± 54 ng/min, P < 0.05). Renal mRNA levels and total cellular Npt2 protein did not differ between the animal groups or in response to the changes in diet. Renal brush-border membrane (BBM) expression of Npt2 protein, however, was lower in NHERF-1 (–/–) mice compared with wild-type. In addition, with both the high- and low-phosphate diets, there was increased detection of Npt2 in submicrovillar domains that were particularly prominent in NHERF-1 (–/–) mice compared with wild-type animals. On the other hand, a change from a low-phosphate diet to a high-phosphate diet was associated with a similar increase in the urinary excretion of phosphate in wild-type and NHERF-1 (–/–) animals. These experiments demonstrate that full renal adaptation to a low-phosphate diet requires NHERF-1, which serves to increase BBM expression of Npt2.
- renal electrolyte transport
- PDZ adaptor proteins
- gene deletion
- phosphate metabolism
the renal tubular reabsorption of filtered phosphate is mediated by a family of phosphate transporters in the apical membrane of the renal proximal tubule (11; for a review, see Refs. 8, 12, 17). Elegant studies by Murer and colleagues (2–4) established that the sodium-dependent phosphate transporter IIa (Npt2, NaPi IIa) is the transporter regulated by hormones and changes in the dietary intake of phosphate. By contrast to some other transport proteins and membrane receptors, the activity of Npt2 is not regulated by phosphorylation of reactive sites in the COOH terminus of the protein and Npt2 undergoes limited or no endosomal recycling. Accordingly, the reabsorption of phosphate in the renal proximal tubule is determined by the abundance of Npt2 in the brush-border membrane (BBM) and, in turn, is regulated by metabolic processes that control the trafficking of this transporter.
Animals such as the rat rapidly adjust the urinary excretion of phosphate in response to changes in the dietary intake of phosphate (9). For example, administration of a low-phosphate diet to rats is associated with a rapid decrease in the urinary excretion of phosphate. By 2 h on a low-phosphate diet, there is an increase in BBM expression of Npt2 but no change in Npt2 mRNA. The short-term response is believed to represent either increased translation of Npt2 protein and/or the translocation of a cellular pool of Npt2 to the apical membrane. The mechanisms involved in the rapid translocation of Npt2 to the apical membrane of renal tubular cells are not known, but Biber and colleagues (3) demonstrated that a protein containing multiple PSD-95/Dlg/ZO-1 (PDZ) protein interaction domains, NaPi-Cap1 (Diphor-1), is upregulated in response to dietary restriction of phosphate. Subsequent yeast two-hybrid studies using the COOH terminus of Npt2 as bait revealed the interaction of Npt2 with not only NaPi-Cap1 but also with the PDZ-containing proteins Na+-H+ exchanger regulatory factor-1 and -2 (NHERF-1 and NHERF-2; 5, 19, 20). Additional experiments documented that NHERF-1, which is abundant in the renal proximal tubule BBM, colocalizes with Npt2 and that disruption of this association results in decreased expression of Npt2 in the apical membrane (7). We recently reported initial findings in the NHERF-1 (–/–) mouse (16). In this model, expression of NHERF-2 is the same as in wild-type animals. Nonetheless, the NHERF-1 (–/–) mouse manifests a decrease in the serum concentration of phosphate, an increase in the urinary excretion of phosphate, and a decrease in Npt2 expression in the BBM of the renal proximal tubule while on a normal rodent chow diet. Collectively, these results suggest that NHERF-1 regulates Npt2 expression in the BBM of the kidney proximal tubule.
The role of NHERF-1 in the acute regulation of phosphate metabolism in an intact animal has not been examined due, in part, to the absence of an appropriate experimental model. To begin the study of factors required for the modulation of phosphate transport and Npt2 trafficking, the renal adaptation to changes in the dietary intake of phosphate was determined in wild-type and NHERF-1 (–/–) mice. The results indicate that 24 h on a low-phosphate diet are associated with a rapid decrease in phosphate excretion and an increase in Npt2 expression in the BBM in the absence of changes in Npt2 mRNA in both groups of animals. NHERF-1 (–/–) mice, however, failed to adapt fully to the low-phosphate diet and, compared with the wild-type mice, had higher rates of phosphate excretion in the urine and lower BBM expression of Npt2. These data provide the first evidence in an intact animal that NHERF-1 plays a role in the acute renal response to phosphate deprivation.
Male NHERF-1 (–/–) mice bred into a C57BL/6 background for six generations and parental wild-type inbred control C57BL/6 mice age 12–16 wk were housed in metabolic cages and maintained on 12:12-h light-dark cycles (16). To determine the adaptive response to a low-phosphate diet, animals were stabilized on a high-phosphate diet (1.2%) for 48 h and then changed to a low-phosphate diet (0.1%). Urine was collected during the last 24 h on the high-phosphate diet and while on the low-phosphate diet. When changed to the low-phosphate diet, the initial 4 h of urine were discarded and the ensuing 24 h were collected for analysis. In separate experiments, animals were fed the low-phosphate diet for 24 h and then changed to the high-phosphate diet. Twenty-four-hour urine collections were obtained while on the low- and high-phosphate diets. Urine sodium, phosphate, and creatinine concentrations were determined by automated methods.
BBMs were isolated from mouse kidneys using previously described methods (19). Western immunoblotting was used to determine the abundance of Npt2, ezrin, NHERF-1, NHERF-2, and NaPi-Cap1 in the BBM and Npt2 in whole kidney homgenates. The protein concentrations of the samples were determined, and 10 μg of protein were loaded per lane. The proteins were resolved using 10% SDS-PAGE, transferred to nitrocellulose, and immunoblotted using the indicated antibodies (16, 19). The immunocomplexes were detected using enhanced chemical luminescence (Amersham, Arlington Heights, IL). Densitometry measurements of the Ponseau S-stained transfers and the intensity of ezrin immunoblots were used to ensure equal loading of the lanes. Total RNA was extracted, and Northern blotting was performed for Npt2 mRNA using the rat cDNA and GAPDH as described previously (20). The immune complexes and mRNA bands were quantiated using laser densitometry. For immunocytochemistry, the mice were anesthetized with metofane and the kidneys were perfused-fixed as previously described (16, 18). Appropriate species-specific secondary antibodies were coupled to Alexa 488 or 568 dyes (Molecular Probes) and used in a concentration of 10 μg/ml.
Antibodies to ezrin, Golgi Matrix Protein 130 (GM130), and clathrin were obtained from commercial sources. Well-characterized antibodies to NHERF-1 and NHERF-2 were also used as previously described by this laboratory (16, 18). Dr. M. Knepper provided antibodies to Npt2 and Drs. J. Biber and H. Murer provided antibodies to NaPi-Cap1. Protein concentrations were determined by the method of Lowry et al. (10). Peritz analysis of variance was used for statistical analyses (6).
To determine the role of NHERF-1 in the acute adaptation to a low-phosphate diet, wild-type and NHERF-1 (–/–) animals were initially fed a high-phosphate diet (1.2%) for 48 h. Urine was collected during the second 24-h period. The animals were then changed to a low-phosphate diet (0.1%). In initial experiments, 4-h urine samples were obtained after the change in diet. There was great variability in the urinary output during this 4-h period, and, in most animals, the volumes were too small for analysis by autoanalyzer. Accordingly, urine collected during the initial 4 h on the low-phosphate diet was discarded and the urine excreted over the ensuing 24 h was collected and analyzed. As summarized in Table 1, the urinary volumes did not differ between the groups and were not affected by the diet. The urinary excretion of sodium also did not differ between the groups or by the type of diet. With a high-phosphate diet, there was a tendency toward higher rates of phosphate excretion in the NHERF-1 (–/–) mice but these differences were not statistically significant. There was a rapid decrease in phosphate excretion in wild-type animals when they were changed from the high- to the low-phosphate diet (n = 9), expressed as the rate of phosphate excretion (8,794 ± 772 to 255 ± 54 ng/min, P < 0.05) or the urine phosphate/creatinine ratio (19.1 ± 1.0 to 0.7 ± 0.2, P < 0.05). In a similar manner, the NHERF-1 (–/–) mice (n = 9) also demonstrated a rapid decrease in phosphate excretion when changed from the high- to the low-phosphate diet, expressed as the rate of phosphate excretion (10,006 ± 663 to 1,097 ± 356 ng/min, P < 0.05) or the urine phosphate/creatinine ratio (21.9 ± 1.0 to 2.1 ± 0.5, P < 0.05). However, compared with wild-type animals, NHERF-1 (–/–) animals had significantly higher rates of phosphate excretion (1,097 ± 356 vs. 255 ± 54 ng/min, P < 0.05) and higher urine phosphate/creatinine ratios (2.1 ± 0.5 vs. 0.7 ± 0.2, P < 0.05) 24 h after receiving the low-phosphate diet. The 24-h excretion of creatinine was 0.8 ± 0.1 and 1.0 ± 0.2 mg [P = not significant (NS)] in wild-type mice and 0.9 ± 0.1 and 0.8 ± 0.2 mg (P = NS) in NHERF-1 (–/–) mice on the high- and low-phosphate diets, respectively. These results indicate that NHERF-1 (–/–) mice do not adapt as readily as wild-type mice to an acute decrease in the dietary intake of phosphate over the initial 24 h.
The processes that regulate renal BBM Npt2 expression in response to a low-phosphate diet are likely different from those that mediate adaptation to a high-phosphate diet. To determine whether NHERF-1 (–/–) mice adapted normally to a change from a low-phosphate diet to a high-phosphate diet, animals were fed a low-phosphate diet for 24 h and then changed to the high-phosphate diet. As summarized in Table 2, there were no differences in the urinary volume or sodium excretion between the groups or in response to the diets. The 24-h excretion of creatinine was 0.7 ± 0.1 and 0.8 ± 0.2 mg (P = NS) in wild-type mice and 0.9 ± 0.3 and 0.8 ± 0.1 mg (P = NS) in NHERF-1 (–/–) mice on the low- and high-phosphate diets, respectively. The urinary phosphate excretion was significantly higher in the NHERF-1 (–/–) animals compared with wild-type animals on the low-phosphate diet. There were no significant differences, however, between the wild-type and NHERF-1 (–/–) mice on the high-phosphate diet. Thus, by contrast to the maladaptive response to a low-phosphate diet, the NHERF-1 (–/–) mice increased their urinary excretion of phosphate on a high-phosphate diet appropriately.
Northern blot analysis was used to determine whether there were changes in Npt2 mRNA expression in response to the dietary manipulations. As shown in Fig. 1, Npt2 mRNA abundance was not significantly different in either wild-type or NHERF-1 (–/–) mice receiving the high- or low-phosphate diet for 24 h. Western immunoblots were used to determine Npt2 and ezrin abundance in the BBM and Npt2 abundance in the whole kidney. These results are shown in Table 3. BBM Npt2 abundance did not differ between wild-type and NHERF-1 (–/–) mice on a high-phosphate diet. In response to the low-phosphate diet, both the wild-type mice (903 ± 106 vs. 2,958 ± 258 arbitrary units, P < 0.05) and the NHERF-1 (–/–) mice (683 ± 133 vs. 1,795 ± 311 arbitrary units, P < 0.05) significantly increased expression of Npt2 in the BBM. Our prior studies indicated that total Npt2 abundance and BBM abundance of ezrin were not altered in the NHERF-1 (–/–) mouse (16, 19). This was confirmed herein in that the BBM expression of ezrin and Npt2 expression in the whole kidney homogenate were not different between wild-type and NHERF-1 (–/–) mice and were not influenced by diet in either group.
We also examined BBM abundance of NHERF-1, NHERF-2, and NaPi-Cap1 in wild-type and NHERF-1 (–/–) mice in response to the different phosphate diets. There was no statistical difference between NHERF-1 expression between wild-type animals on a high-phosphate diet (2,991 ± 225 arbitrary units, n = 6) or a low-phosphate diet (3,307 ± 249 arbitrary units, n = 5; P = NS; Fig. 2A). As anticipated, NHERF-1 was not detected in the NHERF-1 (–/–) animals. There were also no differences in NHERF-2 abundance between wild-type mice on the high-phosphate diet (1,562 ± 110 arbitrary units, n = 6) or the low-phosphate diet (1,512 ± 183 arbitrary units, n = 6) or between the NHERF-1 (–/–) mice on the high-phosphate diet (1,607 ± 120 arbitrary units, n = 5) or the low-phosphate diet (1,620 ± 59 arbitrary units, n = 5; Fig. 2B). Thus the phosphate content of the diet did not regulate NHERF-1 abundance in the wild-type animals or NHERF-2 abundance in either the wild-type or knockout mice. Moreover, as previously reported, there was no difference in NHERF-2 expression in the wild-type compared with the NHERF-1 (–/–) animals (18, 20). By contrast, BBM expression of NaPi-Cap1 was about threefold higher in both wild-type and NHERF-1 (–/–) mice on the low-phosphate diet compared with the high-phosphate diet. There were no differences, however, between wild-type and NHERF-1 (–/–) mice. On the low-phosphate diet, BBM expression of NaPi-Cap1 averaged 9,326 ± 640 arbitrary units (n = 4) in wild-type mice compared with 9,316 ± 152 arbitrary units (n = 4) in the NHERF-1 (–/–) animals (P < NS; Fig. 2C).
We previously reported that NHERF-1 (–/–) mice on a normal rodent chow diet that is relatively high in phosphate have decreased BBM expression of Npt2 in the BBM using confocal microscopy (16). To examine changes in the distribution of Npt2 in greater detail, immunolocalization studies were carried out in animals fed the high-phosphate diet and compared with animals receiving the low-phosphate diet for 24 h. As shown in Fig. 3, there was a significant increase in the abundance of Npt2 in the BBM of both wild-type and NHERF-1 (–/–) animals on the low-phosphate compared with the high-phosphate diet. However, consistent with the immunoblot findings described above, the abundance of Npt2 in the BBM was less for NHERF-1 (–/–) mice on the low-phosphate diet (Fig. 3D) compared with wild-type animals (Fig. 3C). Although the increase in BBM expression of Npt2 in response to the low-phosphate diet was striking, there was considerable variability in Npt2 expression between individual tubules in animals on the low-phosphate diet as shown in Fig. 3C. In addition to the expected localization in the BBM, Npt2 was also identified at intracellular sites. One site of intracellular Npt2 accumulation was just beneath the microvilli. These vesicles aligned as a “collar” or “string of beads” just below the BBM (arrows, Fig. 3D). Although Npt2 labeling at this site was seen in NHERF-1 (–/–) mice on a high-phosphate diet (Fig. 3B) and occasionally in wild-type mice on a low-phosphate diet (Fig. 3C), it was clearly most prominent in NHERF-1 (–/–) mice on a low-phosphate diet. This submicrovillar region was also labeled strongly with clathrin. When sections were colabeled with antibodies to Npt2 and clathrin, however, it was apparent that only a minority of the vesicles were positive for both markers (arrowhead, Fig. 4C). Most of the Npt2-labeled vesicles (arrows, Fig. 3) were located just below the band of clathrin labeling. The amount of vesicular Npt2 at this site did not appear to be affected by the phosphate content of the diet but was consistently more prominent in the NHERF-1 (–/–) mice compared with wild-type animals. Another site of less intense but definitive Npt2 labeling was detected at some distance from the apical surface. These structures colabeled with the Golgi marker GM130, whereas the subapical vesicles did not (arrows, Fig. 5). Npt2 labeling at the Golgi site did not appear to be altered in the NHERF-1 (–/–) mice or affected by the dietary content of phosphate. Npt2 antibody labeling of the BBM and all intracellular sites were blocked completely by coincubation of the antibody with a peptide representing the COOH terminus of Npt2 (data not shown).
Npt2 is the major regulated phosphate transporter in the proximal tubule of the kidney, and studies in the rat have indicated that hormones such as parathyroid hormone and dietary restriction of phosphate are associated with decreases and increases in Npt2 expression in the BBM, respectively (8, 9, 11, 12, 17). Because the activity of individual transporters is not altered in these circumstances and given the findings that Npt2 undergoes minimal endosomal recycling, it has been concluded that these maneuvers alter the trafficking of the Npt2 protein. A major insight into the renal response to phosphate restriction were the observations of Biber and co-workers (3) that phosphate restriction in the rat was associated with upregulation of a multi-PDZ domain containing protein called NaPi-Cap1. Although the physiological role of NaPi-Cap1 remains unknown, these observations focused attention on the role of proteins that interact with Npt2. The COOH terminus of Npt2 contains a PDZ-binding motif, and recent experiments have provided evidence that Npt2 has the potential to interact not only with NaPi-Cap1 but also with NHERF-1 and NHERF-2 (3, 20, 21). Additional evidence for a role of PDZ adaptor proteins in the renal handling of phosphate and Npt2 expression derives from initial studies of the NHERF-1 (–/–) mouse that are characterized by an increase in the urinary excretion of phosphate and a decrease in BBM abundance of Npt2 (16).
The role of NHERF-1 in the acute regulation of phosphate transport in the kidney and in the dynamic regulation of Npt2 expression has not been examined. The present experiments, therefore, were designed to determine whether NHERF-1 had a role in the rapid renal adaptation to restriction of the dietary intake of phosphate by comparing wild-type mice with NHERF-1 (–/–) mice. In the rat, 2 h of dietary phosphate restriction are associated with a decrease in the urinary excretion of phosphate and an increase in BBM Npt2 expression (9). This rapid response is not associated with changes in Npt2 mRNA levels and is thought to reflect either increased translation of Npt2 mRNA and/or movement of a cellular pool of transporters to the apical membrane. More prolonged dietary phosphate deprivation alters the abundance of Npt2 mRNA. We initially examined 4 h of phosphate deprivation in the mouse so as to approximate the protocols developed for the rat, but the urinary measurements proved too variable to yield meaningful results. Accordingly, we discarded urine excreted in the initial 4 h after the change in diet and then collected and analyzed urine excreted during the ensuing 24 h. In contrast to the rat, mouse Npt2 mRNA levels were not affected by the change to the low-phosphate diet for 24 h in either group of mice and did not differ between wild-type and NHERF-1 (–/–) animals. While these findings in the mice approximate the acute short-term response to phosphate deprivation in the rat, we recognized the present studies might encompass some elements of the more chronic response to the change in dietary phosphate consumption. In both wild-type and knockout mice, there was a large decrease in the urinary excretion of phosphate in response to the decrease in the phosphate content of the diet. Compared with the wild-type mice, however, the NHERF-1 (–/–) mice had a higher urinary excretion of phosphate, indicating that the adaptive response was incomplete. There was, however, a marked increase in BBM expression of Npt2 in both wild-type and NHERF-1 (–/–) animals when changed to the low-phosphate diet. Consistent with the urinary excretion data, BBM Npt2 expression was less in the NHERF-1 (–/–) animals compared with the wild-type animals. Wild-type animals increased Npt2 BBM expression 3.3-fold while on a low-phosphate diet compared with a high-phosphate diet, whereas the NHERF-1 (–/–) mice increased 2.6-fold. These results were confirmed using confocal microscopy by the demonstration that BBM expression of Npt2 was less in NHERF-1 (–/–) mice on the low-phosphate diet compared with wild-type animals on the same diet. These findings indicate that NHERF-1 is required for the full adaptive response to dietary phosphate restriction and the associated increase in BBM expression of Npt2.
Npt2 is somewhat unique among membrane proteins in that its activity is not altered by biochemical modification of transporters already resident in the BBM (8, 12, 17). Moreover, Npt2 appears to undergo limited endosomal recycling. In principle, then, the decreased BBM expression of Npt2 in the NHERF-1 (–/–) animals could be the result of NHERF-1 interacting with the distinct processes involved in Npt2 insertion, retention, or retrieval to and from the BBM. To examine the role of NHERF-1 in the regulation of Npt2 trafficking, additional studies were performed using confocal microscopy. Npt2 was detected in distinct intracellular sites in addition to the BBM. One intravesicular pool of Npt2 colabeled with the protein GM130, a marker for Golgi. There were no apparent differences in Golgi staining for Npt2 in the wild-type compared with NHERF-1 (–/–) mice. A new and more striking observation was the detection of Npt2 in a submicrovillar domain, forming a “collar” or “string-of-beads” configuration. The detection of Npt2 in the submicrovillar domain of the cell was not affected by the phosphate content of the diet but was clearly more prominent in NHERF-1 (–/–) animals than in wild-type controls. Within this submicrovillar region, some of the Npt2 colabeled with clathrin. The majority of Npt2, however, appeared to be located in a region just beneath the clathrin vesicles. Due to the lack of species-specific antibody pairs for endosome or lysosome identification in the mouse, it was not possible to definitely identify the cellular location of the submicrovillar nonclathrin-associated Npt2, but we suspect that it represents endosomal Npt2. Thus, while NHERF-1 may be involved in the biochemical pathways that insert Npt2 into the apical membrane in response to a low-phosphate diet, we currently favor the interpretation that NHERF-1 acts as a membrane retention signal for Npt2. In the absence of NHERF-1, Npt2 trafficked to the apical membrane is not efficiently retained, thereby reducing the BBM amount and increasing its detection in submicrovillar vesicles. This conclusion is supported by recent studies in opossum kidney (OK) cells demonstrating that NHERF-1 and Npt2 colocalize at the apical membrane and that NHERF-1 dominant-negative reagents disrupt the apical expression of Npt2 (7). We recognize that additional experiments will be required to fully define the role of NHERF-1 in Npt2 trafficking and that such studies will require other techniques and methods. Nonetheless, the present experiments reveal for the first time a role for NHERF-1 in the acute regulation of phosphate metabolism in the kidney of an intact animal.
We think it worth highlighting that the defect in Npt2 trafficking in response to the decrease in the dietary intake of phosphate in NHERF-1 (–/–) mice is relative rather than absolute, indicating perhaps that other factors in addition to NHERF-1 are important or that other proteins can compensate for the absence of NHERF-1, at least in part. Two candidate proteins are NHERF-2 and NaPi-Cap1. NHERF-2 expression is normal in the NHERF-1 (–/–) mice and is unchanged by alterations in the phosphate content of the diet (16). Despite the identification of NHERF-2 in a yeast two-hybrid screen using Npt2 as the bait, other evidence would suggest that the interaction between NHERF-2 and Npt2 is indirect and requires NHERF-1 and/or other associated proteins (5; Wade and Weinman, unpublished observations). It is also of interest to note that the rat responds to limitation of the dietary intake of phosphate like the wild-type mouse, although the rat proximal tubule expresses NHERF-1 but no detectable amounts of NHERF-2 (18). In a similar manner, there is an increase in BBM expression of Npt2 in response to a decrease in the phosphate concentration of the media in OK cells, a renal proximal tubule cell line that expresses NHERF-1 but not NHERF-2 (1, 14). When considered together, these findings suggest that NHERF-2 is not a major factor in the regulatory response to changes in the dietary intake of phosphate. On the other hand, cofactors such as NaPi-Cap1 might play an important role along with NHERF-1 in the alterations in BBM Npt2 expression and the renal response to dietary phosphate restriction. NaPi-Cap1 was initially identified in rat as a protein whose abundance was increased in response to dietary limitation of phosphate intake (3). Our results indicate that NaPi-Cap1 abundance is also increased in mice on the low-phosphate diet, but we found no differences in BBM expression of NaPi-Cap1 between wild-type and NHERF-1 (–/–) mice. The possibility of a direct or indirect association between NaPi-Cap1 and NHERF-1 has not yet been examined but, given that other PDZ proteins interact with one another, it is plausible to consider that both NHERF-1 and NaPi-Cap1 function in a cooperative manner to regulate the trafficking of Npt2 (2, 4). Finally, there may be a quantitative disparity between changes in the urinary excretion of phosphate and BBM abundance of Npt2. One possible explanation is that other phosphate transporters in the renal proximal tubule compensate to increase transepithelial phosphate transport by processes that are independent of NHERF-1. Miyamoto and colleagues (13, 15) recently reported identification of the Na-Pi IIc transporter in rat and mouse proximal tubule apical membrane and the increased expression of this sodium-dependent phosphate transporter in animals fed a low-phosphate diet. It remains to be determined, however, whether Na-Pi IIc is upregulated in NHERF-1 (–/–) animals.
These studies were supported by grants from the National Institutes of Health [DK-55881 (E. J. Weinman and S. Shenolikar), DK-32839 (J. B. Wade)] and Research Service, Department of Veterans Affairs (E. J. Weinman).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society