In the kidney, ∼80% of the filtered phosphate (Pi) is reabsorbed along the proximal tubule. Changes in renal Pi reabsorption are associated with modulation of the sodium-dependent Pi cotransporter type IIa (NaPi-IIa) and type IIc (NaPi-IIc) protein abundance in the brush-border membrane (BBM) of proximal tubule cells. NaPi-IIa is mainly regulated by dietary Pi intake and parathyroid hormone (PTH). The purpose of the present study was to elucidate the effect of alteration in dietary magnesium (Mg2+) intake on renal Pi handling. Urinary Pi excretion and renal expression of NaPi-IIa and NaPi-IIc were analyzed in rats fed a normal (0.2%) or high-Mg2+ (2.5%) diet. A high-Mg2+ diet resulted in decreased renal Pi excretion and increased protein expression of NaPi-IIa and NaPi-IIc. Serum FGF-23 (fibroblast growth factor 23) levels were unchanged under a high-Mg2+ diet. Serum PTH levels were slightly decreased under a high-Mg2+ diet. To examine whether the observed changes in renal Pi reabsorption are PTH dependent, expression of NaPi-IIa and NaPi-IIc was also analyzed in parathyroidectomized (PTX) rats fed a normal or high-Mg2+ diet. In PTX rats, Mg2+ had no significant effect on renal Pi excretion or NaPi-IIa protein expression. Mg2+ increased NaPi-IIc protein expression in PTX rats. This experiment shows for the first time on the molecular level how Mg2+ stimulates renal Pi reabsorption. Under a high-Mg2+ diet, NaPi-IIa expression is dependent on PTH levels, whereas NaPi-IIc expression seems to be independent of PTH levels.
- parathyroid hormone
magnesium (Mg2+) is the second most abundant intracellular cation in vertebrates. It interacts with calcium (Ca2+) and phosphate (Pi) homeostasis in different ways. Mg2+ activates like Ca2+ the Ca2+-sensing receptor (CaSR) of the parathyroid glands (2). Activation of the CaSR results in decreased parathyroid hormone (PTH) secretion. Although Mg2+ is less potent than Ca2+ in stimulating the CaSR, genetic diseases caused by mutations in the CaSR gene suggest that CaSR serves as an Mg2+ sensor in vivo. Individuals with autosomal-dominant hypoparathyroidism (ADH) due to activating mutations of the CaSR show, besides hypercalciuric hypocalcemia, tendencies toward hypomagnesemia and hyperphosphatemia (10). Heterozygous inactivating mutations of the CaSR cause familial hypocalciuric hypercalcemia (FHH), which is characterized by hypocalciuric hypercalcemia often associated with hypomagnesiuric hypermagnesemia (26). Affected individuals may display hypophosphatemia due to decreased renal Pi reabsorption (16). Homozygous or compound heterozygous inactivating mutations of the CaSR result in neonatal severe hyperparathyroidism (NSHPT), which is generally lethal unless a parathyroidectomy is performed (26).
High serum Mg2+ levels may have a suppressive role in PTH secretion in humans (5). In peritoneal dialysis patients, serum Mg2+ concentration is inversely correlated with PTH levels (23, 33). Taken together, these observations point to an influence of Mg2+ on PTH secretion in vivo.
Thus the aim of this study was to elucidate the role of Mg2+ in renal Pi handling. Overall Pi homeostasis is mainly regulated by intestinal absorption and renal excretion of Pi. Approximately 80% of the filtered Pi is reabsorbed along the proximal tubule. Transport of Pi across the proximal tubular cell membrane is mainly initiated by the sodium dependent Pi cotransporter NaPi type IIa (NaPi-IIa) (21). Less is reabsorbed by NaPi-IIc. The abundance of NaPi-IIa protein in the brush border membrane (BBM) reflects the capacity of the kidney to reabsorb Pi. Acute phosphaturic stimuli like PTH decrease the amount of NaPi-IIa protein in the BBM by inducing the endocytosis and lysosomal degradation of NaPi-IIa (3).
In this study, the effect of Mg2+ on renal Pi excretion and expression of NaPi-IIa and NaPi-IIc was analyzed in rats fed a normal (0.2%) or high-Mg2+ (2.5%) diet for 4 wk. Mg2+ increased renal Pi reabsorption by enhancing protein expression of NaPi-IIa and NaPi-IIc. Slightly decreased serum PTH levels were measured under a high-Mg2+ diet. In a second set of experiments, renal Pi excretion and NaPi-IIa and NaPi-IIc expression were analyzed in parathyroidectomized (PTX) rats fed a normal or high-Mg2+ diet. In PTX rats, Mg2+ did not significantly change renal Pi excretion or NaPi-IIa expression. NaPi-IIc expression was increased. These data indicate that Mg2+ stimulates renal Pi reabsorption via NaPi-IIa in a PTH-dependent manner, whereas NaPi-IIc expression seems to be influenced independently by Mg2+ and PTH levels.
MATERIALS AND METHODS
The experiments were performed with male Sprague-Dawley rats (300–350 g). Animals were purchased from Charles River Laboratories: 14 PTX rats and 14 control rats. All animal studies were performed according to the APS Guiding Principles in the Care and Use of Laboratory Animals and were approved by the local institutional office (Landesamt für Arbeitsschutz, Gesundheitsschutz und technische Sicherheit Berlin; Number 0001/06). Four different groups (7 animals each) were investigated: group 1 (control), nonoperated rats that received normal drinking water and standard pelleted chow (R/M-F extrudate from ssniff, 0.2% Mg2+) ad libitum; group 2 (control + Mg2+), nonoperated rats that received a high-Mg2+ diet (drinking water supplemented with 2.5% Mg2+-sulfate and standard pelleted chow ad libitum); group 3 (PTX), PTX rats that received normal drinking water and standard pelleted chow (0.2% Mg2+) ad libitum; and group 4 (PTX + Mg2+), PTX rats that received a high-Mg2+ diet (drinking water supplemented with 2.5% Mg2+-sulfate and standard pelleted chow ad libitum). The feeding period lasted for 28 days. At the end, rats were housed in metabolic cages for 24-h urine collection. Animals were anesthetized with ketamine and xylazine injected intraperitoneally. Blood samples were taken. Kidneys were perfused with ice-cold PBS solution via the abdominal aorta and removed.
Serum concentrations of Mg2+, Ca2+, Pi, Na+, K+, and alkaline phosphatase were determined. Twenty-four-hour urine samples were analyzedfor pH, Mg2+, Ca2+, Pi, Na+, and K+. Intact PTH serum levels were measured by an immunoradiometric assay specific for rat intact PTH (Immutopics, San Clemente, CA). FGF-23 serum levels were measured by an immunoradiometric assay specific for human FGF-23 with cross-reactivity for rat FGF-23 (Kainos Laboratories) (22).
RNA isolation and quantitative PCR.
Total RNA was extracted from kidneys using the RNeasy-total-RNA-kit (Qiagen). cDNA was synthesized by reverse transcription of RNA, using a cDNA synthesis kit (Invitrogen, Carlsbad, CA). Primers used for amplification of NaPi-IIa were forward 5-cgataatcatgggctccaacat-3, reverse 5-caaaagcccgcctgaag-3, and probe 5-FAM-acaccattgtggccctgatgca-TAMRA-3. Expression of NaPi-IIa and GAPDH, as an endogenous control, were determined by quantitative real-time PCR on an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA).
Kidneys were homogenized in iced lysate buffer (20 mmol/l Tris, pH 7.4, 5 mmol/l MgCl2, 1 mmol/l EDTA, 0.3 mmol/l EGTA) enriched with protease inhibitor cocktail complete mini (Boehringer Mannheim) according to the manufacturer's instructions. Insoluble material was removed by centrifugation at 250 g for 5 min at 4°C. For obtaining the membrane protein fraction, the supernatant was then centrifuged at 43,000 g for 30 min at 4°C. The pellet was resuspended in iced lysate buffer. Protein concentration was determined using a BCA protein assay kit (Pierce, Rockford, IL).
Aliquots of 25 μg (for NaPi-IIa and NaPi-IIc) and 50 μg protein (for β-actin) were separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane (NEN Life Science Products, Boston, MA). After the blots were blocked for 2 h in 5% milk powder in PBS and overnight in 5% bovine serum albumin in PBS containing 0.1% Tween 20, they were incubated with the primary antibodies (mouse anti-β-actin, Sigma-Aldrich, St. Louis, MO, 1:100,000, rabbit anti-NaPi-IIa, 1:1,000, and rabbit anti-NaPi-IIc, 1:1,000) for 3 h at room temperature (RT). Secondary peroxidase-conjugated goat anti-rabbit and goat anti-mouse IgG antibodies and the chemiluminescence detection system Lumi-LightPlus (Roche) were used to detect bound antibodies. Densitometric comparison was performed on the same immunoblot.
Kidneys were fixed in a solution containing 4% formalin and 10% sucrose in PBS for 1 h at RT. After fixation, kidneys were stored in 20% sucrose in PBS for 1 h at RT and in 30% sucrose in PBS overnight at 4°C. Kidneys were then frozen in liquid isopentane cooled with liquid nitrogen. For immunofluorescence staining, 5-μm-thick sections were pretreated with PBS containing 0.5% Triton X-100 in PBS for 15 min. After being blocked with 5% goat serum for 1 h, sections were incubated with the primary antibody NaPi-IIa (1:1,000) in 5% goat serum for 3 h at 37°C. After being washed with 5% goat serum, sections were incubated with the secondary antibody Alexa Fluor 594 (Invitrogen) diluted 1:500 in 5% goat serum for 45 min at RT. Fluorescence images were obtained using a confocal scanning microscope (LSM 510, Carl Zeiss, Jena, Germany). For semiquantitative determination of protein abundance, images were analyzed with the internal software, resulting in quantification of the protein levels as the mean of integrated optical density (IOD) in arbitrary units.
Data are expressed as means ± SD. All data were tested for significance using ANOVA followed by an unpaired Student's t-test. P values <0.05 were considered significant.
Metabolic changes induced by magnesium.
The metabolic data obtained under a normal diet (0.2% Mg2+) and high-Mg2+ diet (2.5% Mg2+) over 4 wk in control and PTX rats are shown in Table 1 and Fig. 1. A high-Mg2+ diet had no effect on serum Na+ concentration in control and PTX rats. Urinary Na+ excretion in PTX rats under a high-Mg2+ diet was significantly reduced compared with control rats under a normal diet. A high-Mg2+ diet had no effect on serum or urinary K+ levels. Serum alkaline phosphatase levels were not altered under a high-Mg2+ diet in control and PTX rats. Urinary pH was unchanged in the different groups. Total Mg2+ serum concentration and urinary Mg2+ excretion were significantly elevated in control and PTX rats under a high-Mg2+ diet. Urinary Ca2+ excretion was also significantly elevated in control and PTX rats under a high-Mg2+ diet. Total serum Ca2+ levels in PTX rats were significantly lower than in control rats. Under a high-Mg2+ diet, total serum Ca2+ levels increased in PTX rats, whereas total serum Ca2+ levels in control rats were unchanged under a high-Mg2+ diet. PTX rats had significantly lower urinary Pi excretion compared with control rats. Serum Pi concentration was significantly higher in PTX rats than in control rats. Under a high-Mg2+ diet, urinary Pi excretion was significantly reduced in control rats. A high-Mg2+ diet had no effect on serum Pi concentration. In PTX rats, urinary Pi excretion was not significantly decreased under a high-Mg2+ diet. Serum PTH levels in PTX rats were significantly lower compared with control rats. A high-Mg2+ diet tended to result in lower PTH levels in control rats, whereas low PTH levels in PTX rats were not affected by magnesium. Serum FGF-23 (fibroblast growth factor 23) levels were significantly lower in PTX rats compared with control rats. A high-Mg2+ diet had no effect on serum FGF-23 levels in both groups.
Expression of NaPi-IIa in normal vs. PTX rats under a normal diet.
NaPi-IIa protein expression was analyzed by Western blotting (Fig. 2) and immunohistochemistry (see Figs. 3–⇓5). NaPi-IIa protein expression in the whole kidney membrane protein fraction was significantly higher in PTX rats compared with control rats (Fig. 2). In control rats under a normal diet, proximal tubules of midcortical and superficial nephrons showed weak NaPi-IIa staining (Fig. 3) as described previously (6). While NaPi-IIa protein staining in the BBM was moderate, there was also staining in the cytoplasm of the proximal tubule cells, most likely reflecting NaPi-IIa protein residing in the Golgi apparatus (Fig. 4). Lowered PTH levels increased NaPi-IIa staining especially in proximal tubules of midcortical and superficial nephrons (Fig. 3). Under these conditions, NaPi-IIa protein was exclusively located in the BBM; no signal in the cytoplasm was detectable (Fig. 4). Differences in NaPi-IIa staining in juxtamedullary nephrons were less conspicuous (Fig. 5). NaPi-IIa mRNA expression was analyzed by quantitative real-time PCR in whole kidney and was unchanged in PTX rats compared with control group (Table 2).
Expression of NaPi-IIa in normal vs. PTX rats under a high-Mg2+ diet.
A high-Mg2+ diet significantly increased NaPi-IIa protein expression in whole kidney membrane protein fraction in control rats (Fig. 2). Under a high-Mg2+ diet, NaPi-IIa protein-related staining was particularly increased in midcortical und superficial nephrons (Fig. 3). NaPi-IIa was primarily located in the BBM of proximal tubule cells; no intracellular signal was detectable (Fig. 4). Changes in NaPi-IIa staining in juxtamedullary nephrons were more moderate (Fig. 5). In PTX rats, Mg2+ had no effect on total NaPi-IIa protein expression analyzed by Western blotting and immunohistochemistry (Figs. 2–5). There was no change in NaPi-IIa staining intensity in the proximal tubule of rat renal cortex (Figs. 3–5). NaPi-IIa mRNA expression was unchanged under a high-Mg2+ diet in control and PTX rats (Table 2).
Expression of NaPi-IIc in normal vs. PTX rats under a normal and high-Mg2+ diet.
NaPi-IIc protein expression was analyzed by Western blotting (Fig. 6). NaPi-IIc protein expression in the whole kidney membrane protein fraction was significantly higher in PTX rats compared with control rats. A high-Mg2+ diet significantly increased NaPi-IIc protein expression in whole kidney membrane protein fractions in control rats and PTX rats (Fig. 6).
In the present study, the effect of Mg2+ on renal Pi handling in rats was investigated. A high-Mg2+ diet increased renal Pi reabsorption by enhancing protein expression of NaPi-IIa and NaPi-IIc transporters. Increased NaPi-IIa protein expression in rats during high Mg2+ intake was dependent on serum PTH levels. NaPi-IIc protein expression increased independently from serum PTH levels, suggesting an alternative mechanism of action.
The effect of either Mg2+ supplementation or depletion on Ca2+-Pi homeostasis in rats (7, 8, 12, 13, 25, 28, 30) and other species (19, 31) has been investigated before. In these studies, changes observed in Ca2+-Pi homeostasis were not congruent. One reason for these differences might be variances in dietary Mg2+ content and duration of the feeding period. In the present study, urinary Ca2+ excretion increased significantly with no change in total serum Ca2+ concentration under a high-Mg2+ diet. Increased urinary Ca2+ excretion might be due to competitive paracellular Ca2+/Mg2+ reabsorption in the thick ascending limb of Henle mediated by claudin 16 (11), stimulation of the CaSR in the thick ascending limb of Henle, inhibition of PTH release (2), or competition for the TRPV5 calcium channel (4).
Urinary Pi excretion decreased significantly under a high-Mg2+ diet with no change in serum Pi levels in our study. This observation is in accordance with the study of Ginn et al. (8): under Mg2+ deprivation, urinary Pi excretion increased significantly without affecting serum Pi levels. Fiore et al. (7) also found no change in serum Pi level under Mg2+ supplementation; urinary Pi excretion was not investigated. Massry et al. (19) found elevated serum Pi levels and decreased fractional Pi excretion under Mg2+ supplementation. In the present study, serum PTH levels seemed to be lower under Mg2+ supplementation, but this was not statistically significant probably due to large variances in the control group. In the above-mentioned studies, the effect of Mg2+ on serum PTH levels varied widely. Riond et al. (28) and Katsumata et al. (13) observed no changes in serum PTH levels under high Mg2+ diet, whereas Fiore et al. (7) found decreased serum PTH levels under a high-Mg2+ diet. Mg2+ deficiency in rats caused suppressed, unchanged, or elevated serum PTH levels (12, 25, 30). Slatopolsky et al. (31) showed that hypermagnesemia in dogs inhibits the renal action of PTH, resulting in less urinary Pi excretion.
In all these studies, the underlying molecular mechanisms responsible for the observed changes in Pi metabolism under Mg2+ deprivation or supplementation were not elucidated. Renal Pi reabsorption is regulated by a variety of hormonal and metabolic factors. Insulin and Pi deprivation stimulate renal Pi reabsorption, whereas PTH, FGF-23, and metabolic acidosis reduce Pi reabsorption (1, 17, 21). The rate-limiting step of transcellular Pi reabsorption is the apical entry of Pi in the proximal tubule cell. This is mainly initiated by NaPi-IIa. NaPi-IIa expression is heterogeneously distributed within the renal cortex. Under basal conditions, i.e., normal dietary phosphate intake, NaPi-IIa protein abundance in the BBM is highest in juxtamedullary nephrons, whereas protein abundance in superficial and midcortical nephrons is low (3). Enhanced renal Pi reabsorption is caused by an increased abundance of NaPi-IIa protein in the BBM especially in superficial and midcortical nephrons of proximal tubules (18, 29). The regulation of renal NaPi-IIc expression has been studied less. Metabolic acidosis increases NaPi-IIc protein expression without changing the distribution within the renal cortex (24). The influence of dietary Mg2+ on the expression of NaPi-IIa and NaPi-IIc has not been investigated to date. In our study, a high-Mg2+ diet stimulates renal protein expression of NaPi-IIa and NaPi-IIc in the whole kidney membrane protein fraction (Figs. 2 and 6). Immunohistological analysis showed increased NaPi-IIa protein expression particularly in midcortical and superficial nephrons, where NaPi-IIa protein was exclusively located in the BBM of the proximal tubule under a high-Mg2+ diet (Figs. 3–5).
Mg2+ could directly influence NaPi-IIa expression or affect its abundance via an alternative indirect pathway. The animals in our experiment received Mg2+-sulfate, which can cause mild metabolic acidosis and acidification of urine (9). Acidosis is known for enhancing Pi excretion (1). In our experiment, no change in urinary pH under Mg2+ supplementation was observed. The phosphatonin FGF-23 also stimulates renal Pi excretion by decreasing NaPi-IIa expression (17). In our experiment, a high-Mg2+ diet had no effect on serum FGF-23 levels. Thus it is unlikely that Mg2+ stimulates NaPi-IIa expression through changes in acid-base balance or changes in FGF-23 secretion.
Mg2+ can activate like Ca2+ but less potently than the CaSR of the parathyroid glands (2). Activation of the CaSR decreases PTH secretion. In our study, Mg2+ seemed to decrease PTH levels, although PTH changes did not reach significance. To investigate whether Mg2+ stimulates Pi reabsorption in a PTH-dependent or -independent manner, a second experiment with PTX rats was performed. Apparently, parathyroidectomy was only partial as PTH was still detectable in serum although at significantly lower levels than in control rats. Serum Pi concentration was elevated, while urinary Pi excretion was reduced compared with control rats as expected. Total serum Ca2+ concentration was decreased in PTX rats. Taken together, these changes in Ca2+-Pi homeostasis indicate an effective reduction of biologically active PTH in the PTX rats (27).
Serum FGF-23 levels were significantly reduced in PTX rats in our study. Correlation between PTH and FGF-23 secretion is not completely clarified. While Kobayashi et al. (15) found elevated FGF-23 levels in subjects with primary hyperparathyroidism, Tebben et al. (32) found no changes in FGF-23 serum levels in subjects with primary hyperparathyroidism compared with healthy subjects. Recent animal studies indicate that 1,25-(OH)2 vitamin D3 stimulates FGF-23 synthesis, while activation of 25-OH vitamin D3 by the 1-α-hydroxylase enzyme is under the control of PTH (20). Thus PTH may control FGF-23 levels via 1,25-(OH)2-vitamin D3. Consistently, in our study deprivation of PTH was associated with decreased serum FGF-23 levels.
A high-Mg2+ diet in PTX rats resulted in increased urinary Ca2+ excretion like in control rats. However, total Ca2+ serum concentration was elevated under Mg2+ supplementation in PTX rats. These findings cannot only be explained by competitive paracellular Ca2+/Mg2+ reabsorption in the thick ascending limb of Henle (11) or stimulation of the CaSR in the thick ascending limb of Henle and inhibition of PTH release (2). Alternatively, Bonny et al. (4) recently suggested that high urinary Mg2+ may compete for reabsorption with Ca2+ at the TRPV5 channel in the distal convoluted and connecting tubule. The mechanism responsible for elevating serum Ca2+ levels in this case is not known.
NaPi-IIa protein expression in the whole kidney membrane protein fraction increased after parathyroidectomy in our study (Fig. 2). NaPi-IIa protein abundance, particularly in midcortical and superficial nephrons, was enhanced (Figs. 3 and 4). Changes in expression of NaPi-IIa mRNA were not observed. These findings are in accordance with Kempson et al. (14), who observed increased NaPi-IIa protein expression in the BBM fraction without changes in mRNA expression in PTX rats. Supplementation of Mg2+ after parathyroidectomy had no additional effect on total NaPi-IIa protein expression or distribution of NaPi-IIa protein within the renal cortex (Figs. 2–5). Under a high-Mg2+ diet, urinary excretion of Pi seemed to be lower in PTX rats. Mg2+ supplementation increased NaPi-IIc expression in PTX rats. This fact could explain the slightly increased Pi reabsorption under a high-Mg2+ diet after parathyroidectomy. This suggests that Mg2+ regulates NaPi-IIc expression independently from PTH levels. Maybe Mg2+ has an additional effect on renal Pi excretion, besides lowering PTH levels. It could directly stimulate the CaSR of proximal tubule cells or modulate the action of other phosphaturic or antiphosphaturic factors, including PTH (31). These data indicate that the main effect of Mg2+ is not directly influencing renal Pi reabsorption. Rather, Mg2+ inhibits PTH secretion, which results in increased renal Pi reabsorption by increasing NaPi-IIa protein expression. Therefore, these experiments show for the first time on the molecular level how Mg2+ influences renal Pi excretion. Moreover, these data also confirm previous findings that Mg2+ influences PTH secretion not only in vitro but also in vivo (2, 5, 10, 16, 23, 26, 33).
This work was supported by the 6th Framework Programme of the European Union (EuReGene FP6005085) to J. Biber, H. Murer, C. A. Wagner, and D. Müller.
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 © 2008 the American Physiological Society