Phosphate transport in the kidney

Effects of Excessive Dietary Phosphorus Intake on Bone Health

PURPOSE OF REVIEW

The purpose of this review is to provide an overview of dietary phosphorus, its sources, recommended intakes, and its absorption and metabolism in health and in chronic kidney disease and to discuss recent findings in this area with a focus on the effects of inorganic phosphate additives in bone health.

RECENT FINDINGS

Recent findings show that increasing dietary phosphorus through inorganic phosphate additives has detrimental effects on bone and mineral metabolism in humans and animals. There is new data supporting an educational intervention to limit phosphate additives in patients with chronic kidney disease to control serum phosphate. The average intake of phosphorus in the USA is well above the recommended dietary allowance. Inorganic phosphate additives, which are absorbed at a high rate, account for a substantial and likely underestimated portion of this excessive intake. These additives have negative effects on bone metabolism and present a prime opportunity to lower total phosphorus intake in the USA. Further evidence is needed to confirm whether lowering dietary phosphorus intake would have beneficial effects to improve fracture risk.

The regulation of renal phosphate transport

Renal phosphate transport is mediated by the abundance and activity of the sodium-dependent phosphate transporters, Npt2a, Npt2c, and PiT-2, present within the apical brush border membrane of the proximal tubule. Recent studies have demonstrated differential expression and activity of these sodium-dependent phosphate transporters within the proximal tubule. In general, phosphate transport is regulated by a variety of physiological stimuli, including parathyroid hormone, glucocorticoids, vitamin D3, estrogen, and thyroid hormone. Phosphatonins are now recognized as major regulators of phosphate transport activity. Other factors that affect phosphate transport include dopamine, dietary phosphate, acid-base status, lipid composition, potassium deficiency, circadian rhythm, and hypertension. Studies have shown that the PDZ-containing sodium/hydrogen exchanger regulatory factor (NHERF) proteins, specifically NHERF-1 and NHERF-3, play a critical role in the physiological regulation of phosphate transport, particularly in response to dietary phosphate. In addition, recent studies have found that NHERF-1 is also important in both the parathyroid hormone- and dopamine-mediated inhibition of phosphate transport. This review will detail the various hormones and agents involved in the regulation of phosphate transport as well as provide a brief summary of the signaling pathways and cytoskeletal proteins active in the transport of phosphate in the renal proximal tubule.

1alpha,25-Dihydroxyvitamin D3 upregulates FGF23 gene expression in bone: the final link in a renal-gastrointestinal-skeletal axis that controls phosphate transport

Fibroblast growth factor (FGF)23 is a phosphaturic hormone that decreases circulating 1alpha,25-dihydroxyvitamin D(3) [1,25(OH)(2)D(3)] and elicits hypophosphatemia, both of which contribute to rickets/osteomalacia. It has been shown recently that serum FGF23 increases after treatment with renal 1,25(OH)(2)D(3) hormone, suggesting that 1,25(OH)(2)D(3) negatively feedback controls its levels by inducing FGF23. To establish the tissue of origin and the molecular mechanism by which 1,25(OH)(2)D(3) increases circulating FGF23, we administered 1,25(OH)(2)D(3) to C57BL/6 mice. Within 24 h, these mice displayed a dramatic elevation in serum immunoreactive FGF23, and the expression of FGF23 mRNA in bone was significantly upregulated by 1,25(OH)(2)D(3), but there was no effect in several other tissues. Furthermore, we treated rat UMR-106 osteoblast-like cells with 1,25(OH)(2)D(3), and real-time PCR analysis revealed a dose- and time-dependent stimulation of FGF23 mRNA concentrations. The maximum increase in FGF23 mRNA was 1,024-fold at 10(-7) M 1,25(OH)(2)D(3) after 24-h treatment, but statistically significant differences were observed as early as 4 h after 1,25(OH)(2)D(3) treatment. In addition, using cotreatment with actinomycin D or cycloheximide, we observed that 1,25(OH)(2)D(3) regulation of FGF23 gene expression occurs at the transcriptional level, likely via the nuclear vitamin D receptor, and is dependent on synthesis of an intermediary transfactor. These results indicate that bone is a major site of FGF23 expression and source of circulating FGF23 after 1,25(OH)(2)D(3) administration or physiological upregulation. Our data also establish FGF23 induction by 1,25(OH)(2)D(3) in osteoblasts as a feedback loop between these two hormones that completes a kidney-intestine-bone axis that mediates phosphate homeostasis.

Fibroblast growth factor-23 relationship to dietary phosphate and renal phosphate handling in healthy young men

The renal handling of inorganic phosphate (Pi) is controlled not only by PTH, but also by hitherto undetermined mechanisms dependent on phosphate intake. Recently, fibroblast growth factor (FGF)-23 was identified as a novel phosphaturic factor in tumor-induced osteomalacia and autosomal-dominant hypophosphatemic rickets. We hypothesized that phosphate intake could influence FGF-23 concomitantly to the changes in renal Pi handling. Twenty-nine healthy males were subjected to a 5-d low-phosphate diet and a phosphate binder, followed by a high-phosphate diet including supplements. Concomitant modifications in calcium intake allowed minimizing PTH changes in response to dietary phosphate. Serum FGF-23 levels significantly decreased on the low-phosphate diet, then increased with the oral phosphate load. Changes in FGF-23 were positively correlated with changes in 24-h urinary Pi excretion and negatively correlated with changes in the maximal tubular reabsorption of Pi and 1,25(OH)(2)D(3) (calcitriol), whereas PTH was not. In multivariate analysis, changes in FGF-23 remained the most significantly correlated to changes in 1,25(OH)(2)D(3) and maximal tubular reabsorption of Pi. Moreover, FGF-23 was positively correlated to serum osteocalcin, a marker of osteoblastic activity. In summary, FGF-23 was inversely related to renal Pi transport and serum calcitriol levels in healthy young men. These data suggest that FGF-23 may be implicated in the physiological regulation of Pi homeostasis in response to dietary phosphate changes, independent of PTH.

Molecular mechanisms in proximal tubular and small intestinal phosphate reabsorption (plenary lecture)

Renal and small intestinal (re-)absorption contribute to overall phosphate(Pi)-homeostasis. In both epithelia, apical sodium (Na+)/Pi-cotransport across the luminal (brush border) membrane is rate limiting and the target for physiological/pathophysiological alterations. Three different Na/Pi-cotransporters have been identified: (i) type I cotransporter(s)--present in the proximal tubule--also show anion channel function and may play a role in secretion of organic anions; in the brain, it may serve vesicular glutamate uptake functions; (ii) type II cotransporter(s) seem to serve rather specific epithelial functions; in the renal proximal tubule (type Ila) and in the small intestine (type IIb), isoform determines Na+-dependent transcellular Pi-movements; (iii) type III cotransporters are expressed in many different cells/tissues where they could serve housekeeping functions. In the small intestine, alterations in Pi-absorption and, thus, apical expression of IIb protein are mostly in response to longer term (days) situations (altered Pi-intake, levels of 1.25 (OH2) vitamin D3, growth, etc), whereas in renal proximal tubule, in addition, hormonal effects (e.g. Parathyroid Hormone, PTH) acutely control (minutes/hours) the expression of the IIa cotransporter. The type II Na/Pi-cotransporters operate (as functional monomers) in a 3 Na+:1 Pi stoichiometry, including transfer of negatively charged (-1) empty carriers and electroneutral transfers of partially loaded carriers (1 Na+, slippage) and of the fully loaded carriers (3 Na+, 1 Pi). By a chimera (IIa/IIb) approach, and by site-directed mutagenesis (including cysteine-scanning), specific sequences have been identified contributing to either apical expression, PTH-induced membrane retrieval, Na+-interaction or specific pH-dependence of the IIa and IIIb cotransporters. For the COOH-terminal tail of the IIa Na/Pi-cotransporter, several interacting PDZ-domain proteins have been identified which may contribute to either its apical expression (NaPi-Cap1) or to its subapical/lysosomal traffic (NaPi-Cap2).

Proximal tubular phosphate reabsorption: molecular mechanisms

Renal proximal tubular reabsorption of P(i) is a key element in overall P(i) homeostasis, and it involves a secondary active P(i) transport mechanism. Among the molecularly identified sodium-phosphate (Na/P(i)) cotransport systems a brush-border membrane type IIa Na-P(i) cotransporter is the key player in proximal tubular P(i) reabsorption. Physiological and pathophysiological alterations in renal P(i) reabsorption are related to altered brush-border membrane expression/content of the type IIa Na-P(i) cotransporter. Complex membrane retrieval/insertion mechanisms are involved in modulating transporter content in the brush-border membrane. In a tissue culture model (OK cells) expressing intrinsically the type IIa Na-P(i) cotransporter, the cellular cascades involved in "physiological/pathophysiological" control of P(i) reabsorption have been explored. As this cell model offers a "proximal tubular" environment, it is useful for characterization (in heterologous expression studies) of the cellular/molecular requirements for transport regulation. Finally, the oocyte expression system has permitted a thorough characterization of the transport characteristics and of structure/function relationships. Thus the cloning of the type IIa Na-P(i )cotransporter (in 1993) provided the tools to study renal brush-border membrane Na-P(i) cotransport function/regulation at the cellular/molecular level as well as at the organ level and led to an understanding of cellular mechanisms involved in control of proximal tubular P(i) handling and, thus, of overall P(i) homeostasis.

Effect of iron(III) chitosan intake on the reduction of serum phosphorus in rats

Because of the widespread use of aluminium- and calcium-containing phosphate binders for the control of hyperphosphataemia in patients with end-stage renal failure, an iron(III) chitosan complex was synthesised and fed to rats to measure its effect on serum phosphorus and calcium, intestinal phosphate binding and phosphate absorption. Thirty-six Wistar rats were randomly selected and distributed into a baseline group (n = 6), a control group (n = 8 (days 0-15), n = 8 (days 16-30)) and a treatment group (n = 8 (days 0-15), n = 8 (days 16-30)). The control groups ingested AIN-76 diet mix with a 1% w/w fibre content; however, the treatment groups had the fibre content completely substituted with iron(III) chitosan. The mean weights of the treated rats were slightly lower from 15 days (not significant); but overall, rat growth was not stunted in the treatment groups. The serum phosphorus levels of the treated group (n = 8) were significantly reduced after 15 days (P = 0.004; control: 5.7+/-0.9 mg dL(-1); treatment: 4.4+/-0.5 mg dL(-1); 95% CI of difference: 0.5-2.2) and 30 days (P = 0.002; control: 5.5+/-0.9 mg dL(-1); treatment = 4.1+/-06 mg dL(-1); 95% CI of difference: 0.6-2.3) as compared with the respective control group. The serum calcium-phosphorus product was 62.0+/-12.1 mg2 dL(-2) for the control and 45.1+/-6.6 mg2 dL(-2) for the treatment group after 30 days (P = 0.004). The serum iron concentration of the treatment group did not differ from the baseline value after 15 and 30 days, but the treatment group was significantly higher than the control group (P<0.05) after 30 days. The faeces phosphorus levels (mg day(-1)) were higher (P<0.01) and its iron content was much higher (P<0.01) for the treated group. The urine phosphorus (mg kg(-1)) was not significantly reduced for the treated group, but the mean was consistently less. The kidney and liver weights of both groups were similar, but the phosphorus content of the kidney (mg (g kidney)(-1)) was higher for the treated group after 30 days (P = 0.041; control, 4.2+/-1.2 mg g(-1) vs treatment, 5.6+/-1.4 mg g(-1). Because iron(III) chitosan had a high phosphorus-binding capacity of 308 (mg P) per gram of Fe3+ for both the in-vitro (pH 7.5) and in-vivo studies, which is greater than nearly all commonly used phosphate binders, and a small net phosphorus absorption difference of 3.7 mg day(-1), it is an efficient phosphate binder for lowering serum phosphate levels without increasing serum calcium levels.

Posttranscriptional regulation of the proximal tubule NaPi-II transporter in response to PTH and dietary P(i)

The rate of proximal tubular reabsorption of phosphate (P(i)) is a major determinant of P(i) homeostasis. Deviations of the extracellular concentration of P(i) are corrected by many factors that control the activity of Na-P(i) cotransport across the apical membrane. In this review, we describe the regulation of proximal tubule P(i) reabsorption via one particular Na-P(i) cotransporter (the type IIa cotransporter) by parathyroid hormone (PTH) and dietary phosphate intake. Available data indicate that both factors determine the net amount of type IIa protein residing in the apical membrane. The resulting change in transport capacity is a function of both the rate of cotransporter insertion and internalization. The latter process is most likely regulated by PTH and dietary P(i) and is considered irreversible since internalized type IIa Na-P(i) cotransporters are subsequently routed to the lysosomes for degradation.

Membrane topography of the renal phosphate carrier NaPi-2: limited proteolysis studies

The rat sodium/phosphate cotransporter NaPi-2 is a 70 kDa polypeptide (p70) for which eight transmembrane segments have been predicted. We have shown that p70 exists predominantly as p45 and p40 fragments which are linked by disulfide bonds. In this work, the p40 fragment, corresponding to the C-terminus of NaPi-2, was purified from renal brush-border membranes using non-reducing and then reducing column electrophoresis followed by enzymatic deglycosylation and SDS-PAGE. The N-terminal sequence obtained for this fragment, VEAIG, indicates that the formation of p45 and p40 arises from the cleavage of p70 between arginine-319 and valine-320. In order to determine the membrane topography of NaPi-2, brush-border membrane vesicles were digested with various proteases and the transporter-derived proteolytic peptides were subsequently identified by Western blotting using N- and C-terminal-directed antibodies. Our results lead us to propose an alternative topographical model in which p45 and p40 possess three transmembrane domains each and indicate that the processing site of p70 for the generation of p45 and p40 is localized in a large protein core facing the extracellular milieu. This localization of the cleavage site indicated that NaPi-2 could either be processed intracellularly by vesicular proteases or extracellularly by secretory proteases or by brush-border membrane ectoenzymes.

Parathyroid hormone and dietary phosphate provoke a lysosomal routing of the proximal tubular Na/Pi-cotransporter type II

BACKGROUND

A decrease of proximal tubular reabsorption of phosphate (Pi), which can be provoked by parathyroid hormone (PTH) or by a high Pi-diet, has been shown to correlate with a decrease of the number of type II Na/Pi-cotransporters residing in the brush border membrane. While both PTH and a high Pi-diet lead to an internalization of type II cotransporters, the further cellular routing of internalized cotransporters has not been established unequivocally.

METHODS

To prevent lysosomal degradation, rats were treated with leupeptin prior to the injection of PTH or feeding acutely with a high Pi-diet. Kidney cortex were recovered and used for immunohistochemistry. In parallel, brush border membranes and lysosomes were isolated and analyzed by Western blotting.

RESULTS

Under both conditions (PTH and high Pi-diet), a strong overlap of internalized type II cotransporters with the late endosomes/lysosomes was observed by immunohistochemistry. In agreement, the content of type II Na/Pi-cotransporters was increased in lysosomes isolated from the corresponding tissues.

CONCLUSIONS

These results suggest that in proximal tubular cells type II Na/Pi-cotransporters internalized due to the action of PTH and acute high Pi-diet are routed to the lysosomes, and likely do not enter a recycling compartment.

Involvement of disulphide bonds in the renal sodium/phosphate co-transporter NaPi-2

The rat renal brush border membrane sodium/phosphate co-transporter NaPi-2 was analysed in Western blots with polyclonal antibodies raised against its N-terminal and C-terminal segments. Under reducing conditions, proteins of 45-49 and 70-90 kDa (p45 and p70) were detected with N-terminal antibodies, and proteins of 40 and 70-90 kDa (p40 and p70) were detected with C-terminal antibodies. p40 and p45 apparently result from a post-translational cleavage of NaPi-2 but remain linked through one or more disulphide bonds. Glycosidase digestion showed that both polypeptides are glycosylated; the cleavage site could thus be located between Asn-298 and Asn-328, which have been shown to constitute the only two N-glycosylated residues in NaPi-2. In the absence of reducing agents, both N-terminal and C-terminal antibodies detected p70 and a protein of 180 kDa (p180), suggesting the presence of p70 dimers. Much higher concentrations of beta-mercaptoethanol were required to produce a given effect in intact membrane vesicles than in solubilized proteins, indicating that the affected disulphide bonds are not exposed at the surface of the co-transporter. Phosphate transport activity decreased with increasing concentrations of reducing agents [beta-mercaptoethanol, dithiothreitol and tris-(2-carboxyethyl)phosphine] and was linearly correlated with the amount of p180 detected. The target sizes estimated from the radiation-induced loss of intensity of p40, p70 and p180 were all approx. 190 kDa, suggesting that NaPi-2 exists as an oligomeric protein in which the subunits are sufficiently close to one another to allow substantial energy transfer between the monomers. When protein samples were pretreated with beta-mercaptoethanol [2.5% and 5% (v/v) to optimize the detection of p40 and p70] before irradiation, target sizes estimated from the radiation-induced loss of intensity of p40 and p70 were 74 and 92 kDa respectively, showing the presence of disulphide bridges in the molecular structure of NaPi-2.

Adaptation to changes in dietary phosphorus intake in health and in renal failure

Phosphate (Pi) homeostasis is maintained by the ability of the kidneys to adjust the tubular reabsorption of Pi to changes in the dietary intake of phosphorus. Renal tubular Pi reabsorption increases with the ingestion of a low-phosphorus diet (LPD) and decreases when a high-phosphorus diet (HPD) is consumed. A similar adaptive mechanism is also operative at the intestinal microvillus. The adaptive changes in Pi reabsorption are independent of parathyroid hormone production and are paralleled by similar changes in the Na+-dependent Pi transport at the brush border membrane (BBM). Type II Na+-Pi cotransporters (NaPi-2) are mainly involved in such regulatory mechanisms. Chronic dietary phosphorus restriction leads to increased Na+-Pi cotransport rate, along with increased NaPi-2 protein and mRNA abundance. In acute dietary phosphorus restriction, transport rate and NaPi-2 protein are also increased, but mRNA abundance remains unchanged. A shuttling mechanism involving translocation of cotransporters from intracellular pools to the BBM is involved in the rapid proximal tubular adaptation. The intestinal adaptation to changes in dietary phosphorus are similar to those described for the renal Pi transport, but the molecular structure of the intestinal Na+-Pi cotransporter is not known. When nephron mass is reduced, phosphate homeostasis is maintained through enhanced Pi excretion by residual nephrons. The adaptation to renal mass reduction is mediated by increased parathyroid hormone (PTH) production and by PTH-independent mechanisms, including increased intrarenal dopamine production. The adaptive changes of Pi transport to dietary phosphorus restriction can counteract the effect of dietary phosphorus reduction often prescribed in patients with renal failure. However, because of the reduced filtered load of Pi, the overall impact on serum Pi concentration is minimal.

Phosphate deprivation induces overexpression of two proteins related to the rat renal phosphate cotransporter NaPi-2

Polyclonal antibodies were raised in rabbits against the C-terminal portion of the rat renal brush-border membrane sodium/phosphate cotransporter NaPi-2. Antibody specificity and molecular sizes of proteins related to NaPi-2 were assayed by Western blot analysis. Proteins of 40 and 70-75 kDa (p40 and p70) were immunodetected in rat and mouse brush-border membranes and proteins of 72 and 82 kDa were detected in rabbit. The absence or presence of beta-EtSH in the samples before electrophoresis greatly influenced the immunodetection profile of the rat proteins. Since the 40 kDa protein (p40) can only be detected under reducing conditions, it probably originates from reduction of disulfide bonds in p70. Tryptic cleavage of p40 and p70 revealed identical protein fragments showing the close structural identity of those proteins. Both proteins were more abundant in the outer cortex portion of the rat kidney than in the juxtamedullary portion. Furthermore, rats fed a low-phosphate diet for 24 h showed a 20- and 14-fold increase in the amount of p40 and p70, respectively, compared to control rats, showing that the adaptation to P(i) deprivation by increasing renal phosphate reabsorption is not only the result of overproduction of p70, as previously shown, but is also due to the novel p40 which most probably derives from p70.

Factors affecting the stability of the renal sodium/phosphate symporter during its solubilization and reconstitution

Phosphate is reabsorbed across the brush-border membrane of the proximal tubule by a specific sodium-dependent symporter. Like the other brush-border membrane transport proteins of the kidney, the phosphate carrier remains to be isolated in a functional state. To establish a set of parameters that allow to preserve its biological activity, the phosphate carrier was solubilized under systematically varied conditions and reconstituted into proteoliposomes. Successful reconstitution was achieved only when the extraction buffer contained lipids extracted from the renal brush-border membrane. Glycerol, an osmolyte which reduces the water activity of the solution, was also required. It could however be replaced by 150 mM sodium or potassium phosphate. Below this concentration and in the presence of glycerol, the ionic strength of the solution had little effect on the stability of the transporter, but sodium phosphate could not be replaced by sodium chloride. Phosphate transport in reconstituted vesicles depended on the concentration of detergent and pH of the extraction buffer. Finally, transport activity was increased when solubilization was carried out in the presence of a reducing agent, dithiothreitol. These results should be helpful during the purification and further characterization of the renal phosphate symporter.

Molecular size of the functional complex and protein subunits of the renal phosphate symporter

The oligomeric structure of the rabbit renal brush-border membrane sodium/phosphate cotransporter was examined with the radiation inactivation and fragmentation technique. The size of its functional complex (its "radiation inactivation size") was estimated from the rate of decay of its sodium-dependent transport activity as a function of the radiation dose. A radiation inactivation size of 223 +/- 42 kDa was obtained. The polypeptide constituting the monomeric unit of the Na1+/Pi symporter was detected by immunoblotting with polyclonal anti-peptide antibodies directed against the 14 amino acid C-terminal portion of the symporter molecule. Its apparent molecular size estimated by comparison with standards following SDS-polyacrylamide gel electrophoresis was 64,000. This value is in good agreement with its known molecular mass of 51,797 Da calculated from the amino acid sequence deducted from the nucleotide sequence of its gene since this protein is probably glycosylated. The loss of labeling intensity of the polypeptide of M(r) = 64,000 was also measured as a function of radiation dose. The molecular size calculated from these data (its "target size") was 165 +/- 20 kDa. The target size estimated for the rat phosphate cotransporter was 184 +/- 46 kDa, and its previously reported radiation inactivation size was 234 +/- 14 kDa. These results strongly suggest that the renal Na1+/Pi cotransporter exists as an oligomeric protein, probably a homotetramer. The fact that the values obtained for the target size are about 3/4 those obtained for the radiation inactivation size of these cotransport proteins indicates that their subunits are closely associated since most of their subunits appear to be fragmented by a single ionizing radiation hit.

Immunodetection and characterization of proteins implicated in renal sodium/phosphate cotransport

Polyclonal antibodies raised against the 14-amino acid C-terminal portion of the rabbit renal brush-border membrane Na+/Pi cotransporter, as deduced from the nucleotide sequence of the cloned NaPi-1 gene, were used for Western blot analysis of renal brush-border membrane proteins from rat, rabbit and beef. Proteins of 65 kDa from the rat, 64 kDa from the rabbit, and 38, 66, 77, 92, 110, 176 and 222 kDa from the beef were specifically labelled. The affinity of the antibodies was much greater, however, for the proteins of the rat and rabbit than for those of the beef. The rat 65-kDa antigen was readily detected in brush-border membranes isolated from kidney cortex, but was absent from the basolateral membrane and the cytosolic and microsomal fractions of this tissue, in agreement with the subcellular localization of the Na+/Pi cotransporter. This antigen was however several-fold more abundant in the juxtamedullary portion of the cortex than in the outer portion. Despite a strong stimulation in phosphate transport, a low-phosphate diet had little influence on the amount of antigen detected. An additional peptide-displaceable band corresponding to a protein of 250 kDa appeared when beta-mercaptoethanol was omitted during electrophoresis, in agreement with the possibility that disulfide bonds may be involved in the regulation of renal phosphate transport activity.

Expression of Na/Pi cotransport from opossum kidney cells in Xenopus laevis oocytes

Xenopus laevis oocytes have been used for the expression of Na/Pi-cotransport activity by injections of poly(A)+ RNA (mRNA) isolated from an established renal cell line (OK cells). 3-5 days after mRNA injection, Na-dependent phosphate (Pi) uptake by oocytes was increased in a dose-dependent manner; there was no increase in Na-independent Pi uptake. Sucrose density-gradient fractionation indicated that the mRNA species encoding this activity is 2.4-2.8 kb in length. In Northern blots, using a cDNA probe related to human kidney-cortex Na/Pi-cotransport activity (NaPi-3), hybridization with a mRNA-species of 2.4-2.6 kb was obtained. Kinetic characterization ([Pi], [Na]) showed that expressed transport activity has properties similar to apical Na/Pi cotransport in OK cells.

Localization of NaPi-1, a Na-Pi cotransporter, in rabbit kidney proximal tubules. I. mRNA localization by reverse transcription/polymerase chain reaction

We have recently isolated from a rabbit cortex cDNA library a cDNA clone (NaPi-1), which, after in vitro transcription (cRNA) and injection into Xenopus laevis oocytes, expresses Na-dependent Pi uptake [Werner A, et al. (1991) Proc Natl Acad Sci USA 88:9608-9612]. The aim of the present work was to study the nephron location of the NaPi-1-related mRNA(s) by combining nephron microdissection procedures, reverse transcription (RT) and amplification of the resultant cDNA by the polymerase chain reaction (PCR). RT-PCR using NaPi-1-specific primers (different combinations) and either total kidney cortex RNA or microdissected proximal tubule segments resulted in two PCR products, both of approximately the expected length (but differing by about 30 base pairs). Restriction-enzyme analysis and nucleotide sequencing confirmed that both PCR products are related to NaPi-1 and that the "longer" PCR product has an insert of 26 base pairs containing an AluI restriction site. Nephron microdissection documents expression of NaPi-1-related mRNA(s) in superficial and deep proximal tubules (S1, S2 and S3 segments) and their absence in glomeruli, thin descending limb and thick ascending limbs of Henle's loop, distal convoluted tubules and cortical and inner medullary collecting ducts. These experiments suggest a "microheterogeneity" of NaPi-1-related mRNA(s) (which is not detected in Northern blot analysis) and proximal tubular expression of NaPi-1.

High salt alone does not influence the kinetics of the Na(+)-H+ antiporter

During a high-salt diet, tubular sodium reabsorption is decreased. This study concerns the effect of a high-salt diet on the proximal tubular (PT) Na+ influx pathways. Brush-border membrane vesicles (BBMV) were prepared from rats on normal-salt (NS) and rats on high-salt (HS) diets. The initial uptake rates of Na+ were the same in NS and HS rats, both in the absence and the presence of 1 mM amiloride. Vmax and Km for the amiloride-sensitive Na+/H+ antiporter were also the same in the NS (Vmax 3.69 +/- 0.31 nmol mg prot-1 10 s-1, Km 6.13 +/- 0.58 mM) and HS groups (Vmax 3.54 +/- 0.28 nmol mg prot-1 10 s-1, Km 6.18 +/- 0.64 mM). There was no difference in the initial uptake rates of the Na(+)-glucose and the Na(+)-alanine symporters in NS and HS. Vmax and Km for the L-dopa-Na+ symporter were also the same in NS (Vmax 72 +/- 2.5 pmol mg prot-1 20 s-1, Km 98 +/- 14 microM) and HS groups (Vmax 78 +/- 6.0 pmol mg prot-1 20 s-1, Km 106 +/- 4 microM). In summary, HS diet does not change the kinetics of the Na+ transporters in the brush-border membrane of PT cells.

Reconstitution and characterization of a Na+/Pi co-transporter protein from rabbit kidney brush-border membranes

A protein with Na+/Pi co-transporter activity has been extracted from rabbit brush-border membranes with chloroform/methanol and purified by hydroxyapatite chromatography. The protein has been incorporated by the dilution method into liposomes formed from different types and ratios of lipids. The greatest reconstitution has been achieved into liposomes prepared from cholesterol (20%), phosphatidylcholine (20%), phosphatidylethanolamine (30%) and phosphatidylserine (30%) (CH/PC/PE/PS). Pi uptake by these proteoliposomes had the following characteristics: (i) the initial rate was markedly greater in the presence of an inwardly directed Na+ gradient (600 pmol/10 s per mg) than with a K+ gradient (65 pmol/10 s per mg); (ii) maximal uptake was increased 8-fold above the equilibrium value ('overshoot') when a Na+ gradient was applied; (iii) Pi was not merely bound to proteoliposomes but was transported intravesicularly; and (iv) Na(+)-dependent Pi uptake was sensitive to the known phosphate transport inhibitors. This first successful attempt of reconstitution of Na+/Pi transport activity into proteoliposomes led us to isolate and characterize physico-chemically the protein responsible. Its isoelectric point was about 5.8, and urea/SDS gel electrophoresis revealed a broad band of molecular mass ranging from 63 to 66 kDa under both reducing and non-reducing conditions. In the native form, the molecular mass analysed by gel filtration was estimated to be 170 +/- 10 kDa, suggesting that the protein is a polymer, probably stabilized by hydrophobic bonds. Endoglycosidase F treatment decreased the molecular mass to approx. 50 kDa. It is postulated that this acidic glycoprotein might represent a subunit of the intact Na+/Pi co-transporter from rabbit kidney brush-border membranes.

Sequential and precise in vivo measurement of bone mineral density in rats using dual-energy x-ray absorptiometry

In the design of new strategies for the treatment of osteoporosis, noninvasive, precise, and sensitive bone mass measurement capable of detecting changes over short periods of time in small animals is essential. Most of the models described thus far require the sacrifice of the animals and/or display low reproducibility. Using a dual-energy x-ray absorptiometer (DEXA; Hologic QDR-1000) in an ultrahigh-resolution mode, we measured bone mineral density (BMD) in rats at the levels of lumbar spine (L1-4), proximal tail (caudal vertebrae C2-4), and tibia. Accuracy was evaluated by measuring the mineral content of bone powder capsules (within the range of rat vertebrae BMD), under 0.5-3 cm water to mimic variations in soft tissue thickness. The bone powder capsule mineral content was highly correlated with chemically determined hydroxyapatite content (r = 0.999). In vivo reproducibility was evaluated by calculating the coefficient of variation (CV = 100 x SD/mean) of four to six BMD measurements, each time with repositioning, in seven rats (220-500 g body weight). CV was 1.36 +/- 0.32% (x +/- SD) for lumbar spine, 0.66 +/- 0.50% for proximal tail, and 1.12 +/- 0.45% for tibia. The ability to detect BMD changes was investigated by measuring BMD before and every 4 weeks after ovariectomy (OVX) in 270 g rats, pair fed during the whole experiment. Compared with sham-operated control animals, a highly significant difference in lumbar spine BMD was observed 4 weeks after OVX, which reached a maximum by 8 weeks and remained stable thereafter. At the level of the proximal tibia, the difference was maximal 4 weeks after OVX.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of Mg2+ and ATP on the phosphate transporter of sarcoplasmic reticulum

The extra uptake of Ca2+ by vesicles of sarcoplasmic reticulum (SR) observed in the presence of Pi, attributable to transport of Pi by the Pi-transporter, has been studied. It has been shown that the Pi transporter is stimulated by ATP. Single channel conductance measurements have shown that the Cl- channel in the SR membrane is impermeable to Pi. It is suggested that the transporter could be an ion antiporter system. Studies of uptake as a function of pH and Mg2+ concentration suggest that transport of MgHPO4 and H2PO-4 are faster than transport of HPO2-4. For oxalate and pyrophosphate, Mg2+ binding inhibits transport. It is suggested that protonation of lysine residue(s) at the anion binding site increase the rate of transport.

Cellular mechanisms in proximal tubular reabsorption of inorganic phosphate

Filtered inorganic phosphate (Pi) is largely reabsorbed in the proximal tubule. Na-Pi cotransport, with a stoichiometry of at least 2:1, mediates uphill transport at the apical membrane; at the basolateral membrane different types of transport systems can be involved in efflux and uptake of Pi from the interstitium. Regulation of transcellular Pi flux involves alteration of the apical Na-Pi cotransport; at least three different cellular control/sensing systems seem to participate in this regulation and are exemplified by parathyroid hormone (PTH)-dependent inhibition, Pi deprivation-dependent increase, and insulin-like growth factor I (IGF-I)-dependent increase in Na-Pi cotransport. For PTH inhibition, recent evidence suggests a role of the phospholipase C/protein kinase C-dependent regulatory cascade in inhibition of Na-Pi cotransport, at least at low PTH concentrations. In addition, an endocytic mechanism seems to be involved in this PTH action. Little is known of the cellular mechanisms in Pi deprivation-dependent and/or IGF-I-dependent increases in Na-Pi cotransport; they are dependent on de novo protein synthesis. Recent experiments involving an expression in Xenopus laevis oocytes led to the identification of an approximately 50 kDa membrane protein that is a good candidate for being involved in brush-border membrane Na-Pi cotransport activity.

Familial renal hypophosphatemia, minor facial anomalies, intracerebral calcifications, and non-rachitic bone changes: apparently new syndrome?

We report on two brothers with renal hypophosphatemia, intracerebral calcifications, minor facial anomalies, and short distal phalanges. The children presented with recurrent dental abscesses; one had premature closure of the anterior fontanelle. Biochemical findings included hypophosphatemia and elevated serum alkaline phosphatase with normocalcemia. Blood levels of parathyroid hormone, 1,25(OH)2 and 25(OH) vitamin D levels were normal; TRP (the fractional tubular reabsorption of PO4) and TmP/GFR (the tubular maximum rate of PO4 reabsorption in relation to GFR) were low. Both parents had a normal serum phosphate and brain CT scan without evidence of calcifications. This apparently new syndrome of renal hypophosphatemia associated with intracerebral calcifications appears to be inherited as either an autosomal recessive or an X-linked trait.

Thiol redox and phosphate transport in renal brush-border membrane. Effect of nicotinamide

In the present study, the effect of thiol redox and its possible role in the inhibitory effect of nicotinamide on renal brush-border membrane (BBM) phosphate uptake was examined. Addition of thiol reducing agent, dithiothreitol (DTT, 5 mM), caused an increase, while addition of thiol oxidant, diamide (DM, 5 mM) caused a reversible decrease in sodium-dependent BBM phosphate uptake. Kinetic analyses revealed an increase in both Vmax and Km by DTT, and a decrease in Vmax by DM. These results suggest that thiol redox influences BBM phosphate uptake with sulfhydryl (SH) groups relate to its capacity and disulfide (SS) groups to its affinity for phosphate. Since changes in cytosolic NAD levels may affect BBM thiol redox through changes in redox states of NADP and glutathione systems, we have examined such possibility by studying the effect of nicotinamide (NM). Incubation of proximal tubules with NM (10 mM) induced an oxidative effect on redox states of cytosolic NAD, NADP systems as inferred from decreased cellular lactate/pyruvate, malate/pyruvate, respectively. Measurements of cytosolic glutathiones and BBM thiols also revealed that NM pretreatment shifted the cytosolic glutathione redox (GSH/GSSG) and BBM thiol redox (SH/SS) toward more oxidized state. On the other hand, incubation of proximal tubules with NM suppressed phosphate uptake by the subsequently isolated BBM vesicles. The lower phosphate uptake by NM-pretreated BBM vesicles was reversed by DTT and was resistant to the inhibitory effect of DM. These results thus suggest that BBM thiol oxidation may be involved in the inhibitory effect of NM on BBM phosphate uptake.

Insulin in the acute renal adaptation to dietary phosphate restriction in the rat

Dietary phosphate restriction produces a rapid increase in tubular reabsorption of phosphate. To evaluate whether insulin is important in the acute renal adaptation following a low phosphate meal, four groups of conscious rats were studied by renal clearance methods, following a single meal by gavage. Group A received a normal (0.8%) phosphate meal, followed by saline infusion; Group B, a low (0.03%) phosphate meal, followed by saline infusion; Group C, a low phosphate meal, followed by infusion of somatostatin to suppress endogenous insulin secretion; and Group D, a low phosphate meal, followed by infusion of somatostatin plus insulin. Baseline plasma phosphate, insulin, glomerular filtration rate, and fractional excretion of phosphate were similar in all four groups. Following a low phosphate meal in Groups B, C, and D, there was a decrease in plasma phosphate, as compared with Group A. Whereas fractional excretion of phosphate decreased when plasma phosphate fell in Group B, administration of somatostatin (Group C) prevented the drop in fractional excretion of phosphate, despite a lower plasma phosphate. The addition of exogenous insulin (Group D) restored the antiphosphaturic effect of the low phosphate meal. These results suggest that insulin contributes to the acute decrease in phosphate excretion following a low phosphate meal.

Cellular aspects of proximal tubular phosphate reabsorption

In vivo manipulations to alter renal Pi reabsorption and the subsequent isolation of proximal tubular brush border membrane vesicles have greatly increased our knowledge about the regulation of renal Pi reabsorption via the Na+/Pi cotransport system. Only recently, direct biochemical and cell-biological access has become possible by the use of established and primary cell cultures. Based on the results obtained with isolated brush border membranes and cultured cells, a model has been presented, which might serve as a basis for future research of the regulatory control mechanisms of the renal Na+/Pi cotransport. At present, a major drawback is the fact that the molecular identity of the Na+/Pi cotransport system is still unknown. The identification of this transport system would certainly be a great step and would allow to verify or falsify one or the other hypotheses postulated in the past few years for the regulatory control mechanism(s) of the renal Na+/Pi cotransport.

The Na+/Pi-cotransporter of OK cells: reaction and tentative identification with N-acetylimidazole

Using an established renal epithelial cell line (OK cells) the effect of the amino-acid side-chain modifying reagent N-acetylimidazole (NAI) upon the sodium-dependent transport of phosphate (Pi) was investigated. After an incubation with 10 mM NAI for 20 min, cellular Na+/Pi uptake was inhibited by 70%. The presence of 5 mM Pi protected this transport function from being affected by NAI by 80 to 100%. Since the presence of sulfate was unable to protect the Na+/Pi transport inactivation by NAI and since the presence of Pi did not affect NAI inhibition of other transport systems, it is suggested that NAI interacts with the Pi transporter directly. The protective effect of Pi was used as a criterion to identify Pi-protectable [3H]NAI labelling of OK cell plasma membrane proteins. Pi protection was observed in four molecular mass regions: 31, 53, 104 and 176 kDa. Since the incorporation of [3H]NAI into these proteins was also affected by parathyroid hormone at 10(-10) M, it is concluded that the identified proteins represent possible candidates for the renal Na+/Pi cotransporter.

Vitamin D3 and the renal handling of phosphate in American eels

Cholecalciferol (Vitamin D3) increased plasma inorganic phosphate concentration in American eels,Anguilla rostrata, in a dose-dependent fashion. This response was more marked in phosphate loaded fish. In control as well as phosphate loaded eels the hyperphosphatemic response to D3 was associated with a sharp reduction in renal phosphate clearance relative to(14)C-polyethelene glycol (PEG) clearance. Glomerular filtration and urine flow rates were not affected by D3. As renal phosphate clearance, even in phosphate loaded eels, never significantly exceeded that of PEG, it is suggested that D3 reduced the relative clearance rate of phosphate by increasing renal phosphate reabsorption rather than by reducing the tubular secretion of phosphate.

Glutathione-dependent inactivation of sodium-dependent phosphate transport across rat renal brush-border membrane

Thiol/disulfide is fundamental in protein function; we previously observed an inhibitory effect of thiol oxidants on the Na-dependent phosphate (Pi) uptake into renal brush border membrane vesicles (BBMV). We examined whether oxidation of glutathione (GSH) is involved in the mechanism. Vesicular thiols were measured by liquid chromatography. BBMV were incubated with reagents before an influx of Pi. Diamide (5 mM) reduced the capacity of the Pi uptake. Subsequent treatment with dithiothreitol (5 mM) blocked the inhibitory effect of diamide. Vesicular GSH was not modified only by the incubation, whereas it was oxidized by the treatment with diamide, and reduced by dithiothreitol. Furthermore, in vivo treatment with cAMP provided GSH-depleted BBMV without any influence on Pi uptake. Diamide did not inhibit the transport of Pi into GSH-depleted vesicles, but it did inhibit the uptake when GSH was introduced into the vesicles. In conclusion, a GSH-dependent mechanism is involved in the inhibitory effect of diamide on sodium-dependent Pi transport across the renal brush-border membrane.

Factor derived from human lung carcinoma associated with hypercalcemia mimics the effects of parathyroid hormone on phosphate transport in cultured renal epithelia

A decrease in renal tubular reabsorption of inorganic phosphate (Pi) can be observed in hypercalcemia of malignancy. In the present study we investigated the effect of serum-free conditioned medium (CM) from cells, derived from a lung carcinoma (BEN) of a hypercalcemic patient, and of PTH on cyclic AMP (cAMP) production and sodium-dependent Pi transport (NaPiT) in epithelia of two renal cell lines. In opossum kidney cells (OK), PTH is known to enhance cAMP production and inhibit NaPiT; in contrast, in LLC-PK1 cells, PTH has no effect on NaPiT since this kidney cell line is devoid of PTH receptors. In OK cells, BEN CM induced a three- to fourfold increase of cAMP production, which was blunted by the PTH inhibitors bPTH(3-34) and bPTH(7-34). NaPiT, as assessed by measuring the initial rate of Pi uptake, was inhibited in a dose-dependent manner by BEN CM, with an effect maximal between 1h30 and 6 hr of incubation (40 +/- 4% and 47 +/- 4%, respectively), corresponding to the effect produced by 1-3 nM bPTH(1-34). The Na-dependent transport of a glucose analog was affected neither by BEN CM nor by PTH. In LLC-PK1 cells, neither BEN CM nor PTH altered cAMP production nor NaPiT after 1h30 of incubation. At 6 hr, BEN CM caused a slight decrease in NaPiT. In conclusion, these results constitute the first evidence of a direct and selective inhibition by tumor-derived factor(s) of NaPiT in cultured renal epithelia. Most of the renal NaPiT inhibitory activity produced by the lung tumor required the presence of a PTH receptor-adenylate cyclase system.(ABSTRACT TRUNCATED AT 250 WORDS)

Transport of sulfate and phosphate in small intestine and renal proximal tubule: methods and basic properties

1. Isolated brush border membrane vesicles, basolateral membrane vesicles, and cultured renal epithelial cells provide good material for studying transport systems. 2. The vesicle systems have been used to study the transport of labeled phosphate, sodium/phosphate cotransport, sodium/sulfate cotransport, basolateral transport of sulfate and basolateral transport of phosphate via anion exchange. 3. Cultured renal cells show sodium/phosphate cotransport and parathyroid dependent inhibition of phosphate transport.

Medical reversal of acquired hypophosphatemic osteomalacia

Hypophosphatemic osteomalacia may present as severe disability from bone disease. This report describes a patient with long-standing disease and multiple fractures. Medical management of the phosphate loss may be successful in promoting bone healing when it is not possible to establish the cause of the phosphaturia. Judicious increases in calcium, 1,25-dihydroxyvitamin D, and phosphorus supplements were carefully monitored to avoid failure of therapy or hypercalcemic complications from pharmacologic amounts of these supplements.

Phosphate transport in brush-border membranes from control and rachitic pig kidney and small intestine

1. Na-Pi co-transport was analysed using renal cortical and small intestinal brush-border membrane vesicles which were isolated from control (normal, heterozygotes) and rachitic piglets (homozygotes). 2. A kinetic analysis of Na-dependent initial linear uptake of Pi was performed using vesicles obtained from control animals. The results suggest similar kinetic properties for the renal and small intestinal co-transport system. (i) A sigmoidal dependence on Na concentration of Pi uptake suggests the involvement of more than one Na ion in the co-transport. (ii) Increasing Na concentration leads to an increase in the apparent affinity of the transport system for Pi and has minimal effect on the apparent Vmax (maximum velocity of uptake). (iii) Increasing pH leads to an increase in Pi transport rate. 3. The kinetic characteristics of the Na-Pi co-transport system in vesicles obtained from rachitic animals were similar to those in controls. The apparent Vmax, but not the apparent Km (Michaelis constant) for Na and Pi, is reduced in intestinal and renal brush-border membranes isolated from rachitic animals as compared to control animals. Injection of vitamin D3, three days prior to killing of rachitic litter-mates, increased the Na-Pi uptake rate in the brush-border membrane vesicles isolated from these piglets. 4. It is concluded that intestinal and renal brush-border membranes from piglets contain a similar Na-Pi co-transport system and that in vitamin-D-dependent rickets the number of operating transport units is reduced in both membranes.

Intracellular regulatory cascades: examples from parathyroid hormone regulation of renal phosphate transport

The knowledge about intracellular regulatory cascades in hormone action has increased considerably over the last few years. Receptor occupation at the plasma membrane level results in a production of intracellular messengers, such as cyclic nucleotides (cAMP, cGMP), inositoltrisphosphate (IP3), diacylglycerol (DAG) and a rise in cytosolic calcium concentration. These messengers control the activity of different regulatory mechanisms which operate either in sequence or in parallel to generate the final biological response. In PTH-dependent regulation of renal phosphate transport, cAMP-dependent and calcium-dependent mechanisms are involved: Recent experiments with cultured renal epithelial cells have confirmed that activation of adenylate cyclase is the initial event. However, the cAMP signal can be bypassed and direct activation of protein kinase C seems to mimic PTH induced inhibition of phosphate transport. The final event in the regulatory cascade is most likely a removal of the phosphate transport system followed by a degradation.

Current concepts of regulation of phosphate transport in renal proximal tubules

The wealth of new information on BBM transport of Pi which has accumulated in recent years gives an indication of the importance and intellectual challenge that the mechanism of this process poses to investigators. In this brief reflection on the field, we have tried to draw attention to some general principles and features which may be helpful as working hypotheses in the development of the field. To date, a disproportionate amount of effort may have been spent on deciphering putative intracellular regulatory mechanisms, without knowing some essential fundamental properties of the Na+-Pi-COT. We suggest that a major effort should be exerted towards elucidating biogenesis of the Na+-Pi-COT, the possible existence of a membrane cycling mechanism, and a refined analysis of the Na+-Pi-COT in specific subsegments of proximal tubules. Advances in these areas together with studies of both the rapid and long-term adaptive regulation of Pi transport are needed, given the central role of the kidney in total body Pi homeostasis both in health and disease.

Characterization of Na+-dependent phosphate uptake in cultured kidney cells (JTC-12) from monkey

Phosphate uptake was studied in confluent monolayers of an epithelial-cell line (JTC-12) derived from monkey kidney. Phosphate uptake consisted of a saturable, Na+-dependent, component, which accounted for about 80% of the uptake, and a nonsaturable, Na+-independent, component. The saturable component was specifically dependent on the presence of extracellular Na+ and has an apparent Km value for phosphate of 0.12 mM at 137-mM-Na+, which is close to those reported in the brush-border membranes in mammalian kidneys. The presence of Na+ in the uptake solution decreased the Km for phosphate without affecting the Vmax. Phosphate uptake was inhibited by carbonyl cyanide p-trifluoromethoxyphenylhydrazone and ouabain, suggesting that phosphate transport is an active, energy-dependent, process and is dependent on an Na+ gradient across cell membranes. With respect to the effect of external Na+ concentration, a sigmoid relation was seen between the initial velocity of phosphate uptake and Na+ concentrations, and Hill analysis gave a Hill coefficient of 1.8. In the pH range 6.6-7.4, phosphate uptake declined with increasing pH. Phosphate uptake was stimulated when cells were cultured in the presence of insulin, and was also affected by changes in phosphate concentrations in cultured medium. These results indicate that JTC-12 cells have an Na+-dependent phosphate-transport system with many of the features of phosphate transport in the proximal tubule.

Effects of fasting compared to low phosphorus diet on the kinetics of phosphate transport by renal brush-border membranes

Changes in the kinetics of sodium gradient-dependent brush border Pi transport in response to dietary phosphorus deprivation were analysed using initial rate conditions. In rats adapted to low phosphorus diet the apparent Vmax, determined from a double-reciprocal plot, was increased 2-fold but the apparent Km was not different compared to control rats fed normal phosphorus diet. In contrast when renal adaptation to low phosphorus diet was reversed by fasting the apparent Vmax was not significantly different but the apparent Km was increased 5-fold. The results suggest that regulation of renal Pi transport in vivo may occur not only through changes in the apparent Vmax of the brush border Pi transport system but also, in certain circumstances, through changes in the apparent Km.

Regulation of bone mineral loss during lactation

The effects of varying dietary calcium and phosphorus content, vitamin D deficiency, oophorectomy, adrenalectomy, and simultaneous pregnancy on bone mineral loss during lactation were examined in rats. Unless otherwise stated, the diet contained 0.47% calcium and 0.3% phosphorus and the rats were given 26 nmol of vitamin D3. Femur ash weights were determined after 21 days of lactation and on age-matched nonlactating rats. Decreasing dietary calcium to 0.02% caused an increased loss of bone mineral, whereas increasing dietary calcium to 1.4% increased plasma calcium levels to 12 mg/100 ml but did not diminish the bone mineral loss observed during lactation. Varying dietary phosphorus did not have a major effect on bone mineral loss during lactation. In vitamin D-deficient rats, bone mineral loss during lactation was independent of dietary calcium levels and slightly greater than the loss observed in vitamin D-replete rats fed the normal calcium diet. Oophorectomy and adrenalectomy did not produce changes in femur ash weights of nonlactating rats or reduce bone mineral loss during lactation. Rats mated during their postpartum estrus and thus simultaneously pregnant and lactating, lost the same amount of bone mineral as caused by lactation alone.