Role of Thyroid Hormone in Regulation of Renal Phosphate Transport in Young and Aged Rats

Role of Thyroid Hormone in Regulation of RenalPhosphate Transport in Young and Aged Rats*

ANA I. ALCALDE, MANUEL SARASA, DEMETRIO RALDUA, JOSE ARAMAYONA,ROSA MORALES, JURG BIBER, HEINI MURER, MOSHE LEVI, AND

VıCTOR SORRIBAS

Departments of Physiology (A.I.A.), Anatomy (M.S., D.D.), Toxicology (R.M., V.S.), and Pharmacology(J.A.), University of Zaragoza, 50013 Zaragoza, Spain; Institute of Physiology (J.B., H.M.), Universityof Zurich-Irchel, CH-8057 Zurich, Switzerland; and Department of Internal Medicine (M.L.),University of Texas Southwestern Medical Center and Department of Veterans Affairs Medical Center,Dallas, Texas 75216

ABSTRACTIn the present study, we have examined the cellular mechanisms

mediating the regulation of renal proximal tubular sodium-coupledinorganic phosphate (Na/Pi) transport by thyroid hormone (T3) inyoung and aged rats. Young hypothyroid rats showed a marked de-crease in Na/Pi cotransport activity, which was associated with par-allel decreases in type II Na/Pi cotransporter (NaPi-2) protein andmessenger RNA (mRNA) abundance. In contrast, administration oflong-term physiological and supraphysiological doses of T3 resulted insignificant increases in Na/Pi cotransport activity, protein, andmRNA levels. Nuclear run-on experiments indicated that thyroidhormone regulates NaPi-2 mRNA levels by a transcriptional mech-

anism. In aged rats, although there were no changes in T3 serumlevels (when compared with young animals), there were significantdecreases in serum Pi concentration, renal Na/Pi cotransport activity,and NaPi-2 protein and mRNA abundance. These effects were me-diated, at least in part, by a reduction in the transcriptional rate ofthe NaPi-2 gene, probably caused by, among other factors, a smallerresponse to the stimulatory action of T3. Compared with young rats,the old rats exhibited less sensitivity of the Na/Pi cotransporter tothyroid hormone, with decreased effects in both hypothyroid (inhib-itory) and hyperthyroid (stimulatory) animals. (Endocrinology 140:1544–1551, 1999)

HIGHER ORGANISMS use inorganic phosphorus (Pi)for several vital functions, including bone matrix and

phospholipid synthesis, blood buffering, intracellular signaltransduction, synthesis of energetic bonds in nucleotides,and others. These important and multiple functions implythat organisms must have precise and efficient mechanismsfor controlling Pi homeostasis. Intestinal absorption and re-nal excretion/reabsorption of Pi are the two main targets forthe mechanisms of control (e.g. Refs. 1 and 2). Whereas in-testinal absorption has only a role on long-term regulation ofPi homeostasis, the control of renal excretion/reabsorption isimportant for both short-term and long-term regulation of Pi

homeostasis (3).Short-term regulation of Pi reabsorption is mainly medi-

ated by PTH and alterations in dietary Pi intake, both mech-anisms involving shuttling or recycling of Na/Pi cotrans-porter-containing vesicles between the cytoplasm and thebrush border membrane (BBM). Long-term regulation in-cludes many different hormonal and nonhormonal mecha-nisms that alter the cell content of the specific messengerRNA (mRNA) (3). To understand properly the complex net-work controlling Pi renal excretion/reabsorption, it is im-

portant to determine precisely the physiological role of eachone of these regulatory mechanisms.

Because Pi is intensively used in general metabolism, Pi

homeostasis should be regulated by factors controlling therate of metabolism itself. One such factor is thyroid hormone,and its role in Pi reabsorption regulation has been extensivelyanalyzed (4–6). Pharmacological doses of T3 have beenshown to increase Na/Pi cotransport in BBM vesicles fromrat proximal tubules (4, 5). In addition, T3 concentrationsapproximating the association constant (Km) of the thyroidhormone nuclear receptor also elicited a similar increase inPi transport in opossum kidney (OK) cells (6). In both cases,the increase in transport rate was caused by an increase in thecapacity of the transport system, whereas the affinity was notmodified. Recently, Euzet et al. (8, 9) have shown an impor-tant role for T3 in the maturation of the renal Na/Pi cotrans-porter, which was associated with changes in both Km andVmax, as well as in the recently cloned type II Na/Pi cotrans-porter (NaPi-2) (7) protein and mRNA abundance.

In the present study, we have determined the role of phys-iological concentration of thyroid hormone in renal phos-phate transport in vivo; more exactly, the molecular mecha-nisms of enhancement of phosphate reabsorption by thyroidhormone. In addition, we have tried to determine the po-tential role of thyroid hormone in impairment of phosphatereabsorption that accompanies the aging kidney. Our resultsshow that chronically treated hypothyroid rats, using a phys-iological dose of T3, exhibit increases in phosphatemia,NaPi-2 mRNA and protein content, and Na/Pi cotransport in

Received September 9, 1998.Address all correspondence and requests for reprints to: Vıctor Sor-

ribas, Ph.D., Departamento de Toxicologıa, Facultad de Veterinaria,Universidad de Zaragoza, Calle Miguel Servet, 177, E-50013 Zaragoza,Spain. E-mail: [email protected].

* This work was supported by Grant PB93–0585 from the SpanishMinister of Education and Science (to V.S.).

0013-7227/99/$03.00/0 Vol. 140, No. 4Endocrinology Printed in U.S.A.Copyright © 1999 by The Endocrine Society

1544

superficial and juxtamedullary renal cortex, all these effectsby means of enhanced transcription of the correspondingNaPi-2 gene. The stimulatory effect of the hormone is lessevident in the aging kidney, which shows a lower level ofbasal phosphate reabsorption. In our experimental design,only pharmacological hyperthyroidism is able to restore par-tially the level of phosphatemia observed in young animals.

Materials and MethodsAnimals and experimental designs

All studies were conducted in accordance with the European Unionlegislation and the NIH Guide for the Care and Use of LaboratoryAnimals. The experiments were performed with 3-month-old (young) or24-month-old (old) male Wistar rats. Animals were thyroparathyroid-ectomized (TPTX) in our laboratory and were left for 1 month in theircages to reach an approximately zero concentration of T3 in blood serumand kidney. These animals were drinking water containing calciumgluconate (350 mg Ca21/liter). After this time, rats were divided intofour groups: group A, nonoperated control rats; group B, TPTX-hypo-thyroid rats, treated during 20 days with sc 21-day release placebopellets of T3, obtained from Innovative Research of America(IRA, Sara-sota, FL); group C, TPTX-euthyroid rats, treated during 20 days with aphysiological dose of T3, 0.5 mg/100 mg BWzday by using sc pellets of21-day release (IRA); group D, TPTX-hyperthyroid rats, treated for 20days with 5 times the physiological dose, 2.5 mg T3/100 mg BWzday, withsc pellets (IRA). In some experiments, a treatment with 10 times thephysiological dose of T3 was also performed (group E, 5 mg T3/100 mgBWzday). In the experiments with old animals, group C was excluded,because we did not find differences with group A old animals; instead,group E was always used in experiments with old rats. Normal levelsof T3 in kidneys of hypothyroid rats are reached after 12 days of treat-ment with 0.5 mg T3/100 mg BWzday (personal communication,G. Morreale de Escobar, Institute for Biomedical Research-Consejo Su-perior de Investigaciones Cientifica, Madrid, Spain). On the day of theexperiment, 24-h urine was collected, the animals were euthanatized byCO2 narcosis, and blood was drawn from the aorta. TPTX condition wassystematically checked by postmortem inspection of the thyroid gland,weight changes of the animals, and RIA of T3 and T4 serum levels. Urineand serum Pi, calcium, and creatinine were measured as previouslydescribed (10). Fractional excretion of Pi (FEPi) was calculated as (UPi/SPi)/(UCr/SCr), where U is urine, S is serum, and Cr is creatinine.

Thyroid hormone assays

T3 was measured in whole plasma by specific and highly sensitiveRIA, as described (11). The standard curve is prepared using plasmafrom severely hypothyroid thyroidectomized rats, the limit of detectionbeing 15 ng T3/dl.

BBM preparations

On the day of the experiment and after CO2 narcosis, blood wasdrawn from the aorta, and the kidneys were rapidly removed. Thin sliceswere cut, at 4 C, from the superficial and juxtamedullary cortex and werehomogenized with a Disperser DIAX 600 (Heidolph, Kelheim, Ger-many) in a buffer consisting of the following (in mm): 300 DL-mannitol,5 EGTA, 0.5 phenylmethylsulfonyl fluoride, and 16 HEPES (pH 7.5 withTris). BBM were purified from this homogenate by Mg21 precipitationand differential centrifugation, as described (12). The final pellet wasresuspended in a buffer of 300 mm mannitol and 16 mm HEPES-Tris (pH7.5). Purity of BBM preparations was enzymatically assayed as described(12).

Transport assays

Sodium gradient-dependent phosphate transport (Na/Pi cotrans-port) measurements were performed, in freshly isolated BBM vesicles,by uptake of 0.1 mm PO4 (a mix of K2HPO4 plus KH2PO4, pH 7.4), plusK2H32PO4 (DuPont NEN Research Products, Boston, MA) as radio tracer

(4 mCi/ml uptake medium, 3,000 Ci/mmol) and an inwardly directedsodium gradient (120 mm NaCl), followed by rapid filtration. Uptakewas measured at 10 sec, which still represents the initial linear phase ofphosphate transport at 25 C.

SDS-PAGE and immunoblots

Aliquots of superficial and juxtamedullary BBM vesicles were dena-tured for 2 min at 95 C in 2% SDS, 10% glycerol, 0.5 mm EDTA, and 95mm Tris-HCl (pH 6.8) (final concentrations), and 10 mg BBM protein perlane were separated on 10% polyacrylamide gels according to themethod of Laemmli (13) and electrotransferred on nitrocellulose paper(14). Immunodetection with antiserum against NaPi-2 (15) and sodium-sulfate cotransporter NaSi-1 (16) was performed by chemiluminescenceusing the BM Western Blotting Kit from Boehringer Mannheim (Mann-heim, Germany) and visualized with x-ray films (Hyperfilm MP) fromAmersham International (Buckinghamshire, UK). Image analysis andquantification were done with a Gel Doc 1000 Video Gel DocumentationSystem (Bio-Rad Laboratories, Inc., Hercules, CA).

Immunohistochemistry

These experiments were performed as described (10). Briefly, anes-thetized rats were perfused retrogradely with a fixative of paraformal-dehyde and picrinic acid, and 5-mm-thick sections were cut at 220 Cwith a cryostat. Anti-NaPi-2 antibody was used at 1:500 dilution inPBS/milk powder buffer and detected with goat antirabbit IgG antibodylinked to Cy2 (Amersham International) at 1:200 dilution. In some ex-periments, b-actin was visualized with rhodamin-conjugated phalloidin(Sigma Chemical Co., St. Louis, MO) at 1:500 dilution in PBS/milkpowder. Sections were coverslipped using mounting media plus 2.5%1,4-diazabi-cyclo{2.2.2}octane (DABCO; Sigma Chemical Co.) as a fadingretardant and studied with epifluorescence microscopy using a narrow-band filter system for fluorescein isothiocyanate (BX60, Olympus Op-tical Co., Tokyo, Japan) or a confocal Zeiss LSM 410 (Carl Zeiss, Jana,Germany).

RNA isolation and Northern blotting

Superficial and juxtamedullary cortex were cut out of the kidney at4 C and homogenized with a Disperser DIAX 600 in a denaturationsolution containing 4 m guanidium thiocyanate, 25 mm sodium citrate(pH 7.0), 0.5% sarcosyl, and 0.1 m 2-mercaptoethanol. RNA was ex-tracted by the guanidium thiocyanate-phenol acid-chloroform method,as described (10, 17). Twenty micrograms of total RNA were denatured,electrophoresed in a formaldehyde agarose gel, transferred onto Hy-bond N1 nylon membranes (Amersham International) and UV cross-linked (UVC 500, Hoefer Pharmacia Biotech Inc., San Francisco, CA).Full-length complementary DNA (cDNA) probes of NaPi-2, NaSi-1,b-actin, and 18S ribosomic RNA were labeled with [a-32P]-deoxycyti-dine triphosphate (3,000 Ci/mmol) by random priming (RadPrime DNALabeling System; Gibco BRL-Life Technologies, Grand Island, NY), andhybridization was carried out at high stringency, as described (18).Signals were quantified by image analysis with Gel Doc 1000 Video GelDocumentation System (Bio-Rad Laboratories, Inc.) after exposure tox-ray films (Hyperfilm MP).

Nuclear run-on

Cell nuclei were isolated from superficial and juxtamedullary cortex,as described (19). Briefly, 0.5 g tissue was mixed with 5 ml lysis buffer(0.32 m sucrose, 5 mm CaCl2, 3 mm magnesium acetate, 0.1 mm EDTA,0.1% Triton X-100, 1 mm dithiothreitol (DTT), 1 mm Tris-Cl, pH 8.0) andhomogenized in a loose-fitting pestle. After filtration through severallayers of cheesecloth, the filtrate was rehomogenized with 10 strokes ina tightly fitting pestle. The homogenate was mixed with 1 vol 2 m sucrosesolution (2 m sucrose, 3 mm magnesium acetate, 0.1 mm EDTA, 1 mmDTT, 10 mm Tris-Cl, pH 8.0) and centrifuged over a cushion of the samesolution at 30,000 3 g for 45 min at 4 C. Nuclei were resuspended inglycerol storage buffer (40% glycerol, 5 mm magnesium acetate, 0.1 mmEDTA, 5 mm DTT, 50 mm Tris-Cl, pH 8.0) at 100 3 106 nuclei/ml.Usually, about 30 million nuclei were obtained per 0.5 g renal tissue.Labeling of nascent RNA transcripts was performed exactly as described

THYROID HORMONE MODULATES PHOSPHATE TRANSPORT 1545

(19). A final heterogeneous RNA probe of 8–10 3 106 cpm/2 ml hy-bridization solution [10 mm N-Tris[hydroxymethyl]methyl-2-aminoeth-anesulfonic (TES; pH 7.4), 10 mm EDTA, 0.2% SDS, 300 mm NaCl] wasincubated at 65 C for 36 h with single nylon strips containing 5 mg of eachdenatured cDNA (NaPi-2, NaSi-1, b-actin, and linearized pBluescript asnegative control) and 10 mg 18S, that were fixed by slot blotting and UVcross-linking. After posthybridization washes and ribonuclease A treat-ment, nylon strips were exposed to BioMax MS films and their corre-sponding intensifying screens. Specific signals were quantified using theGel Doc 1000 Video Gel Documentation System (Bio-Rad Laboratories,Inc.).

Statistical analysis

All the data are expressed as mean 6 se. A one-way ANOVA, withFisher’s protected least-significant difference as a multiple-comparisonmethod, was used to compare data among groups and was consideredstatistically significant at P , 0.05.

ResultsThyroid hormone increases renal reabsorption ofinorganic phosphate

Preliminary experiments were performed to establish theexperimental conditions for further assays. In these experi-ments, chronic treatment of rats with thyroid hormone wasassayed up to a concentration 10 times the physiological doseof the hormone, as explained in Materials and Methods. Theplasma concentrations of T3 are shown in Table 1, togetherwith phosphate and calcium serum levels and the fractionalexcretion of phosphate. Plasma concentration of phosphatewas progressively increased by T3 and reduced in hypothy-roid animals that also showed the highest fractional excretionof phosphorus.

Renal phosphate reabsorption was also measured in theseanimals by using BBM vesicles from both superficial andjuxtamedullary cortex (Fig. 1). Inward sodium-coupledphosphate transport was increased by the hormone in adose-dependent manner in both superficial and juxtamed-ullary vesicles, although no significant difference was ob-served between both hyperthyroid animals [those receiving2.5 (group D) and 5 (group E) mg T3/100 g bwzday]. Thestimulation of phosphate transport in juxtamedullary BBMvesicles tended to be higher than the stimulation measuredin BBM vesicles from the superficial cortex; however, suchdifferences were not statistically significant.

Interestingly, hypothyroid young rats showed a reduction

of approximately 50% in phosphate transport, with respectto euthyroid rats. Hyperthyroidism elicited a 40% increase inPi transport over euthyroid animals and 175% increase overhypothyroid animals.

Thyroid hormone specifically increases NaPi-2 protein

To understand the effect of thyroid hormone on phosphatereabsorption, we have analyzed the content of Na/phos-phate cotransporter in each of the treatments, by Westernblotting assay, using the same BBM vesicles as in the trans-port experiments. As expected, we have found a correspon-dence between the changes in transport activity elicited bythe hormone and the changes in NaPi-2 protein. Fig. 2 showsthe three conditions analyzed: euthyroid, hypothyroid, andhyperthyroid animals. These results indicate that chronicallyadministered triiodothyronine increased NaPi-2 transporterin a way similar to that of phosphate transport, with moreintensity in juxtamedullary than in superficial renal cortex,although this difference was, again, statistically not distin-guishable. Whereas hypothyroidism reduced NaPi-2 proteinto about 40% of euthyroid animals (60% reduction), hyper-thyroidism increased this protein 350%, compared withhypothyroidism; and 65%, compared with euthyroidism.

FIG. 1. Effect of T3 on phosphate transport in young and old rats.Phosphate transport was measured in BBM vesicles from differentanimal groups, as explained in Materials and Methods. Group A,Nonoperated, control rats; group B, TPTX-hypothyroid rats receivingplacebo pellets; group C, TPTX-euthyroid rats receiving a physiolog-ical dose of T3 (0.5 mg/100 mg BWzday); group D, TPTX-hyperthyroidrats receiving 5 times the physiological dose of T3; group E, TPTXhyperthyroid rats receiving 10 times the physiological dose of T3;black bars, BBM from superficial cortex; lined bars, BBM from jux-tamedular cortex; Y, young rats; O, old rats; *, significant differencefrom the respective animal group A.

TABLE 1. Effect of T3 on serum phosphate concentration and fractional excretion in young and aged rats

Young rats Aged rats

Animals T3 (ng/dl) Pi (mg/dl) Ca (mg/dl) FEPi (%) T3 (ng/dl) Pi (mg/dl) Ca (mg/dl) FEPi (%)

A 79.2 6 8.6 10.5 6 0.8a 9.8 6 0.6 7.2 6 1.1a 81.3 6 6.7 6.6 6 0.3 9.3 6 0.5 13.6 6 0.8B ,15.0b 4.8 6 0.2a,b 8.4 6 0.4 15.2 6 0.9b ,15.0b 4.2 6 1.2b 8.9 6 0.8 16.5 6 2.0b

C 45.0 6 8.3 11.1 6 2.6 9.0 6 0.5 6.8 6 0.4D 481.6 6 10.0c 12.8 6 2.1 8.9 6 0.8 4.8 6 0.6c,a 260.0 6 18.3c 7.2 6 1.7 8.7 6 0.9 9.4 6 1.2c

E .1000c 13.1 6 0.3c 9.3 6 0.6 3.8 6 0.4a,c .1000c 8.0 6 0.8c 8.5 6 1.0 8.7 6 0.9c

Group A, Nonoperated, control rats; group B, TPTX-hypothyroid rats receiving placebo pellets; group C, TPTX-euthyroid rats receivingphysiological dose of T3 (0.5 mg/100 g BW per day); group D, TPTX-hyperthyroid rats receiving 5 times the physiological dose of T3; group E,TPTX hyperthyroid rats receiving 10 times the physiological dose of T3. The treatment was maintained for 20 days. Relevant statistic differences(95%). FEPi, Fractional excretion of inorganic phosphate.

a Significant between the corresponding values of young and aged rats (same lane).b Significant with all other data from the same column.c Significant with value from line A of the same column.

1546 THYROID HORMONE MODULATES PHOSPHATE TRANSPORT Endo • 1999Vol 140 • No 4

Signals corresponding to NaSi are shown as unmodifiedcontrols.

These results were confirmed by immunohistochemistry.As Fig. 3 shows, hyperthyroidism elicited a strong increaseof NaPi-2 protein in the luminal membrane of proximal tu-bules. This increase was also extended to the intracellularstores of the transporter, shown in the figure as fluorescent-dotted cytoplasm of the epithelial cells (2). Fig. 3 also showsb-actin filament labeling of tubular plasma membrane,which is not changed by thyroid hormone.

Thyroid hormone increases NaPi-2 mRNA bytranscriptional stimulation

Most cellular effects of thyroid hormone are mediated byincreases in mRNA transcription and consequent proteinsynthesis. Therefore, several Northern blotting assays wereperformed to check for an increase in NaPi-2 mRNA steady-state in rat kidney cortex. The results are summarized in Fig.4: In parallel with the results of protein content, thyroidhormone induced similar changes in specific NaPi-2 mRNA

FIG. 3. Confocal immunohistochemicalanalysis of T3 effect on NaPi-2 abun-dance in proximal tubules of superficialcortex from young rat kidneys. Hyper-thyroidism elicits an increase in NaPi-2protein content in both apical mem-brane and intracellular stores (dottedpattern) of proximal tubular epithelialcells, in comparison with the hypothy-roid status. b-Actin was stained withphalloidin as an unmodified membraneprotein. Bar, 10 mm.

FIG. 2. Effect of thyroid hormone on NaPi-2 protein content in BBM vesicles from proximal tubules in young and old rats. NaSi protein contentis also shown as a nonmodified control. SC, Superficial cortex; JM, juxtamedullary cortex. Samples, run in duplicate for each condition, are thesame as in Fig. 1. Ratios between optical densities of NaPi-2 and NaSi are shown in bars, and P , 0.001 in all groups. All differences with groupB in the histograms are significant (95%); group D is significant with A and C in young animals, and group E with A in old rats.

THYROID HORMONE MODULATES PHOSPHATE TRANSPORT 1547

abundance in all conditions assayed. In this case, the stim-ulation by T3 on juxtamedullary NaPi-2 mRNA was alsohigher than in superficial cortex mRNA (data not shown). Asunmodified controls, RNA levels of sodium-sulfate cotrans-porter, b-actin, and ribosomic 18S are also shown.

In a previous paper (6), we have shown that the effect ofthyroid hormone on phosphate transport in OK cells is notmediated by changes in the stability of the specific NaPi-4mRNA. Now, we have checked for changes in transcriptionrate of the NPT-2 gene (21). Nuclear run-on experiments wereperformed from several conditions using nuclei isolatedfrom full kidney cortex. Results are shown in Fig. 5A, withcomparison between hypothyroid and hyperthyroid condi-tions: an approximately 2-fold stimulatory effect of T3 inNPT-2 gene transcription was found between these two con-ditions. However, an additional change in the NaPi-2 mRNAstability in the rat cannot be excluded.

Thyroid hormone stimulation of phosphate transport isimpaired in the aged

As expected from our previous results (10), 24-month-oldrats exhibited lower plasma phosphate concentration andincreased fractional excretion of phosphate than young rats(Table 1). This decrease was paralleled by a proportionalreduction in phosphate transport capacity by BBM vesiclesfrom superficial and juxtamedullary proximal tubular cells(Fig. 1). The effect of thyroid hormone to modulate Na/Picotransport activity was less dramatic than in young animals.Preliminary experiments with old rats, including group C,showed (as expected) no differences with group A of thesame age animals. Therefore, this group was omitted whenusing old rats. Instead, we always used group E to find a doseof T3 that resulted in effects similar to those seen in the younganimals. Chronic hypothyroidism elicited a 30% reduction inphosphate reabsorption and plasma concentration, that was

restored with thyroid hormone treatment, whereas chronichyperthyroidism increased phosphate transport by only 20%over control levels and 50% over the hypothyroid status,even when plasma concentration of the hormone was morethan 10 times the physiological levels. Interestingly, thesethyrotoxic doses of T3 in aged rats were still not able to restorethe basal level of phosphate transport present in controlyoung animals (Figs. 1 and 6), although additional factors(different from a reduced sensitivity to T3) may be respon-sible for the blunted stimulatory effect of this hormone.

Western blot analysis of NaPi-2 protein in old rats showeda parallelism with transport (Fig. 2). Densitometric analysisrevealed a 40% reduction of NaPi-2 protein in hypothyroid,compared with euthyroid animals, whereas hyperthyroid-ism increased NaPi-2 protein 110%, with respect to hypo-thyroidism, and 40% over euthyroid animals. Similar resultswere obtained for the specific NaPi-2 mRNA content, byNorthern blotting, in both superficial and juxtamedullarykidney cortex (Fig. 4), that was also caused by an increase intranscription of the NPT-2 gene, although this increase wasless effective than in young animals (Fig. 5).

Figure 6 summarizes the influence of aging on the T3

stimulatory effect on phosphate transport rate, NaPi-2 pro-tein, and mRNA contents, and it compares young-controlnonoperated animals with old-control and old-hyperthroidTPTX-animals. On the one hand, transport rate and NaPi-2protein and RNA contents were less than half in control(nonoperated) old animals, compared with control youngrats. On the other hand, hyperthyroidism in old rats in-creased NaPi-2 RNA content above the level in young ani-mals. However, protein content was similar in hyperthyroidold animals and euthyroid young animals, and in terms ofphosphate transport rate, these old animals exhibited evenless transport rate per protein unit than the young rats.Therefore, these results seem to indicate that the aging pro-

FIG. 4. Effect of T3 on NaPi-2 mRNAcontent in superficial kidney cortex ofyoung and old rats. Samples are run induplicate, and the filters were probedwith the cDNAs indicated at the left ofthe figure. The histograms show ratiosof NaPi-2/NaSi optical densities. Treat-ments are the same as in previous fig-ures. ANOVA, P , 0.001. All differenceswith group B in the histograms are sig-nificant (95%); group D is significantwith A and C in young animals, and Ewith A and D in old rats.

1548 THYROID HORMONE MODULATES PHOSPHATE TRANSPORT Endo • 1999Vol 140 • No 4

cess affects all the steps of the stimulatory effect of thyroidhormone on the cell physiology of phosphate transport, fromgene transcription to protein function.

DiscussionEffect of T3 on phosphate reabsorption

The present study demonstrates a physiological role forthyroid hormone in the control of phosphate homeostasis.We have found that a physiological dose of T3 stimulates Pirenal reabsorption (Fig. 1) to a level that is able to increaseserum Pi concentration (Table 1). Higher (pharmacological)doses of T3 further increase Pi renal reabsorption and serumphosphate level. This effect is mediated by parallel increasesin the amount of Na/Pi cotransporter (NaPi-2) in the brushborder of proximal tubular epithelial cells (Figs. 2 and 3). Thespecific increase in protein content is, in turn, caused by anincrease in the intracellular content of the specific NaPi-2mRNA (Fig. 4), which was produced by stimulation of thetranscription rate of the corresponding gene, NPT-2 (Fig. 5).These results point to the classic mechanism of thyroid hor-mone, acting through intracellular receptors and binding tothyroid hormone response elements (TREs) in the corre-sponding gene promoters (20). However, in spite of the re-cent identification and sequence of several type II (NaPi-2,NPT-2) gene promoters (21), no consensus sequence of classicTREs has been found, but only a putative vitamin D-responseelement (22). Therefore, the stimulatory effect on NaPi-2-mediated transport could be explained by the existence of anovel TRE in the NPT-2 gene or by an indirect T3 stimulation;i.e. T3 could stimulate the synthesis of a protein responsiblefor a direct effect on the NPT-2 gene. Future experiments,using these gene promoters, should clarify this question.

To our knowledge, our study is the first that clearly showsa physiological role for T3 on renal tubular phosphate reab-

sorption: chronic hypothyroidism induces a substantial de-crease in serum phosphate, as well as an inhibition of phos-phate transport, that is reversed by the exogenousphysiological treatment with T3. It is interesting that thereduction in phosphatemia occurs in spite of the existence ofalternative mechanisms for increasing phosphate reabsorp-tion in the rat (e.g. acute and chronic adaptation to dietaryphosphate deprivation, and others), because none of them isable to restore the euthyroid level of serum phosphate inhypothyroid animals. This could be explained as either theexistence of additive effects of all these regulatory mecha-nisms, T3 being a major regulator, and/or as the need ofthyroid hormone presence for a correct functioning of allother additional mechanisms. The second possibility is morelikely to occur because, as it has been shown, euthyroid rats,chronically fed with low (0.1% Pi) phosphate diet vs. control(0.6% Pi) or high (1.2% Pi) phosphate diets, are able to main-tain a normal phosphatemia, thanks to an avid renal adaptivemechanism that induces an increase of almost 100% reab-sorption of phosphate ultrafiltrate (18, 23). In this case, theadaptation consists in both increases in specific NaPi-2mRNA and protein content in tubular cells, although thiseffect is not mediated by NPT-2 gene transcription stimula-tion but by an increase of NaPi-2 mRNA stability (24, 25). In

FIG. 5. Effect of T3 on gene NPT-2 transcription rate, measured innuclear run-on from renal cortex of young and old rats. Total labelednascent RNA was labeled and probed against denaturized-linear plas-mids containing the cDNAs indicated at the left of the figure. HypoT3,Hypothyroid animals, group B; Hyper T3, hyperthyroid animals,group D; pBlu, pBluescript.

FIG. 6. Comparison of T3 effects on superficial kidney cortex fromyoung and old rats: Na/phosphate cotransport, NaPi-2 protein, andNaPi-2 mRNA. A-Y, Control nonoperated young rats (group A, young);A-O, control nonoperated old rats (group A, old); D-O, TPTX-hyper-thyroid old rats (group D, old); *, significant difference with group AY;#, significant difference between AO and DO. Values of RNA andprotein densitometries in histograms are arbitrary densitometricunits. Dens., Densitometric.

THYROID HORMONE MODULATES PHOSPHATE TRANSPORT 1549

addition, though the expected transient hypophosphatemiacaused by dietary phosphate deprivation induces the men-tioned increase in NaPi-2 mRNA and protein, in the case ofhypothyroidism, the effect is just the contrary (that is, re-duction in NaPi-2 mRNA and protein content).

From these results, we can initially conclude that thyroidhormone is a major controller of phosphate homeostasis inlong-term regulation, which should be distinguished fromacute regulation, a less frequent physiological condition. Thisconclusion is also supported by a characteristic of our ex-perimental design: to avoid interferences with PTH, the an-imals were made TPTX, instead of just thyroidectomized.Therefore, our animals possessed hypothyroid and hypo-parathyroid status. As it has been recently shown, hypopar-athyroidism induces an increase in phosphate reabsorptionthrough lack of the phosphaturic hormone (26). Therefore,when the animals have normal levels of PTH in serum, theactual effect of chronic hypothyroidism is, most likely, moredramatic than we have shown in the present paper.

With respect to the stimulatory effect of T3, treatment withT3 (5 and 10 times the physiological dose) further increasedPi reabsorption through the same mechanisms as the phys-iological dose (RNA and protein synthesis). The stimulatoryeffect of chronic hyperthyroidism was already saturated at 5times the physiological dose in young animals (see Fig. 1) andwas similar to the increase obtained with acute thyrotoxicdoses, 400 times the physiological dose (200 mg T3/100 mgBWz12 h, for 3 days) shown in previous papers and in ourown lab (Refs. 4 and 23; data not shown). The equivalence ofour chronic hyperthyroid doses to acute thyrotoxic doses(because of saturability) shown in the present study, withrespect to the stimulation of phosphate reabsorption, couldmake feasible the use of low-dose thyroid hormone as along-term therapy in several disorders involving phosphatehomeostasis unbalance or bone diseases.

Previous works (4) reported that the main effect of T3 takesplace on the renal juxtamedullary cortex (mostly straighttubules), in spite of the higher transport rate and NaPi-2mRNA and protein content of the superficial cortex (mostlyconvoluted tubules; Ref. 15). In the present paper, we havenot found such difference in T3 effect; that again, could becaused by the differences in the experimental design: long-term physiological T3 doses vs. acute thyrotoxic doses in theprevious work.

Finally, two comments must be made with respect to thetwo controls used, the NaSi and b-actin. Two groups havefound discrepancies, in relation to the the effect of T3 onsulfate transport in BBM vesicles: no effect in rat (4, 5); andstimulation in mouse (27). Our results have shown a lack ofT3 effect on transcription and RNA and protein content of theNaSi transporter. With respect to b-actin, it has been shownrecently that thyroid hormone induces a 2-times inhibition inits mRNA levels in skeletal and white muscle tissue (28). Inthe present work, we have also not found any effect onmRNA or protein b-actin levels. This difference could beexplained by tissue-specific effects, but also by the differ-ences in both experimental design and the dose: 0.5 mg T3/100 mg BW vs. 100 mg/100 mg BW in the previous paper.

Effect of aging on thyroid hormone action

There are two main results of the present study in relationto the effect of aging: first, NPT-2 gene transcriptional activityis decreased with the age of the animals; and second, thestimulatory effect of thyroid hormone is also reduced in theaging. In a previous paper (10), we have published that oldrats contain less NaPi-2 mRNA and protein per tissue unitthan young rats. Now, we have shown that such decreasemay be caused by a reduced transcription rate of the corre-sponding gene, which might explain the lower serum phos-phate concentration in old animals. Such a decrease is ageneral characteristic of the aging process in mammals (29).Old rats are also less sensitive to T3 than the young ones, andthis can be observed in both hypothyroid and hyperthyroidanimals. On the one hand, old hypothyroid animals show asmaller reduction in phosphatemia than young hypothyroidanimals (in comparison with the respective euthyroid rats),as well as proportional (smaller) alterations in transport rate,protein, and mRNA levels of the renal phosphate transporterNaPi-2. On the other hand, the stimulatory effect of T3 toinduce hyperthyroidism is also less effective in old than inyoung animals, for all parameters analyzed.

In addition, the reduction in the basal transcriptional rateof NaPi-2 mRNA in aged rats is not caused by an age-dependent reduction in serum levels of T3. It should be notedthat some authors have reported a decrease in serum T3levels with age in humans (30), caused by both decreases inpituitary TSH release and peripheral degradation of T4 that,in return, result in a decline in serum and tissue concentra-tion of T3 (31). Our results are, however, in agreement withother studies (e.g. Ref. 32), which have not found differencesin serum T3 concentrations between young and old rats, andthey suggest that the relative hypothyroid status of the agingrat is not produced by a decreased concentration of T3 but iscaused by a reduced relative response to the hormone. In fact,the effect of the aging can be detected in all the cellular andmolecular processes of phosphate transport regulation: tran-scription, translation, and function of the NaPi-2 protein(Fig. 6).

In summary, we have shown a physiological role of T3 inthe regulation of, mainly, basal level of renal phosphatereabsorption, which is impaired with aging. Furthermore,our study is the first that shows genomic regulation of thecorresponding gene, NPT-2, by thyroid hormone and theaging process.

Acknowledgment

We thank Prof. Gabriela Morreale, from the Institute for BiomedicalResearch-Consejo Superior de Investigaciones Cientifica, for T3 RIA andcomments on the manuscript.

References

1. Biber J, Murer H 1994 Renal epithelial apical Na/Pi cotransporters. CellPhysiol Biochem 4:185–197

2. Murer H, Biber J 1996 Molecular mechanisms of renal apical Na/phosphatecotransport. Annu Rev Physiol 58:607–618

3. Murer H 1995 Renal phosphate transporter: another shuttle? News Physiol Sci10:287

4. Yusufi ANK, Murayama N, Keller MJ, Dousa TP 1985 Modulatory effect ofthyroid hormones on uptake of phosphate and other solutes across luminalbrush border membrane of kidney cortex. Endocrinology 116:2438–2449

1550 THYROID HORMONE MODULATES PHOSPHATE TRANSPORT Endo • 1999Vol 140 • No 4

5. Beers KW, Dousa P 1993 Thyroid hormone stimulates the Na1-PO4 symporterbut not the Na1-SO4 symporter in renal brush border. Am J Physiol265:F323–F326

6. Sorribas V, Markovich D, Verri T, Biber J, Murer H 1995 Thyroid hormonestimulation of Na/Pi-cotransport in opossum kidney cells. Pflugers Arch431:266–271

7. Magagnin S, Werner A, Markovich D, Sorribas V, Biber J, Murer H 1993Expression cloning of human and rat renal cortex Na/Pi cotransport. Proc NatlAcad Sci USA 90:5979–5983

8. Euzet S, Lelievre-Pegorier M, Merlet-Benichou C 1995 Maturation of ratrenal phosphate transport: effect of triiodothyronine. J Physiol (Lond)488:449 – 457

9. Euzet S, Lelievre-Pegorier M, Merlet-Benichou C 1996 Effect of 3,5,39-tri-iodothyronine on maturation of rat renal phosphate transport: kinetic char-acteristics and phosphate transport messenger ribonucleic acid and proteinabundance. Endocrinology 137:3522–3530

10. Sorribas V, Lotscher M, Loffing J, Biber J, Kaissling B, Murer H, Levi M 1996Cellular mechanisms of the age-related decrease in renal phosphate reabsorp-tion. Kidney Int 50:855–863

11. Obregon MJ, Pascual A, Morreale de Escobar G, Escobar del Rey F 1979Pituitary and plasma thyrotropin, thyroxine and triiodothyronine after hy-perthyroidism. Endocrinology 104:1467–1473

12. Levi M, Baird BM, Wilson PW 1990 Cholesterol modulates rat renal brushborder membrane phosphate transport. J Clin Invest 85:231–237

13. Laemmli UK 1970 Cleavage of structural proteins during assembly of the headof the bacteriophage T4. Nature 227:680–685

14. Towbin H, Staehelin T, Gordon J 1979 Electrophoretic transfer of proteinsfrom polyacrylamide gels to nitrocellulose sheets: procedure and some ap-plications. Proc Natl Acad Sci USA 76:4350–4354

15. Custer M, Lotscher M, Biber J, Murer H, Kaissling B 1994 Expression of Na-Pico-transport in rat kidney: localization by RT-PCR and immunohistochemis-try. Am J Physiol 266:F767–F774

16. Lotscher M, Custer M, Quabius ES, Kaissling B, Murer H, Biber J 1996Immunolocalization of Na/SO4-cotransport (NaSi-1) in rat kidney. PflugersArch 432:373–378

17. Chomcyznski P, Sacchi N 1987 Single-step method of RNA isolation by acidguanidium thiocyanate-phenol-chloroform extraction. Anal Biochem162:156–159

18. Levi M, Lotscher J, Sorribas V, Custer M, Arar M, Kaissling B, Murer H, BiberJ 1994 Cellular mechanisms of acute and chronic adaptation of rat renal Pitransporter to alterations in dietary Pi. Am J Physiol 267:F900–F908

19. Farrell RE 1993 RNA Methodologies: A Laboratory Guide for Isolation andCharacterization. Academic Press Inc, San Diego, p 244–251

20. Yen PM, Chin WW 1994 New Advances in understanding the molecularmechanisms of thyroid hormone action. Trends Endocrinol Metab 5:65–72

21. Hartmann CM, Hewson AS, Kos CH, Hilfiker H, Soumoumou Y, Murer H,Tenenhouse HS 1996 Structure of murine and human renal type II Na1-phosphate cotransporter genes (Npt2 and NPT2). Proc Natl Acad Sci USA93:7409–7414

22. Taketani Y, Miyamoto K, Tanaka K, Katai K, Chikamori M, Tatsumi S,Segawa H, Yamamoto H, Morita K, Takeda E 1997 Gene structure and func-tional analysis of the human Na1/phosphate co-transporter. Biochem J 324:927–934

23. Yusufi ANK, Szczepanska-Konkel M, Hoppe A, Dousa TP 1989 Differentmechanisms of adaptive increase in Na1-Pi cotransport across renal brush-border membrane. Am J Physiol 256:F852–F861

24. Markovich D, Verri T, Sorribas V, Forgo J, Biber J, Murer H 1995 Regulationof opossum kidney (OK) cell Na/Pi cotransport by Pi-deprivation involvesmRNA stability. Pflugers Arch 430:459–463

25. Collins JF, Zeng J, Ghishan FK 1997 In vivo adaptation of the mouse renalsodium-phosphate transporter to low Pi diet is not related to gene transcrip-tion. Nutr Res 17:295–304

26. Kempson SA, Lotscher M, Kaissling B, Biber J, Murer H, Levi M 1995Parathyroid hormone action on phosphate transporter mRNA and protein inrat renal proximal tubules. Am J Physiol 268:F784–F791

27. Tenenhouse HS, Lee J, Harvey N 1991 Renal brush border membrane Na1-sulfate cotransport: stimulation by thyroid hormone. Am J Physiol 261:F420–F426

28. Torrance CJ, Devente JE, Dohm GL 1997 Effects of thyroid hormone onGLUT4 glucose transporter gene expression and NIDDM in rats. Endocrinol-ogy 138:1204–1214

29. Kanungo MS 1995 Changes in gene expression during aging of mammals. In:Macieira-Coelho A (ed) Molecular Basis of Aging. CRC Press, Boca Raton, pp183–218

30. Mariotti S, Franceschi C, Cossarizza A, Pinchera A 1995 The aging thyroid.Endocr Rev 16:686–715

31. Mooradian AD, Wong NCW 1994 Age-related changes in thyroid hormoneaction. Eur J Endocrinol 131:451–461

32. Lamberts SWJ, van del Beld AW, van der Lely A-J 1997 The endocrinologyof aging. Science 278:419–424

THYROID HORMONE MODULATES PHOSPHATE TRANSPORT 1551