IT IS WELL KNOWN that the decrease in circulating estradiol

0021-972X/00/$03.00/0 Vol. 85, No. 3 The Journal of Clinical Endocrinology & Metabolism Printed in U.S.A. Copyright 2000 by The Endocrine Society Close Correlation between Estrogen Treatment and Renal Phosphate Reabsorption Capacity HIROKAZU UEMURA, MINORU IRAHARA, NAOTO YONEDA, TOSHIYUKI YASUI, KAORI GENJIDA, KEN-ICHI MIYAMOTO, TOSHIHIRO AONO, AND EIJI TAKEDA Departments of Obstetrics and Gynecology (H.U., M.I., N.Y., T.Y., T.A.) and Clinical Nutrition (K.G., K.-i.M., E.T.), The University of Tokushima, School of Medicine, Tokushima, 770-8503 Japan ABSTRACT To determine the influence of estrogen on the activity of renal proximal tubular reabsorption of inorganic phosphate (Pi) in women, we examined the changes of the renal threshold phosphate concentration (also denoted as TmP/GFR), as well as the changes in the concentrations of mineral components in the circulation in two groups of women one receiving hormone replacement therapy (HRT) and one receiving gonadotropin-releasing hormone agonists (GnRH-a) therapy. We also examined the changes in the concentrations of serum PTH in the GnRH-a group. The patients in the HRT group were continuously treated with 0.625 mg conjugated equine estrogens plus 2.5 mg medroxyprogesterone acetate per day. The patients in the GnRH-a group were treated with a monthly injection of 3.75 mg leuprolide acetate depot for 6 months. The values of TmP/GFR decreased in all of the patients who received HRT. The mean percentage Received September 2, 1999. Revision received November 10, 1999. Accepted November 30, 1999. Address correspondence and requests for reprints to: Hirokazu Uemura, M.D., Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine, 3-18-15. Kuramoto-cho, Tokushima, 770-8503 Japan. E-mail: uemura@clin.med. tokushima-u.ac.jp. change in TmP/GFR was 14.5% (range, 24.3% to 9.6%). In contrast, in all of the patients treated with GnRH-a, the values of TmP/ GFR increased after 6 months of treatment (the mean percentage change was 28.5%; range, 18.2 78.3%) and returned to the preadministration level at 12 weeks after stopping therapy. In these patients, both the values of TmP/GFR and the concentrations of serum Pi correlated significantly with circulating estradiol levels (r 0.767, P 0.01 and r 0.797, P 0.01, respectively), but the concentrations of serum corrected calcium did not correlate. Moreover, in the same patients, the levels of serum intact PTH decreased significantly (P 0.05) after 6 months of treatment, but at 12 weeks after stopping therapy the trends of these levels varied among individual patients. These results suggest that estrogen could act directly to suppress sodium-dependent Pi reabsorption in the renal proximal tubules. (J Clin Endocrinol Metab 85: 1215 1219, 2000) IT IS WELL KNOWN that the decrease in circulating estradiol (E 2 ) in women at menopause is a major trigger of the changes in bone metabolism and causes the loss of bone density (1, 2). Administration of estrogen to postmenopausal women produces many health benefits, such as reduction of climacteric symptoms and prevention of arteriosclerosis (3, 4) and osteoporosis (5, 6). On the other hand, gonadotropinreleasing hormone agonists (GnRH-a) have been recognized to be effective against endometriosis by inducing the decrease in circulating E 2, leading to atrophy of both the endometrium and the endometriotic tissues. However, such long-term estrogen deficiency is likely to have a negative influence on bone density. Bone turnover is high at menopause. Estrogen reduces bone turnover and especially reduces bone resorption (7). By this action, the levels of circulating calcium (Ca) and inorganic phosphate (Pi) decrease in women who receive estrogen replacement (8 10). Renal proximal tubular reabsorption of Pi is a major determinant of the circulating level of Pi and contributes to the maintenance of Pi homeostasis. Pi reabsorption at the apical membrane is carried out by sodium-dependent Pi (Na/Pi) cotransporters, which are classified into types I and II by molecular characterization (11). Several hormonal and nonhormonal factors that control serum Pi homeostasis affect the rate of Pi reabsorption by changing the activities of type II Na/Pi cotransporter (12, 13). However, little is known about the effects of estrogen on renal phosphate reabsorption in women. To clarify this issue, we studied the effects of hormone replacement therapy (HRT) in climacteric women and of GnRH-a therapy in women with endometriosis on the renal phosphate threshold concentration (TmP/GFR) by modulating circulating E 2 levels. Experimental Subjects Patients and clinical treatment HRT treatment. Five Japanese women who had complained of various climacteric symptoms and/or had undergone surgical castration, aged 46 56 yr (mean sd, 52.8 3.7), with body mass indices ranging from 20.6 24.6 kg/m 2 (mean sd, 22.9 1.4) were studied after providing informed consent. They were treated with 0.625 mg conjugated equine estrogens (CEE) (Premarin; Wyeth-Ayerst Laboratories, Inc., Radnor, PA) plus 2.5 mg medroxyprogesterone acetate (Provera; Upjohn Co., Kalamazoo, MI) per day. All patients had no history of estrogen-dependent cancer, hypercortisolism, hyperthyroidism, metabolic bone disease, or renal failure. All patients had both a negative mammogram and negative Papanicolaou smear within 6 months prior to the study. None of them had previously received estrogen replacement therapy or any drug treatment that might have affected Ca or Pi metabolism. GnRH-a treatment. Five Japanese women with endometriosis, aged 30 49 yr and with body mass indices of 17.0 28.8 1215

1216 UEMURA ET AL. JCE&M 2000 Vol 85 No 3 kg/m 2 (mean sd, 22.9 5.2), were studied after providing informed consent. They were treated with a monthly injection of 3.75 mg leuprolide acetate (Leuprin; Takeda Pharmaceutical Co., Tokyo, Japan) depot for 6 months. All patients were in good health except for their endometriosis. None of them had received any other hormonal treatment for their endometriosis. Materials and Methods Mineral and hormone determination Fasted blood and urine samples were collected from the patients during HRT at the beginning and at the 6th month of treatment. They were also collected from the patients treated with GnRH-a at the beginning and end (i.e. at the 6th month) of treatment and at 12 weeks after stopping treatment. Serum was immediately separated after blood collection and promptly frozen at 40 C until the assay. Urine was also frozen and kept at 40 C until it was assayed. Serum E 2 was measured by a combination of high-performance liquid chromatography after extraction and RIA. The high-performance liquid chromatography was performed using a CAPCELL PAK NH2 column (4.6 250, 5 um; Shiseido Co., Tokyo, Japan). RIA was performed using a DSL Kit (Diagnostics Systems Laboratories, Inc., Webster, TX). The assay range was 1.0 50 pg/ml. Intra- and interassay coefficients of variation were 8.4 13.8% and 15.5%, respectively. Serum intact PTH was measured by two-site immunoradiometric assay (Nichols, San Juan Capistrano, CA), with a normal range of 10 65 pg/ml and assay sensitivity of 1 pmol/l. Pi concentration in serum and urine was analyzed by the method of Taussky and Shorr (14). Creatinine concentration in serum and urine was measured using a Creatinine-TEST Wako (Wako Pure Chemical Industries, Ltd., Osaka, Japan). Determination of TmP/GFR The activity of renal tubular reabsorption of Pi is best expressed as the renal threshold phosphate concentration (also denoted as TmP/GFR, or the ratio of maximum rate of renal tubular reabsorption of phosphate to GFR), which is independent of GFR and of net inflow of phosphate. TmP/GFR was calculated using a standard nomogram (15) and was expressed in millimoles per liter. Statistical analysis Data are expressed as the mean sd unless otherwise stated. Statistical significances were determined by t test and linear regression analysis. All P values of 0.05 or less were considered statistically significant. Analyses were carried out using a Stat Works program (Cricket Software, Inc., Philadelphia, PA). Results Changes in clinical values for serum and urine The changes in clinical chemical and hormonal values for serum and urine in patients during HRT are shown in Table 1. In all the patients, serum E 2 levels increased after 6 months of treatment, whereas the concentrations of both serum corrected Ca and serum Pi decreased. The changes in clinical chemical and hormonal values for serum and urine in patients who received GnRH-a therapy are shown in Table 2. In all patients treated with GnRH-a, the concentrations of serum E 2 decreased after 6 months of treatment and returned to the preadministration level at 12 weeks after stopping therapy, and the concentrations of both serum corrected Ca and serum Pi changed inversely to those of serum E 2. Changes in TmP/GFR The changes in the values of TmP/GFR in patients during HRT are shown in Fig. 1. In all of the patients, the values of TABLE 1. Parameters in serum and urine in patients who received HRT (n 5) Before TmP/GFR decreased after 6 months of treatment. The mean percentage changes in TmP/GFR was 14.5% (range, 24.3% to 9.6%). In contrast, in all of the patients treated with GnRH-a, the values of TmP/GFR increased after 6 months of treatment (the mean percentage change was 28.5%; range, 18.2 78.3%) and returned to the preadministration level at 12 weeks after stopping therapy (Fig. 2). Changes in levels of serum intact PTH The concentrations of serum intact PTH in the patients who received GnRH-a therapy decreased significantly (P 0.05) after 6 months of GnRH-a treatment, but at 12 weeks after stopping therapy the trends of these levels varied among individual patients (Fig. 3). Relationship between circulating E 2 and chemical values Because administration of CEE leads mainly to a rise in circulating estrone levels, we analyzed the correlations between circulating E 2 levels and other indices in the patients treated with GnRH-a. As shown in Fig. 4, the values of TmP/GFR correlated significantly with circulating E 2 levels (r 0.767, P 0.01). In these patients, the concentrations of serum Pi also correlated with circulating E 2 levels (r 0.797, P 0.01), but those of serum corrected Ca did not correlate with circulating E 2 levels. 6 M Serum Pi 3.78 0.33 3.16 0.26 Ca 8.90 0.16 8.46 0.38 Albumin 4.38 0.08 4.30 0.23 Corrected Ca 8.52 0.19 8.16 0.21 Creatinine 0.72 0.04 0.78 0.08 Urine Pi 55.02 18.51 41.48 20.69 Creatinine 136.20 89.80 163.76 134.71 TmP/GFR (mg/100 ml) 3.97 0.80 3.35 0.46 TABLE 2. Parameters in serum and urine in patients treated with GnRH-a (n 5) Before 6M (at the end of therapy) 12 weeks after stopping therapy Serum Pi 3.72 0.57 4.20 0.47 3.66 0.23 Ca 8.34 0.26 8.94 0.36 8.72 0.60 Albumin 4.08 0.31 4.46 0.23 4.34 0.27 Corrected 8.26 0.49 8.48 0.24 8.38 0.43 Ca Creatinine 0.66 0.11 0.68 0.14 0.66 0.15 E 2 (pg/ml) 42.74 33.16 5.56 2.47 56.10 32.46 intact PTH 29.44 13.38 22.39 10.27 23.39 10.61 (pg/ml) Urine Pi 55.66 15.12 26.68 8.69 37.73 12.73 Creatinine 98.90 59.97 80.34 57.68 71.40 28.46 TmP/GFR (mg/100 ml) 3.68 1.13 4.73 1.02 3.47 0.68

ESTROGEN AND RENAL PHOSPHATE REABSORPTION 1217 FIG. 1. Changes in the values of TmP/GFR in the patients during HRT. In all patients, the values of TmP/GFR decreased after 6 months of treatment. The mean percentage change in TmP/GFR was 14.5% (range, 24.3% to 9.6%). FIG. 2. Changes in the values of TmP/GFR in patients receiving GnRH-a therapy. In all patients, the values of TmP/GFR increased after 6 months of treatment when therapy was completed (mean percentage change, 28.5%; range, 18.2 78.3%) and returned to the preadministration levels at 12 weeks after stopping therapy. Discussion Pi is essentially necessary to body functions because it is a constituent of the skeleton, membrane phospholipids, nucleic acids, and nucleotides. Moreover, it is necessary in many kinds of metabolic reactions and regulatory processes, including protein phosphorylation. Pi is widely recognized to be one of the most important constituent minerals of bone. The content of phosphorus in FIG. 3. Changes in the values of serum intact PTH in patients receiving GnRH-a therapy. The values of serum immunoreactive PTH decreased significantly (P 0.05) after 6 months of treatment when therapy was completed, but at 12 weeks after stopping therapy the trends of these levels varied among individual patients. the body increases from 4 5 g/kg at birth to 10 12 g/kg in adulthood (16). And although levels of Pi are influenced by several factors, such as diet, intestinal absorption, bone metabolism, and renal reabsorption, Pi homeostasis in the body is mainly maintained by the mechanism of renal proximal tubular reabsorption. Previous studies in vivo and in vitro have reported that physiological regulation of proximal tubular Pi reabsorption is most likely related to the capacity of apical Na/Pi cotransport (13, 17, 19). In cases where bone resorption is accelerated, such as at menopause, phosphate enters the circulation along with Ca from bone. In contrast, administration of estrogen to women at menopause leads to a reduction of bone resorption (20), thereby suppressing the flux of Ca and Pi into the circulation and ultimately decreasing the circulating level of Pi. However, little is known about the effects of estrogen on renal reabsorption of Pi in women. In our studies, the values of TmP/GFR decreased with the increase of estrogen level in all women who received HRT, and it increased with the decrease of estrogen level in all women treated with GnRH-a. Administration of CEE to women leads mainly to a rise in the circulating estrone level, rather than in the E 2 level, and thus we analyzed the correlations between circulating E 2 levels and other indices in the patients treated with GnRH-a. We also studied the changes of serum intact PTH levels in these patients. The concentrations of circulating E 2 correlated to those of serum Pi but did not correlate to those of serum corrected Ca. These results suggest that estrogen regulates the renal proximal tubular Pi reabsorption directly or indirectly, but that estrogen does not regulate the renal Ca reabsorption. Previous studies have suggested that long-term estrogen administration leads to an increase in circulating immunoreactive PTH (21 23), due probably to a slight decrease in serum Ca resulting from inhibition of bone resorption. It is recognized that PTH has an inhibitory effect on the renal proximal tubular Na/Pi cotransporter (13, 19). Based on this

1218 UEMURA ET AL. JCE&M 2000 Vol 85 No 3 fact, we might offer another possible explanation that the effect of estrogen on the renal Na/Pi cotransporter might be mediated by the change of circulating PTH concentrations. However, although short-term administration of estrogen to postmenopausal women reduces circulating PTH, it also reduces the value of TmP/GFR (24). Therefore, the influence of estrogen on the change of TmP/GFR does not always occur through the change of circulating PTH. Moreover, Beers et al. (25) recently reported that in thyroparathyroidectomized and ovariectomized rats 17 -E 2 injection suppressed the capacity for Na/Pi cotransport across the renal brush border membrane. This report strongly suggests that estrogen has a direct inhibitory effect on the renal proximal tubular Na/Pi cotransporter in rats. In our studies, the values of TmP/GFR changed inversely to circulating E 2 levels, but the changes in TmP/GFR were not associated with the changes of serum intact PTH levels. Together with the results of Beers et al. (25), our present findings suggest that estrogen may regulate renal Pi reabsorption by acting directly on E 2 receptors in renal proximal tubular cells in women. However, a typical estrogen response element was not observed in the sequence of 5 -flanking region of a type II human Na/Pi transporter (NaPi-3) (26). Pi is an activator of bone turnover. It is well known that estrogen reduces bone turnover in postmenopausal women by its direct effect on osteoblasts, which have been shown to have estrogen receptors (27), and through several kinds of cytokines or growth factors (28, 29). Our results indicate the further possibility that estrogen may reduce bone turnover by reducing plasma Pi levels via a suppression in the capacity for Na/Pi cotransport across the renal brush border membrane. However, further investigation will be required to define the relationship between the bone turnover rate and circulating Pi level. In summary, we examined the relation between the changes of TmP/GFR and the changes in circulating E 2 levels in women treated with HRT or GnRH-a and found that the changes of TmP/GFR and circulating E 2 were inversely related, whereas those of TmP/GFR and serum intact PTH were not related. Our results suggest that estrogen could act directly to suppress the capacity for Na/Pi cotransport at the renal brush border membrane. To elucidate the detailed mechanisms of estrogens on renal Na/Pi cotransport, more fundamental studies will be needed. References FIG. 4. Correlations between the concentrations of circulating E 2 and the following indices in five patients treated with GnRH-a: TmP/GFR (A), concentration of serum Pi (B), and concentration of serum corrected Ca (C). The values of TmP/GFR correlated significantly with circulating E 2 levels (r 0.767, P 0.01). The concentrations of serum Pi also correlated with circulating E 2 levels (r 0.794, P 0.01), but those of serum corrected Ca did not correlate with circulating E 2 levels. 1. Lufkin EG, Wahner HW, O Fallon WM, et al. 1992 Treatment of postmenopausal osteoporosis with transdermal estrogen. Ann Intern Med. 117:1 9. 2. Raize LG, Shoukri KC. 1993 Pathogenesis of osteoporosis. In: Mundy GR, Martin TJ, eds. Physiology and pharmacology of bone. Heidelberg: Springer- Verlag; 299 323. 3. Walsh BW. 1992 Estrogen replacement and heart disease. Clin Obstet Gynecol. 35:894 900. 4. Collins P, Rosano GM, Jiang C, Lindsay D, Sarrel PM, Poole-Wilson PA. 1993 Cardiovascular protection by oestrogen a calcium antagonist effect? Lancet. 341:1264 1265. 5. Lindsay R, Aitkin JM, Anderson JB, Hart DM, MacDonald EB, Clarke AC. 1976 Long-term prevention of postmenopausal osteoporosis by estrogen. Lancet. 1:1038 1041. 6. Riis BJ, Thomsen K, Strom V, Christiansen C. 1987 The effect of percutaneous estradiol and natural progesterone on postmenopausal bone loss. Am J Obstet Gynecol. 156:61 65. 7. Riggs BL, Jowsey J, Kelly PJ, Jones JD, Maher FT. 1969 Effect of sex hormones on bone in primary osteoporosis. J Clin Invest. 48:1065 1072.

ESTROGEN AND RENAL PHOSPHATE REABSORPTION 1219 8. Stock JL, Coderre JA, Mallette LE. 1985 Effects of a short course of estrogen on mineral metabolism in postmenopausal women. J Clin Endocrinol Metab. 61:595 600. 9. Castelo-Branco C, Martinez-de-Osaba MJ, Pons F, Gonzalez-Merlo J. 1992 The effect of hormone replacement therapy on postmenopausal bone loss. Eur J Obstet Gynecol Reprod Biol. 44:131 136. 10. Stock JL, Coderre JA, Posillico JT. 1989 Effects of estrogen on mineral metabolism in postmenopausal women as evaluated by multiple assays measuring parathyrin bioactivity. Clin Chem. 35:18 22. 11. Murer H, Biber J. 1997 A molecular view of proximal tubular inoganic phosphate (Pi) reabsorption and of its regulation. Pflugers Ach-Eur J Physiol. 433:379 389. 12. Murer H, Biber J. 1992 Renal tubular phosphate transport. In: Seldin GW, Giebisch G, eds. The kidney: physiology and pathophysiology, 2nd ed. New York: Raven; 2481 2509. 13. Bernet TJ, Knox FG. 1992 Renal regulation of phosphate excretion. In: Seldin GW, Giebisch G, eds. The kidney: physiology and pathophysiology. New York: Raven; 2511 2532. 14. Taussky HH, Shorr E. 1953 A microcolorimetric method for the determination of inorganic phophorus. J Biol Chem. 202:675 685. 15. Walton RJ, Bijvoet OL. 1975 Nomogram for derivation of renal threshold phosphate concentration. Lancet. 2:309 310. 16. Forbes GB. 1987 Growth, aging, nutrition and activity. In: Human body composition. New York: Springer Verlag; 170 182. 17. Murer H, Werner A, Reshkin S, Wuarin F, Biber J. 1991 Cellular mechanisms in proximal tubular reabsorption of inorganic phosphate. Am J Physiol. 260:C885 C899. 18. Deleted in proof. 19. Dennis VW. 1992 Phosphate homeostasis. In: Windhager EE, ed. Handbook of physiology. New York: Oxford University Press; 1785. 20. Riggs BL, Jowsey J, Kelly PJ, Jones JD, Maher FT. 1969 Effect of sex hormones on bone in primary osteoporosis. J Clin Invest. 48:1065 1072. 21. Riggs BL, Jowsey J, Goldsmith RS, Kelly PJ, Hoffman DL, Arnaud CD. 1972 Short- and long-term effects of estrogen and synthetic anabolic hormone in postmenopausal osteoporosis. J Clin Invest. 51:1659 1663. 22. Riggs BL. 1979 Postmenopausal and senile osteoporosis: current concepts of etiology and treatment. Endocrinol Jpn. 26(Suppl):31 41. 23. Gallagher JC, Riggs BL, DeLuca HF. 1980 Effect of estrogen on calcium absorption and serum vitamin D metabolites in postmenopausal osteoporosis. J Clin Endocrinol Metab. 51:1359 1364. 24. Stock JL, Coderre JA, Mallette LE. 1985 Effects of a short course of estrogen on mineral metabolism in postmenopausal women. J Clin Endocrinol Metab. 61:595 600. 25. Beers KW, Thompson MA, Chini EN, Dousa TP. 1996 beta-estradiol inhibits Na -P(i) cotransport across renal brush border membranes from ovarectomized rats. Biochem Biophys Res Commun. 221:442 445. 26. Taketani Y, Miyamoto K, Keiko T, et al. 1997 Gene structure and functional analysis of the human Na /phosphate cotransporter. 324:927 934. 27. Eriksen EF, Colvard DS, Berg NJ, et al. 1988 Evidence of estrogen receptors in normal human osteoblast-like cells. Science. 241:84 86. 28. Pacifici R, Rifas L, McCracken R, et al. 1989 Ovarian steroid treatment blocks a postmenopausal increase in blood monocyte interleukin 1 release. Proc Natl Acad Sci USA. 86:2398 2402. 29. Pacifici R, Brown C, Puscheck E, et al. 1991 Effect of surgical menopause and estrogen replacement on cytokine release from human blood mononuclear cells. Proc Natl Acad Sci USA. 88:5134 5138.