THE JOURNAL OF EXPERIMENTAL ZOOLOGY 264:167-176 (1992)
Inhibitory Role of Sex Steroid in the Regulation of Ovarian Follicle-Stimulating Hormone Receptors During Pregnancy KAZUYOSHI TSUTSUI Department of Radiation Biophysics, Kobe University School of Medicine, Chuo-ku,Kobe 650, Japan
ABSTRACT The present study was conducted to determine the role of sex steroids in the regulation of FSH receptors in pregnant rats. In the normal physiological condition, FSH bindings per unit ovarian weight (density of binding) and per 2 ovaries (total binding) increased during days 14-21 gestation. Scatchard plot analyses of the binding suggested that the increase in FSH binding was due to an increase in the number of FSH-binding sites. The plasma FSH concentration in pregnant rats was stable during the receptor change. In contrast, the plasma estradiol-176 concentration continuously increased from gestation day 14 to 21, and the testosterone level showed a large peak on gestation day 18. Estradiol-17P (one silastic plate containing 13 mg crystal)-implanted pregnant rats during 14-21 days of gestation induced significant decreases in the total FSH binding and ovarian weight on gestation day 21. Estradiol administration increased the plasma estradiol level 2.3-fold but did not change the FSH level. Testosterone or 5 a-dihydrotestosterone, a nonaromatizable androgen, did not influence the binding level under the same dose treatment. In contrast, continuous treatment with aminoglutethimide (2 plates containing 20 mg crystal), an inhibitor of adrenocortical steroidogenesis, for 7 days significantly increased the total FSH binding without a significant change in the ovarian weight. The plasma titers of estradiol and testosterone in pregnant rats treated with aminoglutethimide were reduced by 37% and 51%, respectively. Aminoglutethimide did not influence plasma FSH levels. These results suggest that circulating estradiol acts as a negative factor in the regulation of ovarian FSH receptors, a t least during the second half of pregnancy. Other factor(s)that is (are) independent of sex steroids and FSH may contribute to FSH receptor induction. 01992 Wiley-Liss, Inc.
Follicle-stimulating hormone (FSH)promotes the growth of ovarian follicles and stimulates preovulatory ovarian estrogen production (Dahl and Hsueh, '88).FSH binding is mainly restricted to granulosa cells, as ascertained by autoradiographic, in vitro binding and in vivo uptake studies with radioiodinated hormone (Richards and Midgley, '76; Nimrod et al., '76; Ireland and Richards, '78; Carson et al., '79). Granulosa cells of essentially all follicles appear to possess FSH receptors (Richards et al., '76; Richards and Midgley, '76; Richards, '78). During pregnancy the change in follicular growth occurs in murine rodents (Richards, '80).Previous investigators (Richards et al., '78; Richards, '79)reported that in the pregnant rat FSH receptors in the granulosa cell of developing follicles were constant and abruptly increased during late pregnancy. As summarized by Richards ('80), the upregulation of FSH and the synergistic action between FSH (peptide hormone) and estradiol (steroid hormone) are believed to be main causes of the induction of ovarian FSH receptors in the cyclic rat. However, the 01992 WILEY-LISS, INC.
increase in FSH receptors during late pregnancy seems to be independent of FSH, because the change in plasma FSH levels was not pronounced during FSH receptor change (Linkie and Niswender, '72; Cheng, '76). Thus, the mechanism of FSH receptor regulation during the second half of pregnancy is not yet understood in murine rodents. Sex steroids are involved in pregnancy. In murine rodents, estradiol is one of essential hormones involved in a successful pregnancy (Rowlands and Weir, '84; Spies and Chappel, '84; Yen and Lein, ,841, and the circulating level increases during late pregnancy (McCormack and Greenwald, '74; Shaikh, '71; Taya and Greenwald, '81). Androgen, which is an important component of luteotropic process in the ovary, is secreted during pregnancy (Gibori and Keyes, '78; Gibori et al., '78). Circulating testosterone shows a peak during the second half of pregnancy (Gibori et al., '79; Stahl et al., '84).It is Received February 27,1992; revision accepted May 21,1992.
168
K. TSUTSUI
considered that not only the ovary, but also the adrenal and placenta are important sources of sex steroids during pregnancy (Challis et al., '75; Sridaran et al., '81; Macdonald and Matt, '84). On the other hand, 5a-dihydrotestosterone, a nonaromatizable androgen, in the ovary increases at the end of pregnancy (Sridaran and Gibori, '81). Therefore,it is probable that gonadal and nongonadal sex steroids contribute to the regulation of ovarian FSH receptor during pregnancy. To elucidate the role of sex steroids in the regulation of ovarian FSH receptors during pregnancy, pregnant rats were continuously treated with sex steroids, i.e., estradiol, testosterone, and dihydrotestosterone, or aminoglutethimide, an inhibitor of adrenocortical steroidogenesis (Cash et al., '67; Asbury et al., '81) by silastic plate implantation. In the present study not only ovarian FSH receptors, but also circulating levels of sex steroids and FSH were measured to determine the effects of sex steroids.
taramide (Sigma, St. Louis), according to the same schedule as the second experiment.
MATERIALS AND METHODS Experimental schedules Adult females of Wistar/'ltv rats were mated with adult males, and those showing sperms in the vaginal smears at 1O:OO-11:OO a.m. were designated as 0 day of gestation. The gestation period is usually 22 days (Tsutsui and Kawashima, '87). In the first series of experiments, 10 pregnant rats were sacrificed by decapitationbetween 1000 and 11:OO a.m. on the 14th, 18th, and 21st days of gestation to determine the normal changes in ovarian FSH receptors and plasma concentrations of estradiol, testosterone, and FSH. In the second series of experiments, 25 anesthetized pregnant rats were treated with a silastic plate containing testosterone, estradiol-l7P, or 5a-dihydrotestosterone on the 14th day of gestation. A silastic plate (1.5 X 17 X 2 mm; approx. 13 mg crystal/plate), which was made of a mixture of medical silastic adhesive (silicone type, Dow Corning, Michigan) and crystalline sex steroid (Sigma, St. Louis), was implanted into the fat body surrounding the uterus. Sex-steroid-implantedrats were sacrificed by decapitation on the 21st day of gestation, along with control rats that were implanted with only silastic adhesive, to examine the effects of sex steroids on FSH receptors and plasma hormone levels. In the last experiment, 15 anesthetized pregnant rats were treated with 2 silastic plates (1.5 x 15 x 2 mm; approx. 10 mg crystal/plate) containing gluaminoglutethimide, p-(a-amino-phenylla-ethyl
Binding assay
Preparations ofplasma and receptor samples Trunk blood was collected into heparinized glass tubes and centrifuged at 1,800 g for 30 min at 4°C. Plasma was stored at - 20°C until assayed for estradiol, testosterone, and FSH. Immediately after blood collection, the ovaries were removed and weighed. They were snap-frozenon dry ice-ethanol and stored at -80°C until the binding assays for FSH and luteinizing hormone (LH) were performed. For the receptor preparation, the frozen samples were rapidly thawed and homogenized in cold Tris-HC1buffer (0.04 M; pH 7.4) containing MgS04 (5 d) and 0.1% BSA. The homogenates were centrifuged at 11,000 g for 20 min at 4°C. The resulting pellets were resuspended in cold buffer and adjusted t o contain 4 mgEq wet tissue/100 pl. The suspension was used as the receptor preparation.
NIDDK-rat FSH (rFSH)-1-6and NIDDK-rat LH (rLH)-I-6 were radioiodinated with I3'I (Na13'I, Radiochemical Centre, Amersham, United Kingdom) in the presence of lactoperoxidase and hydrogen peroxidase using a method described previously (Tsutsui and Ishii, '78, '80). The specific activities of [1311]iodo-rFSHand [13111iodo-rLHwere 1.2 and 1.6 MBq/pg, respectively. Unlabeled USDA-FSH-I31 and NIDDK-ovine (o)LH-24were used to correct for nonspecific binding throughout the assays of FSH and LH receptors, respectively. For the FSH-binding assay (Tsutsui et al., '85, '88; Tsutsui, '91), receptor preparation (100 pl; 4 mgEq original tissue) and [13111iodo-rFSH(50 p1; 1.45 ng) were incubated at 35°C for 2 hr with or without unlabeled FSH (50 p1; 20 pg USDA-FSHBl). Due to the small amount of membrane available, the saturation binding experiment was performed using a micro-RRA, described previously (Tsutsui and Kawashima, '85; Tsutsui, '91). In the saturation binding experiments, different amounts of [13111iodo-rFSH(20 p1; 0.24-3.87 ng) and receptor preparations (50 pl; 2 mgEq original tissue) were incubated with or without an excess of cold USDAFSH-Bl(20 p1; 5-80 pg). For the LH-binding assay (Tsutsui et al., '85, '89; Tsutsui, ,911, a receptor preparation (100 pl; 4mgEqoriginal tissue) and [1311]iodoLH (50 pl; 1.7 ng) were incubated at 35°C for 2 hr with or without unlabeled LH (50 p1; 1pg NIDDKoLH-24). At the end of incubation, 1ml cold TrisHCl buffer (0.04 M; pH 7.4) containing 5 mM MgSO,
REGULATION OF FSH RECEPTORS DURING PREGNANCY
and 0.1% BSA was added to each tube, and the tubes were centrifuged at 11,000 g for 3 min at 4°C. The pellets were washed twice with cold buffer, and the radioactivity of resultant pellets was counted in an autowell gamma counter. Before the experiments, all reaction tubes had been coated with BSA. Scatchard plots were constructed from the data obtained from the saturation binding experiment. The dissociation constant (Kd) and the number of binding sites were determined from the Scatchard plots. Straight lines were fitted to the plots by the method of least squares. Statistics for linearity, precision, and 95% confidence interval were computed according to the method of Bliss ('52).
RIA Testosterone concentrations were measured by RIA (Tsutsui et al., '88, '89, '91) using an antiserum of testosterone (TeikokuZoki Pharmaceutical Co., Ltd., Tokyo, Japan) and [l,2, 6, 7-3Hltestosterone (Radiochemical Center, Amersham, United Kingdom) after extraction of 50 pl diluted or undiluted plasma samples by ether. The assay was performed without chromatographic purification of testosterone, and the first antiserum used in the present experiment cross-reactedwith 5 a-dihydrotestosterone at about 13.5%. The least detectable amount was 39 pg/ml, and intraassay variation was less than 10%. The estradiol-17p RIA (Itoh et al., '90) was performed with anti-estradiol-17p serum (Teikoku Zoki Pharmaceutical Co., Ltd.) and [2,4, 6,7-3Hlestradiol (Radiochemical Center) after extraction of 100 p1 undiluted plasma samples by ether. The serum cross-reactedwith estriol at about 1.77%,with testosterone at about 0.29%, and with progesterone at less than 0.08%. The least detectable amount was 9.8 pg/ml, and intraassay variation was less than 10%. Separation of bound and free hormones was performed using a double-antibody method for testosterone and estradiol. Plasma FSH was measured by RIA using a double-antibody method (Tsutsui et al., '85, '88; Tsutsui, '91). The FSH concentration assayed in 50 p1 plasma was calculated in terms of nanograms of NIDDK-rFSHRP-2/ml. Reagents for RIA of FSH were kindly provided by Dr. S. Raiti, NIDDK, Rat Pituitary Hormone Distribution Program, NIH (Bethesda, MD), and Dr. A. F. Parlow,Pituitary HormonedAntisera Center, Harbor-Universityof California-Los Angeles. The least detectable amount was 0.49 ng NIDDK-rFSH-RP-2/ml,and intraassay variation was less than 9%.
169
Statistical analyses Results were expressed as the mean -t SEM and were analyzed for significance by Duncan's multiple range test or Student's t test (Bliss, '67).
RESULTS Changes in ovarian FSH receptors and plasma concentrations of sex steroids and FSH during pregnancy Pregnant rats were sacrificed by decapitation on the 14th, 18th, and 21st days of gestation. The weights of 2 ovaries were almost constant up to gestation day 18 and significantly increased by day 21 (gestation day 18 vs. 21: P < 0.05; day 14vs. 21: P < 0.01; Fig. 1A).The specific binding of [13111iodo-rFSH per unit ovarian weight (density of FSH binding) increased during days 14-21 of gestation (gestation day 14 vs. day 21: P < 0.05; Fig. 1B). A significant increase in FSH binding per 2 ovaries (total FSH binding) was also detected during the same period (gestation day 14 or day 18 vs. day 21: P < 0.01; Fig. 10. The increase in FSH binding seemed to take place prior t o that in ovarian weight. To determine whether these changes in the binding of [13111iodo-rFSHto the ovary during pregnancy were due to the changes in the number of binding sites or changes in the affinity of binding, Scatchard plot analyses were performed on ovarian preparations of pregnant rats on gestation days 14 and 21. Scatchard plots showed significantly (P < 0.05) straight lines in both groups, suggesting the presence of a single class of FSH-binding sites (Fig. 2). The equilibrium Kd values calculated from the fitted lines of the plots were 4.37 (95%confidence interval, 3.33-6.33) x 10-loM on gestation day 14 and 4.59 (3.54-6.50) x 1O-loMon gestationday 21. The Kd values of the 2 groups indicated no significant change in the affinity of FSH binding. The mean numbers of FSH-binding sites (capacity) in the ovaries of pregnant rats on gestation days 14 and 21 were 1.23 (95%confidence interval, 1.07-1.53) and 1.76 (1.56-2.15) x 10-15mol/mgtissueequivalent, respectively. There was a significant difference (P < 0.05) in the number of binding sites per milligram tissue equivalent between the 2 groups. The total numbers of FSH-binding sites per 2 ovaries in pregnant rats on gestation days 14 and 21 were 1.04 and 2.01 x mol, respectively. Thus, the increase in FSH binding during the second half of pregnancy was due to an increase in the number of binding sites. As shown in Fig. 3A, the plasma testosterone concentration in pregnant rats markedly increased (I'
K. TSUTSUI
170
I
L
B
I
I
c
"4 OL
'
F
I I
O1
20
40
Fig. 2. Scatchard plots of the binding of rFSH to receptor preparations of pregnant rats on gestation day 14 (0-0) and 21 (0-0). B = Concentration of bound hormone at apparent equilibrium; F = concentration of free hormone at apparent equilibrium.
0
14
L
L
16
18
20
Days of pregnancy
Fig. 1. Changes in the ovaries weight (A) during pregnancy, specific binding of ['3111iodo-rFSHper 4 mg ovarian tissue (B), and specificbinding of [1311]iodo-rFSHper 2 ovaries ( C )in rats. Incubation was for 2 hr at 35°C. Each point depicts the mean k SEM. The number of pregnant rats is indicated in parentheses. Significance ofdifference: * P < 0.05;** P < 0.01 (gestation day18vs.21):IrP
Inhibitory Role of Sex Steroid in the Regulation of Ovarian Follicle-Stimulating Hormone Receptors During Pregnancy KAZUYOSHI TSUTSUI Department of Radiation Biophysics, Kobe University School of Medicine, Chuo-ku,Kobe 650, Japan
ABSTRACT The present study was conducted to determine the role of sex steroids in the regulation of FSH receptors in pregnant rats. In the normal physiological condition, FSH bindings per unit ovarian weight (density of binding) and per 2 ovaries (total binding) increased during days 14-21 gestation. Scatchard plot analyses of the binding suggested that the increase in FSH binding was due to an increase in the number of FSH-binding sites. The plasma FSH concentration in pregnant rats was stable during the receptor change. In contrast, the plasma estradiol-176 concentration continuously increased from gestation day 14 to 21, and the testosterone level showed a large peak on gestation day 18. Estradiol-17P (one silastic plate containing 13 mg crystal)-implanted pregnant rats during 14-21 days of gestation induced significant decreases in the total FSH binding and ovarian weight on gestation day 21. Estradiol administration increased the plasma estradiol level 2.3-fold but did not change the FSH level. Testosterone or 5 a-dihydrotestosterone, a nonaromatizable androgen, did not influence the binding level under the same dose treatment. In contrast, continuous treatment with aminoglutethimide (2 plates containing 20 mg crystal), an inhibitor of adrenocortical steroidogenesis, for 7 days significantly increased the total FSH binding without a significant change in the ovarian weight. The plasma titers of estradiol and testosterone in pregnant rats treated with aminoglutethimide were reduced by 37% and 51%, respectively. Aminoglutethimide did not influence plasma FSH levels. These results suggest that circulating estradiol acts as a negative factor in the regulation of ovarian FSH receptors, a t least during the second half of pregnancy. Other factor(s)that is (are) independent of sex steroids and FSH may contribute to FSH receptor induction. 01992 Wiley-Liss, Inc.
Follicle-stimulating hormone (FSH)promotes the growth of ovarian follicles and stimulates preovulatory ovarian estrogen production (Dahl and Hsueh, '88).FSH binding is mainly restricted to granulosa cells, as ascertained by autoradiographic, in vitro binding and in vivo uptake studies with radioiodinated hormone (Richards and Midgley, '76; Nimrod et al., '76; Ireland and Richards, '78; Carson et al., '79). Granulosa cells of essentially all follicles appear to possess FSH receptors (Richards et al., '76; Richards and Midgley, '76; Richards, '78). During pregnancy the change in follicular growth occurs in murine rodents (Richards, '80).Previous investigators (Richards et al., '78; Richards, '79)reported that in the pregnant rat FSH receptors in the granulosa cell of developing follicles were constant and abruptly increased during late pregnancy. As summarized by Richards ('80), the upregulation of FSH and the synergistic action between FSH (peptide hormone) and estradiol (steroid hormone) are believed to be main causes of the induction of ovarian FSH receptors in the cyclic rat. However, the 01992 WILEY-LISS, INC.
increase in FSH receptors during late pregnancy seems to be independent of FSH, because the change in plasma FSH levels was not pronounced during FSH receptor change (Linkie and Niswender, '72; Cheng, '76). Thus, the mechanism of FSH receptor regulation during the second half of pregnancy is not yet understood in murine rodents. Sex steroids are involved in pregnancy. In murine rodents, estradiol is one of essential hormones involved in a successful pregnancy (Rowlands and Weir, '84; Spies and Chappel, '84; Yen and Lein, ,841, and the circulating level increases during late pregnancy (McCormack and Greenwald, '74; Shaikh, '71; Taya and Greenwald, '81). Androgen, which is an important component of luteotropic process in the ovary, is secreted during pregnancy (Gibori and Keyes, '78; Gibori et al., '78). Circulating testosterone shows a peak during the second half of pregnancy (Gibori et al., '79; Stahl et al., '84).It is Received February 27,1992; revision accepted May 21,1992.
168
K. TSUTSUI
considered that not only the ovary, but also the adrenal and placenta are important sources of sex steroids during pregnancy (Challis et al., '75; Sridaran et al., '81; Macdonald and Matt, '84). On the other hand, 5a-dihydrotestosterone, a nonaromatizable androgen, in the ovary increases at the end of pregnancy (Sridaran and Gibori, '81). Therefore,it is probable that gonadal and nongonadal sex steroids contribute to the regulation of ovarian FSH receptor during pregnancy. To elucidate the role of sex steroids in the regulation of ovarian FSH receptors during pregnancy, pregnant rats were continuously treated with sex steroids, i.e., estradiol, testosterone, and dihydrotestosterone, or aminoglutethimide, an inhibitor of adrenocortical steroidogenesis (Cash et al., '67; Asbury et al., '81) by silastic plate implantation. In the present study not only ovarian FSH receptors, but also circulating levels of sex steroids and FSH were measured to determine the effects of sex steroids.
taramide (Sigma, St. Louis), according to the same schedule as the second experiment.
MATERIALS AND METHODS Experimental schedules Adult females of Wistar/'ltv rats were mated with adult males, and those showing sperms in the vaginal smears at 1O:OO-11:OO a.m. were designated as 0 day of gestation. The gestation period is usually 22 days (Tsutsui and Kawashima, '87). In the first series of experiments, 10 pregnant rats were sacrificed by decapitationbetween 1000 and 11:OO a.m. on the 14th, 18th, and 21st days of gestation to determine the normal changes in ovarian FSH receptors and plasma concentrations of estradiol, testosterone, and FSH. In the second series of experiments, 25 anesthetized pregnant rats were treated with a silastic plate containing testosterone, estradiol-l7P, or 5a-dihydrotestosterone on the 14th day of gestation. A silastic plate (1.5 X 17 X 2 mm; approx. 13 mg crystal/plate), which was made of a mixture of medical silastic adhesive (silicone type, Dow Corning, Michigan) and crystalline sex steroid (Sigma, St. Louis), was implanted into the fat body surrounding the uterus. Sex-steroid-implantedrats were sacrificed by decapitation on the 21st day of gestation, along with control rats that were implanted with only silastic adhesive, to examine the effects of sex steroids on FSH receptors and plasma hormone levels. In the last experiment, 15 anesthetized pregnant rats were treated with 2 silastic plates (1.5 x 15 x 2 mm; approx. 10 mg crystal/plate) containing gluaminoglutethimide, p-(a-amino-phenylla-ethyl
Binding assay
Preparations ofplasma and receptor samples Trunk blood was collected into heparinized glass tubes and centrifuged at 1,800 g for 30 min at 4°C. Plasma was stored at - 20°C until assayed for estradiol, testosterone, and FSH. Immediately after blood collection, the ovaries were removed and weighed. They were snap-frozenon dry ice-ethanol and stored at -80°C until the binding assays for FSH and luteinizing hormone (LH) were performed. For the receptor preparation, the frozen samples were rapidly thawed and homogenized in cold Tris-HC1buffer (0.04 M; pH 7.4) containing MgS04 (5 d) and 0.1% BSA. The homogenates were centrifuged at 11,000 g for 20 min at 4°C. The resulting pellets were resuspended in cold buffer and adjusted t o contain 4 mgEq wet tissue/100 pl. The suspension was used as the receptor preparation.
NIDDK-rat FSH (rFSH)-1-6and NIDDK-rat LH (rLH)-I-6 were radioiodinated with I3'I (Na13'I, Radiochemical Centre, Amersham, United Kingdom) in the presence of lactoperoxidase and hydrogen peroxidase using a method described previously (Tsutsui and Ishii, '78, '80). The specific activities of [1311]iodo-rFSHand [13111iodo-rLHwere 1.2 and 1.6 MBq/pg, respectively. Unlabeled USDA-FSH-I31 and NIDDK-ovine (o)LH-24were used to correct for nonspecific binding throughout the assays of FSH and LH receptors, respectively. For the FSH-binding assay (Tsutsui et al., '85, '88; Tsutsui, '91), receptor preparation (100 pl; 4 mgEq original tissue) and [13111iodo-rFSH(50 p1; 1.45 ng) were incubated at 35°C for 2 hr with or without unlabeled FSH (50 p1; 20 pg USDA-FSHBl). Due to the small amount of membrane available, the saturation binding experiment was performed using a micro-RRA, described previously (Tsutsui and Kawashima, '85; Tsutsui, '91). In the saturation binding experiments, different amounts of [13111iodo-rFSH(20 p1; 0.24-3.87 ng) and receptor preparations (50 pl; 2 mgEq original tissue) were incubated with or without an excess of cold USDAFSH-Bl(20 p1; 5-80 pg). For the LH-binding assay (Tsutsui et al., '85, '89; Tsutsui, ,911, a receptor preparation (100 pl; 4mgEqoriginal tissue) and [1311]iodoLH (50 pl; 1.7 ng) were incubated at 35°C for 2 hr with or without unlabeled LH (50 p1; 1pg NIDDKoLH-24). At the end of incubation, 1ml cold TrisHCl buffer (0.04 M; pH 7.4) containing 5 mM MgSO,
REGULATION OF FSH RECEPTORS DURING PREGNANCY
and 0.1% BSA was added to each tube, and the tubes were centrifuged at 11,000 g for 3 min at 4°C. The pellets were washed twice with cold buffer, and the radioactivity of resultant pellets was counted in an autowell gamma counter. Before the experiments, all reaction tubes had been coated with BSA. Scatchard plots were constructed from the data obtained from the saturation binding experiment. The dissociation constant (Kd) and the number of binding sites were determined from the Scatchard plots. Straight lines were fitted to the plots by the method of least squares. Statistics for linearity, precision, and 95% confidence interval were computed according to the method of Bliss ('52).
RIA Testosterone concentrations were measured by RIA (Tsutsui et al., '88, '89, '91) using an antiserum of testosterone (TeikokuZoki Pharmaceutical Co., Ltd., Tokyo, Japan) and [l,2, 6, 7-3Hltestosterone (Radiochemical Center, Amersham, United Kingdom) after extraction of 50 pl diluted or undiluted plasma samples by ether. The assay was performed without chromatographic purification of testosterone, and the first antiserum used in the present experiment cross-reactedwith 5 a-dihydrotestosterone at about 13.5%. The least detectable amount was 39 pg/ml, and intraassay variation was less than 10%. The estradiol-17p RIA (Itoh et al., '90) was performed with anti-estradiol-17p serum (Teikoku Zoki Pharmaceutical Co., Ltd.) and [2,4, 6,7-3Hlestradiol (Radiochemical Center) after extraction of 100 p1 undiluted plasma samples by ether. The serum cross-reactedwith estriol at about 1.77%,with testosterone at about 0.29%, and with progesterone at less than 0.08%. The least detectable amount was 9.8 pg/ml, and intraassay variation was less than 10%. Separation of bound and free hormones was performed using a double-antibody method for testosterone and estradiol. Plasma FSH was measured by RIA using a double-antibody method (Tsutsui et al., '85, '88; Tsutsui, '91). The FSH concentration assayed in 50 p1 plasma was calculated in terms of nanograms of NIDDK-rFSHRP-2/ml. Reagents for RIA of FSH were kindly provided by Dr. S. Raiti, NIDDK, Rat Pituitary Hormone Distribution Program, NIH (Bethesda, MD), and Dr. A. F. Parlow,Pituitary HormonedAntisera Center, Harbor-Universityof California-Los Angeles. The least detectable amount was 0.49 ng NIDDK-rFSH-RP-2/ml,and intraassay variation was less than 9%.
169
Statistical analyses Results were expressed as the mean -t SEM and were analyzed for significance by Duncan's multiple range test or Student's t test (Bliss, '67).
RESULTS Changes in ovarian FSH receptors and plasma concentrations of sex steroids and FSH during pregnancy Pregnant rats were sacrificed by decapitation on the 14th, 18th, and 21st days of gestation. The weights of 2 ovaries were almost constant up to gestation day 18 and significantly increased by day 21 (gestation day 18 vs. 21: P < 0.05; day 14vs. 21: P < 0.01; Fig. 1A).The specific binding of [13111iodo-rFSH per unit ovarian weight (density of FSH binding) increased during days 14-21 of gestation (gestation day 14 vs. day 21: P < 0.05; Fig. 1B). A significant increase in FSH binding per 2 ovaries (total FSH binding) was also detected during the same period (gestation day 14 or day 18 vs. day 21: P < 0.01; Fig. 10. The increase in FSH binding seemed to take place prior t o that in ovarian weight. To determine whether these changes in the binding of [13111iodo-rFSHto the ovary during pregnancy were due to the changes in the number of binding sites or changes in the affinity of binding, Scatchard plot analyses were performed on ovarian preparations of pregnant rats on gestation days 14 and 21. Scatchard plots showed significantly (P < 0.05) straight lines in both groups, suggesting the presence of a single class of FSH-binding sites (Fig. 2). The equilibrium Kd values calculated from the fitted lines of the plots were 4.37 (95%confidence interval, 3.33-6.33) x 10-loM on gestation day 14 and 4.59 (3.54-6.50) x 1O-loMon gestationday 21. The Kd values of the 2 groups indicated no significant change in the affinity of FSH binding. The mean numbers of FSH-binding sites (capacity) in the ovaries of pregnant rats on gestation days 14 and 21 were 1.23 (95%confidence interval, 1.07-1.53) and 1.76 (1.56-2.15) x 10-15mol/mgtissueequivalent, respectively. There was a significant difference (P < 0.05) in the number of binding sites per milligram tissue equivalent between the 2 groups. The total numbers of FSH-binding sites per 2 ovaries in pregnant rats on gestation days 14 and 21 were 1.04 and 2.01 x mol, respectively. Thus, the increase in FSH binding during the second half of pregnancy was due to an increase in the number of binding sites. As shown in Fig. 3A, the plasma testosterone concentration in pregnant rats markedly increased (I'
K. TSUTSUI
170
I
L
B
I
I
c
"4 OL
'
F
I I
O1
20
40
Fig. 2. Scatchard plots of the binding of rFSH to receptor preparations of pregnant rats on gestation day 14 (0-0) and 21 (0-0). B = Concentration of bound hormone at apparent equilibrium; F = concentration of free hormone at apparent equilibrium.
0
14
L
L
16
18
20
Days of pregnancy
Fig. 1. Changes in the ovaries weight (A) during pregnancy, specific binding of ['3111iodo-rFSHper 4 mg ovarian tissue (B), and specificbinding of [1311]iodo-rFSHper 2 ovaries ( C )in rats. Incubation was for 2 hr at 35°C. Each point depicts the mean k SEM. The number of pregnant rats is indicated in parentheses. Significance ofdifference: * P < 0.05;** P < 0.01 (gestation day18vs.21):IrP
