The Specificity of Gonadotropin Binding by the Human Corpus Luteum*†

FERTILITY AND STERILITY

Vol. 27, No.8, August 1976 Printed in U.S.A .

Copyright < 1976 The American Fertility Society

•

THE SPECIFICITY OF GONADOTROPIN BINDING BY THE HUMAN CORPUS LUTEUM*t FRANCIS E. COLE, PH.D.,:!: JOHN C. WEED, M.D.,§ GEORGE T. SCHNEIDER, M.D.,§ JOHN B. HOLLAND, M.D.,§ WILLIAM L. GEARY, M.D.,§ DONALD L. LEVY, M.D.,§ ROBERT A. HUSEBY, M.D., PH.D.,~ AND BERNARD F. RICE, M.D.:f:

The Richard W. Freeman Research Institute, Division of Endocrine Research, Section II, and Department of Obstetrics and Gynecology, Alton Ochsner Medical Foundation, New Orleans, Louisiana 70121, and the American Medical Center at Denver, Spivak, Colorado 80028 The specificity of gonadotropin binding was studied in fresh and frozen human corpora lutea. Ouine, bovine, and porcine luteinizing hormone (LH) competed with 125/-labeled human LH (1 25/-hLH) and 125/-labeled human chorionic gonadotropin (1 251-hCG) for binding to tissue receptors in homogenates of human corpora lutea frozen for3 to 12 months. In contrast, oLH, bLH, and pLH competed minimally for 1251-hLH and 125/-hCG binding sites in homogenates of fresh human corpora lutea. Ouine follicle-stimulating hormone (FSH) and thyroid-stimulating hormone (TSH) did not compete in homogenates of fresh or frozen tissue. Competition of oLH and hCG for 1251-hCG binding sites at several dose levels in a homogenate ofa fresh corpus luteum was studied. One hundred micrograms of oLH and ten nanograms of hCG gave an equivalent competition--a 1 0,000-fold difference in competitive potency. Only hCG competed with 1251-hCG for binding when the competition of oLH, bLH, pLH, oFSH, oTSH, hCG and hCG subunits, and hCG were compared at the 10-pg level in a homogenate of fresh human corpus luteum. The binding of 125/-labeled homologous human hormones by the corpus luteum was examined in a limited fashion. 1251-Prolactin did not bind to preparations of fresh stroma from a patient with polycystic ovaries nor did it bind to three separate preparations of fresh corpora lutea which did bind 125/-hCG. 125/-hTSH did not show significant binding to a fresh human corpus luteum preparation which did bind 125/-hCG. These studies indicate that the gonadotropin receptor ofthe fresh human corpus luteum possesses a unique species specificity and illustrate the importance of working with human corpora lutea in their most native state. Accepted March 19, 1976. *Supported in part by Grant FR-05518 from the National Institutes of Health, a grant from the Greater New Orleans Cancer Association, and the Starr Research Fund. tPresented in part at the Thirtieth Annual Meeting of The American Fertility Society, April 4 to 6, 1974, Hollywood, Fla. :f:Richard W. Freeman Research Institute, Division of Endocrine Research, Section II, Alton Ochsner Medical Foundation, New Orleans, La. §Department of Obstetrics and Gynecology, Alton Ochsner Medical Foundation, New Orleans, La. ~American Medical Center at Denver, Spivak, Colo.

The hormone binding by a gonadotropin receptor in homogenates of fresh human corpora lutea exhibits a unique species specificity. The human receptor binds 125!-labeled human luteinizing hormone (1 25!-hLH) but not 125!-labeled ovine LH (1 25 !-oLH) or bovine LH (1 25 l-bLH). 1 LH from other species (ovine, bovine, and porcine), as well as other nonhuman trophic hormones, do not compete with 125!-hLH for binding in homogenates of fresh human corpora lutea. In this report, we demonstrate a similar and more ex-

921

COLE ETAL.

922

tensive species specificity for competition with 1251-labeled human chorionic gonadotropin ( 1251-hCG) binding by homogenates of fresh human corpora lutea. Since human corpora lutea are available for study on an irregular basis, many binding studies may be performed on frozen tissues. In contrast to the results obtained with fresh tissue in binding 1251-hLH or 1251-hCG, in this report we observe that homogenates made from frozen corpora lutea lose their species specificity for human luteinizing hormones but retain their specificity among certain other mammalian trophic hormones for those possessing LH activity. MATERIALS AND METHODS

Corpora lutea of the menstrual cycle were obtained from women undergoing elective surgery at the Ochsner Foundation Hospital. The day of the menstrual cycle was determined at the time of operation, within 2 days, from the patient's menstrual history, basal body temperature record, and endometrial histology by one of the authors (B. F. R.) in consultation with the pathology department. Binding studies on fresh corpora lutea were performed within 1 hour of surgery. Midluteal corpora lutea were used for binding studies, since very early (day 14), very late (day 28), or postmenstrual corpora lutea do not bind 125 l-hLH. 1 The frozen corpora lutea studied had been frozen in Earle's buffer (Krebs-Ringerbicarbonate buffer [KRBBI) and stored at - 20" C for 3 to 12 months before use. In specificity studies with frozen corpora lutea, several corpora lutea were pooled and homogenized to provide sufficient material. For fresh corpus luteum studies, individual corpora lutea were used. Corpus Luteum Homogenate Preparation. Fresh human corpora lutea were obtained within 15 minutes of operative removal and placed in ice-chilled KRBB. Excess connective and stromal tissues

August 1976

were removed and the corpus luteum was weighed (±0.02 gm). The corpus luteum was passed one time through an icechilled tissue press (Harvard Instrument Co., Millis, Mass.), followed by dilution to 100 mg/ml of KRBB, aspiration, and expulsion in a syringe with an 18-gauge needle. The resulting homogenates could be reproducibly pipetted by using a 1-ml pipette with its tip cut off. Homogenates of frozen tissue were prepared identically. Incubation Procedure. Typically, corpus luteum homogenates were incubated in triplicate with varying quantities (5 to 100 ng or 105 to 106 dpm, depending on specific activity) of 1251-labeled hormone in glass tubes (15 x 85 mm) containing 1-ml aliquots of homogenate ( -100 mg/ml) in KRBB to which were added 100 ml of KRBB with or without competing hormone; this mixture was shaken gently at 37° C, in an atmosphere of 95% 0 2 -5% C0 2 , for 100 minutes. After equilibrium had been reached (100 minutes), the incubation mixtures were centrifuged at 1700 x g for 30 minutes at 4° C. The supernatants were handaspirated with a disposable pipette, and the precipitates were resuspended in 1.0 ml of cold KRBB by mixing on a Vortex mixer for 2 minutes. After a second centrifugation, the supernatants were again aspirated, and the 1251 in the cell pellet was measured in a Packard AutoGamma spectrometer. Isotope. Low specific activity 1251-hLH, 1251-hCG, 1251-oLH, 1251-hPRL, 1251-hTSH, and 1251-hFSH (<25 JLCiiJLg) were prepared using lactoperoxidase. 2 A typical iodination reaction mixture contained 10 f.Lg of trophic hormone in 10 JLl, 1 mCi of Na 125l (Amersham/Searle Corporation, Des Plaines, Ill.) in 10 JLl, 20 JLl of 0.4 M sodium acetate buffer (pH 5.6), 30 IU of lactoperoxidase (B grade, 122 IU/mg; Calbiochem, Los Angeles, Calif.) in 25 f.Ll of 0.1 M sodium acetate buffer (pH 5.6), and 2.5 ng of H 2 0 2 in 50 f.Ll of distilled water. The latter

Vol. 27, No.8

0 .._,.:.==.u. Added None Hormone

923

SPECIFICITY OF GONADOTROPIN BINDING BY CORPUS LUTEUM

APL

hCG

oLH NIH-517

bLH NIH-B5

pLH LER-568-2

oF5H NIH-53

oT5H NIH-56

IOOI.U.

IOOpg

IOOpg

IOOpg

IOOpg

IOOpg

FIG. 1. Relative competition by trophic hormones for 1251-hLH (dots) and 1251-hCG (lines) in homogenates (-100 mg of tissue/ml) of fresh and frozen corpora lutea. Each bar represents the mean of triplicate determinations. Binding in the absence (None) of unlabeled hormone (80 ) was regarded as 100%. B0 was -7% of the added radioactivity (-5 x 10 5 cpm of specific activity -10 f.LCii(.Lg) with 1251-hLH and -15% of the added radioactivity (-1 x 10 6 cpm of specific activity -20 f.LCii(.Lg) with 1251-hCG.

two reagents were added twice during a 20-minute incubation. Purification of the labeled hormones was accomplished on 1% bovine serum albumin-coated Sephadex G-75 columns poured into 10-ml disposable pipettes. Under identical labeling and incubation conditions, 1251-hCG has a higher specific activity, higher total binding, and lower nonspecific binding than does 1251-hLH when used in the mouse luteoma3 • 4 gonadotropin receptor system. Specific activity of labeled peptide hormones was calculated by assuming that the mass of hormone recovered was proportional to the 1251 recovered. 4 Hormones. Luteinizing hormone (NIH-LH-817 and NIH-LH-B5), folliclestimulating hormone (NIH-FSH-83), hFSH (LER-1575-C), prolactin (NIH-PS9), hPRL (NIH-hPRL-2), thyroid-stimulating hormone (NIH-TSH-S-4), hTSH

(NIH-TSH-HS-3), and human luteinizing hormone (LER-960) were donated by the Pituitary Hormone Distribution Program, National Institute of Arthritis, Metabolism and Digestive Diseases. Human chorionic gonadotropin (hCG-APL, -2500 IU/mg of protein) was provided by Ayerst Laboratories, New York, N. Y. Purified hCG (CR-117, CR-121, -10,000 IU/mg) and its highly purified a and {3 subunits (CR-114) were provided by Dr. R. E. Canfield, Columbia College of Physicians and Surgeons, New York, N. Y. The purified porcine (pLH, LER-568-2, 0.8 x NIH-LH-S1 [ovarian ascorbic acid depletion assay)), bovine (bLH, LER-1373, 1.9 x NIH-LH-S1), and ovine (oLH, LER-1374, 1.78 x NIH-LH-S1) LH preparations were supplied by Dr. L. E. Reichert, Jr., Emory University School of Medicine, Atlanta, Ga.

924

COLE ET AL.

h CRI21 10Jig

APL 20 I.U.

August 1976

bLH Prolactin oGH hCGcr NIH-85 NIH-81 NIH-SIO CRII4 IOJig IOJig IOJig IOJig

H pLH NIH-S3 NIH-S6 LER-568·2 IOJig IOJig IOJig

FIG. 2. Relative competition for binding of 125!-hCG (6 x 105 cpm of ~20 JLCiiJLg) by 10-JLg quantities of various trophic hormones in a homogenate (100 mg/ml) of a fresh human corpus luteum. Each bar is the mean of triplicate determinations. B0 was 14% of the total radioactivity. RESULTS

Data from four experiments comparing the ability of trophic hormones to compete with 125 1-hCG and 1251-hLH binding in homogenates of fresh and frozen human corpora lutea are summarized in Figure 1. As previously reported/ 100

80·~-~

only homoiogous hCG competes with 1251-hLH for binding in homogenates of fresh human corpora lutea. In marked contrast, in homogenates of frozen corpora lutea, oLH, pLH, and bLH compete with 1251-hLH almost as effectively as homologous hLH.

·--.

0~

NIH-LH-S17 " - . .

af 60

' c

i

::r'----------'---~-...L__----'----lng

lOng

lOOng IJig Hormone Added

IOJig

IOOj/g

FIG. 3. Competition for binding of 125!-hCG ( ~ 1 x 106 cpm of ~20 JLCiiJLg) by increasing quantities of hCG and oLH in a homogenate (100 mg/ml) of a fresh human corpus luteum. Each point is the mean of triplicate determinations. B0 was 14% of the added radioactivity.

Vol. 27, No. 8

Using 125!-hCG as the labeled ligand, a similar pattern was observed. Whereas there was minimal competition of the animal gonadotropins in homogenates of fresh corpora lutea, in frozen corpora lute a the competition by the animal gonadotropins was essentially identical with that of hCG. Essentially no competition between oFSH or oTSH and 125!-hLH or 125!-hCG was observed in either fresh or frozen corpora lutea. When the abilities of animal and human trophic hormones and subunit preparations to compete for binding with 125!-hCG were compared at lower (10 p.,g/ ml) concentrations using a fresh human corpus luteum (Fig. 2), only native hCG competed with 125!-hCG for binding. The ability of oLH and hCG to compete with 125!-hCG for binding were compared in a single fresh corpus luteum (Fig. 3), using graded doses of each hormone. Competition with 125!hCG binding was again observed with oLH at the 100-p.,g dose. Ten thousandfold less hCG (10 ng) exhibited equivalent competition. Using a different approach, the ability of 125!-oLH to bind to fresh human corpus luteum and mouse luteoma homogenates was studied. 125!-oLH of low specific activity (~4 p.,Ci/p.,g) failed to bind to the fresh human corpus luteum (Table TABLE 1. Receptor Binding of 125/-oLH (LER-1374) by Human Corpus Luteum and Mouse Luteoma Homogenates" Competing hormone added

1-oLH

125

bound~>

1). When the receptor binding activity of identically prepared low specific activity 125!-oLH was tested using fresh mouse luteoma homogenate as the ganadotropin receptor, the 125!-oLH bound to the mouse luteoma homogenate (Table 1), as shown by the competition exerted by 100 IU of hCG. In testing the effect of short-term freezing and thawing on binding of human corpus luteum homogenates, total and nonspecific binding ofl 25l-hLH by a homogenate of a fresh corpus luteum diminished by nearly one-half when a replicate homogenate was frozen overnight and then studied (Table 2). In both the fresh and frozen homogenates, hCG showed competition but oLH and oFSH did not. TABLE 2. Receptor Specificity of a Human Corpus Luteum Homogenate": Fresh and Frozen Overnight '"1-hLH bound' Competing hormone added

Amount Frozen overnight

Fresh

cpm

108,000 None Human chorionic 100 IU 61,200 gonadotropin (Ayerst APL) 100 p,g 94,300 Ovine luteinizing hormone (NIH-LH-811) Ovine follicle100 p,g 111,100 stimulating hormone (NIH-FSH-83)

59,200 33,500 59,800 63,600

"Each tube contained 100 mg of corpus luteum homogenate in 1.1 ml of KRBB containing 6 x 105 cpm (-54 ng) of 1251-hLH with a specific activity of 10 p,Ci/p,g. The corpus luteum was from day 22 of the menstrual cycle. bMean of duplicate measurements.

Amount

H urn an corpus

luteum

Mouse luteoma

cpm

None hCG(APL)

925

SPECIFICITY OF GONADOTROPIN BINDING BY CORPUS LUTEUM

100 IU

34,500 35,250

28,200 17,000

"Each tube contained 100 mg of fresh tissue homogenate in 1.1 ml of KRBB. Both 1251-oLH preparations had a specific activity of -4 p,Ci!p,g. Approximately 1 x 106 cpm were added to the corpus luteum homogenate and 3 x 10' cpm were added to the mouse luteoma homogenate. bMean of duplicate (human corpus luteum) or triplicate (mouse luteoma) measurements.

The ability of fresh human corpus luteum homogenates to bind 125!-labeled homologous human hormones was also studied in a limited fashion. Data in Table 3 reveal that 125!-hPRL did not bind to a stromal preparation from a patient with polycystic ovaries, nor did it bind to two separate preparations of fresh corpora lutea which did bind 125!-hCG. 1251-hTSH did not show binding to a fresh human corpus luteum preparation which did bind

926

COLE ETAL.

August 1976

TABLE 3. Specificity of Binding of 1251-Labeled Human Trophic Hormone by Fresh Human Ovarian Tissues" Labeled hormone added (cpm)

Competing hormone added

t2:>J

boundb

Tissue

Patient

cpm

1-hPRL (345,157)

None hPRL (2.5 p.,g)

1-hPRL (268,654)

None hPRL (1 p.,g)

7,314 ± 735 9,772 ± 967

1-hCG (335,448)

None hCG (200 IU)

12,451 ± 329 8,147 ± 410

1-hPRL (157,450)

None hPRL (2 p.,g)

9,558 ± 754 9,621 ± 925

125

1-hCG (1,304,280)

None hCG (200 IU)

81,314 ± 382 22,661 ± 1,541

125

1-hPRL (210,280)

None hPRL (2 p.,g)

36,292 ± 1,906 38,167 ± 1,684

1-hCG (2,300,000)

None hCG (200 IU)

392,873 ± 12,052 94,938 ± 3, 781

1-hTSH (643,974)

None hTSH (2.5 p.,g)

48,471 ± 689 48,467 ± 2,453

1-hFSH (743,267)

None hFSH (2.5 p.,g)

16,154 ± 115 12,713 ± 209

1-hCG (805,776)

None hCG (200 IU)

31,133 ± 1,041 8,902 ± 262

125

125

125

125

125

125

125

125

18,029 ± 979 19,017 ± 428

Ovarian stroma, polycystic ovary

HFC no. 122 #222835

Late luteal corpus luteum (day 26c)

HFC no. 124 #355297

Early luteal corpus luteum (day 17c)

HFC no. 127 #194731

Midluteal corpus luteum (day 22c)

HFC no. 125 #526379

Late luteal corpus luteum (day 27c)

HFC no. 126 #527730

"Each tube contained ~100 mg/ml of tissue, isotope, and competing hormone in 1.1 ml of KRBB. bMean ± standard error of the mean. cDay of menstrual cycle.

I-hCG. In this same corpus luteum, I-hFSH appeared to show marginal binding. 125 125

DISCUSSION

These studies indicate that the human corpus luteum gonadotropin receptor, in the fresh state, exhibits a unique species specificity inasmuch as only human gonadotropins compete for binding when 125I-hCG is used for receptor binding experiments. These studies are in agreement with our previously reported study in which 125I-hLH was used as the labeled ligand. 1 In other gonadotropin receptor systems, such as mouse luteoma cells, 3 porcine granulosa cells, 5 luteinizing rat ovaries, 6 rat testes/ and bovine corpora lutea,H· 9 LH from other mammalian species competes with 125 I-hLH or 125 I-hCG

for binding to the gonadotropin receptor. In contrast, freezing human corpora lutea for 3 to 12 months destroys the high degree of specificity shown for human gonadotropins (Fig. 1). After prolonged freezing, ovine, bovine, and porcine LH compete with 125I-hLH and 125 I-hCG for binding to the human corpus luteum receptor almost as well as homologous hCG. Short-term freezing (overnight) does not destroy the high degree of specificity but does lower the total, specific, and nonspecific binding (Table 1), suggesting that the loss of specificity might be due to a lytic deterioration of the receptor elements governing specificity during the long period when tissues are frozen. Although the specificity for nonhuman 1uteinizing hormones is lost in frozen human luteal tissue, oTSH and oFSH do not compete for binding with

Vol. 27, No.8

•

SPECIFICITY OF GONADOTROPIN BINDING BY CORPUS LUTEUM

125 I-hLH or 125 I-hCG (Fig. 1) in fresh or frozen tissue. The hLH-hCG binding sites of nonhuman gonadotropin receptor systems share some specificity properties in common with the human corpus luteum (Figs. 1 and 3), as other nonluteinizing trophic hormones do not compete with 125 I-hLH or 125 I-hCG for binding. 3 - 10 The data of Figures 1, 2, and 3 suggest that the phenomenon of specificity may be more quantitative than absolute. The apparent time-dependent changes that occur in frozen tissue presumably accentuate this quantitative difference. Some competition with 125 I-hLH or 125IhCG was observed with extremely high concentrations (100 J.tg/ml) of nonhuman LH preparations (Figs. 1 and 3). When low concentrations of these gonadotropins, comparable to a human gonadotropin concentration that is saturating, were used, no competition with 125 I-hCG was observed with a fresh human corpus luteum preparation (Fig. 2). Purified subunits have been observed to be inactive in competing with 125I-hCG for the receptor in several gonadotropin receptor systems, 5 - 7 but this is the first report of their inability to compete in a homologous system with the exacting specificity requirements of the human corpus luteum. Although the role of prolactin in maintaining the corpora lutea in some species has been established, a similar role in humans is equivocal. Whereas Saito and Saxena 11 have demonstrated a low level of binding in a plasma membrane preparation of a single human corpus luteum, a recent study by Del Pozo et al. 12 indicates no action of prolactin in the regulation of the human menstrual cycle. Furthermore, there are limited data on the ability of human prolactin to stimulate progesterone synthesis in the human corpus luteum. 13 Although our laboratory was not equipped to assay the biologic activity of the iodinated trophic hormones,

927

a preliminary study was performed to attempt to elucidate the role of prolactin and to demonstrate the presence of separate ovarian receptors for 125 I-hPRL, 125 I-hTSH, and 125 I-hFSH (Table 3). 125 ILabeled hormones with specific activities of less than 25 JLCil JLg failed to bind to corpus luteum preparations which did bind 125I-hCG . Similar observations have been reported for FSH and TSH in other ovarian systems. In luteinized rat ovaries, Lee and Ryan 6 have demonstrated that there was no significant tissue uptake of 125 I-hFSH or 125I-hTSH. Our repeated failure to demonstrate 125 IhPRL binding in either ovarian stroma or three corpora lutea does not eliminate a possible role for prolactin in the human corpus luteum and must be interpreted with due reservation. It is now clear that the species specificity of gonadotropin binding by the human corpus luteum receptor is unique among the animal gonadotropin receptor systems studied. These data may reflect the large differences in the primary sequence of the a subunit of hLH when compared with that of other animal a subunits recently reported by Parlow and Slome. 14 There is great interest in the development of an agent which will compete with native human gonadotropins for the corpus luteum receptor, and data in this report emphasize the necessity of working with human tissues to develop an agent that will block the action of gonadotropin on the human corpus luteum. This specificity is labile and is best studied in unfrozen tissue. Acknowledgment. The authors are grateful for the technical assistance of Ms. Susan Miller for her help in the preparation of several of the iodinated trophic hormones. REFERENCES 1. Cole FE, Weed JC, Schneider GT, Holland JB,

Geary WL, Rice BF: The gonadotropin receptor of the human corpus luteum. Am J Obstet Gynecol117:87, 1973

928

COLEETAL.

2. Miyachi Y, Vaitukaitis JL, Neischlag E, Lipsett MB: Enzymatic radioiodination of gonadotropins. J Clin Endocrinol Metab 34:23, 1972 3. Cole FE, Davis K, Huseby RA, Rice BF: Gonadotropin receptor of a mouse luteoma: interactions with luteinizing hormone (LH) and its a and fJ subunits. Bioi Reprod 8:550, 1973 4. Cole FE, Rice BF: A method for the determination of the specific activity of receptorbound human chorionic gonadotropin. Endocrine Res Commun 2:109, 1975 5. Kammerman S, Canfield RE, Kolena J, Channing CP: The binding of iodinated hCG to porcine granulosa cells. Endocrinology 91:65, 1972 6. Lee CY, Ryan RJ: Luteinizing hormone receptors in luteinized rat ovaries. Adv Exp Med Bioi 37:419, 1973 7. Catt KJ, Dufau ML, Tsuruhara T: Radioligand-receptor assay of luteinizing hormone and chorionic gonadotropin. J Clin Endocrinol Metab 34:123, 1972 8. Haour F, Saxena BB: Characterization and solubilization of gonadotropin receptor of bovine corpus luteum. J Biol Chern 249:2195, 1974

August 1976

9. Rao ChV: Properties of gonadotropin receptors in the cell membranes of bovine corpus luteum. J Bioi Chern 249:2864, 1974 10. Lee CY, Coulam CB, Jiang NS, Ryan RJ: Receptors for human luteinzing hormone in human corpora luteal tissue. J Clin Endocrinol Metab 36:148, 1973 11. Saito T, Saxena BB: Specific receptors for prolactin in the ovary. Acta Endocrinol (Kbh) 80:126, 1975 12. Del Pozo E, Goldstein M, Friesen H, Brun del Re R, Eppenberger V: Lack of action of prolactin suppression on the regulation of the human menstrual cycle. Am J Obstet Gynecol 123:719, 1975 13. Savard K, Marsh J, Rice BF: Gonadotropins and ovarian steroidogenesis. Recent Prog Horm Res 21:285, 1965 14. Parlow AF, Slome B: The primary structure and radioimmunoassay of the a and fJ subunits of rat luteinizing hormone (rLH). Abstract 233, Program of 57th Annual Meeting of the Endocrine Society, New York, June 1975