0021-972X/91/7301-0084$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1991 by The Endocrine Society

Vol. 73, No. 1 Printed in U.S.A.

The Secretion of Human Chorionic Gonadotropin as well as the a- and /? Messenger Ribonucleic Acid Levels Are Stimulated by Exogenous Gonadoliberin Pulses Applied to First Trimester Placenta in a Superfusion Culture System W. E. MERZ, C. ERLEWEIN, P. LICHT, AND P. HARBARTH Department of Biochemistry II, University of Heidelberg, 6900 Heidelberg, Germany

ABSTRACT. It is well documented that the hypothalamic decapeptide gonadoliberin (GnRH) controls the biosynthesis and secretion of pituitary gonadotropins; however, it is still unclear whether GnRH synthesized by the placenta plays the same role with respect to hCG. In the current study we have investigated the acute response of placenta tissue to a single GnRH pulse as well as the influence of GnRH pulses on the secretion of hCG and hCG mRNA concentrations elicited several hours after application of the peptide hormone. For this purpose we have used a superfusion culture model of first trimester placenta tissue (8-12 weeks of gestation). In the first hour after explantation of the tissue, hCG secretion was decreased, and increasing amounts of free subunits were released. Afterward, the original hCG secretion rates were recovered and maintained for several days, attended by decreased levels of free subunits in the culture medium. The superfusion model was superior to static incubations, since it showed approximately a 4-fold higher amount of hCG to be secreted within 24 h (day 3 of cultures).

A single GnRH pulse (1 /nmol/L; 30 min) caused a significantly increased transient release of hCG (P < 0.0001). Two GnRH pulses (concentration range, 0.01-10 /onol/L; 30 min) applied 24 h after (first pulse) and in the interval between 3648 h after (second pulse) the start of the superfusion culture elicited a long-lasting 2-fold increase in the hCG secretion rate, which rose approximately 6 h after the second GnRH pulse. This was correlated with increased mRNA concentrations measured by means of Northern blots of total RNA. At 0.02 fimol/L GnRH, 4-fold higher /3 mRNA levels were observed. The a mRNA levels were 2.5-fold elevated. GnRH pulses of 0.01 and 10 jimol/L, respectively, were ineffective. A further effect of GnRH pulses was augmentation of the episodic character of hCG secretion. Our results suggest that GnRH causes different specific acute and late effects on the amount and pattern of hCG secretion as well as on hCG biosynthesis at the levels of both hCG subunit mRNAs. (J Clin Endocrinol Metab 73: 84-92, 1991)

I

T IS GENERALLY accepted that gonadoliberin (GnRH) plays a central role in regulation of the biosynthesis and secretion of hypophyseal gonadotropins (1, 2). There is also growing evidence for an extrapituitary function of GnRH in the gonads and placenta (3). Considering the action of GnRH as a releasing hormone for gonadotropins, a homology between the hypothalamic-pituitary-endocrine axis and the placenta seems to exist. The first trimester placenta is composed of two neighboring cell layers, the cytotrophoblast and the syncytiotrophoblast. The tissue is able to synthesize a LHreleasing factor (4, 5) that is structurally identical to the hypothalamic GnRH (6). Recently, a cDNA for the GnRH precursor was established from placental origin (7). The site of production of GnRH seems to be the

cytotrophoblast (8). It has been suggested that placental GnRH might be involved in the paracrine regulation of the biosynthesis of hCG (9) in the syncytiotrophoblast (10-13) via low affinity GnRH receptors (14, 15). This hypothesis is supported by the observation that the highest GnRH tissue concentrations are present in the first trimester of pregnancy, coinciding with the optimum of hCG synthesis (16, 17). Exogenous GnRH has been applied in several studies to explants of first trimester placenta (9, 18-23) and term placenta (9, 18, 19, 23-26) in order to study the role of GnRH in the regulation of hCG biosynthesis and secretion. In most cases, static tissue culture systems have been used, with daily exchanges of the culture medium, a procedure that does not allow observation of the dynamics of hCG release in response to GnRH in detail. In the present investigations we have used a superfusion system for cultivation of human first trimester placenta to observe the acute and especially the delayed effects of GnRH pulses, which

Received September 7,1990. Address all correspondence and requests for reprints to: Dr. W. E. Merz, Department of Biochemistry II, University of Heidelberg, Im Neuenheimer Feld 328, 6900 Heidelberg, Germany.

84

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

INFLUENCE OF GnRH ON hCG BIOSYNTHESIS have not been extensively studied so far. We provide evidence that GnRH causes an increase in mRNA levels of both hCG subunits in placental tissue. Materials and Methods Source of reagents Tissue culture medium 199 (M199), BSA, human transferrin, bovine insulin (Zn2+ complex), dexamethasone, and Norit-A charcoal were obtained from Serva Feinbiochemica (Heidelberg, Germany). GnRH and reduced forms of nicotinamide adenine dinucleotide were purchased from Boehringer Mannheim (Mannheim, Germany); streptomycin sulfate, penicillin G, newborn calf serum (NCS), and amphotericin-B were obtained from Biochrom KG (Berlin, Germany). All other reagents were of analytical grade. The First International Standard for hCG (Immunoassay) and the Third International Standard were obtained from WHO International Laboratories for Biological Standards (London, United Kingdom). Tissue culture Placental tissue was obtained from abortions performed on social grounds in healthy women (8-12 weeks of gestation) with written consent of the patients. Immediately after the operative treatment the tissue was placed into about 100 mL ice-cold sterile medium. It contained (per L) 11.05 g M199 with Hanks' salts, 0.6 g sodium bicarbonate, 0.11 g penicillin G (sodium salt), and 0.2 g streptomycin sulfate. The pH of the medium was 7.4. The chorion tissue was freed from clotted blood and nonvillous tissue, thoroughly washed, and dissected with sterile scissors into pieces of about 10-30 mg wet weight. All manipulations were carried out under sterile conditions. The tissue was cultured in a medium (M199/NCS) containing (per L) 11.05 g M199 with Hanks' salts, 1.5 g sodium bicarbonate, 0.11 g penicillin G (sodium salt), 0.2 g streptomycin, 1 mg insulin, 0.39 mg dexamethasone, and NCS to give a final concentration of 10%. The pH was adjusted to 7.4 with 1 N NaOH (before the addition of the proteins). Later, experiments were carried out in serum-free culture medium which contained (per L) 11.05 g M199 with Hanks' salts, 1.5 g sodium bicarbonate, 100 000 IU penicillin-streptomycin, 2.5 mg amphotericin-B, and 10 mg iron-saturated human transferrin. Static incubations Static incubations were performed in 24-well culture plates (16 mm diameter of the wells; Greiner GmbH, Nurtingen, Germany) filled with 1.5 mL/well culture medium M199/NCS or serum-free culture medium, as stated below. Each well contained whole chorion tissue pieces of about 10-30 mg wet weight. The tissue was cultivated at 37 C in 90% relative humidity under an atmosphere of 95% air and 5% carbon dioxide (Assab T305GF Incubator, Assab Medicintechnik, Hannover, Germany). The tissue culture medium was changed every 24 h.

85

Superfusion culture The superfusion apparatus used is depicted in Fig. 1. The culture medium circulated permanently by means of a peristaltic pump (Multiperpex, LKB, Bromma, Sweden; 1.1 L/h) in a closed system between a reservoir (2.5-L content) and a glass vessel in which the culture medium was saturated with a mixture of 95% air and 5% carbon dioxide (passed through a sterile 0.22-^m Millex FG50 filter; Millipore Corp. Dreieich, Germany). Both vessels were equipped with a jacket circulated with water of 37 C. The gas-saturated culture medium was distributed into different streams by capillary glass tees. By means of peristaltic micrometering pumps (model PLG, Desaga, Heidelberg, Germany) GnRH (0.01-10 ^mol/L) was added to the culture medium (0.3 mL/h-culture at intervals given below). Mixing (magnetic stirring, mixer driver 11300, LKB) was performed in glass tubes (1.5-mL volume; closed with Teflon plugs) with 2 inlets and 5 outlets. The incubation chambers (volume of 0.5 mL) were made from 2 screwed together Teflon halves sealed by a silicon ring and were located in thermostatized (37 C) aluminium blocks (11 x 10 X 3 cm). One incubation chamber contained 10-30 mg placental tissue (wet weight) placed on a polyester net (100-/um mesh size; Swiss Silk Bolting Cloth Mfg. Co. Ltd., Zurich, Switzerland). In most cases, 16 cultures were run simultaneously, arranged into groups of 4-5 treated cultures and control cultures, respectively. The tissue was flushed with a constant flow of 3 mL medium/ h-culture, maintained by a 16-channel pump (IPS-16, Ismatec Corp., Zurich, Switzerland). At this flow rate of the culture medium there was no difference in the oxygen content between influent and effluent media. Four or 5 tissue pieces were generally used as parallel cultures per condition unless otherwise stated. The effluent medium was collected in a cold box (2 C) equipped with a fraction collector (Colora, Lorch, Germany). Before use, all pieces of the apparatus in contact with the tissue or the culture medium were autoclaved (30 min at 125 C) and sterilized at 200 C for 6 h, respectively. Measurements of hCG, the free hCG subunits, and lactate dehydrogenase in the culture medium hCG and the free subunits were measured by RIAs. Purified hCG (13,000 IU/mg) and the isolated subunits were obtained as described previously (27). [125I]hCG and 125I-labeled isolated subunits were prepared by the chloramine-T method, as described by Markkanen et al. (28) (the molar ratio of purified proteins vs. radioactive label was 1-1.4; 15- to 17-fold molar excess of chloramine-T based on the amount of radioiodide). The following mean specific radioactivities were achieved: hCG, 52 kBq/pmol; /3-subunit, 33 kBq/pmol; /3-subunit, 22 kBq/pmol (Hilf, G., and W. E. Merz, unpublished). The RIA was performed in microvessels (no. 731055, Sarstedt, Numbrecht, Germany) using Dextran T70-coated Norit-A charcoal (1:10, wt/ wt) for the separation of free and antibody-bound hormone according to the method of Herbert et al. (29). Incubation with the antibody was carried out in a humidified closed chamber at 37 C for 2 h. Polyclonal goat antisera against purified hCG and the free subunits showed the following cross-reactivities: in the anti-hCG system, relative to purified hCG, 1.6 x 10'2 with the purified hCG a-subunit (wt/wt), 3.0 X 10"3 with the purified

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

MERZ ET AL.

86

JCE & M • 1991 Vol73«Nol

FIG. 1. Schematic view of the superfusion system. The tissue culture medium was pumped from a double walled flask (1) by means of a peristaltic pump (2; 1.1 L/h) into an artificial lung (3), and the medium was saturated with a sterile mixture of 95% air-5% CO2 (4) and allowed to flow back into the medium tank. By means of a series of glass distributors (5), the culture medium was divided into different flows according to the experimental protocol. GnRH [placed in double walled tanks (6) chilled to 2 C by a thermostat] was added to the culture medium by micrometering pumps (7), which were connected to an electronic control unit (8). Culture medium and GnRH were mixed in mixing chambers (9; 1.5-mL volume; magnetic stirring) sealed with pierced Teflon plugs to give 2 inlets and 5 outlets. Four or 5 incubation chambers (10; 0.5-mL volume) containing the tissue (10-30 mg wet weight located on a lOO-jim net) were connected to 1 mixing unit. The incubation chambers were placed in a thermoblock. The effluent culture medium was pumped (3 mL/h) by means of a 16-channel pump (11) and collected by a fraction collector (12) placed in a cold box (2 C). The medium tank, the "lung," and the thermo-block containing the incubation chambers were kept at 37 C (13) by means of a thermostat.

hCG /3-subunit (wt/wt); in the anti-a-subunit RIA, relative to the purified hCG a-subunit, 2.0 X 10'3 with hCG (wt/wt) and the purified /3-subunit (wt/wt), respectively. Determinations were carried out in triplicate. The intraassay variabilities for the hCG and hCGa RIA were 5.9% and 8.5%, respectively; the interassay variabilities were 14.1% and 16.0%, respectively. Initially the First and later the Third International Standard for hCG were used as reference. The content of lactate dehydrogenase released from the tissue into the culture medium was determined according to the method of Bergmayer (30). RNA preparation and Northern blots Total RNA was prepared from four pooled placental tissue pieces at the end of the superfusion culture. The tissue was homogenized (20 s; Ultra-Turrax, Janke and Kunkel, Staufen/ Breisgau, Germany) in a buffer, pH 7.0, containing 4 mol/L guanidinium isothiocyanate, 25 mmol/L sodium citrate, 0.1 mol/L 2-mercaptoethanol, and 0.5% iV-lauroyl sarcosine. RNA was separated on prepacked Qiagen anion exchange columns according to the protocol of the manufacturer (The Qiagenologist, Diagen GmBH, Dusseldorf, Germany). Thereafter, RNA was denaturated and blotted on GeneScreen (DuPont-NEN,

Dreieich, Germany) according to the method of Sernia et al. (31). Hybridization with 32P-labeled cDNAs [hCGa (32), hCG/? (33), and 0-actin (rat), respectively] was performed according to the method of Church and Gilbert (34). Labeling of the cDNAs was carried out by means of a random primed DNA labeling kit (Boehringer Mannheim) using [a-32P]dCTP (3000 Ci/mmol; Amersham Buchler, Braunschweig, Germany). The typical specific radioactivity of labeled cDNA was 5-10 X 10'8 cpm/^g. Radioactivity was detected by autoradiography (18 h; -80 C; Kodak XAR-5 intensifying screen, Eastman Kodak, Rochester, NY). Quantitative evaluation was performed by laser densitometry (2202 Ultroscan, LKB) of the x-ray films (three scans each). The scanning data were collected and integrated by means of an Apple He computer equipped with the Adalab interface card using the Chromatochard software (Interactive Microware, Inc., State College, PA). Statistical analysis Analysis of variance, calculation of the standard curve using spline approximation, and calculation of the medium concentrations of hCG and the free subunits, as determined in the RIAs, were performed with our own software (W. Haag, H. J.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

INFLUENCE OF GnRH ON hCG BIOSYNTHESIS Scheuermann, and W. E. Merz, unpublished) on an HP 9925 A computer and printed by means of an HP 9871 A printer. The statistical significance of changes in the hCG medium concentrations in response to single GnRH pulses in different experiments was determined by means of a rank test for complete block designs, as described by Haux et al. (35). Episodic hCG secretion was kindly analyzed by Dr. T. 0. F. Wagner (Medizinische Hochschule, Hannover, Germany) by means of the modified algorithm described by Santen and Bardin (36), the Pulsar algorithm according to Merriam and Wachter (37), and the Cycle Detector program of Clifton and Steiner (38).

87

2000

1500

oi 1000

o 500

Results

hCG secretion in superfusion and static incubations hCG secretion was significantly influenced by the culture conditions, as evidenced by an approximately 4-fold higher secretion rate in superfusion cultures compared with that in static incubations of the same placentae (Table 1). In the superfusion system used (Fig. 1) a typical profile of basal hCG secretion was observed (Fig. 2, top panel) in four placentae of different gestational age which were studied in the same experiment. In the first hours of the culture, hCG secretion was decreased 2- to 3-fold. This diminished level remained constant for about 15 h. Thereafter, the concentration of immunoreactive hCG in the medium was continuously increased (placenta tissues of 8 and 10 weeks gestation) and reached values similar to those measured at the beginning of the experiment. The tissue from the 12th week of gestation showed low hCG secretion from the beginning, which was maintained throughout. The hCG secretion rate reached toward the end of the superfusion experiment was not essentially changed for at least 2 further days (not shown). During the entire incubation time the medium content of lactate dehydrogenase remained constant, indicating that the increased concentrations of hCG and free subunits, respectively, were not caused by the destruction of tissue. The experiments were generally stopped 3 and in some cases 5 days after TABLE 1. Comparison of hCG secretion in superfusion cultures and static incubations of human first trimester placental tissue

Exp Week of no. gestation

1 2 3 4 0

8 9 9 8

hCG secretion rate on day 3 [IU hCG/10 mg tissue (wet weight) • 24 h]° Superfusion n

Static incubations

234.0 ± 92.9 ± 114.2 ± 58.5 ±

58.5 ± 23.2 ± 31.6 ± 13.0 ±

59.4 23.2 14.9 36.2

5 5 4 5

19.5 10 14.9 6 10.2 8 2.8 6

Ratio superfusion/static

4.0 4.0 3.6 4.5

hCG was determined by RIA, using the First International Standard for hCG (Immunoassay) as reference. Values are the mean ± SD. n, Number of cultures in parallel.

FIG. 2. Time course of hCG secretion in superfusion cultures by placental tissue of different gestational age {upper panel). In the lower panel the molar ratio of secreted hCG/a-subunit is depicted. O and • , Eighth week of gestation; A 10th week of gestation; A, 12th week of gestation. Culture medium (M199/NCS) contained 10% NCS. Similar results were obtained in 65 other separate superfusion experiments with culture medium containing NCS as well as in serum-free cultures.

the tissue had been explanted. Coincidently with the decrease in hCG secretion in the first 20 h of superfusion culture, the secretion of the free a-subunit (as well as of the free /?-subunit; not shown) was increased, so that the molar ratio of hCG/free a-subunit was diminished (Fig. 2, bottom panel). Thereafter, in the second phase, the release of free a-subunit was decreased in accordance with recovery of the initial hCG secretion rate, yielding an enlarged hCG/free a-subunit ratio. The molar ratio of hCG/free a-subunit measured at the beginning was reached again or even exceeded by the end of the superfusion culture. GnRH-induced effects on hCG secretion The tissue responded to a single short application of GnRH with an immediate and transient increase in hCG secretion (Fig. 3). The placental tissues of different donors and even tissue pieces of the same placenta showed large quantitative differences in hCG release. However, the quality of the GnRH-induced response was the same in all cultures. A statistical analysis of experiments performed with placentae from different donors by means of a rank test (35) showed that in cultures stimulated with a GnRH pulse, the peak maximum hCG values were significantly increased compared to those in the controls ( P < 0.0001). Two GnRH pulses (the first applied 24 h after the start of the cultures, the second between 36 and 48 h, as indicated below) induced a long-lasting stimulation of the secretion of immunoreactive hCG (Figs. 4 and 5).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

MERZ ET AL.

88

6

1250

FlG. 3. Acute increase in hCG secretion in response to a single short application of GnRH (1 jtmol/L; 30 min; • ; cultures 6-10) of individual tissue pieces of the same placenta (eighth week of gestation) in superfusion culture. The GnRH pulse was applied 24 h after the start of the culture. The control cultures (1-5) were kept under the same conditions, except the GnRH infusion was omitted. Fractions of 8 min were collected. Culture medium (M199/NCS) containing 10% NCS was used. The hCG secretion of these cultures over a longer period is shown in Fig. 5. Similar results were obtained in five other separate superfusion experiments with placentae from other donors.

J C E & M • 1991 Vol 73 • No 1

GnRH

10

1000

/\/ vA

750 500 250

1 2

V

Control

750

500 250

24.5

25.0

25.5 24.5

FlG. 4. hCG secretion pattern of individual placental pieces in response to two GnRH pulses (1 /miol/L; 30 min each) applied 24 and 36 h after the start of the superfusion culture. The response of the tissue in the interval 39-46 h of culture is depicted. The control cultures (14) as well as the GnRH-treated cultures were obtained from the same placenta at the eighth week of gestation. The late effects of GnRH (310 h after the last GnRH pulse) on episodic hCG secretion are shown. Serum-free culture medium was used. Fractions of 30 min were collected. Similar results were obtained in four other experiments.

25.0

24.5 25.5

25.0

25.5 24.5

25.0

24.5 25.5

25.0 25.5 HOURS

GnRH concentrations in the range of 0.01-10 jumol/L caused a broad optimum curve of tissue response. The highest hCG secretion rates (2-fold increase compared to control cultures) were observed between 0.02-0.1 /umol/ L in different experiments. Concentrations of 0.01 and 10 jiimol/L elicited small elevations, however, which were not significantly different from the control values (not shown). Estimation of the mRNA levels for the hCG subunits by means of Northern blot analysis showed that GnRH caused an increase in the a as well as 0 mRNA levels (Fig. 6). The GnRH-induced changes in the mRNA concentrations seemed to be specific, since /3-actin mRNA was not increased. The optimum GnRH action on the hCG subunit mRNAs was observed at 0.02 /imol/ L exogenous GnRH for both subunits. This was in accordance with the GnRH effect on the secretion of immunoreactive hCG. A further phenomenon seemed to be an augmentation of episodic secretion of hCG in response to the application of two exogenous GnRH pulses (Figs. 4 and 5), especially after the second one. The untreated control cultures showed episodic hCG secretion also. However, the episodic pattern of hCG secretion occurred not as frequently and showed a smaller amplitude than in the case of the GnRH-treated cultures. Continuous application of GnRH (1 /zmol/L) in superfusion cultures caused hCG secretion that was not significantly different from or was even less than that in control cultures (not shown).

Discussion The superfusion system used here seems to provide a substantial improvement of the culture conditions, which

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

INFLUENCE OF GnRH ON hCG BIOSYNTHESIS

89

FIG. 5. Stimulation of episodic secretion of hCG by GnRH. A placenta of 8 weeks gestation was used. Two GnRH pulses (1 /imol/L; 30 min) were applied to cultures 6-10, 24 and 48 h (indicated by arrows) after the beginning of the superfusion experiment. Cultures 1-5 served as controls. During the first GnRH pulse, 8-min fractions were collected in order to follow the acute response of the tissue. These values are omitted in the graph and are depicted in Fig. 3 separately. In the other parts of the experiment, fractions of 90 min were collected. The culture medium (M199/NCS) contained 10% NCS. The mean ± SD of triplicate values are given throughout.

72

is evident from the higher hCG production compared to that in static incubations. In the superfusion system, 48 times more medium is provided per culture (1.5 mL/24 h in static incubations vs. 72 mL/24 h in superfusion cultures). This probably means a better supply of nutrients and a more effective removal of metabolic products than is the case for the static incubations. At present it is not clear whether these better culture conditions are alone responsible for the higher hCG production or if the abolishment of a possible autocrine inhibition of hCG biosynthesis also plays a role. In the superfusion system hCG secretion was generally decreased in the first hours, followed by an interval in which the diminished secretion was maintained, and finally by a recovery of hCG biosynthesis and secretion. This seems to be evident from the observation that the incorporation of radioactive amino acids into secreted hCG was increased (not shown) in the phase when the release of free subunits was diminished to the benefit of an increased hCG secretion. Furthermore, in vitro translation, as described by Eisenstein and Harper (39) [adapted to placental tissue (S. Uiker and W. E. Merz, unpublished)] was not effective when performed in the first 24 h after the tissue had been explanted; however, it could be successfully carried out thereafter, providing further support for the assumption that hCG biosynthesis had recovered in the course of the superfusion culture. Based on these results we investigated the effect

0 Hours

12

72

of GnRH on hCG biosynthesis and secretion. The rationale for these studies was the hypothesis of a functional analogy between the central nervous system neuroendocrine axis and the placenta. This had been suggested for the stimulating action of GnRH on hCG biosynthesis (9-14), for inhibition of the human choriosomatomammotropin synthesis by somatostatin (40), and for the positive effect of corticoliberin on human chorionic corticotropin synthesis (41). In the pituitary, two GnRH receptor types have been described, a low affinity, high capacity receptor (Ka = 106 L/mol) and a high affinity receptor (Ka = 0.7 X 109 L/mol) of low capacity (42). The high affinity receptor was supposed to be of physiological significance considering the low GnRH concentrations in the pituitary stalk plasma (~0.1 nmol/L) (42). In the placenta only a single class of GnRH receptors of the low affinity type (Ka = 1-3 x 106 L/mol) has been described (14,15, 26). Application of a single pulse of exogenous GnRH caused an immediate and transient increase in hCG secretion. A GnRH pulse increased hCG secretion between 1.5-4 times basal levels. In the pituitary, gonadotropin secretion may be maximally stimulated about 3to 10-fold by a GnRH pulse of some hundred picomoles, in vivo as well as in vitro. In the case of the placenta, roughly a 10- to 20-fold greater GnRH concentration is necessary. The smaller stimulatory capacity of exogenous GnRH in the placenta, as also observed by others (18,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

MERZ ET AL.

90 GnRH ((jmol/L)

0

10

1

0.1

0.05 0.02 0.01

B hCG-/5 C Actin

r 3

rh

10

1

0.1

0.05

0.02

0.01

GnRH (pmol/L)

FIG. 6. Specific stimulation by GnRH of hCG mRNA content in placental tissue (ninth gestational week). Two GnRH pulses of 30 min have been applied in superfusion culture (serum-free medium) 24 and 43 h after the start of the culture. Preparation of total mRNA was performed 8 h after the second GnRH pulse. The Northern blot was sequentially hybridized with 32P-labeled hCGa cDNA probe (A), hCG/3 cDNA probe (B), and rat /3-actin cDNA probe (C). The autoradiographs were quantitated by densitometry. The fold increase in a and /? mRNAs relative to /3-actin mRNA in correlation with the GnRH dosage applied is shown in D (D, a mRNA/j8-actin mRNA; D, 0 mRNA//3-actin mRNA).

20, 22, 43), might be caused by the presence of endogenous GnRH concentrations, sufficient to allow optimal hCG release, so that the addition of exogenous GnRH might have only a small effect. The dynamics of hCG secretion in response to a GnRH pulse indicate that at least part of the hCG is released from a rapidly available pool. A classical secretory pathway was recently described in first trimester placenta (12), possibly existing in addition to the constitutive way of hCG secretion (44). Exogenous GnRH also exerted a long-lasting stimulation of hCG biosynthesis and secretion (~2 times basal

Vol73-Nol

levels) starting several hours after the two GnRH pulses had been applied. The effects of GnRH on the secretion of immunoreactive hCG as well as the incorporation of radioactive amino acids into hCG have been already described (18, 19, 21, 23, 42-44); however, the mode of GnRH action is still unknown. Here, we describe for the first time that GnRH seems to cause a specific increase in the concentrations of both subunit mRNAs, with an optimum at 0.02 yitmol/L. There is no consistent model at present concerning the GnRH effect on mRNA content of the pituitary gonadotropin subunits. In hypogonadal mice the mRNA levels of the common a-subunit, but not of the LH/? or TSH/3 mRNAs (45), were increased. GnRH-stimulated elevation of LH/? release was not coupled with a change in LH0 gene transcription or cytoplasmic mRNA levels (46) in isolated rat pituitary cells. Short term treatment (4-6 h) of rat pituitary cells with GnRH in vitro increased LH release significantly without any effect on the a- or LH/3 mRNA concentrations, whereas longer incubation with GnRH (24-72 h) enlarged the cellular a-mRNA pool, but not the LH/3 mRNA content (47). Pulsatile application of GnRH to castrate testosterone-replaced male rats raised a as well as LHj8 mRNA levels depending on both the GnRH dose and the pulse interval (48). Others have reported an elevation merely of a mRNA, independent of the mode of stimulation (pulsatile vs. continuous), whereas LH/3 mRNA was unresponsive (49). Our results suggest that the GnRH-induced changes in the gene expression of both gonadotropin subunits in the placenta differ from the regulation in the pituitary. Application of two exogenous GnRH pulses enhanced an episodic type of hCG secretion, especially after the second pulse. Episodic hormone secretion may be analyzed by means of different algorithms [recently reviewed by Urban et al. (50)]. Calculations of our data using the modified Santen and Bardin program, the Cycle Detector, and the Pulsar programs supported the assumption of an episodic hCG secretion pattern in vitro (not shown). A detailed analysis of pulse amplitude and frequency, however, was not the objective of the present study. A GnRH-induced episodic hCG secretion of a completely different type has been reported recently by Barnea and Kaplan (22), who observed a secretory pattern of high frequency occurring immediately after application of a short (1-min) GnRH stimulus. In our case the episodic secretion showed a distinctly lower number of cycles and occurred spontaneously in GnRH-treated cultures many hours after the GnRH stimulus had been applied. This seems to suggest that different episodic secretion patterns may be expressed by first trimester placental tissue. A pulsatile pattern of the hCG plasma concentration in pregnant women has been described by Owens et al. (51). Even in normal nonpregnant women and in men episodic secre-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

INFLUENCE OF GnRH ON hCG BIOSYNTHESIS tion of immunoreactive hCG has been described (52, 53). The hypothesis that GnRH acts on placental gonadotropin synthesis is supported by the findings of Das et al. (54), who observed that in pregnant baboons, which synthesize a CG (bCG) also, bCG as well as progesterone levels were sharply decreased when the animals were given anti-GnRH monoclonal antibodies or injections of a GnRH superagonist, which both resulted in termination of the pregnancies. The pregnancies were not abrogated when the treatment was performed at a time before the appearance of bCG in the circulation. Our results provide further evidence that GnRH regulates hCG biosynthesis and secretion. GnRH seems to exert acute effects on hCG secretion as well as late effects on hCG biosynthesis at the level hCG subunit mRNAs. The significance of GnRH in the control of hCG biosynthesis in concert with other factors that have effects on hCG biosynthesis has to be elucidated in further investigations.

9.

10. 11.

12. 13. 14. 15.

16.

Acknowledgments The authors are greatly indebted to E. Mohr for excellent technical assistance, to Dr. J. C. Fiddes (California Biotechnology, Mountain View, CA) for providing the hCG subunit cDNA clones, and to Dr. D. Werner (German Cancer Research Center, Heidelberg, Germany) for the actin cDNA. B. Safferling performed excellent work in manufacturing the glass vessels for the superfusion apparatus, which is thankfully acknowledged, as is the support of Dr. J. Rafael (Department of Biochemistry I, University of Heidelberg) in connection with the O2 measurements. The authors want to thank Dr. T. 0. F. Wagner (Medizinische Hochschule, Hannover, Germany) for the computerized calculations of the episodic secretion of hCG, and Dr. A. Rosenberg (Psychatric and Biobehavioral Department, University of California-Los Angeles School of Medicine) for critical reading of the manuscript.

17. 18. 19.

20. 21.

22.

References 1. Hazum E, Conn M. Molecular mechanism of gonadotropin releasing hormone (GnRH) action. I. The GnRH receptor. Endocr Rev. 1988;9:379-86. 2. Huckle WR, Conn M. Molecular mechanism of gonadotropin releasing hormone action. II. The effector system. Endocr Rev. 1988;9:387-95. 3. Hsueh AJ, Jones PBC. Gonadotropin releasing hormone: extrapitultary actions and paracrine control. Annu Rev Physiol. 1983;45:83-94. 4. Gibbons JM, Mitnik M, Chieffo V. In vitro biosynthesis of TSHand LH-releasing factors by human placenta. Am J Obstet Gynecol. 1975;121:127-31. 5. Khodr GS, Siler-Khodr TM. Placental luteinizing hormone-releasing factor and its synthesis. Science. 1980;207:315-7. 6. Tan L, Rousseau P. The chemical identity of the immunoreactive LHRH-like peptide biosynthesized in the placenta. Biochem Biophys Res Commun. 1982;109:1061-71. 7. Seeburg PH, Adelman JP. Characterization of cDNA for precursor of human luteinizing hormone releasing hormone. Nature. 1984;311:666-8. 8. Khodr GS, Siler-Khodr TM. Localization of luteinizing hormone-

23.

24.

25.

26.

27.

91

releasing factors in the human placenta. Fertil Steril. 1978;29:5236. Siler-Khodr TM, Khodr GS, Valenzuela G, Rhode J. Gonadotropin-releasing hormone effects on placental hormones during gestation. I. Alpha-human chorionic gonadotropin, human chorionic gonadotropin and human chorionic somatomammotropin. Biol Reprod. 1986;34:245-54. Midgley AR, Pierce GB. Immunohistochemical localization of human chorionic gonadotropin. J Exp Med. 1962;115:289-94. Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss III JF. Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology. 1986;118:1567-82. Morrish DW, Marusyk H, Siy 0. Demonstration of specific secretory granules for human chorionic gonadotropin in placenta. J Histochem Cytochem. 1987;35:93-101. Wide M, Perrson H, Lundkvist 0, Wide L. Localization of mRNA for the 0-subunit of placental hCG by in situ hybridization. Acta Endocrinol (Copenh). 1988;119:69-74. Iwashita M, Evans MI, Catt KJ. Characterization of a gonadotropin-releasing hormone receptor site in term placenta and chorionic villi. J Clin Endocrinol Metab. 1986;62:127-33. Escher E, Mackiewicz Z, Lagace G, et al. Human placental LHRH receptor: agonist and antagonist labeling produces differences in the size of the non-denaturated, solubilized receptor. J Receptor Res. 1988;8:39l-405. Siler-Khodr TM, Khodr GS. Content of luteinizing hormonereleasing factor in the human placenta. Am J Obstet Gynecol. 1978;130:216-9. Siler-Khodr TM, Khodr GS, Valenzuela G. Immunoreactive gonadotropin-releasing hormone level in maternal circulation throughout pregnancy. Am J Obstet Gynecol. 1984;150:376-9. Takemori M, Nishimura R, Ashitaka Y, Tojo S. Release of human chorionic gonadotropin (hCG) and its alpha-subunit (hCG-a) from perifused human placenta. Endocrinol Jpn. 1981;28:757-68. Haning Jr RV, Choi L, Kiggens AJ, Kuzma DL. Effects of prostaglandins, dibutyryl cAMP, LHRH, estrogens, progesterone, and potassium on output of prostaglandin F2a, 13,14-dihydro-15-ketoprostaglandin F2a, hCG, estradiol, and progesterone by placental minces. Prostaglandins. 1982;24:495-506. Biitzow R. Luteinizing hormone-releasing factor increases release of human chorionic gonadotrophin in isolated cell columns of normal and malignant trophoblasts. Int J Cancer. 1982;29:9-11. Mathialagan N, Jagannadha Rao A. Gonadotropin releasing hormone (GnRH) stimulates both secretion and synthesis of human chorionic gonadotropin by first trimester placental minces in vitro. Biochem Int. 1986;13:757-65. Barnea ER, Kaplan M. Spontaneous, gonadotropin-releasing hormone-induced, and progesterone-inhibited pulsatile secretion of human chorionic gonadotropin in the first trimester placenta in vitro. J Clin Endocrinol Metab. 1989;69:215-7. Merz WE, Hilf G. Influence of LHRH and des-Glylo[DD-Ala6]LHRH ethylamide on the biosynthesis of hCG in placental tissue culture. In: Schmidt-Gollwitzer M, ed. New developments in biosciences. Berlin: de Gruyter; 1985;l:69-79. Haning Jr RV, Choi L, Kiggens AJ, Kuzma DL, Summerville JW. Effects of dibutyryl adenosine 3',5'-monophosphate, luteinizing hormone-releasing hormone, and aromatase inhibitor on simultaneous outputs of progesterone, 17/3-estradiol, and human chorionic gonadotropin by term placental explants. J Clin Endocrinol Metab. 1982;55:213-8. Petraglia F, Vaughan J, Vale W. Inhibin and activin modulate the release of gonadotropin-releasing hormone, human chorionic gonadotropin, and progesterone from cultured human placental cells. Proc Natl Acad Sci USA. 1989;86:5114-7. Belisle S, Guevin J-F, Bellabarba D, Lehoux J-G. Luteinizing hormone-releasing hormone binds to enriched human placental membranes and stimulates in vitro the synthesis of bioactive human chorionic gonadotropin. J Clin Endocrinol Metab. 1984;59:119-268. Merz WE, Hilgenfeldt U, Dorner M, Brossmer R. Biological, immunological and physical investigations on human chorionic go-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.

MERZ ET AL.

92

nadotropin. Hoppe Seylers Z Physiol Chem. 1974;355:1035-45. 28. Markkanen S, Tollikko K, Jaaskelainen K, Rajaniemi H. Effect of radioiodination on the structural integrity, receptor and antibodybinding activity and circulatory behavior of human choriogonadotropin. Horm Res. 1980;12:32-45. 29. Herbert V, Lau KS, Gorrlieb CW, Bleicher SJ. Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab. 1965;25:137584. 30. Bergmayer HU. L(+)-Lactatdehydrogenase aus Kaninchenmuskel. In: Bergmayer HU, Gawehn K, eds. Methods of enzymatic analysis, 3rd ed. Weinheim: VCH Verlagsgesellschaft; 1983;l:512-3. 31. Sernia C, Clements JA, Funder JW. Regulation of liver angiotensinogen mRNA by glucocorticoids and thyroxine. Mol Cell Endocrinol. 1989;61:147-56. 32. Fiddes JC, Goodman HM. Isolation, cloning and sequence analysis of the cDNA for the a-subunit of human chorionic gonadotropin. Nature. 1979;281:351-6. 33. Fiddes JC, Goodman HM. The cDNA for the /3-subunit of human chorionic gonadotropin suggests evolution of a gene by readthrough into the untranslated region. Nature. 1980;286:684-7. 34. Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci USA. 1984;81:1991-5.

35. Haux R, Schumacher M, Weckesser G. Rank tests for complete block designs. Biomath J. 1984;5:567-82. 36. Santen RJ, Bardin CW. Episodic luteinizing hormone secretion in man. Pulse analysis, clinical interpretation, physiologic mechanisms. J Clin Invest. 1973;52:2617-28. 37. Merriam GR, Wachter KW. Algorithms for the study of episodic hormone secretion. Am J Physiol. 1982;243:E310-8. 38. Clifton DK, Steiner RA. Cycle detection: a technique for estimating the frequency and amplitude of episodic fluctuations in blood hormone and substrate concentrations. Endocrinology. 1983;112:1057-64. 39. Eisenstein RS, Harper AE. Characterization of a protein synthesis system from rat liver. Translation of endogenous and exogenous messenger RNA. J Biol Chem. 1984;259:9922-8. 40. Watkins WB, Yen SSC. Somatostatin in cytotrophoblast of the immature human placenta: localization by immunoperoxidase cytochemistry. J Clin Endocrinol Metab. 1980;50:969-71. 41. Sasaki A, Liotto AS, Luckey MM, Margioris AN, Suda T, Krieger DT. Immunoreactive corticotropin-releasing factor is present in human maternal plasma during the third trimester of pregnancy. J Clin Endocrinol Metab. 1984;59:812-4. 42. Clayton RN, Catt KJ. Gonadotropin-releasing hormone receptors:

43.

44. 45.

46.

47.

48.

49.

50.

51. 52. 53. 54.

JCE & M • 1991 Vol 73 • No 1

characterization, physiological regulation, and relationship to reproductive function. Endocr Rev. 1981;2:186-209. Ashitaka Y, Nishimura R, Takemori M, Tojo S. Production and secretion of hCG and hCG subunits by trophoblastic tissue. In: Segal SJ, ed. Chorionic gonadotropin. New York: Plenum Press; 1980; 147-75. Hussa RO. Biosynthesis of human chorionic gonadotropin. Endocr Rev. 1980;l:268-94. Stanley HF, Lyons V, Obonsawin MC, et al. Regulation of pituitary a-subunit, /3-luteinizing hormone and prolactin messenger ribonucleic acid by gonadotropin-releasing hormone and estradiol in hypogonadal mice. Mol Endocrinol. 1988;2:1302-9. Salton SRJ, Blum M, Jonassen JA, Clayton RN, Roberts JL. Stimulation of pituitary luteinizing hormone secretion by gonadotropin-releasing hormone is not coupled to /3-luteinizing hormone gene transcription. Mol Endocrinol. 1988;2:1033-42. Hubert JF, Simard J, Gagne B, Barden N, Labrie F. Effect of luteinizing hormone releasing hormone (LHRH) and [D-Trp6,DesGly-NH210]LHRH ethylamide on a-subunit and LH/? messenger ribonucleic acid levels in anterior pituitary cells in culture. Mol Endocrinol. 1988;2:521-7. Haisenleder DJ, Katt JA, Ortolano GA, El-Gewely MR, Duncan JA, Marshall JC. Influence of gonadotropin-releasing hormone pulse amplitude, frequency, and treatment duration on the regulation of luteinizing hormone (LH) subunit messenger ribonucleic acids and LH secretion. Mol Endocrinol. 1988;2:338-43. Weiss J, Jameson JL, Burrin JM, Crowley Jr WF. Divergent responses of gonadotropin subunit messenger RNAs to continuous versus pulsatile gonadotropin-releasing hormone in vitro. Mol Endocrinol. 1990;4:557-64. Urban RJ, Evans WS, Rogol AD, Kaiser DL, Johnson ML, Veldhuis JD. Contemporary aspects of descrete peak-detection algorithms. I. The paradigm of the luteinizing hormone pulse signal in men. Endocr Rev. 1988;9:3-37. Owens OM, Ryan KJ, Tulchinsky D. Episodic secretion of human chorionic gonadotropin in early pregnancy. J Clin Endocrinol Metab. 1981;53:1307-9. Odell WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med. 1987;317:1688-91. Odell WD, Griffin J. Pulsatile secretion of chorionic gonadotropin during the normal menstrual cycle. J Clin Endocrinol Metab. 1989;69:528-32. Das C, Gupta SK, Talwar GP. Pregnancy interfering action of LHRH and anti-LHRH. J Steroid Biochem. 1985;23:803-6.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 12:01 For personal use only. No other uses without permission. . All rights reserved.