Follicular Atresia and Infertility in Rats Treated with a Gonadotropin ...

0013-7227/90/1271-0025$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society

Vol. 127, No. 1 Printed in U.S.A.

Follicular Atresia and Infertility in Rats Treated with a Gonadotropin-Releasing Hormone Antagonist* KATHY L. SHARPE, MARIA C. BERTERO, BENJAMIN P. LYON, KEN N. MUSE, AND MICHAEL W. VERNONt Department of Obstetrics and Gynecology, University of Kentucky, Lexington, Kentucky 40536

was assessed by mating vehicle-and Antide-treated rats at spontaneous proestrus (n = 8) for eight posttreatment cycles as well as after follicular stimulation (n = 2). All vehicle-treated and no Antide-treated rats became pregnant. No oocytes were found in the oviducts of Antide-treated rats after eight cycles, indicating that ovulation had not occurred. The serum FSH and estradiol concentrations in the rats treated with Antide were lower (P < 0.05) on days 6, 12, and 18, but rose to values equal to (days 24 and 30) or greater than (day 42) those in vehicle-treated rats. Serum progesterone levels in rats treated with Antide were lower (P < 0.05) than those in vehicle-treated rats on all days tested. In conclusion, at a dosage sufficient to suppress reproductive cyclicity (and elicit the regression of endometriosis), Antide also caused long term follicular atresia and infertility. {Endocrinology 127: 25-31,1990)

ABSTRACT. Adverse effects of the GnRH antagonist Antide on folliculogenesis and fertility were noted when we were evaluating the therapeutic value of Antide on endometriosis in a rat model. Cyclic rats with (n = 56) and without (n = 18) surgically induced endometriosis received Antide (2 mg/kg) or vehicle at noon on days 0 (proestrus), 3, 6, and 9. Rats were killed at noon on days 0, 6, 12, 18, 24, 30, 42, and 165. The number of antral follicles and the number of atretic antral follicles evaluated did not differ (P > 0.05) between endometriosis and control rats. Antide-treated rats had more (P < 0.05) atretic antral follicles (64.7%) than vehicle-treated rats (15.3%). Antide-treated rat ovaries contained fewer corpora lutea than those of vehicletreated rats on days 6, 12, and 18. No corpora lutea were found in Antide-treated rat ovaries after day 18. The abnormal ovarian morphology of the Antide-treated rats persisted for the duration of the project (165 days). Fertility (rats without endometriosis)

I

Therefore, we are reporting the occurrence of the adverse effects of Antide on reproduction in the rat.

N THE past decade, substantial efforts have been made in the development of GnRH analogs as both diagnostic and therapeutic agents. The actions of these analogs have been suggested as having a practical value in the management of endometriosis, uterine leiomyomata, uterine bleeding, premenstrual syndrome, and polycystic ovarian disease. Both GnRH agonists and antagonists have been examined as potential therapeutic agents; however, only the GnRH agonists are currently available for human use. Although the early GnRH antagonists caused a favorable suppression of gonadotropin secretion, their use was limited due to histamine-related side-effects (1). Drs. K. Folkers and C. Y. Bowers (2) designed and synthesized a GnRH antagonist with prolonged duration of action without the side-effects. This new third generation antagonist, called Antide, has generated great interest for clinical investigations (2-4). We have recently examined the effectiveness of Antide therapy in the reduction of endometriosis in rats with surgically induced disease (4). However, while examining the effects of Antide in this rat model, we noted adverse effects of Antide on follicular development and fertility.

Materials and Methods Reproductive cyclicity All animals were maintained in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Female Harlan (Madison, WI) Sprague-Dawley rats (250 g; -70-72 days of age) were housed in an environmentally controlled room with a 14-h light, 10-h dark cycle. Throughout the course of the experiments, vaginal cytology was evaluated daily as an indirect index of ovarian activity and for detection of successful mating, as evidenced by the presence of sperm and continual diestrous cytology. Vaginal cytology was scored with a morphological grading system. In this grading system, proestrus (rounded nucleated epithelial cells) and estrus (cornified epithelial cells) smears were scored +1 and +2, respectively. Diestrus day I (cornified epithelial cells and leukocytes) and diestrus day II (predominantly leukocytes with few cornified and rounded epithelial cells) smears were scored - 1 and - 2 , respectively. Therefore, a positive score was indicative of estrus and a negative score of diestrus. Cyclic variation above and below 0 reflected the changes in the vaginal milieu that are representative of ovarian cyclicity (Fig. 1A).

Received February 19,1990. * This work was supported by NIH Grant HD-21962. t To whom requests for reprints should be addressed.

To evaluate Antide-associated effects, we designed two ex25

26

GnRH ANTAGONIST-INDUCED REPRODUCTIVE DYSFUNCTION

Estrus

-1 • -2Diestrus

I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—1—1 Estrus

-2-Diestrus I-

t

I

I

I

I

I

I

I

I

I

i

Days

I

I

I

I

I

150

FIG. 1. Reproductive cyclicity in rats after injections of vehicle (A) or the GnRH antagonist Antide (B) on days 0, 3, 6, and 9 (arrows). A morphological grading system was designed and based on vaginal cytology, where +1 and +2 were indicative of proestrus and estrus stages of the 4-day estrous cycle, respectively, and - 1 and - 2 were indicative diestrus day I and diestrus day II, respectively (see Materials and Methods for further explanation). Each point represents the mean score of four animals.

periments to examine follicular development (Exp 1) and fertility (Exp 2). Exp 1: effects of Antide on follicular development The effects of Antide on follicular development were examined in rats with and without surgically induced endometriosis. Briefly, endometriosis was surgically induced by the autotransplantation of 3-mm squares of uterine tissue derived from the distal third of the ablation of the right uterine horn. These uterine squares (implants) were autotransplanted to the arterial cascades of the intestinal mesentery (5). Sham operations were also performed, where only the uterine horn was removed, and no ectopic uterine implants were created. Six weeks after surgical induction of endometriosis or sham operation (controls), reproductively cyclic female rats received sc injections of Antide (iV-Ac-D-2-Nal1,D-4ClPhe2,D-3-Pal3,

Endo• 1990 Voll27«Nol

Nic-Lys5,D-Nic-Lys6,I-Lys8,D-Ala10; 2 mg/kg in 0.5 ml oil vehicle) or vehicle alone at noon on days 0 (proestrus) 3, 6, and 9. Antide was made available by Dr. Marvin Karten of the Contraceptive Development Branch of the NICHHD. Previous work has shown that a single dose of 2 mg/kg successfully caused long term suppression of reproductive cyclicity in the rat, while lower doses (0.5 and 1.0 mg/kg) were not effective (4). Furthermore, following the four-injection (2 mg/kg each) protocol described above, we have previously shown that Antide caused profound regression of endometriotic implant size by more than 85% in rats with surgically induced disease (4). Previous GnRH antagonists have been noted for their deleterious histamine-related side-effects (1, 2). Therefore, the rats in this study were monitored for signs of edema and other related side-effects after each injection. To examine the effects of Antide on follicular development, rats (endometriosis plus vehicle, n = 4/day; endometriosis plus Antide, n = 4/day; control plus Antide, n = 2/day) were killed at noon on days 0, 6, 12, 18, 24, 30, and 42. Additional rats (control plus vehicle, n = 2; control plus Antide, n = 2) were killed 156 days after cessation of treatment (day 165) to examine the long term effects of Antide on follicular development. Rats were anesthetized, serum was immediately collected via heart puncture for RIA, and ovaries were prepared for morphological evaluation. All ovaries were weighed, fixed, set in paraffin, sectioned (5 /an), and stained with hematoxylin and eosin for morphological evaluation. Slides used for morphometric analysis began at a 400-/zm depth into the ovary and continued at 200-^m intervals so that 3 slides from each ovary (400, 600, and 800 ^m) were examined by light microscopy for the overall number of antral follicles and the number of atretic antral follicles present. Only antral follicles that were greater than 300 /tm in diameter and contained a fluid-filled cavity were evaluated. Therefore, due to the methodology of analysis and the criteria established for antral follicle evaluation, the possibility of counting an antral follicle more than once, although infrequent, was equally distributed between the treatment groups and, therefore, did not affect the morphometric analysis. Sections of antral follicles containing more than 10 pyknotic nuclei and/or degenerating oocytes were considered to be undergoing atresia. Measurement (RIA) of FSH, estradiol (E2), and progesterone (P) was performed in our laboratory using previously validated assays. Antigen, antiserum, and standard (RP-2) for the FSH RIA were provided by Dr. A. F. Parlow of the NIDDK. All serum samples were run within the same assay. The coefficients of variation within the FSH, E2, and P assays were 4.7%, 9.3%, and 6.4% respectively. Exp 2: effects of Antide on Fertility To assess the effects of Antide on fertility, rats without endometriosis received Antide or vehicle (injection protocol as in Exp 1; day 0 = proestrus) and were caged with males at first spontaneous proestrus (n = 4 vehicle-treated rats; n = 4 Antidetreated rats) or on day 14 after follicular stimulation (n = 2 Antide-treated rats; day 12 = 10 IU PMSG; day 14 = 5 IU hCG). Coitus was determined by the presence of sperm in the vagina, followed by continuous diestrous vaginal cytology. If

GnRH ANTAGONIST-INDUCED REPRODUCTIVE DYSFUNCTION pregnancy did not occur after coitus, females were placed with a different male on each subsequent proestrus for a total of eight cycles. After eight nonpregnant cycles, females were killed at estrus, and oviducts were flushed for detection of oocytes. Ovaries were prepared for morphological evaluation as described above. Statistical analysis Statistical analysis was performed using Statistical Analysis System's general linear models computer program (SAS, SAS Institute, Cary, NC). When the F test from least square means analysis of variance indicated significance, Student-NewmanKeuls mean separation test was used to evaluate differences in follicular atresia and steroid hormone concentration.

Results Reproductive cyclicity In vehicle-treated animals, the vaginal cytology was indicative of the characteristic 4-day estrous cycle of the rat for the duration of the project (Fig. 1A). However, evaluation of the cytology from the rats receiving Antide showed a rapid and profound period of diestrus beginning on day 2 and continuing until day 30, 21 days after the last Antide injection (Fig. IB). Reproductive cyclicity in Antide-treated rats was erratic from days 30-42, but returned to a more characteristic 4-day pattern after day 42. On day 120, persistent vaginal estrus ensued in the Antide-treated rats and prevailed until death on day 165 (rats were 295 days of age; Fig. IB). GnRH antagonist-related side-effects No signs of histamine-related side-effects were observed in any of the Antide-treated rats. No edema of the face and extremities or cutaneous wheal reaction at the site of injection was seen at any time after any of the four injections.

27

luminal leukocytes (Fig. 2, C and D). Additionally, the lumina of the antral follicles of the Antide-treated rats were frequently enlarged with a diameter greater than 800 /xm (Fig. 3A) and contained leukocytic cells. Even 34 weeks after cessation of Antide therapy (days 30 and 42), the lumina of the antral follicles of the Antidetreated rats were lined by an unusually thin layer of granulosa cells which displayed signs of luteinization (Fig. 3B). Oocytes from the Antide-treated rats frequently appeared fractured or had a granular cytoplasm. A relative paucity of corpora lutea was found in the Antide-treated rat ovary on days 6, 12 (Fig. 2C), and 18. Those corpora lutea present exhibited the invasion of stromal fibroblasts and contained degenerating cells compared to the corpora lutea of the vehicle-treated rats (Fig. 2A). No corpora lutea were found in the ovaries of the Antide-treated rats after day 18 (i.e. days 24, 30, and 42; Fig. 3A). The ovaries from the Antide-treated rats killed on day 165 (156 days from the last injection) contained large (>800 /xm), atretic antral follicles and no corpora lutea compared to the ovaries of the vehicle-treated rats, which contained both healthy follicles and corpora lutea. Morphometric analysis indicated that 40 of the 258 (overall average of 15.3%) antral follicles examined in the ovaries of the vehicle-treated rats (n = 28) displayed signs of atresia (Fig. 4). However, 268 of the 389 (overall average, 64.7%) antral follicles evaluated from the ovaries of Antide-treated rats (n = 36) displayed obvious signs of cell death (Fig. 4). Initially, both serum FSH and E2 concentrations of Antide-treated rats were lower (P < 0.05) on days 6,12, and 18, but then increased to values similar to those in vehicle-treated rats on days 24 and 30 (Table 1). E2 levels were higher (P < 0.05) in Antide-treated than in vehicletreated rats on day 42. Serum P levels in Antide-treated rats were lower (P < 0.05) than those in vehicle-treated rats on all days tested (Table 1).

Exp 1: Antide-induced follicular atresia The numbers of healthy and atretic antral follicles were similar (P > 0.05) in the presence or absence of endometriosis; therefore, results from these groups were pooled. Microscopic evaluation of the ovarian tissue sections from the vehicle-treated rats illustrated the presence of both healthy follicles and newly formed corpora lutea typically found in the rat ovary (Fig. 2, A and B). No difference (P < 0.05) in the number of antral follicles per section was found in the ovarian sections evaluated from the vehicle- and Antide-treated rat ovaries. However, as early as days 6 and 12, sections from the ovaries of the Antide-treated rats displayed antral follicles whose granulosa cell layer contained numerous pyknotic nuclei, vacuoles, an absence of mitotic figures, as well as intra-

Exp 2: Antide-associated infertility All vehicle-treated rats conceived after one mating, as evidenced by their appropriate mating behavior, the presence of sperm in all vaginal smears, and successful parturition. In the Antide-treated rats, no sperm was found in the vaginal smears after their first spontaneous proestrus cocaging with a proven breeder male. For the seven subsequent proestrus cocagings of the Antide-treated rats, sperm was observed in 85% of the vaginal smears. Sperm was found in all of the vaginal smears of the Antide-treated rats that had received follicular stimulation before their first mating. However, none of the Antide-treated rats became pregnant after either spontaneous proestrus or follicular stimulation. Even after

28


Endo• 1990 Voll27«Nol

FlG. 2. Morphology of follicles and corpora lutea from ovaries of control (A, X40; B, X200, original magnification) vs. Antide-treated (C, X40; D, X200) rats on day 12. Healthy antral follicles (F) and newly formed corpora lutea (CL) were present in the ovaries of the vehicletreated rats (A). Healthy antral follicles contained a granulosa cell layer of appropriate thickness which was free of pyknotic nuclei and lacked the presence of leukocytic cells in the follicular lumen (B). The ovaries from the Antide-treated rats displayed atretic antral follicles and few corpora lutea, which, when present, exhibited evidence of fibroblastic invasion (C). Obvious signs of granulosa cell (GC) death, including numerous pyknotic nuclei (P) and vacuoles (V) as well as intraluminal leukocytes (WBC) were seen in these antral follicles of the Antide-treated rats (D).

D mating for eight cycles, no pregnancies ensued in the Antide-treated rats. The oviducts of the rats treated with Antide contained no oocytes. Oviducts from vehicletreated rats were not flushed, since all pregnancies were allowed to go to term. Histologically, the ovaries of the Antide-treated rats were similar to those of the Antidetreated rats of Exp 1. The ovaries of these animals also contained large atretic antral follicles and lacked newly formed corpora lutea.

Discussion Although Antide, as we have previously shown (4), elicits a profound regression of endometriosis without histamine-related side-effects and, as such, may represent a beneficial therapy, these studies indicate that

Antide also produces follicular atresia and infertility in the rat. The fact that no difference was found in the assessment of the overall number or number of the antral follicles of the endometriosis and control rats indicates that the reproductive dysfunction observed in this study was due to the administration of Antide and not to the presence of endometriosis. The deleterious effects of Antide led to an anovulatory state, with an atretic ovarian morphology devoid of corpora lutea. Follicular atresia is an ongoing process whose mechanisms of initiation and control are not clearly understood. Preantral follicular growth depends on the presence of extraovarian FSH and high local concentrations of estrogens (6). The suppression of these hormones, therefore, leads to atresia of antral follicles in a fashion similar to that caused by hypophysectomy. However,


B FIG. 3. Ovary (A; X40) and luteinized granulosa cells of a follicle (B; X400) from an Antide-treated rat on day 42 (33 days after the last Antide injection). The ovary lacked corpora lutea and contained atretic antral follicles (F) more than 800 ^m in diameter (A) which contained luteinized granulosa cells (LGC; B). GE, Germinal epithelium; FF, follicular fluid.

unlike in the hypophysectomized animal, after cessation of Antide therapy, serum FSH concentrations returned to normal values, while the aberrant development of antral follicles persisted. This anomalous follicular development suggests that local paracrine effects may be present and involved in the persistence of the follicular atresia. Intraovarian products, such as enzymes and hormones, from degenerating follicles have a paracrine effect on other follicles (6). The remarkably high percentage of atretic antral follicles in the Antide-treated rats could have been induced in a paracrine fashion whereby the

29

degeneration of neighboring follicles could have adversely acted on adjacent follicles. Furthermore, the absence of paracrine signals (including P) from the relative paucity of corpora lutea present in the Antide-treated rat ovaries may also have contributed to follicular demise. Of particular concern is the long term Antide-associated follicular atresia. Six months after withdrawal from Antide treatment, aberrant ovarian morphology persisted. This morphology, including multiple large (>800 ixm) atretic follicles with no corpora lutea, may be related to the dose of GnRH antagonist used in this study. Kovacs and co-workers (7) found that a similar, yet reversible, aberrant ovarian morphology occurred after the administration of an analog similar to Antide at a 20-fold lower total dose over a 21-day period. At this lower dose, these ovarian anomalies were present for only 14 days after cessation of treatment. Although this lower dose was reported to be antiovulatory, reproductive cyclicity, although irregular, was still present, and serum P was decreased (25.19 ± 3.60 to 9.91 ± 2.01 ng/ml) to only 40% of the control value during the treatment period, suggesting incomplete suppression of gonadal function. Further studies are necessary to determine if the long term deterioration of follicular quality observed after Antide treatment was dose dependent and, thereby, possibly reversible. The degeneration of follicular quality reported here was also consistent with the infertility we have observed in Antide-treated rats. The profound infertility observed in the rats treated with Antide was most likely due to oocyte degeneration, follicular atresia, and anovulation. The absence of oocytes/embryos in the oviducts of the Antide-treated rats was attributed to anovulation, as no corpora lutea were observed. Even after follicular stimulation (PMSG/ hCG), no evidence of ovulation was found. A deviation in the normal mating behavior of the Antide-treated rats was indicated by the absence of sperm in the vaginal lavages of the Antide-treated females after their first posttreatment cocaging with proven breeder males. Normally, the female rat will only exhibit lordosis after exposure to the appropriate cyclic hormonal milieu leading to the period of estrus (8). As Antide caused an interruption of the normal secretory patterns of the gonadotropin and steroid hormones, Antide may have altered mating behavior. Although Antide initially suppressed serum FSH and E2, concentrations did return to normal after cessation of Antide treatment; however, P concentrations remained suppressed for the duration of the study. The absence of cyclic exposure to P may explain the lack of estrous behavior in the Antidetreated rats at their first proestrus. However, mating behavior did occur in 85% of the subsequent cocagings, while P concentrations remained suppressed. Therefore,


30

8 ioo

1

Endo • 1990 Vol 127 • No 1

Antide Vehicle

INJECTIONS

1 1 11

SO-

2

FlG. 4. Percentage of antral follicles examined (n = 647 total follicles examined) that were atretic after Antide (n = 268 atretic follicles) or vehicle (n = 40 atretic follicles) administration. Each point represents the least squares mean ± the least square SEM. Means with different superscripts are statistically different (P < 0.005).

SO-40 20 0

I I I I t I I I I I

0

12

I I I I I

18

24

I I I I I

42

30

Days Post—Injection TABLE 1. Serum FSH, E2, and P in vehicle-treated vs. Antide-treated rats E2 (pg/ml)

FSH (ng/ml) Day

Vehicle 0 6 12 18 24 30 42

24.4 ± 4.1 34.5 ± 1.3 28.6 ± 4.2 23.1 ± 1.5 21.6 ±2.6 26.3 ± 2.9 28.4 ± 2.1

Antide

Vehicle

16.9 ± 1.3* 16.8 ± 0.6" 12.2+1.1 6 26.3 ± 1.1" 21.3 ± 1.9" 29.2 ± 4.6°

41.2 + 5.8 20.4 ± 8.9C 7.2 ± 1.2C 9.2 ± 2.6C 13.3 ± 6.3C 22.8 ± 3.9C 10.1 + 4.3C

P (ng/ml) Antide 1.1 0.0 0.0 8.3 22.9 29.0

± 0.8d ± 0.0" ± 0.0d ± 1.8C ± §.2C ± 3.3d

Vehicle 7.1 ± 0.7 9.8 ± 0.5" 15.6 ±7.1° 18.6 ± 3.0° 25.4 ± 5.9' 14.5 ± 7.3e 14.0 ± 2.7°

Least squares mean ± least squares SEM hormone concentrations in rats injected with Antide or vehicle on days 0, 3, 6, and 9 (day 0 superscripts indicate different means between vehicle- and Antide-treated rats for each hormone on each day (P < 0.05).

the infertility seen in this study was more likely related to Antide-generated ovarian dysfunction than to mating behavior. The mechanism of Antide-induced follicular atresia and infertility in the rat could be a result of the action of this GnRH antagonist on the hypothalamic-pituitary axis, the ovary, or both. However, the exact mechanism by which Antide exerts its effects on reproductive function cannot be directly discerned from these studies. Antide may initially exert its effects via the hypothalamic-pituitary axis, as indicated by the initial decrease in the serum FSH concentration in response to Antide treatment. The return of Antide-suppressed serum FSH concentrations to values equal to those in control rats 2 weeks after Antide withdrawal suggests a rebound of pituitary function. This suppression and recovery of serum FSH of the Antide-treated rats appear to affect ovarian function, since the serum concentration parallels the systemic FSH pattern. The fact that the changes in concentrations mirror the changes in FSH concentrations indicates that the ovary may be responsive to the

Antide 5.6 ± 4.4 ± 5.6 ± 3.9 ± 2.1 ± 4.1 ±

0.2' 1.5' 1.5' 1.0' 0.6' 0.9'

proestrus). Different

gonadotropins and again suggests that the effect of Antide may occur at the level of the hypothalamic-pituitary axis. The frequency of FSH pulses plays a pivotal role in its effectiveness as an endocrine signal. The single daily measurements of FSH in this study did not allow us to measure FSH pulse frequency. Even though the amplitude of FSH secretion appeared to return to the control concentration by day 24 of treatment, we cannot rule out the possibility that Antide modified the pulse frequency of FSH secretion, which, in turn, was responsible for the ovarian dysfunction observed in this study. After Antide-induced suppression and recovery, on day 42, the serum E2 concentration in Antide-treated rats was significantly greater than that in the vehicle-treated rats. This difference in serum E2 concentration could be a result of asynchronous reproductive cyclicity between the vehicle- and Antide-treated rats on day 42. Alternatively, the disparate E2 concentrations of the Antidetreated rats on day 42 could occur because Antide induced premature ovarian failure. Both the high levels of


E2 and the aberrant ovarian morphology of the Antide treated rats on day 42 fit the scenario of premature ovarian aging (6). Although a recovery of FSH and E2 was observed after Antide withdrawal, systemic P concentrations remained dramatically suppressed for the duration of the experiment. The absence of systemic P in Antide-treated rats may be a reflection of the paucity of newly formed Pproducing copora lutea. However, suppressed P production also suggests that Antide exerts its deleterious effects directly at the level of the ovary. The direct actions of GnRH on the ovary have been previously documented in the rat (9-11). Pharmacological doses of GnRH and its agonistic analogs have been shown to inhibit steroidogenesis in rat ovary both in vitro and in vivo (9). Potentially, pharmacological doses of a GnRH antagonist could also inhibit steroid production via direct action with the ovary. The presence of pyknotic granulosa cells and atypical oocytes that we observed in the antral follicles of the Antide-treated rats indicates that Antide may exert a deleterious effect on follicular function, either directly or via alteration of granulosa cell function. Recent evidence indicates that GnRH and its analogs may also act directly on the human ovary. Turek and coworkers (12) noted inhibitory actions of GnRH on P secretion by cultured human granulosa cells. The presence of GnRH receptors or binding sites in human luteal tissue (13, 14) and in the granulosa cell layer of preovulatory follicles (15) has also been recently documented. Unfortunately, conflicting data exist on the direct effects of GnRH and its analogs on the human ovary. Contrasting reports have found that GnRH or a GnRH agonist did not alter FSH- or hCG-stimulated steroidogenesis in vitro (16, 17). Furthermore, other studies have indicated the absence of high affinity GnRH receptors in the human corpus luteum (18). These studies suggest that GnRH and its agonists do not act directly on human granulosa or luteal cells to modulate steroidogenesis. Additional work is needed to clarify the extrapituitary actions of GnRH and its analogs in humans. Further studies are needed to determine if Antide is exerting its effects directly on pituitary gonadotropin production, which results in irregularities of reproductive function, or if Antide is acting in an extrapituitary fashion. Regardless of the mechanism of action, if the effects of Antide observed in the rat are similar to those in humans, Antide would be undesirable as an adjunct to infertility therapy. These studies suggest that before use in clinical trials for endometriosis, further investigation of the effects of Antide on folliculogenesis and fertility in the rat and primate are indicated.

31

Acknowledgments The authors wish to thank Ms. Cindy Meier, Ms. Melissa Seiter, and Mr. Warren Nothnick for their assistance with this project.

References 1. Schmidt F, Sundaram K, Thau R, Bardin CW 1984 [Ac-DNal(2)1,4FD-Phe2,D-Trp3,DArg6]-LHRH, a potent antagonist of LHRH, produces transient edema and behavioral changes in rats. Contraception 29:283 2. Ljungqvist A, Feng D-M, Tang P-FL, Kubota M, Okamoto T, Zhang Y, Bowers CY, Hook WA, Folkers K 1987 Design, synthesis and bioassays of antagonists of LHRH which have high antiovulatory activity and release negligible histamine. Biochem Biophys Res Commun 148:849 3. Leal JA, Williams RF, Danforth DR, Gordon K, Hodgen GD 1988 Prolonged duration of gonadotropin inhibition by a third generation GnRH antagonist. J Clin Endocrinol Metab 67:1325 4. Sharpe KL, Bertero MC, Vernon MW 1990 Rapid regression of endometriosis by a new gonadotropin-releasing hormone antagonist in rats with surgically induced disease. In: Chada DP, Buttram VC (eds) Current Concepts in Endometriosis. Liss, New York, p 449 5. Vernon MV, Wilson EA 1985 Studies on the surgical induction of endometriosis in the rat. Fertil Steril 44:684 6. Peters H, McNatty KP 1980 The Ovary. University of California Press, Berkeley, p 120 7. Kovacs M, Mezo I, Seprodi J, Csernus V, Teplan I, Flerko B 1989 Effects of long-term administration of a superactive agonistic and an antagonistic GnRH analog on the pituitary-gonad system. Peptides 10:925 8. Turner CD, Bagnara JT 1976 General Endocrinology. Saunders, Philadelphia, p 472 9. Hseuh AJW, Erickson FG 1979 Extrapituitary action of gonadotropin-releasing hormone: direct inhibition of ovarian steroidogenesis. Science 204:854 10. Hseuh AJW, Wang C, Erickson FG 1980 Direct inhibitory effect of gonadotropin-releasing hormone upon follicle-stimulating hormone induction of luteinizing hormone receptors and aromatase activity in rat granulosa cells. Endocrinology 106:1697 11. Clayton RN, Harwood JP, Catt KJ 1979 Gonadotropin-releasing hormone analogue binds to luteal cells and inhibits progesterone production. Nature 282:90 12. Turek RW, Mastroianni L, Blasco L, Strauss JF 1982 Inhibition of human granulosa cell progesterone secretion by gonadotropinreleasing hormone agonist. J Clin Endocrinol Metab 54:1078 13. Popkin R, Bramley TA, Currie A, Shaw RW, Baird DT, Fraser HH 1983 Specific binding of LHRH to human luteal tissue. Biochem Biophys Res Commun 114:750 14. Bramley TA, Stirling D, Swanston IA, Menzies GS, McNeilly AS, Baird DT 1987 Specific binding sites for gonadotropin-releasing hormone, LH/chorionic gonadotropin, low-density lipoprotein, prolactin, and FSH in homogenates of human corpus luteum. II. Concentrations throughout the luteal phase of the menstrual cycle and early pregnancy. J Endocrinol 113:371 15. Latouche J, Crumeyrolle-Arias M, Jordan D, Kopp N, AugendreFerrante B, Cedard L, Haour F 1989 GnRH receptors in human granulosa cells: anatomical localization and characterization by autoradiographic study. Endocrinology 125:1739 16. Casper RF, Erickson GF, Yen SSC 1984 Studies of the effect of gonadotropin-releasing hormone and its agonist on human luteal steroidogenesis in vitro. Fertil Steril 42:39 17. Casper RF, Erickson GF, Rebar RW, Yen SSC 1982 The effect of luteinizing hormone-releasing factor and its agonist on cultured human granulosa cells. Fertil Steril 37:406 18. Clayton RN, Huhtaniemi IT 1982 Absence of gonadotropin-releasing hormone receptors in human gonadal tissue. Nature 299:56-8