BIOLOGY OF REPRODUCTION 56, 1565-1569 (1997)
Ovarian Activin Receptor Subtype and Follistatin Gene Expression in Rats: Reciprocal Regulation by Gonadotropins' J.A. Aloi, J.C. Marshall, M. Yasin, J.T. Gilrain, D.J. Haisenleder, and A.C. Dalkin2 Department of Internal Medicine, Division of Endocrinology & Metabolism, University of Virginia Medical Center, Charlottesville, Virginia 22908 ABSTRACT The production of activin, follistatin (FS), and inhibin, proteins present in the ovary and involved in mammalian reproduction, isregulated by gonadotropins and estradiol. We report here gonadotropin regulation of ovarian activin receptor (ActR) subtype and FS mRNAs. Expression of ActRI, ActRIIA, ActRIIB, and FS mRNA was measured on the afternoon of proestrus (1800 h) and the morning of estrus (0800 h). ActRI and ActllA subtype mRNA concentrations fell by -50% (p < 0.05) following the proestrous gonadotropin surge (ActRIIB mRNA was undetectable), while FS mRNA was unchanged. To define the contribution of gonadotropins, hypophysectomized (HYPOX) female rats were given recombinant human (rh) FSH and hCG, which decreased both ActR mRNAs (by -70% and -50% for ActRI and IIA, respectively) and increased FS mRNA by 2-fold. As gonadotropins could act via estradiol (E2), HYPOX rats were given E2; ActRI was decreased, but ActRIIA mRNA was increased. The actions of gonadotropins were preferential, as the combination of rhFSH and hCG with E2 reduced ActRIIA mRNA. FS mRNA was increased to a similar degree by E2 and/or gonadotropins. These data suggest that gonadotropins regulate ActR and FS gene expression via multiple mechanisms. Both a direct action on ActRIIA (inhibition) and an indirect action through E, on ActRI (inhibition) and FS (stimulation) suggest potential physiologic mechanisms for the reciprocal regulation of ActR subtype and FS mRNAs. INTRODUCTION Regulation of ovarian follicular maturation and gamete production is a complex process with multi-level regulation by endocrine, autocrine, and paracrine factors. Classically, the glycoprotein hormones, FSH and LH, are released from the pituitary in response to GnRH. FSH and LH then act in concert at the gonad to regulate folliculogenesis and steroidogenesis, respectively (for reviews see [1-3]). The twocell gonadotropin hypothesis describes this cooperativity of FSH and LH. FSH is the main modulator of ovarian follicular maturation, acting to increase granulosa cell responsiveness both to itself and to other hormones. Specifically, FSH facilitates follicular maturation and development by increasing the content of its own receptor, synergizes with estradiol (E2) to further increase FSH receptor number and aromatase activity, and induces LH receptors. LH promotes the synthesis of androgens by the theca interna, and androgens act as a substrate for the granulosa cell to produce E2 under the influence of FSH. LH also functions to stimulate preovulatory follicular growth and induce ovulation, and
MATERIALS AND METHODS Animals
Accepted February 3, 1997. Received November 22, 1996. 'Supported by effress grant to #J-313 to A.C.D. 2Correspondence: Alan C. Dalkin, 5041 MR4 Building, University of Virginia Health Sciences Center, Charlottesville, VA 22908. FAX: (804) 924-1284; e-mail:
[email protected]
regulates corpus luteum function. Thus, gonadal function is clearly dependent on both FSH and LH signaling. Ovarian responses to gonadotropin stimuli involve the production of both steroid and peptide hormones, the latter including the related inhibins and activins as well as the activin-binding protein follistatin (FS). While these hormones serve classic endocrine feedback pathways in regulating the synthesis and secretion of FSH (for review see [4, 5]), they also may serve paracrine and autocrine functions directly in the ovary (for review see [5, 6]). Data suggest that inhibin, in addition to its endocrine FSH-suppressing role, acts locally to positively regulate LH-induced androgen production in the ovary, while activin tends to antagonize this effect [7]. Activin directly stimulates inhibin subunit mRNA expression, inhibin secretion, and FSH receptors, and, in parallel, augments the actions of FSH to increase cAMP, inhibin and E2 production and LH receptor numbers [8-12]. FS, either by its interactions with activin or by direct effects on progesterone production by granulosa cells, serves to promote luteinization or atresia [13]. Activin acts via binding to specific cell surface receptors, including type I (ActRI) and type II (ActRIIA and IIB) subtypes (for review see [14, 15]). While little is known regarding activin receptor (ActR) expression in the adult ovary, in extragonadal tissues such as the rat pituitary gland, changes in ActR and FS mRNAs have been reported after ovariectomy and during the estrous cycle [16-19]. Moreover, activin ligand itself appears to alter expression of these gene products [20]. Hence, regulation of activin ligand, ActR numbers, and/or FS could serve as a means to modify an activin signal. A number of reports have shown that ovarian mRNAs encoding the inhibin/activin subunits undergo patterned changes during the rat estrous cycle [21-23], and we have recently reported that both gonadotropins and E 2 may serve to regulate the expression of these genes [24]. It was the purpose of these studies to determine whether ovarian ActR and FS mRNA levels varied during the time of the endogenous gonadotropin surge. Thereafter, we aimed to establish the roles of gonadotropins and E2 in the regulation of those mRNAs.
Adult (200-250 g) female Sprague-Dawley rats were used in all experiments. For studies examining mRNA expression on proestrus and estrus, only animals that demonstrated three consecutive 4-day cycles, as confirmed by daily vaginal cytology, were used. Hypophysectomized (HYPOX) animals were purchased from Hilltop Farms (Scottsdale, PA) and used for studies 5-7 days after HYPOX. For the i.v. administration of human recombinant FSH (rhFSH; see below for rationale), right atrial catheters were placed via the external jugular vein under metofane
1565
1566
ALOI ET AL.
anesthesia. Animals were killed by decapitation; ovaries were rapidly removed, decapsulated, and then snap frozen in liquid nitrogen and stored at -70°C until total RNA extraction. All animal experimentation was conducted in accordance with the NIH and University of Virginia guidelines for the care and use of laboratory animals. Hormones For these studies we utilized a regimen of rhFSH and hCG that has been previously shown to restore serum E2 and inhibin to intact levels in HYPOX rats while yielding approximate circulating physiologic concentrations of rhFSH [24]. Recombinant human FSH was kindly provided by Ares Advanced Technology (Randolph, MA). Aliquots were diluted to a concentration of 1000 IU/ml with 0.9% saline/0. 1% BSA and stored frozen until use. Aliquots were diluted 1:10 with 0.9% saline at the start of experiments, and 5 IU of rhFSH was administered every 6 h i.v. for 24 h. As rhLH is not yet available, hCG was obtained from Steris Laboratories (Phoenix, AZ), diluted with bacteriostatic water, and stored at 4°C until use; 10 IU of hCG was administered s.c. at the start of experiments. E2 (estradiol171) was purchased from Sigma Chemical Company (St. Louis, MO) and suspended in sesame oil. Experimental Protocols HYPOX animals received either 1) 10 IU hCG s.c. and 5 IU rhFSH administered i.v. every 6 h for 4 doses; 2) E2 via a silastic implant [24] to reflect physiologic concentrations (E2 Low; 42.3 2.2 pg/ml); 3) a single 50 Vxg E 2 s.c. injection to be 15- to 20-fold higher than plasma E2 and reflect intraovarian E2 levels (E 2 High; 798 + 98 pg/ml); or 4) hCG+rhFSH+E 2 High. Saline-treated animals served as controls. All treatments were continued for 24 h. Messenger RNA Quantitation 1. ActRs. We have previously reported a quantitative reverse transcription polymerase chain reaction (RT-PCR) assay for the ActRI, ActRIIA, and ActRIIB mRNAs that is able to detect low levels of target mRNA using nanogram amounts of native RNA [16, 20]. The rat ActR subtype cDNAs were kindly provided by Dr. K. Mayo (Northwestern University, Evanston, IL). The RT-PCR assay is based on coamplification of a constant amount of native mRNA (pituitary derived) and a standard curve of size-altered and polyadenylated competitive template mRNA (produced in vitro) in a series of reactions (see [16] and [20] for the characterization of the competitive template cDNAs). These assays 1) allow for the use of oligo(dT) priming of both native and competitive template mRNAs, 2) allow for identical primers for native cDNA and competitive template cDNA amplification, and 3) prevent heteroduplex formation between DNA products as a result of their dissimilar midregions. The PCR reactions (quadruplicate measurements with competitive template concentrations of 2, 10, 50, and 300 fg) are performed in the presence of trace amounts of [3 2P]CTP. The primers used for these assays were the following: ActRI-upstream primer 5' CTCGACAGATAACCCTGTTGGAGT 3' (base pairs [bp] 761-784) and downstream primer 5' GTCCGTTCTTCTTCACGAGGATGT 3' (bp 1169-1192), generating a native product of 431 bp and a competitive template product of 719 bp; ActRIIA-upstream 5' GCTCACTGTCAGACTTTCTTAAGG 3' and downstream 5' GTGCAACGAGA-
AGCCAATTCCCAT 3', generating a native product of 387 bp and a competitive template product of 603 bp; ActRIIB-upstream primer 5' GCTCAGCTCATGAACGACTTTG 3' (bp 1070-1092) and downstream primer 5' TGCAACGAGAAACGAGCTCCCA 3' (bp 1634-1655), generating a native product of 585 bp and a competitive template product of 770 bp. After completion of PCR amplification, the DNA products were separated by gel electrophoresis (Visigel, Stratagene Corp., La Jolla, CA), excised, and counted in a scintillation counter. The amount of DNA product generated from each mRNA species (native and competitive template) was calculated (on a molar basis) and the point of equivalence (i.e., point at which the native and competitive template-generated DNA products are equimolar) was determined. As both native and competitive template mRNAs undergo reverse transcription and then polymerase chain reaction simultaneously, molar equivalence at completion of the assay (between cDNA products) reflects molar equivalence at the outset of the assay (between mRNAs), allowing quantification of native mRNA levels. For ActRI mRNA measurements, 0.5 pxg of total ovarian RNA was used. For ActRIIA mRNA measurements, 0.2 Rpg of total ovarian RNA was used. ActRIIB mRNA was not detected in total rat ovarian RNA using starting amounts up to 1 g (35 cycles of amplification). 2. FS. The relative abundance of ovarian FS was found to be sufficient to allow for quantitation using less sensitive assays. Preliminary experiments were performed to establish the kinetics for FS mRNA dot-blot hybridization (data not shown) and determine the optimal amount of total cytoplasmic RNA per dot. Based on these results, 5 jig RNA was spotted to nitrocellulose filters. A saturating amount of [3 2 P]CTP-labeled cDNA (also kindly provided by Dr. K. Mayo) was allowed to hybridize overnight using our standard dot-blot technique [251. Filters were washed and exposed, and dots were excised and quantified in a B counter. Results are expressed as picograms bound per 100 micrograms of ovarian DNA. To control for interassay variability, an aliquot from a pool of ovarian RNA was spotted on each filter, and all samples for each experiment were assayed on the same filter. Interassay variability was 14%, and intraassay variation was < 10 %. Statistical Analysis Data were examined by one-way analysis of variance, with Duncan's Multiple Range test applied to determine significant differences between groups. All data are expressed as the mean SEM. RESULTS ActR Subtype and FS mRNAs during the Estrous Cycle Initial studies were performed to assess whether ovarian ActR subtype or FS mRNAs are regulated between proestrus and estrus. ActRI and ActRIIA receptor subtypes were present on both days, while ActRIIB mRNA was not detected using starting amounts of up to I pLg RNA. Messenger RNA concentrations were measured on the afternoon of proestrus (1800 h) and compared to values on the morning of estrus (0800 h, Fig. 1). ActRI and ActRIIA subtype mRNAs were significantly reduced (ActRI by 54%, ActRIIA by 42%) on the morning of estrus (p < 0.05), following the proestrous gonadotropin surge. In contrast, FS mRNA concentrations were unchanged. These data sug-
1567
OVARIAN ACTIVIN RECEPTORS AND FOLLISTATIN 200
ActRI
T
-T 4-
ActRI
z
100-
z
*
C a 0 O
o0-
o
,-M.
45
600-
4
E 30 *
ActRIIA
E 400-P
15
4 D
o -
.--
4.5
Z
7.5
Follistatin T
o
Follistatin
o 1.5
5.0 cns.0
*
8 0 1i 2.5 -
z 0.
H
0
0
a
*
200 -
0 C 3.0
'
ActRIIA
a
z
n
Proestrus
Estrus
FIG. 1. ActR subtype and FS mRNA concentrations during the estrous cycle. Animals were killed on either proestrus (1800 h) or estrus (0800 h). For each group, n = 7-8 animals. *p < 0.05 vs. proestrus.
gest that endogenous gonadotropin signals are associated with dynamic changes in ovarian ActR subtypes while FS mRNA remains relatively static. Regulation of Ovarian ActR and FS mRNA Concentrations To determine whether the changes in ActR subtype mRNAs resulted from the proestrous gonadotropin surge to the ovary or the subsequent ovarian production of E2, the effects of gonadotropin and/or steroid treatment were studied; the results are shown in Figure 2. Ovarian ActR subtype mRNA concentrations were higher and FS mRNA concentrations were lower in HYPOX animals. Compared to levels at proestrus and estrus (compare Figs. 1 and 2), ActRI levels were 34- and 68-fold higher, ActRIIA levels were 10- and 17-fold higher, and FS levels were 50% and 43% lower in HYPOX animals. Gonadotropin administration for 24 h tended to restore ActRI, ActRIIA, and FS mRNA concentrations toward intact levels. That is, ActRI and ActRIIA mRNA levels fell (85% and 50%, respectively) and FS mRNA levels rose (2.2-fold). E2 replacement alone at either low or high doses decreased ActRI (65% and 87%, respectively) and increased FS mRNA (2.5- and 2-fold, respectively); this was similar to what was observed with gonadotropin treatment. It is noteworthy that the higher concentration of E2 was more effective for reducing expression of ActRI mRNA (p < 0.05 vs. E 2 Low). Surprisingly, either concentration of E2 increased (p < 0.05) ActRIIA mRNA levels above those seen in HYPOX animals (E2 Low, 1.5-fold increase; E2 High, 1.4-fold increase). HYPOX animals also received a combination of gonad-
Z
i
4M 0.
Saline
I
FSH +hCG
E 2 Low
E High
FSH+hCG +E 2 High
HYPOX
FIG. 2. ActR subtype and FS mRNA concentrations: effects of 1) rhFSH (5 IU i.v. every 6 h for 24 h) and hCG (10 IU s.c. at the time of the first rhFSH dose); 2) an E2implant (E2Low: serum E2 -40 pg/ml); 3) a 50-itg E2 injection (E, High: serum E, 700-800 pg/ml) for 24 h; or 4) the combination of E2 High, rhFSH, and hCG in HYPOX female rats for 24 h. For each group, n = 5-6 animals. *p < 0.05 vs. HYPOX, **p < 0.05 vs. E2 Low.
otropin and E2 replacement. For ActRI and FS mRNAs, the actions of hCG+rhFSH+E 2 were similar to those of either gonadotropins or E 2 alone, suggesting that hCG and rhFSH act on these gene products indirectly via their actions in stimulating E2 production. In contrast, for ActRIIA, combined gonadotropin and steroid replacement was similar to treatment with gonadotropins alone, suggesting that ActRIIA gene expression is regulated in a manner distinct from that of ActRI or FS, and that the inhibitory action of gonadotropins is not mediated by E2. DISCUSSION These data demonstrate dynamic expression of ActR subtypes on different days of the rat estrous cycle, with a decrease in ovarian mRNA concentrations for both ActRI and ActRIIA following the proestrous gonadotropin surge. FS mRNA, in contrast, was unchanged. In the rat, mRNA expression of a single type I (ActRI) and two type II (IIA and IIB) receptors has been previously reported [26-29]. The distribution of the type II ActR mRNAs is widespread, in particular in the reproductive tissues, prostate, gonads, and brain [28]. Using in situ techniques in ovarian tissues, Cameron et al. reported overall greater staining for ActRIIA mRNA than for IIB mRNA, with primary staining of the IIA mRNA over granulosa cells, oocytes, and corpus luteum [30]. We were unable to detect ActRIIB mRNA in the adult rat ovary, which agrees with prior studies that also
1568
ALOI ET AL.
found the IIB subtype mRNA to be in low abundance in rat ovarian structures [28, 30]. It is possible that apparent binding with in situ hybridization could reflect the high degree of homology between the ActRIIA and IIB mRNAs, a limitation not encountered with higher-stringency polymerase chain reaction. Few data exist regarding the regulation of ovarian ActRs. Woodruff et al. [31], utilizing in situ ligand binding, reported binding sites in both granulosa and luteal cells and also noted an overall decline in activin-binding sites from proestrus compared to estrus. This technique does not differentiate between activin binding to its receptor and to FS, and our data suggest that the decline in binding reflects a reduction in ActR between proestrus and estrus. As we observed changes in ActR mRNA expression occurring at a time in the rat estrous cycle when gonadotropins and E2 concentrations are in flux, we therefore attempted to determine whether these factors were involved the regulation of ActR subtype mRNAs. The present model was selected for several reasons. First, we used a regimen of FSH and hCG that has been shown to restore the decline in plasma inhibin and E 2 observed in HYPOX rats and to restore ovarian inhibin/activin subunit gene expression toward intact levels [24]. Secondly, while treatment with both gonadotropins for a similar duration may not truly replicate the pituitary signals to the ovary occurring through the estrous cycle, the current methodology allows for assessment of an integrated ovarian response in a "physiologic" system. The use of total ovarian RNA samples precludes comment on whether the changes reflect the sum response from a homogenous population of cells (e.g., granulosa cells) or represent the averaging of responses from differing cell types. However, in situ data suggest that the granulosa cell is the principal cell type in the ovary expressing ActRs [30]. The fall in ActR mRNAs following exogenous gonadotropin treatment of HYPOX animals, or after the proestrous gonadotropin surge, suggests a decline in activin action in the ovary in response to gonadotropins. Similarly, the rise in FS mRNA observed in HYPOX animals given gonadotropin replacement is compatible with this view. Free plasma levels of activin do not vary through the human menstrual cycle [32], supporting the proposal that activin functions primarily as a paracrine/autocrine factor in regulating follicular development and differentiation. To date, data regarding the ovarian actions of activin are conflicting, and activin has been implicated in both promoting and preventing follicular atresia. In vivo administration of activin, by either an intrabursal injection or repeated s.c. injections, has been shown to promote atresia [33], and in a later study, premature ovulation [34]. However, in vitro experiments have suggested that activin stimulates granulosa cell proliferation [35] and follicular development [36]. Additionally, activin has several indirect effects on the growth and development of granulosa cells by synergizing with FSH to stimulate E 2 and inhibin production. Overall, the finding that activin binding is primarily restricted to healthy follicles [31] lends support to the concept of activin's role in allowing for selective action of FSH on a cohort of developing follicles. FS mRNA was unchanged during the proestrous/estrous period examined in the present study. Nakatani et al. [37] reported that all differentiating granulosa and luteal cells expressed FS mRNA, but expression of FS protein was restricted to granulosa cells of dominant follicles throughout the estrous cycle. Indeed, the pattern of staining re-
ported in that study showed little overall change in FS mRNA, with a shift in staining from dominant, preovulatory follicles early in the estrous cycle to the corpus luteum following ovulation. The authors concluded that expression of FS mRNA may be a general consequence of the entry of a primordial follicle into the pool of growing follicles, a process associated with development of a full complement of FSH receptors and E2 synthesis. Our data, indicating that E2 is the primary factor regulating FS mRNA, are consistent with this concept and could account for the stable level of expression of FS mRNA in ovarian RNA during the rat estrous cycle. That is, increasing E 2 during proestrous morning could initially increase FS gene expression, with the gonadotropin surge serving to maintain expression of FS through the day of estrus. As found for FS mRNA, gonadotropins may regulate ActR gene expression (at least in part) through their action to increase E2. ActRI mRNA was decreased toward intact levels by exposure of HYPOX female rats either to gonadotropin or to E2. E2 may be acting directly at the level of gene transcription, but this is uncertain, as the regulatory regions of the ActRI gene have not yet been described. In contrast, E2 increased ActRIIA gene expression, a finding perhaps explained by the presence of an E2-sensitive transcription site in the ActRIIA 5'-regulatory region [38]. Interestingly, this rise in ActRIIA was prevented by coadministration of rhFSH and hCG with E2. This suggests that a gonadotropin-dependent factor (that is not E2) is also involved in ActRIIA mRNA regulation. Progesterone has been shown to prevent induction of transforming growth factor type II receptors [39], and a potential progesterone response element is also present on the rat ActRIIA promoter [40], suggesting that other gonadal steroids may regulate activin responsiveness. Alternatively, E2 may act indirectly to regulate ActR gene expression through its known effects to increase FS [41] or inhibin [24, 42]. This latter possibility is intriguing, as an inhibin receptor has not yet been identified, and suppression of ActR gene expression could be a potential mechanism for inhibin's biologic antagonism of activin. Formal testing of these concepts in vivo awaits adequate availability of FS, activin, and inhibin. It is important to note that in response to E2, the decline in ActRI gene expression may be of greater physiologic significance than the increase in ActRIIA. Both types I and II ActR subunits are needed for full receptor function, and thus a reduction in ActRI could result in activin insensitivity despite higher levels of ActRIIA. Conversely, a second rat type I receptor could exist (homologous to the ActRIB subunit in humans) and serve to maintain activin responsiveness despite a reduction in ActRI. However, as data supporting a second type I receptor subunit in the rat are lacking, a reduction in ActR expression and increased FS likely represent the actual ovarian response to E.2 In summary, ActR mRNAs decrease in response to exogenous gonadotropins with a reciprocal change observed for FS mRNA. This divergent regulation of two regulators of ovarian function could result in the net biologic effect of gonadotropins to decrease activin action, both by increasing its binding protein and by decreasing its receptor number. This could allow for an additional modulating action of gonadotropins on ovarian follicular development by increasing inhibin production and decreasing activin action. REFERENCES 1. Richards JS. Maturation of ovarian follicles: actions and interactions of pituitary and ovarian hormones on follicular cell differentiation. Physiol Rev 1980; 60:51-89.
OVARIAN ACTIVIN RECEPTORS AND FOLLISTATIN 2. Richards JS, Ireland JJ, Rao MC, Bernath GA, Midgley AR, Reichert LE. Ovarian follicular development in the rat: hormone receptor regulation by estradiol, follicle stimulating hormone and luteinizing hormone. Endocrinology 1976; 99:1562-1570. 3. Hsueh AJW, Adashi EY, Jones PBC, Welsh TH. Hormonal regulation of the differentiation of cultured ovarian granulosa cells. Endocr Rev 1984; 5:76-127. 4. Ying SY. Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicle stimulating hormone. Endocr Rev 1988; 9:267-293. 5. DePaolo LV, Bicsak TA, Erickson GF, Shimasaki S, Ling N. Follistatin and activin: a potential intrinsic regulatory system within diverse tissues. Soc Exp Biol Med 1991; 37:500-512. 6. Findlay JK. An update on the roles of inhibin, activin, and follistatin as local regulators of folliculogenesis. Biol Reprod 1993; 48:15-23. 7. Hsueh AJW, Dahl KD, Vaughan J, Tucker E, Rivier J, Bardin CW, Vale W. Heterodimers and homodimers of inhibin subunits have different paracrine action in the modulation of luteinizing hormone-stimulated androgen biosynthesis. Proc Natl Acad Sci USA 1987; 84: 5082-5086. 8. LaPolt PS, Soto D, Su J-G, Campen CA, Vaughan J, Vale W, Hsueh AJW. Activin stimulation of inhibin secretion and messenger RNA levels in cultured granulosa cells. Mol Endocrinol 1989; 3:1666-1673. 9. Sugino H, Nakamura T, Haseqawa Y, Miyamoto K, Igarashi M, Eto Y, Shibai H, Titani K. Identification of a specific receptor for erythroid differentiation factor on follicular granulosa cell. J Biol Chem 1988; 253:15249-15252. 10. Siao S, Findlay JK, Robertson DM. The effect of bovine activin and follicle-stimulating hormone (FSH) suppressing protein/follistatin on FSH-induced differentiation of rat granulosa cells in vitro. Mol Cell Endocrinol 1990; 69:1-8. 11. Hasegawa Y, Miyamoto K, Abe Y, Nakamura T, Sugino H, Eto Y, Shibai H, Igarashi M. Induction of follicle-stimulating hormone receptor by erythroid differentiation factor on rat granulosa cells. Biochem Biophys Res Commun 1988; 156:668-674. 12. Hutchinson LA, Findlay JK, deVos FL, Robertson DM. Effects of bovine inhibin, transforming growth factor-1 and bovine activin-A on granulosa cell differentiation. Biochem Biophys Res Commun 1987; 146:1405-1412. 13. Xiao S, Findlay JK. Interactions between activin and follicle-stimulating hormone-suppressing protein and their mechanisms of action on cultured rat granulosa cells. Mol Cell Endocrinol 1991; 79:99-107. 14. Mathews LS. Activin receptors and cellular signaling by the receptor serine kinase family. Endocr Rev 1993; 15:310-325. 15. Gaddy-Kurten D, Tsuchida K, Vale W. Activins and the receptor serine kinase superfamily. Recent Prog Horm Res 1995; 50:109-129. 16. Dalkin AC, Gilrain JT, Marshall JC. Ovarian regulation of pituitary inhibin subunit and activin receptor type II gene expression: evidence for a non-steroidal inhibitory substance. Endocrinology 1994; 135: 944-949. 17. Kaiser UB, Lee BL, Carroll RS, Unabia G, Chin WW, Childs GV. Follistatin gene expression in the pituitary: localization in gonadotropes and folliculostellate cells in diestrus rats. Endocrinology 1992; 130:3048-3056. 18. Kaiser UB, Chin WW. Regulation of follistatin messenger ribonucleic acid levels in the rat pituitary. J Clin Invest 1993; 91:2523-2531. 19. Halvorson LM, Weiss J, Bauer-Dantoin AC, Jameson JL. Dynamic regulation of pituitary follistatin mRNA during the rat estrous cycle. Endocrinology 1994; 134:1247-1253. 20. Dalkin AC, Haisenleder DJ, Yasin M, Gilrain JT, Marshall JC. Pituitary activin receptor subtypes and follistatin gene expression in female rats: differential regulation by activin and follistatin. Endocrinology 1996; 137:548-554. 21. Woodruff TK, D'Agostino J, Schwartz NB, Mayo KE. Dynamic
22.
23. 24. 25.
26.
27. 28. 29. 30.
31.
32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.
1569
changes in inhibin messenger RNAs in rat ovarian follicles during the reproductive cycle. Science 1988; 239:1296-1299. Meunier H, Cajander S, Roberts R, Rivier C, Sawchenko P, Hsueh A, Vale W. Rapid changes in the expression of inhibin (x, P3A-, and 3Bsubunits in ovarian cell types during the rat estrous cycle. Mol Endocrinol 1988; 2:1352-1363. Woodruff TK, Mayo KE. Regulation of inhibin synthesis in the rat ovary. Annu Rev Physiol 1990; 52:807-821. Aloi JA, Dalkin AC, Schwartz NB, Yasin M, Mann B, Haisenleder DJ, Marshall JC. Ovarian inhibin subunit gene expression: regulation by gonadotropins and estradiol. Endocrinology 1995; 136:1227-1232. Papavasiliou SS, Zmeili S, Herbon L, Duncan-Weldon J, Marshall JC, Landefeld TD. Alpha and LH mRNA of male and female rats after castration: quantitation using an optimized RNA dot blot hybridization assay. Endocrinology 1986; 119:691-698. Legerski R, Zhou X, Dresback J, Eberspaecher H, McKinney S, Segarini P, de Crombrugghe B. Molecular cloning and characterization of a novel rat activin receptor. Biochem Biophys Res Commun 1992; 183:672-679. deWinter JP, Themmen APN, Hoogerbrugge JW, Klaij IA, Grootefoed JA, deJong FH. Activin receptor mRNA expression in rat testicular cell types. Mol Cell Endocrinol 1992; 83:RI-R8. Feng Z-M, Madigan MB, Chen C-LC. Expression of type II activin receptor genes in the male and female reproductive tissues in the rat. Endocrinology 1993; 132:2593-2600. Ebner R, Chen RH, Lawler S, Zionchech TE Derynck R. Determination of type I receptor specificity by the type II receptors for TGF, or activin. Science 1993; 262:900-902. Cameron VA, Nishimura E, Mathews LS, Lewis KA, Sawchenko PE, Vale WW. Hybridization histochemical localization of activin receptor subtypes in rat brain, pituitary, ovary and testis. Endocrinology 1994; 134:799-808. Woodruff TK, Krummen L, McCray G, Mather JP. In situ ligand binding of recombinant human [1251]activin-A and recombinant human [1251]inhibin-A to the adult rat ovary. Endocrinology 1993; 133:29983006. Demura R, Suzuki T, Tajima S, Mitsuhashi S, Odagiri E, Demura H, Ling N. Human plasma free activin and inhibin levels during the menstrual cycle. J Clin Endocrinol & Metab 1993; 76:1080-1082. Woodruff TK, Lyon RJ, Hansen SE, Rice GC, Mather JP. Inhibin and activin locally regulate rat ovarian folliculogenesis. Endocrinology 1990; 127:3196-3205. Erickson, GF, Kokka S, Rivier C. Activin causes premature superovulation. Endocrinology 1995; 136:4804-4813. Doi M, Igarashi M, Hasegawa Y, Eto Y, Shibai H, Miura T, Ibuki Y. In vivo action of activin-A on pituitary-gonadal system. Endocrinology 1992; 130:139-144. Li R, Phillips DM, Mather J. Activin promotes ovarian follicle development in vitro. Endocrinology 1995; 136:849-856. Nakatani A, Shimasaki S, DePaolo LV, Erickson GE Ling N. Cyclic changes in follistatin messenger ribonucleic acid and its protein in the rat ovary during the estrous cycle. Endocrinology 1991; 129:603-611. Suva LJ, Harm SC, Gardner RM, Thiede MA. In vivo regulation of Zif268 messenger RNA expression 173-estradiol in the rat uterus. Mol Endocrinol 1991; 5:829-835. Roy SK, Kole AR. Transforming growth factor-,3 receptor type II expression in the hamster ovary: cellular site(s), biochemical properties, and hormonal regulation. Endocrinology 1995; 136:4610-4620. Martin MM, Bradley A. Structure of the mouse activin receptor type II gene. Biochem Biophys Res Commun 1992; 185:404-413. Shukovski D, Keren-Tal I, Dantes A, Amsterdam A. Regulation of follistatin messenger ribonucleic acid in steroidogenic rat granulosa cell lines. Endocrinology 1995; 136:2889-2895. Rivier C, Vale W. Effect of recombinant inhibin on follicle-stimulating hormone secretion by the female rat: interaction with a gonadotropinreleasing hormone antagonist and estrogen. Endocrinology 1991; 129: 2160-2165.
