UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Autocrine/paracrine regulation of steroidogenesis in human ovarian cells Li, Wei 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1994-894253.pdf [ 4.13MB ]
Metadata
JSON: 831-1.0088168.json
JSON-LD: 831-1.0088168-ld.json
RDF/XML (Pretty): 831-1.0088168-rdf.xml
RDF/JSON: 831-1.0088168-rdf.json
Turtle: 831-1.0088168-turtle.txt
N-Triples: 831-1.0088168-rdf-ntriples.txt
Original Record: 831-1.0088168-source.json
Full Text
831-1.0088168-fulltext.txt
Citation
831-1.0088168.ris

Full Text

AUTOCRINE/PARACRINE REGULATION OF STEROIDOGENESIS IN HUMAN OVARIAN CELLS by Wei Li B.M., Inner Mongolia Medical College, 1983 M.Sc., Beijing Agricultural University, 1985  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUTREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in  THE FACULTY OF GRADUATE STUDIES Reproductive & Developmental Sciences Program Department of Obstetrics & Gynecology University of British Columbia We accept this thesis as confirming to  dar ed/  THE  LUMBIA July 1993 Wei Li, 1993  In  presenting  this thesis  in  partial  fulfilment  of the  requirements  for an  advanced  degree at the University of British Columbia, agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department  of  The University of British Columbia Vancouver, Canada  Date /VO 3V (?  DE-6 (2/88)  ABSTRACT This study investigated the hypothesis that local ovarian substances regulate gonadotropin-dependent steroidogenesis in the human ovary and form a paracrine and autocrine network for granulosa cell responses to endocrine stimulation. The specific substances investigated were inhibin, activin, follistatin, angiotensin H and Ill (A II and A III), prostaglandin F (PGF) and GnRH. Inhibin-related peptides were isolated in follicular fluid and were stimulated by FSH.  Inhibin-A had no effect on basal and  gonadotropin-stimulated progesterone production in human granulosa-luteal cells. Activin-A inhibited hCG/FSH- stimulated but not basal progesterone production, suggesting its role as an inhibitor of luteinization.  Follistatin stimulated basal  progesterone production, suggesting its antagonistic role for activin-A.  Intracellular  cAMP levels were inhibited by activin and stimulated by foffistatin. Follistatin may induce the preovulatory progesterone peak which is necessary to trigger the LH surge. Activin-A and follistatin form a balanced autocrine network to regulate proper progesterone production from granulosa cells around the ovulatory phase. Inhibin BAsubunit mRNA expression was increased by hCG treatment in cultured human granulosa cells, suggesting inhibin and activin production are subject to endocrine regulation. For comparison with another steroidogenic tissue, inhibin subunit mRNA expression in human trophoblast cells was also studied. GnRH and cAMP stimulated inhibin a and BB mRNA expression in human term placental cells. cAMP also stimulated inhibin BA mRNA expression. These results provide evidence for the interaction of local regulators and gonadotropins within the human ovary and placenta.  11  A renin-angiotensin system exists in human ovary and was stimulated by LH. Effects of A II and A Ill on steroidogenesis of granulosa-luteal cells were examined. A II and A ifi bound to the granulosa-luteal cell surface. A Ill but not A II inhibited hCG stimulated progesterone production.  A III had an additive effect on PMA, a  diacylglycerol analog, inhibited progesterone production response to hCG and PMA pretreatment did not block the inhibitory action of A Ill. These suggested that inhibitory effect of A III was not modulated by protein kinase C pathway. A ifi may play a luteolytic role during the luteal phase. Effects of GnRH and PGF on hCG-stimulated progesterone production were studied. GnRH had no effect on basal and hCG-stimulated progesterone production. PGF failed to have consistent effects on basal and hCG- stimulated progesterone production. These results are different from those observed in the rat ovary and may be due to species difference in the cellular response to PGF. Taken together, the results of this study showed that ovarian local regulators play key roles in the control of steroidogenesis by granulosa-luteal cells. The physiological significance of the actions of these local regulators might be that they interact with each other to fine-tune the endocrine response of the ovary to the robust actions of the gonadotropins.  111  TABLE OF CONTENTS ABSTRACT  .  ii  LIST OF TABLES  xv  LIST OF FIGURES  xvi  LIST OF ABBREVIATIONS  xx  PREFACE  xxiii  ACKNOWLEDGEMENTS  xxvi  1.0  INTRODUCTION AND LITERATURE REVIEW  1  1.1  INTRODUCTION  1  1.2  OVARIAN HISTOLOGY  2  1.3  1.2.1  Follicular growth and atresia  3  1.2.2  Morphology of granulosa cells  5 5  OVARIAN PRODUCTS  1.3.2  5  Steroid biosynthesis  1.3.1 1.3.1.1  Estrogen biosynthesis  6  1.3.1.2  Androgen biosynthesis  7  1.3.1.3  Progestin biosynthesis  8  Biosynthesis of Local nonsteroidal substance 1.3.2.1  Biosynthesis of inhibin-related peptides 1.3.2.1.1  10  Isolation and characterization of inhibin related peptides  1.3.2.1.2  9  Regulation  iv  of biosynthesis  10 of inhibin-  related  .  Biosynthesis of intraovarian growth factors  1.3.2.2  1.3.2.2.1  Insulin-like growth factor-I (IGF-I)  1.3.2.2.2  Epiderinal growth factor (EGF) and  ..  ..  11 13 13  transforming growth factor-a (TGF 14  a) 1.3.2.2.3  Transforming  growth  factor-fl 14  (TGF-J31) 1.3.2.2.4  Basic  fibroblast  growth  factor 14  (bFGF) 1.3.2.3  Biosynthesis of renin-angiotensins  1.3.2.4  Biosynthesis  1.3.2.5 1.4  of  prostaglandins  1.4.2  and  leukotrienes  15  Biosynthesis of GnRH-like peptides  16 18  Ovarian regulation 1.4.1  15  Neuroendocrine regulation of the ovary  18  1.4.1.1  GnRH  19  1.4.1.2  Prolactin (PRL) regulating factors  19  Gonadotropic hormone regulation of the ovary  20  1.4.2.1  Chemistry of gonadotropic hormones  20  1.4.2.2  Role of gonadotropin hormones  22  1.4.2.2.1  Role of FSH  V  22  1.4.3  1.4.4  1.4.2.2.2  Role of LH  23  1.4.2.2.3  Role ofprolactin (PRL)  25 26  Role of steroid Hormones 1.4.3.1  Role of estrogen  26  1.4.3.2  Role of progestins  28  1.4.3.3  Role of androgen  29 31  Role of nonsteroidal local regulators 1.4.4.1  1.4.4.2  1.4.4.3  Role of gonadotropin-releasing hormone (GnRH)  30  Role of inhibin-related peptides  34  1.4.4.2.1  Local action of inhibin  34  1.4.4.2.2  Local action of activin  35  1.4.4.2.2  Local action offollistatin  36  Role of intraovarian growth factors  37  1.4.4.3.1  Role of EGF and TGF a  37  1.4.4.3.2  Role of insulin and IGFs  39  1.4.4.3.3  Role of bFGF and TGFJ3  39 40  1.4.4.4  Role of ovarian renin-angiotensin .system..  1.4.4.5  Role of intraovarian prostaglandins  42  1.4.4.6  Intracellular signalling in the ovary  44  1.4.4.6.1  ..  Mechanism of transduction by steroid hormone receptors  vi  46  1.4.4.6.2  Adenylate cyclase-cAMP pathway.... 46  1.4.4.6.3  Calcium  and protein kinase  C 47  pathway 1.5  LOCALIZATION, SECRETION AND ACTION OF INHIBIN IN 49  HUMAN PLACENTA GENERAL OBJECTIVES  51  BACKGROUND AND RATIONALE  51  HYPOTHESIS  52  SPECIFIC OBJECTIVES  52  2.0  2.1  HUMAN GRANULOSA CELL PREPARATION  2.2  HUMAN PLACENTA CELL PREPARATION AND HORMONE  2.3  3.0  53  GENERAL MATERIALS AND METHODS  53  TREATMENT  54  HORMONE ANALYSIS  55  2.3.1  Radioimmunoassay (RIA) for steroids  55  2.3.2  Radioimmunoassay (RIA) for cAMP  57  2.3.3  Statistical analysis  58  EFFECT OF INHIBIN AND ACTIVIN ON PROGESTERONE AND 20-a-OH-P PRODUCTION IN HUMAN GRANULOSA CELLS  60  3.1  INTRODUCTION  60  3.2  TREATMENTS  61  3.2.1  Activin  A-,  inhibin  vii  A-  and  gonadotropin-regulated  61  steroidogenesis 3.2.2  Dose-dependency and time-course study of activin A on 61  hCG stimulated steroidogenesis 3.2.3  3.2.4 3.3  Effects of activin A on cAMP-stimulated progesterone production  61  Effects of activin A on cellular cAMP accumulation  62 62  RESULTS 3.3.1  Effect of activin A and inhibin A on gonadotropin 62  stimulated progesterone and 20o-OH-P accumulation 3.3.2  Dose- and time-dependent inhibition of progesterone and 62  2Oo-OH-P production by activin A 3.3.3  3.3.4 3.4 4.0  Activin A- and cAMP-induced progesterone and 20a-OH-P in accumulation  67  Effect of activin A on cellular cAMP accumulation  67 71  DISCUSSION  STIMULATION  OF  PROGESTERONE  SECRETION  BY  RECOMBINANT FOLLISTATIN-288 IN HUMAN GRANTJLOSA 74  CELLS 4.1  INTRODUCTION  74  4.2  TREATMENTS  75  4.2.1  Recombinant expression of human follistatin-288  75  4.2.2  Follistatin-induced steroidogenesis  75  viii  4.2.3  Dose-dependency and time-course study of follistatin 75  action Effects of follistatin on cAMP-stimulated progesterone  4.2.4  production  75  Effects offollistatin on cellular cAMP accumulation  76  4.2.5 4.3  76  Results 4.3.1  Dose-dependent increase of progesterone secretion by 76  follistatin 4.3.2  Effect of follistatin on basal and hCG-stimulated progesterone 76  production 4.3.3  Dose-dependent increase in 2Oct-OH-P production by 79  follistatin 4.3.4  Effect  of follistatin  on  hCG-stimulated  20a-OH-P 79  production 4.4.5  Time-dependent increase of progesterone production by 79  follistatin 4.4.6  4.4.7 4.5 5.0  Effect offollistatin on cAMP-stimulated progesterone and 20-cc-OH-P production  79  Effect of Follistatin on cellular cAMP accumulation  82 85  Discussion  EFFECT OF ANGIOTENSIN III ON PROGESTERONE PRODUCTION  BY HUMAN GRANULOSA CELLS  ix  88  5.1  INTRODUCTION  88  5.2  TREATMENTS  89  5.2.1  A II and A III conjugation and binding  5.2.2  Effects  of A  II  and  A  III  on  89 hCG-stimulated 90  steroidogenesis 5.2.3  Dose-dependency and time-course study of A III on hCG 90  stimulated steroidogenesis 5.2.4  Effects  of A III on cAMP-stimulated progesterone 90  production 5.2.5  Effects ofPMA on hCG- and cAMP-stimulatedprogesterone 91  production 5.2.6  Dose-dependent  effect  of PMA  on  hCG-stimulated 91  progesterone production 5.2.7  Effect of A III on hCG-stimulated progesterone production 91  in cells pre-treated with PMA 5.2.8  Additive effect of PMA and A III on hCG-stimulated 91  progesterone production 5.3  91  RESULTS  91  5.3.1  A II and A III binding to cultured human granulosa cells  ..  5.3.2  Effect of A III on hCG-stimulated progestin accumulation  ..  5.3.3  Dose- and time-dependent effects of A III  92  5.3.4  Effect of PMA on progesterone production  96  x  92  5.4 6.0  5.3.5  cAMP-stimulated progesterone production  96  5.3.6  Effect of A III in cells pre-treated with PMA  96 103  DISCUSSION  EFFECT OF GnRH ON PROGESTERONE PRODUCTION BY HUMAN GRANULOSA CELLS  108  6.1 INTRODUCTION  108  6.2  109  TREATMENTS 6.2.1  GnRH,  GnRH-agonist,  steroidogenesis 6.3  gonadotropin-regulated  and  .  109  RESULTS 6.3.1  Effect of GnRH on hCG- and cAMP-stimulated progesterone 109  accumulation 6.3.2  Effect of GnRH-agonist on hCG- and cAMP-stimulated progestin 110  accumulation 6.4 7.0  113  DISCUSSION  EFFECTS  109  OF  PROSTAGLANDIN  F  ON  PROGESTERONE  PRODUCTION BY HUMAN GRANULOSA CELLS  114  7.1  INTRODUCTION  114  7.2  TREATMENTS  115  7.2.1  , 2 PGF  -agonist, 2 PGF  and  gonadotropin-regulated 115  steroidogenesis 7.2.2  Dose-dependency  of  xi  2 PGF  on  hCG-stimulated  steroidogenesis 7.3  .  115  Results 7.3.1  agonists on gonadotropin PGF © and 2 2 Effects of PGF 115  stimulated progesterone accumulation 7.3.2  Dose-dependent inhibition of progesterone production by 116  2 PGF 7.4 8.0  115  120  Discussion  HORMONAL REGULATION OF INHIBIN MESSENGER RNA IN HUMAN PLACENTAL CELLS  121  8.1  INTRODUCTION  121  8.2  MATERIALS AND METHODS  123  8.2.1  Total RJ’JA extractions  123  8.2.2  Northern blot analysis 8.2.2.1  8.2.3  ..  123 123  Preparation of plasmid vector DNA 8.2.2.1.1  Transformation  123  8.2.2.1.2  Small scale DNA preparation  123  8.2.2.1.3  Large scale plasmid preparation  ..  124  8.2.2.2  Preparation of specific hybridization probe  125  8.2.2.3  Northern analysis using specific probe  126  8.2.2.4  Hybridization/Washing  ..  Quantitative RT-PCR 8.2.3.1  Quantitation ofInhibin x,fiA andJiB subunit  xii  126 127  mRNA  8.3  Synthesis offirst strand cDNA from mRl’JA  127  8.2.3.3  Primers of inhibin subunits  128  8.2.3.4  Quantitative PCR  128  8.2.3.5  Statistical Analysis  130 130  8.3.1  Expression of human inhibin subunit mRNA  8.3.2  Inhibin cc,JJA and JiB mRNA expression in human placental  Stimulation of inhibin subunit mRNA expression by 8-Br-cAMP and 139  GnRH  9.0  130  134  cells  8.4  127  8.2.3.2  RESULTS  8.3.3  .  139  DISCUSSION  HORMONAL REGULATION OF INHIBIN MESSENGER RNA N HUMAN GRANULOSA CELLS  143  9.1  INTRODUCTION  143  9.2  MATERIALS AN]) METHODS  143  9.3  9.2.1  Human granulosa cell treatment  144  9.2.2  Total RJ’JA extraction  144  9.2.3  Quantification of Inhibin x, f.M and fiB subunit mRNA  144  9.2.4  Statistical Analysis  145 145  RESULTS 9.3.1  Inhibin  x, f.M and fiB mRNA expression in human  xlii  145  granulosa cells 9.3.2  Stimulation of inhibin subunit mRI’JA expression by 8-Br149  cAMP and hCG 9.4 10.  152  DISCUSSION  155  GENERAL SUMMARY  164  11 .FUTURE EXPERIMENTS 11.1  ACTIVIN-A AND FOLLISTATIN  164  11.1.1 Interaction of activin-A and follistatin  164  11.1.2 Regulation of inhibin subunit mRNA expression by FSH and 164  follistatin 11.2  EFFECT OF All ON BASAL AND hCG-STIMULATED ESTRADIOL PRODUCTION AND MECHANISMS OF INHIBITORY EFFECT OF A 165  III 11.2.1 Effects of A II on basal and hCG-stimulated  estradiol 165  production  11.2.2 Mechanisms of inhibitory effect of A III on hCG-stimulated progesterone production 11.3  hCG-REGULAThD INHIBIN SUBUNIT mRNA EXPRESSION IN HUMAN PLACENTAL CELLS  12.  166  166 167  REFERENCES  xiv  LIST OF TABLES Table 1.  Local regulatory ovarian substances  table 2.  FSH-stimulated functional parameters in cultured granulosa cells  Table 3.  LH-modulated functional parameters in cultured rat granulosa  10 .  21  cells  25  Table 4.  Direct effects of estrogens on follicular granulosa cells  27  Table 5.  Direct stimulatory effect of GnRH on granulosa cell functions and 33  reproductive processes  xv  LIST OF FIGURES Figure 1.  The structure of human ovary: showing the process of follicular development and corpus luteum formation  2  Figure 2.  The complete follicular growth trajectory  5  Figure 3.  The scheme for synthesis of progestin, estrogen and androgens.  6  Figure 4.  Prostaglandins and arachidonic acid metabolic cascade  Figure 5.  The  two-cell/two- gonadotropin  of  hypothesis  17  follicular  estrogen 22  production Figure 6.  Mechanism  of  intracellular  signalling  transduction  by  peptide 44  hormones Figure 7.  Effect of inhibin and activin on hCG- or FSH-stimulated progesterone production by human granulosa cells  Figure 8.  Effect of inhibin and activin on hCG-, FSH- and 8-Br-cAMP stimulated 20a-OH-P production by human granulosa cells  Figure 9.  65  Time-course effect of activin A (lOOng/mi) on hCG-stimulated progesterone production by human granulosa cells  Figure 11.  Effect of activin on  66  8-Br-cAMP- and forskolin-stimulated  progesterone production by human granulosa-luteal cells Figure 12.  64  Effect of increasing doses of activin A on hCG (1 IU/ml)-stimulated progesterone production  Figure 10.  63  68  Effect of activin A (100 ng/ml) on hCG-stimulated cellular cAMP production, in the absence or presence of IBMX (0.5 mM)  xvi  69  Figure 13.  Effect of activin A (100 ng/ml) on hCG-stimulated progesterone production, in the absence or presence of IBMX (0.5 mM)  70  Figure 14.  Effect of follistatin (0.1-3 nM) on basal progesterone secretion.  77  Figure 15.  Effect  of  on  follistatin  basal  and  hCG/cAMP-stimulated  progesterone secretion by human granulosa cells  78  Figure 16.  Effect of follistatin (0.1-3 nM) on basal 20a-OH-P production.  80  Figure 17.  Effect of follistatin on basal and hCG/cAMP-stimulated 20x-OH-P production by human granulosa cells  Figure 18.  Time-course effect of follistatin (3 nM) on basal progesterone 83  production Figure 19.  81  Effect of follistatin (3 nM) and hCG (1 RI/mi) on progesterone secretion (upper panel) and cellular cAMP accumulation (lower panel) in the presence of iso-butyl-methyxanthine (IBMX, 0.5 84  mM) Figure 20.  The binding sites of FITC-labelled A II on human granulosa-luteal 93  cells Figure 21.  The binding sites of FITC-labelled A Ill on human granulosa-luteal 94  cells Figure 22.  Effects of A WA III on basal and hCG-stimulated progesterone production by human granulosa cells  95  Figure 23.  Effects of A ifi on hCG-stimulated 20cc-OH-P product  97  Figure 24.  Effects of A ifi (10w  -  l0 M) on hCG-stimulated progesterone  xvii  98  production Figure 25.  Temporal profile of the effects of A III (10-i M) on hCG-stimulated progesterone production by human granulosa cells  Figure 26.  Effects of PMA (1O M) on hCG-stimulated progesterone production by human granulosa cells 106 M) on basal progesterone production  Figure 27.  Effects of PMA (10’°  Figure 28.  -10 6 (10 M ) Effects of A III 5  -  +  Effects of A III and PMA on cAMP-stimulated progesterone  105  Effect of GnRH on basal and hCG-stimulated progesterone production by human granulosa cells  Figure 32.  104  Effects of A ifi (10 M) on hCG-stimulated progesterone production by PMA pre-treated cells  Figure 31.  101  102  production by human granulosa cells Figure 30.  100  PMA (l0 M) on hCG-stimulated  progesterone production Figure 29.  99  111  Effect of GnRH agonist on basal and hCG/cAMP- stimulated progesterone production by human granulosa cells  112  Figure 33.  Effect of PGF a on hCG/cAMP-stimulated progesterone production 117 2  Figure 34.  Effect of Cloprostenol and lutalyse on basal and hCG/cAMP stimulated progesterone production  Figure 35.  Effect of PGF (10-10 M) on hCG (1 UJ/ml)-stimulated 119  progesterone production Figure 36.  118  PCR amplification of inhibin x-, BA-, and BB-subunit mRNA from 129  cultured placental cells  xviii  Figure 37.  Northern analysis of inhibin cx, BA and BB mRNA from term placental cells treated with or without 8-Br-cAMP and GnRH  133  Figure 38.  Influence of PCR cycle number on product formation  136  Figure 39.  Quantitative analysis of inhibin cx-, BA-, and BB-subunit mRNA levels in 8-bromo cyclic AMP (cAMP) and GnRH treated placental 138  cells Figure 40.  PCR assay of inhibin cx-, BA-, and BB-subunit mRNA in cultured human granulosa cell RNA and Southern blot analysis of PCR 146  products Figure 41.  Different amplification cycles of PCR products of inhibin subunits using human granulosa-luteal cell total RNA  Figure 42.  148  Quantitative analysis of inhibin cx-, BA-, and BB-subunit mRNA levels in 8-Br-cAMP (0.15 mM) and hCG (1 lU/mi) treated human 151  granulosa-luteal cells Figure 43.  A model, suggested by the results, for the interaction of inhibin related peptides and gonadotropins in the regulation of intracellular cAMP production, progestin synthesis and inhibin subunit mRNA transcription by human granulosa-luteal cells  Figure 44.  158  A model, suggested by the results, for the interaction of angiotensins, hCG and PKC in the regulation of progesterone production by human granulosa 160  luteal cells  xix  LIST OF ABBREVIATIONS adenosine triphosphate (ATP) angiotensin I (A I), II (A II) and Ill (A III) arachidonic acid (AA) basic fibroblast growth factor (bFGF) cAMP-responsive element (CRE) 8-bromo-cyclic AMP (8-Br-cAMP) cyclic adenosine monophosphate (cAMP) diacylglycerol (DAG) dihydrotesteosterone (DHT) EDTA (ethylenediaminetetraacetate) epidermal growth factor (EGF) equine chorionic gonadotropin (ECG) fluorescein isothiocyanate (FITC) follicle-stimulating hormone (FSH) germinal vesicle breakdown (GVBD) gonadotropin releasing hormone (GnRH) human chorionic gonadotropin (hCG) human menopause gonadotropin (hMG) hydroperoxyeicosateira-enoic acids (HPETEs) hydroxyeicosatetraenoic acids (HETEs) 20a-hydroxyprogesterone (20x-OH-P)  xx  38-hydroxysteroid dehydrogenase (3B-HSD) ) 3 inositol( 1 ,4,5)trisphosphate (Ins( 1 ,4,5)P iso-butyl-methyixanthine (IBMX) insulin-like growth factor I (IGF-I) LB medium (Luria-Bertani medium leukotrienes (LTs) low-density lipoprotein (LDL) luteinizing hormone (LH) messenger ribonucleic acid (mRNA) 3-(N-morpholino) propanesulfonic acid (MOPS) nordihydroguaiaretic acid (NDGA) phorbal 12-myristate 13-acetate (PMA) phospholipase C (PLC) polyethylene glycol (PEG) prolactin (PRL) prostaglandins (PGs) ) 2 2 (PGE prostaglandin E ) 2 prostagladin F (PGF  prostaglandin  12  ) 2 (PGI  protein kinase C (PKC) radioimmunoassay (RIA) reverse transcription-PCR (RT-PCR)  xxi  SDS (sodium didecyl sulfate) somatomedin C (SmC) SSC (sodium chloride/sodium citrate buffer) SSPE (sodium chloride/sodium phosphate and EDTA buffer TE buffer (tris-EDTA) 2 (TXA ) 2 thromboxane A thyroid-stimulating hormone (TSH) thyrotropin releasing hormone (TRH) transforming growth factor B (TGF-B) trichloroacetic acid (TCA)  xxii  PREFACE  Much of the material contained in this thesis has been previously published by the author, or is in preparation for submission to be published. Full details of authorship follow;  1. Wei Li, Eui-Bae Jeung, Basil Ho Yuen, P. C. K. Leung. Regulation of inhibin gene expression in human term placenta by GnRH and CAMP. Molecular Endocrinology 1993. in preparation.  2. Wei Li, Basil Ho Yuen, P. C. K. Leung. Effect of angiotensin III on progesterone production by human granulosa cells. Endocrinology 1993, in preparation.  3. Wei Li, S Khorasheh, Basil Ho Yuen, Nicholas Ling, P. C. K. Leung. Stimulation of progesterone secretion by recombinant follistatin-288 in human granulosa cells. Endocrinology 1993, 124: 1750-1756  4. Wei Li, Basil Ho Yuen and P. C. K. Leung. Inhibition of progesterone accumulation by activin-A in human granulosa cells. Journal of Clinical Endocrinology & Metabolism 1992. 75:285-289  5. W. David Currie, Wei Li, K.G. Baimbridge, Basil Ho Yuen, P.C.K. Leung. Cytosolic  xxiii  free calcium increased by prostaglandin F2x(PGF2a), gonadotropin-releasing hormone, and angiotensin II in rat granulosa cells and PGF2(x inhuman granulosa cells. Endocrinology 1992. 130:1837-1843  6. Wei Li, Basil Ho-yuen and P. C. K. Leung. Effect of recombinant human follistatin on progesterone production by human granulosa cells in vitro. (Abstract) International Conference on Reproductive Endocrinology. Beijing.  Second  September 15-18,  1992.  7. Wei Li, Basil Ho Yuen, P. C. K. Leung. Effect and mechanism of activin-A on progesterone production by human granulosa cells. (Abstract)  The Society of  Obstetricians and Gynecologists of Canada, Vancouver. June 15, 1992.  8.  Wei Li, Basil Ho Yuen, P. C. K. Leung.  (Abstract)  Effect of activin-A on  progesterone production by human granulosa cells. Serono Symposium Molecular Basis of Reproductive Endocrinology, Vancouver, July 25-27, 1991.  9.  Wei Li and Chi-Yu Gregory Lee.  Monoclonal antibodies which inhibit sperm  acrosome reaction and fertilization. (Abstract) International Conference on Advances in Obstetrics and Gynaecology. Taipei, December 2-6, 1990.  10. Wei Li, Ming-sun Liu and Gregory Lee. Monoclonal antibodies against human  xxiv  sperm antigens and their effects on cAMP-induced acrosome reaction. (Abstract) Canada West Society for Reproductive Biology. The 12th. Annual Winter Workshop. Vancouver, February 15-17, 1990.  xxv  A CKNOWLEDGEMENTS I am grateful to Drs. Leung and Ho Yuen (co-supervisor) for their support, advice and outstanding generosity during my four year study. I thank the other members of my Supervisory Committee (Dr. M. Hayden, Dr. R. Rajamahendran, Dr. D. Rurak) for their advise and support. I appreciate Dave Currie’s assistance and advice on the preparation of the thesis. I appreciate the technical assistance and good times Eui-Bae Jeung and Shideh Khorasheh have given me. Thanks to my parents for their encouragement and support in my study. Finally, thanks to my husband Hao Jing and son Tian Jing for making this all worthwhile.  xxvi  1.0  INTRODUCTION AND LITERATURE REVIEW  1.1  INTRODUCTION Postpubertal human ovaries are endocrine tissues in which many compartments  interact to produce a fertilizable ovum and secrete sex steroids during the menstrual cycle. Ovarian functions are regulated by gonadotropins secreted by the anterior pituitary. However, changes in the blood level of gonadotropins cannot account for some phenomena occuring during the menstrual cycle. An example would be the occurrence of ovulation in only one of the two ovaries despite perfusion by identical circulating gonadotropin concentrations. Not all follicles mature to the same extent during a given cycle within the ovary from which ovulation occurs. After ovulation, only the dominant follicle is thought to fonti the corpus luteum.  The mechanisms of maintenance and  luteolysis of corpus luteum remain unclear. Many substances identified in the ovaries of laboratory mammals have local regulatory actions.  These substances may form an  elaborate network of highly integrated paracrine and autocrine systems, although this remains to be proven. Current information suggests that the ovary may play an active role in the initiation and maintenance of reproductive cyclicity. The hypothalamus and pituitary play a permissive and tonic role in this connection. Intra-follicular substances may respond to gonadotropins to integrate a proper biological response by granulosa cells. Therefore, intraovarian substances may act locally to modify gonadotropin functions in the ovary, controlling follicular development and steroidogenesis. The studies of this thesis addressed this hypothesis using human granulosa-lutein cell cultures to study the regulation of sex steroid secretion and transcription of messenger ribonucleic acid  1  (mRNA) coding for some local substances.  1.2  OVARIAN HISTOLOGY The human ovary is divided into a peripheral cortex and an inner medulla. The  cortex contains three parts: 1. The germinal epithelium, which is the outermost cellular layer of the cortex; 2. the stroma, containing connective tissue and interstitial cells; 3. the follicular complex.  The medulla contains the loose connective tissue, nerves, lymph  vessels, blood vessels and interstitial cells. The follicular complex contains follicles and corpora lutea in various stages of development (Fig. 1).  Figure 1.  The structure of human ovary: showing the follicular development and corpus luteum formation. (Modified from Yen S S C & Jaffe R B, Reproductive Endocrinology, 3rd ed. Saunders W B Company 1990).  2  1.2.1  Follicular growth and atresia The follicle is a basic unit of the ovary. An ovarian follicle is composed of the  oocyte associated with a layer or layers of granulosa cells. The oocyte and granulosa cells are separated from the surrounding stroma by a basement membrane, the “basal lamina.” Blood vessels and theca cells are outside the basal membrane. The theca cells are spindle-shaped cells transformed from interstitial cells. Granulosa cell proliferation and differentiation, theca cell hyperirophy and oocyte growth all combine to increase follicular diameter (McNatty, 1981).  Localized fluid appears among the  granulosa cells when the follicular diameter reaches 100-200 microns.  This process  eventually transforms the primary follicle into an antral follicle, then a preovulatory follicle.  This process only occurs during the reproductive lifespan (Harrison 1962;  Lintern-Moore et al., 1974). Antral fluid contains protein-bound and free sex steroids, local substances, plasma proteins, mucopolysaccharides and electrolytes (Edwards 1974). These components aie secreted by granulosa and theca cells or diffuse from vascular spaces outside the basal membrane.  One follicle ovulates and gives rise to a corpus  luteum during each menstrual cycle. Other follicles undergo atresia. More than 99 per cent of follicles are lost by atresia (Ingram, 1962). The functional lifespan of the human corpus luteum is approximately 14± 2 days. The granulosa cells undergo morphological changes (called luteinization) with capillary and theca cell penetration to form the corpus luteum (Harrison, 1962, Fig. 1). Follicular development is separated into three phases: 1)  The preantral growth phase concerns the conversion of primordial follicles to primary follicles. This phase is continuously operational from the fifth month of  3  fetal life to the end of adult life and occurs independent of gonadotropic support. The mechanisms responsible for recruiting some primordial follicles remain a mystery. 2)  The tonic growth phase concerns the conversion of primary fofficles to antral follicles. This phase may be follicle-stimulating hormone (FSH)-dependent.  3)  The gonadotropin-dependent growth phase concerns follicular growth through the antral to preovulatory stage and ovulation. This phase is gonadoiropin-dependent (Fig. 2).  Figure 2.  The complete follicular growth trajectory. M menses; Gn, gonadotropin. ,  Ovul, ovulation.  4  1.2.2  Morphology of granulosa cells The granulosa cells are from the intraovarian rete ovarii (Byskov & Rasmussen,  1973; Byskov, 1978). The differentiation of granulosa cells is not unifonn in a given follicle.  As the antral follicle develops, these cells become organized into  morphologically distinguishable regions with specialized functions. The mural granulosa cells are closer to the basal membrane, while the cumulus cells surround the oocyte. Cumulus cells physically support the oocyte and provide nutrients for oocyte growth. Cytoplasmic extensions from the cumulus cells penetrate the zona pellucida, allowing signal exchange for coordinated follicular and oocyte maturation.  The majority of  granulosa cells are mural, lining the follicular cavity. The mural cells are interconnected by extensive intercellular gap junctions, resulting in cellular electrical coupling to yield an integrated and functional syncytium (Adashi, 1991).  1.3  1.3.1  OVARIAN PRODUCTS  Steroid biosynthesis Ovaries have the capacity to synthesize all three classes of sex steroids (progestin,  estrogen and androgen) from their common precursor, cholesterol (Fig. 3). Cholesterol from both low-density lipoprotein and high-density lipoprotein has been demonstrated to serve as a precursor for steroidogenesis in the ovarian follicle (Gwynne & Strauss, 1982). Lipoprotein uptake from plasma is regulated by the availability of serum lipoproteins and the lipoprotein receptor-dependent uptake system. Cholesterol synthesis by granulosa cell  5  Figure 3.  The scheme for synthesis of progestin, estrogen and androgens.  has also been reported (Schuler et al., 1979; Turek et al., 1982). Estrone and estradiol are the major steroid products of the follicle. Progesterone and l7cx-hydroxyprogesterone are the major products of the corpus luteum. Studies using (labelled or unlabelled C 21 and 19 precursors showed that isolated granulosa cells produce progesterone and estrogens C ) as well as lRx-hydroxyprogesterone (Fowler et al., 1978; Channing 1969; Tsang et al., 1980).  Isolated theca cells produce progesterone, lRx-hydroxyprogesterone, and  androstenedione (Tsang et al., 1980; McNatty et al., 1979).  1.3.1.1  Estrogen biosynthesis  6  Granulosa cells are the cellular source of the two most important ovarian steroids, estradiol and progesterone.  Progesterone synthesis requires only granulosa cells but  estrogen synthesis requires cooperation between granulosa and theca cells. Androgens (mainly androstenedione) produced by luteinizing hormone (LH)-stimulated theca cells are the main substrate for estrogen biosynthesis by FSH-stimulated granulosa cells. FSH induces granulosa cell aromatase activity, which is absent in theca cells. Both FSH and LH action on steroidogenesis require membrane-associated adenyl cyclase (Knecht et al., 1983).  Adenyl cyclase converts intracellular adenosine triphosphate (ATP) to cyclic  adenosine monophosphate (cAMP). cAMP binds to the regulatory subunit of protein kinase A.  The catalytic subunit of protein kinase A is activated by cAMP and  phosphorylates intracellular proteins.  However, the exact proteins involved in the  phosphorylation remain unknown.  1.3.1.2  Androgen biosynthesis The rate-limiting step in androgen biosynthesis by theca cells is catalyzation of  -lyase enzyme complex. 1720 pregnenolone or progesterone by l7cx-hydroxylase/C enzyme complex is located in the microsomal fraction.  The  Comparatively speaking,  -lyase (Short, 1962; Bjersing & 1720 granulosa cells contain almost no 17a-hydroxylase/C Carstensen, 1967).  The lack of these enzymes in granulosa cells indicates that both  granulosa and theca cells participate in androgen and estrogen biosynthesis. The 17o17 position before the side chain is -lyase complex hydrolyses at the C 1720 hydroxylase/C 19 steroids (androgens). This is a key cleaved from the C 21 steroids (progestins) to form C  7  step in hormonal control of follicular steroid secretion.  Androgens synthesized by  theea/interstitial cells are transported to the follicular fluid and to the blood circulation as a precursor of estrogen synthesis.  1.3.1.3  Progestin biosynthesis Granulosa cells synthesize progestins using cholesterol derived from circulating  serum lipoproteins (Brown & Goldstein, 1976). Low-density lipoprotein (LDL) particles bind to specific membrane receptors. LDL-receptor complexes enter the cell by receptormediated endocytosis (Goldstein et al., 1979; Anderson et al., 1977). The endocytotic vesicles fuse to lysosomes and yield free cholesterol (Brown et al., 1975). The free cholesterol is either re-esterified and stored in lipid droplets or is transported to mitochondria for steroidogenesis.  Human follicular fluid contains little LDL but  cholesterol can be generated endogenously and used to synthesize progesterone (Simpson et at, 1980; Schuler et al., 1979; Turek & Strauss, 1982). The rate-limiting step in 21 compound, progestin synthesis is side-chain cleavage of cholesterol to pregnenolone, a C by side-chain cleavage cytochrome P-450 in the inner mitochondrial membrane. Pregnenolone is the key steroidogenic intermediate common to all steroids produced by follicles. Both granulosa and theca cells convert pregnenolone to progesterone by two enzymes, 3B-hydroxysteroid dehydrogenase (3B-HSD) and  ,  -isomerase (Samuels et 4  al., 1951; Cheatum & Warren, 1966). Isomerase activity appears excessive (Philpott & Peron, 1971; Dimino & Campbell, 1976), such that the production of progesterone from pregnenolone is mainly regulated by 3B-HSD. The secretion of progesterone by granulosa  8  cells may also be modulated by conversion of progesterone to 20x-hydroxyprogesterone (2Ooc-OH-P) by the 20x-hydroxysteroid dehydrogenase.  The concentration of 2Ocz-  hydroxysteroid dehydrogenase is high in corpora lutea undergoing luteolysis (Biersing, 1977; Lahav, 1977; Eckstein, 1977). 2Ox-OH-P is less active as a progestational agent than the precursor progesterone. As 20c&reduced steroids are poor substrates for 1720 C lyase, 20a-hydroxysteroid dehydrogenase may play a significant role in determining the available amount of -substrates 21 for androgen conversion. Granulosa and theca cells C provide precursors for androgen synthesis by theca/interstitial cells. Granulosa cells are the main source of progesterone during the early follicular phase, when few LH receptors are on the membrane of theca/intersthial cells.  1.3.2  Biosynthesis of Local nonsteroidal substance Granulosa cells secrete regulatory proteins, prostaglandins and catecholamines.  Table 1 lists some locally produced ovarian substances. Among intraovarian substances, inhibin-related peptides, prostaglandins and neuropeptides such as gonadotropin releasing hormone (GnRH) have been the subject of intense investigation. Most of these agents do not act in the traditional endocrine fashion. They may act locally to modulate the growth and function of somatic and germ cells. Potential intraovarian communication may take on one of the following configurations: 1.  Paracrine communication which involves actions of substances secreted from producer cells on neighbour target cells.  9  Table 1.  LOCAL REGULATORY OVARIAN SUBSTANCES  PEPTIDES  PEPTIDES  OTHERS  Inhibin  Angiotensins  Prostaglandins  Activin  Interleukins  Estradiol  Fibroblast growth factor  Follistatin  Progesterone  Transforming growth factor a/13  Insulin-like growth factor  Catecholamines  Mullerian Inhibiting Factor  Epidermal growth factor  Gonadotropin-releasing hormone  Tumor necrosis factor cx  2.  Autocrine communication which involves the actions of a substance on the cells of origin. Although ovarian cells produce a variety of nonsteroidal products, relatively little  attention has been given to this issue. remains a mystery. A selected few  The function of most of these local substances  are discussed below.  1.3.2.1  Biosynthesis of inhibin-related peptides  1.3.2.1.1  Isolation and characterization of inhibin-related peptides  Inhibin, activin  and follistatin have been isolated from follicular fluid of several  mammalian species. There are two forms of inhibin (A and B), inhibin A composed of a common cx subunit (MW. 18000) and one of two distinct B subunits, fA inhibin B composed of a  (MW. 14000);  and 3B subunit (MW. 14000). Activin also has two forms, one  10  consisting of a homodimer of BA subunit (Activin A) and another a heterodimer B subunit, BA and BB subunits (Activin AB). Activin B, the inhibin BB subunit homodimer, has not yet been isolated from natural sources. The protein family includes transforming growth factor B (TGF-B), mullerian inhibiting factor and several other members (Ling et al., 1986; Vale et al., 1986). Inhibin subunit structures show a high degree of inter species homology, with the a-subunits being 80% homologous among human, bovine, porcine, murine and ovine inhibin. The BA-subunit is 100% homologous, aside from one amino acid substitution in the ovine BA subunit, and the BB subunit is more than 95% homologous (Forage et al., 1987). The a, BA and BB subunit genes have been cloned (Mason et al., 1985) and recombinant inhibin A and activin A recently became available for investigation (Broxmeyer et al., 1988). Follistatin is a single-chain, glycosylated protein with at least three isoforms (molecular weights  32000, 35000 and 39000) which are structurally different from the  inhibins (Robertson et al., 1983; Ueno et al., 1987). Follistatin was first isolated from porcine and bovine follicular fluid (Robertson et al., 1987; Vale et aL, 1986). Natural follistatin suppresses FSH release in vitro with only 5-30% potency in comparison with inhibin (Robertson et aL, 1990).  Follistatin, inhibins and activins are products of  granulosa cells.  1.3.2.1.2  Regulation of biosynthesis of inhibin-related peptides  The major source of ovarian inhibin is the granulosa cell (Bicsak et al., 1986). Inhibin concentration was found to increase with follicular size and estradiol  11  concentrations (Saito et aL, 1991). Inhibin is also produced by the corpus luteum (Davis et al., 1987; Schwall et al., 1990). Cultured human luteinizing cells produced inhibin (Adashi, 1985). Inhibin a- and BA-subunit mRNAs were identified in rat and human corpora lutea (Mock et al., 1983). FSH is the regulator of inhibin production. Low doses of LH following FSH exposure cause inhibin release from granulosa cells (Bicsak et al., 1986; Zhiwen et al., 1988a). Higher doses of LHlhuman chorionic gonadotropin (hCG) inhibit FSH-induced inhibin production in vitro (Zhiwen et al., 1987) and in vivo (Lee, 1983). Several local ovarian substances also affect inhibin production by granulosa cells in vitro.  Insulin-like growth factor I (IGF-I) stimulated inhibin production by rat  granulosa cells in a time- and dose-dependent manner (Zhiwen, 1987). TGFB, an activin homolog, dose-dependently increased basal and FSH- stimulated inhibin production 2 (which increases (Zhiwen, 1988b). 8-bromo-cAMP (8-Br-cAMP) and prostaglandin E cellular adenyl cyclase activity) dose-dependently stimulated inhibin production, as did phosphodiesterase inhibitor and forskolin (Klein et al., 1991). Epidermal growth factor (EGF) inhibits both basal and FSH-stimulated inhibin production (Klein et aL, 1991; Kim et al., 1989; Flanders et aL, 1989). EGF also inhibits -stimulated inhibin production (Flanders et aL, 2 the 8-Br-cAMP- and prostaglandin E 1989). TGFa from theca/interstitial cells has an EGF-like activity in porcine granulosa cells (Hsu et al.). Inhibin production by granulosa cells can be regulated at a local level by other members of the inhibin/TGFB family.  Activin- stimulated and follistatin  suppressed inhibin production by rat granulosa cells in vitro have been reported (Sugino, 1988a; Xiao et aL, 1990).  12  Activin production likely occurs during latter stages of folliculogenesis. Schwall (1990) reported B subunit expression (without x subunit expression) in small antral fofficles in the primate ovary. However, little is known about regulation of activin production. The follistatin gene sequence has been deduced from porcine (Esch 1987) and murine (Shimasaki, 1989) eDNA clones. Northern blot analyses showed that ovarian follistatin mRNA transcription was stimulated by equine chorionic gonadotropin (ECG) treatment of immature rats but not affected by an ovulatory hCG dose (10 lU/rat) after ECG (Shimasaki, 1989). In situ hybridization studies showed that follistatin mRNA was low in primordial follicles but dramatically increased in granulosa cells of antral and preovulatory follicles. Follistatin mRNA was not detected in theca, stroma or interstitial cells and some corpora lutea showed slight hybridization. Granulosa cells, however, gave a strong positive signal indicative of the origin of follicular foffistatin. There are no reports on the regulation of follistatin production in vitro.  Follistatin production by  differentiated granulosa cells from preovulatory follicles is FSH- but not LH-dependent (Klein et al., 1991). Follistatin is an activin binding protein in the rat ovary (Nakamura et al., 1990; Saito et al., 1991).  1.3.2.2  Biosynthesis of intraovarian growth factors  1.3.2.2.1  Insulin-like growth factor-I (IGF-I)  IGF-I is a 70-amino acid polypeptide believed to play many metabolic roles. The  13  ovary is a site of IGF-I production, reception and action (Adashi et aL, 1985). The granulosa cell is the only cellular site of IGF-I gene expression. Granulosa and theca interstitial cells possess specific receptors for IGF-I (Adashi et al., 1985).  1.3.2.2.2  Epidermal growthfactor (EGF) and transforming growthfactor-a (TGF a)  EGF is a single polypeptide chain of 53 amino acids. In situ hybridization of tissue sections indicates that RNA complementary to cloned EGF probes may be present in many tissues. EGF membrane receptors exists on bovine, ovine and murine granulosa cells (Mock & Niswender, 1983). TGF-a, an EGF structural analog, is a single-chain, 50 amino acid polypeptide binding to a common EGF/TGF-a receptor.  TGF-a is  structurally the same as EGF (St-Arnaud et al., 1983), raising the possibility that theca interstitial cell-derived TGF-a may exert paracrine effects on adjacent granulosa cells.  1.3.2.2.3  Transforming growth factor-J)1 (TGF-J)1)  TGF-B1, a homodimeric polypeptide comprised of two identical 112 amino acid chains, is produced in the ovary (Kudilow et al., 1987; Skinner et aL, 1987). Ovarian theca-interstitial and granulosa cells may be sites of TGF-B1 production and action (Magoffin et al., 1989).  1.3.2.2.4  Basic fibroblast growth factor (bFGF)  bFGF, a 146-amino acid polypeptide, is a mitogen for a wide variety of tissues.  14  An N-terminal-truncated form lacking the first 15 residues has been identified in the corpus luteum (Gospodarowicz et aL, 1985). Recent studies indicate that theca-interstitial cells are also a site of bFGF reception and action (Adashi 1991).  1.3.2.3  Biosynthesis of renin-angiotensins The wide variety of tissues identified to have a renin-angiotensin system include  brain, adrenal gland, arterial wall, mesenteric vasculature, heart, testis, uterus and ovary (Weishaar et al., 1991; Sramek et al., 1988). Prorenin, renin-like activity and angiotensin II (A II) have been found in human follicular fluid at concentrations greater than in blood, suggesting that the ovary produces these substances (Itskovitz et al., 1988; Glorioso et al., 1986; Fernandez et al., 1985; Do et al., 1988). Circulating prorenin and renin concentrations were increased 10-fold in gonadotropin-stimulated women before ovulation (Culler et al., 1986).  Renin is an aspartyl protease, cleaving its hepatic substrate  angiotensinogen to form angiotensin I (A I). A I is converted to A II by angiotensin converting enzymes. The A II, in turn, is converted to angiotensin Ill (A III) by the same enzyme (Adashi, 1991).  1.3.2.4  Biosynthesis of prostaglandins and leukotrienes Prostaglandins (PGs) were discovered by Von Euler in the 1930s as active  components of human seminal fluid. They are important products of the ovarian cells and their secretion is under hormonal control (Triebwasser et al., 1978; Clark et al., 1978). 20 polyunsaturated fatty acid. C 4 The PG precursor is arachidonic acid (AA), which is a :  15  Esterification of cell membrane phospholipids in the 2-acyl position allows glycerol and AA to be released. The free AA concentration in cells is less than 10 M. The free acid 6 concentration in tissue represents a balance between acid release by hydrolysis and re esterification. Free AA undergoes two metabolitic pathways as outlined in Figure 3. The 2 (PGH ), 2 cyclooxygenase pathway forms the endoperoxide intermediate prostaglandin H 2 which is then converted by isomerase to biologically active molecules, prostaglandin E ), prostaglandin D 2 (PGE 2 (PGD ), prostaglandin F (PGF), prostaglandin 2  ‘2  ) and 2 (PGI  thromboxane A 2 (TXA ). The letters following the abbreviation PG indicate the position 2 of the oxygen-containing substitutes present in the cyclonptane ring. The 2-series PGs are formed from AA.  The 1-series and 3-series PGs are formed from 8,11,14-  eicosatrienoic and 5,8,11,14,1 7-eicosapentaenoic acid, respectively. Another metabolic pathway of AA is provided by lipoxygenase enzymes. The products of the lipoxygenase enzymes are hydroperoxyeicosatetra-enoic acids (HPETEs) which can then be converted into hydroxyeicosatetraenoic acids (HETEs), leukotrienes (LTs) and lipoxins (Fig. 4). hCG-induced PG production occurs in granulosa and theca cells of preovulatory follicles (Hedin et al., 1987). Cyclooxygenase, the enzyme which forms PGs from AA, is inhibited by nonsteroidal anti-inflammatory drugs such as aspirin and indomethacin (Flower & Vane, 1974).  Lipoxygenase activity is inhibited by compounds such as  nordihydroguaiaretic acid (NDGA) (Salari et al., 1984).  1.3.2.5  Biosynthesis of GnRH-like peplides  Peripheral circulating GnRH concentrations are insufficient to bind to ovarian  16  GnRH receptors (Aten, 1986). Direct physiological actions of GnRH at the ovary may  Figure 4.  Prostaglandins and arachidonic acid metabolic cascade.  depend upon production of GnRH-like substances by the ovary. Murine, bovine and  17  ovine ovaries contain a GnRH-like peptide that competes with GnRH for binding to ovarian membrane receptors but with immunoactivity distinctly different from hypothalamic GnRH (Aten, 1986, 1987). A separate gonadotropin-releasing peptide has been isolated from human follicular fluid (Li, 1987). The amino acid composition and sequence of ovarian GnRH (H-Thr-Asp-Thr-Ser-His-Asp-Gln-Asp-His-Pro-Thr-Phe-Asn ) differs from that of hypothalamic GnRH (Pyro-Glu-His-Trp-Ser-Tyr-Gly-Leu-Arg 2 NH Pro-Gly-NH2). Ovarian GnRH stimulates gonadotropin release from murine pituitary in vitro less potently than does hypothalamic GnRH. The presence of GnRH-like peptides might serve a paracrine or autocrine role in the ovary.  1.4  Ovarian regulation The menstrual cycle is a repetitive operation of the hypothalamic-pituitary-ovarian  axis associated with target tissue changes. The menstrual cycle is divided into follicular, ovulatory and luteal phases (Fig. 2). secretion from the hypothalamus.  The menstrual cycle is controlled by GnRH  GnRH is transported to the anterior pituitary cells  through the hypothalamic-hypophyseal portal system and stimulates gonadotropin secretion from the pituitary. Gonadotropins regulate the production of sex steroids and some local substances by the ovary. However, current information suggests that ovarian products play an active role in the regulation of ovarian function. The hypothalamus and pituitary may play a permissive role.  1.4.1  Neuroendocrine regulation of the ovary  18  GnRH and prolactin (PRL) regulating factors are secreted from the neurons in the anterior hypothalamus and arcuate nucleus.  GnRH and PRL regulating factors are  axonally transported to the median eminence. GnRH and PRL regulating factors are secreted in the median eminence, diffuse into the hypothalamic-pituitary portal vasculature and are delivered to the adenohypophyseal gonadotropes and lactotropes.  1.4.1.1  GnRH GnRH is a decapeptide secreted in a pulsatile manner into the hypothalamic-  pituitary portal circulation. Several neurotransmitters are involved in GnRH secretion. The amplitude and frequency of the pulses are regulated by catecholamines and neuropepetides.  These inputs to the GnRH neurons are affected by the feedback of  estradiol and progesterone (Yen, 1982). Estradiol and dlihydrotestosterone (DHT) are concentrated in the nuclei of catecholaminergic neurons; the target neurons for these two steroids are surrounded by catecholaminergic terminals (Heritage et al., 1980). GnRH binds to specific receptors on gonadotropes and stimulates LH and FSH secretion. Gonadotropin responses to GnRH are maximal when GnRH receptor numbers are highest. Increased GnRH concentration and pulse frequency increases GnRH receptor numbers, a process known as upregulation. GnRH receptors will be downregulated if continuous GnRH administration is substituted for pulsatile stimulation (Clayton & Catt 1981).  1.4.1.2  Prolactin (PRL) regulating factors Human pituitary lactotropes are tonically suppressed by hypothalamic factors.  19  Dopamine, a neurohormone produced in the arcuate nucleus, is an important inhibitor. Dopamine is secreted into the portal circulation in concentrations (0.7 ng/ml) sufficient to inhibit PRL release (Neill J et al., 1961). Dopamine binds to specific membrane receptors on lactotropes. Hypothalamic thyrotropin releasing hormone (TRH) has both PRL and thyroid-stimulating hormone (TSH) releasing function.  1.4.2  Gonadotropic hormone regulation of the ovary The control of gonadal function and reproduction in primates and other  mammalian species is exerted predominantly through three gonadotropin hormones secreted by the pituitary, FSH, LH and PRL  1.4.2.1  Chemistry of gonadotropic hormones FSH and LH have many chemical similarities and differ from PRL. FSH and LH  have two polypeptide subunits bound with a noncovalent, high affinity bond. subunit, designated  X,  One  is common to the glycoproteins LH, FSH, hCG and TSH (Ward,  1978; Jutisz & Tertrin-clary, 1974). The other subunit, designated B, differs in its amino acid sequence in each glycoprotein and carries the specific hormone activity.  The  molecular weight of LH and hCG is about 28,000 while that of FSH is 33,000.  The  genes encoding human LH and hCG 13-subunits have relatively little genetic drift and differ mainly in the expression of the C-terminal translated region. This sequence codes for an additional SER-rich region in hCG that attaches 0-linked carbohydrate chains that are not present in the LH molecule.  The  20  molecular  weight  of  PRL  is  approximately 22,000. PRL is a single polypeptide chain with no carbohydrate moieties. Three internal disuiphide bridges are incorporated into the secondary structure of PRL.  TABLE 2.  FSH-STIMULATED FUNCTIONAL PARAMETERS IN CULTURED GRANULOSA CELLS  Enhancement of steroidogenesis Estrogen biosynthesis a. Induction of aromatases b.  2.  Induction of specific plasma membrane receptors LH receptor formation a. Prolactin receptor formation b. B,,-adrenergic receptor formation (coupling) c. d. e. f.  3.  c.  Plasminogen activator Prostaglandins  d.  Proteoglycans (mucopolvsaccharides)  Stimulation of general functions a. b. c. d.  5.  Lipoprotein receptor formation FSH receptor formation EGF receptor formation  Secretion of nonsteroidal cell products Inhibin a. b.  4.  Progesterone and 20cc-OH-P biosynthesis Induction of cholesterol side-chain cleavage enzymes and mitochondnal cytochrome P450 activity Induction of 3B-hydroxysteroid dehydrogenase Induction of 20cc-bydroxysteroid debydrogenase  DNA synthesis Protein synthesis Glucose uptake and lactate formation Cell roundup and aggregation Gap junction formation Microvilli formation  e. f. Plasma membrane-related processes Adenyl cyclase activation and cAMP formation a. Formation of cAMP binding protein b. c.  Phosphodiesterase activation  21  U  Increases in plasma membrane microviscosity  d.  1.4.2.2  Role of gonadotropin hormones  1.4.2.2.1  Role of FSH  FSH induces ovarian follicular maturation and granulosa cell development. FSH stimulates the granulosa cells to produce estrogens, progestins and various nonsteroidal substances. The action of these granulosa cell products ensures optimal folliculogenesis and oocyte maturation. FSH receptors are present in the granulosa cells. The FSH regulated granulosa cell functions are summarized in Table 2. Based on rat studies, a two-cell, two-gonadotropin hypothesis for ovarian estrogen synthesis is summarized (Fig. 5).  According to this model, LH stimulates androgen  synthesis from cholesterol in theca-interna cells (Tsang et al., 1980; Makris & Ryan, 1975). Androgens diffuse across the lamina basalis and are converted to estrogen by  granulosa cells. FSH increases the granulosa cell aromatase activity in vitro and in vivo. FSH also stimulates progesterone production in cultured granulosa cells (Channing 1974; Veldhuis et al., 1980). FSH regulates granulosa cell progestin synthesis by increasing the activities  of  cholesterol  side-chain  cleavage  enzymes  and  38-hydroxysteroid  dehydrogenase (Jones et al.,1982). FSH increases LH, PRL and EGF receptor content in granulosa cell membrane. FSH increases the cholesterol utilization by increasing LDL receptor numbers in the cell  22  membrane. FSH may increase its own receptor numbers in granulosa cells (Amsterdam et at., 1981; White & Ojeda, 1981). FSH stimulates granulosa cell division in vivo and increases gap junction formation (Amsterdam et al., 1981). cAMP is the second messenger of FSH action in granulosa cells.  FSH stimulates cAMP formation, which activates cAMP-dependent  protein kinase A in granulosa cells (Richards et at., 1983; Conti et al., 1984).  Granulosa Cells  (Follrcula  -  rFo(rce  L  Fluid)  SI,rnolln  Hormone  Figure 5.  The two-cell/two-gonadotropin hypothesis of follicular estrogen production.  1.4.2.2.2  Role of LH  LH stimulates preovulatory follicular growth, induces ovulation and regulates  23  corpus luteum function. hCG has the same effects as LH because LH and hCG interact with the same receptor in luteal, granulosa and theca intema cells. LH effects on cultured granulosa-luteal cells are listed in Table 3. LH increases aromatase activity in cultured granulosa cells exposed to FSH (Wang et al., 1981; Zhuang et al., 1982). LII stimulates progesterone production from the corpus luteum in vivo, dissociated luteal cells and FSH primed granulosa cells in vitro (Zhuang et at, 1982; Schomberg et al., 1967; Mason et at, 1961).  LH stimulates progesterone production by increasing cholesterol esterase  activity and lipoprotein receptors, resulting in increased free intracellular cholesterol (Behrman & Armstrong, 1969; Caffrey et at, 1979). LII also increases the activities of side-chain cleavage and 38-hydroxysteroid dehydrogenase enzymes to convert cholesterol to pregnenolone (Madej, 1980; Caffrey et al., 1979). LII does not affect the activity of 20x-OH-steroid dehydrogenase. LH increases PRL receptors in granulosa cells in vivo and in vitro. LH may increase granulosa cell responsiveness to cL-adrenergic agents (Sheela Rani et al., 1983).  Granulosa and luteal cells secrete many nonsteroidal  substances. Prostaglandin synthesis is stimulated by LII (Marsh et al., 1981; Jones & Hsueh, 1982).  LII also induces oocyte maturation and the release of plasminogen  activator, proteoglycans and relaxin (Yanagashita et al., 1981; Gebauer et al., 1978). LH alters granulosa cell morphology and cell function. LII reduces the size of granulosa cell gap junctions. This may serve to signal the resumption of meiotic maturation and to allow granulosa cell desegregation.  LH stimulates transcription and translation of  steroidogenic enzymes and the enzymes for PG production. LH inhibits DNA synthesis by luteinized bovine and murine granulosa cells (Rao et al., 1978; Clark et at, 1976;  24  Gospodarowicz & Gospodarowicz, 1975). LH stimulates cAMP production (Marsh, 1976).  LH does not affect  phosphodiesterase activity (Marsh, 1970). Microfilament and Ca 2 may be involved in adenyl cyclase activation. TABLE 3. 1  .  LH-MODULATED FUNCTIONAL PARAMETERS IN CULTURED RAT GRANULOSA CELLS  Enhancement of steroido genesis Estrogen synthesis Progesterone and 20x-OH-P synthesis Stimulation of cholesterol esterase activity Induction of cholesterol side-chain cleavage enzyme Induction of 3B-hydroxysteroid dehydrogenase Modulation of plasma membrane receptors Increase in PRL receptor formation a. Increase in 2 adrenergic receptor formation b.  a. b.  2.  c. d. 3.  4.  Down-regulation or stimulation of LH receptors Increases in lipoprotein receptor formation  Secretion of non-steroidal local regulators a.  Plasminogen activator  b.  Mucopolysaccharides  c.  Prostaglandins Relaxin  d. Modulation of general cell functions a. b. c. d.  Increase in protein synthesis Increase in glycolysis Alteration of cell shape Reduction in size of gap junctions  Inhibition of DNA synthesis e. 5. Plasma membrane-related processes b.  Adenyl cyclase activation and cAMP formation Stimulation of cAMP-dependent protein kinase activity  c.  Stimulation of phospholipid turnover  a.  25  1.4.2.2.3  Role of prolactin (PRL)  PRL has little or no luteotropic effect on human ovarian function, although PRL  acts as a luteotropic hormone in rodents and several other mammalian species. However, PRL does have important regulatory effects on pituitary-ovarian function in women.  Hyperprolactinemia decreases gonadotropin secretion by impairing GnRH release.  1.4.3  Role of steroid Hormones  1.4.3.1  Role of estrogen  Estrogens maintain secondary sexual characteristics and exert feedback action on the hypothalamic-pituitary unit to affect synchronized preovulatory gonadotropin release. Table 4 describes various granulosa cell functions regulated locally by iniraovarian estrogens. Estrogens act on granulosa cells. Estradiol increases granulosa cell proliferation and estradiol receptor numbers (Richards, 1975). Estradiol enhances FSH-induced FSH and LH receptors (Richards et al., 1976). Therefore, estradiol exerts anti-atretic effects (Richards, 1975; Harman et al., 1975; Louvet et aL, 1975). Estrogen increases intercellular gap junction formation in granulosa cells (Merk et aL, 1972). Estrogen also enhances gonadotropin-initiated and maintained follicular development (Payne & Runser, 1958). Estrogens also regulate granulosa cell estrogen production.  Various native and synthetic estrogens augment FSH- and LH-induced  26  aromatase activity (Adashi & Hsueh, 1982; Zhuang et aL, 1982). This may partially explain maintenance of the dominant follicle in the ovary. Once a chosen follicle is producing a significant amount of estrogen, this follicle has the capacity to produce more estrogens than neighbouring follicles. However, it is not known how the selected TABLE 4.  DIRECT EFFECTS OF ESTROGENS ON FOLLICULAR GRANULOSA CELLS  1.  Granulosa cell proliferation (growth of follicles)  2.  Antiatretic action  3.  Increases in coupling between granulosa cells through intercellular gap junctions  4.  Increases in FSH induction of antrum formation  5.  Increases in granulosa cell estrogen receptor content  6.  Increases in FSH stimulation of granulosa cell LH and FSH receptors  7.  Increases in FSH-induced cAMP formation  8.  Increases in FSH-induced aromatase  9.  Increases in FSH-induced progesterone production  follicle continues to secrete elevated estrogens in the face of declining FSH concentrations (Zeleznik, 1981). Receptors of EGF and IGF-1 localized on granulosa cells suggest that these growth factors have a role on follicular development (Jones et al., 1982; St-Arnaud et al., 1983; Adashi et at, 1985) and steroidal hormone production of granulosa cells (May et al, 1985). Other local substances such as inhibin related peptides have not been studied in this point. It is possible local substances involve in the continuously elevated estrogen concentration within selected follicles. Estrogens secreted by antral follicles during the early follicular phase  27  suppress pituitary FSH and LH secretion. During the late follicular phase, increased estrogens are essential for the generation of the preovulatory estrogen surge.  This  elevated estrogen exerts positive feedback to increase LH and FSH release and may increase gonadotropin-stimulated estrogen biosynthesis. Estrogens are essential to gonadoiropin-mediated maintenance of corpus luteum function by enhancing gonadotropin-stimulated progesterone synthesis. This suggests that estrogens may act locally to enhance the sensitivity of luteal cells to pituitary gonadotropins.  1.4.3.2  Role of progestins Although the granulosa cells produce large quantities of progesterone, its local  action in modulating follicular growth and granulosa cell functions is not clear. Unilateral  ovarian progesterone implants in monkeys inhibit follicular growth without affecting the contralateral ovary (Goodman & Hodgen, 1979).  Further, follicular growth begins  immediately following lutectomy, suggesting that locally elevated progesterone concentrations inhibit folliculogenesis. The progesterone receptor has been identified in the cytoplasm of rodent, bovine, lapin and human granulosa cells (Schreiber & Erickson, 1979;, Philibert et al., 1977; Jacobs & Smith, 1980; Jacobs et al., 1980; Pasqualini et al., 1980). Progesterone receptors translocate to the nucleus (Naess 1981). The presence of ovarian progesterone receptors and high progesterone concentrations 6 (4x10 M ) in follicular fluid (McNatty, 1975) suggest an intrafollicular regulatory role for progesterone. Progesterone regulates its own production by granulosa cells. Pretreatment with  28  progesterone for 2 days enhanced FSH- and LH-stimulated progesterone and 2Oo-OH-P production in cultured granulosa cells (Fanjul et al., 1975). These findings may relate to the autonomy of luteal cell progesterone production and represent another example of an autocrine control mechanism. Sustained progesterone secretion during the preovulatory stage may exert a local action at the ovarian level to induce ovulation. Progesterone treatment enhanced LH stimulated ovulation in hypophysectomized rats in vivo (Fanjul et aL, 1975). This steroid may also contribute to ovulation in hamster ovaries obtained during proestrus and incubated with progesterone in vitro (Baranczuk & Fainstat, 1976). Progestin inhibits FSH-stimulated estrogen production and LH receptor fonriation in cultured rat granulosa cells (Schreiber et al., 1981; Schreiber et al., 1982).  1.4.3.3  Role of androgen Androgens are produced by theca and interstitial cells of the ovary and play an  important role in follicular development.  Specific androgen binding sites have been  identified in the cytoplasm of granulosa cells and translocate to the nucleus after androgen binding (Schreiber et al., 1976). Androgens stimulate follicular atresia and antagonize estrogen-induced ovarian weight gain. This is accompanied by necrosis of granulosa cells and ovum death (Payne et al., 1958, 1959). Pretreatment of intact rats with dihydrotestosterone (DHT) prevents FSH induction of LH receptors in granulosa cells (Farookhi, 1980). This inhibitory effect of DHT is antagonized by estrogen treatment. Androgens may counteract estrogen actions  29  by decreasing ovarian estrogen receptor content (Saiduddin et al., 1978). Androgen and estrogen concentrations in follicular fluid vary with the stage of follicular development. Elevation of the androgen to estrogen ratio is invariably associated with atresia (McNatty et al., 1975; Kemeter et aL,1975). Although androgen treatment induces atresia, androgen stimulates ovarian aromatase activity. Decreased androgen levels are accompanied by decreased aromatase activity (Katz et al., 1976).  Aromatase activity increases with  restoration of androgen levels (Katz et al., 1979). Androgens  also  stimulate  progestin  biosynthesis  by  granulosa  cells.  Administration of androgens to intact rats enhances the ability of FSH and LH to increase progesterone production by subsequently isolated granulosa cells (Leung et al., 1979). Thus, concomitant treatment with FSH and androgen increases the responsiveness of rat granulosa cells to LH, as measured by progesterone and cAMP production (Goff et al., 1979). Androgens act synergistically with FSH to stimulate progesterone production in cultured rat granulosa cells (Welsh et al., 1982). The mechanism Of the stimulatory effect of androgens on progestin synthesis has been examined.  Androgen and FSH act  synergistically to stimulate lipoprotein utilization by cultured rat granulosa cells (Schreiber et al., 1982).  Androgens have no effect on FSH binding, cAMP production or  phosphodiesterase activity (Nimrod, 1977). Thus, androgens exert both stimulatory and inhibitory effects in granulosa cells. The effect of androgens in vivo may vary according to the stage of follicle development. Androgens in the presence of high FSH concentrations promote estrogen and progestin synthesis and the development of preovulatory follicles. High androgen  30  concentrations in the presence of low FSH concentrations may lead to atresia.  1.4.4  1.4.4.1  Role of nonsteroidal local regulators  Role of gonadotropin-releasing hormone (GnRH) Hypothalamic GnRH acts on the pituitary gland to increase gonadotropin release.  Direct GnRH actions have also been demonstrated in rat granulosa cells (Hsueh et al., 1981a). Treatment with GnRH or GnRH agonists inhibits FSH-stimulated estrogen and progestin production (Hsueh et al., 1979a, b). Concomitant treatment with GnRH or GnRH agonists inhibits FSH-stimulated LH and PRL receptor formation in cultured rat granulosa cells (Hsueh et aL, 1980, 1983). The inhibitory effect of GnRH on LH and PRL receptor formation and FSH-stimulated steroidogenesis is blocked by concomitant treatment with GnRH antagonists (Jones et al., 1981; Navickis et aL, 1982; Clark et al., 1980; Massicotte et al., 1980). These results suggest that GnRH effects are mediated by GnRH binding sites in granulosa cells. GnRH modulates various other FSH-stimulated responses in granulosa cells, such as inhibition of FSH-stimulated cAMP formation (Clark et al., 1980; Massicotte et at, 1980; Knecht et al., 1981) and augments FSH-stimulated prostaglandin production (Clark et at, 1980). GnRH also inhibits FSH-stimulated ovarian EGF receptor content (St-Arnaud et al., 1983). The mechanism by which GnRH inhibits various FSH—dependent processes is unclear however, GnRH decreases FSH receptors in granulosa cells (Renta et al., 1983). This effect of GnRH may partially account for its  31  inhibitory effects. GnRH and GnRH agonists also inhibit the stimulation of steroidogenesis by cholera toxin, PGE 2 and cAMP analogs (Hsueh et al., 1980; Jones et al., 1981, 1983; Knecht et al., 1981a; Hiller et al., 1981). The response of granulosa cells to GnRH (ie. stimulation or inhibition of steroidogenesis) varies with the maturational stage of these cells as well as with the duration of exposure to GnRH (Popkin et al., 1983; Rani et al., 1983). Specific high affinity GnRH binding sites have been demonstrated in granulosa (Pieper et al., 1980; Jones et al., 1981) and luteal (Clayton et al., 1980) cells. Autoradiographic studies demonstrate that these ovarian GnRH receptors are present in luteal, thecal and granulosa cells at all stages of cellular differentiation (Pelletier et al., 1982). Photoaffinity labelling of ovarian GnRH receptors has identified two specific components with apparent molecular weights of 60,000 and 54,000 (Hazum et al., 1982). Since only one form of these GnRH receptors is found in the pituitary, the extra component of the ovarian receptor may be related to the different functions of GnRH in the gonad. Treatment with GnRH in vivo increases or decreases GnRH receptor content in the granulosa cells, depending on the dose and time of treatment (Pieper et al., 1980). FSH maintains GnRH receptor content in cultured granulosa cells (Ranta et al., 1982). GnRH-induced decreases in progesterone production may result from the inhibition of 38-hydroxysteroid dehydrogenase activity and enhancement of 20c-hydroxysteroid dehydrogenase activity (Jones et al., 1982a, 1981). The major effects of GnRH may be decreased pregnenolone production, coupled with increased conversion of progesterone  32  to 2Oct-OH-P.  Modulation of these steroidogenic enzymes by GnRH is blocked by  concomitant treatment with GnRH antagonists. An inhibitory effect of GnRH on FSH induced cAMP production has been observed in cultured porcine and murine granulosa cells (Knecht et aL, 1981b). These effects may be due to inhibition of adenyl cyclase activity and stimulation of phosphodiesterase activity (Knecht et al., 1981c). GnRH inhibited cAMP production by rat granulosa cells may be 2 -Ca dependent (Ranta et al., 1983). GnRH and GnRH agonists inhibit ovarian luteal functions in pregnant and nonpregnant rats (Hsueh et aT., 1980).  GnRH inhibits LII-, PRL- and B adrenergic  stimulated estrogen and progesterone production in rat granulosa cells (Jones et aL, 1982b).  GnRH also inhibits PRL-induced LH receptor content in rat granulosa cells  (Jones et al., 1982b). Other studies have demonstrated direct stimulatory effects of GnRH on various ovarian functions (Table 4). Treatment of granulosa cells with GnRH stimulates estrogen, progesterone and 20x-OH-P production (Dorrington et al., 1982; Clark 1982; Davis et al., 1983). GnRH stimulates phospholipid labelling in granulosa cells (Davis et aT., 1983; Naor et aT., 1982) as well as luteal cells (Leung et aT., 1983). The GnRH effect on phospholipid turnover may serve as a plasma membrane transducing event to initiate steroidogenesis and other events. Additionally, cAMP may be involved in GnRH action in ovarian cells (Knecht et aT., 1981b; Naor et aT., 1982). GnRH also stimulates maturation of follicle-enclosed rat oocytes in vitro (Leung et aT., 1983) and induces ovulation in hypophysectomized rats (Hillensjo et aT., 1980, 1982; Corbin et aT., 1981).  33  A low affinity GnRH binding site appears to be present in the human corpus luteum. GnRH treatment for 2 days is ineffective in modulating progesterone production in human granulosa cells (Clayton et at, 1982; Popkin et at, 1983; Casper et at, 1982). Prolonged treatment with GnRH inhibits human granulosa cell progesterone production in vitro (Tureck et al., 1982). Further, GnRH treatment inhibits steroidogenesis by human ovarian tumor cells in vitro (Lamberts et al., 1982). TABLE 5.  STIMULATORY EFFECT OF GNRH ON GRANULOSA CELL FUNCTIONS AND REPRODUCTIVE PROCESSES  1.  Stimulation of estrogen biosynthesis by increasing aromatase activity  2.  Stimulation of progesterone and 20a-OH-P synthesis by increasing 313hydroxysteroid dehydrogenase activity  3.  Increases in protein synthesis  4.  Increases in follicular lactate formation  5.  Increases in plasma membrane phospholipid turnover  6.  Induction of Oocyte maturation and ovulation in hypophysectomized rats  7.  Stimulation of prostaglandin production  8.  Stimulation of plasminogen activator activity  1.4.4.2  Role of inhibin-related peptides Activin inhibits and inhibin/follistatin stimulate the secretion of FSH by pituitary  gonadotropes (Robertson et at, 1985; Ling et at, 1985; Ling et at, 1986; Vale et at, 1986). 1.4.4.2.1  Local action of inhibin  There is increasing evidence of local actions of inhibin on ovarian and testicular function (Tsafriri et al., 1989; Sum et at, 1989; Ying et at, 1986; Shukovski 1990).  34  Porcine (Ying et al., 1986, 1987) but not bovine (Hutchinson et al., 1987; Hillier et aL, 1990) inhibin may suppress FSH-induced estrogen production by rat granulosa cells in vitro. Using purified bovine inhibin A (MW=32000, >200ng 1 m1 ) , Sugino et al. reported (Hillier et al., 1991a, b; Hsueh et al., 1987). Although granulosa cells possess inhibin only weak suppression of progesterone production by rat granulosa cells in vitro. Inhibin increased LH-stimulated androgen production by human theca and rat testicular cells receptors (Sugino et al., 1988; Woodruff et al., 1990), effects of inhibin on these cells are unclear. However, inhibin may alter the steroidogenic response of murine and bovine granulosa cells to gonadotropins (Xiao et al., 1990). Two paracrine actions of inhibin have been reported. Inhibin has been shown to enhance LH-induced androgen production by rat theca cell preparations. This effect of inhibin can be attenuated by activin (Hsueh et al., 1987). Inhibin A has been found to inhibit spontaneous maturational divisions of rat oocytes (0 et al., 1989). This inhibition was dose-dependent with an IC 50 of 4 Nm, similar to concentrations found in bovine follicular fluid (Burger et al., 1988). These paracrine actions of inhibin are yet to be confirmed in other species.  1.4.4.2.2  Local action of activin  Activin A increased basal and FSH-stimulated aromatase activity, progesterone and inhibin production in undifferentiated murine granulosa cells (Xiao et at, 1990). Erythroid differentiation factor (EDF), which was shown to be activin (Sugino et al., 1988a), also induced FSH and LH receptors on rat granulosa cells in vitro (Sugino et al.,  35  1988b,180). Binding sites for EDF have been demonstrated on granulosa cells and EDF receptor numbers are increased by FSH (Sugino et at, 1988a). The report from the same laboratory provided controversial results.  It suggested that activin inhibited FSH  stimulated progesterone production by rat granulosa cells (Hutchinson et al., 1987). However, activin A has a time- and dose-dependent inhibitory effect on progesterone and oxytocin production in differentiated bovine granulosa cells (Shukovski et al., 1990). Activin may act in a paracrine fashion to suppress LH-induced androstenedione production in rat theca cells and to antagonize the action of inhibin in the presence of LH. These paracrine effects of the granulosa cells would indirectly regulate the supply of androgen substrate required by granulosa cells to produce estrogen.  1.4.4.2.3  Local action offollistalin  Follistatin acts as an activin-binding protein (Kogawa et al, 1991). The detection of follistatin and inhibin-13 mRNA and protein in the rat ovary suggests a local regulatory action involving follistatin and activin interactions (Shimasaki et at, 1989). The effects of follistatin on basal and FSH-stimulated hormone production by rat granulosa cells have been investigated (Xiao et al., 1990). Follistatin decreased FSH-stimulated aromatase activity and inhibin secretion and increased progesterone production. However, follistatin had no effects on basal aromatase activity or progesterone and inhibin production (Xiao et al.,  1990).  Follistatin enhanced LH-stimulated progesterone production in  undifferentiated but not differentiated bovine granulosa cells (Shukovski et al., 1991). Follistatin accelerated the rate of meiotic maturation of rat oocytes in culture (0  36  et aL, 1989). This is opposite to reported effects of inhibin on oocyte maturation. Yet follistatin and inhibin have similar actions on the pituitary gland. The effects of follistatin on the human ovary are unknown. Like activins and inhibins (Li et aL, 1992; Hildj 1979; Hiller et al., 1991b), follistatin may well affect steroid hormone production in the human ovary.  1.4.4.3  Role of intraovariun growth factors The ovarian follicle undergoes cell proliferation during growth. Examination of  endocrine growth factors has centred on their growth-promoting actions in the ovary. However, these growth factors may also modulate cell differentiation. This action may be uncoupled from effects on cellular proliferation (Hilldj 1979; Chait et al., 1980; Johnson et al., 1980; Tashijian et al., 1982; Turo et al., 1982; Schmid et al., 1984; Baird et al., 1985). The following growth factors or growth factor families have been examined relative to their involvement in ovarian physiology.  1.4.4.3.1  Role of EGF and TGF a  Studies of EGF and ovarian function have been performed largely in vitro using granulosa cells.  EGF has been reported to attenuate FSH-induced LH/hCG receptor  formation in rat granulosa cells (Mondschein et al., 1981).  EGF not only decreased  LH/hCG receptor numbers but also affected granulosa steroidogenesis. EGF inhibited FSH-, dibutryl cAMP-, cholera toxin-, and 2 PGE stimulated estrogen production in rat granulosa cells (Hsueh et aL, 1981b). The inhibitory action of EGF does not exclusively  37  reside within granulosa cells. LH-stimulated androgen secretion by rat theca-interstitial cells is also inhibited by EGF (Erickson et al., 1983). Although EGF attenuates FSH action, EGF has been shown to stimulate FSH receptor content of porcine granulosa cells Osterman et al., 1979). Therefore, it has been difficult to formulate a consensus as to the role of EGF in folliculogenesis. The effect of EGF on progestin production is less clear. EGF has been reported to have no effect or to inhibit basal and FSH-stimulated progesterone production by granulosa cells after two days of culture (Jones et aL, 1982c; Knecht et al., 1983). EGF stimulated progestin production and enhanced FSH-stimulated 20x-hydroxysteroid dehydrogenase and 38-hydroxysteroid dehydrogenase activities in estrogen primed immature, hypophysectomized rats (Schomberg et al., 1984). The reason underlying the differences in cell responsiveness to EGF in terms of progestin secretion are unknown but may involve the endocrine status of the animal and culture conditions of granulosa cells. Based on the above information, EGF plays an inhibitory role with respect to granulosa cell function. However, this inhibitory role may be related only to specific periods of cell differentiation or follicular development. EGF is a mitogenic factor in the ovary and is a major growth stimulating factor required for elevated granulosa cell proliferation. Despite the presence of other growth factors, such as IGF-I, insulin, and PDGF, granulosa cell proliferation is impaired in the absence of EGF (Gospodarowicz D et ai., 1979). TGFcz has an amino acid sequence similar to EGF. TGFct binds to EGF receptors. Since TGFa acts via EGF receptors, the actions of TGFa mimic those of EGF. Recent  38  studies, however, suggested that the two growth factors may have different potencies, if not divergent actions.  TGFa was found to be more potent than EGF in stimulating  angiogenesis in the hamster cheek pouch assay. Since the formation of the corpus luteum requires massive blood vessel formation, TGFa may have an effect during the early stages of corpus luteum formation.  1.4.4.3.2  Role of insulin and IGFs  The use of insulin to supplement culture media for ovarian cells and the presence of ovarian receptors and insulin action on granulosa cell metabolism and steroidogenesis have been reviewed (Veidhuis et al., 1983; May et al., 1981). As there appears to be no report of local insulin production, the action of IGFs on ovarian function wifi be addressed. Addition of somatomedin C (SmC)/IGF-I to granulosa cells from small (4-6 m m) bovine follicles caused mitogenic responses in serum-free cultures (Savion et aL, 1981). Likewise, IGF-I and IGF-II or insulin enhanced 3 ][H -thymidine incorporation into DNA of porcine  granulosa cells (Baranao et al., 1984).  Addition of SmC/IGF-I  augmented FSH-stimulated progesterone accumulation but had limited effects on basal progesterone production (Adashi et at, 1985a,b).  Similarly, SmC/IGF-I stimulated  progesterone production by porcine granulosa cell cultures (Baranao et at, 1984). IGF-II synergized with FSH in stimulating progesterone production (Veldhuis et at, 1983). Similarly, SmCIIGF-I synergized with esiradiol-17B in enhancing porcine granulosa progesterone and pregnenolone synthesis (Veldhuis, 1985). SmC/IGF-I synergized with  39  FSH in enhancing aromatase activity of rat granulosa cells but had no effect on basal aromatase activity (Adashi et al., 1985c). Finally, SmC/IGF-I synergized with FSH in inducing LH receptors (Adashi et al., 1985d). Therefore, IGFs may be involved in the regulation of follicular growth and differentiation.  1.4.4.3.3  Role of bFGF and TGFJJ  bFGF, a 146-amino acid polypeptide, is a mitogen for a wide variety of mesoderm-derived and neuroectoderm-derived cells. Although the physiological function of bFGF in the ovary remains unclear, evidence suggests that bFGF may play a central role in growth and development of granulosa-lutein cells (Gospodarowicz 1989). BFGF stimulates the replicative lifespan of cultured bovine, porcine, lapin, rodent and human granulosa cells (Gospodarowicz et al., 1977, 1978, 1979). More recent studies indicated that theca-interstitial cells are also a site of bFGF reception and action. bFGF is a potent attenuator of gonadoiropin-supported androgen synthesis by theca-interstitial cells. TGFB is a polyfunctional regulator in the ovary. TGFB has been shown to alter the proliferation and differentiation of rat granulosa cells (Dorrington et al., 1988, Knecht et al., 1989). Evidence suggests that ovarian theca-interstitial cells, as well as granulosa cells, may be sites of TGFB production and action (Magoffin et al., 1989). The potential significance of TGF-B to ovarian physiology remains unknown (Knecht et al., 1989).  1.4.4.4  Role of ovarian renin-angiotensin system A II may be involved in ovarian neovascularization.  40  A II stimulated bFGF  production in bovine luteal cells, bFGF being a potent angiogenic factor (Sterling et al., 1990). A II has also been shown to regulate steroidogenesis. In vitro studies on rat ovarian fragments from ECG-stimulated rats showed a stimulatory effect of A II on estradiol secretion (Pucell et al., 1987; Bumpus et al., 1988). However, studies with granulosa cells from diethylstilbestrol-treated rats suggested no effect of A II on aromatase activity (Pucell et al., 1988).  A II increased gonadotropin-stimulated  progesterone, testosterone and estradiol production by cultured human granulosa cells (Palumbo et al., 1988a). A II had no effect on basal progesterone secretion by cultured bovine luteal cells and inhibited LH-stimulated progesterone production (Sterling et al., 1990).  It appears that there may be species-dependent and cellular differentiation-  dependent differences in effects of A II on ovarian steroidogenesis.  Recent studies  indicated the involvement of A II in ovulation and oocyte maturation (Pellicer et al., 1988, Kuo et al., 1991). Studies with perfused rat and rabbit ovaries demonstrated that A II induced ovulation in the presence or absence of gonadoiropiñs. An A II antagonist, saralasin, reversed the effect of A II (Kuo et al., 1991, Peterson et al., 1993, Yoshimura et al., 1992). A II also induces oocyte maturation in rats. Saralasin dose-dependently inhibited hCG-stimulated germinal vesicle breakdown (GVBD) (Palumbo et al., 1988a). Prorenin concentration correlates with follicular development, oocyte-cumulus complex maturity and oocyte viability in humans (Ward 1978). Studies of other species have shown that active renin content of the ovary is not correlated with oocyte maturity (Daud et al., 1988). Angiotensins may play a role in follicular development through modulation of steroidogenesis. Since estrogens predominate in the follicular fluid with  41  meiotically arrested oocytes, 1713-estradiol has been reported to inhibit GVBD in porcine oocytes (McGaughey, 1977). Progesterone and testosterone can also inhibit GVBD in mouse oocytes (Smith et al., 1980). Seibel et al. reported that estradiol concentrations were lower in fofficular fluid containing degenerated oocytes than in that containing healthy oocytes (Seibel et at, 1989).  Thus, angiotensin concentrations in dominant  follicles regulate estradiol concentrations of follicular fluid, which could effect oocyte maturation. Although many factors, including inhibin-related peptides, growth factors, prostaglandins and other substances affect steroidogenesis, growing evidence indicates that angiotensins play a significant role in this process.  1.4.4.5  Role of infraovaruxn prostaglandins Prostaglandins play an important role in follicular rupture in a variety of  mammalian species. PG synthesis inhibitors block ovulation. Afitiserum against PGF inhibits ovulation. Elevated PG concentrations occur in preovulatory follicles (Behrman, 1979). PGE 2 and PGF are increased in follicular fluid but it has not been established whether PGE , PGF or both induce follicular rupture. 2  Plasminogen activator from  granulosa cells is secreted into the follicular fluid in response to gonadotropins, inducing digestion of the follicular wall by plasmin (Beers et al., 1975). The role of PGs in this process is unclear (Strickland et at, 1976). The role of PGs in human luteal regression is controversial, whereas in non-primate mammalian species PGF is the major physiological luteolysin (Behrman, 1979).  42  PGF treatment is directly luteolytic,  immunization against PGF prolongs the life span of the corpus luteum (as does treatment with cyclooxygenase inhibitors) and increased circulating PGF of uterine origin occurs at the approximate time of luteal regression in non-primate mammals. PGF2(x induces functional regression of the corpus luteum by a receptor-mediated process in rats, independent of changes in ovarian or luteal blood flow (Pang et al., 1981). PGF uncouples the occupied LH receptor from adenylate cyclase, which decreases LH dependent cAMP accumulation and steroidogenesis (Thomas et al., 1978). The mediators of PGF effects are unknown but do not appear to involve Ca or protein kinase C. Yet these specific mediators produce effects similar to that of PGF (Pepperell et aL, 1989; Musicki et al., 1990). PGF may be luteolytic in humans but this conclusion must be tempered by the relative insensitivity of the human corpus luteum to PGF. The human uterus plays no role in the luteolytic process. However, PGF is produced by the human corpus luteum and PGF receptors are present in human luteal tissue (Powell et al., 1974). Estrogens induce premature luteal regression in humans (Johansson et al., 1971) and monkeys (Karsch et al., 1973). Based on studies in monkeys, this action of estrogen appears to be directly within the ovary (Karsch et al., 1976). The luteolytic action of estrogen in monkeys is linked to PG production; estrogen increases PGF concentrations in ovarian blood and indomethacin blocks estrogen-induced luteolysis (Auletta et al., 1976). Nonprimate mammals are refractory to PGF during the early luteal phase in vivo and in vitro (Hall et al., 1981). An endogenous PGF antagonist or inadequate PGF  receptor complement may protect the corpus luteum from regression in the early luteal  43  phase.  1.4.4.6  Intracellular signalling in the ovary  Ovarian cellular functions are regulated by peptide hormones, neurotransmitters, other nonsteroidal substances and steroids. These hormones regulate ovarian cells via intracellular second messengers after the binding of hormone and receptor. Hormone receptors in the ovary are separated into two categories: membrane receptors and intracellular receptors. Most peptide hormones interact with membrane receptors while most steroid hormones interact with intracellular receptors. The actions of hormones may be mediated by one or more of several intracellular signalling pathways. Although the mechanism of action of many ovarian hormones is unclear, various signalling pathways which may be involved will be described. Fig. 6 illustrates the general pathways and second messengers generated by hormone-receptor interaction.  44  Figure 6.  Mechanism of intracellular signalling transduction by peptide hormones. Two major pathways appear to be involved in the peptide hormone actions. G proteins (GP) regulate adenylate cyclase, latter hydrolyzes ATP to cAMP. cAMP activates protein kinase A (PKA), PKA phosphorylates substrate proteins and stimulates steroidogenesis. Other hormones such as GnRH-receptor complex activates phospholipase A 2 and phospholipase C and produce AA and inositol triphosphate (1P3)/DAG by hydrolysis of polyphosphoinositides (PIP2). DAG activates protein kinase C (PKC), 1P3 releases intracellular Ca , and AA metabolizes to prostaglandins. PKC, 2 2 and prostaglandin F inhibit the enzymes involved in steroidogenesis. Ca  1.4.4.6.1  Mechanism of transduction by steroid hormone receptors  Steroid hormones are hydrophobic and can diffuse through cellular membranes to bind with specific intracellular receptors. Steroid hormone-receptor complexes induce gene expression by binding to specific DNA sequences which act as transcriptional enhancers. Direct induction of transcription of a few specific genes is known as the primary response. The products of these genes may then activate or inactivate other genes to produce a secondary response. Thus, a steroid hormone can cause complex  45  changes of ovarian cells. The same receptor protein regulates different genes in different target cells, as DNA-binding proteins required for specific gene transcription differ between cell types.  1.4.4.6.2  Adenylate cyclase-cAMP pathway  A large number of hormones exert their effects by increasing intracellular cAMP concentration. Membrane receptors may be associated with two classes of G proteins; Gs protein activates adenylate cyclase, while Gi protein inhibits this enzyme. The G proteins are heterotrimers consisting of a-, B- and ‘c-subunits (Neer et aL, 1990). Activation of G proteins occurs when agonists bind to receptors, releasing GDP from G protein and allowing GTP to bind in its place. Activation triggers the a-subunit to dissociate from the B-’r-subunits, rendering the a-subunit free from the receptor. The a-subunit exhibits intrinsic GTPase activity which hydrolyzes GTP to GDP. Released a-subunit binds to and activates the catalytic subunit of adenylate cyclase with energy generated from the hydrolysis of GTP.  Activated adenylate cyclase hydrolyzes ATP to cAMP, which  subsequently activates protein kinase A in the cytoplasm. The activation of protein kinase A phosphorylates proteins involved in steroidogenesis. Alternatively, receptor-mediated inhibition of adenylate cyclase may occur by activation of Gi proteins. Two bacterial toxins which stimulate Gs and inhibit Gi (cholera toxin and pertussis toxin, respectively) are useful tools for the study of the adenylate cyclase-cAMP pathway and its involvement in intracellular signalling. Intracellular cAMP is degraded to 5’-AMP by the action of phosphodiesterase.  cAMP may be the second messenger for LII, FSH and hCG in  46  ovarian cells (Kolena et al., 1992; Goff et al.; 1977, Marsh et al., 1966; Tsang et at, 1979). PGE 2 also increases cAMP in cultured granulosa cell differentiation (Behrman, 1979; Kolena et al., 1992; Goff et al., 1977).  1.4.4.6.3  Calcium and protein kinase C pathway  Many hormone receptors are coupled to G proteins which promote activation of phospholipase C (PLC). PLC hydrolyzes membrane-bound polyphosphoinositides, the products of which act as second messengers in Ca 2 mobilization and activation of calmodulin-dependent enzymes and protein kinase C (PKC).  The various second  messengers derived from inositol lipids has been reviewed (Can et at, 1991). Briefly, PLC activation results in rapid metabolism of phosphatidylinositol(4,5)bisphosphate ) in the membrane, releasing inositol( 1 ,4,5)trisphosphate (Ins( 1 ,4,5)P 2 (Ptdlns(4,5)P ) and 3 diacyiglycerol (DAG) into the cytosol. Ins(1,4,5)P 3 releases Ca 2 from the endoplasmic reticulum (Furuichi et al., 1989; Ross et at, 1989) and calsequestrin-containing organelles (termed calciosomes) to stimulate Ca 2 release (Volpe et at, 1988).  Receptors for  3 bind with high specificity and affinity and have been estimated to release at Ins(1,4,5)P least 20 Ca 2 ions (Joseph et al., 1984). The extent of the initial Ca 2 response may be partially due to positive feedback of Ca 2 on PLC activity.  Additional steps in the  degradative and synthetic pathways of inositol metabolism include generation of additional signal molecules, InsP 5 and InsP , prior to recycling of inositol into membrane 6 phosphatidylinositols (Valejo et al., 1987). A second branch of the PLC pathway originates from the generation of DAG.  47  DAG regulates the activity of the PKC family of calcium/phospholipid-dependent enzymes by controlling their affinity for Ca 2 and for phosphatidylserine (Nishizuka 1984). Therefore, DAG stimulates PKC activation in the absence of a change in 2 [Ca j . Further, simultaneous elevation of [Caj and Ins(1,4,5)P 3 enhances PKC activity if DAG concentrations are suboptimal (Catt et at, 1989). Phorbol esters, compounds which bear physical homology to DAG, activate PKC by binding to the regulatory region of the enzyme.  Phorbol esters are useful tools for the study of hormones which act by  stimulating the PLC pathway. PKC may also exert negative feedback through inhibition of polyphosphoinositide hydrolysis and promotion of Ins(1,4,5)P 3 metabolism (Catt et al., 1989). PKC may alter receptor affinity through phosphorylation of membrane receptors. Seven subspecies of PKC have been reported (Nishizuka, 1988). GnRH is a peptide hormone and its effects are mediated by specific receptors. The mechanism of GnRH action on the ovary has been investigated. GnRH and GnRH agonists stimulate the breakdown of polyphosphoinositides into inositol phosphates and DAG in the ovary (Leung et al., 1983; Ma et al., 1985; Minegishi et at, 1985). Ca 21 is required for GnRH action in granulosa cells (Ranta et al., 1983) and protein kinase C has been characterized in the ovary (Noland et at, 1984). DAG may be hydrolysed by DAG lipase resulting in the generation of arachidonic acid, another second messenger. It has been proposed that some subspecies of PKC are also activated by phospholipid metabolites, such as DAG, arachidonic acid and lipoxin A (Nishizuka, 1988; Hansson et al., 1986).  48  1.5  Localization, secretion and action of inhibin in human placenta The placenta functions as an endocrine organ during pregnancy. It maintains the  pregnancy after the luteo-placental shift. The synthesis of steroid and peptide hormones are regulated by local placental and fetal factors. hCG is a polypeptide hormone that regulates progesterone production from the corpus luteum and placenta (Simpson et aL, 1981).  A placental GnRH has been proposed as a local modulator of hCG secretion  (Butzow, 1985; Kim et al., 1987; Peiraglia et aL, 1987). The presence of inhibin-like bioactivity and immunoreactivity in placenta of several mammalian species suggests a role for inhibin in the endocrinology of pregnancy (McLachlan et al., 1986; Petraglia et al., 1987). Further evidence that inhibin functions in pregnancy is the observation that inhibin was increased rather than declined during the late luteal phase of women who haxl conceived  following  in  vitro  fertilization  (McLachlan  et  al.,  1987a).  Immunohistochemical staining showed that inhibin-like immunoactivity was localized in centrally situated cytotrophoblast cells; peripheral syncytiotrophóblast were unstained. However, fluorecescent-labelled antisera specific to each subunit indicated that inhibin a. was localized to cytotrophoblast. Inhibin BB-subunit was identified in the syncytial layer of the villi. Inhibin BA-subunit was present in the cytotrophoblast and syncytial layer (Petraglia et al., 1991). Placental inhibin synthesis has been confirmed using molecular techniques. lihibin cc-subunit mRNA was cloned and sequenced from a human term placental cDNA library (Mayo et al., 1986). Inhibin BA- and BB subunits have been detected in the placenta by S 1-nuclease analysis (Meunier et al., 1988a). Northern blot analysis of placenta inhibin mRNAs revealed that the BA subunit was present in the  49  highest concentration, suggesting local synthesis of activin-A (Meunier et al., 1988a). Inhibins and activins regulate hCG and GnRH synthesis and secretion in human placental cells (Petraglia et al., 1987). The ability of activin but not inhibin to stimulate hCG release was confirmed using placental cell perfusion (Currie et al., 1993).  50  GENERAL OBJECTIVES This study examined (1) the effect of local regulators on progesterone production by human granulosa cells, (2) the regulation of synthesis of these local substances in granulosa and placental cells and (3) the mechanisms of action of these local substances.  BACKGROUND AND RATIONALE The physiologic processes of the ovarian cycle have classically been attributed to changes in hormonal output of the hypothalamic-pituitary axis. The following phenomena suggest that local substances play key roles in the regulation of the microenvironment of the ovary or individual follicles which react to gonadotropins: 1.  Ovulation occurs in one ovary, but gonadotropin concentrations are the same in the blood perfusing the two ovaries.  2.  Generally, only one follicle responding to LH during a given cycle ovulates.  3.  The hormonal profiles of follicular fluid differ among follicles and determine follicular development.  4.  The maintenance and lysis of the corpus luteum are not explained by gonadotropin regulation. Substances in ovarian tissues and follicular fluid act locally, either alone or by  modulating the actions of the gonadotropins, to change the functions of ovarian cells. The inhibin-related peptides, as well as GnRH, PGF and angiotensins are synthesized in the ovary. Their roles as local regulators in human ovaries are uncertain. As an experiment model, a granulosa-luteal cell culture system that is responsive to hCG and  51  FSH stimulation was used. As progesterone production is stimulated by FSH and LH in cultured granulosa cells, progesterone production served as a functional marker. The results of these studies will advance the understanding of the endocrine function of the hypothalamic-pituitary-ovarian axis.  HYPOTHESIS This study tested the hypothesis that specific intraovarian substances (activin, inhibin, follistatin,  A II & A III, PGF and GnRH) are involved in gonadotropin  dependent processes in the human ovary and, more specifically, on production of ovarian steroidal hormones.  SPECIFIC OBJECTIVES 1.  To examine the regulatory effect of inhibin and activin on basal and gonadotropin stimulated progesterone production and the mechanisms underlying the effects of inhibin and activin on human granulosa cells.  2.  To examine the regulatory effect of follistatin on basal and hCG- stimulated progesterone production and the mechanisms underlying the effects of follistatin on human granulosa cells.  3.  To examine the regulatory effects of GnRH, PGF and angiotensins on basal and hCG-stimulated progesterone production.  4.  To examine inhibin-subunit gene expression in human granulosa and placental cells.  52  2.0  GENERAL MATERIALS AND METHODS  2.1  HUMAN GRANULOSA CELL PREPARATION Human granulosa cells were harvested during oocyte collection in the University  of British Columbia In Vitro Fertilization Program.  Ovulation was induced using a  combination of clomiphene citrate and exogenous gonadotropins. Clomiphene citrate (100 mg, Serono) was taken daily by mouth between days 3-7 of the menstrual cycle. Menotropins (75-150 IU im, Pergonal, Serono) were given from day 6 and rapid assay monitoring of circulating estradiol (Diagnostic Products, Los Angeles, CA) and luteinizing hormone (Serono) was begun.  The follicle was tracked with ukrasonography and  menotropin stimulation stopped when at least 2 leading follicles were 18 mm or greater in diameter.  Human chorionic gonadotropin (10,000 IU im, Serono) was given  approximately 32 h before aspiration of 10-20 follicles using a transvaginal approach. Granulosa cells were harvested from the follicular fluid contents after the oocyte was identified for in vitro fertilization.  The use of human granulosa cells was approved by  the Clinical Screening Committee for Research and Other Studies Involving Human Subjects of the University of British Columbia. Follicular contents were centrifuged at 1000 x g and supematant was decanted. Cells were suspended in 1-2 ml Hank’s balanced salt solution (Gibco, Burlington, ON) containing 0.5% bovine serum albumin (BSA, Sigma, St. Louis, MO), layered onto 40% Percoll solution (Sigma) and centrifuged at 1700 x g for 20 mm at 20 C (Davis et al., 1989). The cells were then washed and resuspended at the density of i0 5 cells/mi media  53  199 supplemented with 10% fetal bovine serum (FBS), sodium penicillin (100 NJ/mi) and streptomycin (100 jig/mi) (M199, Gibco). The cells were plated and cultured in 48-well plates in replicates of 8 wells per treatment. The incubation media were replaced with M199 containing 2% FBS at 48 h and subsequently with M199 containing 0.5% BSA and the various treatments at 96 h after plating. Cells were incubated at 37 C in humidified air with 5% CO . Incubation media were collected every 24 h during incubation for 2 progesterone measurement.  Progesterone concentration was highest in first 24 h  incubation and gradually decreased until fifth day incubation. The basal progesterone concentration increased when cell concentration increased.  The basal progesterone  concentration at same cell concentration also varied with different patients. Incubation after 96h was chosen for treatment time and progesterone and 2Oct-OH-P were measured by radioimmunoassay (RIA) after 24 h treatment. The cultured human granulosa cells in this study have been exposed to the superovulated dose of human menopause gonadotropin (hMG) (Serono) and hCG, most of these cells will be luteinized during the culture. The rat granulosa cells mentioned during discussions in this thesis were collected by superstimulating immature rats with ECG and hCG.  2.2  HUMAN  PLACENTA  CELL  PREPARATION  AND  HORMONE TREATMENT Use of human placental cells was approved by the Clinical Screening Committee for Research and Other Studies Involving Human Subjects of the University of British  54  Columbia.  Term placentae were from elective cesarean section.  Membranes and  connective tissues were removed by dissociating term trophoblast in Media 199 (0.5% BSA, 100 EU penicillin G sodium and 100 pg streptomycin/nil supplemented M199, Gibco, Burlington, Ontario) through a 150 p-screen (Sigma, St. Louis, MO). Hematocytes were removed by centrifuging (1700 x g for 20 mm) the suspensions on 40% Percoll in M199 as above. Cell viability exceeded 90% as determined by Trypan Blue exclusion test and cell concentration was adjusted to 10 /ml. The purified cell suspension (20-30 8 ml) was cultured in 50 ml culture tubes (Corning Inc., Corning, New York) in a 37 C shaking water bath (30 cycles/mm) for 6 h with or without 8-Br-cAMP (150 pM) and GnRH (100 nM) (Sigma).  2.3  HORMONE ANALYSIS Concentrations of progesterone and 20(x-OH-P in media were determined by  validated RIA with specific antisera provided by Dr. D. T. Armstrong of the University of Western Ontario. The cells were resuspended in 6% trichloroacetic acid (TCA) and sonicated 3 times at 90 watts for 10 seconds for intracellular cAMP assay. cAMP concentrations were assayed using a commercial kit (Rianen cAMP [‘I] RIA kit, DuPont, Sarnia, ON).  2.3.1  Radioimmunoassay (RIA) for steroids  Reagents and buffer:  55  1) The assay buffer (PBSG) was 0.1 M phosphate buffered saline (pH 7.4): NaC1 80 gm, KC1 2 gm, Na 4 H 2 PO 11.5 gm, KH 4 P 2 O 2 gm, 0.1%  (w/v) thimerosal,  0.1% (w/v) gelatin, stored at 4C. 2) The progesterone (Sigma) standards were serially diluted in glass distilled ethanol from 0.32 mM stock solution. A standard curve was set up with 8 reference concentrations ranging from 0.5 nM to 64 nM. 3) The 2Oct-OH-P (Sigma) standards were serially diluted in glass distilled ethanol from 0.32 mM stock solution. A standard curve was set up with 8 reference concentrations ranging from 0.5 nM to 64 nM. 4) The antisera were rabbit anti-progesterone and 20-OH-P antisera powders provided by D. T. Armstrong, raised against 4-pregnen-68-ol-3,20-dione hemisuccinate:bovine serum albumin conjugate and 4-pregnen-20c-ol-3-one:bovine serum albumin conjugate (Steraloids, Wilton, NH). These were used at a final dilution of 1:12500 w/v in PBSG and gave approximately 50% binding of tracer. 5) The radio-tracers were 3 H-progesterone (NET-724, NEN DuPont, Mississauga, ON) and 3 H -2Oct-OH-P. 6) The dextran charcoal solution was 0.025% dextran and 0.25% charcoal dissolved in PBSG. 7) Scintiverse (Fisher). Intra-assay coefficients of variation were 6.8% for progesterone and 7.8% for 20aOH-P. respectively.  Sensitivity of assays was determined by the detectably lowest  hormone concentration plus 2 x standard deviation (SD). The sensitivity of both assays  56  was 0.5nM. Protocol:  1)100 p1 progesterone standard or sample solution was added to each tube, each standard in triplicate and each sample in duplicate. 2) 100 p1 diluted antibody solution was added to each tube. H-progesterone (10000 cpm, 10 nM) was added in each tube. 3)100 p 13 4)100 p1 PBSG was added in all tubes to make final volume 400 p1. 5) Incubation at 4 C for 16-24 h. 6) 500 p1 dextran charcoal solution was added to each tube. 7) Incubation for 10 mm at 4 C. 8) Centrifugation 2500 x g for 10 mm at 4 C. 9) Supernatant was transferred to vials. 10) Scintiverse (3 ml) was added to each vial which was counted by LKB B-counter. 11) The progesterone concentrations of samples were calculated from the standard curve.  2.3.2  Radioimmunoassay (RIA) for cAMP  Reagents and buffer: Reagents and buffer were provided in the kit. Protocol:  1) 300 p 1 TCA was added to each cell culture well and cells were sonicated at 30 watts for 30 sec. 2) The cell suspension was collected and centrifuged at 10000 x g for 5 mm.  57  3) The supernatant was transferred to 10 ml glass tube for ether extraction. 4) Ether (3 ml) was added in the tube and vortex for 1 mm. 5) Ether phase was removed and this procedure was repeated 2 more times. 6) The supernatant (sample) was used for intracellular cAMP assay. 7) The sample was acetylated by adding 5 p1 acetylation reagent to each sample and then immediately vortexed. The sample was incubated for at least 3 mm at room temperature, then 900 p1 of assay buffer was added to each tube. 8) 100 p1 of each standard solution and sample was added to test tubes. 9)100 p1 working tracer solution was added to all tubes. 10) 100 p1 antiserum complex was added to all tubes. 11) The tubes were incubated overnight (16-18 h) at 2-8 C. 12) 0.5 ml cold cAMP precipitator was added to all tubes except total count tubes. 13) The tubes were centhfuged for 15 mm at 1200 x g after vortexing. 14) The supematant were decanted into a radioactive waste container by gently inverting all tubes once. The tube was kept inverted for 1 mm to facilitate removal of remaining droplets. 15) All tubes were counted on LKB gamma counter and the cAMP concentrations were calculated from the standard curve.  2.3.3  Statistical analysis The experiments were performed with eight wells per treatment. The human  granulosa cells were from one patients in each experiment.  58  The experiments were  repeated by using the cells from different patients. Statistical significance of the data were determined by analysis of variance followed by Sheffe’s test.  59  3.0  EFFECT OF INHIBIN AND ACTIVIN ON PROGESTERONE AND 2O-o-OH-P PRODUCTION IN HUMAN GRANULOSA CELLS  3.1  INTRODUCTION There is increasing evidence of local actions of inhibin and activin on ovarian and  testicular function in addition to their known actions on FSH secretion at the pituitary (Tsafriri et at, 1989; Sum et al., 1989; Ying et al., 1986; Shukovski, 1990). Inhibin and activin increased and decreased, respectively, LH-stimulated androgen production by human theca and rat testicular cells (Hillier et at, 1991a&b; Hsueh et at, 1987). Although granulosa cells possess inhibin and activin receptors (Sugino et al., 1988a; Woodruff et at, 1990), their effects on these cells are unclear. However, inhibin and activin may alter the steroidogenic response to gonadotropins in murine and bovine granulosa cells (Shukovski et at, 1990; Woodruff et at, 1990; Xiao et al., 1990). The genes encoding inhibin subunit alpha, beta A and beta B have been cloned and recombinant inhibin A and activin A recently became available for investigation (Mason et at, 1985; Broxmeyer et aL, 1988). Autocrine regulation of human ovarian function by inhibin and activin has not been elucidated.  This study investigated effects of  recombinant inhibin A and activin A on progesterone and 20x-OH-P and cAMP production in human granulosa cells.  60  3.2  TREATMENTS  3.2.1  Activin A-, inhibin A- and gonadotropin-regulated steroidogenesis The effects of activin A and inhibin A on the steroidogenic response of cultured  granulosa cells to gonadotropins were investigated by adding to the culture medium hCG (1 lU/mi) (Sigma) or FSH (0.5 jig/mI ovine FSH, NIADDK-oFSH-16, NIH, Bethesda, MD) with or without 100 ng/ml recombinant human activin A or inhibin A (Genentech, San Francisco, CA).  3.2.2  Dose-dependency and lime-course study of acivin A on hCG stimulated steroidogenesis Increasing doses of activin A (1 to 100 ng/ml) were added with hCG (1 lU/mi)  to the cultured cells and the cells were incubated for 16 h. The incubation media were collected for measurement of progesterone and 20a-OH-P. A time-course analysis was performed by adding hCG (1 lU/mi), with or without activin A (100 ng/ml) to the cells and harvesting the media at 6, 12, 18 and 24 h of incubation for progesterone analysis.  3.2.3  Effects of activin A on cAMP-stimulated progesterone production 8-bromo-cAMP (8-Br-cAMP, 1.5 mM, Sigma) and forskolin 6 (50xl0 M , Sigma),  with or without activin A (100 ng/ml), were added to cells. The incubation media were collected 16 h later and analyzed for progesterone.  61  3.2.4  Effects of activin A on cellular cAMP accumulation Activin A (100 ng/ml) was added to the cells, with or without hCG (1 Ri/mi) and  in the presence or absence of iso-butyl-methyixanthine (IBMX, 0.5 mM).  Medium  progesterone and intracellular cAMP were measured after 24 h incubation.  3.3  RESULTS  3.3.1  Effect of activin A and inhibin A on gonadotropin-stimulated progesterone and 2Oc-OH-P accumulation Recombinant human activin A and inhibin A at a concentration of 100 ng/ml did  not affect basal progesterone synthesis (P>O.05). increased 2-6 fold by hCG or FSH (P<0.0l).  Progesterone concentrations were Activin A inhibited hCG-stimulated  progesterone production (P<0.0 1) and FSH-stimulated progesterone production (P<0.0 1) (Fig. 7). Inhibin A had no effect (P>0.05). Activin A suppressed hCG-stimulated 20ctOH-P accumulation (P<0.01, Fig. 8).  3.3.2  Dose- and time-dependent inhibition ofprogesterone and 2Oci-OH-P production by activin A The effect of activin A on progesterone and 20x-OH-P production in response to  hCG was dose-dependent. The effect of activin A was significant at a dose of 3.7 ng/ml (Fig. 9). Time-course experiments revealed that progesterone production was decreased by activin A at 18 and 24 h of treatment (Fig. 10).  62  8  4.0  .5  C.)  U)  C) C.) U, 0  J.W  •1  o  0  a 2.0 2  -  1.0  C)  3  0.J  T  2-  control  Figure 7.  FSH+ activin  control Inhibin actMnhCG  hCG+  hCG+  inhibin activin  Effect of 100 ng/ml human recombinant inhibin-A and activin-A on 1 lU/mi hCG- or 0.5 jig/mi ovine-FSH-stimulated progesterone production by human granulosa cells. Cells were incubated for 24 h.  Values are  mean ± sem (n=8). Similar results were obtained in other 5 experiments. Different superscripts (above the sem bars) denote statistical differences (P<O.O1).  63  7. 2  6  1 j T  control Inhibin  Figure 8.  activifl  hCG  hCG+ activln  cAMP  cAMP+ activin  Effect of 100 ng/m1 human recombinant inhibin-A and activin-A on 1 EU/mi hCG-, 0.5 ig/m1 ovine-FSH- and 1.5 mM 8-Br-cAMP-stimulated 20a-OH-P production by human granulosa cells. Cells were incubated for 24 h. Values are mean ± sem (n=8). Similar results were obtained in other 2 experiments. Different superscripts (above the sem bars) denote statistical differences (P<O.01).  64  12  II C)  10  U)  C 1  0  E C  C 0 G) (0 0) C) 0 0..  6  4  10  1  0  100  Activin (ng/ml) Figure 9.  Effect of increasing concentrations of recombinant human activin A (3-100  ng/ml) on hCG (1 IU/ml)-stimulated progesterone production. Cells were incubated for 24 h. Values are mean ± scm (n=8). Similar results were obtained in other 2 experiments. *  P<0.05,  **  P<0.01  65  °  30  hG hCG÷Activin  0 C) It) **  0  20  1  0  E  C 0) C 0 C) 0 0) 0) 0  10  L..  0  0  12  6  0  18  24  Time (hr) Figure 10.  Time-course effect of recombinant human activin A (lOOng/mi) on hCG (1 RJ/ml)-stimulated progesterone production by human granulosa cells. Cells were incubated for 24 h. *  P<0.05,  **  P<0.0l  66  Values are mean ± sem (n=8).  3.3.3  Activin A- and cAMP-induced progesterone and 20cc-OH-P in accumulation  Progesterone stimulation by 8-Br-cAMP (1.5 mM) was not affected by 100 ng/ml activin A (P>0.05, Fig. 11). Activin A did not affect forskolin-stimulated progesterone accumulation (Fig. 11; P>O.05). The accumulation of 20cc-OH-P in response to 8-BrcAMP was not affected by activin A (Fig. 8; P>O.05).  3.3.4  Effect of acivin A on cellular cAMP accumulation  Incubation of human granulosa cells with hCG, with or without activin A, increased intracellular cAMP concentrations after 16 h of treatment.  There was no  difference in the intracellular cAMP concentrations between hCG and hCG  +  activin A  treated groups (Fig. 12). The level of cAMP stimulated by hCG was enhanced in the presence of IBMX (P<0.01). Activin A attenuated the cAMP response to hCG in the presence of IBMX (P<0.01) (Fig. 12). Progesterone concentrations were reduced by activin A, irrespective of the presence or absence of TBMX (P<001, Fig. 13).  67  4  12  2  2  0 G) 0 If  3. 2  C) C.) If) 0  ‘z8  r  0  0  E  E  C) 0  C) C 0  C)  0)  0  0 C)  C  1  2  0 I  2 0•  I  -]--  I L control  Figure 11.  01 CAMP  activin  control  cAMP +activin  ;-  forskolin  forskolin +activin  Effect of recombinant human activin A (100 ng/ml) on 8-Br-cAMP (1.5 M)-stimulated progesterone production by 5 mM)- and forskolin (5 x10 human granulosa-luteal cells. -Cells were incubated for 24 h. Values are mean ± sem (n=8). Similar results were obtained in other 4 experiments. Different superscripts (above the sem bars) denote statistical differences (P<0.0 1).  68  3  2000  &51500 G) 0  It)  0  !b000  2  500  >  o o  C  0 0  ><  ><  ><  x  •  C)  0  Cu +  C  0 0 -c  + >  +  0 0 -c  4-’  + C  > 44  C,  0  Cu  +  0 0 -c Figure 12.  hCG (1 lU/mi) Effect of recombinant human activin A (100 ng/ml) on presence of 113 MX stimulated cellular cAMP production, in the absence or  (0.5 mM).  Cells were incubationed for 24 h.  Intracelular cAMP was  (n=8). Similar extracted according to the text. Values are mean ± sem rscripts (above results were obtained in 3 other experiments. Different supe the sem bars) denote statistical differences (P<0.0l).  69  6  Cl)  -a;C)  2  2  51  U,  0  4. 3  E  C  31  ci)  C  0  2  ci)  -I-  (‘I  ci)  1  1•  0 0 0  ><  ><  ><  + 0  +  o  (U  +  0 0 Figure 13.  Effect of recombinant human activin A (100 ng/ml) on hCG (1 lU/mi) stimulated progesterone production, in the absence or presence of JBMX  (0.5 mM). Cells were incubationed for 24 h. Values are mean ± sem (n=6). Similar results were obtained in 2 separate experiments. Different superscripts (above the sem bars) denote statistical differences (1 vs 2, P<0.01; 1 vs 3, 2 vs 3, P<0.05).  70  3.4 DISCUSSION Peptides of the inhibin family, secreted from granulosa cells into the follicular fluid, regulate pituitary gonadotropin secretion and may also exert paracrine/auto crine functions within the ovary (Sum et al., 1989; Ying et al., 1986; Hsueh et al., 1987, Sugino et aL, 1988a; Woodruff et al., 1990). Inhibin and activin were originally identified as an inhibitor and stimulator, respectively, of FSH secretion. Inhibin and activin were first isolated from ovarian follicular fluid and subsequently shown to be products of granulosa cells. Since then, inhibin has been shown to potentiate and activin to suppress androgen production in response to LH in human theca and rat testis cells (Hillier et al., 1991a; Hsueh et al., 1987).  This provides evidence of a paracrine action of these gonadal  peptides. Reports of autocrine effects of inhibin and activin on granulosa cells have been inconsistent. Purified bovine activin was shown to suppress FSH-stimulated progesterone production in rat granulosa cells, while inhibin had no effect (Hutchinson et al., 1987). Another report suggested that activin potentiated FSH-stimulated progesterone production in rat granulosa cells (Xiao et al., 1990; Sugino et al., 1988b). Here, it was found that recombinant activin A (100 ng/ml) inhibited FSH- and hCG- stimulated progesterone production by 90% and 70% respectively, compared with the gonadotropins alone. This supports the hypothesis that activin, but not inhibin, inhibits gonadotropin-induced ovarian progesterone production. Dose-dependency experiments revealed an ED 50 value of about 3.7 ngfml (approximately 10’° M).  This may be within the range found in ovarian  follicular fluid, even though the concentration of activin in follicular fluid is low (about 10%) compared to that of inhibin (Zhiwen et al., 1987; Sugino et al., 1988b; Robertsonet  71  al., 1986, McLachlan et al., 1987b). Similar doses of recombinant inhibin A had no effects on basal or hCG-stimulated progesterone production. This observation was similar to that of a previous study of granulosa cell differentiation (Hutchinson et al., 1987). It was reported that higher levels (60 lU/mi) of inhibin were produced by rat granulosa cells when stimulated with FSH (Zhiwen et al., 1987). Thus, generation of endogenous inhibin by the cells may have affected the results from the present experiments where exogenous gonadotropins were added. Sugino et al. observed a suppressive effect of inhibin on progesterone secretion in rat granulosa cells (Sugino et al., 1988b). The possible mechanisms underlying inhibitory effects of gonadotropin-stimulated progesterone production by activin A were investigated. Inhibition of hCG-stimulated progesterone accumulation by activin A takes several hours. A similar inhibition of 20cchydroxy-progesterone accumulation by activin A suggests that activin attenuates progesterone production rather than promoting conversion of progesterone to its major 2Ocx-hydroxylated metabolite.  Possible mechanisms may involve inhibition of gene  transcription and translation of rate-limiting enzymes in the steroidogenic pathway (such as cytochrome P450 scc).  Alternately, activin may interfere with steps between the  coupling of gonadotropin receptor and the adenylate cyclase-cAMP cascade. Experiments using 8-Br-cAMP and an adenylate cyclase activator, forskolin, with or without activin, suggest that inhibitory effects of activin occur prior to gonadoiropin-induced cAMP accumulation. Using IBMX to block phosphodiesterase activity, cAMP levels in the hCG +  IBMX group was higher than the hCG group, suggesting metabolic rate of cAMP is  relatively high. hCG-induced cAMP, in the presence of IBMX, was reduced by activin  72  A.  This suggests that activin A may have reduced the cAMP response to hCG by  inhibiting cAMP synthesis rather than by enhancing phosphodiesterase-induced cAMP catabolism (Fig. 12). Although hCG-stimulated cAMP accumulation was enhanced by IBMX, progesterone accumulation was not affected (Fig. 8).  This implies that hCG  treatment resulted in a maximal progesterone response which cannot be increased by blockade of cAMP degradation, or hCG-stimulated progesterone production needs on slight increase of intracellular cAMP concentration. Progesterone accumulation in the activin A  +  hCG and activin A  +  compared to hCG alone and hCG  hCG +  +  IBMX groups were equally attenuated as  IBMX, respectively (Fig. 13). These observations  questioned that activin inhibits gonadotropin-stimulated progesterone production by blocking cAMP synthesis, therefore inhibitory effect of activin on Gn-stimulated progesterone by other intracellular pathway should also be considered. The present results suggest that activin A may serve as a negative regulator of gonadotropin-stimulated progesterone production in human granulosa cells. It is generally agreed that activin stimulates FSH secretion which, in turn, increases estrogen production in the ovary. Thus, it is plausible that the overall action of activin may be to promote follicular differentiation and delay the onset of luteinization. This could be achieved through the stimulation of FSH secretion at the pituitary and the suppression of gonadotropin-stimulated progesterone production at the ovary via an autocrine/paracrine mechanism.  73  4.0  STIMULATION OF PROGESTERONE SECRETION BY RECOMBINANT  FOLLISTATIN-288  IN  HUMAN  GRANULOSA CELLS  4.1  INTRODUCTION Follistatin production in the rat ovary occurs exclusively within granulosa cells  (Nakatani et al., 1991). Follistatin production by differentiated bovine granulosa cells harvested from preovulatory follicles is FSH- but not LH-dependent (Klein et aL, 1991). Follistatin is an activin binding protein (Nakamura et al., 1990).  Activin  suppresses gonadotropin-stimulated progesterone production by human granulosa cells (Li et al., 1992 & Chapter 3.0).  Follistatin may also locally regulate steroid hormone  production in the human ovary.  Follistatin augments FSH-stimulated progesterone  production by rat granulosa cells (Xiao et al., 1990) and enhances LH-stimulated progesterone production in undifferentiated but not in differentiated bovine granulosa cells (Shukovski et al., 1991).  Taken together with the above findings, the presence of  follistatin (Shimasaki 1989; Nakatani et al., 1991) and inhibin-B (Ling et al., 1986; Vale et al., 1986; Woodruff et al., 1988; Meunier et al., 1988) mRNA within the ovary suggests local functions of inhibin, activin and follistatin (Xiao et al., 1990; DePaolo et aL, 1991). A recombinant human follistatin (follistatin-288) has been developed (lnouye et al., 1991). The present experiments were designed to test the hypothesis of direct actions of inhibin-related peptides in human granulosa cells collected from superovulated  74  preovatory follicles.  4.2  TREATMENTS  4.2.1  Recombinant expression of human follistatin-288 Recombinant human follistatin with 288 amino acids was expressed in Chinese  hamster ovary cells under the control of the simian virus-40 promoter, as detailed in a previous report (Inouye et al., 1991).  4.2.2  Follistatin-induced steroidogenesis Recombinant human follistatin-288 (3 nM) was added in M199 with 0.5% BSA,  with or without hCG (1 UJ/ml, Sigma). Progesterone and 2Ocz-OH-P were measured.  4.2.3  Dose-dependency and lime-course study offollistatin action Increasing doses of follistatin (0.1 to 3 nM) were added to the cultured cells and  the cells were incubated for 24 h. Time-course experiments examined follistatin (3 nM) effects. Progesterone concentrations were measured in incubation media after 6, 12, 24 and 36 h.  4.2.4  Effects offollistatin on cAMP-stimulated progesterone production 8-Br-cAMP (1.5 mM, Sigma), with or without follistatin (3 nM), was added to the  cells. Progesterone and 2Oct-OH-P concentrations were measured in incubation media  75  after 24 h. Intracellular cAMP concentrations were measured.  4.2.5  Effects offollistatin on cellular cAMP accumulation Follistatin (3 nM) and hCG (1 lU/mi) were added to the cells in the presence of  0.5 mM IBMX. Progesterone and intracellular cAMP were measured.  4.3  Results  4.3.1  Dose-dependent increase of progesterone secretion by follistatin Treatment of human granulosa cells with follistatin stimulated progesterone  secretion. The effect of follistatin on progesterone secretion was dose-dependent and was significant at 0.33 nM (P<0.01) and remained elevated at 1 and 3 nM (Fig. 14).  4.3.2  Effect offoiistatin on basal and hCG-stimulated progesterone production The effect of recombinant human follistatin was examined in the presence or  absence of exogenous gonadotropin (Fig. 15). Follistatin (3 nM) increased (P<0.0l) basal progesterone secretion by approximately 3-fold in the absence of exogenous gonadotropin (Fig. 15). Progesterone concentrations were increased 3.5 fold by hCG (1 lU/mi, P<0.01). Follistatin did not affect hCG-stimulated progesterone secretion (ie. in cells where follistatin and hCG were tested together, follistatin did not augment hCG- stimulated progesterone secretion (P>0.05).  76  11  **  **  **  U)  0 L()  0  7.  E U) C  0  ci)  5  (I) ci) 0)  0  /1  03.  1  0  1  [0  Follistatin (nM) Figure 14.  Effect of recombinant human follistatin-288 (0.1-3 nM) on basal progesterone secretion. Cells were incubated for 24 h. Values are mean ± sem (n=8). Similar results were obtained in 3 separate experiments.  P<o.01.  77  **  3O 2 2  0 G) C.)  °20  olO  o a  control  hCG  follistatin  hCG+ follistatin  Figure 15.  cAMP  cAMP+ follistatin  Effect of recombinant human follistatin-288 (3 nM) on basal and hCG (1 IU/ml)/8-Br-cAMP (1.5 mM)-stimulated progesterone secreti on by human granulosa cells. Cells were incubated for 24 h. Values are mean ± sem (n=8).  Similar results were obtained in other 4 experiments. Different  superscripts (above the sem bars) denote statistical differences (P<O.Ol).  78  4.3.3  Dose-dependent increase in 20a-OH-P production by folliskitin To investigate the possible effect of follistatin on progesterone metabo lism, the  concentration of 2Oct-OH-P in the culture medium following a 24 treatm ent period was determined.  Similar to the effect on progesterone accumulation, basal 20a-OH -P  secretion was markedly enhanced in the presence of increasing doses of follistatin (Fig. 16). The stimulation by follistatin became significant at a dose of 1 nM.  4.3.4  Effect offollistatin on hCG-stimulated 2Oo-OH-P production Treatment of the cells of hCG (1 lU/mi) for 24 h significantly enhanc ed the  accumulation of 2Oct-OH-P in the culture medium (Fig. 17). A similar stimulation was observed with follistatin (3 nM).  In cells that were treated with follistatin and hCG  concomitantly, the accumulation of 2Oct-OH-P was not stimulated further when compared with either treatment alone (P>-O.05).  4.4.5  Time-dependent increase of progesterone production by follistatin Follistatin stimulated progesterone secretion at 24 h in time-course experim ents  (Fig. 18, P<O.O1).  4.4.6  Effect of follistatin on cAMP-stimulated progesterone and 20-ct-O H-P production 8-br-cAMP-stimulated progesterone and 2Oct-OH-P secretion were not affecte  d by  follistatin (3 nM) (Figs. 15 and 17, P>O.05).  79  12 *  a -;  11-  CI)  If)  0  10  E  C  G)  C 0  a) (I) G)  0) 0  8  1  aI I 0 ed  6  I  1  10  Follistatin(flM)  -  Figure 16.  .1  0  Effect of recombinant human follistatin-288 (0.1-3 nM) on basal 20-OHP production. Cells were incubated for 24 h. Values are mean ± sem (n=8).  *  P<0.05.  80  (I)  ci) C) LI)  C  0  E C  ci C  0 0) (I)  0) z:y) 0  0  0 C  c”J  control  hCG  follistatin hCG+ follistatin  Figure 17.  cAMP  cAMP+ follistatin  Effect of recombinant human follistatin-288 (3 nM) on basal and hCG (1 IU/ml)/8-Br-cAMP (1.5 mM)-stimulated 2Oc-OH-P production by human granulosa cells. Cells were incubated for 24 h. Valu es are mean ± sem (n=8).  Similar results were obtained in other 4 experiments. Different  superscripts (above the scm bars) denote statistical differences (P<O.05).  81  (n=8). Similar results were obtained in other 4 experiments. Different superscripts (above the sem bars) denote statistical differenc es (P<O.05).  4.4.7  Effect of Follistatin on cellular cAMP accumulation  Incubation of human granulosa cells with follistatin or hCG, in the presence of IBMX, resulted in increased cellular cAMP concentrations after 24 h incubation (Fig. 19, P<zO.O1). Cellular cAMP concentrations between hCG and follistatin treated groups were similar (P>O.05). In these experiments, progesterone conc entrations in the incubation media were increased by follistatin or hCG in the presence of IBMX (P<O.O1). The effects of follistatin and hCG on the accumulation of cAM P and progesterone were not additive (P>O.05).  82  50  **  control  I  Follistatin  40 **  U)  ci) C.) If)  CD  30  1  0  E ci)  20  0 G) U)  ci) 0 0  10•  0 0  10  I  I  20  30  40  Follistatin (hr) Figure 18.  Time-course effect of recombinant human follistatin-288 (3 nM) on basal progesterone production. Cells were incubated for 24 h. Values are mean ± sem (n=8).  **  P<O.O1.  83  (I)  30  C) C-)  CD  0  C) C 0  20  10  C) C,)  C) 0) 0  0  a-  control  hCG  follistatin  hCG+ follistatin  hCG  hCG+ follistatin  200 0  2  C])  150 LI) 1  control follistatin  Figure 19.  Effect of recombinant human follista tin-288 (3 nM) and hCG (1 lU/mi) on progesterone secretion (upper pan el) and cellular cAMP accumulat ion (lower panel) in the presence of iso-but yl-methyxanthine (IBMX, 0.5 mM). Cells were incubated for 24 h Intracellular cAMP was extracted according to the text. Values are mea n ± sem (n=8). Similar results were obtained in other 3 experiments. Different superscripts (above the sem bars) denote statistical differences (P< 0.01).  84  4.5  Discussion Follistatin-288 is 8-10 times more potent than the native protein and similar to the  potency of inhibin-A in vitro, but more potent and longer acting than inhibin-A in vivo (Xiao et al., 1990). The physiological role of follistatin in the human ovary has not been reported. Although follistatin has thus far been measured only in human ovarian follicular fluid (Schneyer et at, 1992, Robertson et al., 1990), a number of smdies have localized follistatin message and/or protein in the granulosa cells of ovaries from other species (Shimasaki, 1989; Nakatani et at, 1991; Ying 1988; Findley et al., 1990). The present observation that follistatin stimulates basal progesterone and 20a-OH-P accumulation in human granulosa cells provides the first evidence for a direct action of follistatin in the human ovary and suggests the potential for an autocrine role. It has been demonstrated that activin-A decreased hCG-stimulated progesterone and 20o-OH-P production in human granulosa cells (Saito et al., 1991). Follistatin is a binding protein of activin (Robertson et al., 1990; Nakamura et al., 1990). Thus, it can be postulated that a balance between follistatin and activin-A activity contributes to the regulation of the granulosa cell response to gonadotropins. Follistatin increased basal progesterone concentrations with equal potency to hCG, suggesting that follistatin may be an important regulator in the ovary. Follistatins are produced from differentiated granulosa cells in an FSH-dependent manner (Klein et at, 1991).  Increased  FSH is necessary for development of antral  preovulatory stage (Meunier et al., 1988c),  follicles to the  a phase in follicular maturation when  synthesis of follistatins is stimulated (Klein et al., 1991). Increased basal progesterone  85  secretion from human preovulatory granulosa cells by follistatin suggests that follistatin could in part contribute to the rise in progesterone before the preovulatory LH surge. Higher progesterone concentrations in larger follicles than in smaller follicles before the LH surge (Schwa!! et al., 1990), as well as an elevated circulating progesterone concentration, may facilitate estrogen stimulation of the LH surge (Woodruff et al., 1988). The ED 50 for the increase in progesterone secretion by follistatin was about 10 M, supporting the hypothesis that the effects of follistatin were physiological.  The  mechanism of action of follistatin is not known. Our results show that increased basal progesterone is accompanied by an increase in cAMP accumulation. In the presence of IBMX, which blocks phosphodiesterase activity, cellular cAMP concentrations in the follistatin group were higher than in the control group (Nakatani et al., 1991). Although the receptor of follistatin has not been identified, follistatin-induced cAMP accumulation suggests that the putative follistatin receptor could be coupled with a GJadenylate cyclase complex. Whether or not cAMP serves as a second messenger for follistatin-stimulated progestin accumulation cannot be determined from the present experiments.  It is  interesting to note that in the presence of hCG or exogenous cAMP, follistatin falls to affect progestin accumulation.  This may have resulted from the already maximal  progesterone production in the presence of hCG (1 lU/mi) or cAMP (1.5 mM). The timecourse studies demonstrated that the follistatin-induced increase in progesterone secretion was observed at 24 h, which was delayed compared to hCG-stimulated progesterone production. The slower time-course of action of follistatin may implicate an indirect action such as through the binding of activin (Robertson et al., 1990; Nakamura et al.,  86  1990) and disinhibition of activin suppression of progesterone synthesis (Saito et aL, 1991).  This potential mechanism of action remains to be investigated.  The other  possibility that follistatin might have blocked progesterone metabolism was examined by measuring 20a-OH-P concentrations in the follistatin-treated cells. The observation of parallel stimulatory effects of follistatin on progesterone and 20a-OH-P concentrations argues against a suppression of the 20x hydroxylase step. The results of this study differ from report of the effect of follistatin on bovine granulosa cells (Hillier et al., 1991b, Hsueh et aL, 1987). Follistatins had no effect on differentiated bovine granulosa cells but augmented LH-stimulated progesterone production in the undifferentiated stage of such cells (Hillier et al., 1991b).  In rat  granulosa cells, follistatin augments FSH-stimulated but not basal progesterone production (Xiao et al., 1990). The cause for this apparent species difference in the granulosa cell response to the direct action of follistatin is not known. It might be due to different degrees of the granulosa cell differentiation in different studies. In summary, follistatin increased basal but not hCG- or cAMP- stimulated progesterone secretion by differentiated human granulosa cells. Increased basal progesterone concentrations before the LH surge may partly result from increased follistatin production within preovulatory follicles.  Although the mechanism of follistatin is unknown, the increase of basal  progesterone secretion by follistatin is accompanied by an elevated intracellular cAMP formation. The stimulatory effects of follistatin on progesterone and 20a-OH-P secretion suggest that follistatin may be a potential regulator of ovarian steroidogenesis.  87  5.0  EFFECT OF ANGIOTENSIN III ON PROGESTERONE PRODUCTION BY HUMAN GRANULOSA CELLS  5.1  INTRODUCTION Prorenin, renin and A II were found within the stromal, thecal and luteal cells  of human and rat ovaries using immunochemical methods. Granulosa cells of preovulatory follicles and atretic follicles stain intensely for both renin and A II (Palumbo et al., 1989). The principal function of the renin-angiotensin system may involve regulation of cardiovascular homeostasis. However, increasing evidence suggests other roles for this regulatory system in endocrine tissues. Specifically, components of the renin-angiotensin system appear to alter ovarian steroidogenesis in several mammalian species (Bumpus et al., 1988; Fernandez et al., 1985; Culler et al., 1986; Jerry et al., 1988). A II has no effect on basal progesterone production by cultured bovine granulosa cells and inhibits LH-stimulated progesterone production (Sterling et at, 1990). A II has a brief half life, being rapidly cleaved by angiotensinase to the active septapeptide, A Ill. It appears that there are no reports about the effect of A III on steroidogenesis in the ovary from different species. The angiotensin-converting enzyme which converts decapeptide A I to the actapeptide A II and the angiotensinase which converts the hepapeptide A II to A III have been reported to exist in rat ovarian follicles (Daud et al.,1990; Speth et al., 1988). This suggests that ovary may secrete its own A Ill. The presence of  88  angiotensin receptors in granulosa and thecal cells of different species supports the postulate of a biological role for angiotensins in the ovary (Pucell et al., 1987; Speth et al., 1988; Husian et al., 1987; Miyazaki et al., 1988). In addition, both A II and A III have been reported to bind to the similar receptors (Goodfriend, 1991). However, the function(s) of A II and A III in the human ovary are unclear. This study examined the role of angiotensins in progesterone production by human granulosa-lutein cells. These cells are presumably programmed to undergo atresia or luteinization and are uniquely suited to the study of the autocrine and paracrine role of angiotensins in human ovary.  5.2  TREATMENTS  5.2.1  A II and A III conjugation and binding Fluroisothiocyamate labelling of A II and A Ill was performed as reported by  Woodruff 1991 with the following modifications. A II and A III (1 mg, Peninsula, Belmont, CA) were mixed with 100 pg fluorescein isothiocyanate (FITC, Sigma) in 1 ml 1 M sodium carbonate, pH 9.5. The mixture was stirred for 1 h and pH balanced to 9.0-9.5 at 20 C. Conjugated and unconjugated FITC were separated using a G-25 sephadex column eluted with 0.01 M sodium phosphate buffer, pH 7.4. The FITC conjugated A II and A III were determined by measuring the final protein concentration and fluorescence with a spectrometer (280 nm) and fluorometer (excitation 488 nm, emission 525 nm). Binding activity was measured by incubating  89  granulosa cells (in M199 with 0.5% BSA) with FITC-conjugated A II and A III for 30 mm at 37 C. Cells were washed 3 times with M199 containing 0.5% BSA. Nonspecific binding was determined in the presence of 100-fold excess of unlabelled A II and A Ill. Binding is reported as florescent intensity versus nonspecific binding using a Nikon florescent microscope (Nikon type 108).  5.2.2  Effects of A II and A III on hCG-stimulated steroidogenesis Effects of A II and A III on steroidogenic response to hCG were examined by  adding 1 lii hCG/ml (Sigma) with or without 10 M A II and A Ill. Progesterone and 5 2Ocx-OH-P were measured by RIA.  5.2.3  Dose-dependency and time-course study of A III on hCG-stimulated steroidogenesis Increasing doses of A III M 5 9 (10 10 ) were added with hCG (1 UJ/mi) to the  cultured cells and incubated for 24 h. The incubation media were collected for progesterone and 20a-OH-P RIA. hCG (1 lU/mi) was added to cells with or without A Ill 5 (10 M ) and incubation media were collected at 6, 12, 24 and 36 h for progesterone time-course analysis.  5.2.4  Effects of A III on cAMP-stimulated progesterone production 8-Br-cAMP (1.5 mM, Sigma) was added to cells with or without A III 5 (10 M ).  Progesterone was determined.  90  5.2.5  Effects of PMA on hCG- and cAMP-stimulated progesterone production Phorbol 1 2-myristate 13-acetate (PMA, 1 0 M, Sigma) was added to the cells 7  with or without hCG (1 LU/mi) and cAMP (1.5 mM). Progesterone was measured.  5.2.6  Dose-dependent effect of PMA on hCG-stimulated progesterone production PMA 6 (10’°-10 M ) was added to cell culture media with hCG (1 lU/mi).  Progesterone was measured.  5.2.7  Effect of A Ill on hCG-stimulated progesterone production in cells pre treated with PMA (1(r M PMA 7 ) was added to cell culture media. Cells were rinsed 3 times in  M199 with 0.5% BSA at 24 h. A III 5 (10 M ) with or without hCG (1 lU/mi) was added to the cells. Media were collected at 24 h for progesterone determination.  5.2.8  Additive effect of PMA and A III on hCG-stimulated progesterone production hCG (1 lU/mi), hCG with PMA 7 (1(Y M ), with or without A Ill (106 and 5 10 M )  were added to the cells. Progesterone were measured.  5.3  RESULTS  5.3.1  A II and A III binding to cultured human granulosa cells The binding activity of A II and A III was examined using UV light after  91  incubation of FITC-labelled A II and A III with human granulosa cells. autofluorescence was observed for unlabelled proteins.  No  The binding specificity was  determined by incubation with 100 fold higher concentrations of unconjugated A II and A III. Results showed that A II and A III specifically bound to human granulosa cells (Fig. 20, 21).  5.3.2  Effect of A III on hCG-stimulated progestin accumulation  Treatment of the cells with A U or A III 5 (10 M ) did not affect basal progesterone synthesis (P>0.05, Fig. 22). Progesterone concentrations were increased 2-3 fold by hCG (P<0.01). A Ill suppressed hCG-stimulated progesterone production (P<0.01). A II was not effective in this regard (P>0.05). A III also had a trend to suppress hCG-stimulated 2Oct-OH-P accumulation (Fig. 23).  5.3.3  Dose- and time-dependent effects of A III  A III suppression of hCG-stimulate progesterone production was dose-dependent. Effect of A Ill on hCG-stimulated P 4 production was observed at 10 M (Fig. 24). The 7 50 for the effect of A Ill on hCG-stimulated progesterone production was 3.2 ± 0.38 ED x 10 M (based on 2 separate experiments). 8  Time-course experiments revealed that  progesterone production was inhibited by A III at 24 and 36 h of treatment (P<0.0l, Fig. 25).  92  -  -  --:: ..-  . ..  .••  •s..  Figure 20.  •  The binding sites of FITC-labeled A II on human granulosa-luteal cells. The cultured granulosa cells were incubated with (1,2) or without (3,4) 100 x concentration of A II for 30 mm at 37 C before incubation with FITC labeled A II.  The cells were washed with 0.5% BSA M199 for three  times. The cells were photographed using Nikon microscope (x 400) by normal light (1,3) or ultraviolet dark fields (2,4).  93  . • ..  ..  ,  4I . ‘  Figure 21.  ..  The binding sites of FITC-labeled A III on human granulosa-luteal cells. The cultured granulosa cells were incubated with (1,2) or without (3,4) 100 x concentration of A Ill for 30 mm at 37 C before incubation with FITC labeled A III. The cells were washed with 0.5% BSA M199 for three times. The cells were washed with 0.5% BSA M199 for three times. The cells were photographed using Nikon microscope (x 400) by normal light (1,3) or ultraviolet dark fields (2,4).  94  16  14 C,)  C)  0  1,2  12  U)  100  E  1,2  8 C) 1  0 1 C)  1  6  (1)  C)  0 0  4.  2  control  Figure 22.  All  Aill  hCG  hCG+All hCG+Alll  Effects of 10 M A IT/A III on basal and hCG (1 IU/ml)-stimulated 5 progesterone production by human granulosa cells. Cells were incubated for 24 h. Values are mean  +  sem (n=8). Similar results were obtained in  other 4 experiments. Different superscripts (above the sem bars) denote statistical differences (P<O.O 1).  95  5.3.4  Effect of PMA on progesterone production PMA 7 (10 M ) did not affect basal progesterone synthesis (P>0.05, Fig. 26).  Progesterone concentrations were increased 2-3 fold by hCG (P<O.O1). PMA inhibited hCG-stimulated progesterone production (P<0.O1). The effect of PMA on progesterone  production in response to hCG was dose-dependent (Fig. 27). The minimum effective concentration of PMA was 10 M (P<0.01). The ED 8 50 for the effect of PMA on hCG stimulated progesterone production was 8.7 ± 0.6 nM (based on 2 separate trials. The inhibitory effects of A Ill and PMA was additive (P<O.01), (Fig. 28).  5.3.5  cAMP-stimulated progesterone production 8-Br-cAMP-stimulated progesterone production (1.5 mM) was not affected by A  III 10 M or PMA 10 5 M (P<0.05, Fig. 29). 7  5.3.6  Effect of A III in cells pre-treated with PMA A III suppressed hCG-stimulated progesterone production after 24 h of PMA pre  treatment (P<0.01, Fig. 30).  96  30  **  (I)  C) LC)  0 1  0  E  zL  20  C 0 I-. C) C,) C) C) 0 L. 0  10  0I 0  control  Figure 23.  hCG+Alll hCG+All  hCG  ) on hCG (1 IUIml)-stimulated 20a-OH-P (l(Y M Effects of A Ill 5 production. Cells were incubated 24 h. Values are mean **  p<o.ol. 97  +  sem (n=8).  18  16  I  ‘ii  U)  15 14 C)  to 0  0 C **  a)  C 0  G)10  ** **  Cl)  a) 0  8  6  //—  0  .  .01  1  .1  10  100  Angiotensin Ill (lOOnM) Figure 24.  Effects of A UI (1O  -  5 M) on hCG (1 IU/ml)-stimulated progesterone i0  production. Cells were incubated for 24 h. Values are mean Similar results were obtained in other 2 experiments.  98  **  +  sem (n=8).  p<O.Ol.  70  hCG  o—  hCG+AIII  60  Co C)  50  IC)  0 0  E  C 0  40  30  Cl) 0) 0 L. 0  20  10 0  10  30  20  40  Time (hr) Figure 25.  Temporal profile of the effects of A III (1O- M) on hCG (1  IlJIml)  stimulated progesterone production by human granulosa cells. Cells were incubated for 24 h. Values are mean obtained in other 2 experiments.  99  **  +  sem (n=8). Similar results were  p<O.Ol.  0 0 U)  0  E G) 0 G) I  4-.  (I) 1  0  1  0’ I  0  PMA  control Figure 26.  hCG  hCG+PMA  Effects of PMA (1O M) on hCG (1 IU/ffil)-stimulated progesterone production by human granulosa cells. Values are mean  +  Cells were incubated for 24 h.  sem (n=8). Similar results were obtained in other 3  experiments. Different superscripts (above the sem bars) denote statistical differences (P<O.O 1).  100  12  c-/I  10U)  a) C) LC)  **  8  1  0  E ‘G) C  **  0  C) Cl) C)  0  L.  0.  **  4  2  //  1  .1  0  10  100  1000 10000  PMA (nM) Figure 27.  Effects of PMA (10’° are mean  +  experiments.  -  106 M) on basal progesterone production. Values  sem (n=8). **  Similar results were obtained from other 2  p<O.Ol.  101  2  3  0 G) C) U) 0  3  0  E 1  G) 1  0 (I)  1  G)  0 0 L.  control  hCG  hCG+PMA  hCG+PMA hCG+PMA  hCGi-Alll  +AlIl1OM  Figure 28.  ) Effects of A Ill 5 -10 6 (10 M  +  M 5 +A11110  PMA (lOg M) on hCG (1 IU/ml)-stimulated  progesterone production. Values are mean  +  sem (n=8). Similar results  were obtained in other 3 experiments. Different superscripts (above the sem bars) denote statistical differences (P<OO 1).  102  5.4 DISCUSSION Result with fluorescent-labelled A II and A III indicated that human granulosa cells are sites of both A II and A III action. This result agrees with the previous finding that the A II receptors existed in murine and bovine ovary or follicles (Pucell et aL, 1987; Daud et al.,1990; Speth et aL, 1988; Husian et al., 1987; Miyazaki et al., 1988) and A II and A Ill may share the same receptors (Goodfriend, 1991). Dose-dependent suppression of hCG-stimulated progesterone production by A III suggested that A III, but not A II, is an active inhibitor of progesterone production in the human ovary. The possibility that the des-aspartic acid’ derivative of A II (A III) is an agonist with potency exceeding its progenitors in certain specific sites of action has been proposed (Goodfriend 1991). Whereas A III is a less potent vasoconstrictor than A II, A III is as potent as A II in the adrenal glomerulosa (Cook et aL, 1990). A III may also be the active angiotensin in the brain (Wright et al., 1990; Einspanier et al., 1991). Tn the adrenal gland, A III is a powerful stimulator of aldosterone secretion, and stimulates prostaglandin formation. The biological effects of A II in ovarian cells are obscure. A II secretion may be episodic in pre- and post-ovulated porcine follicles but the episodes appear to have no effect on steroidogenesis (Lacy et al., 1989). The effects of A II in this study showed inconsistent results. Other studies have shown that A II interferes with steroidogenesis in the ovary of other mammals and amphibians (Kuo et al., 1991; Lightman et al., 1987).  This  suggests species differences or different rates of degradation of the angiotensins in different species. Alternatively, A Ill may be formed by the sequential action of the aminopeptidase and converting enzyme on the decapeptide angiotensin I.  103  8.0W  0 C)  6.0  LC)  0 0  E C)  4.0  0 C) 0 C) 0  L..  0  2.0-  control  Figure 29.  All!  cAMP cAMP+ All  PMA  cAMP+ PMA  ) on cAMP (1.5 mM)-stimulated 0 7 (1 ) and PMA M 0 5 (1 Effects of A III M d granulosa cells. Cells were incubate progesterone production by human for 24 h. Values are mean  +  ained in sem (n=8). Similar results were obt  ote erscripts (above the sem bars) den other 5 experiments. Different sup statistical differences (P<O.05).  104  2  (0 C) C.) L()  0 1  0  E  1 1  C)  0 C) U)  C) D) 0 0  control Figure 30.  hCG  Alli  hCG+Alll  Effects of A UI (l0- M) on hCG-stimulated progesterone production by ) pre-treated cells. Cells were pretreated with PMA for 24 h. (10 M PMA 7 After pretreatment, cells were washed three times, and treated with A III and hCG for 24 h. Values are mean+sem (n=8).  Similar results were  obtained in other 3 experiments. Different superscripts (above the sem bars) denote statistical differences (P<0.0l).  105  The angiotensin converting enzyme has been reported to exist in the rat ovary (Sterling et aL, 1990; Bumpus et aL, 1988). The ED 50 for A Ill suppression of hCG stimulated progesterone production was about 10 fold of the KDa (3 nM) for A Ill binding (Auletta et al., 1988), which suggested the physiological significance of the action of A Ill. It is possible that A II is secreted into the follicle and then converted to A III by an angiotensinase and A Ill but not A II, suppresses hCG-stimulated progesterone production by the human ovary. LH treatment increased renin-like activity and A II immunoactivity in human follicular fluid, suggesting that LH is a major factor in the regulation of the ovarian renin-angiotensin system (Carrithers et al., 1990). Therefore, there may be feedback loop between renin-angiotensin system and LH, which regulates their actions in human ovary. Intracellular mechanisms underlying the inhibitory actions of A UI on hCG stimulated progesterone production are unknown. The A II receptor has been reported to couple with adenylate cyclase and PLC and A II and A ifi increased inositol polyphosphates in a dose-dependent manner (Audinot et al., 1991; Leung et al., 1988; Davis et al., 1989). DAG is generated by the activation of PLC. The phorbol ester PMA, like DAG, is an activator of PKC. PMA has been reported to inhibit FSH-stimulated progesterone production in the rat (Spat, 1988) and hCG-stimulated cAMP production (Kojima et al., 1985) in cultured human granulosa cells. This study examined the effect of PMA on hCG-stimulated progesterone production in cultured human granulosa cells. The data suggested that PMA treatment of granulosa cells mimicked the effect of A III in human granulosa cells.  106  A II has been shown to stimulate polyphosphoinositide turnover and increase formation of DAG and inositol trisphosphate in several tissues (Kojima et aL, 1985; Wang et al., 1985).  The possibility that the inhibition of hCG-stimulated progesterone  production by A UI is mediated via the activation of PKC was examined. The results of this study suggested that suppression of hCG-stimulated progesterone production by A III and PMA were additive.  Further, the inhibitory effects of A Ill on hCG-stimulated  progesterone production were observed in cells pre-treated with PMA. This suggested that the inhibitory action of A III may not involve PKC activation, but may occur via  other signal pathways such as the adenylate cyclase system and calcium and calciumbinding protein system. Indeed, A II has been reported to have some effect on [Ca j in 2 rat granulosa and luteal cells (Wang et al., 1989; Currie et al., 1992). Further studies are necessary to elucidate the intracellular signalling mechanisms of A III. In summary, A III inhibits hCG-stimulated progesterone production in human granulosa cells. A II at similar concentrations does not produce a consistent effect. This suggested that A III may be the dominant effector on steroidogenesis in the human ovary. Binding sites of A II and A Ill, as detected by fluorescent-labelling, provided further evidence that human granulosa cells are targets of angiotensin action. Although PMA inhibited hCG-stimulated progesterone production in human granulosa cells, the action of A Ill may not involve the activation of PKC.  107  6.0  EFFECT OF GnRH ON PROGESTERONE PRODUCTION BY HUMAN GRANULOSA CELLS  6.1 INTRODUCTION Circulating GnRH concentrations in non-pregnant animals are considered too low to interact with extrapituitary tissues (Marchetti et al., 1988; Eskay et al., 1977; Nett et al., 1974). However, membrane-bound GnRH receptors and GnRH-lilce peptides appear to be present in the ovary of some mammalian species, including the human.  Such  locally produced GnRH potentially modulates ovarian activity via paracrine and autocrine pathways (Hsueh et al., 1981a; Pieper et al., 1981; Jones et al., 1980; Hillensjo et al., 1980; Latouche et al., 1989).  GnRH activity within the ovary may be inhibitory or  stimulatory. Notably, GnRH is thought to suppress gonadotropin-induced steroidogenesis (Latouche et al., 1989) hut stimulates oocyte maturation and ovulation (Hsueh et al., 1981a; Latouche et al., 1989).  A GnRH agonist suppressed basal, LH- and FSH  stimulated cAMP accumulation and progesterone secretion in porcine granulosa cells (Hsueh et al., 1988; Harwood et al., 1980). LH/hCG exposure may induce FSH-sensitized granulosa cells to produce a GnRH like peptide (Hsueh et al., 1988). Sufficient, perhaps continuous, exposure to such a factor could act at pituitary GnRET receptors to desensitize gonadotropes, explaining the absence of an endogenous LH surge after treatment with a GnRH agonist (Hsueh et al., 1988). GnRH  may  directly  suppress  FSH-induced  108  adenylate  cyclase  activity,  steroidogenesis and gonadotropin receptor expression in rat, pig, hamster, rabbit and chicken granulosa and luteal cells (Latouche et aL, 1989; Demoulin et al., 1991). GnRH agonists also suppressed progesterone production from spontaneously luteinized human granulosa cells in vitro (Knecht et aL, 1983a) and granulosa-lutein cells from preovulatory follicles aspirated after hCG administration to in vitro fertilization patients (Parinaud et aL, 1988).  Conflicting studies suggest that GnRH does not act at the ovary to alter  gonadotropin-stimulated steroidogenesis in vitro (Knecht et al., 1981a; Bramley et al., 1985), and undifferentiated granulosa cells from small human follicles were not susceptible to GnRH-suppression (Casper et al., 1982).  6.2  TREATMENTS  6.2.1  GnRH, GnRH-agonist, and gonadotropin-regulated steroidogenesis Effects of GnRH (Sigma) and GnRH-agonist J 6 ([D-A1 -LHRHa , Sigma) on  steroidogenic response to gonadotropins were investigated by adding hCG (1 lU/mi), with or without GnRH 7 (10 M ), or GnRH-agonist.  Human granulosa-luteal cells were  incubated with GnRH for 24 or 48 h. Progesterone was measured by RIA.  6.3 RESULTS  6.3.1  Effect of GnRH on hCG- and cAMP-stimulated progesterone accumulation  109  Treatment of the cells with GnRH 7 (10 M ) for 24 h did not affect basal progesterone synthesis (P>O.05, Fig. 31). Progesterone concentrations were increased 2-3 fold by hCG (P<O.O1). GnRH had no effect on hCG-stimulated progesterone production (P>O.05, Fig. 31).  Progesterone concentrations were increased 2-3 fold by a cAMP  analog, 8-Br-cAMP. GnRH had no effect on cAMP-stimulated progesterone production (P>O.05, Fig. 31). Increased incubation time to 48 h had same result as that of 24 h incubation.  6.3.2  Effect of GnRH-agonist on hCG- and cAMP-stimulated progestin accumulation Treatment of the cells with GnRH-agonist 7 (lO M ) did not affect basal  progesterone synthesis (P>O.05, Fig. 32). Progesterone concentrations were increased 2-3 fold by hCG (P<O.O1). GnRH-agonist had no effect on hCG-stimulated progesterone production (P>O.05, Fig. 32). Progesterone concentrations were increased 2-3 fold by a cAMP analog, 8-Br-cAMP.  GnRH-agonist had no effect on cAMP-stimulated  progesterone production (P>O.05, Fig. 32).  110  3.  2  U)  2  2  C,  control  hCG  hCG+  GnRH  cAMP+ GnRH  GnRH  Figure 31.  CAMP  ) on basal and hCG (1 RJ/ml)-stimulated (1O M Effect of GnRH 7 progesterone production by human granulosa cells. Cells were incubated for 24 and 48 h. Values are mean ± sem (n=24). Different superscripts (above the sem bars) denote statistical differences (P<O.O 1).  111  U) 0 C) U, 0  0  E C 0 1.  0 (I) 0 3) 0  4-  0  Control  Figure 32.  hCG  hCG + GnRH Analog  CAMP CAMP + GnRH Analog  M) on basal and hCG (1 IU/ml)/cAMP (1.5 7 Effect of GnRH agonist (10 mM)-stimulated progesterone production by human granulosa cells. Values are mean ± sem (n=6). Different superscripts (above the sem bars) denote statistical differences (P<O.O 1).  112  6.4 DISCUSSION LH/hCG treatment induced GnRH-like peptide secretion from FSH- sensitized granulosa cells (Hsueh et al., 1981a). The effect of GnRH on ovarian steroidogenesis is inconsistent.  The results from murine, porcine, hamster, rabbit, chicken and human  granulosa and luteal cells showed that GnRH suppress FSH-induced steroidogenesis and gonadotropin receptor expression (Massicotte et al., 1980; Hillensjo et aL, 1980; Demoulin et al., 1991; Wickings et aL, 1990; Behrman et at, 1980).  Other studies  showed that GnRH did not have effects on ovarian steroidogenesis (Bramley et al., 1985; Casper et at, 1982). Results obtained here support the contention that GnRH does not affect progesterone production by human granulosa cells. The lack of effect of a GnRH agonist suggested that the lack of effects of GnRH was not because of its short half life. Incubation for 48 h with GnRH did not alter the results. This suggested that GnRH does not act at the ovary to alter gonadotropin-stimulated steroidogenesis. GnRH also lacked the ability to affect intracellular Ca 2 in human granulosa cells (Currie et al., 1992). Ca 2 is required for GnRH-suppression of cAMP and steroid production during prolonged incubation of rat ovarian cells (Sharpe et al., 1990). The reason for the conflicting results is unknown. Individual variation of granulosa cells from different patients and different pretreatment could be the explanation.  113  7.0  EFFECTS OF PROSTAGLANDIN F ON PROGESTERONE PRODUCTION BY HUMAN GRANULOSA CELLS  7.1  INTRODUCTION Studies using dispersed luteal cells from a variety of non-primate mammals have  revealed that PGF is a potent antagonist of LH/hCG action (Niswender et al., 1985; Lahav et al., 1976; Behrman et al., 1979; Richardsonm 1980; Kenny et al., 1986; Harrison et al., 1987; Rogers et al., 1983, 1985). Evaluation of data in primates shows that the model used (dispersed vs. whole tissue), the concentration of PGF used and the age of the corpus luteum are important variables in establishing conclusions.  Studies using  granulosa cells from preovulatory follicles and minced luteal tissue from monkeys and humans (Channing et al., 1975; Rogers et al., 1985; McNAtty et al., 1975; Handerson et al., 1977; Auletta et aL, 1983; Dennefors et al., 1982; Hamburger et aL, 1979; Patwardhan et al., 1974) have shown that PGF inhibits basal and LH-stimulated progesterone production.  Incubations of dispersed luteal cells produced conflicting results. Using  dispersed human luteal cells and lower concentrations of PGF, PGF was found to promote hCG-stimulated progesterone production during early and mid-luteal phases (Richardson et al., 1980). This stimulatory effect of PGF was lost in the late luteal phase. Thus, a clear-cut effect of PGF in the human ovary is lacking. The studies here were designed to examine effects of PGF on steroidogenesis in human granulosa cells.  114  7.2  TREATMENTS  7.2.1  PGF , 2 ,2 PGF agonist, and gonadotropin-regulated steroidogenesis Effects of PGF (Sigma) and the PGF-agonists, cloprostenol and lutalyse (Gifts  from Dr. Rajamahendran, Univ. of British Columbia) on steroidogenic response to gonadotropins were investigated by adding hCG (1 lU/mi) with or without PGF  7M) 1 ( 0  or agonist. Progesterone were measured.  7.2.2  Dose-dependency of PGF a on hCG-stimukxted steroidogenesis 2 Increasing doses of PGF 6 (l(Y’°-10 M ) were added with hCG (1 lU/mi) to the  cultured cells and the cells were incubated for 24 h. The incubation media were collected for progesterone RIA.  7.3  Results  7.3.1  Effects of PGF -agonists on gonad4tropin-stimulated progesterone 2 a and PGF 2 accumulation PGF 7 (10 M ) did not affect basal progesterone synthesis (P>O.05). Progesterone  concentrations were increased 2-6 fold by hCG (P<O.Ol). PGF did not inhibit hCG stimulated progesterone production (Fig. 33).  Cloprostenol inhibited hCG-stimulated  progesterone production after 6 h incubation (P>O.05, Fig. 34). Lutalyse increased basal and hCG-stimulated progesterone production after 24 h incubation (Fig. 34, P<O.05).  115  7.3.2  Dose-dependent inhibition of progesterone production by PGF a 2 c at different doses had variable effects on progesterone production in 2 PGF  response to hCG. PGF 2 numerically decreased hCG-stimulated progesterone production in 6 trials, the decrease being significant in only 2 trials at PGF (1(Y ) and (106), 9 respectively (Fig. 35).  116  I  OCD  C)  9)  CD  C -t  CD  3  Cd)  CD  CD  -  9)  ‘-‘  .  ‘d CD -t Cd) C)  Cd)  -  CD  CD C) CD  -t  CD  -  -h 0  0  Cl)  ‘:-‘  00  ‘-S  CD  Cl)  CD  e  <  C)  CD  C  C  (Q CD  CD i  d  CD ;<  9’  C)  5-.-  .•  C)  —.  C  I  rn  I  C)  p  ,  -o  p o ,  1.3  •ti  C)  -a  1.3 C, +  p  0  C,  0  r  9  T1 1.3  C,  9+  C,  0  0  0  0  C  ‘,‘,  ,  1%)  5%555%”’  i—1  p  ceIls nmol/1O 5  ’..’.’C\’s’s’s’s”s’s”S ’ 5 ‘  ,‘z’z’,,  ‘‘SN’’’’’ titt/t’’’  Progesterone  -L  C)  5 cells Progesterone nmol/10  C,  0  0  0  0  p  1.3  -n  C,  -o  9+  C, 0 ‘1 1.)  9  1.3+  11  0 C,  0  0 0  0  C.)  0  -  Progesterone nmol/1O  -  cells  5 cells Progesterone nmol/10  00  I.  p  C) CD  CD CD  ‘—  00  II  ,—‘  -i  CD  CD  -  Cd,  +  CD p  p  CD  t’.)  p  a  CD  Cd,  CD  0  0  e3_  C)  p  o  CD  CD  Cd,  CD  CD  0  Cd,  C)  -i Cd,  CD  -  C)  0  CD  CD  0  0  C)  0  -o  C)  Co  0  C)  C.) C)  0  0 0  0  C)  0  0  C)  o  1+’  r.3  cells  I’)  —  IF  -‘  Progesterone nmol/1O  0  C  0  C  0 •0  0.  C)  C) C)  1•-  Progesterone nmol/10 5 cells  0•  •0  C,  L0 C  0  0  r  8  0  0  0  0  0  Progesterone nmol/10 5 cells  4-  5. 3 0,  C) O  0,  In  C) C)  C) I-)  -  In  U-)  23-  4-  C)  0  0  E  ‘ll  0  C  C  C2-  a)  C) C 0  C) C  22-  **  3-  C)  0) C) 0) 0  0)  0) CI 0  0 I  0  100  10  I  2000  C) 0  I  0.8  0.6  0)  IT  5  In  C)  0  0) ci,  C) C 0  **  -  0  0.0  C) C 0  -  2-  c_i-  I  I  20) C) C) 0 I  1- ••••••••T II  I  I  PGF2 co(nM)  Figure 35.  5-  C  C, C) 0  0) 0  10000  E  ci,  0.2  1000  0 -  E  -  200  04-  4  C  04  I 20  In  o  0  I I  C) 0  0)  -  .1  PGF2 a(nM)  PGF2C (nM)  -  ci,  I),  0  110100)00010000  0  10000  PGF2 cx (nM)  1.0  1-  2-  I  0  PGF2 a(nM)  A  I  20  100  1000 10000 0000.  PGF2 a(nM)  Effect of PGF (1O--lO M) on hCG (1 TU/ml)-stimulated progesterone production. Cells were incubated for 24 h. Values are mean+sem (n=8). Results from six experiments are presented.  119  **  P<O.O1.  7.4  Discussion Examination of the effect of PGF on steroidogenesis in human granulosa cells  provided inconsistent results. PGF agonists, having longer metabolic half-lives, were used to prolong PGF effects in culture. The same inconsistent results with cloprostenol and lutalyse suggested that metabolism of natural PGF was not the reason for the inconsistencies. Again, patient variability may partially explain the variability in PGF effects observed.  The results suggested that the effects of PGF on steroidogenesis  depend on unknown local factors but not on gonadotropin concenirations.  The  concentration of the unknown factor(s) might alter the reaction of human granulosa to PGF. This provides evidence that effects of gonadoiropins on steroidogenesis can be influenced by the cellular environment. The inconsistent results may also be caused by cell conditions related to the early luteal phase. An additive effect of PGF on hCG stimulated progesterone production during the early and mid-luteal phases was reported (Mason et al., 1986) in human luteal cells. PGF M 9 5 (10 1(Y ) increased free cytosolic 2 in individual human granulosa cells. This suggests that changes in intracellular Ca Ca 2 concentration can not be translated to changes in steroidogenesis by human granulosa cells.  120  8.0  HORMONAL REGULATION OF INHIBIN MESSENGER RNA IN HUMAN PLACENTAL CELLS  8.1 INTRODUCTION Inhibins and activins were first isolated from gonads of both sexes and various species (Ling et al., 1986; Vale et al., 1986; Mason et al., 1985). Their functions in gonads have been studied in many species (Tsafriri et al., 1989; Sum et aL, 1989; Ying et a1., 1986; Shukovski 1990). Inhibin a, BA and BB subunit mRNA from human ovary have been sequenced (Mason et aL, 1986).  The study of inhibin gene regulation in  human granulosa cells is difficult as there are not enough cells available for total RNA extraction. The regulation of inhibin gene expression in human term placental cells was investigated. The possibility that inhibin plays a role in pregnancy emerged from several studies.  Bioactive and immunoactive inhibin levels have been reported in human  placental extracts (McLachlan et al., 1986; Hochberg et al., 1981). These results were supported by the cloning and sequencing of inhibin cDNA from human term placenta (Mayo et al., 1986). Petraglia and Merchenthaler localized inhibin to the cytotrophoblast layer of the human placental villi by immunofluorescent staining method and showed that placental cultures can produce inhibin (Petraglia et al., 1987; Merchenthaler et al., 1987). Activin A may be synthesized by human term placenta as suggested by the presence of high levels of inhibin BA subunit mRNA (Meunier et al., 1988). In vivo studies showed that inhibin levels rose in first trimester with a further increase in the third trimester  121  (Yohkaichiya et al., 1989). Inhibins and activins have been reported to regulate hCG and GnRH synthesis and secretion in human placental cells (Petraglia et at, 1987). Stimulation of hCG release by activin but not inhibin was confirmed using a placental cell perfusion system (Steele et al., 1993). However, little is known about the regulation of inhibin and activin secretion in human placenta.  The inhibin (x-subunit gene in rat  contains a cAMP-responsive element (CRE) in the promoter region. Such a CRE has not been identified in rat BA- and BB-subunit genes (Chen et al., 1992). However, three CREs have been found in the human inhibin BB subunit gene (Mason et at, 1986). Inhibin BA mRNA expression appeared to be stimulated by 8-Br-cAMP in human fibrosarcoma cells (Tanimoto et al., 1992). Inhibin a-subunit increased during incubation of rat Sertoli cells in the presence of follicle-stimulating hormone (FSH) (Klaij et at, 1990). hCG, 8-Br-cAMP and adenylate cyclase activators (forskolin and cholera toxin) increased the release of inhibin from cultured human placental cells (Petraglia et at, 1987). Placental GnRH may stimulate hCG secretion (Sailer-Cheater et at, 1981, 1984). GnRH stimulation of hCG secretion suggests a possible role for GnRH in the regulation of inhibin synthesis during pregnancy. This study investigated regulation of inhibin a, BA and BB subunit mRNA expression by GnRH and cAMP. A sensitive assay, reverse transcription-polymerase chain reaction (RT-PCR) for detection of inhibin subunit mRNA was employed to measure inhibin mRNA expression.  122  8.2  MATERIALS AND METHODS  8.2.1  Total RNA extractions Trophoblast cells harvested after 6 h culture were centrifuged at 1000 x g for 5  mm and cell pellets were resuspended in 4 M guanidinium isothiocyanate and homogenized using a polyiron for 30 sec. Total RNA was prepared using the method of Chomezynski et at, 1987 and Glisin et at, 1974. RNA concentration was determined by UV-spectrophotometry.  8.2.2  Northern blot analysis  8.2.2.1  Preparation of plasmid vector DNA  8.2.2.1.1  Transformation  Ligated recombinant plasmid (10 j.il, 1-5 ng/il) was mixed with 200 p1 of competent cells. After 30 mm incubation on ice, the mixture was incubated at 42 C for 1 mm, then 0.5 ml LB medium was added and incubated at 37 C for 1 h to allow expression of the antibiotic resistant gene. The cell mixture (0.2 ml) was spread over an LB-agar plate containing ampicihin (50 mg/L) for overnight culture. The positive colony was selected by small scale DNA preparation.  8.2.2. 1.2  Small scale DNA preparation  123  The plasmid vectors (PGEM 4 containing inhibin Hc4, HBA-4 and GBB-5, kindly offered by Dr. Meunier H) was propagated in E. coil strain DH5cx in 5 ml LB. The E. coli cells transformed with plasmid vector were incubated at 37 C until the exponential phase. The cells were collected by centrifugation and resuspended in 150 p1 solution I (50 mM glucose, 25 mM Tris.C1 (pH8.0), and 10 mM EDTA (ethylenediaminetetraacetate, pH 8.0) and chilled on ice. Five mm later, 200 p1 of freshly prepared solution 11(0.2 N NaOH and 1 % SDS (sodium didecyl sulfate) was added and mixed by inversion. Another 300 p1 of ice-cold solution III (5 M potassium acetate 60 ml, glacial acetic acid 115 ml, 1120 28.5 ml) was added and mixed for 5 mm on ice. The mixture was extracted with equal volumes of phenollchloroform. The supernatant was aspirated and dispensed into a fresh tube. Two volumes of 99% ethanol was added into supernatant for 5 mm incubation at -20 C. The supematant was centrifuged at 10000 x g for 10 mm. The supernatant was discarded. The pellet was washed with 70% ethanol, then dissolved in 10 p1 TE buffer (tris-EDTA, 10 mM Tris.C1, 1 mM EDTA, pH 8.0). This plasmid mmni-prep’ method was used for a rapid analytical restriction digest analysis for specific eDNA insert before large scale plasmid production.  8.2.2.1.3  Large scale plasmid preparation  E. coli cells transformed with plasmid vector were incubated with 250 ml LB medium (Luria-Bertani medium, bacto-tryptone, 10 g; bacto-yeast extract, 5 g; Naci 10 g in 1 liter deionized 2 H 0 ) containing 12.5 mg ampicillin overnight at 37 C. Cells  124  were harvested at 5000 x g at 4 C for 15 mm and resuspended in 6 ml freshly prepared ice-cold lysis buffer (25 mM Tris-HCI, pH 8.0, 10 mM EDTA, 15 % sucrose and 2 mg/mi lysozyme) at 4 C for 20 mm. Twelve ml freshly prepared 0.2 N NaOH, 1 % SDS were added, mixed by inversion and kept on ice for 10 mm. 7.5 ml of 3 M sodium acetate (pH4.6) was added for 20 mm incubation on ice. The solution was centrifuged at 10000 x g for 15 mm to allow separation of the DNA layer. The supernatant was removed and incubated with 50 iii of RNAse A (stock 1 mg/mi) for 20 mm at 37 C. One volume of phenol:chloroform(l:1) saturated with TE buffer (pH7-8) was added and vortexed for 5 mm, then centhfuged at 10000 x g for 10 mm. This step was performed twice. The upper phase containing DNA was transferred into a fresh tube and 1 volume of chloroform:isoamyl alcohol (24:1) was added and vortexed. The phases were separated at 10000 x g for 10 mm. The upper phase was transferred to a fresh tube containing 2 volumes of 95% ethanol. DNA was precipitated at 10000 x gfor 20 mm after 30 mm incubation at -20 C. Sterile water (1.6 ml) was added to dissolve the pellet. 4 M NaC1 (0.4 ml) and 2 ml of 13% polyethylene glycol (PEG, MW 6,000) were added for separating plasmid and bacterial DNA. Following 1 h incubation at 4 C, plasmid DNA was precipitated at 10000 x g for 10 mm, washed with 70% ethanol and dissolved with distilled water to about 1mg DNA/mi.  8.2.2.2  Preparation of specific hybridization probe The recombinant cDNA insert was isolated by restriction enzyme digestion,  125  followed by electrophoresis in 1% agarose gel. cDNA (25 ng) insert was labelled with P]deoxycytid 3 [ 2 ine triphosphate by a random primer labelling technique according to the procedure describe in the BRL random primer labelling system (BRL, Burlington, ON). DNA (25 ng) dissolved in distilled water was boiled at 95 C for 5 mm, followed by rapid chilling on ice to denature the DNA. Two p1 of 0.5 mM dATP, dTTP and dGTP, 15 jil of random primer buffer, 50 pCi of 32 [aP}dCT P and 1 p1 of Kienow fragment (3U) were added and the mixture was incubated at 25 C for 1 h. The solution was heated at 100 C for 5 mm followed by rapid chilling at 4 C. The probe was labelled to a specific activity of iO dpm/pg DNA.  8.2.2.3  Northern analysis using specific probe  Total RNA (40 jig) was denatured at 70 C for 15 mm. The samples were electrophoresed through 1.0% (w/v) agarose, 2.2 M formaldehyde denaturing gel using 20 mM 3-(N-morpholino) propanesulfonic acid (MOPS) buffer containing 8 mM sodium acetate and 1 mM EDTA (pH 8.0). Ribosomal RNAs (28S and 18S) were visualized under UV light to ensure the integrity of RNA samples. RNA was transferred to Hybond-N membranes (Amersham, Oakville, ON) by capillary transfer using 3 x SSC (sodium chloride/sodium citrate buffer, 1 x, 8.8 g NaCl; 4.4 g sodium citrate in 1 liter 2 H 0 , pH 7). Following transfer, RNA was fixed to the membrane by UV light for 2 mm.  8.2.2.4  Hybridization/Washing  126  Prior to hybridization, the nylon membrane was soaked in a solution of prehybridization solution (50% formarnide, 5 x SSPE (sodium chloride/sodiu m phosphate and EDTA buffer, pH7.4), 5 x Denhardt’s reagent, 0.1% SDS, 100 jig/mi denatured salmon sperm in 0.1% diethyl pyrocarbonate-treated 2 H 0 ) at 42 C for at least 12 h. The filter was subsequently hybridized with a [ P]dCT 3 2 P labelled DNA probe overnight at 42 C. The filter was washed 3 times at 42 C for 20 mm with about 100 ml each of lx SSC containing 0.1% SDS and rinsed once with 0.1X SSC containing 0.1% SDS at 50 C for 20 mm. The blot was then air dried and autoradiographed by using intensifying screens at -70 C.  8.2.3  8.2.3.1  Quantitative RT-PCR  Quantitation of Inhibin a, JJA and JiB subunit mRNA Total RNA (5 jig) was electrophoresed on formaldehyde ágarose gels.  Ethidium bromide (1 jig) was added to each sample and the gel was photographed under UV-light following electrophoresis. Appearance of ribosomal RNA was used to assess the integrity and to confirm equal loading of RNA.  8.2.3.2  Synthesis offirst strand cDNA from mRNA Total RNA (5 jig) was subjected to first strand cDNA synthesis using a eDNA  synthesis kit (Pharmacia, Bale d’urfe, QUE). Avian Myeloblastosis Virus (AMV ) Reverse Transcriptase and 1.0 pg oligo d(T)  12-18  127  were added. The reaction buffer for  eDNA synthesis contained 10 mM Tris-HC1 (pH 8.3), 6 mM Mg, 40 mM KC1, 50 mM DTT, 1mM dNTPs and 25 U RNase inhibitor. The reaction mixtures (15 p1) were incubated at 32 C for I h,then 95 C for 10 mm and then chilled on ice. Synthe sized cDNA (1 p1 from total 15 pi) was used for PCR at standard conditions.  8.2.3.3  Primers of inhibin subunits Primers derived from human inhibin a, BA and BB cDNA sequences (Mason et  al., 1986) were used for amplification. The cx primers correspond to nucleotides 807828 (5’-CCCAGTTrCATCTTCCACTAC) and 1169-1188 (5’-TTCACH’CTGT GCC TATGGG) of a subunit cDNA. The BA primers correspond to nucleotides 1186-1 206 (5’ -AGGTCAACATCTGCTGTAAG) and 1544-1564 (5 ‘-TI’CTCTGGACAACTC ITG CT) of BA subunit cDNA. The BB primers correspond to nucleotides 748-768 (5’TGTfGCAGGCAACAGTFCTTC) and 1150-1170 (5’-GAATGACTGTACTT AGCCC AC) of fiB subunit cDNA. The identity of the amplified products was determined by sizing and Southern blot hybridization with cDNA probes (Fig. 36).  8.2.3.4  Quantitative PCR Amplification was performed in the presence of 0.04 pCi [ct P] dCTP. The 32  maximal incorporation of [a P] dCTP into specific PCR products was about 20% of 32 total counts. Amplification of fi-actin mRNA served as an internal control. Optim al conditions of PCR were chosen by determining the ranges of exponential increas e of product formation with the numbers of amplification cycles. Optimal cycle number  128  Figure 36.  PCR amplification of inhibin a-, BA-, and BB-subunit mRNA from cultured placental cells. (A) Ethidium bromide stained PCR produc ts separated in 1% agarose gel. Lanes: MW, molecular weight marker (1 kb DNA ladder, BRL, Toronto, ON); a, inhibin a-subunit PCR produc t; BA, inhibin BA-subunit PCR product; BB, inhibin BB-subunit PCR product. 5 ig of total RNA from human placental cells was used to synthesize cDNA. cDNA was amplified for 32 cycles. (B) Southe rn blot analysis. PCR products were blotted to Hybond nylon filter and was hybridized with 3 P]-laa [ beled 2 specific inhibin subunit cDNA probes.  129  was determined for each batch of cDNA synthesized. Products were electrophoresed on 6% polyacrylamide gels. Following staining with ethidium bromide, products were excised from the gel and the radioactivity was counted by Cherenkow counting.  8.2.3.5  Statistical Analysis The ratio of radioactivity (cpm) of inhibin subunit and B-actin was calculated  for each sample. Each treatment group contained three samples in cell culture, and the experiment was repeated four times. Data were pooled from four experiments. mRNA values of inhibin subunits in treatment groups are reported as mean ± sem compared to control group. The n values, referring to the number of cell preparation, was 12. Area under the inhibin subunit mRNA expression response was compared by One-Way Analysis of Variance. Student-Newman-Keuls Multiple Comparisons Test was used to differentiate among means following a significant F-test.  8.3  RESULTS  8.3.1  Expression of human inhibin subunit mRNA  The yield of total RNA from human term placental cell culture is low (about 1 8 jig/b  placental cells comparing with about 1 pg/b 6 granulosa cells). Northern  analysis of RNA was performed using the inhibin a, BA  ,  and 13B clones as probes.  Inhibin BA mRNA were detectable with 40 pg total RNA in term placenta, but inhibin a and BB mRNA were not detectable (Fig. 37). Fig. 37 shows the autoradiography of  130  a Northern blot using 40 jig of total RNA from term placental cells treated with or without 8-Br-cAMP (1.5 mM) and GnRH ( 10 M). The Inhibin BA cDNA probe 7 hybridized specifically to 7, 4.6, and 2.3 kb RNA species in human RNA. The level of inhibin BA mRNA in 8-Br-cAMP treated cells was higher than that in control cells (equal RNA loading was confirmed by hybridization of the blot with B-actin probe and autoradiography) (Fig. 37).  When the same blot was probed with inhibin  cDNA, no signal was detected (Fig. 37).  131  and BB  Figure 37.  Northern analysis of inhibin cc BA and BB mRNA from term placental cells treated with or without 8-Br-cAMP (1.5 mM) and GnRH  M). 7 ( 10  Figure 37A shows ethidium bromide staining of 40 pg total RNA in RNA denature gel.  C, control group, cAMP, 8-Br-cAMP (1.5 mM)-treated  group; GnRH, GnRH 7 (10 M )-treated group; rat uterus, 20 pg total RNA from rat uterus.  Human term placenta were treated for 6 h at 37 C.  Figure 37B shows the autoradiography of a Northern blot.  The RNA  denature gel was transferred to the Hybond nylon membrane which was hybridized with 6-actin, inhibin x, BA, and BB P]dCT 3 [ P-labe 2 lled cDNA probe overnight at 42 C. After hybridization, the membrane was washed 3 times at 42 C for 20 mm with about 100 ml each of lx SSC containing 0.1% SDS and rinsed once with 0.1X SSC containing 0.1% SDS at 50 C for 20 mm. The blot was then air dried and autoradiographed by using intensifying screens at -70 C.  132  w  r  O  w  p  O  8.3.2  Inhibin a, JJA and JJB mRNA expression in human placental cells The levels of a and BB mRNA are too low for Northern blot analysis using cDNA  probes.  Therefore, a more sensitive assay, quantitative RT-PCR for the detection of  inhibin subunit mRNAs has been employed for measuring the levels of inhibin mRNA expression. Oligonucleotide primers were selected to 5’ regions of a, BA and BB subunits mRNAs where their sequences are not homologous. Fig. 36 shows the product formation for each subunit amplification. The size corresponded to the predicted fragment and authenticity was confirmed by Southern blot with the respective cDNA probes. Fig. 38 shows the inhibin a, BA and BB PCR products from a pool of reverse transcribed cDNA samples at different cycles. The amount of PCR products of the three subunits increased logarithmically with cycle number until 40 cycles (Fig. 38).  The  results suggested that the levels of BA and a subunits mRNA expression were higher than that BB subunit.  134  Figure 38.  Influence of PCR cycle number on product formation. (A) Ethidium bromide stained PCR products separated on 1% agarose gel. cc PCR products of inhibin a-subunit; BA, PCR products of inhibin BA-subunit; BB, PCR products of inhibin 13B-subunit. The cycle number are shown on the x-axis.  (B) PCR products were quantitated by counting the  radioactivity of the excised bands.  The counting number (cpm) were  plotted against the number of cycles.  (A) MWa  MWiAMwfB  lO8lbp— 516bp— 394bp— 34 4bp-  (B)  a  f3A  S 135  J3B  (A)  cx  5 O6p 394bp 344bp 506bp 394bp 344bp  — ‘  5O6bp 394bp— 344bp  113 5  —0--(B)  4 io  10 15 20 25 30 35 40 cycles  a fA  —C--  0)  .2 E  %  iü  0  102 0  10  30  20  Cycle numberof PCR 136  40  50  Figure 39.  Quantitative analysis of inhibin u-, BA-, and BB-subunit mRNA levels in 8-bromo cyclic AMP (cAMP) and GnRH treated placental cells.  (A)  Ethidium bromide stained PCR products separated on 6% polyacrylamide gel (data from 1 experiment). Lanes: MW, molecular weight marker; 1-3, inhibin subunit PCR products from control group; 4-6, inhibin subunit PCR products from cAMP treated group; 7-9, inhibin subunit PCR products from GnRH treated group; 10, control reaction without cDNA template. (B) 3 samples were pooled from 4 experiments in each treatment group, to a total nz12. The values on Y axis represent the ratio of treatment to control group.  The data were analyzed by Student-Newman-Keuls  Multiple Comparisons Test. (P<0.05, n=12);  **  *  Significantly different from control group  Significantly different from control group (P<0.01,  n=12).  137  138  (A) MW 1  2  3  4  5  6  7  8  9 10  a  (381 bp  13A  • 378 bp  13B  “422bp  .4 524bp  f-Actin  (B) a 4.’  C  **  f3A f313  **  3.0 **  **  L.  Cu .0  2.0  U) 4-’  C .0  1.0  (0 C .0 .C C 0.0  Control  cAMP  GnRH  I  8.3.3  Stimulation of inhibin subunit mRNA expression by 8-Br-cAMP and GnRH Treatment with 8-Br-cAMP stimulated inhibin a-, BA- and BB-subunit mRNA  expression (Fig. 39). GnRH stimulated inhibin a- and BB-subunit but not BA-subunit mRNA expression (Fig. 39). Each treatment group contained 12 samples, pooled from  four experiments. As shown in Fig. 39, PCR products in B-actin groups did not change when 8-Br-cAMP or GnRH were added. The PCR products of inhibin a- and BB-subunit increased 2-3 fold by 8-Br-cAMP (150 1 iM) (P<0.01) and GnRH (100 nM) (p<O.Ol) (Fig. 39). The PCR product of inhibin BA-subunit slightly increased 1.3 fold by 8-Br-cAMP (150 jiM) (P<0.05, Fig. 39).  8.4  DISCUSSION Inhibins and activins are found in human placenta (Petraglia et at, 1987;  Merchenthaler et al., 1987), but little is known about their regulation of synthesis. Quantitative RT-PCR is a sensitive technique for evaluation of ñiRNA comparing with Northern blot analysis.  The high level of inhibin BA mRNA in human placenta is  consistent with a previous report (Meunier et al., 1988a). The presence of inhibin BA precursors of 7.0, 4.6, and 2.3 kb in Northern blot analysis is also consistent with previous reports (LaPolt et al., 1989). The study of regulation of inhibin subunit mRNA expression showed that cAMP stimulated inhibin BA precursor rnRNA expression. In order to study inhibin a and BB subunit mRNA expression, we used a more sensitive method, the quantitative RT-PCR. The quantification of mRNA levels by RT-PCR is affected by the efficiencies of both cDNA synthesis and PCR amplification. About 10-  139  30% variation was observed from eDNA synthesis and PCR.  Each treatment was  triplicated and the experiment was repeated four times for statistic analysis. Quantitative analysis (Fig. 38) showed that the amount of inhibin subunit mRNA expression was BA >  a> BB. This result is in agreement with Northern blot data (Meunier H, et aL, 1988a). 8-Br-cAMP has been reported to stimulate inhibin production from human  placental cells (Petraglia et al., 1987). The results of this study of inhibin a-, BA- and BB-subunit at the mRNA level supports this observation. It suggests that cAMP may be a second messenger, inducing inhibin a- and BB-subunit expression from human placental cells. It has been shown that the inhibin a-subunit gene contains a cAMP- and phorbol ester-inducible DNA fragment (Klaij et al., 1990; Feng et al., 1989). Inhibin a-subunit mRNA levels increased when dibutyryl cAMP was administrated in rat Sertoli cells (Klaij  et al., 1990), even though its promoter region does not appear to have a CRE (Feng et al., 1989).  Inhibin BB-subunit was also stimulated by 8-Br-cAMP in this study, which  supports the finding that human inhibin BB subunit gene has thiee CRE (Mason et aL, 1986). The expression of inhibin BB-subunit mRNA has been reported to increase by treatment of rat Sertoli cells with cycloheximide, an activator of adenyl cyclase (Klaij et al., 1990). The increase of inhibin BB mRNA expression by cAMP may increase inhibin B or activin AB synthesis in human placenta. The higher mRNA level of inhibin BAsubunit relative to that of inhibin a- and fiB-subunit suggests that the expression of inhibin a- and fiB-subunit mRNA, may be rate-limiting steps during inhibin synthesis in human placental cell. The findings that cAMP slightly increased and GnRH did not have any effect on inhibin BA-subunit mRNA levels suggests that the mechanisms of regulation  140  of inhibin (aBA and aBB) and activin (BABA and BABB) synthesis may be different. Previous reports have suggested a stimulating action of hCG on placental adenylate cyclase and cAMP formation (Davis et al., 1989; Bicsak et at, 1986) and a stimulating action of placental GnRH on hCG secretion (Sailer-Cheater et at, 1981; Currie et al., 1981; Seeburg et al., 1984). The possibility that the effect of GnRH involves stimulation of mRNA expression of inhibin subunits was investigated.  Increases of mRNA  expression of inhibin a- and BB-subunits by GnRH suggest a possible stimulating action of GnRH on placental inhibin production. The same effects of GnRH and cAMP on inhibin a- and BB-subunit mRNA expression suggest that the effects of GnRH may depend on an increase of intracellular cAMP level. GnRH-stimulated hCG production by human term placental cells also can be mimicked by cAMP (Currie et al., 1993). Therefore, it is possible that the intracellular pathway of GnRH in human placental cells is adenyl cyclase-cAMP-dependent. Petraglia et al (1987) reported that anti-inhibin serum increased GnRH-like immunoreactivity (GnRH-LI) in human placental cell cultures (Petraglia et at, 1987). The presence of a feedback loop between inhibin and GnRH may have physiological significance. Maternal plasma inhibin levels are low during first trimester, increased in second trimester, peak in third trimester (Healy et al., 1990). The highest concentrations of GnRH-LI were seen in first half pregnancy, then were sustained in the second half (Siler-Khodr et al., 1984). It is possible that a low concentration of placental inhibin may contribute to an increased GnRH-LI production during early pregnancy,  and high  concentration of inhibin suppress the continuous increase of GnRH-LI production from  141  placenta during late pregnancy. The high mRNA levels of inhibin BA subunit relative to that of inhibin  X  subunit  suggest the formation of activin A in human placenta (Meunier et at, 1988a). Activin A may play a role in the control and regulations of growth and developmental functions according to its function in other tissues (DePaolo et at, 1991). Activin-A might be regulated by human placental lactogen (hPL) or other hormones in placental cells but not GnRH.  Further experiments need to be done to investigate regulation of activin-A  expression. In summary, quantitative RT-PCR confirmed the existence of mRNA expression of inhibin and activin subunits in human placenta at term. cAMP and GnRH stimulated inhibin  x- and BB-subunit niRNA expression.  GnRH-stimulated inhibin mRNA  expression may be cAMP-dependent. Activin-A may not be regulated by GnRH and cAMP.  142  9.0  HORMONAL REGULATION OF INHIBIN MESSENGER RNA IN HUMAN GRANULOSA CELLS  9.1 INTRODUCTION Inhibin and activin have paracrine roles modulating ovarian thecal and granulosa cell gonadotropin-mediated steroidogenesis (Tsafriri et aL, 1989; Sum et a1., 1989; Ying et al., 1986; Shukovski et al., 1990; Sugino et al., 1988b).  McLachlan et al.(1987)  developed a RIA for inhibin and reported the pattern of inhibin secretion during the normal menstrual cycle and in gonadotropin-induced cycles. Additionally, inhibin was  detected in follicular fluid and placenta (Robertson et al., 1985; McLachlan et al., 1986). Detection of human inhibin a- and BA-gene expression by standard Northern blotting techniques has been reported for placenta and corpus luteum (Petraglia et al., 1991; Meunier et al., 1988c). In  Situ  hybridization localized inhibin BA mRNA to the granulosa  cells of follicles; the a mRNA is expressed at low levels in other ovarian cell types (Woodruff et al., 1988; Meunier et al., 1988b). The regulation of inhibin secretion by steroid and polypeptide hormones has been examined in rat granulosa cells. The levels of secreted inhibin and inhibin mRNA are regulated by FSH in rat granulosa culture, consistent with in vivo data that FSH increases inhibin gene expression (Woodruff et aL, 1988; Meunier et al., 1988). FSH increases inhibin a and BA mRNA transcription rates in rat primary granulosa cells (Woodruff et al., 1988). LH stimulates inhibin production only after induction of LH receptor with FSH (Bicsak et al., 1986; Zhiwen et al., 1988).  143  cAMP also stimulates inhibin production by granulosa cells, suggesting that effects of FSH and LH on inhibin production are cAMP-mediated. This study examined regulation of mRNA expression of 3 inhibin subunits by cultured human granulosa cells. This was accomplished using quantitative reverse transcription and polymerase chain reaction.  9.2  MATERIALS AND METHODS  9.2.1  Human granulosa cell treatment hCG (1 lU/mi) or cAMP (1.5 mM) were added 96 h after initiation of culture.  Each treatment group had 4 wells.  9.2.2  Total RNA extraction Media were collected, following 24 h incubation with 0.5% BSA containing  hormone treatment, for progesterone assay.  The cells in each well were collected  separately by adding 0.5 ml lysis buffer to each well. Total RNA was prepared using the RNaid kit method (Bio/Can Scientific). RNA concentrations were determined by UV spectrophotometry.  9.2.3  Quantification of Inhibin a, JJA andJ)B subunit mRNA Total RNA (0.5 jig) was electrophoresed on formaldehyde agarose gels. Ethidium  bromide (1 jig) was added to each sample and the gel was photographed under UV-light following electrophoresis. The appearance of ribosomal RNA was used to assess the  144  integrity of the RNA and to confirm equal loading. Total RNA (0.5 zig) was subjected to first strand cDNA synthesis using a cDNA synthesis kit (Phannacia, Baie D’Urfe, QUE).  Synthesized cDNA (1 p1 from total 15 iii) was used for PCR at standard  conditions. Primers derived from human inhibin cx, BA and BB eDNA sequences (Mason et al., 1986) were as described in section 7.2.4. The identity of the ampiified products was determined by sizing and Southern blot hybridization with cDNA probes (Fig. 40). Amplification by PCR was performed as described in section 7.2.4.  9.2.4  Statistical Analysis The data were analyzed as described in section 7.2.4.5.  9.3  RESULTS  9.3.1  Inhibin cx, JJA and JiB mRNA expression in human granulosa cells The yield of total RNA from human granulosa cell cultures is  6 cells. 1 pg/b  A more sensitive assay, RT-PCR, has been employed for measuring inhibin subunit mRNA expression using smaller cell numbers. Fig. 40 shows the product formation for each subunit with PCR amplification. The size coffesponded to the predicted fragment and authenticity was confirmed by Southern blot with the respective cDNA probes. Fig. 41 shows the inhibin a, BA and BB PCR products from a pool of reverse transcribed eDNA samples at different cycles. The amount of PCR products of the 3 subunits increased logarithmically with cycle number 145  cc  M  1AM  1BM  422bp— 3Blbp-* 378bp-  422bp— 381bp378bp-  Figure 40.  PCR assay of inhibin cx-, BA-, and BB-subunit mRNA in cultured human granuloa cell RNA and Southern blot analysis of PCR products.  Top:  Ethidium bromide staining of PCR products separated in 1% agaros e gel. Lanes: MW, molecular weight markers; cc, inhibin cc-subunit PCR produc t; BA, inhibin BA-subunit PCR product; BB, inhibin BB-subunit produc t. 500 ng of total RNA from 5 day cultured human granulosa-luteal cells was used to synthesize cDNA. cDNA was amplified for 30 cycles. Bottom: Southern blot analysis. PCR products were blotted to Hybond nylon filter and hybridized with P]-la [c 3 2 beled x specific inhibin subunit cDNA probes.  146  Figure 41.  Different amplification cycles of PCR products of inhibin subunits using human granulosa-luteal cell total RNA. Top: Ethidium bromide staining of PCR products separated in 1% agarose gel. o, PCR products of inhibin a-subunit; BA, PCR products of inhibin BA-subunit; fiB, PCR products of inhibin BB-subunit. Cycle numbers are shown. Bottom: The inhibin a-, BA-, and fiB-subunit PCR products of different cycles were quantitated by counting the radioactivity of the excised gel bands. Counts (cpm) were plotted against the number of amplification cycles.  147  M  40  35  30  25  20  15  10  5  (cycles)  381bp-*  378bp4nhi bin 6B—subunit  422bp—  —0— 10000  a  C-)  a  —7—  13A  •  f3B  1000  100 0  20  10  PCR  30  40  cycle  148  J  until 40 cycles (Fig. 41). The results suggested that a and BB subunit mRNA expression were greater than for BA mRNA.  9.3.2  Stimulation of inhibin subunit mRNA expression by 8-Br-cAMP and hCG Treatment with 8-Br-cAMP stimulated inhibin a- and BB-subunit mRNA  expression (Fig. 42).  hCG stimulated inhibin BA-subunit but not a and BB-subunit  mRNA expression (Fig. 42).  Each treatment group had 16 samples, pooled from 4  experiments. PCR products in B-actin groups did not change when 8-Br-cAMP or GnRH were added (Fig. 42). The PCR products of inhibin a- and BB-subunit were increased 1.8-3 fold by 150 jiM 8-Br-cAMP (P<0.01, Fig. 42). The PCR product of inhibin BAsubunit was increased 1.8 fold by 1 IU hCG (P<0.05, Fig. 42).  149  Figure 42.  Quantitative analysis of inhibin a-, BA-, and BB-subunit mRNA levels in 8-Br-cAMP (0.15 mM) and hCG (1 llJIrnI) treated gTanulosa-luteal cells. Top: Ethidium bromide staining of PCR products separated in 1% agarose gel. Lanes: MW, molecular weight markers; hCG, hCG (1 IIJ/ml) treated group; cAMP, 8-Br-cAMP (1.5 mM) treated group;  ,  control without  eDNA template. Bottom: Quantitative analysis of inhibin a-, BA-, and BB subunit mRNA levels in 8-Br-cAMP and hCG treated granulosa cells. Sixteen samples were pooled from 4 experiments in each treatment group and quantitated by counting the radioactivity of the excised gel bands. The values on the Y axis represent the ratio., of treatment to control group. Data were analyzed by Student-Newman-Keuls Multiple Comparisons Test. Different superscripts (above the sem bars) denote statistical differences (P<0.01).  150  control  M  hCG  cAMP  0  inhibin a-subunit  381bp->  inhibin hA-subunit  37 8bp—i  inhibin SB-subunit  422bp-*  3—actin  524bp-*  i  2  2.0 C  C  C  C.)  0  0  4 a  a  ZL  02  a C  1 a 1.0  C .0  C .0  fll -C  -C C  -c  C  C  0.0  control  hCG  cAMP  control hCG  151  cA M P  9.4  DISCUSSION Inhibin a—, iSA- and fiB-subunit mRNA expressed in human granulosa  luteal cells provides evidence for the local activity of inhibins and activins. Inhibin a— subunit mRNA is highly expressed in human granulosa-luteal cells in vitro, the BA- and fiB-subunit mRNA are also expressed in the human granulosa-luteal cells but their expression levels are much lower. This result is consistent with reported findings in rat and human (Woodruff et al., 1987; Turner et al., 1989; LaPolt et al., 1989; Lapolt et al., 1990; Eramaa et al., 1993). Studies in monkey and human showed that a—, BA-subunit mRNA, were detectable, but fiB-subunit mRNA was wealdy- or non-detectable in the granulosa-luteal cells (Erammaa et al., 1993; Schwall et al., 1990).  The inhibin fiB  subunit immunoreactivity reported in human granulosa-luteal cells supports the finding in this study (Yamoto et al., 1991). Thus, the lower level of fiB mRNA that could not be detected by Northern blot analysis but was detected by RT-PCR analysis is efficiently translated to protein. Although the significant biological differences of inhibin A and B, and activin A and AB has not been well understood, the fmding of fiB-subunit mRNA expression in human granulosa-luteal cells indicates that except inhibin A and activin A, inhibin B and activin AB also secrete from these cells and may play a role during ovulation and corpus luteum formation. Stimulatory effects of FSH on inhibin secretion and subunit mRNA levels were mimicked by cAMP and cAMP-generating reagents, including forskolin and cholera toxin  (Bicsak et al., 1986; Zhang et al., 1987; Lapolt et al., 1990). Inhibin a- and fiB-subunit mRNA expression in human granulosa cells were increased by 8-bromo cAMP but not  152  hCG. These results suggest that stimulation of a- and BB-subunit gene expression is through the cAMP-dependent protein kinase A pathway.  The finding that cAMP  increased inhibin a mRNA was consistent with reports that several potential cAMPresponsive elements (CRE) have been identified in the 5’-flanking regions of the rat and human inhibin a- and BB- subunit genes (Feng et al., 1989; Mason et al., 1986). Inhibin a-subunit mRNA expression also increased when dibutyryl cAMP was administrated in rat Sertoli cells (Klaij et al., 1990). The increase of inhibin BB mRNA expression by cAMP may increase inhibin B or activin AB synthesis in human granulosa-luteal cells. It is interesting that the effect of hCG on inhibin subunit gene expression differs with 8bromo cAMP in this study. Although hCG increases the intracellular cAMP level and both hCG and FSH stimulate progesterone production in human granulosa-luteal cells (Li et al., 1992; Li et al., 1993), the effect of hCG on inhibin production may differ from FSH and cAMP. The mechanism causing different effects of hCG and cAMP on inhibin subunit gene expression is unknown.. hCG may stimulate inhibin BA mRNA expression by intracellular pathways other than adenylate cyclase-cAMP, such as protein kinase C pathway, which may attenuate the intracellular cAMP effect on inhibin a- and BB-subunit gene expression and trigger inhibin BA-subunit gene expression. This result is supported by the previous studies that LH at low doses stimulated inhibin protein and mRNA from rat granulosa cells after incubation with FSH (Bicsak et al., 1986; Turner et al., 1989; Zhang et al., 1988), whereas high doses of LHJhCG inhibited PMSG/FSH-induced inhibin production in vitro (Zhang et al., 1988; Michel et al., 1991) and in vivo (Lee, 1983). hCG increased BA-subunit mRNA level when those of the a- and BB-subunit mRNA  153  unchanged.  An increase in the I3AIct mRNA ratio by hCG may reflect the increased  inhibin A and activin-A synthesis. The physiological effects of inhibin A and activin A have been shown in rat and human follicles (Woodruff et al., 1990). Activin A promoted cell proliferation, inhibited progesterone production induced by gonadotropins, and  stimulate basal oestradiol production in cultured human granulosa-luteal cells (Li et al., 1992; Khorasheh et al., unpublished data; Rabinovici et aL, 1990). In order to elucidate the roles of different inhibin and activin dimers as regulators of human ovarian function, The study for the regulation of inhibin each subunit gene expression in human granulosa cells is the key step. This study shows that the granulosa-luteal cells of IVF patients can be used successfully in mRNA regulation studies of inhibin/activin subunits with highly sensitive quantitative RT-PCR analysis. The results suggest that inhibin x-, BA- and BB subunit mRNA are expressed in cultured human granulosa-luteal cells and hCG and cAMP stimulate three subunit mRNA expression in the different mechanism.  154  10.  GENERAL SUMMARY  Physiological Roles of Activin and Follistatin  The hypothesis that local substances mediate human granulosa-lutal cell responses to gonadotropin has been examined.  Activin inhibited gonadotropin-stimulated  progesterone production in these studies, inhibin had no effect on basal or hCG- stimulated progesterone production, while follistatin increased basal progesterone production. Activin, inhibin and follistatin have been isolated from mammalian follicular fluid (Ling et al., 1986; Ling et al., 1987, Robertson et al., 1987).  Suppression of hCGIFSH  stimulated progesterone production by activin in cultured human granulosa-luteal cells suggests a role for activin in prevention of luteinization of preovulatory or large antral follicles in vivo. Higher levels (60 lU/mi) of inhibin were produced by rat granulosa cells when stimulated with FSH (Zhiwen et al., 1987). Thus, generation of endogenous inhibin by the cells may have affected the results from the present experiments where exogenous gonadotropins were added. The possible mechanisms underlying inhibitory effects of gonadotropin-stimulated progesterone production by activin A may involve in the interference with the coupling of gonadotropin receptor and adenylate cyclase-cAMP cascade. IBMX significantly increased intracellular cAMP accumulation, but did not change basal and hCG-stimulated progesterone production. This result suggested that blocking intracellular cAMP degradation to AMP had no effect on progesterone production.  Although activin inhibited hCG-stimulated intracellular cAMP level,  inhibition of gonadoiropin- stimulated progesterone production by activin through  155  additional intracellular pathway has not been excluded. It is generally agreed that activin stimulates FSH secretion which, in turn, increases estrogen production in the ovary. Thus, it is plausible that the overall action of activin may be to promote follicular differentiation and delay the onset of luteinization. The stimulatory effect of follistatin on progesterone production in cultured human granulosa-luteal cells suggests a regulatory role of follistatin in the human ovary and its potential for an autocrine role. Follistatin is a binding protein of activin (Robertson et al., 1990; Nakamura et al., 1990). Thus, it can be postulated that a balance between follistatin and activin-A activity contributes to the regulation of the granulosa-luteal cell response to gonadotropins.  Increased  FSH is necessary for development of aniral  follicles to the preovulatory stage (Meunier et al., 1988c), a phase in follicular maturation when synthesis of follistatin is stimulated (Klein et al., 1991).  Increased basal  progesterone secretion from human preovulatory granulosa cells mediated by follistatin suggests that follistatin could in part contribute to the rise in progesterone before the preovulatory LH surge.  The results showed that increased basal progesterone is  accompanied by an increase in cAMP accumulation. The action of follistatin may also implicate an indirect action such as through the binding of activin (Robertson et al., 1990; Nakamura et al., 1990) and disinhibition of activin suppression of progesterone synthesis (Saito et al., 1991). This potential mechanism of action remains to be investigated. The in vitro data indicate that follistatin, rather than inhibin, may function as an intra-ovarian antagonist of activin. Follistatin, as a binding protein of activin, may bind with activin to neutralize each others’ action(s) on granulosa-luteal cells. The concentrations and/or  156  the ratio of follistatin and activin may play a key role in regulation of steroidogenesis by granulosa-luteal cells during different stages of cell differentiation. Inhibin a-, BA-, and BB-subunit mRNA expression in human granulosa cells may be evidence for local inhibin and activin effects.  This study showed that activin and  follistatin act in an autocrine fashion. Inhibin a-subunit mRNA expression is greater than inhibin BA and BB mRNA expression. This is consistent with reported findings (Mayo et al., 1986; Mason et al., 1986; Davis et aT., 1986). The results suggest that BA- and BB subunit expression is the rate-limiting step for inhibin and activin synthesis. Furthermore, hCG may increase inhibin A and activin A synthesis by stimulating inhibin BA-subunit  mRNA expression. Inhibin a- and BB-subunit mRNA levels were increased by 8-BrcAMP but not by hCG. This suggests that effects of hCG and cAMP on inhibin subunit gene expression are not mediated through a common second messenger system. Hence, hCG may stimulate inhibin BA mRNA expression by intracellular pathways other than adenylate cyclase-cAMP pathway, while not increasing intracellular cAMP concentration enough to increase inhibin a and BB mRNA expression. With this hCG-altered inhibin BA subunit expression, it is possible that hCG may increase inhibin A or activin A production from these cells. Activin A may act as a dominant follicle selector by maintaining high estrogen secretion (Khorasheh et al., unpublished data) and by inhibiting gonadotropin-stimulated progesterone production in given follicles. Furthermore, activin A and hCG/LH may feedback on activin-A and progesterone production from human granulosa-luteal cells. Since follistatin is an activin binding protein, the ratio of follistatin to activin may be critical to steroid production by  157  individual follicles. The hCG-stimulated intracellular cAMP concentration was decreased by activin, while the basal intracellular cAMP concentration was increased by follistatin. Thus follistatin may increase inhibin a and BB subunit mRNA production by the increase of intracellular cAMP. This further suggested a balanced action between the opposing roles of activin—A and follistatin in other functions of granulosa-luteal cells regulated by intracellular cAMP (Fig. 43).  FSHILH  Activin  Fol  atm  Inhibin A  L Cholesterone  Figure 43.  Progesterone  A model, suggested by the results, for the interaction of inhibin-related peptides and gonadotropins in the regulation of intracellular cAMP production, progestin synthesis and inhibin subunit mRNA transcription by human granulosa-luteal cells. AC, adenylate cyclase. Solid lines express  158  the established conclusion. Dash lines express the possible actions.  Physiological Roles of A III  A II and A III bind to, and are locally produced by human granulosa-luteal cells (Husain et al., 1987). LH, but not FSH, increases the concentration of renin-angiotensin in human follicular fluid (Carruthers et aL, 1990). This suggests that angiotensins have local actions  iii  the human ovary. A Ill may be induced by the LH surge during the early  luteal phase, thus, the renin-angiotensin system may be involved in neovascularization and steroidogenesis during corpus luteum formation. In these studies A III inhibited hCG stimulated progesterone production dose-dependently in cultured human granulosa-luteal cells. Thus, a negative feedback ioop between the renin-angiotensin system and LH may regulate progesterone production by the human corpus luteum in the early phase. A III may also act as a luteolytic factor during the late luteal phase. This hypothesis would be supported by evidence of higher renin-angiotensin bioactivity in late luteal phase than that in early luteal phase. The intracellular mechanism of A III effect on steroidogenesis is unclear and the study did not support the involvement of the DAG-protein kinase C system (Fig. 44).  159  ANG III  FSHJLH  I  ANG II  I DAG  / / /  cAMP  ATP  ÷  Figure 44.  A model, suggested by the results, for the interaction of angiotensins, hCG and PKC in.the regulation of progesterone production by human granulosa luteal cells.  AC, adenylate cyclase; PJP2, phosphatidylinositol 4,5-  bisphosphate. DAG, Diacyl glycerol. Solid lines express the established conclusion. Dash lines express the possible actions.  2 Physiological Roles of GnRH and PGF GnRH and PGF are present in human follicular fluid. GnRH had no effect on  160  basal or hCG-stimulated progesterone production in this study. PGF had inconsistent effects on basal and hCG-stimulated progesterone production. GnRH and PGF had luteolytic effect on non-humai mammalian granulosa-luteal cells, suggesting differences between species. The reason for the inconsistent effects of PGF in this study may have been related to unknown factors secreted by granulosa cells or differences in cell pretreatment or differentiation. The effect of PGF might be altered by these unknown factors. The different results reported in other studies in human and other mammals suggest that the effects of PGF on progesterone production, by granulosa-luteal cells, depends on the species studied and the precise experimental conditions.  Inhibin subunit mRNA expression in human term placenta Both inhibins and activins are found in human placenta (Petraglia et al., 1987; Merchenthaler et al., 1987), a finding confirmed in the present study, by the presence of specific mRNA for each subunit. synthesis.  However, little is known about the regulation of their  Using quantitative PCR, the relative amounts of inhibin subunit mRNA  expression was shown to be BA> c> BB. This finding agreed with the results obtained from Northern blot analysis (Meunier et al., 1988a). The high level of inhibin BA mRNA and the presence of inhibin BA precursors of 7.0, 4.6, and 2.3 kb in Northern blot analysis were consistent with previous reports (LaPolt et al., 1989; Meunier et al., 1988a), and suggested the preferential formation of activin A in the human placenta (Meunier et al., 1988a).  Activin A may play a role in control of fetal growth and development, as  suggested by the multiple functions of activin in diverse tissues (Depaolo et al., 1991).  161  The finding that 8-bromo cAMP increased inhibin a-, BA- and BB-subunit mRNA expression in cultured human placental cells was supported by a previous report (Petraglia et al., 1987).  Furthermor, it has been shown that inhibin a- and BB-subunit gene  contains the cAMP- and phorbol ester-inducible DNA fragment (Mason et al., 1986; Feng et aL, 1989). The addition of 8-Br-cAMP led to higher levels of inhibin a- and BB-, than inhibin BA-subunit mRNA, suggesting that the expression of inhibin a- and BB-subunit mRNA may be a hormone-dependent and rate-limiting step during inhibin synthesis in human placental cell. Previous reports have suggested a stimulatory action of hCG on placental adenylate cyclase and cAMP formation (Moore et al., 1992; Davis et al., 1986), and a stimulatory action of placental GnRH on hCG secretion (Sailer-Cheater et al., 1981; Currie et al., 1992; Currie et al., 1981; Seeburg et al., 1984). The present finding of GnRH-increased inhibin a- and BB-subunit mRNA expression, suggested a stimulatory action of GnRH on placental inhibin B production.  The same effects of GnRH and  cAMP on both inhibin a- and BB-subunit mRNA expression raises the possibility that the effect of GnRH is secondary to increased hCG synthesis and intracellular cAMP level. Alternately, it is possible that the action of GnRH in human placental cells is independent of the hCG-induced adenyl cyclase-cAMP pathway. Treatment with anti-inhibin serum increased GnRH-like immunoactivity in human placental cell cultures (Petraglia et al., 1987), suggesting the presence of a regulatory feedback loop between inhibin and GnRH. Maternal plasma inhibin levels are low during the first trimester, increased in second trimester and peak in the third trimester (Healy et  162  al., 1990). The highest serum levels of immunoactive GnRH are seen in the first half pregnancy and become sustained in the second half (Siler-Khodr et al., 1984).  It is  possible that an increased concentration of placental inhibin may contribute to an increased GnRH production during pregnancy, and high concentration of inhibin may suppress the continuous increase of GnRH production from placenta during late pregnancy (Petraglia et al., 1987). The present finding that GnRH stimulates inhibin subunit mRNA expression at term further supports the notion of a complex interplay between GnRH and inhibin in the human placenta.  163  11.  FUTURE EXPERIMENTS  11.1 ACTIVIN-A AND FOLLISTATIN  11.1.1 Interaction of aclivin-A and follistatin Further experiments need to be done for the investigation of interaction of follistatin and activin-A. As they are binding proteins for each other, it can be postulated that a balance between follistatin and activin-A activity contributes to the regulation of the granulosa-luteal cell response to gonadotropins.  The experiment with follistatin  incubated with activin-A could show the neutralized action between follistatin and activin A. The binding ratio of follistatin to activin should be about 4 to 1 when activin reach the maximum concentration (Takanori et aL, 1990). In order to allow time for binding, a preincubation of follistatin with activin could also be attempted. These experiments should indicate if follistatin can antagonize the action of activin-A in progesterone production.  11.1.2 Regulation of inhibin subunit mRNA expression by FSH and follistatin Inhibin a-, BA-, and BB-subunit mRNA is expressed in human granulosa cells and regulated by hCG and cAMP.  Results in this thesis have shown that hCG increased  inhibin BA-subunit mRNA expression and 8-Br-cAMP increased inhibin a and BB mRNA levels. These results suggest that the effects of hCG and cAMP on inhibin subunit gene expression are not related. FSH and hCG had different effects on inhibin subunit mRNA  164  expression and inhibin production (Bicsak et aL, 1986; Turner et aL, 1989; Zhang et al., 1988; Michel et al., 1991). Therefore, it is worthwhile to study FSH effect on inhibin subunit mRNA expression in these cells.  Follistatin may increase inhibin a and BB  subunit mRNA production by increasing intracellular cAMP. The mechanism of such stimulatory effects of hCG on inhibin BA mRNA expression in human granulosa cells is unknown. Therefore it would be worthwhile to study the role of the PKC pathway on inhibin subunit gene expression in these cells in order to elucidate the mechanism of hCG stimulatory effect on inhibin BA mRNA expression.  11.2 EFFECT OF A II ON BASAL AND hCG-STIMULATED ESTRADIOL  PRODUCTION  AND  MECHANISMS  OF  INHIBITORY EFFECT OF A III  11.2.1 Effects of A II on basal and hCG-stimulated estradiol production Results within this thesis have shown that A II binds with human granulosa-luteal cells, but its effects on basal and hCG-stimulated progesterone production is not clear. Further experiments should focus on basal and hCG-stimulated estradiol production in these cells. By examining both progesterone and estrodiol production. Concurrently, it is expected that the autocrine/paracrine role of angiotensins in the ovary can be better defined.  165  11.2.2 Mechanisms of inhibitory effect of A III on hCG-stimulated progesterone production  The results within this thesis have shown that the inhibitory effect of A ifi is not through the DAG—PKC second messenger pathway.  The possibility that A III inhibits  2 concentration hCG-stimulated intracellular cAMP, or increases intracellular free Ca should be examined.  11.3 hCG-REGULATED INHIBIN SUBUNIT mRNA EXPRESSION IN HUMAN PLACENTAL CELLS The results within this thesis have shown that both cAMP and GnRH stimulated inhibin ct and BB mRNA expression. Previous reports have suggested a stimulatory action of hCG on placental adenylate cyclase and cAMP formation (Moore et al., 1992; Davis et al., 1986) and a stimulatory action of placental GnRH on hCG secretion (Sailer Cheater et al., 1981; Curri et al., 1992; Currie et al., 1981; Seeburg et at, 1984). It will be worthwhile to study the effect of hCG on inhibin subunit mRNA expression in these cells.  To provide evidence for a stimulatory effect of GnRH on inhibin cr and BB mRNA  expression, intracellular cAMP levels after incubation with GnRH should be measured and compared with controls.  166  12.  REFERENCES  Adashi E Y, Hsueh A J W 1982 Estrogens augment the stimulation of aromatase activity by follicle-stimulating hormone in cultured rat granulosa cells. J Biol Chem 257: 6077  Adashi E Y, Resnick C E, D’Ercole A J, Svoboda M E, Van Wyk 11 1985 Insulin-like growth factors as intraovarine regulators of granulosa cell growth and function. Endocr Rev 6: 400  Adashi E Y, Resnick C E, Svoboda M E, Van Wyk J I 1985a A novel role for somatomedin-C in the cytodifferentiation of the ovarian granulosa cell. Endocrinololgy 115: 1227  Adashi E Y, Resnick C E, Svoboda M E, Van Wyk J J 1985b Somatomedin-C synergizes with follicle-stimulating hormone in the aquisition of progestin biosynthetic capacity by cultured rat granulosa cells. Endocrinololgy 116: 2135  Adashi E Y, Resnick C E, Brodie A M H, Svoboda M E, Van Wyk 3 J 1985c Sometomedin-C-mediated potentiation of follicle-stimulating hormone-induced aromatase activity of cultured rat granulosa cells. Endocrinology 117: 2313  Adashi E Y, Resnick C E, Svoboda M E, Van Wyk I J 1985d Sometomedin-C enhances  167  induction of luteinizing hormone receptor by follicle-stimulating hormone in cultured rat granulosa cells. Endocrinology 116: 2369  Adashi E Y 1991 The ovarian life cycle. In Yen S S C & Jaffe R B, Reproductive Endocrinology, Wonsiewicz M J ed. pp183-236  Amsterdam A, Knecht M, Catt K 1 1981 Hormonal regulation of cytodifferentiation and intercellular communication in cultured granulosa cells. Proc Nail Acad Sci USA 78: 300  Anderson R G, Brown M 5, Goldstein J L 1977 Role of the coated endocytotic vesicle in the uptake of receptor-bound low density lipoprotein in human fibroblasts. Cell 10: 351  Aten R F, Williams A T, Behrman H R  1986 Presence of gonadotropin-releasing  hormone-like protein(s):demonstrations and characterization. Endocrinology 118: 961  Aten R F, Treland J J, Weems C W, Behrman H R 1987 Presence of gonadotropin releasing hormone-like protein in bovine and ovine ovaries. Endocrinology 120: 1727  Audinot V. Rasolonjanahary R, Bertrand P, Priam M, Kordon C, Enjalbert A 1991 Involment of protein kinase-C in the effect of angiotensin-il on adenosine 3’, 5’monophosphate production in lactotroph cells. Endocrinology 129: 2231  168  Auletta F 3, Caldwell B V, Sperroff L 1976 Estrogen-induced luteolysis in the rhesus monkey reversal with indomethacin. Prostaglandins 11: 745  Auletta F J, Flint A P F 1988 Mechanisms controlling corpus luteum function in sheep, cows nonhuman primates and women especially in relation to the time of luteolysis. Endocr Rev 9: 88  Auletta F J, Kamps D L, Wesley M, Gibson M 1983 Primate corpus luteum progesterone content and in vitro responsiveness to gonadotropin throughout the luteal phase of the menstrual cycle and during PGF-induced luteolysis. Program of the 30th Annual Meeting of the Society for Gynacology Investigation, Washington, DC Abstract 299  Baird A, Morméde P, Ying S Y, Wehrenberg W B, Ueno N, Ling N, Guillemin R 1985 A nonmitogenic pituitary function of fibroblast growth factor: regulation of thyrotropin and prolactin secretion. Proc Nati Acad Sci USA 82: 5545  Baranao J L S, Hammond J M 1984 Comparative effects of insulin and insulin-like growth factors on DNA synthesis and differentiation of porcine granulosa cell. Biochem Biophys Res Commun 124: 484  Baranczuk R 3, Fainstat T 1976 Progesterone-induced ovulation of the hamster ovary in vitro. J Endocrinol 70: 317  169  Beers W H, Strickland S, Reich E 1975 Ovarian plasminogen activator: Relationship to ovulation and hormonal regulation. Cell 6: 387  Behrman H R, Preston S L and Hall A K 1980 Cellular mechanisms of the antigonadotropic action of luteinizing hormone-releasing hormone in the corpus luteum. Endocrinology 107: 656  Behrman H R 1979 Prostaglandins in hypothalamopituitary and cellular mechanisms of the antigonadotropic action of ovarian function. Annu Rev Physiol 41: 685  Behrman H R, Luborsky-Moore J L, Pang C Y, Wright K, Dorflinger L J 1979 Mechanisms of PGF action in functional luteolysis Adv Exp Med Biol 112: 557  Behrman H R, Armstrong D T 1969 Cholesterol esterase stimulation by luteinizing hormone in luteinized rat ovaries Endocrinology 85: 474  Bendell J J, Dorrington J 1988 Rat theca-interstitial cells secrete a transforming growth factor-B like factor that promotes growth and differentiation in rat granulosa cells Endocrinology 123: 941  Bicsak T A, Tucker E N, Cappel S, Vaughan J, Rivier J, Vale W, and Hseuh A W 1986  170  Hormonal regulation of granulosa cell inhibin synthesis. Endocrinology 119: 2711  Biersing L 1977 Ovarian Histochemistry. In Zuckerman, L.Weir, B. J. (eds) The Ovary. Vol 1, Academic press, New York, p303  Bramley T A, Menzies G 5, Baird D T 1985 Specific binding of gonadotropin-releasing hormone and an agonist to human corpus luteum homogenates: Characterization, properties and luteal phase levels J Clin Endocr Metab 61: 834  Broxmeyer H E, Li L, Cooper S, Schwall R H, Mason A 3, Nilcolics K 1988 Selective and indirect modulation of human multipotenfial and erythroid hematopoietic progenitor cell proliferation by recombinant human activin and inhibin. Proc Natl Acad Sci USA 85: 9052  Brown M 5, Dana S E, Goldstein I L 1975 Receptor-dependent hydrolysis of cholesterol esters contained in plasma low density lipoprotein. Proc Nati Acad Sci USA 72: 2925  Brown M 5, Goldstein I L 1976 Receptor-mediated control of cholesterol metabolism. Science 191: 150  Bumpus F M, Pucell A G, Daud A I, Husian A 1988 Angiotensin II: An intraovarian regulatory peptide. Am J Med Sci 295: 406  171  Burger H G, Zhiwen Z, Davis S, Lee V, Carson R S. Herington A, Findlay J K, Farnworth P, Wang  Q F, Au S, de Krester D M, Robertson D M 1988 Inhibin-regulation  and mechanism of action. In ‘First Congrad International Workshop on “Non-steroidal gonadal factors: physiological roles and posibilities in contraceptive development” (Eds Hodgen GD, Rosenwaks Z, Spieler JM) pp 137 (Jones Inst Press: Virginia)  Butzow R 1985 Luteinizing hormone releasing factor increases release of human chorionic gonadotropin in isolated cell columns of normal and malignant trophoblasts.  mt J Cancer  29: 9  Byskov A G, Rasmussen G 1973 Ultrastructure studies of the developing follicle. In: The Development and Maturation of the Ovary and Its Function, International Congress Steries No. 267 edted by H. Peters. ppSS  Byskov A G 1978 The anatomy and ulturastructure of the rete system in the fetal mouse ovary. Biol Reprod 19: 720  Caffrey J L, Fletcher P W, Diekman M A, O’Callaghan P L, Niswender G D 1979 The activity of ovine luteal cholesterol esterase during several experimental conditions. Biol Reprod 21: 601  Caffrey J L, Nett T M, Abel Jr J H, Niswender G D 1979 Activity of 38-hydroxy- 5-  172  steroid 4 5 dehydrogenase/B isomerase B in the ovine corpus luteum. Biol Reprod 20: 279  Carrithers M D, Raman V K, Masuda S. Weyhenmeyer J A 1990 Effect of angiotensin II and III on inositol polyphosphate production in differentiated NG1O8-15 hybrid cells. Biochem Biophys Res Commun 167: 1200  Carroll R S, Corrigan A Z, Gharib S D, Vale W, Chin W W 1989 Inhibin, activin, and follistatin: regulation of follicle-stimulating hormone messenger ribonucleic acid levels. Mol Endocrinol 3:1969  Casper R F, Erickson G F, Rebar R W, Yen S S C 1982 The effect of luteinizing hormone-releasing factor and its agonist on cultured human granulosa cells Fert Steril 37: 406  Catt K J, Hunyady L, Balla T 1991 Mini review: second messengers derived from inositol lipids. 3 Bioenerg Biomembr 23: 7  Catt K J, Balla T 1989 Phosphoinositide metabolism and hormone action. Ann Rev Med 40: 487  Chait A, Ross R, Albers 3 J, Bierman E 11980 Platelet-derived growth factor stimulates activity of low density lipoprotein receptors. Proc Natl Acad Sci USA 77: 4084  173  Channing C P 1972 Stimulatory effects of prostaglandins upon luteinization of rhesus monkey granulosa cell cultures Prostaglandins 2: 331  Channing C P 1969 Steroidogenesis and morphology of human ovarian cell types in tissue culture. J Endocrinol 45: 297  Channing C P 1974 Temporal effects of LH, hCG, FSH and dibutyryl cyclic AMP upon luteinization of rhesus monkey granulosa cells in culture. Endocrinology 94: 1215  Cheatum S G, Warren J C 1966 Purification and properties of 313-hydroxysteroid dehydrogenase and  -3-ketosteroids isomerase from bovine corpus luteum. Biochem 5  Biophys Acta 122: 1  Chen C -L, Feng Z -M 1992 Regulation of inhibin/activin genes. Ninth International Congress of Endocrinology S-133  Chomczynski P, Sacchi N 1987. Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156  Clark M R, Marsh 3 M, LeMaire W 1 1976 The role of protein synthesis in the stimulation by LH or prostaglandin accumulation in preovulatory follicles in vitro. Prostaglandins 12: 209  174  Clark M R, Marsh 3 M, LeMaire W 3 1978 Mechanism of luteinizing hormone regulation of prostaglandin synthesis in rat granulosa cells 3 Biol Chem 253: 7757  Clark M R, Thibier C, Marsh 3 M, LeMaire W J 1980 Stimulation of prostaglandin accumulation by luteinizing hormone-releasing hormone (LHRH) and LHRH analogs in rat granulosa cells in vitro. Endocrinology 107: 17  2 accumulation by Clark M R 1982 Stimulation of progesterone and prostaglandin E luteinizing hormone-releasing hormone (LHRH) and LHRH analogs in rat granulosa cells. Endocrinology 110: 146  Clayton R N, Harwood J P, Catt K J 1980 Ovarian gonadotropin-releasing hormone (GnRH) receptors: characterization, distribution and induction by GnRH. Endocrinology 108: 1148  Clayton R N, Catt K 1 1981 Gonadotropin-releasing hormone receptors: Characterization, physiological regulation, and relationship to reproductive function Endocrin Rev 2: 186  Clayton R N, Huhtaniemi I T 1982 Absence of gonadoiropin-releasing hormone receptors in human gonadal tissue. Nature 299: 56  Conti M, Kasson B G, Hsueh A 3 W 1984 Hormonal regulation of 3’, 5’-adenosine  175  monophosphate phosphodiesterases in cultured rat granulosa cells. Endocrinology 114: 2361  Cook V I, Wright J W, Wright S A, Harding J W 1990 Comparison of angiotensin metabolism by brain membranes from SHR and WKY rats. Brain Res 529: 320  Corbin A, Bex F J 1981 Luteinizing hormone releasing hormone agonists induce ovulation in hypophysectomized proestrous rats: direct ovarian effect. Life Sci 29: 185  Culler M D, Tarlatzis B C, Lightman A, Naftolin F 1986 Angiotensin TI-like immunoreactivity in human ovarian follicular fluid. J Clin Endocrinol Metab 62: 613  Cunie A J, Fraser H M, Sharpe R M 1981 Human placental receptors for luteinizing hormone releasing hormone. Biochem Biophys Res Commun 99: 332  Currie W D, Li W, Baimbridge K G, Yuen B H, Leung P C K 1992 Cytosolic free calcium increased by prostaglandin F in human granulosa cells. Endocrinology 130: 1837  Currie W D, Steele G L, Ho-yuen B, Leung P C K 1993 GnRH- and cAMP-stimulated human chorionic gonadotropin secretion from perifused placental cells. Molecular Basis of Reproductive Endocrinology, Serono Symposia, USA 1993  176  Daud A I, Bumpus F M, Husain A 1990 Characterization of angiotensin I-converting enzyme (ACE)-containing follicles in the rat ovary during the esirous cycle and effects of ACE inhibitor on ovulation. Endocrinology 126: 2927  Daud A I, Bumpus F M, Husain A 1988 Evidence for selective expression of angiotensin II receptors on atretic follicles in the rat ovary: AN autoradiographic study. Endocrinology 122: 2727  Davis J S, Farese R V, Clark M R 1983 Gonadotropin-releasing hormone stimulated phosphatidylinositol metabolism in rat granulosa cells: mechanism of action of GnRH. Proc Natl Acad Sci USA 80: 2049  Davis J 5, Tedesco T A, West L A, Maroulis G B, Weakland L L 1989 Effects of human chorionic gonadotropin, prostaglandin F and protein kinase C activators on the cyclic AMP and inositol phosphate second messenger system in cultured human granulosa-luteal cells. Mol Cell Endocr 65: 187  Davis S R, Dench F, Nikolaidis I, Clements J A, Forage R G, Krozowski Z, Burger H G 1986 Inhibin A-subunit gene expression in the ovaries of immature female rats is stimulated by pregnant mare serum gonadotropin. Biochem Biophys Res Commun 138: 1191  177  Davis S R, Krozowski Z, McLanchlan R I, Burger H G 1987 Inhibin gene expression in the human corpus luteum. I Endocrinol 1987, 115: R21  Demoulin A, Dubois M, Gerday C, Gillain D, Lambotte R, Franchimont P 1991 Variations of luteinizing hormone serum concentrations after exogenous human chorionic gonadotropin administration during ovarian hyperstimulation. Fert Steril 55: 797  Dennefors B L, Sjogren A, Hamberger L 1982 Progesterone and 3’,5’-monophosphate formation by isolated human corpora lutea of different ages: Influence of human chorionic gonadotropin and prostaglandins. I Clin Endocrinol Metab 55: 102  DePaolo L V. Bicsak T A, Erickson G F, Shimasaki 5, Ling N 1991 Follistatin and activin: a potential intrinsic regulatory system within diverse tissues. Proc Soc Exp Biol Med 198: 500  Deschepper C F 1991 The renin-angiotensin system in the pituitary gland. Trends Endocrinol Metab 2:104  Dimino M I and Campbell M D 1976 Progesterone synthesis by luteal mitochondria in vitro. Pro Soc Exp Biol Med 152:54  Do Y 5, Sherrod A, Lobo R A, Paulson R I, Shinagawa T, Chen 5, Kjos 5, Hsueh A I  178  W 1988 Human ovarian theca cells are a source of renin. Proc Nati Acad Sd USA 85:1957  Dorrington J, Mceracher H, Munshi S, Gore-langton R 1982 LHRH independently stimulates steridogenic enzymes in granulosa cell cultures. Endocrinology 110 (Suppi): 178  Dorrington I, Chuma A V, Bendell J 3 1988 Transforming growth factor B and follicle stimulatin g hormone promote rat granulosa cell proliferation. Endocrinology 123: 353  Eckstein B, Raanan M, Lerner N, Cohen S, Nimrod A 1977 The appearance of 2Oxhydroxysteroid dehydrogenase activity in preovulatory follicles of immature rats treated with pregnant mare serum gonadotropin. 3 Steroid Biochem 8: 213  Edwards R G 1974 Follicular fluid. 3 Reprod Fertil 37: 189  Einspanier A, Jarry H, Pitzel L, Holttz W, Wuttke W 1991 Determination of secretion rates of estradiol, progesterone, oxytocin, and angiotensin II from tertiary follicles and freshly formed corpora lutea in freely moving sows. Endocrinology 129: 3403  Erammaa M, Heikinheimo K, Tuuri T, Hilden K, Ritvos 0 1993 Inhibin/activin subunit mRNA expression in human granulosa-luteal cells. Mol Cell Endocrinol 92:R15  179  Erickson G F, Case E 1983 Epidermal growth factor antagonizes ovarian theca-interstitial cytodifferentiation. Mol cell Endocrinol 31: 71  Esch F, Shimasaki S, Mercado M, Cooksey K, Ling N, Ying S Y, Ueno N, Guillemin R 1987 Structural characterization of follistatin: a noval follicle-stimulating hormone releaseinhibiting peptide from the gonad. Mol Endocrinol 1:849  Eskay R L, Mical R S, Porter J C 1977 Relationship between luteinizing hormone releasing hormone concentration in hypophyseal portal blood and luteinizing hormone release in intact, castrated, and electrocheniically stimulated rats. Endocrinology 100: 263  Fanjul L F, Ruiz de Galarreta C M, Hsueh A 3 W 1983 Progestin augmentation of gonadotropin- stimulated progesterone production by cultured rat granulosa cells. Endocrinology 112: 405  Farookhi R 1980 Effects of androgen on induction of gonadotropin receptors and gonadotropin- stimulated adenosine 3 ‘,5 ‘-mono-phosphate production in rat ovarian granulosa cells. Endocrinology 106: 1216  Feng Z M, Li Y P, Chen C 1989 Analysis of the 5’-flanking regions of rat inhibin x- and B-B-subunit genes suggests two different regulatory mechanisms. Mol Endocrinol 3: 1914  180  Fernandez L A, Tarlatzis B C, Rzasa P J, Caride V 3, Laufer N, Negro-Vilar A F, Dechemey A H, Naftolin F 1985 Renin-like activity in ovarian follicular fluid. Fertil Steril 44: 219  Findiley J K, Xiao S, Shakovski L 1990 Role of inhibin-related peptides as intragonadal regulators. Reprod Fertil Dev 2: 205  Flanders K C, Thompson N L, Cissel D S. Van-Orberghen-Schiling E, Baker C C, Kass M E, Ellingworth L R, Roberts A B, Sporn M B 1989 Transferming growth factor 81: Histochemical localization with antibodies to different epitopes. J Cell Biol 108: 653  Flower RJ, Vane JR 1974 Some pharmacologic and biochemical aspects of prostaglandin biosynthesis and its inhibition. In: Robinson HJ & Vane JR (eds) Prostaglanclin Synthesis Inhibitors. Raven Press, New York. pp 9-18  Fowler R E, Fox N L, Edwards R G, Walters D E, Steptoe P C 1978 Steroidogenesis by cultured granulosa cells aspirated from human follicles using pregnenolone and androgens as precursors. J Endocrinol 77: 171  Forage R G, Brown R W, Oliver K 3, Atrache B T, Devine P L, Hudson G C, Goss N H, Bertram K C, Tolstoshev P, Robertson D M, de Kretser D M, Doughton B, Burger H G, Findley 3 K 1987 Immunization against an inhibin subunit produced by recombinant  181  DNA techniques results in increased ovulation rate in sheep. J Endocrinol 1 14:R1-R4  Furuichi T, Yoshikawa S, Miyawake A, Wada K, Maeda N, Mikoshiba K 1989 Primary structure and functional expression of the inositol 1,4,5-triphosphate-binding protein P400. Nature 342: 32  Gebauer H,  Lindner H R,  Amsterdam A  1978  Synthesis of heparin like  glycosaminoglycans in rat ovarien slices Biol Reprod 18: 350  Glisin V, Crkvenjakov R, Byus C 1974 Ribonucleic acid isolated by cesium chloride centrifugation. Biochemistry 13: 2633  Glorioso N, Atlas SA, Laragh JH, Iewelewicz R, Sealey JE Prorenin in high concentrations in human ovarian fofficular fluid. Science 1986, 233:1422  Goff A K, Armstrong D T 1977 Stimulatory action of gonadotropins and prostaglandins on adenosine 3’, 5’—monophosphate production by isolated rat granulosa cells. Endocrinology 101: 1461  Goff A K, Leung P C K, Armstrong D T 1979 Stimulatory effect of follicle-stimulating hormone and androgens on the responsiveness of rat granulosa cells to gonadotropins in vitro. Endocrinology 104: 1124  182  Goldstein J L, Anderson R G, Brown M S 1979 Coated pits, coated vessels, and receptor mediated endocytosis. Nature 279: 679  Goodfriend T L 1991 Angiotensins: A family that grows from within. Hypertension 17: 139  ) 1 Goodfriend T L, Peach M J 1975 Angiotensin III: (DA-aspartic acid  -  angiotensin II.  Circulation Res 36 (Suppi I): 1  Goodman A, Hodgen G 1979 Between-ovary interaction in the regulation of follicle growth: Corpus Luteum function and gonadotropin secretion in the primate ovarian cycle II Effects of lutectomy and hemiovarectomy during the luteal phase in cynomolgus monkeys. Endocrinology 104: 1310  Gospodarowicz D, III C R, Birdwell C R 1977 Effects of fibroblast and epidermal growth factors on ovarian cell proliferation in vitro: I Characterization of the response of granulosa cells to FGF and EGF. Endocrinology 100: 1108  Gospodarowicz D, Bialecki H 1978 The effect of epidermal and fibroblast growth factor on the replicative lifespan of bovine granulosa cells in culture. Endocrinology 103: 854  Gospodarowicz D, Cheng J, Lui G-M, Baird A, Esch F, Bohien P 1985 Corpus luteum  183  angiogenic factor is related to fibroblast growth factor. Endocrinology 117: 2283  Gospodarowicz D, Gospodarowicz F 1975 The morphological transformation and inhibition of growth of bovine luteal cells in tissue culture induced by luteinizing hormone and dibutyryl cyclic AMP. Endocrinology 96: 458  Gospodarowicz D, Bialecki H 1979 Fibroblast and epidermal growth factors are mitogenic agents for cultured granulosa cells of rodent, porcine and human origin. Endocrinology 104: 757  Gospodarowicz D 1989 Fibroblast growth factor:Involvement in early embryonic development and ovarian function. Semi Reprod Endocrinol 7: 21  Gwynne I T and Strauss Ill J F 1982 The role of lipoproteins in steroidogenesis and cholesterol metabolism in steroidogenic glands. Endocr Rev 3:299  Hall A K, Preston S L, Behrman H R 1981 Purine amplification of luteinizing hormone action in ovarian luteal cells. J Biol Chem 256: 10390  Hamberger L, Dennerfors B, Khan I, Sjogren A 1979 Cyclic AMP formation of isolated human corpora lutea in response to hCG-interference by PGF. Prostaglandins 17: 615  184  Handerson K M, McNatty K P 1977 A possible inter-relationship between gonadotropin stimulation and prostaglandin F inhibition of steroidogenesis by granulosa-luteal cells in vitro. J Endocrinol 73: 71  Hansson A, Serhan C N, Haegstrom I, Ingelman-Sundberg M, Samuelsson B 1986 Activation of protein kinase C bu lipoxin A and other eicosanoids Intracellular action of oxygenation products of arachidonic acid. Biochem Biophys Res Commun 134: 1215  Harman S M, Louvet J-P, Ross G T 1975 Interaction of estrogen and gonadotropins on  follicular atresia. Endocrinology 96: 1145  Harwood I P, Clayton R N, Can K 1 1980 Ovarian gonadotropin-releasing hormone receptors I. Properties and inhibition of luteal cell function. Endocrinology 107: 407  Harrison L M, Kenny N, Niswender G D 1987 Progesterone production, LH receptors, and oxytocin secretion by ovine luteal cells types on days 6, 10 and 15 of the oestrous cycle and day 25 of pregnancy I Reprod Fertil 79: 539  Harrison R 1 1962 The structure of the ovary, In Zuckerman, S., A. M. Mantle, and P. Eckstein, (eds.), The Ovary, Academic Press, London, pp143  Harrison RI 1962 The structure of the ovary. In Zuckerman S A M, Mandle, Eckstein P  185  (eds.) The ovary. Academic press, London ppl.  Hasegawa Y, Miyamoto K, Abe Y, Nakamura T, Sugino H, Eto Y, Shibai H, Igarashi M 1988 Induction of follicle stimulating hormone receptor by erythroid differentiation factor on rat granulosa cell. Biochem Biophys Res Commun 156: 668  Hazum E, Nimrod A 1982 Photoaffinity-labeling and fluorescence-distribution studies of gonadotropin-releasing hormone receptors in ovarian granulosa cells. Proc Nail Acad Sci USA 79: 1747  Healy D L, Poison D, Yohkachiya T, de Kretser D 1990 Inhibin and related peptides in pregnancy. Baillière’s Clin Endocrinol Metab 4: 233  Heritage A S, Stumpf W E, Sar M, Grant L D 1980 Brain stem catecholamine neurons are target sites for sex steroid hormones. Science 207: 1377  Hiildj 1979 Stimulation of cartilage zones of the calf costochondral growth plate in vitro by growth hormone-dependent rat plasma somatomedin activity. I Endocrinol 83: 219  Hillensjo T, LeMaire W 1 1980 Gonadotropin-releasing hormone agonists stimulate meiotic maturation of follicle-enclosed rat oocytes in vitro. Nature 287: 145  186  Hillensjo T, LeMaire W J, Clark M R, Ahren K 1982 Effect of gonadotropin-releasing hormone (GnRH) and GnRH agonists upon accumulation of progesterone, cAMP and prostaglandin in isolated preovulatory rat follicles. Acta Endocrinol 101: 603  Hiller S G, Reichert L E Jr Van Hall E V 1981 Modulation of FSH-controled steroidogenesis in rat granulosa cells: Direct in vitro effects of LHRH and ICI-1 18630. Mol Cell Endocrinol 23: 193  Hillier S G, Yong E L, Illingworth P 1, Baird D T, Schwall R H, Mason A J 1991a Effect of recombinant inhibin on androgen synthesis in cultured human thecal cells. Mol Cell Endocr 75: Ri  Hillier S G, Yong E L, Illingworth P J, Baird D T, Schwall R H, Mason A J 199 lb Effect of recombinant activin on androgen synthesis in cultured human thecal cells. J Clin Endocr Metab 72: 1206  Hochberg, Z, Weiss, J and Richman, R A Inhibin-like activity in extracts of rabbit placenta. Placenta 1981, 2: 259  Hsu C J, Holmes S D, Hammond J M 1987 Ovarian growth factor-like activity. Concentrations in porcine follicular fluid during follicular enlargement. Biochem Biophys Res Commun 147:242  187  Hsueh A J W, Jones P B C 1981a Extrapituitary actions of gonadotropin-releasing hormone. Endocr Rev 2: 437 192 Hsueh A I W, Welsh T H, Jones P B C 198 lb Inhibition of ovarian and testicular steroidogenesis by epidermal grwth factor. Endocrinology 108: 2002  Hsueh A J W, Erickson G F 1979a Extrapituitary action of gonadotropin-releasing hormone: Direct inhibition of ovarian steroidogenesis. Science 204: 854  Hsueh A I W, Ling N C 1979b Effect of an analog of gonadotropin-releasing hormone upon ovarian granulosa cell function. Life Sci 25: 1223  Hsueh A J W, Wang C, Erickson G F 1980 Direct inhibitory effect of gonadotropin releasing hormone upon follicle-stimulating hormone induction of luteinizing hormone receptor and aromatase activity in rat granulosa cells. Endocrinology 106: 1697  Hsueh A I W, Adashi E Y, Tucker E, Valk C, Ling N C 1983 Relative potencies induction of luteinizing hormone agonists and antagonists on ovarian and pituitary functions. Endocrinology 112: 689  Hsueh A J W, Dahi K D, Vaughan I, Tucker E, Rivier I, Bardin C W, Vale W 1987 Heterodimers and homodimers of inhibin subunits have different paracrine action in the modulation of luteinizing hormone-stimulated androgen biosynthesis. Proc Nail Acad Sci  188  USA 84: 5082  Hsueh A J W, Lieu X Y, Cajander S. Peng X R, Dahl K, Kristensen P. Ny T 1988 Gonadotropin-releasing hormone induces ovulation in hypophysectomized rats: Studies on ovarian tissue-type plasminogen activator activity, messenger RNA content and cellular localization. Endocrinology 122: 1486  Husain A, Bumpus F M, De Silva P. Speth R C 1987 Localization of angiotensin II receptors in ovarian follicles and the identification of angiotensin II in rat ovaries. Proc Nail Acad Sci USA 84: 2489  Hutchinson L A, Findley J K, de Vos F L, Robertson D M 1987 Effects of bovine inhibin, transforming growth factor-B and bovine activin-A on granulosa cell differentiation. Biochem Biophys Res Commun 146: 1405  Ingram D L 1962 Atresia. In Zuckerman S A M, Mandlle, Eckstein P (eds.) The Ovary. Accademic Press, London pp247  Inouye 5, Guo Y, DePaolo L, Shimonaka M, Ling N, Shimasaki 5 1991 Recombinant expression of human follistatin with and 288 amino acids: chemical and biological comparison with native porcine follistatin. Endocrinology 129: 815  189  Itskovitz J, Sealey J E, Glorioso N, Laragh J H, Rosenwaks Z 1988 The ovarian prorenin angiotensin system. Ann N Y Acad Sci 1988, 541:179  Johansson E D B, Gemzell C 1971 Plasma levels of progesterone during the luteal phase in normal women with synthetic estrogens. Acta Endocrinol 68: 551  Johnson L K, Baxter 3 D, Vlodavsky I, and Gospodarowicz D 1980 Epidermal growth factor and expression of specific genes: effects on cultured rat pituitary cells are dissociable from the mitogenetic response. Proc Nati Acad Sci USA 77: 394  Jones P B C, Hseuh A 3 W 1982 Pregnenolone biosynthesis by cultured rat granulosa cells: Modulation by follicle-stimulating hormone and gonadotropin-releasing hormone. Endocrinology 109: 1347  Jones P B C, Hsueh A J W 1981 Direct effects of gonadotropin-releasing hormone and its antagonist upon ovarian functions stimulated by FSH, prolactin and LH. Biol Reprod 24: 747  Jones P B C, Hsueh A J W, Welsh T H 1982 Regulaiton of ovarian progesterone production of by EGF in cultured rat granulosa cells. J Biol chem 257: 11268  Jones P B C, Hsueh A J W 1982a Regulation of 3B-hydroxysteroid dehydrogenase by  190  gonadotropin-releasing hormone and follicle-stimulating hormone in cultured rat granulosa cells. Endocrinology 110: 1663  Jones P B C, Hsueh A J W 1981 Direct stimulation of ovarian progesterone metabolizing enzyme by gonadotropin-releasing hormone in cultured granulosa cells. J Biol Chem 256: 1248  Jones P B C, Hsueh A J W 1982 Pregnenolone biosynthesis by cultured rat granulosa cells: Modulation by follicle-stimulating hormone and gonadotropin-releasing hormone. Endocrinology 111: 713  Jones P B C, Conn P M, Marian J, Hsueh A J W 1980 Binding of gonadoiropin-releasing hormone agonist to rat granulosa cells. Life Science 27: 2125  Joseph S K, Thomas A P, William R 3, Irvine R F, Williamson J R 1984 Myo-inositol 2 1,4,5-triphosphate: a second messenger for the hormonal mobilization of intracellular Ca in liver. J Biol Chem 259: 3077  Jutisz M, Tertrin-clary C 1974 luteinizing hormone and chorionic gonadotropin: Structure and activity. Cuff Topics Exp Endocrinol 2: 195  Jacobs B R, Smith R G 1980 Evidence for receptor-like protein for progesterone in  191  bovine ovarian cytosol. Endocrinology 106: 1276  Jacobs B R, Suchocki 5, Smith RG 1980 Evidence for human ovarian progesterone receptor. Am J Obstet Gynecol 138: 332  Jerry H, Meyer B, Halzapfel G, Hinney B, Kuhn W, Wuttke W 1988 Angiotensin Il/HI and substance P in human follicular fluid obtained during IVF: relation of the peptide content with follicular size. Acta Endocrinol (Copenh) 119: 277  Karsch F I, Sutton G P 1976 An intraovarian site for the luteolytic action of estrogen in the rhesus monkey. Endocrinology 98: 553  Karsch F J, Krey L C, Weick R F, Knobil E 1973 Functional luteolysis in rhesus monkey: Role of estrogen. Endocrinology 92: 1148  Katz Y, Aniistrong D T 1976 Inhibition of ovarian estradiol secretion by luteinizing hormone in prepubertal, pregnant mare serum-treated rats. Endocrinology 99: 1442  Katz Y, Leung P C K, Armstrong D T 1979 Testosterone restores ovarian aromatase activity in rats treated with a 17,20-lyase inhibitor. Mol Cell Endocrinol 14: 37  Kemeter P, Salzer H, Breitnecker A, Friedrich F 1975 Progesterone, estradiol-17B, and  192  testosterone levels in the follicular fluid of tertiary follicles and Graafian follicles of human ovaries. Acta Endocrinol 80: 686  Kenny N, Robinson J 1986 Prostaglandin F-induced luteolysis: Interactions of LH, prostaglandin F and forskolin in cyclic AMP and progesterone synthesis in isolated rat luteal cells J Endocrinol 111: 415  Kim, I-C, Schomberg, D W 1989 The production of transforming growth factor B activity by rat granulosa cell cultures Endocrinology 124: 1345  Kim S J, Namkoong S E, Lee J W, Jung J K, Kang B C, Park J S 1987 Response of human chorionic gonadotropin to luteinizing hormone-releasing hormone stimulation in the culture media of normal human placenta, choriocarcinoma cell lines and in the serum of patients with gestational trophoblastic disease. Placenta 8: 257  Klaij I A, Toebosch A M W, Themmen A P N, Shimasaki S, de Jong F H, Grootegoed J A 1990 Regulation of inhibin a- and BB-subunit mRNA levels in rat Sertoli cells. Mol Cell Endocrinal 68: 45  Klein R, Robertson D M, Shukovski L, Findlay J K, de Kretser  D M 1991 The  radioimmunoassay of follicle-stimulatory hormone (FSH) suppressing protein (FSP): stimulation of bovine granulosa cell FSP secretion by FSH. Endocrinology 128: 1048  193  Knecht M, Katz M 5, Catt K J 1981a Gonadoiropin-releasing hormone inhibits cyclic nucleotide accumulation in cultured rat granulosa cells. J Biol Chem 265: 34  Knecht M, Amsterdam A, Catt K J 198 lb The regulatory role of cyclic AMP in hormoneinduced granulosa cell differentiation. J Biol Chem 256: 10628  Knecht M, Catt KJ 1981c Gonadotropin-releasing hormone: regulation of adenosine-3’:5’cyclic monophosphate in ovarian granulosa cells. Science 214: 1346  Knecht M, Nanta T, Katz M 5, Catt K J 1983a Regulation of adenyl cyclase activity by follicular-stimulating hormone and a gonadotropin-releasing hormone agonist in cultured rat granulosa cells. Endocrinology 112: 1247  Knecht M, Feng P. Catt K 1987 Bifunctional role of transforming growth factor-B during granulosa cell development. Endocrinology 120: 1243  Knecht M, Feng P. Catt K 1989 Transforming growth factor-B: Autocrine, paracrine, and endocrine effects in ovarian cells. Semin Reprod Endocrinol 7: 12  Knecht M, catt K J 1983b Modulation of cAMP-mediated differentiation in ovarian granulosa cells by epidermal growth factor and platelet derived growth factor. J Biol Chem 258: 2789  194  Kogawa K, Nakamura T, Sugino K, Titani K, Sugino H 1991 Activin binding protein is present in pituitary. Endocrinology 128:1434  Kojima I, Kojima K, Rasmussen H 1985 Role of calcium fluxes in the sustained phase of angiotensin TI-mediated aldosterone secretion from adrenal glomerulosa cells. J Biol Chem 260: 9177  Kolena I, Channing C P 1972 Stimulatory effects of LH, FSH and prostaglandin upon 3’,5’-AMP levels in porcine granulosa cells. Endocrinology 90: 1543  Kudlow J E, Kobrin M S, Purchio A F, Twarclizk D R, Hernandez E R, Asa S L, Adashi E Y 1987 Ovarian transforming growth factor-o gene expression: Immunohistochemical localization to the theca-interstitial cells. Endocrinology 121: 1577  Kuo T C, Endo K, Dharmarajan A M, Miyazaki T, Atlas S J, Wallach E E 1991 Direct effect of angiotensin II on in-vitro perfuse rabbit ovary. J Reprod Fertil 92: 469  Lacy M P McIntosh R P, McIntosh J E 1989 Angiotensin II stimulates an endogenous response in Xenopus laevis ovarian follicles. Biochem Biophys Res Commun 159: 658  Lahav M, Lamprecht S A, Amsterdam A, Lindner H R 1977 Supression of 20ahydroxysteroid dehydrogenase activity in cultured rat luteal cells by prolactin. Mol Cell  195  Endocrinol 6: 293  Lahav M, Freud A, Lindner H R 1976 Abrogation of prostaglandin F of LH-stimulated cyclic AMP accumulation in isolated rat corpora lutea of pregnancy Biochem Biophys Res Commun 68: 1294  Lamberts S W J, Timmers J M, Oosterrom R, Verleun T, Rommerts F G, deJong F H 1982 Testosterone secretion by cultured arrhenoblastoma cells: supression by a luteinizing hormone-releasing hormone agonist. J Clin Endocrinol Metab 54: 450  LaPolt P S, Soto D, Su J G, Campen C A, Vaughan J, Vale W, Hsueh A J W 1989 Activin stimulation of inhibin secretion and messenger RNA levels in cultured granulosa cells. Mol Endocr 3:1666  Lapolt P S, Piquette G N, Soto D, Sinicich C, Hsueh A I W 1990 Regulation of inhibin subunit messenger ribonucleic acid levels by gonadotropins, growth factors, and gonadotropin-releasing hormone in cultured rat granulosa cells Endocrinology 127: 823  Latouche I, Crumeyrolle-Arias M, Jordan D, Kopp N, Augendre-Ferrante B, Cedard L, Haour F 1989 GnRH receptors in human granulosa cells: Anatomical localization and characterization by autoradiographic study. Endocrinology 125: 1739  196  Ledwitz-Rigby F 1990 Gonadotropin-releasing hormone inhibition of LH-stimulated progesterone secretion by porcine granulosa cells in vitro. Dom An Endocr 7: 265  Lee V W K 1983 PMSG-treated immature female rat-a model system for studying control of inhibin secretion. In ‘Factors Regulating Ovarian Function’. (Eds G S Greewald and P F Terranova) pp 157-61 (Raven Press: New York)  Leung P C K, Raymond V, Labrie F 1983 Stimulation of phosphotidic acid and phosphotidyl inosital labeling in luteal cells by luteinizing hormone releasing hormone. Endocrinology 112: 1138  Leung P C K, Goff A K, Armstrong D T 1979 Stimulatory effect of androgen administration in vivo on ovarian responsiveness to gonadotropins. Endocrinology 104: 1119  Leung P C K, Minegishi T, Wang J 1988 Inhibition of follicle-stimulating hormone- and cyclic monophosphate-induced progesterone production by calcium and protein kinase C in the ovary. Am J Obstet Gynecol 158: 350  Li CH, Yamashiro RI), Chung D Gonadotropin-releasing peptide from human follicular fluid: isolation, characterization, and chemical synthesis. Proc Nail Acad Sci USA 1987, 84:959-962  197  Li W, Ho Yuen B H, Leung P C K 1992 Inhibition of progestin accumulation by activin A in human granulosa cells. J Clin Endocrinol Metab 75: 285  Li W, Khorasheh 5, Yuen B H, Ling N, Leung P C K 1993 Stimulation of progesterone secretion by recombinant follistatin-288 in human granulosa cells. Endocrinology 132: 1750  Lightman A, Tarlatzis B C, Rzasa P J, Culler M D, Caride V J, Negro-Vilar A F, Lennard D, DeCherney A H, Naftolin F 1987 The ovaryian renin-angiotensin system: Renin-lilce activity and angiotensin Il/Ill immunoreactivity in gonadotropin-stimulated and unstimulated human follicular fluid. Am J Obstet Gynecol 156: 808  Lightman A, Palumbo A, DeCherney A H, Naftolin F 1989 The ovarian renin-angiotensin system Semin Reprod Endocrinol 7: 79  Ling N, Ying S-Y, Ueno N, Esch F, Denoroy L, Guilemin R 1985 Isolation and partial characterization of a Mr 32,000 protein with inhibin activity from porcine ovarian follicular fluid. Proc Nail Acad Sci USA 82: 7217  Ling N, Ying S-Y, Ueno N, Shimasali 5, Esch F 1986 Pituitary FSH is released by a heterodimer of the B subunits from the two forms of inhibin. Nature 321: 779  198  Lintem-Moore S, Peters H, Moore G P M, Faber M 1974 Follicular development in the infant human ovary. J Reprod Fertil 39: 53  Louvet J-P, Harman S M, Ross G T 1975 Effects of human chorionic gonadotropin, human interstitial cell stimulating hormone, and human fofficle-stimulating hormone on ovarian weights in estrogen-primed hypophysectomized immature female rats. Endocrinology 96: 1179  Ma F, Leung P C K 1985 Luteinizing hormone-releasing hormone enhances polyphosphoinositide breakdown in rat granulosa cells. Biochem Biophys Res Commun 130: 1201  Madej E 1980 Effect of exogenous hormones on the activity of  5-hydrxysteroid  dehydrogenase in cultured granulosa cells from proestrous and preovulatory rat ovarian follicles. Acta Histochem 67: 253  Magoffin D A, Gancedo B, Erickson G F 1989 Transforming growth factor-B promotes differentiation of ovarian thec al-interstitial cells but inhibit androgen production Endocrinology 125: 1951  Makris A, Ryan K J 1975 Progesterone, androstenedione, testosterone, estrone, and estraclial synthesis in hamster ovarian follicle cells. Endocrinology 96: 694  199  Marchetti B, Cioni M 1988 Opposite changes of pituitary and ovarian receptors for LFIRH in ageing rats: Further evidence for a direct neural control of ovarian LHRH receptor activity. Neuroendocrinology 48: 242  Marsh J M, Yang N S T, LeMaire W J 1974 Stimulation of prostaglandin synthesis by LII in Graafian follicles in vitro. Prostaglandins 7: 269  Marsh J M 1976 The role of cyclic AMP in gonadal steroidogenesis. Biol Reprod 14: 30  Marsh J M 1970 The stimulatory effect of luteinizing hormone on axienyl cyclase in the bovine corpus luteum. J Biol Chem 245: 1596  Marsh J M, Savard K 1966 The stimulation of progesterone synthesis in bovine corpora lutea by adenosine-3 ‘,5 ‘-monophosphate. Steroids 8: 133  Mason N, Marsh I, Savard K 1961 An action of gonadotropin in vitro. J Biol Chem 237: 1801  Mason A J, Hayflick 1 5, Ling N, Esch F, Ueno N, Ying S Y, Guillemin R, Niall H, Seeburg P H 1985 Complementary DNA sequence of ovarian follicular fluid inhibin show precursor structure and homology with transforming growth factor-B. Nature 318: 659  200  Mason, A J, Niall, H D and Seeburg, P H 1986 Structure of two human ovarian inhibins. Biochem Biophys Res Commun 135: 957  Massicotte 3, Veilleux R, Lavoie M, Labrie F 1980 An LHRH agonist inhibits FSH induced cyclic AMP accumulation and steroidogenesis inporcine granulosa cells in culture. Biochem Biophys Res Commun 94: 1362  May J V, Schomberg D W 1981 Granulosa cell differentiation in vitro: effect of insulin on growth and functional integrity. Biol Reprod 25: 421  May 3 V, Schomberg D W, Gordon S, Amsterdam A 1985 Concerted effect of insulin and FSH on porcine granulosa cell differentiation: structure function relationships. Soc Study of Reprod, Abst 32  Mayo K E, Cerelli G M, Spiess J, Rosenfeld M G, Evans R M, Vale W 1986 Inhibin Asubunit cDNAs from porcine ovary and human placenta. Proc Nail Acad Sci USA 83: 5849  McGaughey R M 1977 The culture of pig oocytes in minimal medium, and the influence of progesterone and estradiol-17B on meiotic maturation. Endocrinology 100: 39  McLachlan R I, Healy D L, Lutjen P 1, Firniley I K, de Krester M, Burger H G 1987a  201  The maternal ovary is not the source of circulating inhibin levels during human pregnancy. Clin Endocr 27: 663  McLachlan R I, Healy D L, Robertson D M, Burger H G, de Krester D M 1986 The human placenta: a novel source of inhibin. Biochem Biophys Res Commun 140: 485  McLachlan R I, Robertson D M, Healy D L, Burger H G, de Kretser D M 1987b Circulating immunoreactive inhibin levels during the normal human menstrual cycle. J Clin Endocr Metab 65: 954  2 McNAny K P, Henderson K M, Sawers R S 1975b Effects of prostaglandin F and E on the production of progesteone by human granulosa cells in tissue culture 3 Endocrinol 67: 231  McNatty K P, Hunter W M, McNeily A S, Sawers R S 1975a Changes in the concentration of pituitary and steroid hormones in the follicular fluid of human Graafian follicles throughout the menstrual cycle. J Endocrinol 164: 555  McNatty K P 1981 Hormonal correlates of follicular development in the human ovary. Aust I Biol Sci 34: 249  McNatty KP,  Smith DM, Maicris A,  Osathanondh R, Ryan U  202  1979 The  microenvironment of the human antral follicle: Inter- relationships among the steroid levels in antral fluid, the population of granulosa cells, and the status of the oocyte in vivo and in vitro. 3 cm Endocrinol Metab 49:851-860.  McNatty K P, Makris A, DeGrazia C, Osathanondh R, Ryan K 1 1980 The production of progesterone, androgens, and estrogens by granulosa cells, theca tissue, and stromal tissue from human ovaries in vitro. I Cim Endocrinol Metab 51: 1407  Merchenthaler, I, Culler, M D, Petruz, P, and Negro-vilar 1987 A Immunocytochemiccal localization of inhibin in rat and human reproductive tissues. Mol Endocrinol, 54: 239  Merk F B, Botticelli C R, Aibright I T 1972 An intracellular response to estrogen by granulosa cells in the rat ovary; an electron microscope study. Endocrinology 90: 992  Meunier H, Rivier C, Evans R M, Vale W 1988a Gonadal and extragonadal expression of inhibin a, BA and BB subunits in various tissues predicts diverse functions. Proc Nail Acad Sci USA 85: 247  Meunier H, Roberts V I, Sawchenko P E, Hsueh A I W, Vale W 1988b Rapid changes in the expression of inhibin a-, BA-, BB-subunits in hormonally induced immature female rats. Mol Endocrinol 3: 2062  203  Meunier H, Cajander S B, Roberts V J, Rivier C, Sawchenko P E, Hsueh A J W, Vale W 1988c Rapid changes in the expression of inhibin a-, BA-, and BB-subunits in ovarian cell types during the rat estrous cycle. Mol Endocrinol 2: 1352  Michel U, Krozowski Z, McMaster 3, Yu 3 H, Findlay 3 K 1991 The biphase modulation of inhibin mRNA levels and secretion by PMSG in rat granulosa cells in vitro. Repro Fertil Dev 3: 215  Minegishi T, Leung P K C 1985 Luteinizing hormone-releasing hormone stimulates arachidonic acid metabolism in rat luteal granulosa cell. Endocrinology 117: 2001  Miyazaki H, Kondoh M, Ohnishi 3, Masuda Y, Hirose S. Murakami K 1988 High-affinity angioten sin II receptors in the bovine ovary are different from those previously identified in other tissues. Biomed Res 9: 281  I-labeled Mock E 3, Niswender G D 1983 Differences in the rates of internalization of 1 human chorionic gonadotropin, luteinizing hormone, and epidermal growth factor by ovine luteal cells. Endocrinology 113: 259  Mondschein J 5, Schomberg D W 1981 Growth factors modulate gonadoiropin receptor induction in granulosa cell cultures. Science 211: 1179  204  Moore C C, Hum D W, Miller W L 1992 Identification of positive and negative placentaspecific basal elements and a cydicadenosine 3’, 5’-monophosphate response element in the human gene for P45Oscc. Mol Endocrinol 6:2045  Musicki B, Aten R F, Behrman H R 1990 The antigonadotropic actions of PGF and phorbol ester aare not mediated separate processes in rat luteal cells. Endocrinology 126: 1388  Naess 0 1981 Characterization of cytoplasmic progesterone receptors in rat granulosa cells: Evidence for nuclear translocation. Acta Endocrinol (Capenh), 98: 288  Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H 1990 Activin-binding protein from rat ovary is follistatin. Science 247: 836  Nakatani A, Shimasaki S, DePaolo L V, Erickson G F, Ling N 1991 Cyclic changes in follistatin messenger RNA and its protein in the rat ovary during the estrous cycle. Endocrinology 129: 603  Naor Z, Yavin E 1982 Gonadotropin releasing hormone stimulates phospholipid labeling in cultured granulosa cells. Endocrinology 111: 1615  Navickis R J, Jones P B C, Hsueh A J W 1982 Modulation of prolactin receptors by  205  follicle-stimulating hormone in rat granulosa cells by FSH, LH and GnRH. Mol Cell Endocrinol 27: 77  Neer E J, Chow Y K, Garen-Fazio S. Michel T, Schmidt C J, Silbert S 1990 The family of G proteins. In Vanderhoek JY (ed) Biology of Cellular Transducing Signals Plenum Publishing, New York p83  Neil J, Frawley S, Plosky P M, Tindall G T 1981 Dopamine in hypophysial stalk blood of rhesus monkey and its role in regulating prolactin secretion. Endocrinology 108: 489  Nett T M, Akbar A M, Niswender G D (1974) Serum levels of luteinizing hormone and  gonadotropin releasing hormone in cycling, castrated and anestrous ewes. Endocrinology 94: 713  Nimrod A 1977 Studies on the synergistic effect of androgen on the stimulation of progestin secretion by FSH in cultured rat granulosa cells: A search for the mechanism of action. Mol Cell Endocrinol 8: 201  Nishizuka Y 1984 The role of protein kinase C in cell-surface signal transduction and tumor promotion. Nature 308: 693  Niswender G D, Schwall R H, Fitz T A, Farm C, Sawyer H R 1985 Regualation of luteal  206  function in domestic ruminants: New concepts Rec Prog Horm Res 41: 101  Noland T A, Dimino M J 1986 Characterization and distribution of protein kinase C in ovarian tissue. Biol Reprod 35: 863  o W S, Robertson D M, de Krester D M 1989 Inhibin as an oocyte meiotic inhibitor. Mol cell Endocrinol 62: 307  Osterman J, Hammond J M 1979 Effecta of epidermal growth factor, fibroblast growth factor and bovine serum albumin on omithine decarboxylase activity of porcine granulosa cells. Horm Metab Res 11: 485  Palumbo A, Alma M, Lightman A, DeCherney A H, Naftolin F 1988a Angiotensin II affects in vitro steroidogenesis by human granulosa-lutein cells. 70th Annual Meeting of the Endocrine Society, New Orleans, USA  Palumbo A, Jones C, Lightman A, Carcangiu M L, DeCherney A H, Naftolin F 1989 Immunol histochemical localization of renin and angiotensin II in human ovaries. Am J Obstet Gynecol 160: 8  Palumbo A, Pellicer A, DeChemey A H, Naftolin F 1988b Angiotensin action in oocyte maturation in the rat. 35th Annual Meeting of the Society for Gynecologic Investigation,  207  Baltimore, Maryland, 1988, Abstract #107  Pang C, Behrman H 1981 Acute effects of PGF on ovarian and luteal blood flow, luteal gonadotropin uptake in vivo and gonadotropin binding in vitro. Endocrinology 108: 2239  Parinaud, J, Beaur, A, Bourreau, E, Vieitez, G and Pontonnier, G 1988 Effect of a luteinizing hormone-releasing hormone agonist (buserelin) on steroidogenesis of cultured human preovulatory granulosa cells Fert Steril 50: 597  Pasqualini J R, Nguyen B J 1980 Progesterone receptors in fetal uterus and ovary of the guinea pig: Evolution during fetal development and induction and stimulation in estradiol primed animals. Endocrinology 106: 1160  Patwardhan VV, Lanthier A 1974 Effect of prostaglandin F, on the hCG-stimulated  progesterone production in human corpora lutea. Prostaglandins 27: 465  Payne R W, Runser R H 1958 The influence of estrogen and androgen on the ovarian response of hypophysectomized immature rats to gonadotropins. Endocrinology 62: 313  Payne R W, Heilbaum A A, Owens Jr I N 1956 The effect of androgens on the the ovaries  and  uterus of the  estrogen-treated hypophysectomized immature rat.  Endocrinology 59: 306  208  Pellicer A, Palumbo A, DeCherney A H, Naftolin F 1988 Blockage of ovulation by angiotensin antagonist. Science 240: 1660  Pelletier G, Seguin C, Dube D, St-Amaud R 1982 Disthbution of LHRH receptors in the rat ovary. Biol Reprod (Suppl) 26: 151A  Pepperell 3, Preston S 1, Behrman H R 1989 The antigonadotropic action of prostaglandin F is not mediated by elevated cytosolic calcium levels in rat luteal cells. Endocrinology 125: 144  Peterson C M, Zhu C, Mukaida T, Butler T A, Woessner J F, LeMaire W 3 1993 The angiotensin II antagonist, saralasin, inhibits ovulation in the perfused rat ovary. 3 Obstet Gynecol 168: 242  Petraglia F, Lim A T W, Vale W 1987 Adenosine 3’,5’-monophosphate, prostaglandins, and epinephrine stimulate the secretion of immunnoreactive gonadotropin-releasing hormone from cultured human placental cells. Clin Endocrinol Metab 65: 1020  Petraglia F, Sawchenko Paul, Lim A T W, Rivier 3, Vale, W 1987 Localization, secretion and action of inhibin in human placenta. Science 237: 187  Petraglia F, Garuti GC, Caiza L, Roberts V, Giardino L, Genazzani A R, Vale, W,  209  Meunier H 1991 Inhibin subunits in human placenta: Localization and messenger ribonucleic acid levels during pregnancy. Am J Obstet Gynecol 165: 750  Pieper D R, Richards J S, Marshall J C 1981 Ovarian LHRH receptor content: characerization, distribution, and induction by GnRH. Endocrinology 108: 1148  Philibert D, Ojasoo T, Raynaud J-P 1977 Properties of the cytoplasmic progestin-binding protein in the rabbit uterus. Endocrinology 101: 1850  Philpott J E and Peron F G 1971 A microassay procedure for 3-8-hydroxysteroid dehydrogenase based on substrate depletion. Endocrinology 88:1082  Popkin R, Fraser H M, Jonassen 3 1983a Stimulation of androstened.ione and progesterone release by LHRH and LHRH-agonist from isolated rat preovulatory follicles. Mol Cell Endocrinol 29: 169  Popkin R, Bramley T A, Currie A, Shaw R W, Baird D T, Fraser H M 1983b Specific binding of Luteinizing hormone releasing hormone to human luteal tissue. Biochem Biophys Res Commun 114: 750  Powell W S, Hammmarstrom 5, Samuelesson B, Sjoberg B 1974 Prostaglandin F receptor in human corpora lutea Lancet 1: 120  210  Pucell A G, Bumpus F M, Husian A 1987 Rat ovarian angiotensin II receptors: Characterization and coupling to estrogen secretion. J Biol Chem 262: 7076  Pucell A G, Bumpus F M, Husian A 1988 Regulation of angiotensin II receptors in cultured rat ovarian granulosa cells by follicle-stimulating hormone and angiotensin II. J Biol Chem 263: 11954  Rabinovici J, Spencer S  j,  Doldi N, Goldsmith P C, Schwall R, Jaffe R B 1990 J Clin  Endocrinol Metab 7 1:1396  Rani C S S, Ekholm C, Billig H, Magnusson C, Hillensjo T 1983 Biphasic effect of gonadotropin-releasing hormone on progestin secretion by rat granulosa cells. Biol Reprod 28: 591  Rao M C, Midgley Jr A R, Richards J S 1978 Hormonal regulation of ovarian cellular proliferation. Cell 14: 71  Ranta T, Baukal A, Knecht M, Korhonen M, Catt K I 1983a Inibitory actions of a gonadotropin-releasing hormone agonist on ovarian fofficle stimulating hormone receptors and adenylate cyclase in vitro. Endocrinology 112: 956  Ranta T, Knecht M, Darbon J-M, Baukal A.J, Catt K.J 1983b Calcium dependence of the  211  inhibitory effect of gonadotropin-releasing hormone on luteinizing hormone-induced cyclic AMP production in rat granulosa cells. Endocrinology 113: 427  Ranta T, Knecht M, Kody M, Catt K J 1982 GnRH receptors in cultured rat granulosa cells: mediation of the inhibitory and stimulatory actions of GnRH. Mol Cell Endocrinol 27: 233  Richards J S 1975 Estradiol receptor content in rat granulosa cells during follicular development: modification by estradiol and gonadotropins. Endocrinology 97: 174  Richards J S, Ireland J J, Rao M C, Bemath G A, Midgley Jr A R, Reichert Jr L E 1976 Ovarian follicular development in the rat: Hormone receptor regulation by estradiol, follicle-stimulating hormone and luteinizing hormone. Endocrinology 99: 1562  Richards J S, Sehgal N, Tash J S 1983 Changes in content and cAMP-dependent phosphorylation of specific proteins in granulosa cells of preantral and preovulatory ovarian follicles and in corpus lutea. 3 Biol Chem 1983, 258: 5227  Richardson M C, Masson G M 1980 Progesterone production by dispersed cells from human corpus luteum: Stimulation by gonadoiropins and prostaglandin F: Lack of response to adrenaline and isoprenaline J Endocrinol 87: 247  212  Richardson M C 1986 Hormonal control of ovarian luteal cells. Oxford Rev Reprod Biol 8: 321  Robertson D M, Farnworth P G, Clarke L, Jacobsen J, Cahir N F, Burger H G, de Kretser D M 1990a Effects of bovine 35 KDa FSH-suppressing protein on FSH and LH in rat pituitary cells in vitro: Comparison with bovine 31 KDa inhibin. J Endocrinol 124: 417  Robertson D M, Foulds L M, de Vos F, Leversha L, DeKrester D M 1990b Identification of inhibin and inhibin-related proteins in human follicular fluid. Reprod Fertil Dev 2: 327  Robertson D M, de Vos F L, Foulds L M, Mclachlan R L, Burger H G, Morgan F J, Heam M T W, de Kretser D M 1986 Isolation of a 31 kda form of inhibin from bovine fofficular fluid. Mol Cell Endocr 44: 271  Robertson D M, Foulds L M, Leversha L, Morgan F J, Hearn M T W 1985. Isolation of inhibin from bovine follicular fluid. Biochem Biophys Res Commun 126: 220  Robertson D M, Klein R, de Vos F L, McLachlan R I, Wettenhall R E H, Hearn M T W, Burger H G, de Kretser D M 1987 The isolation of polypeptides with FSH suppressing activity from bovine follicular fluid which are structurally different to inhibin. Biochem Biophys Res Comm 149: 744  213  Rogers RJ, O’Shea JO, Findlay JK 1983 Progesterone production in vitro by small and large ovine luteal cells J Reprod Fertil 69: 113  Rogers RJ, O’Shea ID, Findilay 1K 1985 Do small and large luteal cells of the sheep interact in the production of progesterone?J Reprod Fertil 75: 85  Ross C A, Meldolesi J, Milner T A, Sotah T, Supattapone S, Snyder S H 1989 Inositol 1,4,5-triphosphate receptor localized to endoplasmic reticulum in cerebellar Purkinje neurons. Nature 339:468  Saiduddin S, Zassenhaus H P 1978 Effect of testosterone and progesterone on the esiradiol receptor in the immature rat ovary. Endocrinology 102: 1069  Sailer-Cheater T M, Cheater G S 1981 Dose response analysis of GnRH stimulation of hCG release from human term placenta. Biol Reprod 25: 353  Sailer-Cheater T M, Cheater G S, Vickery B H, Nestor J J 1984 Inhibition of hCG, a hCG and progesterone release from human placental tissue in vitro by a GnRH antagonist. Life Sci 32: 2741  Saito S, Nakamura T, Titani K, Sugino H 1991 Production of activin-binding protein by rat granulosa cells in vitro. Biochem Biophys Res Commun 176: 413  214  Salari H, Braquet P, Borgeat P 1984 Comparative effects of indomethacin, acetylenic acid, 15-HETE, nordihydroguaiaretic acid and BW 755C on the metabolism of arachidonic acid in human leukocytes and platelets. Prostaglandins Leukotriences Med, 13:53  Savion N, Lui G M, Laherty R, Gospodarowicz D 1981 factors controlling proliferation and progesterone production by bovine granulosa cells in serum free medium. Endocrinology 109: 409  Schmid C, Steiner T, Froesch E R 1984 Insulin-like growth factor I supports differentiation of cultured osteoblast-like cells. FEBS lett 173: 48  Schneyer A L, O’Neil D A, Crowley Jr W F 1992 Activin-binding proteins in human serum and follicular fluid. J Clin Endocrinol Metab 74: 1320  Schomberg D W, Cunn E B 1984 Growth factors and ovarian function In: Gonadal proteins and peptides and their biological significance. Eds Sairam MR. Atkinson LE pp2O9-l4 World Scientific Publishing Co, Singapore  Schomberg D W, Coudert S P. Short R V 1967 Effects of bovine luteinizing hormone and human chorionic gonadotropin on the bovine corpus luteum in vivo. J Reprod Fertil 14: 277  215  Schreiber I R, Nakamura K, Truscello A M, Erickson G F 1982a Porgestins inhibit FSH stimulated steroidogenesis in cultured rat granulosa cells. Mol Cell Endocrinol 25: 113  Schreiber J R, Nakamura K, Weinstein D B 1982b Androgens and FSH synergistically stimulate rat ovary granulosa cell utilization of rat and human lipoproteins In: Greenwald G, Terranova PF (eds) Factors Regulating Ovarian Function Raven Press, New York, p 311  Schreiber I R, Erickson G F 1979 Progesterone receptor in the rat ovary: further characterization and localization in the granulosa cell. Steroids 34: 459  Schreiber J R, Nakamura K, Erickson G F 1981 Progestins inhibit FSH-stimulated granulosa cell estrogen production at a post cyclic AMP site. Mol Cell Endocrinol 21: 161  Schreiber I R, Ross G T 1976 Further characterization of a rat ovarian testosterone receptor with evidence for nuclear translocation. Endocrinology 99: 590  Schuler L A, Scavo L, Kirsch T M, Flickinger G L, Strauss J F 1979 Regulation of de novo biosynthesis of cholesterol and progestins, and formation of cholesteryl ester in rat corpus luteum by exogenous sterol. I Biol Chem 254: 8662  Schwall R H, Mason A I, Wilcox I N, Bassett S G, and Zelezinik Al 1990 Localization  216  of inhibin/activin subunit mRNAs within the primate ovary. Mol Endocrinol 4: 75  Seeburg P H, Adelman J P 1984 Nature (London) 311: 666 Seibel M M, Smith D, Diugi A M, Levesque L 1989 Periovulatory folliclular fluid hormone levels in spontaneous human cycles. J Clin Endocrinol Metab 68: 1073  Sharpe K L, Bertero M C, Lyon B P, Muse K N, Vernon M W 1990 Follicular atresia and infertility in rats treated with a gonadotropin-releasing hormone antagonist Endocrinology 127: 251  Sheela Rani C 5, Nordenstorm K, Norjavaara E, Ahren K 1983 Development of catecholamine responsiveness in granulosa cells from preovulatory rat follicles Depandence on preovulatory luteinizing hormone surge. Biol Reprod 28: 1021  Shimasaki 5, Koga M, Buscagklia M L, Simmons D M, Bicsak T A, Ling N 1989 Follistatin geneexpression in the ovary and extragonadal tissue. Mol Endocrinol 3: 651  Shukovski L, Findlay J K, Robertson D M 1991 The effect of follicle-stimulating hormone-suppressing protein or follistatin on luteinizing bovine granulosa cells in vitro and its antagonistic effect on the action of activin. Endocrinology 129: 3395  Shukovski L, Findilay J K 1990 Activin-A inhibits oxytosin and progesterone production  217  by preovulatory bovine granulosa cell in vitro. Endocrinology 126: 2222  Siler-Khodr T M, Khodr G S, Valenzuela G 1984 Immunoreactive gonadotropin-releasing hormone level in maternal circulation throughout pregnancy. Am J Obstet Gynecol 150: 376  Simpson E R, McDonald P C 1981 Endocrine physiology on the placenta. Annu Rev Physiol 43: 163  Simpson E R, Rochelle D B, Carr B R, Mcdonald P C 1980 Plasma lipoproteins in follicular fluid of human ovaries. 3 Clin Endocrinol Metab 51: 1469  Skinner M K, Keski-Oja J, Osteen K G, Moses H L 1987 Ovarian thecal cells producetransferming growth factor-B which can regulate granulosa cell growth Endocrinology 121: 786  Smith D M, Tenney D Y 1980 Effect of steroids on mouse oocyte maturation in vitro. 3 Reprod Fertil 60: 331  Spat A 1988 Stimulus-secretion coupling in angiotensin stimulated adrenal glomerulosa cells. 3 Steroid Biochem 29: 443  218  Speth R C, Husain A 1988 Distribution of angiotensin-converting enzyme and angiotensin TI-receptor binding sites in the rat ovary. Biol Reprod 38: 695  Sramek SJ, Wallow III, Day RP, Ehrlich EN 1988 Ocular renin-angiotensin: immunohistochemical evidence for the presence of prorenin in eye tissue. Invest Ophtalmol Vis Sci, 29: 1749  St-Arnaud R, Walker P, Kelly P A, Labrie F 1983 Rat ovarian epidermal growth factor receptors: characterization and hormonal regulation. Mol Cell Endocrinol 31: 43  Steele G 5, Currie W D, Ho Yuen B, Leung P K C 1993 Acute stimulation of hCG secretion by recombinant human activin-A in first trimester human trophoblast. J Clin Endocr Metab 1993, in press  Sterling D, Magness R R, Stone R, Waterman M R, Simpson E R 1990 Angiotensin II inhibits LH-stimulated cholesterol side chain cleavage expression and stimulates basic fibroblast growth factor expression in bovine luteal cells in primary culture. J Biol Chem 265: 5  Strickland S, Beers W H 1976 Studies on the role of plasminogen activators in ovulation. J Biol Chem 251: 5694  219  Sugino H, Nakamura T, Hasegawa Y, Miyamoto K, Igarashi M, Eto Y, Shiba H, Titani K 1988a Identification of a specific receptor for erythroid differentiation factor on follicular granulosa cell. J Biol Chem 263: 15249  Sugino H, Nakamura T, Hasegawa Y, Miyamoto K, Igarashi M, Eto Y, Shibai H, Titani K 1988b Erythroid differentiation factor can modulate follicular granulosa cell functions. Biochem Biophys Res Commun 153: 281  Sugino H, Nakamura T, Hasegawa Y, Miyamoto K, Abe Y, Igarashi M, Eto Y, Shibai H, Titani K 1988 Erythroid differentiation factor can modulate follicular granulosa cell functions. Biochem Biophys Res Commun 153, 281  Sum 0, Robertson, D M, de Kretser D 1989 Inhibin as an oocyte meiotic inhibitor. Mol Cell Endocr 62: 307  Sumuels L T, Helmreich M L, Lasater M B, Reich H 1951 An enzyme in endocrine -3-hydrxysteroids to x, B unsaturated ketones. Science 113: 490 tissues which oxidizes 5  Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H 1990 Activin-binding protein from rat ovary is follistatin. Science 247: 836  Tanimoto K, Tamura K, Ueno N, Usuki 5, Murakami K, Fukamizu A 1992 Regulation  220  of activin beta A mRNA level by cAMP. Biochem Biophys Res Commun 182: 773  Tashijian A H, Hohmann E L, Antoniades H N, Levine L 1982 Platelet-derived growth factor stimulates bone resorption via a prostaglandin-mediated mechanism. Endocrinology 111: 118  Thomas J P, Dorflinger L I, Behrman H R 1978 Mechanism of the rapid antigonadotropic action of prostaglandins in cultured luteal cells. Proc Natl Acad Sci USA 75: 1344  Triebwasser WF, Clark MR, LeMaire WJ, Marsh JM 1978 Localization and in vitro synthesis of prostaglandins in components of rabbit preovulatory Graafian follicles. Prostaglandins, 16: 621  Tsafriri, A, Vale, W and Hsueh A J W 1989 Effect of transforming growth factors and inhibin-related proteins on rat preovulatory Graafian follicles in vitro. Endocrinology 125: 1857  Tsang B K, Moon Y S, Simpson C W, Armstrong D T 1979 Androgen biosynthesis in human ovarian follicles: cellular source, gonadotropic control, and adenosine 3’,S’monophosphate mediation. J Clin Endocrinol Metab 48: 153  Tsang B K, Armstrong D T, Whitfield J F 1980 Steroid biosynthesis by isolated human  221  ovarian follicular cells in vitro. J Clin Endocrinol Metab 51: 1407  Tureck R W, Mastroianni Jr L, Blasco L, Strauss Ill, J F 1982 Inhibition of human granulosa cell progesterone secretion by a gonadotropin-releasing hormone agonist. J Clin Endocrinol Metabl 54: 1078  Tureck R W, Strauss J F 1111982 Progesterone synthesis by luteinized human granulosa cells in culture: The role of de novo sterol synthesis and lipoprotein-carried sterol. 3 Clin Endocrinol Metab 54: 367  Turner I M, Saunders P T K, Shimasaki 5, Hiller S G. 1989 Regulation of inhibin expression by FSH and estradiol in cultured rat granulosa cells. Endocrinology 125:2790  Turo K, Florini J 1982 Hormonal stimulation of myoblast differentiation in the absence of DNA synthesis. Am J Physiol 243: C278  Ueno N, Ling N, Ying S-Y, Esch F, Shimasaki 5, Guillemin R 1987 Isolation and partial characterization of follistatin : a single-chain Mr 35,000 monomeric protein that inhibits the release of follicle-sthnulating hormone. Proc Nail Acad Sci USA 84: 8282  Vale W, River J, Vaughan J, McClintock R, Corrigan A 1986 Purification and characterization of an FSH-releasing protein from porcine ovarian follicular fluid. Nature  222  321: 776  Valejo M, Jackson M T, Lightman S, Hanley M R 1987 Occurrence and extracellular actions of inositol pentakis-and hexakisphosphate in mammalian brain. Nature 330: 656  Veldhuis J, Kolp L A, Toaff M E, Stauss III J F, Demmers L M 1983 Mechanisms subserving the tropic actions of insulin on ovarian cells In vitro studies using swine granulosa cells. J Clin Invest 72: 1046  Veldhuis J D 1985 Role of estradiol as a biological ampliffier of hormone action in the ovary: synergistic interaction of estradiol with insulin-like growth factor tyoe I (IGF I/sometomedin-C). Endocrinology 1 16(suppl)[Abstr 817]  Veidhuis ID, Kiase PA  1982a Mechanisms by which calcium ions regulate the  steroidogenic actions of luteinizing hormone in isolated ovarian cells in vitro. Endocrinology, 111: 1  Veldhuis J D, Kiase P A, Strauss J F III, Hammond J M 1982b The role of esiradiol as a biological amplifier of the actions of follicle-stimulating hormone: in vitro studies in swine granulosa cells. Endocrinology 111: 144  Volpe P, Krause K H, Hashimoto S, Zorzato F, Pozzan T, Meldolesi J, Lau D P 1988  223  “Calciosom&’, a cytoplasmic orgenelle: the inositol 1,4,5-triphosphate-sensitive Ca 2 store of nonmuscle cells. Proc Natl Acad Sci USA 85: 1091  Wang C, Hsueh A J W, Erickson G F 1981 LH stimulation of estrogen secretion in cultured granulosa cells Mol Cell Endocrinol 24: 17  Wang J, Baimbridge K G, Leung P C K 1989 Perturbation of intracellular calcium ion concentration in single rat granulosa cells by angiotensin II. Endocrinology 124: 1094  Ward D N 1978 Chemical approaches to the structure-function relationships of luteinizing hormone (lutropin). In Mckerns, K W (ed), Structure and function of gonadotropins, Plenum Press, New York, p31  Weishaar RE, Panek RL, Major TC, Simmerman J,Rapundalo ST, Taylor DG Jr 1991 Evidence for a functional tissue renin-angiotensin system in the rat mesenteric vasculature and its involvement in regulating blood pressure. J Pharmacol Exp Ther, 256: 568  Welsh Jr T H, Zhuang L Z, Hsueh A J W 1983 Estrogen augmentation of gonadotropin stimulated progestin biosynthesis in cultured rat granulosa cells Endocrinology 112: 1916  Welsh Jr T H, Jones P B C, Ruiz de Galarreta C M, Fanjul L F, Hsueh A J W 1982 Androgen regulation of progestin biosynthetic enzymes in FSH-treated rat granulosa cells  224  in vitro. Steroids 40: 691  White S 5, Ojeda S R 1981 Changes in ovarian luteinizing hormone and folliclestimulating  hormone  receptor  content  and  in  gonadotropin-induced  omithine  decarboxylase activity during prepubertal and pubertal development of the female rat. Endocrinology 109: 152  Wickings E J, Eidne K A, Dixson A F, Hillier S G 1990 Gonadotropin-releasing hormone analogs inhibit primate granulosa cell steroidogenesis via a mechanism distinct from that in the rat. Biol Reprod 43: 305  Woodruff T K, Meunier H, Jones P B C, Hsueh A 3 W, Mayo K E 1987 Rat inhibin: molecular cloning of a and B subunit complementary deoxyribonucleic acids and expression in the ovary. Mol Endocrinol 1: 561  Woodruff T K, D’agostino J B, Schwartz N B, and Mayo K E 1988 Dynamic changes in inhibin messenger RNAs in rat ovarian follicles during the reproductive cycle. Science 239: 1296  Woodruff T K, Lyon R J, Hansen S E, Rice G C, Mather J P 1990 Inhibin and activin locally regulate rat ovarian folliculogenesis. Endocrinology 127: 3196  225  Woodruff T K, Battaglia J, Borree J, Rice G C, Mather J P 1991 Labeling inhibin and identifying inhibin binding to cell surface receptor. Methods In Enzymology Vol 198: 347  Xiao S. Findley J K, Robertson D M 1990 The effect of bovine activin and folliclestimulating hormone (FSH) suppressing protein/follistatin on FSH-induced differentiation of rat granulosa cells in vitro. Mol Cell Endocr 69: 1  Yamoto M, Minami S, Nakano R 1991 Immunohistochemical localization of inhibin subunits in human corpora lutea during menstrual cycle and pregnancy. J Clin Endocrinol Metab 73:470  Yanagashita M, Hascall V C, Rodbard D 1981 Biosynthesis of proteoglycan by rat granulosa cells cultured in vitro: Modulations by gonadotropins, steroidal hormones, prostaglandins, and a cyclic nucleotide. Endocrinology 109: 1641  Yen S S C 1982 Neuroendocrine regulation of gonadotropin and prolactin secretion in women: disorders in reproduction. In Vaitukaitis, I L (ed), Current Endocrinology, Clinical Reproductive Neuroendocrinology Section, Elsevier Biomedical Press, New York, pp 137  Ying S-Y, Becker A, Ling N, Ueno N, Guillemin R 1986 Inhibin and B type transforming growth factor (TGF-B) have opposite modulating effects on the follicle stimulating  226  hormone (FSH)-induced aromatase activity of cultured rat granulosa cells. Biochem Biophys Res Commun 136: 969  Ying S-Y, Becker A, Swanson G, Tan P, Ling N, Esch F, Ueno N, Shimasaki 5, Guillemin R 1987 Follistatin specifically inhibits pituitary follicle stimulating hormone release in vitro. Biochem Biophys Res Commun 149: 133  Ying S-Y, Becker A, Swanson G, Tan P, Vadleigh D, Czvik J, Ling N, Ueno N, Esch F, Chiang T C, Hu R, Dong M H, Sato K, Munegumi T, Guillemin R 1987 Reciprocal relationships between the secretion of follicle stimulating hormones (FSH) and production of inhibin in the rat. In ‘Inhibin-Non-Steroidal Regulation of Follicle Stimulating Hormone Secretion’ (Eds Burger HG, de Krester DM, Findley JK, and Igarashi M) Vol 42 pp 135 (Raven press: New York)  Ying S Y 1988 Inhibins, activins and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 9: 267  Yohkaichiya T, Bishop 5, Healy D 1989 Serum inhibin levels in human pregnancy and hydatidiform mole. Proceedings of the 71st Annual Meeting of the US Endocrine society, p4.18 (abstract 1584)  Yoshimura Y, Karube M, Koyama N, Shiokawa, Nanno T, Nakamura Y 1992 Angiotensin  227  II directly induces follicle rupture and oocyte muturation in the rabbit. FEBS 307: 305  Zeleznik A J 1981 Premature elevation of systemic estradiol reduces serum levels of follicle-stimulating hormone and lengthens the follicular phase of the menstrual cycle in rhesus monkeys. Endocrinology 109: 352 Zhang Z W, Carson R, Herington A, Lee V and Burger H 1987 Follicle-stimulating hormone and somatomedin-C stimulated inhibin production by rat granulosa cells in vitro. Endocr 120: 1633  Zhang Z W, Lee V W K, Carson R S. Burger H G 1988 Selective control of the rat granulosa eel inhibin production by FSH and LH in vitro. Mol Cell Endocrinol 56: 35  Zhiwen, Z, Carson, R S, and Burger, H G 1988a Selective control of rat granulosa cell inhibin production by FSH and LII in vitro. Mol Cell Endocrinol, 54: 213  Zhiwen Z, Findlay J K, Carson R 5, Herington A C, Burger H F 1988b Transforming growth factor-B enhances basal and FSH-stimulated inhibin production by rat granulosa cells in vitro. Mol Cell Endocrinol 58: 161  Zhuang L Z, Adashi E Y, Hsueh A I W 1982 Direct enhancement of gonadoiropin stimulated ovarian  estrogen biosynthesis by  Endocrinology 110: 2219  228  estrogen  and clomiphene citrate.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088168/manifest

Comment

Related Items