- Email: [email protected]
In vivo and in vitro effects of androgens on rat ovarian granulosa cells
In vivo and in vitro effects of androgens on rat ovarian granulosa cells Everett Anderson, PhD Boston, Massachusetts In this communication attention is called to some features of ovarian cyst formation in rats after exposure to 6 mg/100 gm body weight of dehydroepiandrosterone for 21 days. Cysts were identified by the transformation of mural granulosa cells into epithelial cells. The invaginations of the basal plasma membrane of the epithelial cells of the cyst are thought to be indications of the cells' involvement in active endocytosis. This endocytosis may be one strategy for moving fluid into the cyst. The filamentous actin of granulosa cells cultured in the presE1nce of 10- 5 mol/L dehydroepiandrosterone was found to be deficient when compared with that of controls. This lack of filamentous actin coupled with other organelle degeneration is believed to lead to early atresia of granulosa cells in vitro in the presence of high concentrations of androgens. (AM J OBSTET GVNECOL 1989;160:782-8.)
Key words: Androgen, granulosa cell, atresia, cyst formation Since Stein and Leventhal' discovered polycystic ovarian disease in women, numerous studies have been made in an effort to clarify this multidimensional malady! In certain mammalian models, cystic ovaries have been produced by numerous treatments. These treatments include estradiol valerate," 4 dehydroepiandrosterone (DHEA)5 treatment with antibodies to luteinizing hormone-releasing hormones," and treatment of hypothyroid rats with human chorionic gonadotropin. 7 Singh described other methods for producing polycystic ovarian disease. s In addition to producing cysts, the aforementioned regimens also induce follicular atresia. With the production of ovarian cysts with many different hormonal manipulations came the excitement of having a model to analyze certain aspects of the etiologic factors of the disease. Some authors" have described the unusual histologic and cytologic features. Ovaries obtained from patients with polycystic ovarian disease have, for the most part, shown increased thickness of the collagen compromising the tunica albugenia, multiple atretic follicles, follicular cysts, and a hyperplasia of cells of the theca interna. Other investigators lO have concerned themselves with the hormonal profiles of the disease. In the
From the Department of Anatomy and Cellular Bzology and the Laboratory of Human Reproduction and Reproductive Biology, Harvard MedIcal School. ThIS tnvestigation was supported by National Institutes of Health Grants HD06645 (Center Grant of Laboratory of Human Reproduction and ReproductIve Biology) and HD14574. Presented as the Joseph Pnce Oratzon at the Seventh Annual Meettng of the Amencan Gynecological and Obstetrical Society, Napa, California, September 8-10, 1988. Reprint requests: Dr. Everett Anderson, Harvard Medical School, Department of Anatomy and Cellular Biology, LHRRB. 45 Shattuck St., Boston, MA 02115.
782
hormonal studies it has been found that there is an abnormal gonadotropin secretion with an elevated acyclic circulating concentration of androgens and/or estrogens. In 1985, Billiar et aI., II using a primate, the rhesus monkey (Macacca mulatta), were able to elevate levels of androstenedione or estrone. Under these conditions, polycystic changes (structural) in the ovary were also observed. In women with polycystic ovarian disease, Mahesh and Greenblatt l2 found that de hydroepiandrosterone and androstenedione and / or estrone are usually elevated, whereas testosterone or estradiol may not be. In this communication attention is called to some observations of the in vivo and in vitro effects of dehydroepiandrosterone on rat granulosa cells. * Results and comment
In vivo studies. The general histologic features of the rat ovary are well known l3 and need not be detailed here. Suffice it to say that in a section of the mature ovary there are nonantral and antral follicles in different stages of development and in different stages of atresia. Initially, the oocytes of nonantral follicles are surrounded by a squamous layer of granulosa cells that rest on a basement membrane. Once the antrum is formed, the formation of which is due to the action of gonadotropins, the inner granulosa cells are rather polyhedral in shape and are associated with the oocyte, whereas the outer ones are rather columnar and rest *Animals used in this study were maintained in accordance with the guidelines of the Committee on Animals of the Harvard Medical School and those prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Resources, National Research Council (DHHS publication no. (NIH) 85-23, revised, 1985).
Androgens on ovanan granulosa cells
Volullle HiO 1\ lImher 4-
783
Figs. 1 and 2. Electron nllcrographs of portions of epithelial cells lining ova nan cyst. BM, Basement membrane; anuw, tight junction; GJ. gap junction; C, cy,t lumen; Sf, smooth surface 1l1vagination (2). (Original magnificatIon x SO,()OO )
on a basement membrane. Granulosa cells are coupled to each other and to the oocyte via gap (communicating) junctions. 11 ", Another feature of the antral follicle is the presence of an inner vascularized steroid-secreting theca that is situated between the basement membrane and the theca externa, which is non vascularized and is composed of fusiform-shaped cells. Each ovary is covered with an ovarian epithelium that is composed of cuboidal-columnar cells and rests on a basement membrane. This epithelium is continuous with the mesothelium. Beneath the ovarian epithelium are some collagen fibers and some stromal cells. In humans," 'h and monkeys (unpublished observations), the collagen fibers are organized into a definitive layer known as the tunica albugenia. As indicated previously, it is well known that patients with polycystic ovarian syndrome produce an excessive amount of androgens.' Mahesh and Greenblatt" believe that androgens appear to be a contributing component to the syndrome." It has been shown that relatively large amounts of DHEA and ~' androstenedione can be isolated from patients with polycystic ovaries." In 1962 Roy et al.'" injected DHEA and ~' androstenedione into immature female rats over a 21day period and observed precocious ovulation, with cystic and atretic follicles in the ovarv; the rats became acyclic.""·"' After the injection, the serum levels of follicle-stimulating hormone and luteinizing hormone
showed an ovulatory surge. Subsequently, folliclestimulating hormone returned to control levels whereas luteinizing hormone was extremely low. When the treatment was withdrawn, the animals became cyclic over time. Equivalent changes can be induced in mature cycling rats." It is interesting that Billiar et al." used ~ '-androstenedione to induce follicle cysts and atresia in the rhesus monkey. In collaboration with Drs. Marcus Walker and MongTing Lee (unpublished observations), my colleagues and I used an equivalent protocol to induce, among other things, cyst development and atresia. We examined ovaries obtained 10, 15, and 20 days after injecting DHEA. Ovaries removed on day 20 were fixed in Bouin's solution and sectioned for light microscopy; corpora lutea, and follicular cysts like those demonstrated by Knudsen et al.'o were seen. Electron microscopic examination of granulosa cells of some antral follicles showed an increase in smooth endoplasmic reticulum, and the mitochondria had acquired tubular cristae, The aforementioned organelles are characteristics of steroid-secreting cells. 23 The granulosa cells also had blebs, and the blebs were pinched off and removed by macrophages. We believe that the formation of an ovarian cyst (l to 2 mm in diameter) is complete when the granulosa cell ba~al layer becomes (transformed) epithelial (Figs. 1 and 2). In addition to maintaining association with each other via gapjunctions, the "transformed" granulosa cells acquire tight junctions, (Tight
784
Anderson
April 1989 Am .I ObSlel Gynecol
Fig. 3. Electron micrograph of portion of granulosa cell from atretic follicle (monkey). BM, Basement membrane; AG, annular gap junctions. (Original magnification x 60,000.)
junctions are not a feature of granulosa cells.) These epithelial cells rest on an intact basement membrane and acquire many smooth surface invaginations and some coated ones. The non coated invaginations are believed to continue to invaginate and eventually produce a fluid-filled vesicle containing macromolecules from the interstitium and move them into the cyst by transcytosis, a process frequently encountered in the microcirculatory system!' The coated invaginations are believed to remove, from the cell's surface, special molecules for specific cellular functions via receptormediated endocytosis."' During cyst formation the theca interna and theca externa appear to be lost. How this is accomplished remains unexplained. Atretic follicle. As has been described many times, during the regular ovarian cycle numerous follicles un-
dergo atresia!6 We have also observed many atretic follicles when DHEA was administered to immature rats for 20 days. This degenerative condition appears to favor not only the antral follicles but also nonantral follicles. It is unknown which compartment of the follicle degenerates first, the oocyte or the granulosa cells. In any event some of the manifestations of atresia in granulosa cells are pyknotic nuclei and usually a fragmentation of the cytoplasm. We have not seen macrophages within atretic follicles with fragments of degenerating granulosa cells in their cytoplasm. The basement membrane of the atretic follicle becomes detached from the basal layer of granulosa cells and frequently becomes folded. We also saw this phenomenon in atretic follicles of the monkey after treatment with androstenedione (Fig. 3). We have not seen unusual changes of the thecal components. The fate of the theca
Androgens on ovarian granulosa cells
Volume 160 l\umber 4
785
Figs. 4 to 6. Phase-contrast photomicrographs of control cultures. 4, 2 days; 5, 4 days; 6, 6 days. (Original magnification x 450.)
Figs. 7 to 9. Phase-contrast photomicrographs of cells grown in the presence of 10- 5 mollL DHEA. 7, 2 days; 8, 4 days; 9, 6 days. (Original magnification x 450.)
externa is unknown; however, the thecal interna is believed to persist as islands of interstitial tissue.27 In vitro studies. Granulosa cells can be easily expressed from small antral follicles suitable for culturing. We cultured granulosa cells in McCoy's medium with and without the presence of lO-' mollL DHEA or lO-' moUL LlI-androstenedione."8 The reason for using the hormonal concentration was the known circulating and tissue concentrations of the steroids."" Anderson et al."" used this concentration of DHEA or Ll4_
androstenedione and found induced mitochondrial changes (from lamellae cristae to tubular cristae) and the acquisition of smooth endoplasmic reticulum. These organelle changes are similar to those observed in experiments in situ, as pointed out above. These cells showed degenerative changes after 4 days in culture, whereas control cells survived beyond 6 days. We concluded that the aforementioned study supported our hypothesis that high levels of androgens can induce atresia in vitro.
786
Anderson
Apnl1989 Am J Obstet Gvnecol
Figs. 10 to 12. Photomicrographs of cells from control cultures stained with rhodamine phalloidin. 10, 2 days; 11, 4 days; 12, 6 days. Stress fibers are conspicuous longitudinal fluorescent structures. (Original magnification x 450.)
Figs. 13 to 15. Photomicrographs of cells grown in IO-J mollL dehydroepiandrosterone and stained with rhodamine phalloidin. 13, 2 days; 14, 4 days; 15, 6 days. Very few stress fibers are found in cells grown in DHEA. (Original magnification x 450.)
In collaboration with M. Selig, G. Y. Lee, and B. Little (unpublished observations), we studied cultured granulosa cells from small antral follicles for 6 days in a defined serum-free medium with and without the presence of 10- 5 moUL DHEA or 10- 5 Il'androstenedione. 29 In this investigation we examined the cytoskeletal system, which consists of microfilaments, intermediate filaments, and microtubules. In this study particular attention was given to actin. The
presence of actin was assessed by use of phalloidin labeled with rhodamine. We found that equivalent images were obtained whether we used 10- 5 mol/L DHEA or 10- 5 moUL Il'-androstenedione. We report here images obtained when cells were cultured with and without the presence of 10- 5 moUL DHEA. Figs. 4, 5, and 6 are phase-contrast photomicrographs of 2, 4, and 6 days of cells grown in our basic serum-free medium without DHEA and serve as con-
Androgens on ovarian granulosa cells
Volume 160 Kumber 4
trois. As can be seen from these images, cells cultured for 2 days were rather round; however, by day 6 they were completely flattened and somewhat confluent. The day experimental medium was added to the cultures we designated as day O. Figs. 7,8, and 9 are phasecontrast images of cells grown for 2, 4, and 6 days in the presence of DHEA. When compared with control cultures, the cells for the most part were round; some were spindle shaped. They never achieved the flattened configuration like those in control cultures. As indicated above, we assessed the presence of F actin by use of phalloidin labeled with rhodamine. Control cultures show that F actin appears filamentous by day 2 in culture (Figs. 10 to 12). By days 4 and 6 there are not only actin filaments but rather large parallel bundles known as microfilaments and seen by Lewis and Lewis 30 in a number of different cells in vitro. Lewis and Lewis 30 referred to these bundles of micro filaments as "stress fibers." Within the recent past, much work has been done on stress fibers. They have been found to contain actin filaments, myosin, u-actininum, and other actin-binding proteins. The precise function(s) of these microfilament bundles is (are) unknown. 3 ! When one observes photomicrographs of cells stained with rhodamine phalloidin from DHEA-treated cultures, it is obvious that the number of filamentous bundles is few when compared with the number of controls; the cells also have altered shapes (Figs. 13 to 15). In studies by other investigators of cells exposed for a short time to certain biochemical agents such as hormones, there are rapid shape changes. 32 Folkman and Moscana 33 believe that shape changes in their system were indicative of growth control. Just what function, if any, is associated with the observed shape change of granulosa cells grown in the presence of DHEA cannot be ascertained from our study. Obviously, there is a deficiency of actin bundles and perhaps other filamentous actin. It is possible that this alteration or deficiency in actin could reflect some altered function of the cell. In this connection it should be pointed out that actin is involved in structural support, and it has been found that certain glycolytic enzymes are located on actin filaments. 34 3S Anderson et a1. 2 " reported that cells grown in the presence of DHEA are dead by day 4 as evidenced by many heterolysosomes and many large vacuoles. The apparent deficiency of actin filaments and the possibility of altered enzymatic activity coupled with other degenerating compartments may signal early atretic processes of granulosa cells in vitro. REFERENCES 1. Stein IF Sr. Leventhal MC. Amenorrhea association with bilateral polycystic ovaries. AM J OBSTET GYNECOL 1935; 29: 181. 2. Yen SSe. The polycystic ovary syndrome. Clin Endocrinol 1980; 12: 177. 3. Hemmings R, Farookhi R, Brawer JR. Pituitary and ovar-
4. 5.
6.
7. 8. 9. 10. 11.
12.
13. 14.
15. 16. 17. 18. 19.
20. 21. 22.
23.
24.
787
ian responses to LHRH in a rat with polycystic ovaries. Bioi Reprod 1983;29:239. Brawer JR, Munoz M, Farookhi R. Development of the polycystic ovarian condition in the estradiol valeratetreated rat. Bioi Reprod 1986;35:647. Roy S, Mahesh VB, Greenblatt RB. The effect of dehydroepiandrosterone and a"-androstenedione on the reproductive organs of female rats: production of cystic changes in the ovary. Nature 1962;196:42. Popkin RM. Fraser HM, Gosden RG. Effects of LHRH agonist of LHRH immunoneutralization on pituitary and ovarian LHRH receptors in female rats. J Reprod Fertil 1983;69:245. Lee MT, Bruot BC, Adams WC. Hormonal changes during early development of ovarian cysts in the rat. Bioi Reprod 1986;35:542. Singh KB. Persistent estrous: an experimental model of the polycystic ovary syndrome. Obstet Gynecol Soc 1969; 24:2. GreenJA, Goldzieher JW. The polycystic ovary. IV. Light and electron microscope studies. AM J OBSTET GYNECOL 1965;91:173. Rebar R, Judd HL, Yen SSC. Characterization of the inappropriate gonadotropin secretion in polycystic ovary syndrome. J Clin Invest 1976;57: 1320. Billiar RB, Richardson D, Anderson E, Mahajan D, Little B. The effect of chronic and acyclic elevation of circulating androstenedione or estrone concentrations on ovarian function in the rhesus monkey. Endocrinology 1985; 116:2209. Mahesh VB, Greenblatt RB. Isolation of dehydroepiandrosterone and 1713-hydroxy 60S-pregnenolone from the polycystic ovaries of Stein-Leventhal syndrome. J Clin Endocrinol 1962; II :441. Anderson E, Little B. The ontogeny of the rat granulosa cell. In: Toft DO, Ryan RJ, eds. Proceedings of the Fifth Ovarian Workshop. Champaign, Illinois: 1985:102. Albertini DF, Anderson E. The appearance and structure of intercellular connections during ontogeny of the rabbit ovarian follicle with particular reference to gap junctions. J Cell BioI 1974;63:234. Anderson E, Albertini DF. Gap junctions between the oocyte and the companion follicle cells in the mammalian ovary. J Cell Bioi 1976;71:680. Goldzieher JW. GreenJA. The polycystic ovary. 1. Clinical and histologic features. J Clin Endocrinol 1961;22:325. Mahesh VB, Greenblatt RB. Steroid secretions of the normal and polycystic ovary. Recent Prog Horm Res 1964; 20:341. Mahesh VB, Greenblatt RB. Aydar CK, Roy S. Secretions of androgen by the polycystic ovary and its significance. Ferti! SteriI1962;13:513. Roy S, Mahesh VB, Greenblatt RB. The effect of dehydroepiandrosterone and a'-androstenedione on the reproductive organs of female rats: production of cystic changes in the ovary. Nature 1962;196:42. Knudsen JF, Costoff A, Mahesh VB. Dehydroepiandrosterone-induced polycystic ovaries and acyclicity in the rat. Ferti! Steril 1975;26:807. Knudsen JF, Mahesh VB. Initiation of precocious sexual maturation in the immature rat treated with dehydroepiandrosterone. Endocrinology 1975;97:458. Ward CW, Costoff A, Mahesh VB. The induction of polycystic ovaries in mature cycling rats by the administration of dehydroepiandrosterone (DHA). BioI Reprod 1978;18:614. Christensen K, Gillim SW. The correlation of fine structure and function in steroid-secreting cells, with emphasis on those of the gonads. In: McKerns KW, ed. The gonads. New York: Appleton-Century-Crofts, 1969:415. Taylor AE, Granger DN. Exchange of macro-molecules across the microcirculation. In: Handbook of physiology. The cardiovascular system. Circulation. Bethesda, Maryland: American Physiology Society, 1984 vol 2:468.
788
Anderson
25. Albertini DF, Anderson E. Microtubule and microfilament rearrangement during capping of concanavalin A receptors on cultured ovarian granulosa cells. J Cell Bioi 1977;73:111. 26. Wilkinson RF Jr, Byskov AG. Anderson E. In vivo identification and structure of atretic murine ovarian follicles. In: Midgley AR. Sadler WA, eds. Ovarian follicular development and function. New York: Raven Press, 1979: 65. 27. Erickson GF, Magoffin DA, Dyer CA, Hofeditz C. The ovarian androgen producing cells: a review of structure/function relationship. Endocrinol Res 1985;6:371. 28. Anderson E. Little B, Lee GS. Androgen induced changes in rat ovarian granulosa cells in vitro. Tissue Cell 1987; 19:217. 29. Anderson E, Selig M, Lee GY. Androgen induced cellular changes in rat ovarian granulosa cells grown in a serumfree medium. J Cell Bioi 1986;103:263a.
April 1989 Am J Obstet Gynecol
30. Lewis WH, Lewis MR. Behavior of cells in tissue culture. In: Cowdry EV, ed. General cytology. Chicago: The University of Chicago Press, 1924:385. 31. Schliwa M. The cytoskeleton. Cell Bioi Monogr 1986; 13:47. 32. Lawrence TS. Ginzberg RD, Gilula NB, Beers WHo Hormonally induced cell shape changes in cultured rat ovarian granulosa cells. J Cell Bioi 1979;80:21. 33. Folkman J, Moscana A. Role of cell shape in growth control. Nature 1978;273:345. 34. Masters Cj. Interactions between glycolytic enzymes and components of the cytomatrix. J Cell Bioi 1984;99(1 Pt 2):222s. 35. Walsh TP, Masters CJ, Morton DJ, Clarke FM. Binding of aldolase to actin containing filaments. Biochem J 1981;186:89.
Key words: Androgen, granulosa cell, atresia, cyst formation Since Stein and Leventhal' discovered polycystic ovarian disease in women, numerous studies have been made in an effort to clarify this multidimensional malady! In certain mammalian models, cystic ovaries have been produced by numerous treatments. These treatments include estradiol valerate," 4 dehydroepiandrosterone (DHEA)5 treatment with antibodies to luteinizing hormone-releasing hormones," and treatment of hypothyroid rats with human chorionic gonadotropin. 7 Singh described other methods for producing polycystic ovarian disease. s In addition to producing cysts, the aforementioned regimens also induce follicular atresia. With the production of ovarian cysts with many different hormonal manipulations came the excitement of having a model to analyze certain aspects of the etiologic factors of the disease. Some authors" have described the unusual histologic and cytologic features. Ovaries obtained from patients with polycystic ovarian disease have, for the most part, shown increased thickness of the collagen compromising the tunica albugenia, multiple atretic follicles, follicular cysts, and a hyperplasia of cells of the theca interna. Other investigators lO have concerned themselves with the hormonal profiles of the disease. In the
From the Department of Anatomy and Cellular Bzology and the Laboratory of Human Reproduction and Reproductive Biology, Harvard MedIcal School. ThIS tnvestigation was supported by National Institutes of Health Grants HD06645 (Center Grant of Laboratory of Human Reproduction and ReproductIve Biology) and HD14574. Presented as the Joseph Pnce Oratzon at the Seventh Annual Meettng of the Amencan Gynecological and Obstetrical Society, Napa, California, September 8-10, 1988. Reprint requests: Dr. Everett Anderson, Harvard Medical School, Department of Anatomy and Cellular Biology, LHRRB. 45 Shattuck St., Boston, MA 02115.
782
hormonal studies it has been found that there is an abnormal gonadotropin secretion with an elevated acyclic circulating concentration of androgens and/or estrogens. In 1985, Billiar et aI., II using a primate, the rhesus monkey (Macacca mulatta), were able to elevate levels of androstenedione or estrone. Under these conditions, polycystic changes (structural) in the ovary were also observed. In women with polycystic ovarian disease, Mahesh and Greenblatt l2 found that de hydroepiandrosterone and androstenedione and / or estrone are usually elevated, whereas testosterone or estradiol may not be. In this communication attention is called to some observations of the in vivo and in vitro effects of dehydroepiandrosterone on rat granulosa cells. * Results and comment
In vivo studies. The general histologic features of the rat ovary are well known l3 and need not be detailed here. Suffice it to say that in a section of the mature ovary there are nonantral and antral follicles in different stages of development and in different stages of atresia. Initially, the oocytes of nonantral follicles are surrounded by a squamous layer of granulosa cells that rest on a basement membrane. Once the antrum is formed, the formation of which is due to the action of gonadotropins, the inner granulosa cells are rather polyhedral in shape and are associated with the oocyte, whereas the outer ones are rather columnar and rest *Animals used in this study were maintained in accordance with the guidelines of the Committee on Animals of the Harvard Medical School and those prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Resources, National Research Council (DHHS publication no. (NIH) 85-23, revised, 1985).
Androgens on ovanan granulosa cells
Volullle HiO 1\ lImher 4-
783
Figs. 1 and 2. Electron nllcrographs of portions of epithelial cells lining ova nan cyst. BM, Basement membrane; anuw, tight junction; GJ. gap junction; C, cy,t lumen; Sf, smooth surface 1l1vagination (2). (Original magnificatIon x SO,()OO )
on a basement membrane. Granulosa cells are coupled to each other and to the oocyte via gap (communicating) junctions. 11 ", Another feature of the antral follicle is the presence of an inner vascularized steroid-secreting theca that is situated between the basement membrane and the theca externa, which is non vascularized and is composed of fusiform-shaped cells. Each ovary is covered with an ovarian epithelium that is composed of cuboidal-columnar cells and rests on a basement membrane. This epithelium is continuous with the mesothelium. Beneath the ovarian epithelium are some collagen fibers and some stromal cells. In humans," 'h and monkeys (unpublished observations), the collagen fibers are organized into a definitive layer known as the tunica albugenia. As indicated previously, it is well known that patients with polycystic ovarian syndrome produce an excessive amount of androgens.' Mahesh and Greenblatt" believe that androgens appear to be a contributing component to the syndrome." It has been shown that relatively large amounts of DHEA and ~' androstenedione can be isolated from patients with polycystic ovaries." In 1962 Roy et al.'" injected DHEA and ~' androstenedione into immature female rats over a 21day period and observed precocious ovulation, with cystic and atretic follicles in the ovarv; the rats became acyclic.""·"' After the injection, the serum levels of follicle-stimulating hormone and luteinizing hormone
showed an ovulatory surge. Subsequently, folliclestimulating hormone returned to control levels whereas luteinizing hormone was extremely low. When the treatment was withdrawn, the animals became cyclic over time. Equivalent changes can be induced in mature cycling rats." It is interesting that Billiar et al." used ~ '-androstenedione to induce follicle cysts and atresia in the rhesus monkey. In collaboration with Drs. Marcus Walker and MongTing Lee (unpublished observations), my colleagues and I used an equivalent protocol to induce, among other things, cyst development and atresia. We examined ovaries obtained 10, 15, and 20 days after injecting DHEA. Ovaries removed on day 20 were fixed in Bouin's solution and sectioned for light microscopy; corpora lutea, and follicular cysts like those demonstrated by Knudsen et al.'o were seen. Electron microscopic examination of granulosa cells of some antral follicles showed an increase in smooth endoplasmic reticulum, and the mitochondria had acquired tubular cristae, The aforementioned organelles are characteristics of steroid-secreting cells. 23 The granulosa cells also had blebs, and the blebs were pinched off and removed by macrophages. We believe that the formation of an ovarian cyst (l to 2 mm in diameter) is complete when the granulosa cell ba~al layer becomes (transformed) epithelial (Figs. 1 and 2). In addition to maintaining association with each other via gapjunctions, the "transformed" granulosa cells acquire tight junctions, (Tight
784
Anderson
April 1989 Am .I ObSlel Gynecol
Fig. 3. Electron micrograph of portion of granulosa cell from atretic follicle (monkey). BM, Basement membrane; AG, annular gap junctions. (Original magnification x 60,000.)
junctions are not a feature of granulosa cells.) These epithelial cells rest on an intact basement membrane and acquire many smooth surface invaginations and some coated ones. The non coated invaginations are believed to continue to invaginate and eventually produce a fluid-filled vesicle containing macromolecules from the interstitium and move them into the cyst by transcytosis, a process frequently encountered in the microcirculatory system!' The coated invaginations are believed to remove, from the cell's surface, special molecules for specific cellular functions via receptormediated endocytosis."' During cyst formation the theca interna and theca externa appear to be lost. How this is accomplished remains unexplained. Atretic follicle. As has been described many times, during the regular ovarian cycle numerous follicles un-
dergo atresia!6 We have also observed many atretic follicles when DHEA was administered to immature rats for 20 days. This degenerative condition appears to favor not only the antral follicles but also nonantral follicles. It is unknown which compartment of the follicle degenerates first, the oocyte or the granulosa cells. In any event some of the manifestations of atresia in granulosa cells are pyknotic nuclei and usually a fragmentation of the cytoplasm. We have not seen macrophages within atretic follicles with fragments of degenerating granulosa cells in their cytoplasm. The basement membrane of the atretic follicle becomes detached from the basal layer of granulosa cells and frequently becomes folded. We also saw this phenomenon in atretic follicles of the monkey after treatment with androstenedione (Fig. 3). We have not seen unusual changes of the thecal components. The fate of the theca
Androgens on ovarian granulosa cells
Volume 160 l\umber 4
785
Figs. 4 to 6. Phase-contrast photomicrographs of control cultures. 4, 2 days; 5, 4 days; 6, 6 days. (Original magnification x 450.)
Figs. 7 to 9. Phase-contrast photomicrographs of cells grown in the presence of 10- 5 mollL DHEA. 7, 2 days; 8, 4 days; 9, 6 days. (Original magnification x 450.)
externa is unknown; however, the thecal interna is believed to persist as islands of interstitial tissue.27 In vitro studies. Granulosa cells can be easily expressed from small antral follicles suitable for culturing. We cultured granulosa cells in McCoy's medium with and without the presence of lO-' mollL DHEA or lO-' moUL LlI-androstenedione."8 The reason for using the hormonal concentration was the known circulating and tissue concentrations of the steroids."" Anderson et al."" used this concentration of DHEA or Ll4_
androstenedione and found induced mitochondrial changes (from lamellae cristae to tubular cristae) and the acquisition of smooth endoplasmic reticulum. These organelle changes are similar to those observed in experiments in situ, as pointed out above. These cells showed degenerative changes after 4 days in culture, whereas control cells survived beyond 6 days. We concluded that the aforementioned study supported our hypothesis that high levels of androgens can induce atresia in vitro.
786
Anderson
Apnl1989 Am J Obstet Gvnecol
Figs. 10 to 12. Photomicrographs of cells from control cultures stained with rhodamine phalloidin. 10, 2 days; 11, 4 days; 12, 6 days. Stress fibers are conspicuous longitudinal fluorescent structures. (Original magnification x 450.)
Figs. 13 to 15. Photomicrographs of cells grown in IO-J mollL dehydroepiandrosterone and stained with rhodamine phalloidin. 13, 2 days; 14, 4 days; 15, 6 days. Very few stress fibers are found in cells grown in DHEA. (Original magnification x 450.)
In collaboration with M. Selig, G. Y. Lee, and B. Little (unpublished observations), we studied cultured granulosa cells from small antral follicles for 6 days in a defined serum-free medium with and without the presence of 10- 5 moUL DHEA or 10- 5 Il'androstenedione. 29 In this investigation we examined the cytoskeletal system, which consists of microfilaments, intermediate filaments, and microtubules. In this study particular attention was given to actin. The
presence of actin was assessed by use of phalloidin labeled with rhodamine. We found that equivalent images were obtained whether we used 10- 5 mol/L DHEA or 10- 5 moUL Il'-androstenedione. We report here images obtained when cells were cultured with and without the presence of 10- 5 moUL DHEA. Figs. 4, 5, and 6 are phase-contrast photomicrographs of 2, 4, and 6 days of cells grown in our basic serum-free medium without DHEA and serve as con-
Androgens on ovarian granulosa cells
Volume 160 Kumber 4
trois. As can be seen from these images, cells cultured for 2 days were rather round; however, by day 6 they were completely flattened and somewhat confluent. The day experimental medium was added to the cultures we designated as day O. Figs. 7,8, and 9 are phasecontrast images of cells grown for 2, 4, and 6 days in the presence of DHEA. When compared with control cultures, the cells for the most part were round; some were spindle shaped. They never achieved the flattened configuration like those in control cultures. As indicated above, we assessed the presence of F actin by use of phalloidin labeled with rhodamine. Control cultures show that F actin appears filamentous by day 2 in culture (Figs. 10 to 12). By days 4 and 6 there are not only actin filaments but rather large parallel bundles known as microfilaments and seen by Lewis and Lewis 30 in a number of different cells in vitro. Lewis and Lewis 30 referred to these bundles of micro filaments as "stress fibers." Within the recent past, much work has been done on stress fibers. They have been found to contain actin filaments, myosin, u-actininum, and other actin-binding proteins. The precise function(s) of these microfilament bundles is (are) unknown. 3 ! When one observes photomicrographs of cells stained with rhodamine phalloidin from DHEA-treated cultures, it is obvious that the number of filamentous bundles is few when compared with the number of controls; the cells also have altered shapes (Figs. 13 to 15). In studies by other investigators of cells exposed for a short time to certain biochemical agents such as hormones, there are rapid shape changes. 32 Folkman and Moscana 33 believe that shape changes in their system were indicative of growth control. Just what function, if any, is associated with the observed shape change of granulosa cells grown in the presence of DHEA cannot be ascertained from our study. Obviously, there is a deficiency of actin bundles and perhaps other filamentous actin. It is possible that this alteration or deficiency in actin could reflect some altered function of the cell. In this connection it should be pointed out that actin is involved in structural support, and it has been found that certain glycolytic enzymes are located on actin filaments. 34 3S Anderson et a1. 2 " reported that cells grown in the presence of DHEA are dead by day 4 as evidenced by many heterolysosomes and many large vacuoles. The apparent deficiency of actin filaments and the possibility of altered enzymatic activity coupled with other degenerating compartments may signal early atretic processes of granulosa cells in vitro. REFERENCES 1. Stein IF Sr. Leventhal MC. Amenorrhea association with bilateral polycystic ovaries. AM J OBSTET GYNECOL 1935; 29: 181. 2. Yen SSe. The polycystic ovary syndrome. Clin Endocrinol 1980; 12: 177. 3. Hemmings R, Farookhi R, Brawer JR. Pituitary and ovar-
4. 5.
6.
7. 8. 9. 10. 11.
12.
13. 14.
15. 16. 17. 18. 19.
20. 21. 22.
23.
24.
787
ian responses to LHRH in a rat with polycystic ovaries. Bioi Reprod 1983;29:239. Brawer JR, Munoz M, Farookhi R. Development of the polycystic ovarian condition in the estradiol valeratetreated rat. Bioi Reprod 1986;35:647. Roy S, Mahesh VB, Greenblatt RB. The effect of dehydroepiandrosterone and a"-androstenedione on the reproductive organs of female rats: production of cystic changes in the ovary. Nature 1962;196:42. Popkin RM. Fraser HM, Gosden RG. Effects of LHRH agonist of LHRH immunoneutralization on pituitary and ovarian LHRH receptors in female rats. J Reprod Fertil 1983;69:245. Lee MT, Bruot BC, Adams WC. Hormonal changes during early development of ovarian cysts in the rat. Bioi Reprod 1986;35:542. Singh KB. Persistent estrous: an experimental model of the polycystic ovary syndrome. Obstet Gynecol Soc 1969; 24:2. GreenJA, Goldzieher JW. The polycystic ovary. IV. Light and electron microscope studies. AM J OBSTET GYNECOL 1965;91:173. Rebar R, Judd HL, Yen SSC. Characterization of the inappropriate gonadotropin secretion in polycystic ovary syndrome. J Clin Invest 1976;57: 1320. Billiar RB, Richardson D, Anderson E, Mahajan D, Little B. The effect of chronic and acyclic elevation of circulating androstenedione or estrone concentrations on ovarian function in the rhesus monkey. Endocrinology 1985; 116:2209. Mahesh VB, Greenblatt RB. Isolation of dehydroepiandrosterone and 1713-hydroxy 60S-pregnenolone from the polycystic ovaries of Stein-Leventhal syndrome. J Clin Endocrinol 1962; II :441. Anderson E, Little B. The ontogeny of the rat granulosa cell. In: Toft DO, Ryan RJ, eds. Proceedings of the Fifth Ovarian Workshop. Champaign, Illinois: 1985:102. Albertini DF, Anderson E. The appearance and structure of intercellular connections during ontogeny of the rabbit ovarian follicle with particular reference to gap junctions. J Cell BioI 1974;63:234. Anderson E, Albertini DF. Gap junctions between the oocyte and the companion follicle cells in the mammalian ovary. J Cell Bioi 1976;71:680. Goldzieher JW. GreenJA. The polycystic ovary. 1. Clinical and histologic features. J Clin Endocrinol 1961;22:325. Mahesh VB, Greenblatt RB. Steroid secretions of the normal and polycystic ovary. Recent Prog Horm Res 1964; 20:341. Mahesh VB, Greenblatt RB. Aydar CK, Roy S. Secretions of androgen by the polycystic ovary and its significance. Ferti! SteriI1962;13:513. Roy S, Mahesh VB, Greenblatt RB. The effect of dehydroepiandrosterone and a'-androstenedione on the reproductive organs of female rats: production of cystic changes in the ovary. Nature 1962;196:42. Knudsen JF, Costoff A, Mahesh VB. Dehydroepiandrosterone-induced polycystic ovaries and acyclicity in the rat. Ferti! Steril 1975;26:807. Knudsen JF, Mahesh VB. Initiation of precocious sexual maturation in the immature rat treated with dehydroepiandrosterone. Endocrinology 1975;97:458. Ward CW, Costoff A, Mahesh VB. The induction of polycystic ovaries in mature cycling rats by the administration of dehydroepiandrosterone (DHA). BioI Reprod 1978;18:614. Christensen K, Gillim SW. The correlation of fine structure and function in steroid-secreting cells, with emphasis on those of the gonads. In: McKerns KW, ed. The gonads. New York: Appleton-Century-Crofts, 1969:415. Taylor AE, Granger DN. Exchange of macro-molecules across the microcirculation. In: Handbook of physiology. The cardiovascular system. Circulation. Bethesda, Maryland: American Physiology Society, 1984 vol 2:468.
788
Anderson
25. Albertini DF, Anderson E. Microtubule and microfilament rearrangement during capping of concanavalin A receptors on cultured ovarian granulosa cells. J Cell Bioi 1977;73:111. 26. Wilkinson RF Jr, Byskov AG. Anderson E. In vivo identification and structure of atretic murine ovarian follicles. In: Midgley AR. Sadler WA, eds. Ovarian follicular development and function. New York: Raven Press, 1979: 65. 27. Erickson GF, Magoffin DA, Dyer CA, Hofeditz C. The ovarian androgen producing cells: a review of structure/function relationship. Endocrinol Res 1985;6:371. 28. Anderson E. Little B, Lee GS. Androgen induced changes in rat ovarian granulosa cells in vitro. Tissue Cell 1987; 19:217. 29. Anderson E, Selig M, Lee GY. Androgen induced cellular changes in rat ovarian granulosa cells grown in a serumfree medium. J Cell Bioi 1986;103:263a.
April 1989 Am J Obstet Gynecol
30. Lewis WH, Lewis MR. Behavior of cells in tissue culture. In: Cowdry EV, ed. General cytology. Chicago: The University of Chicago Press, 1924:385. 31. Schliwa M. The cytoskeleton. Cell Bioi Monogr 1986; 13:47. 32. Lawrence TS. Ginzberg RD, Gilula NB, Beers WHo Hormonally induced cell shape changes in cultured rat ovarian granulosa cells. J Cell Bioi 1979;80:21. 33. Folkman J, Moscana A. Role of cell shape in growth control. Nature 1978;273:345. 34. Masters Cj. Interactions between glycolytic enzymes and components of the cytomatrix. J Cell Bioi 1984;99(1 Pt 2):222s. 35. Walsh TP, Masters CJ, Morton DJ, Clarke FM. Binding of aldolase to actin containing filaments. Biochem J 1981;186:89.