0021-972X/92/7506-1556$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society
Relative Hormone F. MIRO
AND
Vol. 75, No. 6 Printed in U.S.A.
Effects of Activin and Inhibin on Steroid Synthesis in Primate Granulosa Cells* S. G. HILLIER
Reproductive Endocrinology Laboratory, Department of Obstetrics and Gynaecology, University Edinburgh, Centre for Reproductive Biology, Chalmers Street, Edinburgh EH3 9E W, Scotland
of
ABSTRACT Ovarian granulosa cells produce inhibin and activin, structurally related proteins with potentials to directly modulate follicular steroidogenesis. The aim of the present study was to compare developmentrelated effects of inhibin-A and activin-A on steroidogenesis in marmoset monkey (Callithrixjacchus) granulosa cells. Granulosa cells from “immature” (2 mm diameter) follicles were incubated in serum-free culture medium for 96 h with and without peptide (l-100 ng/mL), in the presence and absence of gonadotropins [human (h) FSH or hLH] (10 ng/mL). Spent medium was collected and stored frozen for progesterone assay. Aromatase activity was determined by incubating cells for a further 6 h in the presence of 1 rmol testosterone and assaying accumulation of oestradiol. Granulosa cells from immature follicles showed characteristically low basal rates of steroid synthesis that were unaffected by treatment alone with either inhibin or activin. Treatment with hFSH stimulated both progesterone production and aromatase activity. Cotreatment
with activin and hFSH further enhanced aromatase activity by up to 4-fold. The progesterone response to activin plus hFSH was related to the effect of hFSH in the absence of activin: high-level responsiveness to hFSH was suppressed by activin while low-level responsiveness was enhanced. Inhibin had no significant effect on FSH-responsive progesterone production, but at high concentrations (>lO ng/mL) it-caused slight CUD to 30%) reduction in FSH-induced aromatase activity. Granu&a &ils from mature follicles showed relatively high basal-rates of steroidogenesis, and treatment with inhibin did not influence either basal or gonadotropin responsive steroidogenesis. Treatment with activin had divergent effects on aromatase activity and progesterone synthesis in that it increased both basal and hLH-responsive aromatase activity (up to ll-fold), had no effect on basal progesterone production, and markedly suppressed (by more than 50%) the progesterone response to hLH. These data reveal development-dependent effects of inhibin and activin on granulosa cell steroidogenesis that are likely to have physiological relevance to ovarian function in uiuo. (J Clin Endocrinol Metab 75: 1556-1561, 1992)
GRANULOSA cells express activin and inhibin subunit messenger RNA (mRNA)s that encode proteins implicated in the paracrine control of follicular development (1, 2) as well as the regulation of FSH release by the pituitary gland (3, 4) (for review see Refs. 5, 6). High-affinity binding sites, thought to be receptors, for activin and inhibin are present on rat granulosa cells (7, 8), and these cells display marked steroidogenic responses to activin in vitro. In cultured rat granulosa cells, responses to activin alter in relation to follicular maturity, with aromatase activity becoming enhanced and progesterone production suppressed as preovulatory development proceeds (9). Effects of inhibin on granulosa cell function are equivocal, since inhibition (10) and no effect (11) of inhibin on steroid synthesis have been reported by the only relevant studies to date. It is obviously important to know how locally produced activin and inhibin influence granulosa function in women. In vitro studies of immunoreactive inhibin production by human granulosa cells have shown that inhibin production by immature granulosa cells is increased by treatment with FSH (12), and that inhibin production increases with preovulatory development (13), paralleling aromatase activity. Mature granulosa cells from preovulatory follicles, therefore,
secrete relatively large amounts of inhibin and estradiol, both being stimulated by LH (12). There is no such information on the regulation of activin production by human granulosa cells, but in macaque ovaries the granulosa cells in immature antral follicles express /3-inhibinlactivin subunit mRNA more strongly than oc-inhibin subunit mRNA (14), suggesting that activin production predominates over inhibin during early stages of follicular development in primates. Recently, it was shown that activin inhibits, whereas inhibin stimulates, androgen synthesis by cultured human thecal cells (15, 16), confirming earlier results from work on cultured rat thecal cells (17). A similar comparison of the effects of activin and inhibin on human granulosa cells has not yet been undertaken, due to limited availability of suitable experimental tissue. As an alternative, we have studied cultured granulosa cells from another nonhuman primate, the common marmoset monkey (Callithrixjacchus). The marmoset ovarian cycle is similar in length to the human cycle, and marmoset and human granulosa cells are comparable in terms of gonadotropin-responsive production of inhibin and steroids in vitro (18). Here, we report the use of this in vitro model to assess relative contributions of activin and inhibin to the paracrine regulation of granulosa cell steroidogenesis in the primary ovary.
Received March 10, 1992. Address all correspondence and requests for reprints to: Dr. F. Mirb, Reproductive Endocrinology Laboratory, Department of Obstetrics and Gynaecology, University of Edinburgh Centre for Reproductive Biology, Chalmers Street, Edinburgh EH3 9EW, Scotland. * Supported by the UK Medical Research Council (Programme Grant 8929853) and the Spanish Ministry of Education and Science.
Subjects
and Methods
Animals Adult female marmosets Reproductive Biology Unit
were from the breeding colony at the MRC Primate Centre in Edinburgh. Late-follicular
1556
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 19:03 For personal use only. No other uses without permission. . All rights reserved.
ACTIVIN
AND
INHIBIN
ACTION
phase tissue was obtained from 12 cyclic animals by inducing luteal regression with a single im injection of 0.5 pg synthetic prostaglandin F2o( analog (“Estrumate”; ICI, Macclesfield, Cheshire, England) given on approximately day 14 of the previous luteal phase (assessed by daily serum progesterone assay). Oophorectomy was carried out 8 or 9 days later (19).
Oophorectomy
and tissue isolation
Both ovaries were removed under general anesthesia and transported to the laboratory in ice-cold culture medium [medium 199 containing Earle’s salts, 25 mmol N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid buffer/L, 50 IU penicillin/ml, 50 PLg streptomycin/ml, 0.1% (wt/ vol) BSA, and extra (2 mmol/L) L-glutamine] (GIBCO, Ltd, Paisley, Strathclyde, U.K.). All follicles greater than or equal to 0.5 mm diameter were dissected free from the ovarian stroma using a zoom-stereomicroscope fitted with an ocular micrometer for visualization. Follicles were classified according to diameter as immature (51.0 mm), or mature (22.0 mm), as described previously (19). Medium-sized follicles (1.0-2.0 mm) were not studied since they are generally atretic at this stage of the ovarian cycle. Granulosa cells from each follicle size category were harvested and pooled as suspensions in culture medium. Large follicles (2-3/animal) were presumed to be preovulatory, with every one yielding an apparently healthy cumulus-enclosed oocyte. No attempt was made to distinguish atretic and nonatretic small follicles. Granulosa cell viability, checked by trypan blue dye exclusion, was greater than or equal to 50%.
ON PRIMATE
GRANULOSA
CELLS Results
Activin on aroma&se follicles
activity in granulosa
During culture for 96 h in the presence of testosterone, estradiol production by granulosa cells from immature follicles increased dose dependently in the presence of FSH, becoming maximal at greater than or equal to 10 ng FSH/ mL (EDs0 - 3 ng FSH/mL). Activin alone, at all doses tested, did not significantly affect basal production of estradiol. However, activin concentrations greater than or equal to 10 ng/mL markedly augmented the action of FSH, and this effect was strongest in the presence of low-dose FSH (1 ng/ mL) (Fig. 1A). The maximum rate of estradiol production in response to FSH (10 ng/mL) plus activin (10 ng/mL) reflected almost quantitative metabolism of testosterone to estradiol; therefore, measurement of estradiol after a 6-h incubation with testosterone at the end of the 96-h culture was used to accurately determine aromatase activity. This approach demonstrated that the level of aromatase activity induced by a maximal stimulatory dose of FSH (10 ng/mL) was increased greater than 4-fold by activin (EDs0 l-10 ng/mL). Activin on progesterone immature follicles
production
FSH dose dependently stimulated by granulosa cells from immature
Tissue culture
cells from immature
by granulosa
cells from
progesterone production follicles, and this effect
Multiwell-plastic dishes (“Linbro” from Flow Laboratories, Rickmansworth, Hertfordshire, U.K.) were precoated with donor calf serum (GIBCO) and washed with Dulbecco’s phosphate-buffered saline before use (20). Culture wells were inoculated with replicate 250-PL portions of a cell suspension containing approximately 2.0 X lo4 live granulosa cells. Gonadotropin and/or peptide was previously added in culture medium to give a final incubation volume of 500 pL. The gonadotropins were human (h)FSH (LER-8/116) or hLH (LER-1972), donated by Dr. L. E. Reichert Jr. Recombinant human activin-A and recombinant human inhibin-A were provided by Genentech Inc. (San Francisco, CA). All treatments were in triplicate. The cultures were incubated at 37 C in a humidified tissue culture incubator gassed with a 95% sir/5% CO* mixture. Incubation was for 96 h with a medium change at 48 h. In some experiments, testosterone (1.0 Fmol) (Sigma Chemical Co., Poole, Dorset, U.K.) was included in the culture medium as an aromatase substrate. Alternatively, aromatase activity was measured at the end of the culture, when the monolayers were washed once with Dulbecco’s phosphate-buffered saline and incubated for a further 6 h in culture medium containing 1.0 pmol testosterone. Medium was stored frozen at -20 C until analysis of progesterone and/or estradiol content, as described below.
Steroid
RIA
Estradiol and progesterone in culture medium RIAs validated for these purposes (21, 22). The precision for each steroid was less than 15%.
Data presentation
were inter-
determined by and intraassay
and analysis
Steroid levels in culture medium are expressed as pmol steroid/1000 cells, related to the number of viable cells used to inoculate each culture well. The data are measurements of steroid accumulation from the 49. to 96-h period of culture, or during the subsequent 6 h incubation to determine aromatase activity. Results were analyzed by analysis of variance for multiple comparisons followed by Neuman-Keul’s test. Student’s t test was used for paired comparisons. Differences with P less than or equal to 0.05 were accepted as statistically significant.
0
1
10 FSH
100
(rig/ml)
FIG. 1. Dose-dependent effect of hFSH and activin on marmoset granulosa cell steroidogenesis from immature (
Relative Hormone F. MIRO
AND
Vol. 75, No. 6 Printed in U.S.A.
Effects of Activin and Inhibin on Steroid Synthesis in Primate Granulosa Cells* S. G. HILLIER
Reproductive Endocrinology Laboratory, Department of Obstetrics and Gynaecology, University Edinburgh, Centre for Reproductive Biology, Chalmers Street, Edinburgh EH3 9E W, Scotland
of
ABSTRACT Ovarian granulosa cells produce inhibin and activin, structurally related proteins with potentials to directly modulate follicular steroidogenesis. The aim of the present study was to compare developmentrelated effects of inhibin-A and activin-A on steroidogenesis in marmoset monkey (Callithrixjacchus) granulosa cells. Granulosa cells from “immature” (2 mm diameter) follicles were incubated in serum-free culture medium for 96 h with and without peptide (l-100 ng/mL), in the presence and absence of gonadotropins [human (h) FSH or hLH] (10 ng/mL). Spent medium was collected and stored frozen for progesterone assay. Aromatase activity was determined by incubating cells for a further 6 h in the presence of 1 rmol testosterone and assaying accumulation of oestradiol. Granulosa cells from immature follicles showed characteristically low basal rates of steroid synthesis that were unaffected by treatment alone with either inhibin or activin. Treatment with hFSH stimulated both progesterone production and aromatase activity. Cotreatment
with activin and hFSH further enhanced aromatase activity by up to 4-fold. The progesterone response to activin plus hFSH was related to the effect of hFSH in the absence of activin: high-level responsiveness to hFSH was suppressed by activin while low-level responsiveness was enhanced. Inhibin had no significant effect on FSH-responsive progesterone production, but at high concentrations (>lO ng/mL) it-caused slight CUD to 30%) reduction in FSH-induced aromatase activity. Granu&a &ils from mature follicles showed relatively high basal-rates of steroidogenesis, and treatment with inhibin did not influence either basal or gonadotropin responsive steroidogenesis. Treatment with activin had divergent effects on aromatase activity and progesterone synthesis in that it increased both basal and hLH-responsive aromatase activity (up to ll-fold), had no effect on basal progesterone production, and markedly suppressed (by more than 50%) the progesterone response to hLH. These data reveal development-dependent effects of inhibin and activin on granulosa cell steroidogenesis that are likely to have physiological relevance to ovarian function in uiuo. (J Clin Endocrinol Metab 75: 1556-1561, 1992)
GRANULOSA cells express activin and inhibin subunit messenger RNA (mRNA)s that encode proteins implicated in the paracrine control of follicular development (1, 2) as well as the regulation of FSH release by the pituitary gland (3, 4) (for review see Refs. 5, 6). High-affinity binding sites, thought to be receptors, for activin and inhibin are present on rat granulosa cells (7, 8), and these cells display marked steroidogenic responses to activin in vitro. In cultured rat granulosa cells, responses to activin alter in relation to follicular maturity, with aromatase activity becoming enhanced and progesterone production suppressed as preovulatory development proceeds (9). Effects of inhibin on granulosa cell function are equivocal, since inhibition (10) and no effect (11) of inhibin on steroid synthesis have been reported by the only relevant studies to date. It is obviously important to know how locally produced activin and inhibin influence granulosa function in women. In vitro studies of immunoreactive inhibin production by human granulosa cells have shown that inhibin production by immature granulosa cells is increased by treatment with FSH (12), and that inhibin production increases with preovulatory development (13), paralleling aromatase activity. Mature granulosa cells from preovulatory follicles, therefore,
secrete relatively large amounts of inhibin and estradiol, both being stimulated by LH (12). There is no such information on the regulation of activin production by human granulosa cells, but in macaque ovaries the granulosa cells in immature antral follicles express /3-inhibinlactivin subunit mRNA more strongly than oc-inhibin subunit mRNA (14), suggesting that activin production predominates over inhibin during early stages of follicular development in primates. Recently, it was shown that activin inhibits, whereas inhibin stimulates, androgen synthesis by cultured human thecal cells (15, 16), confirming earlier results from work on cultured rat thecal cells (17). A similar comparison of the effects of activin and inhibin on human granulosa cells has not yet been undertaken, due to limited availability of suitable experimental tissue. As an alternative, we have studied cultured granulosa cells from another nonhuman primate, the common marmoset monkey (Callithrixjacchus). The marmoset ovarian cycle is similar in length to the human cycle, and marmoset and human granulosa cells are comparable in terms of gonadotropin-responsive production of inhibin and steroids in vitro (18). Here, we report the use of this in vitro model to assess relative contributions of activin and inhibin to the paracrine regulation of granulosa cell steroidogenesis in the primary ovary.
Received March 10, 1992. Address all correspondence and requests for reprints to: Dr. F. Mirb, Reproductive Endocrinology Laboratory, Department of Obstetrics and Gynaecology, University of Edinburgh Centre for Reproductive Biology, Chalmers Street, Edinburgh EH3 9EW, Scotland. * Supported by the UK Medical Research Council (Programme Grant 8929853) and the Spanish Ministry of Education and Science.
Subjects
and Methods
Animals Adult female marmosets Reproductive Biology Unit
were from the breeding colony at the MRC Primate Centre in Edinburgh. Late-follicular
1556
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 19:03 For personal use only. No other uses without permission. . All rights reserved.
ACTIVIN
AND
INHIBIN
ACTION
phase tissue was obtained from 12 cyclic animals by inducing luteal regression with a single im injection of 0.5 pg synthetic prostaglandin F2o( analog (“Estrumate”; ICI, Macclesfield, Cheshire, England) given on approximately day 14 of the previous luteal phase (assessed by daily serum progesterone assay). Oophorectomy was carried out 8 or 9 days later (19).
Oophorectomy
and tissue isolation
Both ovaries were removed under general anesthesia and transported to the laboratory in ice-cold culture medium [medium 199 containing Earle’s salts, 25 mmol N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid buffer/L, 50 IU penicillin/ml, 50 PLg streptomycin/ml, 0.1% (wt/ vol) BSA, and extra (2 mmol/L) L-glutamine] (GIBCO, Ltd, Paisley, Strathclyde, U.K.). All follicles greater than or equal to 0.5 mm diameter were dissected free from the ovarian stroma using a zoom-stereomicroscope fitted with an ocular micrometer for visualization. Follicles were classified according to diameter as immature (51.0 mm), or mature (22.0 mm), as described previously (19). Medium-sized follicles (1.0-2.0 mm) were not studied since they are generally atretic at this stage of the ovarian cycle. Granulosa cells from each follicle size category were harvested and pooled as suspensions in culture medium. Large follicles (2-3/animal) were presumed to be preovulatory, with every one yielding an apparently healthy cumulus-enclosed oocyte. No attempt was made to distinguish atretic and nonatretic small follicles. Granulosa cell viability, checked by trypan blue dye exclusion, was greater than or equal to 50%.
ON PRIMATE
GRANULOSA
CELLS Results
Activin on aroma&se follicles
activity in granulosa
During culture for 96 h in the presence of testosterone, estradiol production by granulosa cells from immature follicles increased dose dependently in the presence of FSH, becoming maximal at greater than or equal to 10 ng FSH/ mL (EDs0 - 3 ng FSH/mL). Activin alone, at all doses tested, did not significantly affect basal production of estradiol. However, activin concentrations greater than or equal to 10 ng/mL markedly augmented the action of FSH, and this effect was strongest in the presence of low-dose FSH (1 ng/ mL) (Fig. 1A). The maximum rate of estradiol production in response to FSH (10 ng/mL) plus activin (10 ng/mL) reflected almost quantitative metabolism of testosterone to estradiol; therefore, measurement of estradiol after a 6-h incubation with testosterone at the end of the 96-h culture was used to accurately determine aromatase activity. This approach demonstrated that the level of aromatase activity induced by a maximal stimulatory dose of FSH (10 ng/mL) was increased greater than 4-fold by activin (EDs0 l-10 ng/mL). Activin on progesterone immature follicles
production
FSH dose dependently stimulated by granulosa cells from immature
Tissue culture
cells from immature
by granulosa
cells from
progesterone production follicles, and this effect
Multiwell-plastic dishes (“Linbro” from Flow Laboratories, Rickmansworth, Hertfordshire, U.K.) were precoated with donor calf serum (GIBCO) and washed with Dulbecco’s phosphate-buffered saline before use (20). Culture wells were inoculated with replicate 250-PL portions of a cell suspension containing approximately 2.0 X lo4 live granulosa cells. Gonadotropin and/or peptide was previously added in culture medium to give a final incubation volume of 500 pL. The gonadotropins were human (h)FSH (LER-8/116) or hLH (LER-1972), donated by Dr. L. E. Reichert Jr. Recombinant human activin-A and recombinant human inhibin-A were provided by Genentech Inc. (San Francisco, CA). All treatments were in triplicate. The cultures were incubated at 37 C in a humidified tissue culture incubator gassed with a 95% sir/5% CO* mixture. Incubation was for 96 h with a medium change at 48 h. In some experiments, testosterone (1.0 Fmol) (Sigma Chemical Co., Poole, Dorset, U.K.) was included in the culture medium as an aromatase substrate. Alternatively, aromatase activity was measured at the end of the culture, when the monolayers were washed once with Dulbecco’s phosphate-buffered saline and incubated for a further 6 h in culture medium containing 1.0 pmol testosterone. Medium was stored frozen at -20 C until analysis of progesterone and/or estradiol content, as described below.
Steroid
RIA
Estradiol and progesterone in culture medium RIAs validated for these purposes (21, 22). The precision for each steroid was less than 15%.
Data presentation
were inter-
determined by and intraassay
and analysis
Steroid levels in culture medium are expressed as pmol steroid/1000 cells, related to the number of viable cells used to inoculate each culture well. The data are measurements of steroid accumulation from the 49. to 96-h period of culture, or during the subsequent 6 h incubation to determine aromatase activity. Results were analyzed by analysis of variance for multiple comparisons followed by Neuman-Keul’s test. Student’s t test was used for paired comparisons. Differences with P less than or equal to 0.05 were accepted as statistically significant.
0
1
10 FSH
100
(rig/ml)
FIG. 1. Dose-dependent effect of hFSH and activin on marmoset granulosa cell steroidogenesis from immature (
