The corpus luteum formed after luteinization of follicular cells secretes progesterone under the control of luteinizing hormone (LH). Binding of LH to its G-protein-coupled receptor leads to the activation of the adenylate cyclase/ cyclic AMP (cAMP)/cAMP-dependent protein kinase (PKA) signalling pathway. The identification of a new class of cAMP-binding proteins termed ‘guanine nucleotide exchange factors’ (cAMP-GEFs) provides a means by which changes in cAMP could yield actions that are independent of PKA. Hence, in this study, we have explored the hypothesis that steroidogenesis in luteinizing cells is mediated in both a cAMP/PKA-dependent and cAMP-dependent, but PKA-independent, manner. Human granulosa cells were isolated from follicular aspirates of women undergoing assisted conception. Luteinizing human granulosa cells were cultured for up to 3 days in the presence of human (h)LH and the adenylate cyclase activator forskolin in the added presence or absence of increasing doses of the PKA inhibitors H89 (N-[2-(4-bromocinnamylamino)ethyl] 5-isoquinoline) and PKI (myristoylated protein kinase A inhibitor amide 14–22) or the cAMP antagonist, Rp-cAMP. Agonist-stimulated progesterone secretion was inhibited in a dose-dependent manner by the PKA inhibitors and the cAMP antagonist, with decreasing sensitivity as luteinization progressed. Pretreatment of granulosa cells for 4 h with human (h)LH reduced the effectiveness of H89 in inhibiting progester-one secretion. Under basal conditions, cAMP-GEFI expression increased progressively throughout culture, and this could be further enhanced when cells were incubated with increasing doses of LH and forskolin. Furthermore, incubation of cells in the presence of increasing concentrations of the novel cAMP-GEF-specific cAMP analogue, 8 CPT-2 ME-cAMP (8-(4-chloro-phenylthio)-2′-0-methyladenosine-3′,5′-cyclic monophosphate), increased progesterone secretion in a dose-dependent manner. The results show that increases in cAMP generated by LH and forskolin, in addition to activating PKA, also induce increases in cAMP-GEFI protein expression in luteinizing human granulosa cells. In addition, activation of cAMP-GEFI results in increased progesterone secretion. Hence, increases in cAMP lead to the activation of PKA-dependent, as well as PKA-independent but cAMP-dependent (via cAMP-GEFI), signalling mechanisms. Since cAMP-GEFs have the capacity to activate the mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase (PI3-K)/protein kinase B (PKB) signalling pathways, these may provide the potential mechanisms by which cAMP-dependent but PKA-independent progesterone synthesis is regulated.

ABSTRACT

Progesterone production by dispersed luteal cells obtained from the marmoset monkey on day 14 after ovulation can be stimulated by both prostaglandin F_2α (PGF_2α) and its structural analogue, cloprostenol. To establish whether these responses can be attributed to cross-reaction with the prostaglandin E₂ (PGE₂) receptor, this study compared the involvement of cyclic adenosine-3′,5′-monophosphate (cAMP) and protein kinase C (PKC) in the luteotrophic responses to PGE₂, PGF_2α and cloprostenol. While all three prostaglandins stimulated similar increases in progesterone production (239·5 ± 7·9% of control; P <0·01), only PGE₂ stimulated a significant increase in cAMP accumulation (373·2 ± 28·4% of control; P <0·01). This study is the first to demonstrate PKC activity in the marmoset ovary. Following down-regulation of PKC with a tumour-promoting phorbol ester, 4β-phorbol 12-myristate 13-acetate (4β-PMA), basal progesterone production was significantly increased (150·9 ± 8·2% of control; P <0·05) and the luteotrophic effects of PGF_2α and cloprostenol were no longer evident, whereas the response to PGE₂ was unaffected. These observations are consistent with the differential involvement of cAMP and PKC in the luteotrophic responses to PGE₂ and PGF_2α/cloprostenol respectively. Hence, we conclude that the luteotrophic actions of prostaglandins E₂ and F_2α on dispersed marmoset luteal cells are mediated via different receptors and signal transduction pathways.

Journal of Endocrinology (1993) 138, 291–298

Luteinization of follicular granulosa cells leads to an increase in progesterone secretion that is regulated by luteinizing hormone (LH). LH acts mainly by elevating intracellular cyclic 3′,5′-adenosine monophosphate (cAMP) and activating cAMP-dependent protein kinase (PKA). In this study, we have examined the role of PKA in relation to progesterone output by luteinizing human granulosa cells. Human granulosa cells were obtained by percoll gradient centrifugation of follicular aspirates of patients undergoing oocyte retrieval for assisted conception. Cells were cultured in serum-supplemented medium for up to 3 days in the presence and/or absence of human (h)LH and other cAMP-elevating agents. Spent medium was assayed for cAMP and progesterone content by specific RIA. Cell lysates were collected and assessed for PKA regulatory (R)IIα/catalytic (C)α expression by Western blotting. Although basal progesterone secretion increased progressively throughout culture, cAMP levels remained unchanged. Under basal conditions, PKA RIIα/Cα expression appeared to increase throughout the 3-day culture period. In the presence of hLH and other cAMP-elevating agents, progesterone secretion increased in a dose-dependent manner coincident with an increase in cAMP. However, despite the increase in both progesterone secretion and cAMP accumulation, there was a dose-dependent decrease in both PKA RIIα and Cα expression. Thus, data presented in this study show that increases in progesterone secretion in luteinizing human granulosa cells can be dissociated from increases in PKA expression. This notion implies that progesterone secretion may be regulated by PKA-dependent as well as PKA-independent mechanisms.

Cortisol–cortisone metabolism is catalysed by the bi-directional NADP(H)-dependent type 1 11β-hydroxysteroid dehydrogenase (11βHSD1) enzyme and the oxidative NAD⁺-dependent type 2 11βHSD (11βHSD2). This study related the expression of 11βHSD1 and 11βHSD2 enzymes (mRNA and protein) to net 11-ketosteroid reductase and 11β-dehydrogenase (11β-DH) activities in bovine follicular granulosa and luteal cells. Granulosa cells were isolated from follicles of < 4, 4–8, > 8 and > 12 mm in diameter in either the follicular or luteal phase of the ovarian cycle. Luteal cells were obtained from corpora lutea (CL) in the early non-pregnant luteal phase. Enzyme expression was assessed by reverse transcription-PCR and western blotting, while enzyme activities were measured over 1 h in cell homogenates using radiometric conversion assays with 100 nM [³H]cortisone or [³H]cortisol and pyridine dinucleotide cofactors. Irrespective of follicle diameter, the expression of 11βHSD2 and NAD⁺-dependent oxidation of cortisol predominated in granulosa cells harvested in the follicular phase. In contrast, in granulosa cells obtained from luteal phase follicles and in bovine luteal cells, expression of 11βHSD1 exceeded that of 11βHSD2 and the major enzyme activity was NADP⁺-dependent cortisol oxidation. Increasing follicular diameter was associated with progressive increases in expression and activities of 11βHSD2 and 11βHSD1 in follicular and luteal phase granulosa cells respectively. In follicular phase granulosa cells from antral follicles < 12 mm, 11βHSD1 migrated with a molecular mass of 34 kDa, whereas in the dominant follicle, CL and all luteal phase granulosa cells, a second protein band of 68 kDa was consistently detected. In all samples, 11βHSD2 had a molecular mass of 48 kDa, but in large antral follicles (> 8 mm), there was an additional immunoreactive band at 50 kDa. We conclude that 11βHSD2 is the predominant functional 11βHSD enzyme expressed in follicular phase granulosa cells from growing bovine antral follicles. In contrast, in bovine granulosa cells from dominant or luteal phase follicles, and in bovine luteal cells from early non-pregnant CL, 11βHSD1 is the major glucocorticoid-metabolising enzyme. The increasing levels of cortisol inactivation by the combined NADP⁺- and NAD⁺-dependent 11β-DH activities suggest a need to restrict cortisol access to corticosteroid receptors in the final stages of follicle development.

ABSTRACT

The actions of forskolin have been investigated to determine to what extent its effects on steroidogenesis in rat adrenal preparations are dependent on activation of adenylate cyclase. In zona glomerulosa preparations, stimulation of both aldosterone and corticosterone production was obtained at concentrations of forskolin between 1 and 10 μmol/l. The effects of 10 μmol forskolin/l were additive with those of low doses (1 pmol/l) of corticotrophin (ACTH), but not with those of high doses (1 nmol/l) of ACTH. In contrast, in zona fasciculata/reticularis cells, doses of forskolin up to 10 μmol/l produced no significant stimulation of corticosterone production either alone or in the presence of ACTH (1 pmol/l and 1 nmol/l). The response to 1 nmol ACTH/l was attenuated in the presence of forskolin (10 μmol/l) in both zona glomerulosa and zona fasciculata/reticularis cell preparations. Cyclic AMP production increased progressively with dose up to 100 μmol forskolin/l in zona glomerulosa cells, whereas corticosterone production was maximal between 10 and 30 μmol forskolin/l and decreased at 100 μmol forskolin/l. In zona fasciculata/reticularis cells, cyclic AMP production was also increased by forskolin (1 and 10 μmol/l).

The stimulation of zona glomerulosa steroido-genesis by forskolin (1–10 μmol/l) and ACTH (1–100 pmol/l) were both reduced by the adenylate cyclase inhibitor, N ⁶-phenylisopropyladenosine (100 μmol/l). The calcium channel inhibitor, nifedipine, only reduced the steroidogenic response to forskolin (3 μmol/l) at doses of 300 μmol/l whereas the response to 8·4 mmol K⁺/l was inhibited at 10 μmol nifedipine/1. Although there is some dissociation between the effects of forskolin on cyclic AMP and steroidogenesis, the results are generally consistent with the view that the effects of forskolin in rat zona glomerulosa cells are mainly dependent on activation of adenylate cyclase. This contrasts with the effects of forskolin in bovine fasciculata cells which are reported to be mediated by activation of voltage-regulated calcium channels.

Journal of Endocrinology (1991) 129, 391–397

ABSTRACT

It is possible that some of the effects of sodium pentobarbitone on the hypothalamo-pituitary-adrenal axis in the intact animal may be attributable to direct actions on the adrenal cortex. The effects of the barbiturate on steroid production by rat adrenal preparations in vitro have therefore been examined. In zona glomerulosa cells, pentobarbitone inhibited basal steroid production in a dose-related fashion. For aldosterone and corticosterone, the doses required for 50% inhibition of production (IC₅₀) were 1·2 mmol pentobarbitone/l and 3·7 mmol/l respectively. Steroidogenesis was inhibited at lower levels of pentobarbitone in the presence of 1 nmol ACTH/l (IC₅₀ = 0·5 mmol pentobarbitone/l for aldosterone and 2·2 mmol/l for corticosterone). In zona fasciculata/reticularis cells, production of corticosterone was similarly reduced with an IC₅₀ of 2·8 mmol pentobarbitone/l for basal production and 1·3 mmol/l for ACTH-stimulated production. The dose-related increases in corticosterone production produced by ACTH (0·1–1000 pmol/l) or dibutyryl cyclic AMP (0·1–1·0 mmol/l) were also eliminated in the presence of 2 mmol pentobarbitone/l. The effects of pentobarbitone (1–4 mmol/l) on the production of pregnenolone and deoxycorticosterone (DOC) were also studied. In zona fasciculata/reticularis cells, the responses of both pregnenolone and DOC were bell-shaped with increases at 1 mmol pentobarbitone/l, which fell back to control levels at 4 mmol pentobarbitone/l. Stimulation of DOC, accompanied by decreases in aldosterone and corticosterone production, was also seen in zona glomerulosa cells at 1 mmol pentobarbitone/l.

The effect of 1 mmol pentobarbitone/l on the conversion of 22(R)-hydroxycholesterol (5-cholestene-3β,22(R)-diol), pregnenolone, progesterone and DOC to corticosterone and aldosterone by zona glomerulosa preparations was studied. There was a comparable reduction in the conversion of these precursors (2 μmol/l) to aldosterone with yields decreased to 20–30% of those found in the absence of pentobarbitone. The dose required for 50% reduction of the conversion of progesterone (2 μmol/l) to aldosterone was 0·55 mmol pentobarbitone/l and for corticosterone the dose was 1·75 mmol pentobarbitone/l.

The results obtained show that pentobarbitone is an effective inhibitor of corticosteroid biosynthesis in rat adrenal cells, and suggest that its effects are brought about by inhibition of cytochrome P450-mediated hydroxylations.

Journal of Endocrinology (1993) 136, 75–83

Abstract

The entire coding region of an ovine endometrial oxytocin receptor (OTR) cDNA was generated by PCR, subcloned into the SV40 major late promoter expression vector pSVLJ and transiently expressed in Cos-7 cells. A specific OTR antagonist, ¹²⁵I-labelled d(CH₂)₅ [Tyr(Me)²,Thr⁴,Tyr-NH₂ ⁹]-vasotocin (OTA), was used to describe the binding kinetics of the expressed receptor which had a K _d of 4·5 nm and B_max of 2·4 nm/mg protein (6·8 × 10⁵ receptor molecules/transfected cell). The functional properties of the expressed OTR were determined by measuring oxytocin-induced phosphoinositide (PI) hydrolysis. Oxytocin increased PI turnover in OTR transfected cells fourfold in excess of residual endogenous activity, and stimulated phospholipase C (PLC) activity in a dose- and time-dependent manner, confirming that the expressed OTR cDNA was functional. Arginine vasopressin also stimulated PI turnover in a dose-dependent manner; thresholds of responses to oxytocin and arginine vasopressin were 10⁻⁹ m and 10⁻⁷ m respectively. OTA did not increase PI turnover and competitively inhibited the oxytocin-induced response. Direct activation of the pathway by aluminium fluoride and guanosine (3′-Othio)-triphosphate (GTPγS) confirmed that the OTR was G-protein linked. Co-incubation of GTPγS with oxytocin shifted the PI-response threshold from 10⁻⁷ m to 10⁻⁹ m and significantly increased the level of response, suggesting that maximum PI turnover was agonist-dependent. The G-protein involved in mediating the signal transduction pathway was pertussis toxin-insensitive and, therefore, probably a member of the Gq subfamily. The PLC inhibitor, U73122, had no effect on oxytocin-induced PI turnover, consistent with the response in endometrial tissue. These data suggest that the signalling pathway mediated by expressed OTR is similar to that attributed to OTR occupancy in ovine endometrium.

Journal of Endocrinology (1996) 149, 389–396

Abstract

Although the role of protein kinases and phosphorylation in steroidogenesis has received much attention, very little is known about the activities of phosphoprotein phosphatases (PP) and dephosphorylation in steroidogenic tissues. The aims of the present study were therefore to identify which of those serine/threonine PPs more commonly involved in intracellular signalling are expressed in rat luteal cells; to quantify, in vitro, the effects of inhibitors on PP activity extracted from purified rat luteal cells; and to measure the effects of PP inhibitors on the phosphorylation of endogenous luteal cell proteins. Polyclonal antibodies raised against the catalytic subunits of PP types 1 and 2A, and a monoclonal antibody raised against the Ca²⁺-binding subunit of PP2B, were used to identify immunoreactive proteins that migrated on SDS-PAGE with approximate molecular masses of 37, 34 and 16 kDa, corresponding well with the reported molecular mass of PP1, PP2A and PP2B respectively. Five selective inhibitors of PP1/PP2A: okadaic acid, calyculin A, cantharidin, tautomycin and microcystin-RR, each caused a dose-dependent decrease in the activity of PPs in luteal cell homogenates, and also enhanced ³²P incorporation into numerous luteal cell proteins; most notably, proteins with approximate molecular masses of 20 and 22 kDa. The results of this study suggest that PPs may play an important role in the regulation of rat luteal cell functions.

Journal of Endocrinology (1996) 150, 205–211

Abstract

The key role of protein kinases and protein phosphorylation in the regulation of luteal steroidogenesis is well documented. However the role of phosphoprotein phosphatases (PP) and dephosphorylation in the regulation of luteal cell progesterone secretion is as yet unknown. We have recently demonstrated the presence and activity of PP1 and PP2A in rat luteal cells and the present study was undertaken to determine the consequences of inhibiting PP activity in terms of progesterone secretion. Three structurally dissimilar inhibitors of PP1/2A, okadaic acid, calyculin A and cantharidin each caused a dose-dependent inhibition of LH-induced progesterone secretion without affecting cyclic AMP accumulation. The less potent derivative of okadaic acid, norokadaone, had no effect on either parameter, suggesting that the inhibitory actions on progesterone secretion are due to their specific actions on PP activity and that this inhibition occurs principally at a locus which is distal to the generation of cyclic AMP. In contrast to the inhibitory effects of PP1/2A inhibitors on progesterone biosynthesis, a PP2B inhibitor, cypermethrin, had no effect on LH-stimulated steroidogenesis. The three PP1/2A inhibitors also caused a concentration-dependent inhibition of dibutyryl cyclic AMP-stimulated progesterone secretion. However, none of the inhibitors affected 22R-hydroxycholesterol-supported steroidogenesis, clearly demonstrating that the inhibitors did not interfere with the activity of steroidogenic enzymes. These results suggest that cycles of phosphorylation/dephosphorylation of specific proteins are required for the sustained production of progesterone. Whilst the precise location and function of putative PP substrates is uncertain, the present results indicate that they are involved in regulating the availability of free cholesterol to steroidogenic enzymes within mitochondria.

Journal of Endocrinology (1996) 150, 213–221

Polyunsaturated fatty acids derived from the diet are incorporated into cell membranes where they act as precursors for prostaglandin (PG) synthesis. Linoleic acid (LA; 18:2 n-6) is a major constituent of plant oils and its consumption in Westernized populations is increasing. This study investigated the influence of LA on PG production by the uterus and placenta. Pregnant ewes were fed a control or an LA-enriched diet. Oxytocin (OT) was injected on day 45 (early) or day 133 (late) of gestation to measure the release of 13,14-dihydro-15-keto PGF_2α (PGFM). Ewes were killed on day 46 or day 138 for collection of uterine intercaruncular endometrium and fetal allantochorion. Basal and stimulated PG release from explant cultures was assessed before and after in vitro treatment with OT, lipopolysaccharide (LPS), dexamethasone (DEX) or calcium ionophore (CaI). Expression of cyclooxygenase (COX)-1 and COX-2 was determined by Western blot in endometrium of late-gestation ewes. Circulating PGFM levels in vivo did not differ according to diet but there were highly significant differences in the release of PGs in vitro. Basal production of PGF_2αand PGE₂ by the endometrium and of PGE₂ by the allantochorion were all higher in tissues from LA-supplemented ewes. Endometrial tissues produced more PG following OT and CaI treatment, whereas DEX inhibited production of both PGs at both stages of gestation. In allantochorion collected at day 46 LPS did not significantly alter PGE₂ release and DEX increased output, whereas at day 138 LPS was stimulatory but DEX was inhibitory. These data show that a high-LA diet can significantly increase the ability of both endometrium and placental tissues to produce PGs in vitro. This effect of diet may only become apparent after a sustained period of PG release, so was not seen following the brief pulse caused by OT treatment in vivo. As COX protein levels were unaltered, the main influence was likely to be via conversion of LA to arachidonic acid, providing an increased supply of precursor. These results support previous studies which suggest that alterations in dietary polyunsaturated fatty acids may influence the time of labour.