Prostaglandins and follicular functions David T. Armstrong M.R.C. Group in Reproductive Biology, Departments of Obstetrics & Gynaecology and Physiology, University of Western Ontario, London, Canada N6A 5A5 Ovulation The involvement of prostaglandins (PGs) in the regulation of ovarian follicular function was first postulated on the basis of the demonstration that inhibitors of prostaglandin synthesis, such as aspirin and indomethacin, were capable of blocking ovulation in rats (Armstrong & Grinwich, 1972; Orczyk & Behrman, 1972). These initial findings were soon confirmed in several other species, including mice (Lau, Saksena & Chang, 1974), rabbits, (Grinwich, Kennedy & Armstrong, 1972; O'Grady, Caldwell, Auletta & Speroff, 1972), rhesus and marmoset monkeys (Wallach, de la Cruz, Hunt, Wright & Stevens, 1975; Maia, Barbosa & Coutinho, 1978), pigs (Ainsworth et al., 1979) and goldfish (Stacey & Pandey, 1975). In two of these species (rabbits and goldfish), the inhibitor was effective when applied locally to the follicle, indicating that the blockade was exerted directly upon the follicle, rather than being mediated via some indirect mechanism, such as through inhibition of gonadotrophin secretion. Further evidence of a role of prostaglandins at the follicular level was provided by the findings that intrafollicular levels of prostaglandins of both the E and F series increased markedly in several of these species shortly before ovulation (Yang, Marsh & LeMaire, 1974; Armstrong, Moon & Zamecnik, 1974; Bauminger & Lindner, 1975; Ainsworth, Baker & Armstrong, 1975; Tsang, Ainsworth, Downey & Armstrong, 1979a); indomethacin, at dosages which prevented ovulation, effectively prevented these increases. The observation that injection of antiserum against PGs blocked the LH-induced ovulation in oestrous rabbits, whether administered systemically (Lau et al., 1974) or via intrafollicular injection (Armstrong et al., 1974), added support to the concept of a role of prostaglandins in ovulation. Antiserum to PGF-2\g=a\appeared to be more effective than that to PGE in these experiments, suggesting that PGF-2\g=a\was the prostaglandin of greater importance in ovulation. Luteinization Numerous subsequent studies raised the possibility that prostaglandins of the E series may play a role in other follicular functions. In the studies of Yang et al (1974) with rabbits, PGE levels remained elevated somewhat longer than PGF levels after ovulation, leading the authors to suggest that while PGF-2a may be most important for follicular rupture, PGEs may play a role in the luteinization process which normally follows ovulation. Further support for this idea was provided by the demonstration in vivo that intrafollicular injection of PGE-2 induced luteinization of rabbit follicles (Phi, Moon & Armstrong, 1977), and by reports of the effects of PGE-2 on progesterone production by follicles or granulosa cells undergoing 'luteinization' in culture (Channing, 1972; Ellsworth & Armstrong, 1974; Neal, Baker, McNatty & Scaramuzzi, 1975). Surprisingly, blockade of ovulation by indomethacin appeared not to be accompanied by blockade of progesterone secretion and corpus luteum formation (Armstrong et al, 1974; 0022-4251/81 /030283-09S02.00/0 © 1981 Journals of Reproduction & Fertility Ltd Downloaded from Bioscientifica.com at 10/01/2021 11:25:14PM via free access Ainsworth et al, 1979), even when the indomethacin treatment was continued well beyond the time of ovulation (Phi et al, 1977). In addition, indomethacin appeared not to interfere with the normal preovulatory LH surge. These findings that ovulation could be blocked without any apparent alteration of gonadotrophin and steroid secretion led to hopes that a new class of anti-fertility drugs could be developed that possessed the desired anti-ovulatory effect but not side-effects associated with the more general endocrine disturbances caused by steroidal contraceptives. Prostaglandin synthesis by human follicles Before embarking on a widespread search for anti-prostaglandin agents more acceptable than indomethacin for ultimate use as ovulation inhibitors in women, it seemed important first to attempt to ascertain whether the concepts developed from the above-mentioned studies in experimental animals were applicable to women. To this end, we began investigation of prostaglandin production by the human follicle, as well as of effects of prostaglandins on human follicle cells. Results of studies with cultured human follicle wall tissue (theca and granulosa cells) indicated significant production of prostaglandin F, which was stimulated by addition of gonadotrophins (human menopausal gonadotrophin and human chorionic gonadotrophin, hCG) to the culture media (Plunkett, Moon, Zamecnik & Armstrong, 1975). Investigations with isolated follicle cell types have indicated that both the theca and granulosa cells have the ability to produce substantial amounts of prostaglandins in culture (unpublished observations). Prostaglandin effects on isolated cell types from human follicles Oestrogen biosynthesis In subsequent studies, human follicles were separated into their two principal cellular components, granulosa and theca cells, in order to examine their possible responsiveness to exogenous prostaglandins. For comparison, their responsiveness to the two classes of gonadotrophins, follicle stimulating hormone (FSH) and luteinizing hormone (hCG), were also determined. A comparison of the ability of granulosa and thecal preparations from a representative pool of 3-5 mm human follicles to secrete oestradiol-17ß when cultured without or with FSH and hCG, is presented in Text-fig. 1(a). In contrast to these low rates of oestrogen secretion by both cell types when cultured in the absence of an aromatizable substrate, addition of testosterone to the culture medium caused a striking 10- to 25-fold stimulation of oestradiol production by granulosa cells from the same pool of follicles (Text-fig. lb). In the presence of testosterone, an ability of purified FSH to stimulate oestradiol secretion by granulosa became evident. HCG was ineffective in stimulating oestrogen production by granulosa cells either in the absence or presence of testosterone. In the absence of testosterone, hCG appeared to stimulate oestradiol production by some thecal preparations, but this was not statistically significant, and in contrast to the results with granulosa cells, addition of testosterone to thecal preparations did not markedly alter oestradiol production, nor did it permit expression of effects of either gonadotrophin on oestradiol production (Moon, Tsang, Simpson & Armstrong, 1978). Cyclic AMP production Since there is abundant evidence that the gonadotrophic hormones exert their actions on their target cells via stimulation of specific receptor-linked adenylate cyclase, with resulting production of cyclic adenosine monophosphate (cAMP) acting as an intracellular 'second messenger' (Marsh, 1976), cAMP production was monitored in granulosa cells exposed to FSH Downloaded from Bioscientifica.com at 10/01/2021 11:25:14PM via free access (a) (b) 250 X I I Control EU FSH (250 ng/ml) 200 ^ hCG (1 ¡.u./ml) ja" 150 100 50 h fm- } rrU; 2 1 0 Granulosa Theca Granulosa Theca Text-fig. 1. Oestradiol-17ß secretion by granulosa and theca preparations from human follicles during culture for 24 h (a) without steroid substrate, and (b) with an aromatizable substrate (0-5 µ -testosterone). Values are mean ± s.e.m. for (a) duplicate and (b) triplicate cultures per treatment, from granulosa and theca preparations from a representative pool of 3-5-mm human follicles. FSH: Papkofflot G4-150C, ovine; hCG: Ayerst. (Data from Moon et al, 1978.) _ Control 2 50 500 1 10 10 100 FSH hCG PGE-2 (ng/ml) (i.u./ml) (ng/ml) Text-fig. 2. Cyclic AMP production by granulosa cells during incubations for 2 h. Values are mean + s.e.m. for quadruplicate incubations per treatment, from granulosa cells isolated from a representative pool of 4-6-mm human follicles. and to hCG. As is evident in Text-fig. 2, FSH, but not hCG, stimulated cAMP production by human granulosa cells. The failure of these cells to respond to hCG with increased production of cAMP or oestradiol suggested that the cells were too immature to have acquired LH (hCG) receptors. Investigations with granulosa cells from numerous species have revealed that LH receptors are acquired late in follicular development, and are indicative of granulosa cell maturity (Zeleznik, Midgley & Reichert, 1974; Richards, 1979). Prostaglandin E-2 at concentrations which appeared to be within the physiological range for other species was highly effective in stimulating cAMP production by these granulosa cells. PGE responsiveness is therefore apparently acquired at an earlier stage of maturity than is LH responsiveness. That the PGE-stimulated cAMP production by granulosa cells is probably of physiological significance is indicated by its effectiveness in stimulating production of both oestradiol (in the Downloaded from Bioscientifica.com at 10/01/2021 11:25:14PM via free access presence of testosterone only) and progesterone (both in the absence and presence of testosterone) (Table 1). Testosterone increased the production of progesterone by granulosa cells cultured with or without the stimulatory agents, FSH and PGE-2. Table 1. Effect of prostaglandin E-2 (PGE-2) on steroid production by human granulosa cells during culture for 24 h Hormone cone, (ng/mg protein) Treatment Progesterone Oestradiol-17ß Control 61 0 + 7-6 1-3 ± 0-6 Testosterone