Adenosine as substrate and receptor agonist in the ovary

ADENOSINE

AS SUBSTRATE AND RECEPTOR AGONIST IN THE OVARY

Hakan Billig, Sten Rosberg, Carl Johanson, and Kurt Ah&n Department ofPhysiology, University S-400 33 Gliteborg, Sweden Corresponding

author:

of Gtiteborg, PO Box 33031,

Hiikan Billig, M.D., Ph.D.

ABSTRACT

In the present study the possible dual effects of adenosine as substrate and adenosine receptor agonist in rat granulosa cells, cumulus-oocyte complexes, luteal cells and ovarian membranes are discussed. Adenosine is an indispensable compound in cell energy metabolism, as precursor to cofactors, second messenger and nucleic acids. Adenosine is also an agonist to adenosine receptors. The adenosine receptor can either inhibit (AI) or stimulate (AZ) adenylate cyclase. Alternatively, in some cells adenosine receptor activation is linked to other cellular events like inhibition of Ca2+ fluxes. Adenosine is taken up by isolated preovulatory ranulosa and luteal cells from pregnant mare serum onadotropin-treate 8 immature rats, but follicle stimulating hormone (FSH7 decreases the uptake by granulosa cells. Adenosine, but not the non-metabolizable adenosine analogs 5’-(Nethyl)carboxamido-adenosine (NECA), 2-chloro-adenosine (2-Clado), N6-(Rphenyl-isopropyl)-adenosine (R-PLA) and N6-(S-phenyl-isopropyl)-adenosine (S-PLA), increase granulosa cell ATP levels. FSH and luteinizing hormone (LH) decrease granulosa cell ATP levels in the presence or absence of adenosine. It has previously been shown that FSH and LH decrease oxygen consumption by cumulus-oocyte complexes and increase their lactate production. These effects have been suggested to be due to a competition of cofactors (e.g. ADP) common to glycolysis and the respiratory chain. The fact that adenosine reverse the gonadotropin-induced effects on oxy en consumption and lactate production support this theory. Adenosine an 8 its analogs increase CAMP accumulation in luteal and granulosa cells only in the presence of gonadotropins, and this effect is anta onized by the adenosine receptor antagonist 8-phenyl-theophy 9line (8-PHT). Furthermore, adenylate cyclase is stimulated by adenosine analogs in membranes from non-luteinited and luteinized ovarian membranes and in luteal cell homogenates. The effect of NECA is antagonized by 8-PHT. In the membranes, the rank order of potency was NECA>2-Clado>R-PLA>SPLA, suggesting adenosine A2 receptors. In summary, it is suggested that adenosine can act both as a substrate to intracellular metabolism and as an adenosine A2 receptor a onist in granulosa and luteal cells. A paracrine short loop positive fee d back model is proposed where extracellular adenosine, derived from a gonadotropin-induced extracellular increase in CAMP and a decrease in cellular ATP, enhances gonadotropin stimulation in granulosa and luteal cells.

STEROIDS54/5

November

1989 (523-542)

523

524

Bilfiget

al: OVARIAN

ADENOSINE

INTRODUCTION

Within the gonads, a multitude of different locally produced factors can influence or regulate the neighboring cells or the producing cell itself (1). One substance that could be regarded as a local regulator is adenosine. Besides being substrate for intracellular metabolism, adenosine acts in many instances as a receptor agonist, regulating adenylate cyclase activity (2.3) and Ca2 + fluxes (4). Several decades ago, it was shown that infusion of adenosine or AMP reduced heart rate, blood pressure, inhibited intestinal movements, and dilated coronary vessels in a number of animals(S). Since then, adenosine has been shown to elicit a number of adenosine receptor-mediated effects in adipose, neural, cardiac, hepatic, nephric, skeletal, endocrine, and vascular tissues (6). The action of adenosine close1 resembles paracrine and autocrine action (7), i.e., adenosine is release dylocally, affects adjacent cells or the releasing cell itself, and the release is locall regulated. Adenosine has also been proposed to be a local hormone (8r or a local “retaliatory metabolite” (9). RESULTSAND DISCUSSION Adenosine as substrate and receptor aqonist in the ovary In the onads, adenosine has been suggested to exert local action: extrain the cellular9y in the testis as a receptor agonist (10) and intracellular . ovary as a substrate (11,12). The testis and the ovary, though drffyermg in functional end roducts, share many metabolic characteristics, such as cAMP production an B steroidogenic responses to onadotropins (13,141 as well as the lactate reduction by supportive cells 9i.e., Sertoli and granulosa cells (15,16)) an 8 substrate requirements of the germ cells (i.e., s ermatocytes and oocytes (17-19)). In the testis, adenosine receptors have 1 een demonstrated (20) and, when activated, these adenosine receptors exhibit functional responses (10). If the action of adenosine in the ovary is autocrine/paracrine then: 1) 2) 3)

Adenosine has to induce metabolic and other effects in neighboring cells or in the adenosine-releasing cell itself. Adenosine has to be produced and released by ovarian cells. The ovarian cells must have a regulatory mechanism for the production, release, and/or response to adenosine.

Adenosine metabolism and uptake into cells The formation of adenosine is accomplished through several pathways like de novo synthesis, dephosphorylation of AMP, and degradation of S-adenosylhomocysteine. AMP is the most likely contributor of the bulk of “recirculating” adenosine within the cell (8). However, the net addition of adenosine and purines in the cell is probably not derived from de novo synthesis in the cells but rather is a result of uptake from the blood (21).

STEROIDS

54/5

November 1989

Billig et al: OVARIAN

525

ADENOSINE

There are three principal wa s in which adenosine is removed from the cell (Fig. 1): deamination, phosp i orylation, or release to the extracellular space as adenosine or adenosine-derived compounds (AMP, CAMP).

Outside

PHypoxanthine

Inside

Figure 1. Schematic presentation of adenosine metabolism and possible routes for transmembrane fluxes of adenosine and related compounds. RGC= gonadotropinadenylate cyclase complex, POE = phosphodies-terase, ADA = adenosine deaminase.

Adenosine and other nucleosides can be transported from the extracellular space across the cell membrane into the cell by facilitated diffusion. At physiological concentrations, specific carrier proteins transport nucleosides. The carrier proteins seem to be present in all mammalian cells (8). Nucleosides can also be transported by simple diffusion. Purine and pyrimidine bases share a common carrier protein distinct from the nucleoside carrier protein (22). The rate of translocation of adenosine, from the exterior to the interior of the cells, is dependent on two processes, the transport mechanism itself and the intracellular metabolization. The intracellular concentration of adenosine is very low, since the nucleoside is rapidly metabolized and a concentration gradient over the cell membrane is maintained (8,22). Adenosine receptors Adenosine receptors are localized at the external surface of the cell membrane and have been demonstrated in a number of tissues. The adenosine receptors are divided into two subclasses (At and AZ) based on the affinity of different adenosine analogs like 5’-(N-ethyl)carboxamidoadenosine (NECA), N6-(R-phenyl-isopropyl)-adenosine (R-PIA), N6-(S-phenylisopropyl)-adenosine (S-PIA), and 2-chloro-adenosine (2-Clado) and on their

STEROIDS

54/5

November

1989

526

Billiget

al: OVARIAN

ADENOSINE

effect on adenylate cyclase activity (3,23). The rank order of potency for the A1 receptor is R-PIA>2-Clado>NECA>S-PIA, and for the A2 receptor, NECA>Z-Clado>R-PIA>S-PIA, with Al inhibiting and A2 activating adenylate cyclase (6). The A1 receptor is considered to be a high affinity receptor with an ICso in the nanomolar range, and the A2 receptor a low affinity receptor with an EC50 in the low micromolar range (6,24). The two functionally and pharmacologically distinct adenosine receptor-mediated effects on adenylate cyclase are GTP-dependent like the stimulating and inhibitin effects of other hormones on aden late cyclase (24). However, not all a%enosine receptor effects are mediate J by adenylate cyclase; it has been shown that adenosine receptor activation is coupled to inhibition of Ca2 + fluxes, e.g., in cardiac muscle (25) and neural tissue (26). Methylxanthines like caffeine, theophylline, and isobutylmethylxanthine are classical phosphodiesterase inhibitors, but they are also adenosine receptor antagonists blocking both Al and A2 receptors (27). Methylxanthines block adenosine receptors at concentrations lower than those required for inhibition of phosphodiesterase (28). and it has been suggested that man of the effects of these a ents in vivo are due to their adenosine receptor- 6 locking activity and not 8 ue to phosphodiesterase inhibition (29). Effects of adenosine in the testis In the testis, several aspects of adenosine receptor action have been documented. Crude homogenates of testis and brain exhibit the highest binding capacity for At receptor agonist of all organs tested (30,31). In the testis, adenosine receptor binding or activation has been claimed to be associated with spermatocytes, Leydig cells, and Sertoli cells, and some controversy exists as to what testicular cell type the physiological action of adenosine should be attributed. Binding of adenosine appeared to be associated with the spermatocytes within the seminiferous tubule epithelium (32). However, binding to Settoli cells with the characteristics of A1 receptors has convincingly been demonstrated (10,33). The functional response of adenosine agonists in the normal testis suggests the presence of an inhibiting Al receptor. In Sertoli cells, FSH-stimulated adenylate cyclase activity is inhibited b adenosine analogs and so is FSHstimulated androgen aromatization r 10.33). Other agents stimulating Settoli cell adenylate cyclase are also inhibited by adenosine (10,34). In early reports, adenosine agonists were shown to stimulate adenosine receptor-mediated aden late cyclase activity in a Leydig tumor-cell line, and these results were useJ as part of the basis for the definition of the stimulating A2 adenosine receptors (2.35). A2 receptors do not seem to be present in the normal testis (10,32,33). However, it has been reported that normal Leydig cells developed responsiveness to adenosine durin culture and that A2 adenosine receptor is a result of the in vitro conditions 936). A physiological role for adenosine receptor-mediated effects has been presented where Sertoli cell adenosine receptors are part of intercellular communication between the Settoli and germ cells whereby Sertoli cells

STEROIDS

5415

November 1989

Billig et al: OVARIAN

monitored cataiyzate,

ADENOSINE

527

the ener y balance in germ cells by sensing adenosine, released Brom the germ cells (33).

Auqmentation

of adenvlate

cvclase activity by adenosine

the ATP

in ovarian ceils

Adenosine can increase cAMP production in two principally different ways, either by increasing substrate availability to adenylate cyclase or by stimulating adenylate cyclase through adenostne A2 receptors. Adenosine and its analogs dose-dependently stimulate basal CAMP accumulation and potentiate gonadotropin-stimulated CAMP accumulation in granulosa cells (37-40), in cumulus-oocyte complexes (41), and in luteal cells (42-46). However, in granulosa, luteal, and Leydig cells (47) the stimulation by adenosine or its analogs alone in ovarian cells is quantitative1 small compared with the augmentation of gonadotropin-stimulate o! adenylate cyclase. Behrman and coworkers (42) suggested that the potentiating effect of adenosine on LH-induced CAMP accumulation in luteal cells was mainly due to an action of adenosine as a substrate for ATP (38). However, later experiments from our group with the adenosine analog NECA indicate that adenosine has an extracellular nonsubstrate action, probably receptormediated (40). As adenosine, the adenosine analog NECA alone stimulates basal CAMP production only slightly in granulosa (40) and luteal cells (46). However, NECA augments gonadotropin-stimulated CAMP accumulation (40,46, Fig. 2). Similar findings have been presented for Leydig cells (47). The EC50 for NECA in FSH-induced CAMP accumulation is around 40 pM in granulosa cells, which is in the same magnitude as for adenosine (40). In LH-stimulated

I

Am%osnIE

NECR

Cprrl

IO

$4”)

loolmo

.Jd05 cc1

.I .*o

.

10

.

lrnlmo q4rll

Figure 2. Dose-dependent potentiation by adenosine, NECA, and 2Clado of FSH-induced cAMP accumulation in granulosa cells. Granulosa cells were incubated for 180 min in the presence of oFSH (100 ng/ml) together with the indicated concentrations of adenosine or adenosine analogs. Each point represents the mean f SEM of triplicate culture wells. The points were fitted with nonlinear regression to four-parametric dose-response curves. Adapted from reference (40) with permission from Endocrinology.

STEROIDS

54/5

November

1989

528

Billig et al: OVARIAN

luteal cells, similar results have been reported for adenosine was reported to be 22 pM (42). The stimulatory effect of accumulation is antagonized by a receptor antagonist in granulosa cells (40) and, to a lesser degree, in LH-stimulated

ADENOSINE

where the EC50 NECA on cAMP FSH-stimulated luteal cells (46).

The EC50 values are at least one order of magnitude hi her than those reported for most other cells with A;2 receptors (6,48). TYlus, it could be questioned whether the effect of NECA is mediated via an adenosine receptor. However, in membranes prepared from preovulatory ovaries from PMSG-treated immature rats, from luteinized ovaries, and from luteal cells, NECA alone stimulated adenylate cyclase activit with EC50values (0.28-0.65 FM), i.e., markedly lower than in granulosa cclrs and more in the expected range for an A2 receptor. The effect of NECA was antagonized by xanthines. The order of potency for adenosine analogs was found to be that of an A2 receptor, i.e., NECA>2-Clado>R-PIA>S-PIA (40,46,49). However, the maximal stimulatory effect of NECA is rather limited, only 30% over basal activity and half of the maximal FSH stimulation (40,49). Adenosine stimulates basal progesterone synthesis in human and rat granulosa cells (37,39) and luteal cells (37,42). Also, the adenosine analog NECA markedly strmulates the progesterone synthesis in luteal cells (46). The stimulatory effect of adenosine and its analogs on steroidogenesis is limited and, which should be stressed, is not in concordance with the marked cAMP response. It might be that the lack of concordance is an in vitro artifact. However, if the adenosine action is paracrine-like, an alternative suggestion could be that adenosine released from the cells occupies a small fraction of adenosine receptors sufficient for maximal steroid0 enic response. This issupported by the findin that submaximally FSH-stimu4ated ranulosa cells exposed to adenosine 8 eaminase in a superfusion system Y50) responded with larger pro esterone production in the presence than in the absence of NECA. When a8 enosine deaminase was not included, NECA did not change FSH-stimulated progesterone production (Fig. 3). This suggests that endogenously released adenosine ’ auto-stimulates” the cells to increase their steroidogenic response to a submaximal FSH stimulation. However, it also implies that only a fraction of the putative adenosine receptors need to be occupied in order to achieve a maximal steroidogenic response. It is perhaps wrong to phrase the question whether the action of adenosine is extracellular or intracellular. Instead, one ma rather ask whether the extracellular receptor-mediated and the intrace rlular substrate actions of adenosine are related to each other and cooperate in the cell. Effects of adenosine on ovarian cell ATP levels Adenosine is taken up by granulosa cells, but compared with luteal cells and most other cells from nongonadal tissue, the uptake of adenosine by granulosa cells is slower (8.22). We found that the K,,, in granulosa cells did not differ from that in luteal cells (15.9 & 4 pM and 19.4 & 8 M, respectively), but the Vmawwas considerabl lower in granulosa cells (1. P k 0.1 pmol/min/lOs cells) than in luteal cells (Lr.l f 0.6 pmol/min/l05 cells).

STEROIDS

54/5

November 1989

Billig et al: OVARIAN

529

ADENOSINE

The rate of uptake of adenosine in granulosa cells was further decreased in the presence of FSH (39). Since the uptake of adenosine is regulated by the intracellular activity of adenosine kinase and adenosine deaminase, it suggests a difference in the intracellular metabolism of adenosine in granulosa and luteaf cells.

0’

0

10 . ..*

2.0

30

40

i@ECA ....” *..... . ..*. C

FS;t Frwttcn

O-

0

10 _ to

JO

iI&. ,“‘1”‘;“““....‘..“, FSH Fr .ct

40 NECA C

, on

Figure 3. Effect of NECA on FSH-stimulated pro esterone production in superfused granulosa cells in absence (left) and presence (right I” of adenosine deaminase. Granulosa cells were superfused for 9 h. Medium, 30 BUmin, was sampled in 15-min fractions. FSH (10 ng/ml) was present from fraction 14 to 20 in all superfusions. NECA (open triangles, 100 PM) was introduced from fraction 9 and was resent until the end of su erfusion. Basal secretion was set to lo%, and was the mean o Pfractions 1-8, i.e., before a 1..drtron of NECA. Basal secretion in absolute terms was 1.33 n fractionimg protein and 3.31 n~ractio~m protein in the absence and presence of a 8 enosine deaminase (ADA), respectively. A II subsequent values are expressed as percent of the basal secretion. Data (mean f SEMI were pooled from two identical experiments with two superfusions for each treatment.

Adenosine increases the ATP level in luteal cells (43,44,51,52), in granulosa cells (38,39), and in cumulus-ooc te complexes (53). The ATP content in granulosa cells is dose-dependent ry decreased by FSH (39), and this effect is even more evident in the presence of adenosine (38,39, Fig. 4). The same phenomenon is also seen with cumulus-oocyte complexes (53). Also, LH decreased the ATP level in prepubertal ovaries (54), in luteal cells isolated from young postovulatory corpora lutea (52), in luteinized ovaries in vivo (55), and in incubated granulosa cells (39), but not in cells isolated from corpora lutea from midluteal phase (52,55). The adenosine analogs NECA, 2Clado, R-PIA, and S-PIA did not significantly Increase ATP levels in the granulosa cells. The decreasing effect of FSH (39) was still present with all analogs since they did not increase the ATP levels (40). This suggests that the augmentation of gonadotropin-induced cAMP response by adenosine analogs is not due to increased cellular ATP levels. Effects of adenosine on ovarian cell qlvcolvsis and oxvqen consumption It has been suggested that lactate production is impo~ant for the oocyte, smce it cannot utilize glucose as energy substrate but requires lactate and

STEROIDS

54/5

November

1989

530

Billig et al: OVARIAN

0-I oM,

.I

I oFSH

10

cng/ml

100

ADENOSINE

1000

1

Figure 4. FSH dose-dependent depression of ATP levels in granulosa cells. Granulosa cells were incubated for 90 min in the absence (open symbols) or presence (filled symbols) of adenosine (50 PM) together with the indicated concentrations of oFSH. Each point represents the mean f SEM of triplicate culture wells. Adapted from reference (39) with permission from Endocrinology.

pyruvate (19,56,57). In analo y, in the testis Sertoli cells produce lactate (18) and the production is stimuBated by FSH (15.18). Testis germ cells can be supported with lactate but not glucose (17,58), and oxy en consumption increases in isolated germ cells after addition of lactate (187 . The preovulato follicle has a high glycolytic capacity (59-61), and lactate accumulation isx igh in gonadotropin-stimulated granulosa cells in spite of hyperphysiological oxygen tension during the incubation (16,62). For example, as much as 80-90% of metabolized lucose is found as lactate in isolated follicles (60). Despite increased meta %olic activity, gonadotropins decreased the oxygen consumption in cumulus-ooc te complexes (63-65). The increased lactate accumulation may indicate a lI ecreased rate of pyruvate entering the citric acid cycle. The concomitant decreased oxygen utilization suggests that the increased lactate accumulation is not due to a lack of available oxy en, but rather reflects an inherent metabolic characteristic of these cells. T4 is has been proposed to be due to a competition for a limited quantity of cofactors in common for both glycolysis and oxidative phos horylation (53,60,64). This mechanism was proposed several decades a o P66) to be the cause of decreased oxygen consumption after addition of g8ucose to certain tumor cells (67). It has been su gested that a relative lack of ADP (68) or inorganic phosphate (69) would % e limiting factors for the decreased rate in the respiratory chain with concomitant decreased oxygen consumption and increased lactate formation. Addition of metabolizable

STEROIDS

54/5

November

1989

Billiget

al: OVARIAN

531

ADENOSINE

adenosine abolished the gonadotropin-induced decrease in oxygen consumption in cumulus-oocyte complexes. Lactate accumulation also decreased, while ATP levels increased (53). Addition of adenosine to granulosa cells also decreased lactate accumulation (Table 1) and increased ATP formation (38,393. The results suggest that, in both stimulated cumulus-oocyte complexes and increased the activity in the respiratory chain pyruvate entering the citric acid cycle, while lactate accumulation decreased. TABLE 1 EFFECT OF ADENOSINE ON LACTATE ACCUMULATION IN CRANULOSA CEtLS IN THE PRESENCE AND ABSENCE OF FSH AND LW, RESPECTIVELY Lactate (nmol/lOs cells) Incubation time (h)

Control -ADO

LH

FSH

+ADO

-ADO

+ADO

-ADO

+ADO

4

1223

421

2822

2022

2921

16k3

24

3522

12+3

121+7

30+2

10325

241t2

Granulosa cells (ZxlOs cells/well) were incubated in humidified air (37 C) for the indicated time in the absence or presence of FSH (100 n ml) or LH (100 nglml) with and without adenosine (ADO; 50 pM). Each value represents tf e mean k SEM, n = 6.

The results support the hypothesis that a relative lack of cofactors (i.e., ADP) common for both glycolysis and the respiratory chain causes the onadotropin-induced decrease in oxygen consumption and increase in ?actate accumulation in cumulus-oocyte complexes. Origin and putative release of ovarian cell adenosine Extracellular adenosine in the ovary may have different origins: The general circulation Ovarian cells Released as adenosine/AMP as a result of hypoxia a) b) Released as adenosine/AMP as a result of decreased levels of cellular ATP not due to hypoxia Released as CAMP/AMP cf

STEROfDS

54/S

November

1989

532

Billig et al: OVARIAN

ADENOSINE

The blood is the probable source for purine homeostasis and for purines required in tissue metabolism (21). However, it is not likely that the blood concentration of adenosine, under normal physiological conditions, would be the regulating factor for adenosine-related effects in the ovary since changes in adenosine levels in the blood do not correlate with ovarian functron (70). Increased ovarian cell adenosine due to hypoxia Adenosine release as a result of hypoxia is a well established mechanism. In blood vessels, adenosine exhibits the well-documented effect of vasodilation (71.72). Increase in adenosine is believed to be caused by decreased oxygen tension and, consequent1 , decreased ATP production. This will result in an egress of AMP and aII enosine from the tissue. The increase in extracellular adenosine stimulates adenosine receptors and dilates the vessels, thereby restoring the blood flow and oxygenation of the tissue (9,731. The high lactate concentration in follicular fluid (74) may intuitively indicate carbohydrate metabolism in low-oxygen tension. Behrman and co-workers (12,381 have suggested that the granulosa cells may be exposed to relatively high concentrations of adenosine caused by hypoxia, residing as they are in an avascular area of the follicle with the distance to oxygenating blood being relatively large. This suggestion is in analogy with adenosine action in vessels, although the ex lanations for the decreased oxygen tension required for increase of a8 enosine efflux differs. In the coronary vessels the explanation is based on changes in metabolism, and in the ovarian follicle it is based on morphology. If the ox gen tension is low in the follicle, then it is probably at the same low leve Y over many hours or even days. The low-oxygen tension is most likely not subject to sudden changes if it is ex lained by the distance to blood vessels and this distance is a function of Pollicular development and growth. If the lower oxygen tension is sustained over a period of time, as assumed in the follicle, then adenosine would not be subject to re-uptake, since the uptake is regulated by the intracellular metabolism. Based on these assumptions, the adenosine level is not likely to be increased in the follicle or newly formed corpus luteum due to low vascularization, even though the oxygen tension may be low, but sustained. Furthermore, the oxygen tension in follicular fluid has been reported to be as high as in venous blood (75). The high lactate production may reflect, instead, a specialized metabolism adapted to the substrate demands of the oocyte since incubated follicular cells do produce a considerable amount of lactate even in atmospheric oxygen tension (16). Increased ovarian cell adenosine due to decreased ATP A decrease in cellular ATP will increase AMP concentrations and, in the resence of S-nucleotidase. also adenosine (8). Conadotropins decreased E0th basal (39,53,54) and adenosine-au mented ATP levels (38,39,52,53) in prepubertal ovaries, follicular cells an 8 luteal cells, from young corpora

STEROIDS

54/5

November 1989

Biliig et al: OVARIAN

ADENOSINE

533

lutea. This decrease in ATP levels may be due to increased consumption of ATP, but perhaps also to a decreased re-phosphorylation (39). Both these mechanisms would increase AMP and lead to an increased release of adenosine/AMP. C clic AMP turnover also consumes ATP and produces AMP by adenylate cycYase and phosphodiesterase action, respectively. It is interesting that FSH decreases adenosine uptake, ATP levels, and oxygen consumption, and increases lactate accumulation (16,38,39,53,64,65), which supports a re-localization of adenosine in gonadotropin-stimulated follicular cells that would increase the extracellular adenosine concentration due to decreased uptake or increased release. Increased ovarian cell adenosine due to increased CAMP Egress of CAMP is well established in a multitude of tissues, including gonadotropin-stimulated gonads, and has been suggested to be carriermediated. Furthermore, CAMP as such is not subject to re-uptake through intact cell membranes before degradation to adenosine or its metabolites. The physiological significance of the marked outflow of CAMP has remained an enigma (76). In many tissues (8). including the ovary (77,78), extracellular CAMP can be degraded b the ectoenzymes phosphodiesterase and 5’-nucleotidase to AMP and adyenosine. Cyclic AMP added to the incubation medium also increased ATP levels in granulosa cells. The ATP increase was dependent on both phosphodiesterase activity and nucleoside transport since a phosphodiesterase inhibitor (Ro 20-1724) and dipyridamole both inhibited cAMP-induced ATP increases in the cells (Tables 2 and 3). These results suggest that extracellular cAMP is converted to AMP by phosphodiesterase and further metabolized to adenosine, which is then transported into the cell where it is phosphorylated to ATP. In this context, it is also of interest to note that FSH stimulates both extracellular and intracellular phosphodiesterase activity, thereby increasing both the CAMP production and its degradation to AMP within and between the ovarian cells (79,80). Also, in the testis, FSH stimulates phosphodiesterase activity (81). It could be concluded that CAMP is released in ovarian tissue and subsequently degraded extracellularly to adenosine. Therefore, an adenosine recirculation and reutilization may be proposed. Extracellular adenosine levels in the ovary The levels of purines have been reported to be hi h in porcine (82) follicular fluid (1.41 mM hypoxanthine) and in murine (83B follicular fluid (0.35-0.70 mM adenosine, 2-4 mM hypoxanthine). However, these values may be overestimations due to the sampling procedure. In human follicular fluid collected during surgery and with prompt addition of an adenosine deaminase inhibitor, the values were much lower (0.3 pM adenosine, 3.7 M hypoxanthine; H. Billig and H. Fredholm, unpublished results), and in 1 ovine follicular fluid no purines could be detected (84). Theoretical calculations, where the gonadotropin-induced cAMP efflux, intracellular ATP decrease, and adenosine uptake have been considered in rat granulosa

STEROIDS

54/5

November 1989

534

Billiget

ceils, indicate an extracellular about l-6 pM (85).

adenosine

al: OVARIAN

concentration

ADENOSINE

in the follicle

of

TABLE 2. EFFECT OF CAMP AND ADENOSINE ON ATP LEVELS IN GRANULOSA CELLS IN THE PRESENCE AND ABSENCE OF ADENOSINE TRANSPORT INHIBITOR

ATP (pmol/l05

Incubation time (min)

;; 180 360

CAMP

Control -DIP

+ DIP

54&Z 77f4 9123 131+4

59*2 71 f3 88+8 107+6

cells)

-DIP

ADO + DIP

602 2 7122 93+ 6 84fl 127+ 3 96f5 280+16137+1

-DIP

+ DIP

lOOf 3 58k 305fl6108fl2 387+20126&15 434218195+

1

7

Granulosa cells (2x10s cells/well) were incubated in humidified air (37 C) for the indicated time in the absence or presence of cAMP (50 PM) and adenosine (ADO, 50 pM) with and without dipyridamole samples.

(DIP, 10 pM). Each value represents the mean f SEM for triplicate

TABLE 3. EFFECT OF cAMP IN THE PRESENCE OF PHOSPHODIESTERASE INHIBITOR OR ADENOSINE TRANSPORT INHIBITOR ON ATP LEVELS IN GRANULOSA CELLS AIP (pmol/lOs

cells)

Control

DIP

Control CAMP

39 f 2 95 f 5

36 f 2 36 It 5

35 + 2 39 f 3

FSH FSH + CAMP

27 + 2 66 f 2

25 f 2 30 f 1

19 f 1 24 + 1

Ro 20-1724

Granulosa cells (2x10s cells/well) were incubated in humidified air (37 C) for 360 min in the absence or presence of cAMP (50 PM), FSH (100 ngml), adenosine transport inhibitor dipyridamole (DIP, 10 PM), and phosphodiesterase Inhibitor (Ro 20-1724, value represents the mean f SEM of triplicate samples.

STEROIDS

54/5

1 mM).

Each

November

1989

Billig et al: OVARIAN

ADENOSINE

535

A model for paracrineiautocrine action of adenosine in the ovary The discussion above on adenosine action in general and in ovarian cells in particular meets many of the requirements for a local and/or paracrine action and reveals that: Adenosine may be produced in and released from ovarian cells (i.e., the enzymes for adenosine formation are present in the same compartment as the substrate for adenosine formation). Adenosine may induce metabolic effects as a substrate (e.g., ATP increase, normalization of oxygen consumption, decreased lactate accumulation, increased substrate availability to CAMP formation, etc). Adenosine may induce metabolic effects as a receptor agonist (e.g., adenylate cyclase stimulation, cAMP production). Adenosine production, release, and uptake may be regulated within the ovary (i.e. regulation by gonadotropins). Adenosine concentrations in the extracellular space, deduced from theoretical calculations, are in the ran e for efficient uptake and receptor action and are not in conflict wit R estimated concentrations in other tissues. If these circumstantial pieces of evidence (derived from incubation experiments) for modulation of ovarian cell activity by adenosine prove to be correct also in vivo, then they favor the adenosine receptor-mediated local regulation rather than regulation via adenosine substrate availability. The net increase in ATP and CAMP due to adenosine only as substrate would require a net increase of ovarian adenosine, and not a relocation of a constant, or decreasing, adenosine content from the intracellular to the extracellular space in the ovarian compartments. Adenosine receptormediated effects on cellular metabolism do not require a net increase in adenosine, but just a metabolically induced relocation of adenosine from within the cell to the extracellular space to increase receptor agonist interactions. This does not imply that adenosine as substrate is of no importance. Adenosine as a substrate is a prerequisite for a number of cellular events but not a regulatory factor like the receptor-mediated events. A speculative and simplified model for paracrine/autocrine action in the ovary may be proposed based on the discussion above (Fig. 5): The adenosine is of ovarian origin, and onadotropin stimulation induces a relocation of adenosine from within til e cell to the extracellular space. The relocation is a result of a gonadotropin-induced decrease of cellular ATP levels and subsequent increase in AMP concentration (due to a relative decrease in the phosphorylation/dephosphorylation ratio of phosphorylated adenine nucleotides) and an increased efflux of CAMP. The CAMP is degraded to AMP by phosphodiesterase. Intracellular and extracellular AMP is further degraded to adenosine. Extracellular adenosine binds to and activates adenosine A2 receptors in an autocrine/paracrine manner. Adenosine receptor activation augments the adenylate cyclase response to gonadotropin and its functional parameters and indirectly increases the adenosine (AMPlcAMP) outflow by a short loop feedback mechanism.

STEROIDS

54/5

November 1989

536

Biiiig et al: OVARIAN

ADENOSlNE

x

Figure 5. A model for autocrinelparacrine action of adenosine in adjacent ovarian cells. Left panel depicts initiation of gonadotropic stimulation of a. Ranulosa cell. Mlddle panel depicts events later on during strmulation, and anel to the ng t represents termmatton of gonadotropin stimulation and reconstitution o P cells. Bottom s uare in each panel depicts the same cell but at different times after stimulation. tona 9 otropm stimulation (open squares, left panel) activates adenylate c ciase with subsequent CAMP outflow and decrease of cellular ATP. Adenosine (close (Y.tnangles) is subject to cellular uptake with subsequent ATP increase and adenosine receptor interaction. For further details, see text.

Adenosine is also subject to uptake into cells. However, the uptake diminished in gonadotropin-stimulated cells.

is

If the gonadotro in~imulation in every cell is note ual, e.g., if the receptor occupancy is di fperent due to submaximal levels o3 hormones, or the distribution of receptor is une ual in a group of cells (e.g., part of the corpus luteum, the follicle, or the w4;,ole follicle), then the adenosine uptake would be unequally distributed. The less gonadotropin-stimulated or the unstimulated cells would take up more adenosine than the stimulated ones. The less-stimulated cells would then increase their ATP levels and be more

STEROIDS

5415

November1989

Billiget

al: OVARIAN

ADENOSINE

537

responsive to a weak gonadotropin stimulation at a later stage. When the gonadotropin stimulation is terminated, the remaining augmented levels of extracellular adenosine are decreased by cellular uptake, and adenosine receptor activation is decreased (Fig. 5). This theoretical model of dual adenosine action as substrate and paracrine receptor agonist in the ovary would imply that the gonadotropin response is more e ually distributed in, for instance, the preovulatory follicle, than the gona a otropin receptor distribution would suggest (86,87). The model also delineates a possibility of prolonging the gonadotropin action when hormone levels fall, of mediating a faster response when hormone levels rise, and of equalizing variations in gonadotropin stimulation due to pulsatility or restraints in diffusion of the hormone. ACKNOWLEDGMENTS This work was supported by grants from the Swedish Medical Research council (nos. 27 and 7314), The Giiteborg Medical Society, the Swedish Society for Medical Research, and the Faculty of Medicine, University of Giiteborg. REFERENCES 1.

2. 3. 4. 5. 6.

7. 8.

9. 10.

STEROIDS

Tsafriri A (1988). Local nonsteroidal regulators of ovarian function. In: The Physiology of Reproduction (Knobil E and Niell JD, eds), Raven Press, New York, pp 527565. Londos C and Wolff J (1977). Two distinct adenosine-sensitive sites on adenylate cyclase. PROC NATL ACAD SCI USA 74:5482-5486. van Calker M, Miiller M, and Hamrecht B (1979). Adenosine regulates via two different types of receptors, the accumulation of cyclic AMP in cultured brain cells. J NEUROCHEM 33:999-1005. Ribeiro JA and Sebastiao AM (1986). Adenosine receptorsand calcium: Basis for proposing a third (A3) adenosine receptor. PROG NEUROBIOL 26: 179-209. Drury AN and Szent-Gyiirgyi A (1929). The physiological activity of adenine compounds with sepecial reference to their action upon the mammalian heart. J PHYSIOL 68:213-237. Daly JW (1983). Role of ATP and adenosine receptors in physiologic processes: Summary and prospectus. In: Ph siolog and Pharmacology of Adenosine Derivatives (Daly JW, Kuro d a Y, PI!.rllis JW, Shimiza H, and Ui M, eds), Raven Press, New York, pp 273-290. Sporn MB and Torado GJ (1980). Autocrine secretion and malignant transformation of cells. NEW ENG J MED 303:878-880. Arch JRS and Newsholme EA (1978). The control of the metabolism and the hormonal role of adenosine. In: Essays in Biochemistry y2a3mpell PN and Aldndge WN, eds), Vol. 14, Academic Press, pp 82Newby AC (1984). Adenosine and the concept of “retaliatory metabolites.” TIBS 9:42-44. Monaco L, Toscano MV, and Conti M (1984). Purine modulation of the hormonal response of the rat sertoli cell in culture. ENDOCRINOLOGY 115:1616-1624.

54/S

November 1989

538

Billig et al: OVARIAN

ADENOSINE

11.

Behrman HR, Hall AK, Gore SD, Dorflinger LJ, Luborsky JL, and Polan ML (1982). Paracrine regulation of the corpus luteum by purines, prostaglandins and peptides. In: Ret Adv Fertil Res, Part A: Developments in Reproductive Endocrinology, Alan R Liss, New York, pp 221238. 12. Behrman HR, Aten RF, Luborsky JL, Polan ML, Miller JGO, and Soodak LK (1986). Purines, prostaglandins and peptides: Nature and cellular mechanisms of action of local assist and assassinagents in the ovary. J ANIM SCI 62 (suppl2):14-24. 13. Baker HWG, Bremner WJ, Burger HG, deKretser DM, Dulmanis A, Eddie LW, Hudson B, Keogh EJ, Lee VWK, and Rennie GC (1976). Testicular control of follicle-stimulating hormone secretion. RES PROG HORM RES32:429-476. 14. Richards JS (1980). Maturation of ovarian follicles: Actions and interactions of pituitary and ovarian hormones on follicular cell differentiation. PHYSIOL REV 60:51-89. 15. Mita M, Price JM, and Hall PF (1982). Stimulation by follicle-stimulating hormone of synthesis of lactate by Sertoli cells from rat testis. ENDOCRINOLOGY 110: 1535-l 541. 16. Billig H, Hedin L, and Magnusson C (1983). Gonadotrophins stimulate lactate production by rat cumulus granulosa cells. ACTA ENDOCRINOL 103 : 562-566. 17. Jutte NHPM, Grootegoed JA, Rommerts FFG, and van der Molen HJ (1981). Exogenous lactate is essential for metabolic activities in isolated rat spermatocytes and spermatides. J REPROD FERTIL 62:399405. 18. Robinson Rand Fritz IB (1981). Metabolism of glucose by Sertoli cells in culture. BIOLREPROD 24:1032-1041. 19. Zeilmaker GH and Verhamme CMPM (1974). Observations on rat oocyte maturation in vitro: Morphology and energy requirements. BIOL REPROD 11: 145-152. 20. Stiles GL, Pierson G, Sunay 5, and Parsons WJ (1986). The rat testicular At adenosine receptor-adenylate cyclase system. ENDOCRINOLOGY 119:1845-1851. 21. Murray AW, Elliott DC, and Atkinson MR (1970). Nucleotide bios nthesis from preformed purines in mammalian cells: Regulatory met 1 anisms and biological significance. PROG NUCLEIC ACID RESMOL BIOL 10:87-l 19. 22. Berlin RD and Oliver JM (1973). Membrane transport of purine and pyrimidine bases and nucleosides in animal cells. In: international Review of Cytology, vol. 42 (Bourne GH and Danielli JF, eds), Academic Press, pp 287-336. 23. Londos C, Cooper DMF, and Wolff J (1980). Subclasses of external adenosine rece tors. PROC NATL ACAD SCI USA 77:2551-2554. 24. Londos C, Wol R J, and Cooper DMF (1983). Adenosine receptors and adenylate cyclase interactions. In: Regulatory Functions of Adenosine gB;?rne RM, Rail TW, and Rubio R, eds), Martinus Nijhoff Publishers, pp 25.

Schrader J, Rubio R, and Berne R (1973). Inhibition of slow action potentials of guinea pi atrial muscle by adenosine: A possible effect on Cal + influx. J MOL C9 LLCARDIOL 7:427-433.

STEROIDS

54/5

November 1989

Billig et al: OVARIAN

26. 27.

28.

29. 30.

31. 32.

33. 34.

35.

36.

37.

38.

39.

40.

41.

STEROIDS

ADENOSINE

539

Phillis JW and Wu PH (1981). The role of adenosine and its nucleotides in central synaptic transmission. PROG NEUROBIOL 16: 187-239. Sattin A and Rail TW (1970). The effect of adenosine and adenine nucleotides on the cyclic adenosine 3’5’-phosphate content of guinea pig cerebral cortex slices. MOL PHARMACOL 6: 13-23. Smellie FW, Davis CW, Daly JW, and Wells JN (1979). Alkylxanthines: Inhibition of adenosine-elicited accumulation of cyclic AMP in brain slices and of brain phosphodiesterase activit . LIFE SCI 24:2473-2482. Fredholm BB (1980). Are methylxanthine ef Yects due to antagonism of endogenous adenosine? TRENDS PHARMACOL SCI 1: 129-132. Williams M and Risley EA (1980). Biochemical characterization of putative central purinergic receptors by using 2-chloro(3H)adenosine. a stable analog of adenosine. PROC NATL ACAD SCI USA 77:68926896. Murphy KMM and Snyder SH (1981). Adenosine receptors in rat testes: Labeling with 3H-cyclohexyladenosine. LIFE SCI 28:917-920. Murphy KMM, Goodman RR, and Snyder SH (1983). Adenosine receptor localization in rat testis: Biochemical and autoradiographic evidence for association with spermatocytes. ENDOCRINOLOGY 113:1299-1305. Monaco L and Conti M (1986). Localization of adenosine receptors in rat testicular cells. BIOL REPROD 35:258-266. Eikvar L, Levy FO, Jutte NHPM, Cervenka J, Yoganathan T, and Hansson V (1985). Effects of adenosine analogs on glucagon-stimulated adenosine 3’,5’-monophosphate formation in Settoli cell cultures from immature rats. ENDOCRINOLOGY 117:488-491. Wolff J and Cook GH (1977). Activation of steroidogenesis and adenylate cyclase by adenosine in adrenal and Leydig tumor cells. J 8lOL CHEM 252:687-693. Rommerts FFG, Molenaar R, Hoogerbrugge JW, and van der Molen HJ (1984). Development of adenosine responsiveness after isolation of Leydig cells. BIOL REPROD 30:842-847. Polan ML, DeCherney AH, Haseltine FP, Mezer HC, and Behrman HR (1983). Adenosine amplifies follicle stimulating hormone action in granulosa cells and luteinizing hormone action in luteal cells of rat and human ovaries. J CLIN ENDOCRINOL METAB 56:288-294. Ohkawa R, Polan ML, and Behrman HR (1985). Adenosine differentially amplifies luteinizing hormone over follicle-stimulating hormone-mediated effects in acute cultures of rat granulosa cells. ENDOCRINOLOGY 117:248-254. Billig H and Rosberg S (1986). Gonadotropin depression of adenosine triphosphate levels and interaction with adenosine in rat granulosa cells. ENDOCRINOLOGY 118:645-652. Billi H, Thelander H, and Rosberg 5 (1988). Adenosine receptorme 8 iated effects by non-metabolizable adenosine analogs in preovulatory rat granulosa cells: A putative local regulatory role of adenosine in the ovary. ENDOCRINOLOGY 122:52-61. Preston SL, Parmer TG, and Behrman HR (1987). Adenosine reverses calcium-dependent inhibition of follicle-stimulating hormone action and induction of maturation in cumulus-enclosed rat oocytes. ENDOCRINOLOGY 120:1346-1353.

54/5

November 1989

540

42.

43.

44.

45. 46.

47.

48.

49. 50.

51. 52. 53. 54.

55.

Billiget

al: OVARIAN

ADENOSINE

Hall AK,. Preston SL, and Behrman HR (1981). Purine amplification of luteinrzrng hormone action of ovarian luteal cells. J BIOL CHEM 256: 10390-10398. Brennan TJ, Ohkawa R, Gore SD! and Behrman HR (1983). Adeninederived purines increase adenosme triphosphate (ATP) levels in the luteal cell: Evidence that cell levels of ATP may limit the stimulation of adenosine 3’S’-monophosphate accumulation by luteinizing hormone. ENDOCRINOLOGY 112:499-508. Behrman HR, Ohkawa R, Preston SL, and MacDonald GJ (1983). Transport and selective utilization of adenosine as a prosubstrate for luteinizing hormone-sensitive adenylate cyclase in the luteal cell. ENDOCRINOLOGY 113: 1132-1140. Sender Baum M and Ah&n K (1986). Adenosine potentiates the effects of LH and isoproterenol on luteal cells but not on isolated intact corpora lutea. ACTA PHYS SCAND 127:373-379. Billig H, Kumai A, and Rosberg S (1989). Adenosine receptor-mediated effects on adenylate cyclase activity in rat luteal tissue. A putative local regulatory role of adenosine in corpus luteum. BIOL REPROD 40:102110. Dix CJ, Habberfield AD, and Cooke BA (1985). Adenosine potentiates lutropin-stimulated cyclic AMP production and inhibits lutropininduced desensitization of adenylate cyclase in rat Leydig tumour cells. BIOCHEMISTRY 230:211-216. Londos C, Wolff J, and Cooper DMF (1981). Adenosine as a regulator of adenylate cyclase. In: Purinergic Receptors: Receptors and Recognition, Series f? (Burnstock G, ed), Vol. 12, Chapman and Hall, London, 289-323. Billi H and Rosberg S (1988). Evidence for A2 adenosine receptorme 8 iated effects on adenylate cyclase activity in rat ovarian membranes. MOLCELL ENDOCRINOL 56:205-210. Johanson C and Johanson V (1988). A technique for superfusion of isolated granulosa cells: dynamics of CAMP production and steroidogenesis in response to gonadotrophins. ACTA ENDOCRINOL 117:497-506. Behrman HR, Polan ML, Ohkawa R, Laufer N, Luborsky JL, Williams AT, and Gore SD (1983). Purine modulation of LH action in gonadal cells. J STEROID BlOCHEM 19:789-793. Sender Baum M and Billig H (1986). The ability of gonadotrophins to depress cellular ATP in rat luteal cells decreases with the age of the corpora lutea. ACTA PHYS SCAND 128:38A. Billig H and Magnusson C (1985). Gonadotropin-induced inhibition of oxygen consumption in rat oocytes complexes: Relief by adenosine. BIOL REPROD 33:890-898. Ah&n K, Beviz A, Hamberger L, and Lundholm L (1968). Mechanisms for the action of luteinizing hormone (LH) on the carbohydrate metabolism of the prepubertal rat ovary. ACTA PHYSIOL EXCERPTUM 73:23a-24a. Soodak LK MacDonald GJ, and Behrman HR (1988). Luteolysis is linked to luteinizmg hormone-induced depletion of adenosine triphosphate in vivo. ENDOCRINOLOGY 122:187-193.

STEROIDS

54/5

November 1989

Billig et al: OVARIAN

56. 57. 58.

59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70.

71. 72. 73.

ADENOSINE

541

Biggers JD, Whittingham DF, and Donahue RP (1967). The pattern of energy metabolism in the mouse oocyte and zygote. PROC NATL ACAD SCI USA 58:560-567. Kennedy JF and Donahue RP (1969). Human oocytes: Maturation in chemically defined media. SClENCE 164:1292-1293. Le Gac F, Attramadal H, Borrebaek B, Horn R, Froysa A, Tvermyr M, and Hansson V (1983). Effects of FSH, isoproterenol, and cyclic AMP on the production of lactate and pyruvate by cultured Sertoli cells. ARCH ANDROL 10: 149-I 54. Nilsson L (1974). Acute effects of gonadotrophins and prostaglandins on the metabolism of isolated ovarian follicles from PMSG-treated immature rats. ACTA ENDOCRINOL 77:540-558. Hillensjo T (1976). Oocyte maturation and glycolysis in isolated preovulatory follicles of PMS-injected immature rats. ACTA ENDOCRINOL 82:809-830. Tsafriri A, Lieberman ME, Ahren K, and Lindner H (1976). Dissociation between LH-induced aerobic glycolysis and oocyte maturation in cultured Graafian follicles of the rat. ACTA ENDOCRINOL 81:362-366. Hillier SG, Purohit A, and Reichert Jr LE (1985). Control of granulosa cell lactate production by follicle-stimulating hormone and androgen. ENDOCRINOLOGY 116: 1163-l 167. Hillensjo T, Hamberger L, and Ahren K (1973). Respiratory activity of oocytes isolated from ovarian follicles of the rat. ACTA ENDOCRINOL 78:731-739. Dekel N, Hultborn R, Hillensjo T, Hamberger L. and Kraicer P (1976). Effect of luteinizing hormone on respiration of the preovulatory cumulus oophorus of the rat. ENDOCRINOLOGY 98:498-504. Magnusson C and Hillensjo T (1981). Further studies on the gonadotrophin-induced inhibition of respiration in preovulatory rat cumulus oophorus. ACTA PHYS SCAND 113: 17-21. Belitzer WA (1936). Ueber die umgekehrte Pasteursche Reaktion. BlOCHEM 2 283:339-342. Crabtree HG (1929). Observations on the carbohydrate metabolism of tumours. BIOCHEM J 23:536-545. Ibsen KH (1961). The Crabtree effect: A review. CANCER RES 21:829841 Koobs DH (1972). Phosphate mediation of the Crabtree and Pasteur effects. SCIENCE 178: 127-133. desaanchez VC, Hernandez-Munoz R, Dias-Munoz M, Villalobos R, Glender W, Vidrio 5, Suarez J, and Yanez L (1983). Circadian variations of adenosine level in blood and liver and its possible physiological significance. LtFESCI 33:1057-f064. Berne RM (1964). Regulation of coronary blood flow. PHYSIOL REV 44: I-29. Fredholm BB and Sollevi A (1986). Cardiovascular effects of adenosine. CLIN PHYSlOL6:1-21. Bruns RF, Lu GH, and Pugsley TA (1987). Adenosine receptor subtypes: Binding studies. In: Adenosine Research (Gerlach E and Becker BF, eds), Springer-Verlag, Berlin, Heidelberg, pp 59-73.

STEROIDS54/S

November1989

542

74.

75. 76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

Billiget

al: OVARIAN

ADENOSINE

Zeilmaker GH and Verhamme CMPM (1977). Lactate concentrations in pre-ovulatory follicles of pro-oestrous rats before and after onset of ooc te maturation. ACTA ENDOCRINOL 86:380-383. Sha r.gr R, Kraicer PF, and Soferman N (1972). Gases and electrolytes of human follicular fluid. J REPROD FERTIL 28:335-340. Barber R and Butcher WR (1983). The egress of cyclic AMP from metazoan cells. In: Advances in Cyclic Nucleotide Research (Greengard P and Robison GA, eds), Raven Press, New York, pp 119-138. Rosberg 5, Selstam G, and Isaksson 0 (1973). Characterization of the metabolism of exogenous cyclic AMP by perfused rat heart and incubated prepubertal rat ovary. ACTA PHYS SCAND 94:522-535. Knecht M and Catt K (1981). Gonadotropin-releasing hormone: Re ulation of adenosine 3’,‘5-monophosphate in ovarian granulosa ccles. SCIENCE 214:1346-1348. Selstam G and Rosberg S (1976). Stimulatory effect of FSH in vitro on the extracellularly active cyclic AMP phosphodiesterase in the prepubettal rat ovary. ACTA ENDOCRINOL 81:563-573. Conti M, Kasson BG, and Hsueh AJW (1984). Hormonal regulation of 3’S’-adenosine monophosphate phosphodiesterase in cultured rat granulosa cells. ENDOCRINOLOGY 114:2361-2368. Conti M, Geremia R, Adamo 5, and Stefanini M (1981). Regulation of Sertoli cell cyclic adenosine 3’:S’monophosphate phosphodiesterase activity by follicle stimulating hormone and dibutyryl cyclic AMP. BIOCHEM BIOPHYS RES COMMUN 98: 1044-1050. PF, and Eppig JJ (1985). Downs SM, Coleman DL, Ward-Baile Hypoxanthine is the principal inhibitor o Y murine oocyte maturation in a low molecular weight fraction of porcine follicular fluid. PROC NATL ACAD SCI USA 82:454-458. Eppig JJ, Ward-Bailey PF, and Coleman DL (1985). Hypoxanthine and adenosine in murine ovarian follicular fluid: Concentrations and activity in maintaining oocyte meiotic arrest. BIOL REPROD 33:10411049. Eppig JJ and Downs SM (1987). The effect of hypoxanthine on mouse oocyte growth and development in vitro: Maintenance of meiotic arrest and gonadotropin-induced oocyte maturation. DEV BIOL 119:313-321. Billig H (1988). Adenosine and ovarian function. Studies on adenosine as substrate and receptor agonist in the rat ovary. (Thesis, ISBN 917900-439-3). University of Goteborg, Goteborg, pp l-46. Amsterdam A, Koch Y, Lieberman ME, and Lindner HR (1973). Distribution of binding sites for human chorionic gonadotropin in the preovulatory follicle of the rat. J CELL BIOL 67:894-900. Uilenbroek JTJ and Richards JS (1979). Ovarian follicular development during the rat estrous cycle: Gonadotropin receptors and follicular responsiveness. BIOL REPROD 20: 1159-l 165.

STEROIDS

54/5

November 1989