Effect of alterations in follicular steroidogenesis on the nuclear and cytoplasmic maturation of ovine oocytes

Transcription

1 /. Embryol. exp. Morph. 98, (1986) Ig7 Printed in Great Britain The Company of Biologists Limited 1986 Effect of alterations in follicular steroidogenesis on the nuclear and cytoplasmic maturation of ovine oocytes j. c. OSBORN Regional TVF Unit, St Mary's Hospital, Hathersage Road, Manchester M13 OJH, UK R. M. MOOR AND I. M. CROSBY AFRC Institute of Animal Physiology, Animal Research Station, 7 Huntingdon Road, Cambridge CB3 OJQ, UK SUMMARY The effects of inhibitors of follicular steroidogenesis on biochemical changes occurring in oocytes maturing in vitro were studied using radiolabelling and polyacrylamide gel electrophoresis. These effects were correlated with previously investigated developmental abnormalities induced by the same inhibitors. The most severe effects were generated by inhibition of 17ar-hydroxylase with the drug SU10603 which resulted in a greatly increased ratio of progesterone to testosterone and oestrogen. Such treatment halved the rate of meiotic maturation. Treated oocytes were analysed individually on SDS-PAGE gels and quantitative analysis showed that the drug had induced synthetic abnormalities even in those oocytes that resumed meiosis. This conclusion was confirmed by separation of oocyte proteins on two-dimensional gels. The effects of SU were reduced by delaying addition of the drug until 6 h after the beginning of maturation but were not alleviated by the addition of exogenous oestrogen to the culture medium. When oocytes from SU-treated follicles were transferred to inseminated, recipient ewes and recovered 24h later, two-dimensional electrophoresis again revealed abnormalities in their protein synthetic patterns. Almost total abolition of steroid secretion by aminoglutethimide (AG) had much less effect on oocyte protein synthesis, although the proportion of oocytes maturing was reduced from 65 % to %. The aromatase inhibitor, androstatriendione (AST) although eliminating follicular oestrogen secretion, had no effect on the rate of maturation and very little effect on protein synthesis. These results correlate well with the effects of steroid inhibitors on fertilization and early cleavage. INTRODUCTION The involvement of steroids in the maturation of mammalian oocytes is controversial. On the one hand, it has been proposed that steroids are not required for the preovulatory resumption of meiotic maturation in rat oocytes (Lieberman et al. 1976; Billig et al. 1983). By contrast, other workers have shown that steroids are important for oocytes of many species for the completion of maturational changes in both the nucleus and cytoplasm (e.g. Soupart, 1974; Thibault, 1977; McGaughey, 1977). A more specific demonstration of the involvement of steroids in sheep oocytes undergoing maturation in vitro has been obtained by selectively Key words: ovine oocytes, steroidogenesis, nuclear maturation, cytoplasmic maturation, follicle.

2 188 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY altering the profile of steroids secreted by the preovulatory follicle using specific inhibitors of follicular steroidogenesis (Moor, Polge & Willadsen, 1980). Their results showed clearly that alterations to the steroid profile during oocyte maturation induced intracellular changes in the oocyte which were expressed as gross abnormalities at fertilization. The incidence and severity of these aberrations reflected the type of steroid environment within the follicle during maturation. In these experiments, pronuclear development 24 h after transfer of the oocytes to the oviducts of inseminated recipients was used as a measure of the effects of the inhibitors. Although these developmental criteria provide an extremely sensitive biological assessment of the lesions induced in oocytes by the treatments, their origin is uncertain. Nevertheless, it is probable that these developmental abnormalities may either result from, or are associated with, anomalies in the normal maturation of the oocyte. It is well established that extensive reprogramming of protein synthesis occurs in mammalian oocytes during maturation (Golbus & Stein, 1976; Schultz & Wasserman, 1977; Van Blerkom & McGaughey, 1978; Moor, Osborn, Cran & Walters, 1981). However, although many of these changes occur spontaneously after the oocyte is released from the inhibitory influence of the follicle and undergoes GVBD, it is clear that the precise pattern of changes associated with the acquisition of full developmental competence requires follicle cell support (Crosby, Osborn & Moor, 1981). The aim of the present investigation is therefore to correlate the developmental effects on ovine oocytes of alterations in follicular steroidogenesis during maturation with specific biochemical changes within the oocyte. MATERIALS AND METHODS Preparation and culture of follicles Ovaries were obtained from Welsh Mountain ewes injected on day of the oestrous cycle with 10 mg FSH-P (Armour, Omaha, Nebraska) in three injections 48, 40 and 24h before slaughter. Intact, nonatretic follicles were dissected from the ovaries and cultured for 21 h using techniques and media described previously (Moor & Trounson, 1977). Cultured follicles were divided into seven treatment groups, outlined in Table 1, which varied with respect to the types of gonadotrophins, steroids and steroid enzyme inhibitors added to the basic culture medium. Gonadotrophins were added in the following standard combination: FSH (NIH-FSH-S12, Sjugml" 1 ), LH (NIH-LH-S18, 3/igml" 1 ) and prolactin (NIH-P-S9, 0-02/xgrnP 1 ). Steroid synthesis in follicles was blocked at one of three sites in the biosynthetic pathway (Fig. 1): (i) by inhibiting the conversion of cholesterol to pregnenolone using the 20or-cholesterol oxidase inhibitor, aminoglutethimide (AG, Elipten) at a concentration of 10~ 2 M (Kahnt & Neher, 1966), (ii) by inhibiting the 17ar-hydroxylase enzyme system with 7-chloro-3,4-dihydro-2(3-pyridyl)- l-(2h)-naphthalenone (SU10603) at a concentration of 10" 4 M (Chart, Sheppard, Mowles & Howie, 1962) or (iii) by inhibiting the aromatizing enzymes with l,4,6-androstatrien-3,17-dione (AST) at a concentration of 10~ 6 M. Steroid supplementation consisted of the addition in one treatment group of oestradiol-17/3 (1 /ig ml" 1 ) to media containing SU Inhibitors that were used in preliminary tests but not included in the main study because of incomplete enzyme inhibition were the aromatizing enzyme inhibitors 5ar-androstan-17j3-ol-3-one (dihydrotestosterone, A2570) and 4-hydroxyandrostene-3,17-dione.

3 Cholesterol precursors 22a,22-dihydroxycholesterol Steroids and oocyte maturation Cholesterol 1 20a--hydroxycholesterol 189 A^pregnenolone Progesterone 17-hydroxyprogesterone Dehydroepiandrosterone n-hydroxy-a'^-pregnenolone- Androstenedione- Testosterone Enzyme 1 20a-hydroxycholesterol dehydrogenase 2 17a'-hydroxylase 3 Aromatizing enzymes Estradiol-17/3 Inhibitor Aminoglutethimide (AG) SU ,4,6 androstatrien-3,17dione (AST) Fig. 1. Diagrammatic representation of steroidogenic pathways in the follicle. The numbers indicate the points of action of the inhibitory drugs used in this study. Measurement of steroids in the culture media The progesterone and total unconjugated oestrogen content of the medium after culture were measured using methods and antisera previously described (Moor, Hay, Mclntosh & Caldwell, 1973). The radioimmunoassay for testosterone was validated by Moor (1977). Electrophoretic analysis of oocyte proteins Cumulus-enclosed oocytes were labelled at 37 C for 3 h in 50 jul of incubation medium (Moor et al. 1981) containing SOOjuCiml" 1 of [ 35 S]methionine (specific activity >1000Cimmol" 1 ; Amersham). After incubation, the cumulus-oocyte complexes were washed and denuded of cumulus cells. The oocytes were then briefly washed in 10mM-Tris-HCl, ph7-4, collected individually in a small volume of Tris buffer, lyophilized and frozen at -70 C until required for electrophoresis. The quantity of labelled methionine incorporated into protein was measured by TCA precipitation (Moor et al. 1981). Labelled proteins were analysed in one dimension as described by Moor et al. (1981) or in two dimensions according to O'Farrell (1975), as described by Osborn & Moor (1983a). Fluorography was as described by Moor et al. (1981) following the methods of Bonner & Laskey (1974) and Laskey & Mills (1975). One-dimensional gels were scanned using a microdensitometer and analysed statistically as described by Moor et al. (1981). Transplantation and labelling of cultured oocytes In one experiment, cumulus-oocyte complexes were removed from follicles cultured for 24 h with either gonadotrophins alone or gonadotrophins supplemented with SU10603 and transferred to the oviducts of recipient ewes in oestrus (Moor & Trounson, 1977). Freshly ejaculated semen was introduced into each uterine horn of recipients immediately before transfer of oocytes. Eggs were recovered from the recipients 24 h after transfer and were labelled with [ 35 S]methionine prior to electrophoresis.

4 Treatment group No hormones Hormones only SU SU 6-21 SU + E2 AG AST Table 1. Outline of treatment groups and effects of inhibitors on steroidogenesis p o Supplements Gonadotrophins Inhibitor None FSH, LH, Prolactin FSH, LH, Prolactin FSH, LH, Prolactin FSH, LH, Prolactin FSH, LH, Prolactin FSH, LH, Prolactin None None SU10603 SU10603 SU oestradiol Aminoglutethimide 1,4,6-androstatrien- 3,17-dione Period of inhibition (h) None None No. follicles Steroid secretion (ng mg of tissue) Progesterone Testosterone Oestrogen 0-1 ± ± ± ± ± ± ± ± ± ±3-9 Not measured 5-0 ± ±80-1* 22-9 ± ± ± ± ± ±0-9 * AST crossreacted with testosterone antibody. Figure unreliable. The following supplements were added in different combinations to follicles in the seven treatment groups: (a) gonadotrophins (FSH, 5jUg; LH, 3//g; prolactin, 0-02jugml~ 1 medium); (b) steroid enzyme inhibitors (aminoglutethimide, 10~ 3 M; SU10603, 10~ 4 M; 1,4,6-androstatrien- 3,17-dione, 10~ 6 M) and (c) oestradiol at l^gml" 1. O O

5 Steroids and oocyte maturation 191 Morphological analysis of oocytes After culture, oocytes were fixed for 24h in ethanol/acetic acid (3:1) before staining with Lacmoid and examining using phase contrast microscopy. RESULTS Effects of enzyme inhibitors on follicular steroidogenesis Previous studies have shown that the addition of either AG or SU10603 to cultured follicles in vitro reduces the overall concentration of steroids in the follicular fluid but that the two inhibitors have different effects on the relative quantities of the major steroids secreted by the follicle (Moor et al. 1980). In the present study, the measurement of steroid secretion into the culture medium (Fig. 2; Table 1) confirmed the alterations in steroid metabolism and secretion induced by the two drugs. Inhibiting the conversion of cholesterol to pregnenolone with AG reduces the rate of secretion of all major classes of steroids synthesized by the follicle, whereas inhibition of the 17or-hydroxylase system with SU10603 markedly depresses the synthesis of both androgens and oestrogens (Table 1) but greatly increases the relative level of progesterone (Fig. 2). As expected, the addition of the aromatizing inhibitor, AST, at explantation had no apparent effect on the output of progesterone but markedly reduced oestrogen secretion and appeared to increase dramatically androgen production. This latter finding is o No hormones GTN GTN + AG 0-21 GTN + SU 0-21 GTN + SU 6-21 GTN + AST 0-21 Fig. 2. Graphical representation of the data in Table 1 showing the effects of gonadotrophins and inhibitors on the relative amounts of progesterone (black), testosterone (white) and oestrogen (hatched) secreted by follicles.

6 1 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY readily explained by the high crossreactivity of AST with our testosterone antibody. The effects of delaying the addition of SU10603 for 6 h is also shown in Fig. 2 and Table 1. In this case, the deleterious effects of the inhibitor on oestrogen secretion was considerably reduced and resulted in a more normal steroid profile. This result is consistent with the observations of Moor (1977) and shows that most of the oestrogen secreted by the follicle occurs during the first 6h after gonadotrophin stimulation. Steroid enzyme inhibitors and nuclear maturation We have previously shown that the pattern of protein synthesis observed in oocytes after culture is closely correlated with their meiotic status (Crosby et ah 1981; Osborn & Moor, 1983a). Thus, oocytes showing an unchanged or prematurational pattern of protein synthesis can be classified as being at the germinal vesicle stage whereas oocytes showing a changed or postmaturational pattern have undergone germinal vesicle breakdown (GVBD); the latter would include oocytes at metaphase I as well as metaphase II (see Fig. 4; Crosby etal. 1981, plate 1). On the basis of these findings, we have used the patterns of proteins observed in individual oocytes to assess the effects of enzyme inhibitors on the proportion of oocytes undergoing GVBD. The results of this analysis (Table 2) show that in the absence of inhibitors the addition of gonadotrophins increased the proportion of oocytes undergoing GVBD from 8% to 65%. By contrast, the additional presence of SU10603 reduced the percentage maturing to 29 %. This effect was partially overcome by delaying the addition of SU for 6 h but was not affected by the presence of oestrogen in the medium. The second inhibitor, AG, also reduced the number of oocytes undergoing GVBD, to %, whereas AST had no effect. That this method of assessing nuclear breakdown from patterns of protein synthesis is a valid procedure is shown in Table 3. In this experiment, the nuclear development of oocytes obtained from follicles cultured with either gonadotrophins alone or in the presence of SU from different times after the onset of culture was directly assessed from lacmoid-stained oocytes. The results show that GVBD is severely inhibited by the presence of SU from the beginning of culture and this effect is alleviated only if the addition of SU is delayed for 9 h. Steroid enzyme inhibitors and protein synthesis Previous work has shown that the incorporation of labelled methionine into TCA-insoluble material by oocytes from cultured follicles is significantly increased by the addition of LH or FSH to the culture medium and that this stimulatory effect is prevented by the presence of SU but not by AG (Moor et al. 1981; Osborn &Moor, 19836). By contrast, quantitative analysis of the effect on protein synthesis of selectively modifying steroid biosynthesis during oocyte maturation showed that the alterations in steroid secretion induced by both these inhibitors are associated with substantial changes in the patterns of protein synthesis by the maturing oocyte (Osborn & Moor, 19836). However, in all of these experiments, analyses were

7 No hormones Hormones only SU SU 6-21 SU + E 2 AG AST Steroids and oocyte maturation 193 Table 2. The effects of steroid inhibitors on maturation as assessed by visual examination of protein synthetic pattern Treatment group No. of oocytes % showing changed pattern Hormones were added to all but the first treatment group as described in Table 1. Table 3. The effects of the addition ofsu10603 at various times after the beginning of culture on nuclear maturation, assessed by lacmoid staining of whole oocytes Treatment group Total GV Promet MI/MII (%) Hormones only (61) SU0-24h (36) SU6-24h (22) SU9-24h (52) made using small groups of oocytes and the variability in synthetic activity in individual oocytes within each treatment group was not taken into account. That this variability is an important factor when quantitatively analysing changes in protein synthesis is apparent from our findings in the previous section of this paper (see Table 2). We have, therefore, re-examined the effects of gonadotrophins and steroid enzyme inhibitors on protein synthesis by analysing both the levels of incorporation of methionine and the patterns of proteins synthesized by individual oocytes cultured and treated in the same way as the groups. In this way, we have been able to make more valid comparisons between oocytes showing changed or 'postmaturationap patterns of protein synthesis and have also been able to assess the effects of both gonadotrophins and inhibitors on the synthetic properties of oocytes that have not undergone maturation. The effects of enzyme inhibitors on the incorporation of labelled methionine into TCA-insoluble material in oocytes showing either changed or unchanged patterns of protein synthesis are shown in Table 4. The results confirm that oocytes from follicles exposed to gonadotrophins alone incorporate significantly more methionine than those obtained from either untreated follicles or follicles exposed to gonadotrophins and SU when added at explantation. Unexpectedly, however, there was also a marginal effect (0-05 <P<0-1) of AG on the level of incorporation. This differs from our previous observations (Osborn & Moor, 1983b) but is consistent with the effects of AG on maturational changes shown in Table 4. It is important to note that in all of these groups there is no significant difference in the levels of incorporation between oocytes showing pre- and postmaturational

8 Treatment group No hormones Hormones only SU SU 6-21 SU + E2 AG AST Table 4. Effects of inhibitors on incorporation of [ 35 Sjmethionine into protein by oocytes Control ±1437 (34) ±1773 (23) ±1647 (13) ±1906 (12) ±22 (17) ±2047 (15) MI-MII (changed pattern) + Inhibitor 9678±1245(27) ±1908 (23) 6417 ±626 (9) 87 ±891 (34) ±16 (15) /-test P<0-05 N.S. P<0-05 N.S. N.S. GV (unchanged pattern) ±51 (64) ±7 (42) ±11 (28) ±689 (24) 9041 ±959 (25) ±1984 (7) Figures are expressed as N.S. not significant. counts min" 1 oocyte" 1 ± S.E.M. (n). O O CO O 50 z o 50 > Z o o 50 O

9 Steroids and oocyte maturation 195 patterns of protein synthesis. These results suggest that the increased incorporation of methionine observed during oocyte maturation is gonadotrophinregulated and is independent of nuclear and protein-synthetic changes. Since both SU and, perhaps, AG inhibit this increase in methionine incorporation, it is possible that steroids are involved in its regulation. It is also evident that the presence throughout culture of exogenous oestrogen does not overcome the suppressive effect of SU on methionine incorporation. On the other hand, it was found that the inhibitory effect was reduced by delaying the addition of the inhibitor for 6h. The addition of AST at explantation had no inhibitory effect on methionine incorporation. The data presented in Tables 2 and 3 show that steroid enzyme inhibitors can have a broadly inhibitory effect on nuclear maturation and the associated changes in protein synthesis. This conclusion was derived from visual assessment of protein synthetic profiles on one-dimensional SDS gels such as those shown in Fig. 4. In order to detect more subtle effects on the pattern of synthesis, fluorograms were scanned and the bands quantified. Fourteen bands were selected for statistical analysis of synthesis in six different treatment groups. Fig. 3 is a two-dimensional representation of the relationships between the treatment groups. The table of distances (Table 5) gives the distance between the centroids of each group. The diagram shows that oocytes from follicles exposed to SU differed most from untreated oocytes. Two groups of SU oocytes are included in the analysis: one group that showed unchanged, prematurational-type patterns and a second comprising oocytes that appeared to have undergone the maturational changes. Both of these groups are well separated (by 6 units) from both pre- and postmaturational controls. The AG- and AST-treated groups are also separated from the postmaturational control group, although this is not very well represented in the figure. The distance table shows that these groups are 4 units from the control group and 3-5 units from one another. The oocytes used for these groups were all visually assessed as being postmaturational and the analysis confirms that they are much closer to the postmaturational than the prematurational controls. The results of the canonical variate analysis of one-dimensional gels show that there are quantitative differences between the protein profiles of oocytes cultured in the presence of inhibitors of steroidogenesis and those of oocytes cultured with gonadotrophins alone. Furthermore, in the case of SU, these differences occur both in oocytes undergoing maturation and in those remaining at the dictyate stage. However, since each of the protein bands separated on one-dimensional gels may represent several different proteins, we have used two-dimensional gel electrophoresis in an attempt to detect steroid-dependent changes in the synthesis of specific proteins. The use of 2D gel electrophoresis for analysing oocyte proteins must, however, be approached with some caution. Difficulties in detecting and correctly interpreting changes in protein synthesis arise from the technical requirements of the system, which necessitate the pooling of large numbers of oocytes, and the variability in synthetic activity between individual oocytes. It is evident that the pooling of oocytes for 2D analysis is acceptable only if cellular

10 196 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY homogeneity can first be demonstrated. We have found that acceptable homogeneity can be obtained by first assessing the patterns of protein synthesis of individual oocytes on ID gels and then combining those oocytes showing either pre- or postmaturational patterns for subsequent analysis on 2D gels. The advantages of 6-- +SU (unchanged) /******^ + su (changed) 4_. No hormones -6-- Fig. 3. Canonical variate analysis of the effects of gonadotrophins and steroid inhibitors on protein synthesis by oocytes, separated by one-dimensional SDS-PAGE. The first two canonical variates are plotted. Each small dot marks a single oocyte and the large dots the centroids of the treatment groups. Table 5. Generalized distances (the Mahalanobis D statistic, see Rao (1952)) between the treatment groups shown in Fig. 3 No hormones Hormones only SU (unchanged) SU (changed) AG AST No Hormones hormones only 9-97 SU (unchanged) SU (changed) AG AST Hormones were added to all but the first treatment group as described in Table 1. The distances reflect the degree of difference between the patterns of protein synthesis.

11 Steroids and oocyte maturation 197 using this approach are clearly shown in Figs 4-7. The characteristic ID patterns of synthesis associated with oocytes at the dictyate stage and those undergoing meiotic maturation are shown in Fig. 4. Two-dimensional analyses of oocytes showing only these pre- or postmaturational patterns are presented in Figs 5 and 6, respectively. Not only do these profiles confirm that maturation is accompanied by major changes in the patterns of protein synthesis, they also allow the firm identification of stage-specific polypeptides. For example, polypeptides A and I are only observed in dictyate oocytes whereas polypeptides 5 and 7 are characteristic of oocytes undergoing maturation. The difficulties involved in interpreting 2D patterns obtained by combining a mixture of oocytes showing both pre- and postmaturational patterns are clearly illustrated in Fig. 7. As expected, the profile is of an intermediate type with polypeptides characteristic of both stages being present. It is clear, therefore, that comparisons between patterns that reflect differing proportions of dictyate and maturing oocytes are invalid. We have used this approach of prescreening oocytes on ID gels to analyse by 2D electrophoresis the effects of enzyme inhibitors on the patterns of proteins synthesized by dictyate and maturing oocytes. Figs 8-11 illustrate the effects of SU on the 2D protein profiles of oocytes cultured in intact follicles for 21 h. In each case, oocytes were prescreened on ID gels and only those showing postmaturational patterns were used for 2D analysis. The effects illustrated are, therefore, quite distinct from the major inhibitory effect of SU on germinal vesicle breakdown. Fig. 9 shows the effects of continuous exposure to SU throughout maturation, the arrows indicating the main differences from the control pattern shown in Fig. 8. Careful,comparison of these patterns suggests that the SU-treated oocytes have failed to undergo completely the maturational changes in polypeptide synthesis. The arrowed polypeptides are generally intermediate in intensity between the corresponding spots in treated and untreated control oocytes. Figs 10 and 11 show, respectively, the effects of delaying SU exposure for 6h and of attempting to negate the effects of SU by adding exogenous oestrogen. The results confirm those obtained from the ID and morphological studies, namely that delaying exposure to the drug for 6 h reduces but does not abolish its effects, and that added oestrogen has very little beneficial effect. Previous studies have shown that the exposure of follicles to SU during maturation leads to abnormalities at fertilization; even oocytes that have progressed to metaphase II are not fertilized successfully, failure of sperm to penetrate being a common occurrence (Moor et al. 1980). In order to detect any relationship between protein synthesis and these developmental abnormalities, oocytes from SU-treated follicles were transferred to ewes, inseminated in vivo and recovered after 24 h in utero for radiolabelling and electrophoresis. Figs 12 and 13 compare the patterns of polypeptide synthesis in these oocytes with those of control oocytes fertilized in the same way. Once again, oocytes were prescreened on ID gels before pooling but it was not possible to detect differences between individual oocytes, so all were combined for analysis. The arrows mark differences between the two patterns.

12 198 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY M r A B x10" Intact GV GVBD x10" 3 IEF \ \c 10 / 9-6 GVBD

13 Steroids and oocyte maturation IEF 199 x10' 3 SDS 14-3 J 5 Intact GV IEF SDS ? o GVBD Figs 4-7. Fluorographs of electrophoretic separations of labelled polypeptides synthesized by oocytes. Fig. 4 is a one-dimensional SDS gel of GV (A) and Mil (B) oocytes demonstrating the distinctly different patterns of synthesis. Figs 5, 6 are 2D separations of GV and Mil oocytes, respectively, previously screened on ID gels to ensure homogeneity. Fig. 7 shows the 2D pattern produced when a mixture of GV and Mil oocytes is separated.

14 200 Mr x10" 3 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY IEF

15 Steroids and oocyte maturation 201 M r xicr 3 IEF SDS SDS Figs The effects of SU10603 on oocyte protein synthesis. Fig. 8 shows the pattern of synthesis in control, untreated oocytes. SU was added to follicles in the other groups at the onset of culture (Fig. 9), after 6 h (Fig. 10) and in conjunction with oestrogen (Fig. 11).

16 202 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY IEF M r x10" 3 SDS SDS Figs 12, 13. Patterns of protein synthesis in 1-cell embryos 24 h after insemination in vivo. The embryos in Fig. 12 were from normally matured oocytes whereas those in Fig. 13 were from follicles exposed to SU during maturation.

17 Steroids and oocyte maturation 203 The two-dimensional patterns of polypeptide synthesis in oocytes from follicles matured in vitro in the presence of AG and AST are shown in Figs 14 and 15. In general, these patterns are very similar to those of control, in vitro -matured oocytes. x10" 3 IEF SDS SDS Figs 14,15. The effects of AG (Fig. 14) and AST (Fig. 15) on oocyte protein synthesis.

18 204 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY DISCUSSION Work on the regulation of oocyte maturation by steroids has been largely restricted to studying the inhibitory effects of the hormones on the progression of meiosis in vitro. This was, perhaps, partly due to the observation that oocytes could mature in follicles in vitro in the absence of steroidogenesis (Liebermann et al. 1976). McGaughey (1977) reported that oestrogen could inhibit the nuclear maturation of denuded pig oocytes and more recently it has been shown that testosterone synergizes with dbcamp (Rice & McGaughey, 1981; Racowsky, 1983) and cholera toxin (Schultz, Montgomery, Ward-Bailey & Eppig, 1983) in inhibiting GVBD and that oestrogen similarly potentiates the action of FSH (Eppig, Freter, Ward-Bailey & Schultz, 1983). However, few studies have examined the positive effects of steroids on maturation although McGaughey (1977) found that progesterone and oestrogen together reduced the number of chromosomal abnormalities in pig oocytes. Earlier work showed that steroids could improve the decondensation of the sperm head in human and rabbit oocytes (Soupart, 1974; Thibault, 1977). The work described in this report followed previous studies in this laboratory that had shown that inhibitors of particular steroidogenic enzymes induced specific abnormalities in oocyte maturation and fertilization (Moor et al. 1980). These experiments were designed to investigate these effects in more detail by examining endocrinological, morphological and biochemical changes in the oocyte and follicle, in order to identify the mechanisms susceptible to disruption by abnormal steroid signals. The processes chosen for particular attention were steroid secretion by the follicle, morphological changes in the oocyte nucleus as it progresses from the dictyate stage to metaphase II and changes in both the rate and patterns of protein synthesis by the oocyte. All of these have previously been studied in this laboratory and are well characterized both in vivo and in vitro, making it possible to identify readily any steroid-induced abnormalities. Three enzyme inhibitors were selected for use, two of which, SU10603 and AG, had been used in the earlier study; the third, AST, was chosen because it provided a means of selectively reducing oestrogen secretion without abolishing androgen production. The effects of these drugs on follicular steroid secretion were quantified by analysis of steroids present in the medium after culture, rather than in follicular fluid as previously. The results for SU and AG were largely confirmatory of the earlier study; AG almost totally abolishing secretion and SU producing a grossly distorted profile with progesterone accounting for 86 % of total production. When SU addition was delayed for 6h, its effect was less marked, supporting earlier studies that showed that most oestrogen secretion occurs in the first few hours of maturation. As expected, AST also greatly suppressed oestrogen production whilst having almost no effect on progesterone; unfortunately, AST crossreacted with the testosterone antibody and made the accurate measurement of testosterone impossible.

19 Steroids and oocyte maturation 205 Having established that the three inhibitors had major but quite distinct effects on the steroid environment of the oocyte, the next step was to relate these to abnormalities of maturation and fertilization. In each case, exposure of the follicle to the drug resulted in morphological and biochemical lesions in the maturing oocyte, although the degree of disruption varied according to the inhibitor used. The most marked abnormalities were induced by SU The presence of this inhibitor throughout maturation in vitro prevented GVBD and the associated changes in protein synthesis in 67 % of oocytes. Methionine incorporation was reduced by 40 % and protein synthetic patterns were different from control oocytes, both in those oocytes remaining at GV and those in which nuclear maturation had begun. It was, however, not possible to identify any polypeptides that were specifically induced or totally inhibited by the presence of SU. The patterns revealed by the 2D analysis were intermediate between GV and normal, mature patterns. Recent studies in this laboratory (Moor & Crosby, 1986) have revealed that the major maturation-associated changes in protein synthesis occur at GVBD although further changes take place during progression from MI to MIL It seems probable that the 'postmaturational' SU groups were in fact composed of a mixture of MI and Mil oocytes. The differences between the protein patterns of SU-treated and control oocytes after transfer and fertilization might also be attributed to some SU-treated oocytes being at MI. This hypothesis is supported by the previous finding that some oocytes were at MI after 24 h in vivo (Moor et al. 1980). The deleterious effects of SU could be reduced by delaying addition of the drug for 6h, thus allowing much more oestrogen secretion by the follicle, but inhibition of GVBD was still marked. Attempting to restore the steroid balance by adding exogenous oestrogen to the culture medium had very little beneficial effect. This was in contrast to the earlier observation that exogenous steroids could overcome the block at GVBD (Moor et al. 1980), though in the earlier study, testosterone and 17ar-hydroxyprogestins were added as well as oestrogens. The effects of AG were significantly less severe, even though half the oocytes remained at GV (compared with 35 % of controls) and methionine incorporation was again inhibited by 50%. The effects on protein synthesis, as measured by the canonical variate analysis, were less marked and there were no readily identifiable, consistent differences on the 2D gels. The aromatase inhibitor, AST, had the least effect of the three drugs tested. Germinal vesicle breakdown and methionine incorporation were unaffected and the effects on protein synthesis were similar to those of AG. The inability to detect major differences at the 2D level, even though the statistical analysis of the ID gels showed separated groups, emphasizes an important weakness of the 2D system, namely its failure to provide a quantitative result. Furthermore, the separating system used in this investigation resolves only those polypeptides with pis between 4 and 8. The use of alternative firstdimension gels would be necessary to analyse a greater range of molecules.

20 206 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY It is, of course, possible that some of the observations attributed in this study to selective inhibitory actions of the drugs on specific cellular mechanisms do in fact result from direct cytotoxic effects on the follicle and oocyte. While such nonspecific activity cannot be totally discounted, the following comments argue against this possibility. First, we have shown previously that the deleterious effects of both aminoglutethimide and SU on oocyte development can be considerably reduced by either delaying the addition of the inhibitors until the second, synthetic phase of oocyte maturation or by supplementing the culture medium from explantation with exogenous steroids (Moor et al. 1980). Second, morphological studies of the cumulus, granulosa and theca of follicles treated with these inhibitors have provided no evidence of cell damage (Hay & Moor, unpublished observations cited in Moor et al. 1980) while two-dimensional gel analysis of labelled cumulus polypeptides shows few differences in synthesis caused by the addition of enzyme inhibitors (Osborn, Moor & Crosby, unpublished observations). Third, oocytes undergoing germinal vesicle breakdown in the presence of SU undergo most of the associated changes in the patterns of protein synthesis indicating that protein synthesis is not nonspecifically affected by the drug, while oocytes from SU-treated follicles and cultured for 24 h in vivo resume meiosis (see also Moor et al. 1980) and show patterns of protein synthesis similar to those of control oocytes. We therefore conclude that the abnormalities observed in oocytes exposed to the inhibitors used in this study are the result of subtle changes in the oocyte caused by specific alterations in the steroid environment pertaining during maturation rather than a nonspecific depression of cellular metabolism. The results of this study reveal more about the relationships between the many events of maturation and the way in which they are controlled. Steroids appear to influence two independent sets of maturational events. The first consists of GVBD and the major changes in protein synthesis that are very tightly coupled to it. The second comprises the totally separate increase in methionine incorporation. Previous studies have demonstrated that not only are all these processes gonadotrophin-regulated but that they are primarily under the control of LH (Moor et al. 1981). A second group of maturational events is principally initiated by the action of FSH. These include the expansion and mucification of the cumulus cells and the associated decrease in intercellular communication between cumulus and oocyte (Moor etal. 1981). Although these have not been studied in this investigation, there is less evidence for the involvement of steroids in their regulation. For example, oocytes in which GVBD has been inhibited by SU still exhibit the normal morphological changes in their cumulus cells (authors' unpublished observations). Overall, these investigations provide an insight into the basis for the previous observations on the effects of steroid inhibitors on fertilization. As predicted, the gross distortion of the steroid profile by SU was found to be more damaging than the total shutdown of secretion induced by AG. The fact that addition of exogenous oestrogen failed to alleviate the effects of SU, coupled with the ob-

21 Steroids and oocyte maturation 207 servation that selectively decreasing oestrogen output with AST was the least disruptive of the treatments, suggests that it is the high level of progesterone generated by SU during the early part of maturation, rather than the depression of oestrogen, that is most damaging to the oocyte. Aminoglutethimide and SU10603 were generously donated by Dr D. H. M. Burley, CIBA Ltd, Horsham, Sussex. Purified gonadotrophins were a gift from the National Institute of Arthritis, Diabetes and Digestive and Metabolic Diseases, Bethesda, Maryland, USA. This work was supported by a Medical Research Council project grant to J. C. Osborn and R. M. Moor. REFERENCES BILLIG, H., HILLENSJO, T., TSAFRIRI, S., MAGNUSSON, C. &BRODIE, A. (1983). Nuclear maturation of follicle-enclosed rat oocytes during inhibition of steroidogenesis. Gamete Res. 8, BONNER, W. M. & LASKEY, R. A. (1974). Afilmdetection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem., CHART, J. J., SHEPPARD, J., MOWLES, T. & HOWIE, N. (1962). Inhibitors of adrenal corticosteroid 17cf-hydroxylation. Endocrinology 71, CROSBY, I. M., OSBORN, J. C. & MOOR, R. M. (1981). Follicle cell regulation of protein synthesis and developmental competence in sheep oocytes. /. Reprod. Fert. 62, EPPIG, J. J., FRETER, R. R., WARD-BAILEY, P. F. & SCHULTZ, R. M. (1983). Inhibition of oocyte maturation in the mouse: participation of camp, steroid hormones and a putative maturation-inhibitory factor. Devi Biol. 100, GOLBUS, M. S. & STEIN, M. P. (1976). Qualitative patterns of protein synthesis in the mouse oocyte. /. exp. Zool. 198, KAHNT, F. W. & NEHER, R. (1966). Uber die adrenale Steroid-Biosynthese in vitro. III. Selective Hemmung der Nebennierenrinden-Funktion. Helv. chim. Ada 49, LASKEY, R. A. & MILLS, A. D. (1975). Quantitative film detection of 3 H and 14 C in polyacrylamide gels by fluorography. Eur. J. Biochem. 56, LIEBERMAN, M. E., TSAFRIRI, A., BAUMINGER, S., COLLINS, W. P., AHREN, K. & LINDNER, H. R. (1976). Oocytic meiosis in cultured rat follicles during inhibition of steroidogenesis. Ada endocr. (Kbh). 83, MCGAUGHEY, R. W. (1977). The culture of pig oocytes in minimal medium and the influence of progesterone and oestradiol-17/3 on meiotic maturation. Endocrinology 100, MOOR, R. M. (1977). Sites of steroid production in ovine Graafian follicles in culture. /. Endocr. 73, MOOR, R. M. & CROSBY, I. M. (1986). Protein requirements for germinal vesicle breakdown in ovine oocytes. /. Embryol. exp. Morph. 94, MOOR, R. M., HAY, M. F., MCINTOSH, J. E. & CALDWELL, B. V. (1973). Effect of gonadotrophins on the production of steroids by sheep ovarian follicles cultured in vitro. J. Endocr. 58, MOOR, R. M., OSBORN, J. C, CRAN, D. G. & WALTERS, D. E. (1981). Selective effect of gonadotrophins on cell coupling, nuclear maturation and protein synthesis in mammalian oocytes. /. Embryol. exp. Morph. 61, MOOR, R. M., POLGE, C. & WILLADSEN, S. M. (1980). Effect of follicular steroids on the maturation and fertilization of mammalian oocytes. J. Embryol. exp. Morph. 56, MOOR, R. M. & TROUNSON, A. O. (1977). Hormonal and follicular factors affecting maturation of sheep oocytes in vitro and their subsequent developmental competence. /. Reprod. Fert. 49, O'FARRELL, P. H. (1975). High resolution, two dimensional electrophoresis of proteins. /. biol. Chem. 250, OSBORN, J. C. & MOOR, R. M. (1983a). Time-dependent effects of ar-amanitin on nuclear maturation and protein synthesis in mammalian oocytes. /. Embryol. exp. Morph. 73,

22 208 J. C. OSBORN, R. M. MOOR AND I. M. CROSBY OSBORN, J. C. & MOOR, R. M. (1983&). The role of steroid signals in the maturation of mammalian oocytes. /. Steroid Biochem. 19, RACOWSKY, C. (1983). Androgenic modulation of Cyclic Adenosine Monophosphate (camp)- dependent meiotic arrest. Biol. Reprod. 28, RICE, C. & MCGAUGHEY, R. W. (1981). Effect of testosterone and dibutyryl camp on the spontaneous maturation of pig oocytes. /. Reprod. Fert. 62, SCHULTZ, R. M., MONTGOMERY, R., WARD-BAILEY, P. F. & EPPIG, J. J. (1983). Regulation of oocyte maturation in the mouse: possible roles of intercellular communication, camp and testosterone. Devi Biol. 95, SCHULTZ, R. M. & WASSARMAN, P. M. (1977). Specific changes in the pattern of protein synthesis during meiotic maturation of mammalian oocytes in vitro. Proc. natn. Acad. Sci. U.S.A. 74, SOUPART, P. (1974). The need for capacitation of human sperm; Functional and ultrastructural observations. In Biology of Spermatozoa (ed. E. S. E. Hafez & C. Thibault), pp Basle: Karger. THIBAULT, C. (1977). Are fouicular maturation and oocyte maturation independent processes. /. Reprod. Fert. 51, VAN BLERKOM, J. & MCGAUGHEY, R. W. (1978). Molecular differentiation of the rabbit ovum. 1. During oocyte maturation in vivo and in vitro. Devi Biol. 63, (Accepted 12 August 1986)