Individual gonadotrophic hormones were used to examine the degree to which changes in intercellular coupling between somatic and germ cells initiate meiotic maturation, regulate protein synthesis or alter the ultrastructure of the ovine oocyte. Follicle Stimulating Hormone (FSH; 50 ng ml−1) suppressed intercellular coupling to the same extent as that observed during oocyte maturation in vivo. At low concentrations FSH did not, however, initiate resumption of meiosis. By contrast, luteinising hormone (LH; 100-500 ng ml−1) invariably initiated meiosis in oocytes cultured within the follicle but did not disrupt intercellular coupling. We conclude that nuclear maturation is not dependent upon the disruption of cell contact between the oocyte and the surrounding follicle cells.

The profile of proteins synthesized by untreated oocytes differed greatly from that of oocytes matured for 18 h in follicles treated with a combination of FSH and LH. Pretreatment of follicles with either FSH or LH at low concentrations resulted in the synthesis of an intermediate and more variable pattern of proteins. No correlation was found between changes in protein synthesis and the extent of junctional communication between the cumulus cells and oocyte.

Membrane vesiculation and lysosomal change in the transzonal processes are early structural changes associated with the suppression of intercellular coupling in oocytes. These changes in coupling probably result in the relocation of intracellular organelles in the final stages of oocyte maturation.

A variety of different cellular components in the oocyte undergo change after the resumption of meiosis. Apart from the characteristic structural changes in the nucleus, alterations also occur in cell metabolism, protein synthesis, membrane transport and in the localization of organelles within the cytoplasm (comprehensively reviewed by Masui & Clarke, 1979). To gain a clear appreciation of the functional aspects of the maturational process it is, however, necessary to move from the study of individual events to studies on interactions and interdependency between different compartments within the oocyte. The existing data, although limited, nevertheless already shows that while some changes are interdependent, others can occur in an apparently independent manner. Thus, nuclear maturation can be induced under certain’conditions without inducing all the changes in protein synthesis that characterize the fully mature oocyte (Thibault, 1977; Warnes, Moor & Johnson, 1977). Similarly, amino-acid fluxes across the oocyte membrane can be increased to those of the mature oocyte without inducing comparable changes in the nucleus or in protein synthesis (Moor & Smith, 1979).

Recent experiments have shown that low molecular-weight compounds pass freely between the cytoplasm of follicle cells and that of the oocyte (Gilula, Epstein & Beers, 1978; Moor, Smith & Dawson, 1980). These compounds enter the oocyte through permeable junctions formed at points of contact between the membranes of the somatic and germ cells. The term cell coupling refers in this paper to passage of molecules from the follicle cells into the oocyte through permeable junctions. This intercellular passage of small molecules is sharply reduced after the resumption of meiosis. A favoured current hypothesis suggests that the decline in intercellular coupling acts as the stimulus for the breakdown of the germinal vesicle (Dekel & Beers, 1978). In this paper we test this hypothesis and report on the relationship that exists between cell-coupling, nuclear maturation and protein synthesis in ovine oocytes.

Tissue preparation

Four separate experimental studies were undertaken (see Results). The follicles used in each study were obtained from the ovaries of sheep that had been injected on day 10–12 of the cycle with 1250 i.u. pregnant mare serum gonadotrophin (PMSG) and slaughtered 40 h later. Follicles that had responded to the exogenous gonadotrophin were dissected from the ovaries and cultured for 9-18 h using similar techniques and culture media to those described previously (Moor & Trounson, 1977).

At the end of the culture period follicles were either fixed for structural studies (expts 2 and 3) or were opened, washed to remove follicular fluid and then further dissected to remove the entire cumulus-oocyte cell complex (expts 1, 3 and 4). These preparative procedures were carried out at 37 °C in an incubation medium consisting of Dulbecco’s phosphate-buffered saline supplemented with bovine serum albumin (4 mg ml-1), pyruvate (0·36 mM), lactate (23·8 mM) and glucose (5·5 mM).

Measurement of intercellular coupling in oocytes

Intercellular coupling was quantified in oocytes obtained from follicles cultured for 15 h in medium containing purified hormones (see Table 1). Cumulus-oocyte complexes were placed in the wells of microtitre test plates and maintained at 37 °C for 1 h in incubation medium containing 36 μM methyl-[3H] choline chloride, 5–15 Ci/m-mole, Radiochemical Centre, Amersham. Incubations were terminated by transferring the cell complexes into unlabelled medium at 4 °C. Thereafter, oocytes were rapidly denuded of cumulus cells using finely graded pipettes. Individual denuded oocytes were transferred to coverslips, disrupted using 10 μL sodium dodecyl sulphate (SDS) buffer and counted with appropriate corrections for quenching and counting efficiency. Results are expressed for convenience as the concentration of labelled choline in the oocyte; it is, however, clear that most of the choline in oocytes is present as phosphocholine or lecithin (Moor et al., 1980). The validity of using [3H]choline for the measurement of intercellular coupling between cumulus cells and oocytes has previously been discussed in detail (Moor et al., 1980).

Table 1.

Entry of choline via permeable junctions and the associated degree of meiotic maturation in oocytes obtained from follicles cultured for 15 h in the presence of different gonadotrophins

Entry of choline via permeable junctions and the associated degree of meiotic maturation in oocytes obtained from follicles cultured for 15 h in the presence of different gonadotrophins
Entry of choline via permeable junctions and the associated degree of meiotic maturation in oocytes obtained from follicles cultured for 15 h in the presence of different gonadotrophins

Structural studies in oocytes

Groups of follicles (n = 8), removed 9, 12 or 15 h after culture in medium containing either NIH-FSH-S9 (100 μgml−1 or NIH-LH-S18 (100 μgml−1), were fixed in 4 % glutaraldehyde in collidine buffer (pH 7·2). The portion of follicle wall bearing the cumulus-oocyte complex was removed and processed for electron microscopy as described previously (Hay, Cran & Moor, 1976).

Hormonally induced changes in nuclear structure were examined in follicles cultured for 15 h in medium containing different gonadotrophins (see Table 1). The follicles were fixed and processed as for electron microscopy and 1 μm serial sections cut through the nucleus. Sections were stained in 1 % toludine blue in 1 % borax.

Electrophoretic separation of polypeptides in oocytes

Protein profiles were examined in oocytes obtained from follicles cultured for 18 h with different purified hormones. Groups of six to ten dissected and washed oocyte-cumulus complexes were placed in microtitre plates and incubated at 37 °C for 3 h in 50 μl incubation medium containing 1 mCi ml−1 of L-[35S] methionine (1000 Ci/m-mole; Radiochemical Centre, Amersham). After incubation, the oocyte-cumulus complexes were washed and denuded of cumulus cells. Groups of three to five denuded oocytes were then briefly washed in 0·01 M-Tris-HCl, pH 7·4, collected in a small volume of Tris buffer ( < 5 μl) and frozen at −75 °C until required for electrophoresis.

Labelled oocytes were prepared for electrophoresis by adding 25–30 μl SDS buffer (O’Farrell, 1975) and then freezing and thawing the samples twice. After heating for one minute at 100 °C, a 5 μl aliquot of each sample was used to determine incorporation of radioactivity into TCA-precipitable material as described by Van Blerkom (1978). A part of the remainder of each sample was applied to an 8–15 % linear gradient SDS polyacrylamide slab gel (14 cm wide, 10 cm long and 0–15 cm thick) such that an equal number of TCA-precipitable cpm (40000) was placed in each well.

Electrophoresis was carried out in the standard manner (Van Blerkom, 1978) using the discontinuous SDS-glycine-Tris buffer system of Laemmli (1970). Polypeptides were separated for 3 h at a constant current of 20 mA per gel. Fluorography was carried out using the technique of Bonner & Laskey (1974). Gels were dried on a Hoefer gel dryer and exposed to preflashed Kodak X-Omat H film at −70 °C for 48 h (Laskey & Mills, 1975). Molecular weight determinations were made using as standards:phosphorylase b (94K), albumin (67K), ovalbumin (43K), carbonic anhydrase (30K), trypsin inhibitor (20·1K) and lactalbumin (14·4K).

Sixteen polypeptide bands were selected for the quantitative analysis of changes in protein synthesis by oocytes exposed to different hormone treatments. Microdensitometer scans were made of each fluorogram and the relative amount of protein in each marker band was obtained from the densitometer tracings by determining the area under each peak using planimetric integration. The results are presented as the amount of protein in each band expressed as a percentage of the total amount of protein in the fluorogram.

Statistical analysis

An analysis was made of the pattern of proteins synthesized by oocytes after treatment with one of four hormonal regimes (expt 4). The data used for analysis consisted of 16 variates, being the percentage of protein in 16 individual marker bands, obtained from each of the six or seven groups of oocytes (five oocytes/ group) that comprised the separate experimental treatments. The appropriate mean values and standard errors for each of the individual marker bands are shown in Table 2. The associated ‘F’ ratios were calculated on each variate by univariate analysis.

Table 2.

Relative amount of labelled protein in each of 16 marker bands expressed as a percentage of the total protein synthesis in oocytes obtained from untreated and gonadotrophin-treated groups of follicles. Each value represents the mean of six or seven groups of oocytes (five oocytes / group) incubated in [35S] methionine for 3 h

Relative amount of labelled protein in each of 16 marker bands expressed as a percentage of the total protein synthesis in oocytes obtained from untreated and gonadotrophin-treated groups of follicles. Each value represents the mean of six or seven groups of oocytes (five oocytes / group) incubated in [35S] methionine for 3 h
Relative amount of labelled protein in each of 16 marker bands expressed as a percentage of the total protein synthesis in oocytes obtained from untreated and gonadotrophin-treated groups of follicles. Each value represents the mean of six or seven groups of oocytes (five oocytes / group) incubated in [35S] methionine for 3 h

The data was subjected to multivariate analysis for a more thorough statistical investigation. The purpose was to condense the original data, which in complete diagrammatic representation would be in 16-dimensional space, into as few dimensions as possible so that any underlying pattern could be recognized. The canonical variate analysis proved particularly useful in this regard (Rao, 1952). This analysis selects the axis along which the inter-group differences are greatest and then another axis, perpendicular to the first, which contains the maximum amount of the remaining inter-group variation and so on to as many variates as the data contains. This technique is, therefore, no more than a transformation of the multiple variates onto a scale which maximizes the discrimination between groups. The selection of the transformation takes account of the variability of variates and the co-variation between each pair of variates.

Since a high proportion of the inter-group variation in our data was accounted for by the first two canonical variates, a two-dimensional plot of the canonical variate values provided a fairly accurate picture of the relative positions of the treatment groups. A measure of the ‘distance’ between the treatment groups was calculated as the Mahalanobis D statistic, which is the standardized difference between the group mean values, on the variate taken collectively. It is these ‘distances’ which are approximated by the inter-group differences in the canonical variate plot.

Due to irregularities in the error structure of the variates we found that a logarithmic transformation of the data gave a more regular pattern in the results. The data in Table 2, is for purposes of clarity, nevertheless still presented in the original variable, whilst the canonical variate, analysis and the distance statistics have been calculated on the log transformed variable.

(1) Measurement of intercellular coupling in oocytes

Studies in this paper are based upon the finding that FSH suppresses cell coupling between oocytes and cumulus cells whilst LH is ineffective in this respect (Moor et al. 1980). Tritiated choline has been used to measure changes in the amount of intercellular coupling between these two cell types. This compound is rapidly taken up and phosporylated by follicle cells but cannot enter the oocyte except through permeable junctions (Moor et al., 1980). The rate at which labelled choline metabolite enters the oocyte therefore provides a functional measurement of the extent of cell coupling between the somatic and germ cells.

The object of the experiments in this first section has been to define the lowest concentration of FSH which consistently depresses intercellular passage of labelled marker into oocytes. To establish this minimal hormonal requirement, groups of follicles were cultured for 15 h in 50, 100 or 500 ng of FSH ml-1 culture medium. The mean concentration of labelled choline in oocytes from these groups of cultured follicles was 287 ± 82, 212 ±43 and 138 ± 34 μM respectively. These concentrations were statistically indistinguishable from those of 175 ±26 and 160 ± 46 μlM found in oocytes obtained (i) 15 h after the initiation of maturation in vivo (Moore et al., 1980) or (ii) after 15 h culture in follicles treated with a combination of high concentrations of FSH (5 μg ml−1) and LH (3 μg ml−1).

In a parallel series of experiments, follicles were cultured for 15 h in medium containing 100 or 500 ng LH ml−1. The mean concentration of choline in oocytes from these two treatments was 541 ± 78 and 675 ± 188 μM. These concentrations were similar to those observed in oocytes cultured in the absence of hormones (516 ±46 μM) but differed significantly from the concentrations in FSH-treated oocytes (P < 0 01).

These results demonstrate that FSH, even at low concentrations, selectively suppresses intercellular passage of low molecular weight compounds into oocytes. On the basis of these findings it was decided to use FSH and LH at concentrations of 100 ng ml−1 medium in all subsequent experiments described in this paper.

(2) Time-dependent effects of FSH and LH on the structure of intercellular junctions in oocytes

Intercellular transmission of compounds into oocytes occurs through slender extensions of the cumulus cells which traverse the zona pellucida and become enlarged on contact with the oolemma (Zamboni, 1972). Before gonadotrophin treatment, each process contains numerous 6 nm microfilaments orientated along its axis. Contact between the process and oolemma is maintained by prominent intermediate junctions (Fig. 1). Occasional non-lysosomal vesicular structures are found within the bulbous end of the process.

Fig. 1.

Terminal regions of cumulus cell processes (C) in an oocyte of a follicle cultured without hormonal supplementation. Contact between the two cell types is characterized by prominent intermediate junctions (arrows). Note tangentially sectioned processes (P) in the zona pellucida. × 23500; bar = 1 μm.

Fig. 1.

Terminal regions of cumulus cell processes (C) in an oocyte of a follicle cultured without hormonal supplementation. Contact between the two cell types is characterized by prominent intermediate junctions (arrows). Note tangentially sectioned processes (P) in the zona pellucida. × 23500; bar = 1 μm.

Addition of low levels of FSH to the culture medium resulted in the progressive degeneration of the cumulus cell processes. Within 9 h of FSH treatment lysosomes were observed within some of the processes (Fig. 5) and clusters of vesicles, some 50 nm in diameter, became visible on the surface of some processes (Fig. 2). The electron micrographs showed clearly that these vesicles were derived directly from the membrane of the cumulus cell process (Fig. 3). By 12 h after the addition of FSH most of the cumulus cell processes showed clear signs of degeneration and after 15 h no intact processes were present. The compact arrangement of cumulus cells at 9 h after FSH was similar to that observed in untreated oocytes. However, by 12 h all of the cumulus cells except for those abutting onto the zona pellucida (corona radiata) had undergone marked dispersal (Fig. 4). Dispersal of the corona cells and separation from the zona pellucida by 10–20 μm had occurred by 15 h after FSH treatment. Nevertheless, these cells remained connected to the zona by cytoplasmic extensions containing abundant microfilaments. At this stage, few processes could be detected in the zona pellucida at either the light or electron microscope level.

Fig. 2.

Cumulus cell process (C) after 9 h culture with FSH. It is surrounded by numerous small vesicles and intermediate junctions are absent. × 57000; bar = 0·5 μm.

Fig. 2.

Cumulus cell process (C) after 9 h culture with FSH. It is surrounded by numerous small vesicles and intermediate junctions are absent. × 57000; bar = 0·5 μm.

Fig. 3.

Part of a process as in Fig. 2. Continuity between a vesicle and the process plasma membrane is visible (arrow). × 70000; bar = 0·5 μm.

Fig. 3.

Part of a process as in Fig. 2. Continuity between a vesicle and the process plasma membrane is visible (arrow). × 70000; bar = 0·5 μm.

Fig. 4.

Part of cumulus after treatment with FSH for 12 h. Those cells abutting upon the zona pellucida (Z) are closely packed while the remainder are dispersed. × 2500; bar = 10 μm.

Fig. 4.

Part of cumulus after treatment with FSH for 12 h. Those cells abutting upon the zona pellucida (Z) are closely packed while the remainder are dispersed. × 2500; bar = 10 μm.

Fig. 5.

Cumulus cell process (C) after treatment with FSH for 9 h. The end of the process contains prominent lysosomes (L). Z, zona pellucida. × 20500; bar = 1 μm.

Fig. 5.

Cumulus cell process (C) after treatment with FSH for 9 h. The end of the process contains prominent lysosomes (L). Z, zona pellucida. × 20500; bar = 1 μm.

The addition of LH to the medium had no effect on the processes in most oocytes examined 9–12 h later. The terminal ends of a small proportion of processes, while retaining normal junctional contact with the oolemma, contained lysosome-like structures (Fig. 6). After 15 h exposure to LH, about half the oocytes showed signs of process degeneration, involving the appearance of lysosomes, loss of junctional integrity and the retraction of processes from the surface, of the oocyte. In addition, the corona cells were more dispersed than at 12 h, and although difficult to quantify, there appeared to be fewer processes traversing the zona than in the untreated group. It is noteworthy that the membrane vesicles which regularly formed after FSH treatment were seldom seen after treatment with LH.

Fig. 6.

Process (C) within the cortical cytoplasm of an oocyte after 12 h LH. Although a lysosome (L) is present, intermediate junctions (arrows) are intact. × 37000; bar = 0·5 μm.

Fig. 6.

Process (C) within the cortical cytoplasm of an oocyte after 12 h LH. Although a lysosome (L) is present, intermediate junctions (arrows) are intact. × 37000; bar = 0·5 μm.

An additional feature seen at 15 h after treatment with either FSH or LH was an increase in the surface area of the oocyte membrane resulting from invaginations which varied from simple clefts (Fig. 7) to complex anastomosing networks apparently isolating small regions of the oocyte cytoplasm (Fig. 8). These specializations of the oocyte membrane, which were more apparent after LH than FSH treatment, were readily penetrated with lanthanum which emphasized their discontinuous nature (Fig. 9). In addition, the oocyte cytoplasm adjacent to these areas changed structurally, containing a fine fibrillar material which was particularly evident in lanthanum-impregnated oocytes (Fig. 10).

Fig. 7.

Plasma membrane of an oocyte after 15 h LH. It has invaginated to form deep clefts (arrows). × 54500 ; bar = 0·5 μm.

Fig. 7.

Plasma membrane of an oocyte after 15 h LH. It has invaginated to form deep clefts (arrows). × 54500 ; bar = 0·5 μm.

Fig. 8.

Complex surface foldings after LH. Note the close apposition of cortical granules to the plasma membrane (arrows). × 12000; bar = 1 μm.

Fig. 8.

Complex surface foldings after LH. Note the close apposition of cortical granules to the plasma membrane (arrows). × 12000; bar = 1 μm.

Fig. 9.

Surface foldings (arrows) impregnated with lanthanum demonstrating their discontinuous nature. × 7000; bar = 1 /μm.

Fig. 9.

Surface foldings (arrows) impregnated with lanthanum demonstrating their discontinuous nature. × 7000; bar = 1 /μm.

Fig. 10.

Surface invaginations after lanthanum impregnation. The associated cytoplasm contains finely filamentous material (arrows). × 18000; bar = 1 μm.

Fig. 10.

Surface invaginations after lanthanum impregnation. The associated cytoplasm contains finely filamentous material (arrows). × 18000; bar = 1 μm.

We conclude that the electron microscopical observations provide a structural basis for the biochemical results which showed that FSH selectively depressed junctional passage of molecules between cumulus cells and oocytes. From the ultrastructure it is clear that the degenerative changes induced by FSH are not only more rapid but are also more widespread and involve more components of the processes than those induced by LH. By contrast, LH appears to act most effectively upon the nonjunctional regions of the oocyte membrane.

(3) Intercellular coupling and nuclear maturation in oocytes

In the first section of this paper we showed that the intercellular passage of small molecules between cumulus cells and oocytes is disrupted by low levels of FSH but is unaffected by LH. The purpose of the experiments in the present section was to relate these changes in cell coupling to those obtained in parallel studies on the morphology of the oocyte nucleus. The most important observations made in the study are summarized in Table 1. After 15 h culture with no hormonal additions, intercellular passage of molecules was high and the great majority of oocytes had intact germinal vesicles. Breakdown of cell coupling and of the germinal vesicles occurred in oocytes cultured in follicles exposed to high levels of FSH plus LH. That the two events are unrelated was, however, clearly seen in follicles cultured with single hormones. Despite the suppression of cellular coupling, FSH at low levels did not initiate germinal vesicle breakdown and nuclear maturation. By contrast, LH initiated nuclear maturation in oocytes in which the degree of intercellular coupling remained at the same high level as that recorded in untreated cells.

(4) Intercellular coupling and protein synthesis

There is almost no information available at present about the mechanisms which induce changes in protein synthesis in mammalian oocytes during maturation. It has, therefore, been of interest to examine the patterns of protein synthesis in oocytes in which selective changes have been induced in some of the other intracellular compartments.

The effect of hormones on the incorporation of labelled methionine into TCA-insoluble material in oocytes was examined in the first part of the study. Oocytes obtained from untreated follicles incorporated [35S]methionine at a mean level (±S.E.M.) of 6·44 ±0·85 f-mole methionine per oocyte h−1. The level of incorporation was significantly increased (P < 0·05) in oocytes obtained from follicles treated with FSH and LH to 8·62 ±0·44 f-mole methionine per oocyte h−1. There was, however, no significant difference in the level of incorporation of methionine between untreated oocytes and those obtained from follicles after treatment with low levels of FSH (6·30 ± 1·4 f-mole per oocyte h−1) or LH (4·91 ±0·89 f-mole per oocyte h−1).

Illustrated in Fig. 11A is the pattern of proteins synthesized by oocytes obtained from follicles cultured in the absence (untreated) or presence of hormones.

Fig. 11.

Fig. 11A. Fluorograph of [35S]methionine-labelled proteins synthesized by oocytes previously cultured for 18 h within follicles in the presence of the following gonadotrophins (i) untreated controls, (ii) FSH, 5 μg ml−1 plus LH, 3 μg ml−1, (iii) FSH; 100 ng ml−1 (iv) LH, 100 ng ml−1. Polypeptide separation was by SDS gradient acrylamide gel electrophoresis. The 16 marker bands selected for analysis are indicated and numbered sequentially from the low to high molecular weight regions.

Fig. 11 B. Densitometer tracers of fluorographs of [35S]methionine-labelled proteins from oocytes obtained from follicles after 18 h culture in medium devoid of gonadotrophins (— ; untreated controls) or in the presence of FSH 5µgml-1 plus LH 3 μg ml−1 (…… FSH/LH group). The numbering of the 16 selected marker bands corresponds to that shown in Fig. 11 a.

Fig. 11.

Fig. 11A. Fluorograph of [35S]methionine-labelled proteins synthesized by oocytes previously cultured for 18 h within follicles in the presence of the following gonadotrophins (i) untreated controls, (ii) FSH, 5 μg ml−1 plus LH, 3 μg ml−1, (iii) FSH; 100 ng ml−1 (iv) LH, 100 ng ml−1. Polypeptide separation was by SDS gradient acrylamide gel electrophoresis. The 16 marker bands selected for analysis are indicated and numbered sequentially from the low to high molecular weight regions.

Fig. 11 B. Densitometer tracers of fluorographs of [35S]methionine-labelled proteins from oocytes obtained from follicles after 18 h culture in medium devoid of gonadotrophins (— ; untreated controls) or in the presence of FSH 5µgml-1 plus LH 3 μg ml−1 (…… FSH/LH group). The numbering of the 16 selected marker bands corresponds to that shown in Fig. 11 a.

Protein synthesis in untreated and FSH/LH-treated oocytes

A visual comparison of the fluorograms from untreated oocytes and oocytes obtained after culture with high levels of gonadotrophin (FSH/LH group) showed numerous changes in their protein profiles. Sixteen prominent bands (identified in Fig. 11 A), which showed consistent changes in each of the six replicate experiments in the study, were selected as markers of protein change during maturation. The effect of hormones on these marker bands was analysed quantitatively by scanning densitometry. Typical densitometer tracings from fluorograph tracks of untreated and FSH/LH-treated oocytes are shown in Fig. 11B. The numerically labelled densitometer peaks correspond to the 16 bands selected as markers of protein change (see Fig. 11 A). It should be noted that in a few instances the resolution of the densitometer has been insufficient to separate a very dense band from an immediately adjacent faint band. Where this has occurred the area of the equivalent band combinations has been determined in both the control and treated oocyte groups.

The rates of protein change in untreated and FSH/LH-treated oocytes are shown in Table 2. When compared with the rates of change in untreated oocytes, addition of gonadotrophin significantly altered the proportion of label incorporated into a number of bands (P < 0 01). The gonadotrophin-induced changes involved a substantial increase of incorporation into bands 1, 4, 6, 10 and 15 and a substantial reduction of incorporation into bands 3, 5, 8 and 12.

The canonical variate analysis provided a more rigorous examination of the differences in synthesis between untreated and FSH/LH-treated oocytes. Since the first two canonical variates accounted for 91·2 % of the intergroup variation, the two-dimensional plot of these variates in Fig. 12 represents a fairly accurate picture of the relative position of the two groups. The table of ‘differences’ included in the figure shows clearly that the pattern of proteins in oocytes treated with FSH and LH differed very substantially from that in untreated oocytes.

Fig. 12.

Analysis of the effect of hormones on protein profiles in oocytes obtained from follicles cultured without gonadogrophins (untreated controls), with either FSH or LH alone or with both gonadotrophins (FSH/LH group). The plot represents the first two canonical variates for 130 oocytes in the four treatments. The table gives the standardized ‘distance’ calculated as the Mahalanobis D statistic (Rao, 1952), between the centroids (*) of the treatment groups.

Fig. 12.

Analysis of the effect of hormones on protein profiles in oocytes obtained from follicles cultured without gonadogrophins (untreated controls), with either FSH or LH alone or with both gonadotrophins (FSH/LH group). The plot represents the first two canonical variates for 130 oocytes in the four treatments. The table gives the standardized ‘distance’ calculated as the Mahalanobis D statistic (Rao, 1952), between the centroids (*) of the treatment groups.

Individual gonadotrophins and protein synthesis

The rates of protein change in oocytes obtained from follicles treated with low concentrations (100 ng ml−1) of either FSH or LH are shown in Table 2. Direct comparisons between individual bands did not show the same large differences as seen between untreated and FSH/LH treated oocytes. Nevertheless the ‘F’ ratio statistic indicated particularly marked heterogeneity between the four treatment groups in bands 1, 3, 4 and 6.

The canonical variate analysis showed that oocytes treated with either FSH or LH formed fairly compact and discrete clusters; there was no overlap with either the untreated or FSH/LH-treated groups (Fig. 12). Moreover, the table of ‘distances’ confirmed that both the FSH and LH groups occupied positions roughly equidistant from the untreated and FSH/LH-treated groups. The individual gonadotrophins nevertheless altered protein synthesis in substantially different ways, thereby accounting for the relatively large ‘distance’ between the centroids of the FSH and LH groups.

This study is based upon the observation that signals generated in the follicle cells are transmitted to the oocyte and form an integral part of the regulatory system during maturation. Steroids and cyclic AMP are the best known examples of these regulatory substances and probably influence the maturation of both amphibian and mammalian oocytes (Masui, 1967; Mailer & Krebs, 1977; Cho, Stern & Biggers, 1974; Moor, Polge & Willadsen, 1980). Amongst the other suggested regulators is a low molecular weight inhibitor protein, secreted by mammalian granulosa cells which possibly maintains oocytes in the dictyate state in non-activated follicles (Chang, 1955; Foote & Thibault, 1969; Tsafriri, Pomerantz & Channing, 1976). It has been our interest to examine the mechanisms involved in the transmission of such putative signal molecules into the oocyte (Moor & Smith, 1978; Moor et al., 1980).

The passage of small molecules through permeable junctions in most somatic cells is thought to regulate and synchronize cell function in many multicellular systems (reviewed by Loewenstein, 1979). However, it was not until the elegant experiments of Lawrence, Beers & Gilula (1978) that unequivocal evidence was obtained to show that hormonally induced signals are rapidly transmitted between somatic cells by contact-dependent mechanisms. That a similar signal transmission system may operate between follicle cells and oocytes is suggested by the high degree of coupling that exists between these heterologous cells (Gilula et al., 1978; Moor et al., 1980). The experiments described in this paper were designed to examine further this form of signal transmission and to relate changes in the transmission system with intracellular events in the oocyte. The finding that purified FSH, but not LH, selectively depresses intercellular coupling in oocytes provided the means by which the different relationships have been investigated. Our present results indicate that FSH levels as low as 50 ng ml−1 effectively suppress intercellular coupling whilst 10–20 times higher LH levels are ineffective in this regard. By using relatively low levels (100 ng ml−1) of the individual gonadotrophins, the problems of cross-contamination with other hormones has been minimized. The upper limit of contamination of NIH-LH-S18 with FSH is less than 5 % while the LH contamination of NIH-FSH-S9 is under 3 %.

The potential interpretative difficulties that could arise in the use of choline as a marker of intercellular coupling in oocytes have been exhaustively discussed in earlier papers (Moor et al., 1980). However, it is highly improbable that difficulties in interpretation have been experienced in this study because the marker uptake measurements have been confirmed in all cases by the results of parallel ultrastructural studies. The differential effects of FSH and LH on the cumulus cell processes and junctions are clearly discernable in the electron microscope. Exposure to FSH causes rapid and widespread changes both within the processes and on the membrane of the cell processes. By contrast LH induces a slow activation of lysosomes within some of the processes and does not induce membrane vesiculation. It is however interesting that LH is far more effective than FSH at inducing structural and functional changes in the non-junctional regions of the oocyte membrane (Moor & Smith, 1978).

Our ultrastructural studies indicate that the suppression of functional coupling between cumulus cells and oocytes by FSH involves two distinct morphological events, namely degeneration and withdrawal of processes. The degenerative process is characterized by the localized formation of lysosomal bodies in the ends of processes which initially remain in junctional contact with the oolemma. The disappearance of the intermediate junctions and the withdrawal of the processes occur about 6 h after the initiation of the early degenerative changes. It is possible that the disruption of these intermediate junctions is a necessary precondition for the dispersal of the corona-radiata cells and the associated retraction of their transzonal processes. The requirement for junctional disruption could explain why the large mass of cumulus cells undergo dispersal about 3 h before the corona cells.

Szollosi (1980) has postulated that the disruption of intercellular coupling in oocytes from the mouse, rabbit, pig and cow provides the stimulus for the intracellular migration of cortical granules to a position directly beneath the oolemma. In the sheep, cortical granules move towards the periphery of the oocyte during follicular growth (Cran, Moor & Hay, 1980). However, between 15 and 18 h after the resumption of meiosis, or 3 to 5 h after the suppression of cell coupling, the cortical granules in this species also became very closely aligned beneath the plasma membrane. While a causal relationship between the two events has still to be determined, it is possible that the disruption of cell coupling may initiate changes in the cortical cytoplasm of the oocyte which result in the redistribution of the organelles in this zone.

The protein results show that while both FSH and LH induce substantial changes in the profile of labelled polypeptides synthesized by oocytes, only FSH has the capacity to suppress intercellular coupling. Our overall conclusion from these results is that the majority of changes in protein synthesis are not regulated by changes within the permeable junctions. However, our experiments do not exclude the possibility that subtle variations in synthesis may be induced by the disruption of cell coupling during maturation. Difficulties in detecting and correctly interpreting minor changes in protein synthesis limit the degree to which this possibility can be fully explored. Variability in synthetic activity between different oocytes provides the first difficulty in detecting subtle intracellular changes. This variability arises from differences in the hormonal status of donor animals at ovariectomy, differences in size, degree of differentiation and extent of atresia of individual follicles and the conditions chosen for the culture and labelling of the oocytes. For example, the variability in biosynthetic activity in individual oocytes removed from follicles and cultured in an extra-follicular environment is directly influenced both by hormones in the medium and by cell interactions with surrounding cumulus cells (I. M. Crosby, J. C. Osborn & R. M. Moor, unpublished observations). It is therefore evident that pooling of oocytes for electrophoretic or other biochemical studies is acceptable only if cellular homogeneity can first be demonstrated. We have found that acceptable homogeneity (Fig. 12) can be obtained by using oocytes from intact cultured follicles which, at explanation, were non-atretic, 3·0 to 5·0 mm in diameter, and from animals that had been treated previously with a standardised hormonal regime. Additional factors which limit the capacity to identify subtle changes in synthesis are the technical requirements of the different protein separation systems. More homogenous oocytes are required for a valid analysis of both the basic and acid proteins using the highly sensitive two-dimensional method of polypeptide separation than in the one-dimensional system. Despite its reduced sensitivity, we have chosen the single dimensional form of separation because of its advantages in providing, from very small numbers of oocytes, quantitative data on both the acid and basic proteins. The problems of making statistical comparisons between protein profiles containing large numbers of different bands have been overcome by using the canonical variate analysis.

Dekel & Beers (1978) have proposed a model for oocyte regulation in mammals in which the dictyate state is maintained by elevated levels of cyclic AMP. This putative meiotic inhibitor, after synthesis in the follicle cells, is thought to enter the oocyte through contact-dependent mechanisms. The abolition of intercellular coupling between the cumulus cells and oocyte before ovulation would, according to the model, result in a fall in the intracellular cyclic nucleotide levels in the oocyte and the consequent resumption of meiosis. The use of low levels of individual gonadotrophins provides a direct means of testing this model. It is clear from the results presented in section 1 and 3 of the present paper that FSH suppresses intercellular coupling between the cumulus cells and oocytes without inducing nuclear maturation. Conversely, LH induces nuclear maturation in oocytes in which full intercellular coupling with the cumulus cells is retained. We are therefore unable to support that part of the model which postulates that the resumption of meiosis in mammalian oocytes is initiated by the disruption of cell contact between the cumulus cells and oocyte. We presently favour the view that during maturation gonadotrophins act primarily to alter the nature or intensity of the signals from the follicle cells. Changes in the intercellular transmission system appear to be secondary to changes in signal generation.

We thank Mrs Linda Musk and Mr Ian Crosby for valuable assistance in many aspects of this work. The gonadotrophins were generously donated by the National Institute of Arthritis. Metabolism and Digestive Diseases, National Institutes of Health, Bethesda, Maryland. One of us (J.C.O.) is indebted to the Medical Research Council for financial support.

Bonner
,
W. M.
&
Laskey
,
R. A.
(
1974
).
Film detection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels
.
Eur. J. Biochem
.
46
,
83
88
.
Chang
,
M. C.
(
1955
).
The maturation of rabbit oocytes in culture and their maturation, activation, fertilization and subsequent development in the fallopian tubes
.
J. exp. Zool
.
128
,
379
405
.
Cho
,
W. K.
,
Stern
,
S.
&
Biggers
,
J. D.
(
1974
).
Inhibitory effect of dibutyryl cAMP on mouse maturation in vitro
.
J. exp. Zool
.
187
,
383
386
.
Cran
,
D. G.
,
Moor
,
R. M.
&
Hay
,
M. F.
(
1980
).
Fine structure of the sheep oocyte during antral follicle development
.
J. Reprod. Fert
.
59
,
125
132
.
Dekel
,
N.
&
Beers
,
W. H.
(
1978
).
Rat oocyte maturation in vitro: relief of cyclic AMP inhibition by gonadotrophins
.
Proc. natn. Acad. Sci., U.S.A
.
75
,
4369
4373
.
Foote
,
W. D.
&
Thibault
,
C.
(
1969
).
Recherches experimentales sur la maturation in vitro des oocytes de truie et de veau
.
Annls Biol. anim. Biochim. Biophys
.
9
,
329
349
.
Gilula
,
N. B.
,
Epstein
,
M. L.
&
Beers
,
W. H.
(
1978
).
Cell-to-cell communication and ovulation. A study of the cumulus-oocyte complex
.
J. Cell Biol
.
78
,
58
75
.
Hay
,
M. F.
,
Cran
,
D. G.
&
Moor
,
R. M.
(
1976
).
Structural changes occurring during atresia in sheep ovarian follicles
.
Cell Tiss. Res
.
169
,
515
529
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature, Lond
.
227
,
680
685
.
Laskey
,
R. A.
&
Mills
,
A. D.
(
1975
).
Quantitative film detection of 3H and 14C in polyacrylamide gels by fluorography
.
Eur. J. Biochem
.
56
,
335
341
.
Lawrence
,
T. S.
,
Beers
,
W. H.
&
Gilula
,
N. B.
(
1978
).
Transmission of hormonal stimulation by cell-to-cell communication
.
Nature, Lond
.
272
,
501
506
.
Loewenstein
,
W. R.
(
1979
).
Junctional intercellular communication and the control of growth
.
Biochim biophys. Acta
560
,
1
65
.
Maller
,
J. L.
&
Krebs
,
E. G.
(
1977
).
Progesterone stimulated meiotic cell division in Xenopus oocytes
.
J. biol. Chem
.
252
,
1712
1718
.
Masui
,
Y.
(
1967
).
Relative roles of the pituitary follicle cells and progesterone in the induction of oocyte maturation in Rana pipiens
.
J. exp. Zool
.
166
,
365
375
.
Masui
,
Y.
&
Clarke
,
H. J.
(
1979
).
Oocyte maturation
.
Int. Rev. Cytol
.
57
,
186
282
.
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
,
319
335
.
Moor
,
R. M.
&
Smith
,
M. W.
(
1978
).
Amino acid uptake into sheep oocytes
.
J. Physiol. Lond
.
284
,
68
69P
.
Moor
,
R. M.
&
Smith
,
M. XV
. (
1979
).
Amino acid transport in mammalian oocytes
.
Expl Cell Res
.
119
,
333
341
.
Moor
,
R. M.
,
Smith
,
M. W.
&
Dawson
,
R. M. C.
(
1980
).
Measurement of intercellular coupling between oocytes and cumulus cells using intracellular markers
.
Expl Cell Res
.
126
,
15
29
.
Moor
,
R. M.
&
Trounson
,
A. O.
(
1977
).
Hormonal and follicular factors affecting maturation of sheep oocytes in vitro and their subsequent developmental capacity
.
J. Reprod. Fert
.
49
,
101
109
.
O’Farrell
,
P. H.
(
1975
).
High resolution two-dimensional electrophoresis of proteins
.
J. biol. Chem
.
250
,
4007
4021
.
Rao
,
C. R.
(
1952
).
Advanced Statistical Methods in Biometric Research
.
New York
:
John Wiley and Sons
.
Szollosi
,
D.
(
1980
).
Interaction between oocyte and follicle in vivo
.
Proceedings 9th Intern. Congr. Anim. Reprod. & A. I
., pp.
95
99
, Madrid.
Thibault
,
C.
(
1977
).
Are follicular maturation and oocyte maturation independent processes?
J. Reprod. Fert
.
51
,
1
15
.
Tsafriri
,
A.
,
Pomerantz
,
S. H.
&
Channing
,
C. P.
(
1976
).
Inhibition of oocyte maturation by porcine follicular fluid: partial characterization of the inhibitor
.
Biol. Reprod
.
14
,
511
516
.
Van Blerkom
,
J.
(
1978
).
Methods for the high resolution analysis of protein synthesis: applications to studies of early mammalian development
.
In Methods in Mammalian Reproduction
(ed.
J. C.
Daniel
), pp.
67
109
.
London
:
Academic Press
.
Warnes
,
G. M.
,
Moor
,
R. M.
&
Johnson
,
M. H.
(
1977
).
Changes in protein synthesis during maturation of sheep oocytes in vivo and in vitro
.
J. reprod. Fert
.
49
,
331
335
.
Zamboni
,
L.
(
1972
).
Comparative studies on the ultrastructure of mammalian oocytes
.
In Oogenesis
(ed.
J. D.
Biggers
&
A. W.
Schuetz
), pp.
5
45
.
Baltimore
:
University Press
.