| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a The Fels Institute for Cancer Research and Molecular Biology and Department of Biochemistry, Temple University School of Medicine, Philadelphia, Pennsylvania 19140
b The Jackson Laboratory, Bar Harbor, Maine 04609
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
As antral follicle development proceeds, mural granulosa cells and cumulus cells develop some distinguishing characteristics. For example, just before ovulation, gonadotropins stimulate cumulus cells to produce and secrete hyaluronic acid that disperses the cumulus cells and embeds them in a mucus-like matrix. This process is called cumulus expansion or mucification [1–3]. Mural granulosa cells do not undergo expansion but instead become luteal cells. Cumulus and mural granulosa cells also differ in the distribution of LH receptors [4–6] and mRNA coding for LH receptors (LHR) [7–9], steroidogenic capabilities [10–12], and mRNAs encoding cholesterol side-chain cleavage cytochrome P450 [13] and cytochrome P450 aromatase [9], insulin-like growth factor-1 mRNA [14], Müllerian inhibiting substance [15], lectin-binding [16], and other uncharacterized molecules [17]. At least some of the differences between cumulus cells and mural granulosa cells are probably due to interactions with oocytes. For example, oocytes promote the ability of granulosa cells to produce hyaluronic acid in response to hormonal stimulation [18–20]. In addition, oocytes suppress the expression of LHR mRNA [21]. However, the extent of the molecular and functional differences between cumulus cells and mural granulosa cells is not known. Here, high-resolution 2-dimensional PAGE (2D-PAGE) was used to obtain a global assessment of these differences and to establish a reference database for the analysis of granulosa cell differentiation in vitro.
Throughout follicular development, including antral follicle development, oocytes undergo dramatic growth and accumulate molecules essential for the continuation of meiosis and for supporting early embryogenesis. Complex communication between granulosa cells and oocytes involving paracrine factors and gap junction-mediated signals is essential for the development of both oocytes and granulosa cells [22, 23]. Culture systems for oocyte development likely depend upon the establishment of conditions that promote appropriate differentiation of oocyte-associated granulosa cells. The system established in this laboratory to study oocyte growth and development in vitro consists of oocyte-granulosa cell complexes isolated from the preantral follicles of 12-day-old mice by collagenase digestion. Almost all of the theca cells are removed, and the basal lamina encompassing the granulosa cells and oocytes is degraded [24–26]. The oocytes at this stage are in mid-growth phase and are incompetent to resume meiosis without further development. Between 200 and 300 oocyte-granulosa cell complexes are cultured attached to a collagen-impregnated membrane for 10 days, a period that spans the time required in vivo for antrum formation and acquisition of competence to resume meiosis, undergo fertilization, and support preimplantation development. The effect of FSH and insulin on oocyte development was assessed using this culture system. Neither FSH nor insulin treatment of cultured complexes appeared to have a beneficial affect upon oocyte growth, the percentage of oocytes acquiring competence to resume meiosis, or the percentage of oocytes competent to undergo fertilization and preimplantation development. However, treatment of complexes with both FSH and insulin produced an unexpected, highly deleterious effect on competence to undergo development from the 2-cell stage to the blastocyst [27]. This deleterious effect of FSH+insulin was correlated with an inappropriate differentiation of the oocyte-associated granulosa cells (OAGCs) during the early stages of the oocyte culture period. In these previous studies, the expression of LHR mRNA was used as a marker of the mural granulosa cell phenotype, because normal mouse cumulus cells do not express this transcript [27]. However, the overall impact of treatment with FSH, insulin, or both on the differentiation of the OAGCs was not known. Indeed, if conditions could be established that would promote the differentiation of these cells as true cumulus cells, it is reasonable to assume that this would have a beneficial effect on oocyte development. Therefore, high-resolution 2D-PAGE was used to compare the development of OAGCs cells treated with FSH, insulin, or both to that of cumulus cells and mural granulosa cells that developed in vivo.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals were bred and maintained in research colonies at The Jackson Laboratory under a controlled light cycle (16L:8D). They were killed by cervical dislocation. Oocyte-cumulus cell complexes and mural granulosa cells were isolated from the ovaries of 18-day-old (C57BL/6J x SJL)F1 mice as described previously [21]. Complexes and clumps of mural granulosa cells were collected individually with micropipettes and were washed manually by three serial passages through dishes of fresh medium to avoid contamination with other cell types. Proteins undergoing de novo synthesis were then metabolically labeled with [35S]methionine, as described below. Oocyte-granulosa cell complexes were isolated from the ovaries of 12-day-old mice and cultured for 6 days as described in detail elsewhere [27]. Four culture conditions were used: control (no insulin, no FSH), insulin alone (5 µg/ml), FSH alone (5 ng/ml), and insulin and FSH together. FSH (ovine FSH-20) was generously provided by the National Hormone and Pituitary Program of the NIDDK (Rockville, MD). It is important to note that this preparation of FSH is highly potent. According to the specifications provided by the NIDDK, it is 175-strength NIH-oFSH S1, or 4463 IU/mg. All groups were cultured in serum-free medium (Waymouth medium MB752/1) supplemented with 0.23 mM pyruvic acid, 50 mg/L streptomycin sulfate, 75 mg/L penicillin G (Sigma Chemical Co., St. Louis, MO), 5 µg/ml transferrin, 5 ng/ml selenium (Collaborative Research, Inc., Bedford, MA), and 3 mg/ml BSA (crystallized; ICN Biochemicals, Aurora, OH). After the 6-day culture period, the oocyte-granulosa complexes were dislodged from the collagen-impregnated membrane, upon which the complexes were grown, by sharply jolting the membrane insert with a snap of a finger against the side of the membrane and then labeled as described below. We refer to the granulosa cells of these cultured complexes as oocyte-associated granulosa cells (OAGCs) because their state of differentiation as cumulus cells or mural granulosa cells is unknown.
Cell Labeling and Lysis
Cells and complexes were incubated with 1 mCi/ml [35S]methionine (1175 Ci/mmol; Dupont NEN Research Products, Boston, MA) in BSA-free Whitten's medium [28] for 3 h. Cumulus cells and OAGCs were stripped from oocytes by repeatedly drawing the complexes in and out of a Pasteur pipette until the oocytes were denuded. The oocytes were then collected and discarded, and somatic cells were pelleted using an Eppendorf (Hamburg, Germany) microfuge. For each gel, cumulus cells and OAGCs from approximately 150 complexes and an equivalent volume of mural granulosa cells were lysed in 30 µl of hot dSDS solution and heated as described previously [29–31]. After treatment with DNase and RNase, the lysates were frozen and lyophilized and redissolved in an equal volume of sample buffer as previously described [29–31].
Two-Dimensional Protein Gel Electrophoresis and Analysis
Approximately 1 x 106 or 2 x 106 dpm (decays per minute) of labeled protein from each sample was applied to individual isoelectric focusing gels for 2D-PAGE as previously described [29, 32, 33]. High-resolution 2D-PAGE was performed in the Gel Laboratory at Cold Spring Harbor Laboratory (Cold Spring Harbor, NY). After electrophoresis, gels were dried and imaged by phosphorimaging along with calibration chips as previously described [31]. Protein spots were then detected, quantified, and matched using the PDQuest software package (Protein Database Inc., Huntington Station, NY).
Two gels from independently prepared samples were obtained for each of the following cell types: cumulus cells and mural granulosa cells isolated from 18-day-old mice; OAGCs from oocyte-granulosa cell complexes cultured without insulin or FSH; OAGCs cultured with FSH; and OAGCs cultured with insulin+FSH. Four gels from four independent samples were obtained from OAGCs cultured with insulin, for a total of 14 gels in the matchset. Three exposures were obtained for all gels, and the different exposures were merged into a single gel image file to avoid limitations in quantification due to saturation of the phosphorimaging screen by the most intense spots. Spot detection was performed after background subtraction and gel calibration as previously reported [32–35]. Spot intensity data from the merged gel image files for all proteins in all gels were exported into a Microsoft Excel (Redmond, WA) spreadsheet for further analysis. Protein spots with a maximum spot intensity of at least 10 dpm were initially selected for analysis. Spot intensity values were then normalized according to the total number of counts contained in all of the spots.
Spots that differed reproducibly by at least 2-fold between certain cell types or culture conditions were then identified and grouped into spot sets (Table 1). In generating the spot sets, proteins were included only if they differed by at least 2-fold in all pair-wise comparisons for all available gels for the two cell types being compared. Thus, spots had to differ by at least 2-fold in at least four separate gel-gel comparisons in order to be included in the spot sets. This stringent criterion avoided the inclusion in spot sets of proteins that did not differ reproducibly, which if included would result in inaccurate measures of sample relatedness. This method of generating spot sets is commonly used to identify proteins that change reproducibly (e.g., [34–36]). Set graphs, which depict the expression of each spot set as a whole among the various culture conditions, were generated as follows. For each member spot within a set, spot intensity values were reexpressed as a fraction of the highest intensity value measured in the group of 14 gels. The average expression of all the gels for each cell type and culture condition was then calculated for each spot, and the averages of all the spots in the set were used to generate the set graphs. By reexpressing spot intensity values as a fraction of the most intense value in the set of gels for each spot, the overall expression patterns derived for each spot set were not dominated by the most intense spots but instead reflected the overall pattern of change for the entire set of spots among the various cell types and culture conditions.
|
A standard gel map was generated using the PDQuest software (Fig. 1). A standard gel map is a synthetic gel image that is created from one of the matchset member gels, which initially contains most of the spots detected in the matchset and thus can be easily matched to all of the member gels. To this gel image are added any additional spots detected in any of the other gels within the matchset. The standard gel map then serves as a tool that allows the expression of any spot to be followed through the entire matchset, and as a convenient reference gel image on which to indicate the positions of spots of interest, such as members of particular spot sets or individual annotated spots that have important and interesting characteristics. The standard image can also be used as a link to future experimental matchsets, so that data from multiple experiments can be analyzed together.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mural versus Cumulus Gene Expression Patterns
Mural granulosa cells from small antral follicles and cumulus cells were expected to be somewhat similar in their overall gene expression patterns, particularly since the mural granulosa cells at this stage have not differentiated to express LHR. However, the two cell types were also expected to exhibit a number of differences in their gene expression patterns associated with their distinct cellular phenotypes. To evaluate the overall degree of similarity between the two cell types and to identify specific protein spots on our 2D gels that correlated with cell type, two sets of proteins were created that differed reproducibly by at least 2-fold between the cumulus and mural granulosa cells isolated directly from 18-day-old mice and labeled in vitro (Figs. 1 and 2). Figure 1 shows the locations of spots in these two sets within a standard gel map. One set (designated MUR) was synthesized at least 2-fold more highly in mural cells than in cumulus cells, and included 22 spots with an average rate of synthesis in mural cells ranging from 6 ppm to 1204 ppm (median = 99 ppm). On average, MUR proteins were synthesized about 4-fold more highly in mural cells than in the cumulus cells (Fig. 2A). The MUR set included one protein (#1405) that migrated at the position expected for tropomyosin 1 (Figs. 1 and 2B). Whether this spot indeed corresponds to tropomyosin 1, however, remains to be verified. The other set of differentially synthesized proteins (designated CUM) included 94 proteins that were more highly synthesized in cumulus cells than in the mural cells, and these had an average level of synthesis in cumulus cells ranging from about 6 ppm to 332 ppm (median = 45 ppm). On average, these proteins were synthesized nearly 6-fold more highly in cumulus cells than in mural granulosa cells (Fig. 2A). The two sets of proteins, combined, encompassed a total of only 116 proteins, which represents approximately 10.5% of the total number of proteins analyzed. Only a comparatively small number of proteins differed reproducibly by 10-fold or more (Table 1) between cumulus and mural cells. Thus, as expected, the cumulus and mural cells were somewhat similar to each other, but they exhibited a number of reproducible differences in their protein synthesis patterns.
|
Effect of Culturing Without FSH or Insulin
To evaluate the effect of culture on the differentiation of OAGCs, the pattern of proteins synthesized in these cells cultured with no FSH or insulin (control cultures) was compared to the pattern of proteins synthesized in freshly isolated cumulus cells and mural granulosa cells. The effect of culture on synthesis of MUR and CUM proteins was evaluated (Fig. 2). Three of the MUR proteins exhibited increased rates of synthesis in OAGCs as compared with cumulus cells; the rates of synthesis for two of these proteins are compared in Figure 2B. One of these proteins (#4402) was increased in synthesis under all culture conditions. As a group, the MUR proteins were synthesized, on average, more highly (67%) in the OAGCs cultured without insulin or FSH than in cumulus cells (Fig. 2A). The average rate of synthesis for the CUM proteins as a group was, by contrast, nearly 2-fold less in cultured OAGCs as compared with cumulus cells (Fig. 2A); and when examined individually, 26 of the 94 CUM proteins were synthesized at least 2-fold less in OAGCs cultured without insulin or FSH (Fig. 2C). These data indicate that the predominant effect of simply culturing the OAGCs, without the addition of FSH or insulin, was reduced synthesis of about one quarter of the proteins that exhibited greater rates of synthesis in cumulus cells than in mural granulosa cells, so that development of the cumulus cell phenotype appeared to be partially retarded. Along with this retardation, a small fraction (3 of 22) of proteins that were more highly synthesized in mural granulosa cells became elevated in the OAGCs, indicative of partial development of a mural granulosa cell phenotype (Fig. 2C). The net result was an intermediate, or ambiguous, phenotype for the cultured OAGCs.
Beyond these effects on the CUM and MUR proteins, cultured OAGCs reproducibly exhibited at least 2-fold reductions in rates of synthesis, compared to cumulus cells, of an additional 55 proteins, for a total of 81 proteins (spot set designated VITRO1) exhibiting reduced rates of synthesis compared to those of cumulus cells (Fig. 3A). The average rate of synthesis of these proteins was reduced by about 6-fold in OAGCs cultured without FSH or insulin added to the medium. Only one protein was reproducibly reduced by more than 10-fold. The suppression in rates of synthesis of this many proteins (7.4% of those analyzed) indicates that, in the absence of FSH or insulin, culture adversely affects the production of the cumulus cell phenotype, and with regard to these VITRO1 proteins, produces a cellular phenotype distinct from either the mural granulosa cell or the cumulus cell phenotype. OAGCs also reproducibly exhibited increased rates of synthesis of 38 proteins in addition to the MUR proteins, for a total of 41 proteins (spot set designated VITRO2) with increased rates of synthesis in OAGCs relative to cumulus cells (Fig. 3A). Only one protein was reproducibly elevated by 10-fold or more. The average rate of synthesis for the VITRO2 group of proteins was somewhat increased in mural granulosa cells versus cumulus cells, even though only three proteins were included in the MUR set. This again indicated a partial differentiation of OAGCs toward the mural cell phenotype that accompanies the suppression in synthesis of many additional proteins.
|
Effect of Addition of FSH to the Culture Medium
The addition of FSH to the culture medium has been used in an attempt to improve the developmental competence of oocytes obtained from our culture system [27]. Therefore, the effect of FSH on the protein synthesis pattern in OAGCs was investigated. Inclusion of FSH in the culture medium had a dramatic effect on the pattern of proteins synthesized by OAGCs. First, of the 81 VITRO1 proteins repressed in OAGCs cultured in control medium, 70 were synthesized at nearly normal rates (i.e., < 2-fold different in comparison to cumulus cells) in OAGCs cultured with FSH, and one other was actually elevated to a rate of synthesis more than 2-fold above that seen in cumulus cells (Fig. 3B). This elevation to near cumulus cell rates of synthesis for most of the VITRO1 proteins was evident in the average rate of synthesis for this group (Fig. 3A), which was only 33% less in FSH-treated OAGCs than in cumulus cells. Likewise, the rates of synthesis of CUM proteins were also largely elevated to near the rates observed for cumulus cells (Fig. 2A). Only 5 CUM proteins remained repressed and only 9 CUM proteins were elevated above cumulus cell synthetic rates in OAGCs treated with FSH (Fig. 2C). The rates of synthesis of 27 of the 46 VITRO2 proteins were also brought to differ by less than 2-fold from the rates of synthesis observed in cumulus cells (Fig. 3B). The average rate of synthesis of the other 14 VITRO2 proteins remained nearly 5-fold greater in FSH-treated OAGCs than in cumulus cells (Fig. 3A).
In addition to these effects, the rates of synthesis of a number of other proteins were reproducibly elevated above cumulus cell rates by FSH treatment of OAGCs, so that a total of 63 proteins were synthesized at least 2-fold more highly in FSH-treated OAGCs than in cumulus cells (designated spot set FSH1). The average rate of synthesis for these proteins was about four times greater in FSH-treated OAGCs than in cumulus cells (Fig. 4A). Only one protein in this set was reproducibly elevated by at least 10-fold. This set of proteins included 3 MUR proteins (Fig. 2C). Another group of 33 proteins (designated FSH2) was reproducibly repressed in OAGCs cultured with FSH relative to cumulus cells, 9 of which were also reduced in OAGCs cultured without FSH (Fig. 4A). One protein was reproducibly repressed by at least 10-fold. This protein was also a member of the CUM set. Overall, it appears that FSH was able to elevate to normal cumulus cell values the rates of synthesis of about 88% of the proteins that were reduced by culturing in control medium. For the FSH1 and FSH2 groups combined, about 8.7% of the proteins analyzed were altered by at least 2-fold in rates of synthesis in the FSH-treated cells relative to cumulus cells. These data indicate that the general effect of including FSH in the culture medium was to elevate to near cumulus cell values the rates of synthesis of many, but not all, of the proteins reduced by culturing in control medium, and to reduce to near cumulus cell rates the rates of synthesis of many of the proteins elevated by culturing without FSH or insulin, but to concurrently increase the rates of synthesis of many other proteins above those observed in cumulus cells and suppress the rates of synthesis of a number of other proteins. Overall, the protein synthesis pattern of FSH-treated OAGCs was slightly more like the cumulus cell pattern than was the pattern for OAGCs cultured in control medium (8.7% versus 11.1%). Thus, FSH treatment appeared to elicit a phenotype that was intermediate between the normal mural and cumulus cell phenotypes but distinct from the phenotype manifested by cells cultured in control.
|
Effect of Addition of Insulin to the Culture Medium
In our previous studies we also attempted to use insulin in the culture medium to improve oocyte quality [27]. Therefore, the effect of insulin on the protein synthesis pattern of OAGCs was investigated, and, as with FSH treatment, there were dramatic effects. Of the 81 VITRO1 proteins showing reduced rates of synthesis in OAGCs cultured without FSH or insulin, all but 2 were elevated to rates of synthesis less than 2-fold different from those observed in cumulus cells (Fig. 3B). The average rate of synthesis of the VITRO1 proteins was only about 20% lower in insulin-treated OAGCs than in cumulus cells (Fig. 3A).
Although the majority of the VITRO1 proteins were elevated to rates of synthesis near those observed for cumulus cells, other proteins were increased to rates of synthesis above those observed for cumulus cells, with a total of 51 proteins (spot set designated INS1) being reproducibly increased by at least 2-fold in rates of synthesis by insulin treatment (Fig. 5A). These spots included 6 that were reproducibly elevated by at least 10-fold. The INS1 set included only one of the MUR spots (#4402). The average rate of synthesis of the INS1 proteins was similar between mural granulosa cells and cumulus cells (Fig. 5A). The INS1 set included only 11 of the 41 VITRO2 spots and 15 of the FSH1 spots (Fig. 3B). Thus, most of the VITRO2 proteins were reduced to near cumulus cell rates of synthesis by insulin treatment. About one quarter of the FSH1 proteins elevated by FSH treatment were increased by insulin treatment as well, and the average rate of synthesis for the FSH1 proteins was elevated by about 2-fold in insulin-treated cells as compared with cumulus cells (Fig. 4B). As a group, the INS1 proteins tended to be aberrantly elevated by insulin treatment, but less so with FSH treatment (Fig. 5, A and B). Of the 33 FSH2 proteins suppressed by FSH treatment, all but 3 differed by less than 2-fold in rates of synthesis from cumulus cells in the insulin-treated OAGCs (Fig. 4B); and the FSH2 proteins were more highly synthesized on average in insulin-treated OAGCs (Fig. 4A), indicating specific effects of the two hormones. Only 2 of the CUM proteins remained below cumulus cell rates of synthesis in the insulin-treated OAGCs, and in fact 6 of the CUM proteins were elevated above cumulus cell rates of synthesis (Fig. 2C). A small set of 7 proteins (INS2) were synthesized less highly in insulin-treated OAGCs than in cumulus cells. These proteins on average exhibited lower rates of synthesis under all culture conditions, but the majority were reduced by 2-fold or more only in insulin-treated OAGCs (Fig. 5, A and B). These data indicate that the predominant effect of including insulin in the culture medium was to increase the rates of synthesis of many VITRO1 proteins to cumulus cell rates, but also to increase significantly the rates of synthesis of many other proteins above the rates observed in cumulus cells by an average of more than 9-fold. Overall, 2-fold or greater differences in rates of synthesis were observed for only 5.1% of the proteins analyzed when insulin-treated OAGCs are compared with cumulus cells, indicating that the insulin-treated OAGCs, like FSH-treated OAGCs, were more like cumulus cells than OAGCs cultured in control medium with neither hormone added. In fact, the insulin-treated OAGCs exhibited the protein synthesis pattern most like that of cumulus cells. However, while the beneficial effects of insulin included some responses not obtained with FSH treatment, they were accompanied by other disruptions in the normal gene expression pattern. As with FSH treatment, the phenotype obtained with insulin treatment appeared to be intermediate between the mural and cumulus phenotype, but distinct from that observed for control cultures or cultures treated with FSH alone.
|
Effect of Addition of FSH+Insulin to the Culture Medium
The above data indicate that either FSH or insulin alone can bring the rates of synthesis of many proteins in cultured OAGCs to near the rates seen in cumulus cells, and that the two hormones have many effects in common on the protein synthesis pattern of OAGCs but hormone-specific effects as well. In our previous studies, we attempted to use simultaneous treatment with FSH and insulin to improve oocyte quality, but this combination paradoxically had a serious negative effect on oocyte developmental competence [27]. We therefore asked whether this dramatic decrease in oocyte developmental competence might be the result of a small number of discrete alterations in gene expression as compared with the other culture conditions, or whether there was a more extensive alteration in molecular phenotype. Inclusion of both insulin and FSH in the culture medium produced a more extensive deviation from the cumulus cell pattern of protein synthesis than the addition of either hormone alone.
Of the 81 VITRO1 proteins repressed by culturing OAGCs in the absence of either FSH or insulin, all but 11 were brought to near the rates of synthesis observed for cumulus cells (Fig. 3B). Of the 41 proteins elevated by culture in control medium without added FSH or insulin (VITRO2), 13 were elevated in cultures given FSH+insulin, and 7 were elevated under all culture conditions (Fig. 3B).
Beyond the elevation of rates of synthesis of the majority of VITRO1 to near cumulus cell rates, the combination of FSH+insulin resulted in 2-fold or greater increases in rates of synthesis of many additional proteins, with a total of 85 proteins exhibiting reproducible 2-fold or greater increases (designated FI1). Only about half of the FI1 proteins were elevated by 2-fold or more in any other culture condition, indicating that many of these alterations were specific effects of the combination of FSH+insulin (Fig. 6B). The average rate of synthesis of the FI1 proteins was about six times greater in OAGCs treated with both hormones than in cumulus cells, about 43% greater than in OAGCs treated with insulin alone, and about two times greater than in OAGCs treated with FSH alone (Fig. 6A). Two proteins were elevated reproducibly by 10-fold or more, one of which was also elevated by about 10-fold in FSH-treated OAGCs. As a group, these proteins were synthesized on average about 50% more highly in mural granulosa cells than in cumulus cells (Fig. 6A). The average rate of synthesis of the MUR proteins was greater in OAGCs treated with insulin+FSH than in cumulus cells (Fig. 2A). Six of the MUR spots were elevated by at least 2-fold in the FSH+insulin-treated OAGCs, more than were affected by either hormone alone; and three of these were elevated only in cells treated with FSH+insulin (Fig. 6B). Five of the CUM proteins were also elevated by the combination of hormones, and three of these were only so affected by treatment with FSH+insulin (Fig. 6B). Fifteen CUM proteins were repressed by at least 2-fold, and four of these were repressed only in OAGCs treated with FSH+insulin (Figs. 2C and 6B).
|
Not only were many proteins elevated by the combination of FSH and insulin, but a set of 55 proteins (FI2) were reproducibly synthesized at reduced rates in OAGCs treated with FSH+insulin, relative to cumulus cells (Fig. 6A). These proteins were synthesized most highly in cumulus cells. These same proteins were only moderately reduced in FSH-treated cultures and in insulin-treated OAGCs, as compared with cumulus cells (Fig. 6A). Four proteins in the FI2 set were reproducibly reduced by at least 10-fold, but only one of these was reduced by at least 10-fold in FSH-treated OAGCs (Table 1). These observations indicate that neither FSH nor insulin alone led to repression of synthesis of these proteins. Even so, the combination of FSH and insulin resulted in more than a 5-fold lower rate of synthesis overall for this group of proteins than in cumulus cells, about one third the rate of synthesis observed in FSH-treated OAGCs, and a more than 3-fold lower rate of synthesis than in OAGCs treated with insulin (Fig. 6A).
Interestingly, only 19 of the INS1 proteins and only 24 of the FSH1 proteins were elevated in OAGCs treated with both hormones (Figs. 4B and 5B). This indicates that many of the proteins increased in OAGCs by treatment with either hormone alone were not overexpressed when the two hormones were used in combination, but that many additional proteins not increased by either hormone alone were overexpressed when OAGCs were treated with FSH+insulin. The overall effect of using FSH+insulin, therefore, was a pronounced elevation in rates of synthesis of many proteins, combined with an abrogation of many of the beneficial effects (i.e., increase to the level of synthesis in cumulus cells) of either hormone alone and an elimination of some of the other disruptive effects of either hormone alone.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It should be noted that because 2D-PAGE is superbly capable of resolving posttranslationally modified forms of proteins from one another, the apparent differences in protein synthetic rates revealed by 2D-PAGE can reflect a combination of differential rates of polypeptide synthesis and differences in posttranslational modifications, e.g., phosphorylation or glycosylation. However, whether individual observed differences result from differential transcription, translation, or posttranslational modification, the relative intensities of individual spots that are resolved are indicative of the overall molecular phenotypes of the cells being compared. Consequently, the ability to examine differences in expression of approximately 1100 separate protein spots provides an opportunity for a global assessment of differences in phenotype that arise due to different experimental conditions or that exist between different types of cells.
One result to come from this analysis is the observation of a number of reproducible quantitative differences between cumulus cells and the mural granulosa cells taken from the small antral follicles of 18-day-old mice. Mural granulosa cells from small antral follicles do not express LHR [21]. Nevertheless, the protein synthesis patterns of these cell types differ at the level of about 10.5% of the proteins having a 2-fold or greater difference. For comparison, previous analyses of REF52 fibroblast lysates revealed that duplicate gels of a given sample differ from one another at the level of about 2% of the analyzed spots, while gels of morphologically similar transformed REF52 fibroblast lines differ by between 4.6% and 12.5% 2-fold differences [35]. About 8% of the analyzed proteins change by 2-fold or more during the 4-cell stage of the mouse embryo once the major reprogramming associated with the maternal-to-embryonic transition has occurred [29]. About 30% of the analyzed proteins exhibit at least a 2-fold difference in their rates of synthesis in more highly divergent cell types (e.g., proliferating versus quiescent cells) [35]. The similarity in protein synthesis patterns of cumulus cells and mural granulosa cells, together with the specialized functions of the two cell types, indicates that the comparatively small number of reproducible differences in gene expression observed may be critical for correct cumulus cell functioning and hence for the formation of developmentally competent oocytes. Within this context, the ability of FSH and insulin treatment to alter the synthesis some of these proteins is highly significant. It is therefore interesting that either FSH or insulin alone was largely able to elicit in OAGCs rates of synthesis of the CUM and MUR proteins approximating the rates observed in cumulus cells. Thus, these two hormones both have a beneficial effect with respect to the CUM and MUR proteins. The beneficial effects of either hormone, however, are accompanied by elevation in rates of synthesis of many other proteins. Either insulin or FSH alone produced a number of beneficial effects on rates of synthesis of other proteins within the OAGCs; but in some cases insulin elicited some effects that were not elicited by FSH, and in other cases FSH produced effects not observed with insulin alone. Insulin treatment produced a protein synthesis pattern most like that of cumulus cells, but this pattern was nevertheless unlike the authentic cumulus pattern.
The failure of either hormone alone to elicit a protein synthesis pattern identical to that observed in cumulus cells, combined with the ability of either hormone to elicit effects not elicited by the other, might be construed as evidence that the two hormones added in combination should result in further mimicking of the pattern of protein synthesis in cumulus cells. However, the experimental evidence is clearly to the contrary. The greatest alteration in protein synthesis pattern was observed following the addition of FSH and insulin in combination. The addition of FSH+insulin likewise produces the most severe defect in oocyte developmental competence [27]. This indicates that the reduction in oocyte developmental competence correlates with the alteration in OAGC protein synthesis pattern. The effects of the combination of FSH and insulin are complex. A larger number of the CUM and MUR spots were altered by the addition of both hormones together, and a larger number of other proteins were increased or decreased in rates of synthesis (Fig. 2C, Table 1). Some proteins appeared to be elevated in rates of synthesis by insulin or FSH treatment alone but even more highly elevated when both hormones were added. Other proteins were uniquely elevated only with the combination of FSH+insulin (Fig. 6B). Perhaps more significantly, the combination of FSH+insulin negatively affected the expression of some proteins that were not altered by culture without added hormones or by treatment with either FSH or insulin alone. This negative effect is most apparent in the FI2 group of proteins, many of which were poorly expressed only in the presence of both hormones together (Fig. 6, A and B). Previous studies have indicated that insulin augments the effects of FSH on the differentiation of granulosa cells [37–39]. Our data indicate that the combination of FSH+insulin can also have a uniquely negative effect on the expression of some genes.
The negative effect of the two hormones together and some of the other effects of either hormone alone may be artifacts of the culture system. Our culture system obviously does not expose the developing OAGCs or oocytes to the same complex extracellular milieu that exists in vivo. One other important difference between our culture system and the situation in vivo is that products secreted by the oocyte are unable to accumulate to the same high concentrations in vitro as they are within intact follicles in vivo [27]. It is possible either that one or more oocyte-derived factors are required to operate in conjunction with FSH or insulin alone in order to impose a correct cumulus cell phenotype on the OAGCs, or that oocyte-derived factors are required to prevent the negative effect of the two hormones together. Oocyte-derived factors might also have a restraining influence upon the response of the OAGCs to the hormones. For example, treatment with either FSH or insulin alone or both in combination often resulted in significant overexpression of some proteins, increasing their rates of synthesis to well above the rates observed in either mural granulosa cells or cumulus cells. Likewise, the tendency of either hormone alone or in combination to increase the rates of synthesis of some of the MUR proteins may be deterred by oocyte factors, and this may constitute a vital aspect of the oocyte control over cumulus cell differentiation.
We previously reported that culture of oocyte-granulosa cell complexes in medium containing FSH (without insulin) did not have a beneficial effect on oocyte growth, competence to resume meiosis, or competence to develop to the blastocyst stage after fertilization in vitro [27]. With these as endpoints, there was no difference in oocytes grown in control medium (without either FSH or insulin) and in medium supplemented with either FSH or insulin alone. However, in the present study, culture in medium supplemented with either insulin or FSH resulted in the development of OAGCs that were more similar to cumulus cells than when the complexes were cultured in control medium. It is possible that these hormones had effects on oocyte developmental potential that were beneficial but too subtle to be detected using the endpoints assessed, and that other aspects of oocyte developmental competence should be evaluated, including fetal development following embryo transfer. Finally, assuming that optimal developmental competence of the oocyte is related to intercommunication between the oocyte and OAGCs expressing the cumulus cell phenotype, or not expressing the mural granulosa cell phenotype, it seems obvious that improvements in the culture system for oocytes should focus on conditions that promote the cumulus cell phenotype. The high-resolution 2D-PAGE approach described here should be a valuable tool in this approach because cellular phenotype can be evaluated globally through the display of over 1000 newly synthesized proteins rather than through reliance on the expression of just a few genes, which may not be suitable indicators of the differentiated state achieved by OAGCs in culture.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence. FAX: 207 288 6073; jje{at}jax.org
Accepted: March 23, 1999.
Received: December 16, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. Acevedo, J. Ding, and G. D Smith Insulin Signaling in Mouse Oocytes Biol Reprod, November 1, 2007; 77(5): 872 - 879. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Zheng, B. Patel, M. McMenamin, E. Moran, A. M. Paprocki, M. Kihara, R. D. Schramm, and K. E. Latham Effects of Follicle Size and Oocyte Maturation Conditions on Maternal Messenger RNA Regulation and Gene Expression in Rhesus Monkey Oocytes and Embryos Biol Reprod, April 1, 2005; 72(4): 890 - 897. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yao, X. Ren, J. J. Ireland, P. M. Coussens, T. P. L. Smith, and G. W. Smith Generation of a bovine oocyte cDNA library and microarray: resources for identification of genes important for follicular development and early embryogenesis Physiol Genomics, September 16, 2004; 19(1): 84 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. E. Latham, K. Wigglesworth, M. McMenamin, and J. J. Eppig Stage-Dependent Effects of Oocytes and Growth Differentiation Factor 9 on Mouse Granulosa Cell Development: Advance Programming and Subsequent Control of the Transition from Preantral Secondary Follicles to Early Antral Tertiary Follicles Biol Reprod, May 1, 2004; 70(5): 1253 - 1262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. O'Brien, J. K. Pendola, and J. J. Eppig A Revised Protocol for In Vitro Development of Mouse Oocytes from Primordial Follicles Dramatically Improves Their Developmental Competence Biol Reprod, May 1, 2003; 68(5): 1682 - 1686. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Chang, C. W. Brown, and M. M. Matzuk Genetic Analysis of the Mammalian Transforming Growth Factor-{beta} Superfamily Endocr. Rev., December 1, 2002; 23(6): 787 - 823. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Mitchell, C. R. Kennedy, and G. M. Hartshorne Effects of varying gonadotrophin dose and timing on antrum formation and ovulation efficiency of mouse follicles in vitro Hum. Reprod., May 1, 2002; 17(5): 1181 - 1188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Eppig and K. Wigglesworth Development of Mouse and Rat Oocytes in Chimeric Reaggregated Ovaries after Interspecific Exchange of Somatic and Germ Cell Components Biol Reprod, April 1, 2000; 63(4): 1014 - 1023. [Abstract] [Full Text]
|
|
|
|
|
|
R. De La Fuente, M. J. O'Brien, and J. J. Eppig Epidermal growth factor enhances preimplantation developmental competence of maturing mouse oocytes Hum. Reprod., December 1, 1999; 14(12): 3060 - 3068. [Abstract] [Full Text] [PDF]
|
|
| ||||
