Partial characterization of a fraction from human follicular fluid that initiates the human sperm acrosome reaction in vitro

A human follicular fluid (HFF) fraction prepared by Sephadex G-75 column chromatography has been previously shown by this laboratory to initiate the human sperm acrosome reaction (AR) in vitro. In the present report, the apparent molecular weight (MW) of this AR activity determined by a longer G-75 column than was used in the previous work was 50,000 ± 5,106. The G-75 Sephadex void volume fractions of some but not all HFF samples were also found to contain some AR-initiating activity. The occasional void volume activity was less potent than that of the 50,000 MW fraction and was not studied further. Further characterization of the 50,000 MW fraction was carried out. A time-course study demonstrated that maximum AR were obtained within 5 min following the addition of the 50,000 MW fraction. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by silver staining revealed that the 50,000 MW fraction was still a relatively crude preparation. Treatment of the 50,000 MW fraction with chloroform:methanol did not extract the AR-initiating activity into the lipid phase. The AR-initiating activity of the untreated 50,000 MW fraction was precipitated when it was boiled, but the activity was partially resistant to boiling after overnight incubation. Treatment of the 50,000 MW fraction with pronase E or with several glycosaminoglycan hydrolases did not destroy the activity. Pronase treatment resulted in a higher amount of boiling-resistant AR-initiating activity. The AR-initiating activity of the untreated 50,000 MW fraction was partially dialyzable, but the activity of an undialyzed fraction did not pass through an ultrafiltration membrane with a 10,000 MW cut-off. However, treatment of the 50,000 MW fraction with protease, peptide:N-glycosidase F, and to a lesser extent chondroitinase ABC yielded an active lower MW activity which could pass through such an ultrafiltration membrane. The lower MW activity released by peptide:N-glycosidase F eluted in the included volume (5,000–1,000) of a Sephadex G-25 column. Neutral hexose but not protein or peptide was detected in the G-25 peak of AR-initiating activity. These results suggest that the AR-initiating activity present in the 50,000 MW fraction of HFF: (1) is present either as two different AR factors (a high-MW factor and a low-MW, noncovalently bound factor) or as a single factor responsible for both the nondialyzable and dialyzable AR-initiating activities (the latter being enzymatically released from the former), and (2) may be at least partially associated with N-linked oligosaccharides of a glycoprotein or proteoglycan.     

Comparison of acrosome reaction-inducing activities of human cumulus oophorus, follicular fluid and ionophore A23187 in human sperm populations of proven fertilizing ability in vitro

Significant differences in the frequency of spontaneous and induced acrosome reactions (identified by electron microscopy) were detected between 5 individual sperm samples. Maximal stimulation of the acrosome reaction was achieved with ionophore A23187, no differences being detected between the two preincubation times used (5 and 15 h). The cumulus oophorus and follicular fluid caused a similar increase in the proportion of acrosome-reacted spermatozoa after 5 h of preincubation. It is concluded that specific products of the cumulus and/or granulosa cells may contribute to the acrosome reaction-inducing activity of human follicular fluid.     

Secretion of cumulus expansion-enabling factor (CEEF) in porcine follicles

The objective of this study was to find out whether porcine cumulus and mural granulosa cells can secrete cumulus expansion-enabling factor (CEEF). Culture drops of M-199 medium were conditioned with denuded porcine oocytes (1 oocyte/μl), cumulus cells from oocytectomized complexes (1 OOX/μl), pieces of mural granulosa isolated from preantral to preovulatory follicles (1000 cells/μl), or oviductal cells (1000 cells/μl) for 24 hr. The production of CEEF was assessed by the addition of mouse OOX and follicle-stimulating hormone (FSH) (1 μg/ml) to microdrops of the conditioned medium. After 16–18 hr, expansion of the mouse OOX was scored on a scale of 0 to 4 by morphologic criteria. Mouse OOX did not expand in nonconditioned FSH-supplemented medium. Immature porcine oocytes produced +3 to +4 expansion of the mouse OOX. Granulosa cells isolated from preantral and early antral follicles and cumulus cells isolated from all stages of follicle development constitutively secreted CEEF under in vitro conditions. Mural granulosa cells of small, medium, and preovulatory (PMSG) follicles also secreted CEEF in vitro; however, FSH or leutenizing hormone (LH) stimulation was essential for this secretion. Hormonally induced secretion of CEEF was accompanied by expansion of the mural granulosa itself. Granulosa cells isolated from follicles of gilts 20 hr after PMSG and human chorionic gonadotropin (hCG) administration did not produce CEEF and did not expand in response to FSH and LH in vitro. CEEF activity also was found in the follicular fluid of small antral follicles, was reduced in medium follicles, and was not detectable in PMSG-stimulated follicles. However, CEEF activity was reestablished in the follicular fluid of preovulatory follicles by hCG injection, conceivably due to increased production of CEEF by cumulus cells. We conclude that (1) porcine cumulus and mural granulosa cells are capable of CEEF production in vitro and (2) autocrine secretion of CEEF by cumulus cells is involved in regulation of porcine cumulus expansion both in vitro and in vivo. Mol. Reprod. Dev. 49:141–149, 1998. © 1998 Wiley-Liss, Inc.     

Localization and synthesis of hyaluronic acid in the cumulus cells and mural granulosa cells of the preovulatory follicle.

Mural and cumulus granulosa cells synthesize hyaluronic acid (HA) and expand in vitro in response to follicle-stimulating hormone and a soluble factor(s) produced by fully grown oocytes. In the present study we examined HA synthesis and extracellular matrix organization by the two cell populations in vivo during the preovulatory period. After injection of human chorionic gonadotropin into pregnant mares' serum gonadotropin-primed animals, a progressive increase in HA synthesis was observed by the cumulus cell-oocyte complex (COC), and by the mural granulosa cells adjacent to the antrum (antral granulosa cells). The outermost layers of mural granulosa cells (peripheral granulosa cells) did not synthesize HA. Net HA synthesis was approximately 4 pg/cell for COCs isolated after full expansion induced either in vivo or in vitro, whereas the total HA content and cell number in the ovulated COC (approximately 11 ng HA and approximately 3000 cells per COC) were about threefold higher than for COCs expanded in vitro (approximately 4 ng HA and approximately 1000 cells per COC). The increased cell content of ovulated COCs appears to be primarily the result of inclusion of proximal mural granulosa cells which synthesize HA in response to the oocyte factor(s) and become incorporated in the expanded COC extracellular matrix mass. Media conditioned by oocytes enclosed in the cumulus cell mass (intact COCs) contained only 10-20% of the HA-stimulatory activity of media conditioned by an equal number of isolated oocytes when tested on mural granulosa cell cultures. Further, HA-stimulatory activity of media conditioned by isolated oocytes was dramatically reduced (approximately 70%) by preincubation for 5 hr with cumulus cells compared to preincubation in the absence of cells. The results suggest that differences in HA synthesis between subregions of membrana granulosa depend on a diffusion gradient of the oocyte factor(s).     

Cumulus oophorus extracellular matrix: its construction and regulation.

Cumulus oophorus, an investing structure unique to oocytes of higher mammals, is induced to synthesize an extensive extracellular matrix by ovulatory stimulus, leading to the characteristic preovulatory expansion of the cumulus-oocyte complex. The extracellular matrix consists of cumulus cell-secreted hyaluronan, proteoglycans and proteins, as well as extrafollicularly originated SHAPs (serum-derived hyaluronan-associated proteins) that are bound covalently to hyaluronan. The secretion and assembly of matrix molecules by cumulus cells are temporally regulated by factors derived from both mural granulosa cells and oocyte, which synchronize the deposition of the cumulus oophorus matrix with other intrafollicular ovulatory events. The cumulus oophorus matrix is essential for ovulation and subsequent fertilization. Recently, taking advantage of animal models with defined genetic modifications, it has become possible to investigate in vivo the structure of the cumulus oophorus matrix, the regulatory mechanism for matrix deposition and its biological functions. This review focuses on the recent findings on the construction of the cumulus oophorus matrix and the regulation.     

An influx of extracelluar calcium is required for initiation of the human sperm acrosome reaction induced by human follicular fluid

The role of Ca²⁺ in the human sperm acrosome reaction was investigated using the fluorescent calcium indicator fura-2. Previous experiments have shown that a Sephadex G-75 column fraction of human follicular fluid can stimulate the human sperm acrosome reaction [Suarez SS, Wolf DP, Meizel S (1986): Gamete Res 14:107–121]. Using fura-2, we demonstrated that this Sephadex G-75 fraction also stimulates a rapid, transient increase in intracellular free Ca²⁺. This Ca²⁺ transient is blocked either by chelation of extracellular calcium or by addition of the Ca²⁺ antagonist La³⁺. We have also been able to stimulate the acrosome reaction in human sperm without significant loss of motility, using the divalent cation ionophore ionomycin. Acrosome reactions stimulated by whole follicular fluid, the G-75 fraction, or ionomycin are all blocked by removal of extracellular Ca²⁺. These results strongly suggest that an influx of extracellular Ca²⁺ is responsible for intiating the acrosome reaction in human sperm treated with human follicular fluid. This is the first demonstration in mammalian sperm that a potentially physiological stimulus can cause an increase in intracellular Ca²⁺ concomitant with the acrosome reaction.     

Acrosome reaction in the cumulus oophorus revisited: involvement of a novel sperm-released factor NYD-SP8

Fertilization is a process involving multiple steps that lead to the final fusion of one sperm and the oocyte to form the zygote. One of the steps, acrosome reaction (AR), is an exocytosis process, during which the outer acrosome membrane fuses with the inner sperm membrane, leading to the release of acrosome enzymes that facilitate sperm penetration of the egg investments. Though AR has been investigated for decades, the initial steps of AR in vivo, however, remain largely unknown. A well elucidated model holds the view that AR occurs on the surface of the zona pellucida (ZP), which is triggered by binding of sperm with one of the ZP glycosylated protein, ZP3. However, this model fails to explain the large number of ‘falsely’ acrosome-reacted sperms found within the cumulus layer in many species examined. With the emerging evidence of cross-talk between sperm and cumulus cells, the potential significance of AR in the cumulus oophorus, the outer layer of the egg, has been gradually revealed. Here we review the acrosome status within the cumulus layer, the cross-talk between sperm and cumulus cells with the involvement of a novel sperm-released factor, NYD-SP8, and re-evaluate the importance and physiological significance of the AR in the cumulus in fertilization.     

Patterns of [3H] thymidine incorporation differ in immature rats and mature, cycling rats

By the time follicular development has progressed to the preovulatory stage, granulosa cells abutting the basement membrane no longer incorporate [3H] thymidine (3H-TdR). The purpose of this experiment was to determine when, during the course of follicular growth, cell proliferation in these mural granulosa cells ceases. Autoradiographs were prepared following continuous 3H-TdR infusion in vivo, or incubation with 3H-TdR in vitro. In cycling rats, the concentration of silver grains over mural regions of the granulosa layer was lower than over antral regions of most follicles with greater than 1000 cells in the largest cross section (LCS). This centripetal labeling pattern became more striking as follicular size increased. By proestrus, only the cells of the discus proligerus (cumulus and the portion of the follicular wall supporting the cumulus oocyte complex) continued to incorporate 3H-TdR. In contrast to cycling rats, centripetal labeling patterns were not seen in ovaries of prepubertal rats, even in follicles of the same size. The difference in follicular growth patterns between these two types of animals suggests an influence of cyclic gonadotropin surges on the control of granulosa cell proliferation.     

Induction of the acrosome reaction in human spermatozoa by a fraction of human follicular fluid

Human ejaculated spermatozoa were washed through a Percoll gradient, preincubated for 10 hr in a defined medium containing serum albumin, and then induced to undergo rapid acrosome reactions by addition of human follicular fluid or a Sephadex G-75 column fraction of the fluid. Induction by follicular fluid did not occur when the spermatozoa were preincubated for only 0 or 5 hr. The reactions were detected by indirect immunofluorescence using a monoclonal antibody directed against the human sperm acrosomal region. The percentage of acrosomal loss counted by transmission electron microscopy agreed with that counted by immunofluorescence. The apparent molecular weight of the Sephadex G-75 fraction containing the peak of acrosome reaction-inducing activity was 45,000 ± 4,200 (SD). The occurrence of physiological acrosome reactions was supported by: assessing motility (no significant loss of motility occurred during the treatment period when sperm were preincubated with bovine serum albumin), transmission electron microscopy (the ultrastructural criteria for the acrosome reaction were met), and zona-free hamster oocyte binding and penetration (spermatozoa pretreated with the active fraction of follicular fluid, then washed and incubated with oocytes, showed significantly greater binding to and penetration of oocytes). The stimulation of the acrosome reaction by follicular fluid is apparently not due to blood serum contamination; treatment of preincubated spermatozoa with sera from the follicular fluid donors had no effect on the spermatozoa. The nature of the active component(s) in that fraction is currently being investigated.     

Granulosa cells stimulate in vitro the expansion of isolated mouse cumuli oophori: involvement of prostaglandin E2

Granulosa cells (2 X 10(6) per ml) obtained from pregnant mare's serum gonadotropin (PMSG)-primed mice induce within 24 h the expansion in vitro of cocultured mouse cumuli oophori. Experiments with conditioned media showed that the expansion-promoting action of granulosa cells is due to diffusible factor(s) released into the culture medium. Studies with prostaglandin synthetase inhibitors and direct measurements of prostaglandin E2 (PGE2) released by granulosa cells in the culture medium have also been performed. The results strongly suggest that the cumulus oophorus expansion-promoting action of granulosa cells is mediated by PGE2, and support the hypothesis (Downs and Longo, 1983) that granulosa cells might play a similar role in the mechanism of cumulus expansion in vivo. The suggestion is advanced that coculture with granulosa cells might be of help to allow physiologic expansion in culture of immature cumuli obtained from preovulatory follicles in in vitro fertilization programs.     

Cumulus cells of oocyte-cumulus complexes secrete a meiosis-activating substance when stimulated with FSH

The effect of the different follicular cell types on resumption of meiosis was studied during stimulation with FSH. Cumulus enclosed oocytes (CEO), denuded oocytes (DO), and cumulus and mural granulosa cells were used. The resumption of meiosis and oocyte maturation were assessed by the determination of the germinal vesicle breakdown (GVBD) and polar body formation (PB) at the end of a 24 hr culture period in the presence of 4 mM hypoxanthine (HX). The effects of recombinant LH (r-LH) and hCG were also evaluated. Oocyte exposure to the gonadotrophins varied from 5 min to 24 hr (i.e., priming time). Oocytes were obtained from immature gonadotrophin-stimulated and -unstimulated mice. 1. FSH (1 IU/L-75 IU/L) provoked a dose-dependent increase in GVBD and PB in CEO, but not in DO, in stimulated and unstimulated mice. Eight IU/L was sufficient for inducing resumption of meiosis. In contrast, LH and hCG (both 1 IU/L-1500 IU/L) were without effect on GVBD and PB in CEO and DO of oocytes from stimulated and unstimulated mice. A combination of 8IU/L FSH and 4–8 IU/L hCG produced an additive effect, whereas combinations with LH and higher concentrations of hCG had no such effect. 2. A 2 hr priming with FSH (8 IU/L-75 IU/L) induced a dose-dependent oocyte maturation in CEO. Thirty minutes of priming with FSH (75 IU/L) was sufficient for induction of meiotic resumption in CEO. 3. Priming CEO with FSH for 2 hr followed by the separation and repooling of oocytes and cumulus cells induced oocyte maturation. GVBD of new, unprimed DO added to cumulus cells of primed CEO increased slightly but was significant, whereas GVBD in DO isolated from the primed CEO only increased marginally. DO cocultured with FSH-primed cumulus masses seem to be prevented from resuming meiosis. 4. Priming a coculture of granulosa cells and DO with FSH for 2 hr caused a significant increase in GVBD compared to the control, evaluated after 24 hr. In contrast, a 24 hr FSH-priming of a coculture of granulosa cells and DO was without effect on GVBD. 5. A spent medium in which unstimulated cumulus cells or mural granulosa cells had grown was without effect on GVBD in DO. However, a small fraction of the DO resumed meiosis after culture in a spent medium derived from a 2 hr priming of CEO and spent media from 24 hr priming of CEO induced a 2–3 times higher GVBD frequency in the DO compared to the controls. Heat treatment of spent media (70°C, 30 min) from a 24 hr FSH-priming of CEO still induced GVBD in naive DO. The results showed that FSH, in a concentration of as little as 8 IU/L, but not r-LH and hCG, induced within 30 minutes the cumulus cells to produce and after 2 hr to secrete a diffusible heat stable meiosis activating substance. This substance overcame, in a paracrine fashion, the inhibiting effect of HX and induced oocyte maturation directly in DO. The production of this substance, however, was dependent on the initial connection between the cumulus cells and the oocyte, indicating an important 2-way communication between these 2 cell types. The mural granulosa cells did not produce a meiosis inducing activity by stimulation with FSH, but significantly, more DO matured after coculture with the nonstimulated granulosa cells for 24 hr than for 2 hr. It is proposed that the heat stable meiosis activating component of the spent media from the FSH-stimulated CEO belongs to the meiosis activating sterols, MAS, previously isolated from human follicular fluid and from adult bull testes. Mol. Reprod. Dev. 46:296–305, 1997. © 1997 Wiley-Liss, Inc.     

Factors affecting the acrosome reaction in human spermatozoa

Large pieces of human cumulus oophorus were exposed for 20-30 min to washed spermatozoa or to spermatozoa recovered after a swim-up procedure, and then fixed for electron microscopy. Spermatozoa of both populations penetrated deeply into the cumulus within that time, and none of 48 observed clearly had undergone an acrosome reaction (AR). As measured by fluorescence microscopy, an AR rate of 12% in spermatozoa obtained at 4 h following a swim-up increased to about 25% in samples incubated in culture dishes for approximately 20 h. However, this latter AR rate was no different in the presence or absence of a cumulus/oocyte complex, and was only moderately greater in 50% follicular fluid. Nor was it affected to any degree by the absence of calcium or by a low (26 degrees C) temperature, both of which are regulators of the physiological AR in other species. By contrast, a clear dose-related enhancement of the AR by the calcium ionophore A23187 was almost completely Ca2(+)-dependent. We conclude that the human cumulus oophorus does not rapidly induce an AR in spermatozoa capacitated in vitro and, unlike the situation in some other mammals, that washed human spermatozoa do not first require a period of capacitation in order to penetrate it. The results also point to the likelihood that ARs monitored in free-swimming human spermatozoa are for the most part spurious or artefactual, and they show that in-vitro AR rates in such populations do not parallel their fertilizing ability.     

Differential modulation of rat follicle cell gap junction populations at ovulation

The hypothesis proposed in the late 1970s that meiotic resumption in mammalian oocytes might result from the disruption of gap junction communication between follicle cells and the oocyte has not been supported by metabolic cooperation experiments which demonstrate that exogenous tracer transfer from the cumulus oophorus to the oocyte does not decrease until several hours after germinal vesicle breakdown (GVBD). Since these studies utilized isolated cumulus-oocyte complexes for their measurements, however, they excluded from consideration the possible effect of separation of the cumulus oophorus from the membrana granulosa which was required for this assay. We considered the possibility that the disruption of cumulus junctions within the intact follicle could mimic this experimental manipulation and previously reported that cumulus gap junctions were dramatically down-regulated during the period of GVBD in vivo. In the present study, we have utilized quantitative morphometric techniques to analyze the responses of other gap junction populations in intact preovulatory rat follicles to an ovulatory stimulus and demonstrate now that membrana granulosa, cumulus, and cumulus-oocyte gap junctions are down-regulated at different times and rates during the preovulatory period. Although membrana gap junctions are down-regulated during the period of meiotic resumption, their loss is not as rapid or as complete as in the cumulus oophorus. Cumulus-oocyte gap junctions are down-regulated after meiosis resumes but during the same period other investigators have demonstrated a reduction in metabolite transfer between the cumulus oophorus and the oocyte. Our results are interpreted to suggest that the cumulus oophorus may regulate the conduction of meiosis inhibitory signals between the membrana granulosa and the oocyte.     

Intrafollicular content of luteinizing hormone receptor, alpha-inhibin, and aromatase in relation to follicular growth, estrous cycle stage, and oocyte competence for in vitro maturation in the mare

The intrafollicular content of LH receptor, alpha-inhibin, and aromatase are known good indicators of follicular status. We investigated the amounts of these proteins in granulosa and cumulus cells in relation to oocyte competence for in vitro maturation, follicular growth, and estrous cycle stage in the mare. Follicular punctures were performed 34 h after an injection of crude equine gonadotropins, either during the follicular phase, at the end of the follicular phase, or during the luteal phase. The cumulus-oocyte complex, granulosa cells, and follicular fluid of follicles larger than 5 mm were collected. The nuclear stage of the oocytes after in vitro culture was determined microscopically. Granulosa and cumulus cell amounts of LH receptor, alpha-inhibin, and aromatase were assessed by the semiquantitative Western blot method and image analysis. Follicular fluids were assayed for progesterone (P4) and estradiol-17beta (E2). The three factors were expressed in mural granulosa and cumulus cells from all follicles from the gonadotropin-independent growth period until the preovulatory stage. Considering all the follicles punctured, the amounts of LH receptor and alpha-inhibin in granulosa cells were not different for the three physiological stages studied. The amounts of aromatase in granulosa cells, as well as the E2:P4 ratios, were higher for follicles punctured during the follicular phase than for the two other groups (p < 0.05). Considering the data from the three groups, the E2:P4 ratio and the LH receptor and aromatase contents, but not alpha-inhibin, in granulosa cells increased with an increase in follicular diameter (p < 0.01). The E2:P4 ratios and the amounts of LH receptor, alpha-inhibin, and aromatase in granulosa cells were lower in follicles 5-9 mm in diameter than in larger ones (p < 0.05). In cumulus cells, the amounts of the three factors were different neither between the three groups nor between the follicular diameters. Although we could not establish any obvious relationship to oocyte competence for in vitro maturation, the influence of the follicle diameter on the content of LH receptors, alpha-inhibin, and aromatase in granulosa cells was similar to the influence of follicle diameter on oocyte competence. Therefore, one can hypothesize that, in the mare, there is a link between the acquisition of oocyte competence and the expression of these factors in the follicular cells.     

Oocyte regulation of anti-Müllerian hormone expression in granulosa cells during ovarian follicle development in mice

In the ovarian follicle, anti-Müllerian hormone (Amh) mRNA is expressed in granulosa cells from primary to preovulatory stages but becomes restricted to cumulus cells following antrum formation. Anti-Müllerian hormone regulates follicle development by attenuating the effects of follicle stimulating hormone on follicle growth and inhibiting primordial follicle recruitment. To examine the role of the oocyte in regulating granulosa cell Amh expression in the mouse, isolated oocytes and granulosa cells were co-cultured and Amh mRNA levels were analysed by real-time RT-PCR. Expression in freshly isolated granulosa cells increased with preantral follicle development but was low in the cumulus and virtually absent in the mural granulosa cells of preovulatory follicles. When preantral granulosa cells were co-cultured with oocytes from early preantral, late preantral or preovulatory follicles, and when oocytes from preovulatory follicles were co-cultured with cumulus granulosa cells, Amh expression was increased at least 2-fold compared with granulosa cells cultured alone. With oocytes from preantral but not preovulatory follicles, this was a short-range effect only observed with granulosa cells in close apposition to oocytes. We conclude that stage-specific oocyte regulation of Amh expression may play a role in intra- and inter-follicular coordination of follicle development.     

Effect of granulocyte-macrophage colony-stimulating factor deficiency on ovarian follicular cell function

Granulocyte-macrophage colony-stimulating factor (GM-CSF), a cytokine secreted by lymphohaemopoietic and other cell lineages, is known to influence ovarian cyclicity and embryo development. The aim of this study was to examine the effect of GM-CSF on ovarian follicular cell function using GM-CSF-deficient (GM -/-) mice. Immature GM -/- and GM +/+ mice were stimulated with eCG, and cumulus-oocyte complexes and mural granulosa cells were collected 48 h later. Expression of GM-CSF receptor (GM-CSFR) alpha and beta mRNA subunits by cumulus-oocyte complexes and mural granulosa cells was examined using RT-PCR. Cumulus-oocyte complexes from both genotypes were found to express mRNA for the GM-CSFRalpha-subunit only, while the mural granulosa cells expressed both the alpha and beta receptor subunits. Cumulus-oocyte complexes recovered from GM -/- mice had approximately twice the number of cumulus cells per cumulus-oocyte complex than did those of GM +/+ mice (P < 0.05), even though the growth-promoting activity of denuded GM -/- oocytes was found to be equivalent to that of wild-type oocytes. GM-CSF deficiency was associated with marginally increased DNA synthesis in cumulus cells and significantly (P < 0.05) lower progesterone production by mural granulosa cells recovered from GM -/- compared with those recovered from GM +/+ mice. The addition of rec-mGM-CSF in vitro did not affect DNA synthesis in either cell type or progesterone production by mural granulosa cells, irrespective of GM-CSF status. There was no effect of GM-CSF deficiency on the capacity of FSH and insulin-like growth factor I to stimulate DNA synthesis in cumulus-oocyte complexes (approximately 15- and threefold, respectively) and in mural granulosa cells (approximately two- and threefold, respectively). Taken together, these data show that GM-CSF influences events associated with follicular maturation in mice. The effects of GM-CSF are not exerted directly in granulosa or cumulus cells, but appear to be mediated indirectly, perhaps through the agency of steroidogenesis-regulating secretions of local macrophage populations residing in the theca.     

Hyaluronic acid as an anti-angiogenic shield in the preovulatory rat follicle

Angiogenesis in the preovulatory follicle is confined to the theca cell layers, and penetration of capillaries through the basement membrane into the granulosa cell layers does not occur until after ovulation. However, elevated expression of the angiogenic growth factor (VEGF) has been reported in the cumulus cells surrounding the oocyte, which are expelled from the follicle during ovulation. This spatial and temporal discrepancy between VEGF expression and angiogenesis was studied here in the rat ovarian follicle, and we showed that cumulus cells secrete to the follicular fluid, in addition to VEGF, material with antiangiogenic activity that blocks endothelial cell proliferation, migration, and capillary formation in vitro. Hyaluronic acid produced by the cumulus cells can account for this antiangiogenic activity. Degradation of hyaluronic acid by hyaluronidase restored proliferation and migration of endothelial cells directed toward the cumulus. Inhibition of hyaluronic acid synthesis with 6-diazo-5-oxo-1-norleucine restored endothelial proliferation and migration in vitro, and it also resulted in early penetration of capillaries across the follicular basement membrane in vivo. These results support the role of hyaluronic acid produced by the cumulus cells as a high-molecular-weight, antiangiogenic shield that prevents premature vascularization of the preovulatory follicle by blocking endothelial cell migration and proliferation.