| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular article |
a Department of Biological Regulation, The Weizmann Institute of Science, Rehovot, 76100, Israel
ABSTRACT
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.
cumulus cells, follicular development, hCG, LH, ovary, ovulation
INTRODUCTION
Angiogenesis, a relatively rare process in normal adult organs, occurs during normal development of the ovarian follicle. During follicular growth, angiogenesis is restricted to the theca cell layer. New blood vessels penetrate the basement membrane of the follicle only after ovulation [1]. The newly formed capillaries surrounding the preovulatory follicle are highly permeable [1]. This elevated permeability can result from the effect of vascular endothelial growth factor (VEGF/VPF), which is highly expressed in the developing follicle [2–5].
A marked increase (~8-fold) in steady-state levels of the transcripts for VEGF120 and VEGF164 was measured between 1 and 4 h after administration with hCG in whole ovaries [2]. Increases were detectable mainly in the cumulus cells but also in the granulosa cells and in thecal/stromal tissue [4, 6, 7]. The high level of expression was maintained at 10 and 18 h (Day 1 corpus luteum) [2, 8, 9]. The pattern of expression in the preovulatory follicle suggests that hypoxic stress may contribute to VEGF expression in the inner cumulus cells surrounding the oocyte [2, 10], as has been reported previously for other systems [3, 6].
Based on the elevated expression of VEGF in the cumulus cells and the vascular sprouting observed in vivo at the theca cell layer of the maturing follicles, the follicular fluid should be proangiogenic. In vitro, media conditioned either by isolated intact follicles from unstimulated or hCG-stimulated rabbits increased endothelial cell migration. The chemoattractant activity was associated with a molecular weight of greater than 30 000, and it was not correlated with the steroid concentration in the medium [11]. Human follicular fluid, aspired during retrieval of oocytes for in vitro fertilization, has enhanced DNA synthesis in endothelial cells [12]. Angiogenic activity has also been demonstrated in the conditioned medium of cells from luteinized rat ovaries [13] and in extracts of nonluteal mice and porcine ovaries [14, 15]. In the latter study, no activity was observed in the follicular fluid.
The conflicting data regarding angiogenicity of the follicular fluid may be explained, in part, by the fact that some angiogenic activity may be modulated by oxygenation. Indeed, conditioned medium derived from oxygen-depleted cultures of rat granulosa cells (2% oxygen) caused elongation and alignment of bovine aortic endothelial cells and also promoted their proliferation [16], whereas in medium conditioned at 20% oxygen, endothelial cells showed a typical cobblestone morphology. In addition, some angiogenic activity of the follicular fluid has been attributed to glycosaminoglycans and degradation products of hyaluronic acid secreted into the antrum by the surrounding granulosa and cumulus cells [17].
Controversially, whereas elevated VEGF expression is detected in cumulus cells at the center of the follicle 4 h after the LH surge, angiogenesis during this period is restricted to the theca cells. Vessel penetration into the granulosa cell layers occurs later, after ovulation, when the VEGF-expressing cumulus-oocyte complex (COC) is expelled to the oviduct. Thus, secretion of VEGF is not followed by directed invasion of endothelial cells through the basement membrane into the follicle center. Moreover, such angiogenic activity, if it were to occur, could potentially interfere with ovulation by restricting the release of the oocyte.
This spatial and temporal discrepancy between VEGF expression and angiogenesis led us to suggest that within the preovulatory follicle, angiogenesis is blocked by a high-molecular-weight constituent of the follicular fluid. The goal of this study was to test the possibility that hyaluronic acid, which is produced by cumulus cells and secreted into the follicular fluid [18–20], plays a role as an intrafollicular antiangiogenic shield. High-molecular-weight hyaluronic acid inhibits angiogenesis during development [21] and in tumors [22], whereas its products of degradation by hyaluronidase are angiogenic [23]. Hyaluronic acid is a major glycosaminoglycan of the extracellular matrix and is secreted specifically by the cumulus cells in the preovulatory follicle. Inhibition of hyaluronic acid synthesis by 6-diazo-5-oxo-1-norleucine (DON) prevents ovulation [18, 19 ].
MATERIALS AND METHODS
Materials
Calf serum was obtained from Hyclone (Logan, UT). Biogro 1, fetal calf serum, Dulbecco's Modified Eagle Medium (DMEM), penicillin, and streptomycin were purchased from Biological Industries (Kibbutz Beth Haemek, Israel). L-Glutamine and fungizone were obtained from Biolab Ltd. (Jerusalem, Israel), and ovine LH and hFSH were obtained from NHPP (Torrance, CA). [3H]Methyl-thymidine was purchased from Rotem Industries Ltd. (Beer Sheva, Israel). Collagen type I, eCG, DON (an inhibitor of hyaluronic acid synthesis), and hyaluronidase were obtained from Sigma Chemical Co. (St. Louis, MO), and hCG was from Organon (Holland). 4-Di-10-Asp [4-(4-(didecylamino)styryl)-N-methylpyridinium iodide] was obtained from Molecular Probes Inc. (Eugene, OR), and the scintillation fluid (ULTIMA GOLD) was from Packard Instrument Co. (Meriden, CT).
Endothelial Cell Culture
Fresh A19 bovine aortic endothelial cells [24] were kindly provided by Prof. I. Vlodavsky (Hadassah Medical Center, Jerusalem, Israel). These cells were cultured at 37°C in DMEM low glucose (1 g/L) supplemented with 10% calf serum, 1% L-glutamine, 1% Biogro-1, and antibiotics (penicillin, 100 U/ml; streptomycin, 0.1 mg/ml). Cells were maintained at 37°C in a 5% CO2 humidified incubator. Cells were dissociated with a solution of 0.05% trypsin, 0.02% EDTA, and 0.01 M sodium phosphate (pH, 7.4) and then were subcultured at a "split" ratio of 1:6 (v:v). The cells were used from passage 14 to 20.
Thymidine Incorporation
Cellular DNA was labeled with [3H]methyl-thymidine (5 µCi/ml) for 6 h. The medium was removed, and the cells were rinsed sequentially with methanol, chilled with 5% trichloroacetic acid and double-distilled water, and finally lysed with NaOH (0.5 M). Radioactive label was counted in an LKB Wallac counter (Turku, Finland) using scintillation fluid.
Plate Coating with Collagen
Rat tail collagen type I (3 mg/ml) was dissolved in 0.0175 M acetic acid in double-distilled water at 4°C. The collagen solution was dialyzed (24 h, 4°C) against excess (1:1100) of DMEM:double-distilled water (1:10; pH, 4). The collagen was then mixed with 0.15 M NaHCO3 and 10x DMEM in ratio of 7:2:1. Culture plates were coated with a thin layer of collagen, and the collagen was then solidified by incubation at 37°C for approximately 10 min.
Animal Handling
Immature, 23-day-old female Wistar rats (from the Wistar colony of the Department of Biological Regulation, Weizmann Institute, Israel) were stimulated to superovulate by eCG (15 U subcutaneously) as described elsewhere [25]. Forty-eight hours later, follicular maturation was induced by injection of human hCG (5 U intraperitoneally), mimicking the action of the endogenous surge of LH in the induction of ovulation. Thus, all the follicles will synchronously differentiate and mature, and the time of ovulation will be predictable (9–12 h after administration of hCG). Animal studies were conducted in accordance with the Guiding Principles for the Care and Use of Research Animals.
Isolation of Follicular Fluid
Five hours after hCG stimulation, rat ovaries were excised and punctured in DMEM culture medium (glucose, 1 g/L; 1% glutamine; 1% Biogro-1; penicillin, 100 U/mL, and streptomycin, 0.1 ng/ml; two ovaries in 1 ml). The medium was collected and centrifuged for 5 min (1000 rpm), and the supernatant was added to culture wells (160 µl/well in a total volume of 0.5 ml of medium; ~30% of the follicular fluid extracted from one ovary was added per well).
Isolation of COCs
Immature COCs from eCG-primed rats or maturing COCs from eCG/hCG-primed female rats were recovered by puncturing selectively the largest follicles in Leibovitz L15 culture medium supplemented with 5% fetal calf serum and antibiotics (penicillin, 50 IU/ml; streptomycin, 50 µg/ml; fungizone, 125 µg/ml). Matured COCs were also extracted from the ampula of the oviduct 24 h after hCG stimulation. The COCs were then aspirated with a capillary under the binocular and were passed several times into new medium to reduce granulosa cell contamination.
Fluorescent Staining of the Endothelial Cells
4-Di-10-Asp, a lipophilic tracer, was dissolved in ethanol at 1 mg/ml. Endothelial cells were incubated at 37°C with 4-Di-10-Asp (40 µg/ml) for 35 min. Excess dye was removed by washing the cells with culture medium. The cells remained fluorescent for several days.
Coculture of Endothelial Cells with COCs
Bovine aortic endothelial cells were seeded in 96- or 24-well plates (70 or 350 cells/mm2 in DMEM medium) and incubated for 6 h to allow cells to adhere to the plate. The COCs (n = 1–10) were then added to the wells. Glutamine (10-2 mg/ml) was present in sufficient quantity to allow hyaluronic acid synthesis by the cumulus cells. Human FSH (12.5 µg/ml) or ovine LH (5 µg/ml) was used for promoting the expansion of COCs [26]. Treatments included hyaluronidase (600 µg/ml) and DON (58 µM). The cocultures were studied by thymidine uptake and direct microscopic observation of the endothelial cells. For direct observation, the endothelial cells were prestained with the vital fluorescent dye as described previously.
In Vivo Inhibition of Hyaluronic Acid Synthesis
Immature eCG-primed rats were injected intraperitoneally with DON (35 µg/g BW) at two time-points: first at 1 h before hCG injection, and then again 6–7 h after administration of hCG. Rats were killed 7–9 h after hCG injection. Histological sections (5 µm) of the ovaries were stained with eosin, hematoxylin, and light green for visualization of the cytoplasm and connective tissue. In addition, adjacent sections were used for endothelial cell-specific BS-1 staining [27].
Statistical Methods
Data are reported as the mean ± SEM. Statistical significance was determined by Student's t-test or ANOVA (Fisher's post hoc probabilistic least significant difference). Experimental values were considered to be significant at P < 0.05.
RESULTS
Follicular Fluid Can Induce and Suppress Endothelial Cell Proliferation
Follicular fluid derived 5 h after administration of hCG affected the proliferation of bovine aortic endothelial cells in a complex manner that depended on the proliferative status of those cells (Fig. 1A). Endothelial cell proliferation was measured by [3H]thymidine incorporation 24 or 48 h after addition of the follicular fluid. In growth-arrested endothelial cells obtained by incubation with serum-free medium containing 0.5% BSA (serum starvation condition), follicular fluid significantly stimulated the proliferation of serum-starved endothelial cells (P = 0.02; Fig. 1A). However, when follicular fluid was applied to proliferating endothelial cells cultured in presence of 10% calf serum, it significantly inhibited their proliferation (P = 0.002 and 0.016 for 24 and 48 h, respectively; Fig. 1A).
|
Inhibition of Endothelial Cell Proliferation by Maturing COCs
To test whether the antiangiogenic activity resulted from the cumulus cells that surround the oocyte, endothelial cells were cocultured with COCs (10 ± 2 per well; Fig. 1B). Immature COCs from eCG-primed rats were cocultured with endothelial cells for 24 h in the presence of serum. Both immature COCs, which were triggered by gonadotropins to mature in vitro (experiments 1, 2, 4, 6) or in vivo (experiment 3), and matured COCs extracted from the ampula (experiment 5) significantly reduced thymidine incorporation in the endothelial cells (P = 0.005, two-tailed paired Student's t-test, six independent experiments, 3–4 wells per treatment in each experiment; Fig. 1B). Control experiments verified that thymidine incorporation was undetectable in COCs cultured under identical conditions in the absence of endothelial cells.
Hyaluronic Acid Dictates the Spatial Distribution of Endothelial Cells Cocultured with COCs
Endothelial cells prestained with the fluorescent vital dye 4-Di-10-Asp were cultured with COCs from eCG/hCG-primed rats (2 h after hCG, triplicates in two experiments). After 24 h, endothelial cells cultured with COCs surrounded the COCs at a distance and did not approach the cumulus cells (Fig. 2, A and B). These results further support the production of antiangiogenic material by COCs. After stimulation with hCG/LH, COC cells secrete high-molecular-weight hyaluronic acid. To test whether the antiangiogenic activity could be attributed to hyaluronic acid, we checked the distribution of endothelial cells among cocultures in which the levels of hyaluronic acid were perturbed either by its degradation with hyaluronidase or by inhibition of hyaluronic acid synthesis using DON. Endothelial cells cocultured with COCs and treated with either hyaluronidase or DON proliferated and migrated toward the cumulus cells (Fig. 2, C–F). Similar results were obtained using COCs derived from eCG-primed rats that were stimulated in vitro with ovine LH (5 µg/ml; data not shown).
|
The distance of the nearest 15 endothelial cells from the center of the oocyte was measured 24 h after treatment (one COC per well, 12 wells per treatment in two experiments). From these data, we calculated the density of endothelial cells (i.e., cells/unit area) as a function of the distance from the oocyte or the COC center (Fig. 3 and Table 1). In the absence of LH, the endothelial cell distribution was equal from 1 to 5 mm from the oocyte center. Thus, endothelial cells cocultured with control, nonmaturing COCs showed a homogeneous distribution, and the average endothelial distance (for the closest 15 endothelial cells) was 2.9 ± 0.1 mm. In the presence of LH, the endothelial cells tended to stay far from the COC center, and the mean cell distance was 3.2 ± 0.08 mm (P = 0.0078). When either DON or hyaluronidase were added, a significant decrease occurred in the mean distance (2.6 ± 0.1 mm and 2.19 ± 0.13 mm, respectively; P = 0.0001 for both). Treatment with DON restored the homogeneous endothelial cell distribution that was lost in the presence of LH, whereas hyaluronidase induced endothelial cells to migrate and proliferate toward the COC (Fig. 3 and Table 1).
|
|
Endothelial Cell Migration and Capillary Formation Are Regulated by COC Secretion of Hyaluronic Acid
Endothelial cells prelabeled with the vital stain 4-Di-10-Asp were cultured on collagen, an extracellular matrix that permits them to migrate, differentiate, and form capillary-like structures. Endothelial cells were observed by fluorescent microscopy every 24 h for 4 days (Fig. 4). At 24 h, endothelial cells cocultured with maturing COCs already remained at a distance and were tangential to the COCs (Fig. 4). However, inhibition of hyaluronic acid synthesis with DON, or its degradation with hyaluronidase, resulted in endothelial cell migration and formation of multiple elongated, capillary-like structures directed toward the cumulus (Fig. 4).
|
Hyaluronic Acid Synthesis Suppresses Follicular Angiogenesis In Vivo
The effect of hyaluronic acid in remodeling of the preovulatory follicular vasculature was also analyzed by perturbing its synthesis in vivo. Histological ovarian sections from rats treated with DON were obtained 3 h before ovulation (7 h after hCG administration), which is before the normal reported time of basement membrane breakdown (Fig. 5). In sections of control, untreated ovaries, we observed that the cumulus oophorus was dispersed, indicating the presence of hyaluronidase-sensitive material [26]. The basement membrane (shown in green) was intact and displayed a well-defined barrier between the granulosa and theca cell layers. In contrast, ovaries of rats treated with DON contained a dense cumulus oophorus, and the basement membrane was partially degraded, showing vague separation between the granulosa and theca cell layers (Fig. 5). Endothelial capillaries that penetrated the basement membrane were counted as well, and the mean number of lectin-positive capillaries per follicle was significantly lower in control follicles (1.51 ± 0.4, n = 13) than in follicles treated with DON (3.25 ± 0.9, n = 9, P = 0.03, Student's t-test). Thus, inhibition of hyaluronic acid synthesis resulted in significant premature angiogenesis toward the center of the follicle.
|
DISCUSSION
The present work aimed to explain the temporal and spatial discrepancy between the induced expression of VEGF and the restricted angiogenesis in the preovulatory follicle. After the LH surge, VEGF expression is elevated in the cumulus cells at the center of preovulatory follicles, and during this period, angiogenesis is confined to the theca interna. Subsequent infiltration of capillaries to the center of the follicle occurs only after the COC was expelled. In late preovulatory follicles and the postovulatory corpus luteum, where angiogenesis is ongoing, both VEGF and angiopoietin 2 are up-regulated, whereas angiopoietin 1 expression persists [7, 9, 28]. Angiopoietin 2 expression appears to be punctate, or focal. Transcripts of angiopoietin 2 were clustered in close association with blood vessels in the theca interna of the late preovulatory follicle. Angiopoietin 2 may collaborate with VEGF at the front of invading vascular sprouts by blocking a constitutive stabilization or maturing function of angiopoietin 1, thus allowing vessels to revert to, and to remain in, a more plastic state, in which they may be more responsive to a sprouting signal provided by VEGF.
These facts suggest that the new sprouts should invade the follicle even before ovulation. One possibility is that angiogenesis is blocked by the basement membrane. However, VEGF-activated endothelial cells can degrade basement membrane, and in fact, collagenolytic degradation of the basement membrane is an integral part of follicular rupture leading to ovulation [20, 29]. We hypothesize here that hyaluronic acid produced by the cumulus cells provides a high-molecular-weight antiangiogenic shield that prevents premature vascularization of the preovulatory follicle by blocking endothelial cell migration and proliferation. To test this hypothesis, we analyzed endothelial cells cultured in vitro with COCs and monitored angiogenesis in vivo among rats in which hyaluronic acid synthesis was suppressed.
Follicular fluid stimulated the proliferation of serum-starved endothelial cells, but it also suppressed the proliferation of endothelial cells cultured with serum. The induced proliferation of serum-starved endothelial cells is consistent with the presence of VEGF [30] and plasma proteins in the follicular fluid [31]. The inhibitory effect of follicular fluid on proliferating endothelial cells suggests that follicular fluid contains, in addition to proangiogenic molecules, a negative regulator of angiogenesis. Thus, the follicular fluid, which is a serum exudate, probably contains both pro- and antiangiogenic molecules. That COCs inhibited endothelial cell proliferation and migration suggests that COC cells are the source for the antiangiogenic activity. The antiangiogenic activity observed in vitro in the presence of COCs could be suppressed by hyaluronidase, which degrades hyaluronic acid, or by DON, which inhibits hyaluronic acid synthesis.
DON is an effective inhibitor of glucosamine synthesis. It binds with and inactivates the aminotransferase that transfers an amino group from glutamine to fructose-6-P, thus inhibiting the formation of glucosamine-6-P [32]. The volumetric increase of the cumulus mass in the presence of the substrates of hyaluronic acid synthesis is directly proportional to the amount of hyaluronic acid synthesized by these cultured COCs [19]. Cumulus expansion in the eCG/hCG-stimulated rat occurs between 2.5 and 5 h after the ovulatory stimulus [33], and by approximately 5 h after stimulation, the volume of the cultured COC increases almost 40-fold. This expansion can be inhibited both in vitro or in vivo by the administration of DON [19].
Endothelial cell migration and capillary formation on collagen were inhibited by COCs, and this inhibitory effect could be suppressed using hyaluronidase or DON. Thus, capillary-like endothelial structures remained at a distance and tangential to the control COC, whereas capillary-like structures were radial and infiltrated into COCs treated with hyaluronidase or DON. Likewise, in vivo administration of DON led to early infiltration of endothelial cells through the basement membrane and into the granulosa cell layers.
Angiogenesis is frequently controlled by a balance of pro- and antiangiogenic regulators [34]. In the preovulatory follicle, we suggest that the spatial and temporal regulation of angiogenesis is maintained through a balance between secretion of diffusible proangiogenic VEGF, which stimulates angiogenesis and permeability in the perifollicular vasculature, and secretion of high-molecular-weight antiangiogenic hyaluronic acid, which prevents vascular sprouting locally into the growing follicle. Induction of VEGF expression is essential for improving perfusion to the follicular periphery and reducing metabolic stress for the oocyte [35]. The VEGF-activated endothelial cells also may contribute to collagenase activity and dissolution of the basement membrane in preparation for ovulation. We argue here that along with the induction of VEGF and initiation of vascular sprouting, it is essential to inhibit angiogenesis locally within the growing follicle to ensure that the oocyte remains unrestrained. Local inhibition of angiogenesis is achieved by secretion of hyaluronic acid, which is a solid, high-molecular-weight gel that can specifically inhibit vascular infiltration in the confined space of the follicular fluid.
ACKNOWLEDGMENTS
We would like to thank Prof. Nava Dekel and Prof. Alex Tsafriri for helpful discussions. Gonadotropin hormones were obtained through the National Hormone and Pituitary Program.
FOOTNOTES
1 Supported by a research grant from the MINERVA Foundation and by NIH grant CA75334-01A1. M.N. is incumbent of a Career Development Award from the Israel Cancer Research Foundation. 
2 Correspondence. FAX: 972 8 9344116; lhneeman{at}wicc.weizmann.ac.il
Accepted: February 22, 2000.
Received: April 14, 1999.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  L. Krishnan, J. B. Hoying, H. Nguyen, H. Song, and J. A. Weiss Interaction of angiogenic microvessels with the extracellular matrix Am J Physiol Heart Circ Physiol, December 1, 2007; 293(6): H3650 - H3658. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Shiftan, T. Israely, M. Cohen, V. Frydman, H. Dafni, R. Stern, and M. Neeman Magnetic Resonance Imaging Visualization of Hyaluronidase in Ovarian Carcinoma Cancer Res., November 15, 2005; 65(22): 10316 - 10323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Gruemmer, L Klein-Hitpass, and J Neulen Regulation of gene expression in endothelial cells: the role of human follicular fluid J. Mol. Endocrinol., February 1, 2005; 34(1): 37 - 46. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Zhang, S. L. Baader, M. Sixt, J. Kappler, and U. Rauch Neurocan-GFP Fusion Protein: A New Approach to Detect Hyaluronan on Tissue Sections and Living Cells J. Histochem. Cytochem., July 1, 2004; 52(7): 915 - 922. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schoenfelder and R. Einspanier Expression of Hyaluronan Synthases and Corresponding Hyaluronan Receptors Is Differentially Regulated During Oocyte Maturation in Cattle Biol Reprod, July 1, 2003; 69(1): 269 - 277. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
(r22 - r12)).) Without LH, the endothelial cells were distributed randomly. With LH, endothelial cells remained distant from the COC, with the maximal endothelial cell density occurring at 3–4 mm. With DON or hyaluronidase, the endothelial cells were found closer to the COC (n = 5). Note that the low density of cells at the outer rings of COCs treated with DON or hyaluronidase results from only the 15 closest cells being counted in each case (see 
