Proliferation of Rat Granulosa Cells during the Periovulatory Interval
http://www.100md.com
内分泌学杂志 2005年第1期
Department of Physiology, Medical College of Georgia, Augusta, Georgia 30912
Address all correspondence and requests for reprints to: Charles L. Chaffin, Ph.D., Department of Physiology, Medical College of Georgia, CA2098, 1120 15th Street, Augusta, Georgia 30912. E-mail: cchaffin@mail.mcg.edu.
Abstract
Granulosa cell proliferation during luteinization and terminal differentiation has historically been assumed to decline rapidly after an ovulatory stimulus. In contrast, terminal differentiation in other cell types has recently been associated with a transient increase in proliferation, suggesting that this may occur in the ovarian follicle. The goal of the current study was to test the hypothesis that an ovulatory stimulus to rats results in additional granulosa cell proliferation before cell cycle arrest. Immature rats were given a single injection of pregnant mare serum gonadotropin (PMSG) followed by human chorionic gonadotropin (hCG) to initiate periovulatory events. The proportion of granulosa cells in S phase did not change until 12 h after hCG, although the majority of the post-hCG proliferation was localized to cumulus granulosa cells for up to 10 h after hCG. The expression of cyclin D2 mRNA did not decline until 12 h after hCG, although both cyclin-dependent kinase (Cdk)4 and Cdk6 mRNA increased at 6 h. Protein levels of cyclin D2 and Cdk4 did not change as a result of hCG, whereas cyclin E increased 6 h after hCG. Kinase activity of Cdk2 dropped markedly by 4 h after hCG, but a slight increase in activity was evident 6–8 h after hCG. These data suggest that cumulus granulosa cells continue to proliferate for up to 10 h after an ovulatory stimulus, possibly via cyclin E/Cdk2. It is concluded that proliferation is maintained in granulosa cells in the proximity of the oocyte during luteinization of the rat follicle.
Introduction
THE PERIOVULATORY INTERVAL is defined as the onset of an ovulatory gonadotropin stimulus until follicle rupture. In rats, this interval lasts approximately 12–16 h, during which time there are rapid changes in steroid synthesis, increased expression of proteolytic enzymes, as well as the production of angiogenic factors (1, 2). These and other functional cascades work in concert to ensure the extrusion of a fertilizable oocyte and formation of a corpus luteum (CL). Because the CL has characteristics that suggest terminal differentiation, e.g. growth arrest and hypertrophy, it has long been assumed that luteinization represents the terminal differentiation of granulosa into luteal cells (3). Part of this terminal differentiation process during the periovulatory interval has been hypothesized to consist of a rapid and irreversible arrest of granulosa cell proliferation immediately after the ovulatory surge (2, 4, 5). However, few studies have been undertaken to characterize the cell cycle during the periovulatory interval as a test of this hypothesis.
In essentially all cell types, cell cycle control involves the complex orchestration of a number of core stimulatory and inhibitory components (6). Extracellular mitogenic stimulation is mediated by the D-type cyclins (D1, D2, and D3), which associate and form complexes with the cyclin-dependent kinases (Cdks) Cdk4 or Cdk6. Thus, although the cyclin D/Cdk4,6 complexes control entry into G1 phase, regulators of entry and progression through S phase are cyclin E/Cdk2 complexes. The aforementioned cyclin/Cdk complexes, once activated, phosphorylate retinoblastoma (RB) and its family members, p107 and p130, thus releasing sequestered E2F transcription factors, allowing transcription of genes required for cell cycle progression (7). In contrast to cyclin D expression, which is induced by mitogenic signals, cyclin E is expressed periodically, with maximal cyclin E/Cdk2 activity occurring in late G1 and at the G1-S boundary. The central importance of cyclin E/Cdk2 has been established through the observation that overexpression of cyclin E obviates the need for cyclin D/Cdk4,6 and can shorten the length of G1 (8). Inhibitory components of the cell cycle include the Cdk inhibitors (CKIs), which bind and inhibit the activity of cyclin-Cdk, contributing to cell cycle arrest. Two families of inhibitors are currently known: the Cip/Kip family (p21, p27, and p57) and the INK4 family (p16, p15, p18, and p19) (9).
Luteal cells derived from antecedent granulosa cells have very low levels of proliferation. For example, Rao et al. (3) showed a marked decline in proliferative activity of granulosa-derived luteal cells from rats 24 h after administration of LH. Quiescence of luteal cells in mice is associated with decreased expression of Cdk2 and cyclin D1 and increased levels of p27 and cyclin D3 (10). Similarly, studies in the nonhuman primate ovary during the early luteal phase showed that 3?-hydroxysteroid dehydrogenase-positive cells are essentially nonproliferative (11). Examination of proliferation in the human CL indicates that the majority of cycling cells are stromal, primarily vascular, with only 5–15% of luteal cells staining positive for Ki-67 by the end of the early luteal phase (12). Interestingly, several studies in humans and primates suggest that periovulatory progesterone is a causative factor in granulosa cell cycle arrest (13, 14, 15), whereas others suggest that progesterone plays a later role by maintaining quiescence of granulosa-lutein cells (16).
Although the aforementioned studies have been primarily long-interval (>24 h after an ovulatory stimulus) studies, a few studies have examined granulosa cell proliferation at much earlier and shorter time intervals. Robker and Richards (4) suggested that proliferation of rat granulosa cells is arrested, and cyclin D2 expression decreased, within 4 h of an ovulatory human chorionic gonadotropin (hCG) stimulus, consistent with the hypothesis that rat granulosa cells are irreversibly programmed to become luteal cells within 5–7 h after an ovulatory stimulus (17). However, Hirshfield et al. (18) and Agarwal et al. (19) both found that hCG treatment induces a transient increase in the proportion of rat granulosa cells in S phase. Importantly, the expression of c-myc increases in rat granulosa cells after hCG treatment, suggesting that myc-mediated signaling may play an important role in periovulatory events (19, 20). In rhesus macaques undergoing controlled ovarian stimulation, granulosa cells express significantly less Ki-67 (a marker of proliferating cells) within 12 h of an ovulatory hCG bolus (5). However, this same study also reported an hCG-induced increase in cyclin D2 mRNA and unchanged cyclin E mRNA, whereas cyclin B1 mRNA levels were reduced 12 h after hCG. Furthermore, in vitro luteinization of macaque granulosa cells is associated with transient increases in DNA synthesis and myc expression, suggesting that hCG stimulates additional proliferation before cell cycle arrest occurs (21).
The goal of the present study was to establish the temporal dynamics of granulosa cell proliferation after an ovulatory stimulus to pregnant mare serum gonadotropin (PMSG)-primed immature rats. It is hypothesized that hCG stimulates a limited period of granulosa cell proliferation before the onset of cell cycle arrest.
Materials and Methods
Reagents
PMSG and hCG were obtained from Sigma Chemical Co. (St. Louis, MO). DMEM/F-12 culture medium was from Life Technologies, Inc. (Rockville, MD). Complete protease inhibitors were purchased in tablet form from Roche Molecular Biochemical (Mannheim, Germany). Bromodeoxyuridine (BrdU) was purchased from Sigma, and anti-BrdU from Fitzgerald Industries (Concord, MA). Antibodies against Cdk2, Cdk4, Cdk6, cyclin D2, and tissue inhibitors of metalloproteinases (TIMP)-1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), anti-cyclin E was from LabVision (Fremont, CA), antibody against ERK2 and protein A-agarose were obtained from Upstate (Charlottesville, VA), and anti-glyceradehyde-3-phosphate dehydrogenase was from Abcam (Cambridge, UK). Secondary antibodies were from Sigma or Jackson ImmunoResearch Laboratories (West Grove, PA). SuperSignal West Dura Extended Duration Substrate was from Pierce (Rockford, IL). Histone III S (HIIIS) was from Sigma.
Animals and tissue collection
All animal procedures were approved by the Medical College of Georgia Animal Care and Use Committee and were in accordance with the National Institutes of Health Guide to the Care and Use of Laboratory Animals. Immature (21-d-old) Sprague Dawley rats obtained from Harlan (Madison, WI) were kept in a 12-h light/dark regimen with food and water ad libitum. On postnatal d 26–27, rats were hormonally stimulated with 10 IU PMSG followed 48 h later by 10 IU hCG. Animals were killed at specific time points before (0 h) or up to 12 h after an ovulatory hCG bolus. At the time of killing, trunk blood was collected and serum obtained by centrifugation at 400 x g for 10 min. Ovaries were harvested into cold DMEM/F-12 medium, shed using 25-gauge needles, and then filtered through 150-μm-pore nylon mesh (Sefar America, Inc., Depew, NY) to collect granulosa cells (22). Enriched granulosa cells were pelleted and frozen for isolation of RNA and protein or prepared for flow cytometric analysis.
Flow cytometry
Isolated granulosa cells were washed (2000 rpm for 10 min at 4 C) two times in ice-cold fluorescence-activated cell-sorting (FACS) sample buffer (0.1% glucose/PBS) and resuspended in 100–200 μl FACS sample buffer to obtain a single-cell suspension. Cells were fixed by drop-wise addition of 1 ml ice-cold 70% ethanol while vortexing. Ethanol-fixed cells were stored at 4 C for at least 24 h before propidium iodide (PI) staining. Cells were centrifuged, all but 100–200 μl of ethanol removed, and then treated with 1 ml of PI staining solution (0.1 mg/ml PI and 0.5 mg/ml RNase A in FACS sample buffer). Stained cells were held at room temperature for at least 1 h before FACS analysis. Immediately before analysis, cells were passed through a Falcon 35-μm nylon mesh cell strainer cap (BD Biosciences, Bedford, MA) to remove aggregated cells. Flow cytofluorometric measurements of forward scatter, side scatter, and PI fluorescence were made using a three-color FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA) to determine DNA content. Single cells were chosen for DNA content analysis by gating PI fluorescence on a graph of PI pulse area vs. PI pulse width. Data acquisition was performed using CellQuest (version 3.3) software and data analysis with ModFit LT for Macintosh (version 2.0) software (Verity Software House, Inc., Topsham, ME).
BrdU uptake and immunohistochemistry
Immature rats were hormonally stimulated as above and injected (ip) with 5 mg/kg BrdU 2 h before killing, at which time they were perfused with 4% paraformaldehyde. Ovaries and spleens were removed and stored overnight in 4% paraformaldehyde, paraffin embedded, and sectioned at 3 μm. Sections were deparaffinized and rehydrated in toluene and sequential ethanol washes, respectively. For BrdU immunohistochemistry, slides were trypsinized and treated with HCl for denaturation of the DNA. Sections were then incubated in a 1:100 dilution of sheep polyclonal anti-BrdU for 2 h at room temperature, rinsed in 10 mM PBS, and then incubated in a 1:1000 dilution of Alexa Fluor 488 donkey antisheep (Molecular Probes, Eugene, OR) for 1 h at room temperature. Adjacent sections were stained for 10 min with hematoxylin (Vector Laboratories, Inc., Burlingame, CA) and for 5 min with Eosin (Sigma) and then dehydrated, cleared with Citrisolv (Fisher Scientific, Pittsburgh, PA), and coverslipped. For TIMP-1 immunohistochemistry, deparaffinized, rehydrated slides were pretreated with Target Retrieval Solution (pH 6.0; Dako Corp., Carpinteria, CA.) using a steamer (Black and Decker rice steamer), followed by a distilled water rinse. Endogenous peroxidase was quenched with 0.3% H2O2 in distilled water for 5 min followed by distilled water for 2 min. Slides were incubated in Power Block (Biogenex Laboratories Inc., San Ramon, CA.), rinsed in distilled water, and placed in 1x PBS for 5 min, followed by anti-TIMP-1 (Santa Cruz Biotechnology) at 1:100 for 1 h at room temperature and then peroxidase-conjugated AffinityPure secondary donkey antirabbit for 1 h, and rinsed in two changes of PBS. Detection was with diaminobenzidine substrate (Dako). Slides were counterstained with hematoxylin (Richard-Allan Scientific, Kalamazoo, MI) and visualized using the Olympus IX71 microscope and the MicroFire Imaging System (Olympus).
RNA isolation and RT-PCR
RNA was isolated using the RNAqueous-Micro Isolation Kit (Ambion, Austin, TX) and reverse transcribed, and 20 ng cDNA was used for multiplex real-time PCR (Cepheid, Sunnyvale, CA) using the 60 S ribosomal protein L32 (RPL32) as an internal standard (23). Primers and probes were designed using Primer Express Software (Applied Biosystems, Foster City, CA), and sequences are listed in Table 1.
TABLE 1. Primer/probe sequences
Protein isolation and Western blot analysis
Granulosa cell protein was isolated using F-buffer [10 mM Tris (pH 7.05), 50 mM NaCl, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, 5 mM zinc chloride, 100 mM sodium orthovanadate, 1% triton X-100, and mixed protease inhibitors (one tablet/25 ml complete protease inhibitors)] (24). Cell pellets were resuspended in F-buffer by rapid pipetting and held on ice for 10 min followed by vortexing for 30 sec. Lysates were briefly sonicated and centrifuged at 14,000 rpm at 4 C for 15 min. The supernatant was pipetted into a clean tube, and the protein concentration determined using the BCA protein assay kit (Pierce Chemical Co., Rockford, IL). Equal amounts of protein were separated by PAGE, transferred onto a polyvinylidene difluoride membrane (Millipore, Bedford, MA), blocked with 5% dry nonfat dairy milk, and then probed with respective primary and secondary antibodies. Immunoreactive bands were visualized using SuperSignal West Dura Extended Duration Substrate. Precision Plus Protein Standards (Bio-Rad, Hercules, CA) were run on each gel to identify the protein/band of interest based on size. Blots were reprobed with total ERK2 as a loading control. Total ERK2 was first tested as a loading control by comparing with glyceradehyde-3-phosphate dehydrogenase protein levels on the same blot. Both proteins showed similar profiles (data not presented). In addition, this protein is expressed at equivalent levels in rat granulosa cells before and after hCG (25), and the molecular mass is different from the cyclins/Cdks examined in the current study; thus blots did not require stripping before reprobing.
Cdk2 activity
Protein was isolated before or after hCG using a commercially available lysis buffer (Upstate). Lysates were immunoprecipitated with anti-Cdk2 (2 μg) for 2 h at 4 C, followed by protein A-agarose overnight at 4 C. Precipitates were washed twice in lysis buffer and once in kinase buffer (Upstate). The resulting Cdk2 immunocomplexes were incubated for 1 h at 37 C in kinase buffer, 10 μCi [32P]ATP (Perkin-Elmer Life Sciences, Inc., Boston, MA), and 10 μg HIIIS. 32P-labeled HIIIS was visualized using PAGE on a Typhoon 8600 phosphorimager (Amersham, Piscataway, NJ). The gel was then stained with Coomassie blue, and total HIIIS was used as a loading control.
RIA
Serum progesterone concentrations were determined using the Coat-A-Count progesterone RIA kit per the manufacturer’s specifications (Diagnostic Products Corp., Los Angeles, CA).
Statistical analysis
All data are presented as mean ± SEM. Bartlett’s 2 was used to test for heterogeneity of variance, and data were subsequently logarithmically transformed. Data were analyzed by one-way ANOVA followed by a Student Newman-Keuls’ means test. Differences were considered significant if P < 0.05. Time points with different superscript letters or symbols were significantly different.
Results
Validation of luteinization in the PMSG-primed rat model
Several markers of luteinization were used to ensure response to the hormonal treatment protocol. Serum progesterone concentrations increased significantly (P < 0.05) by 4 h and peaked 6 h after an ovulatory hCG bolus (Fig. 1). Progesterone receptor (PR) mRNA similarly increased (P < 0.05) within 2 h of hCG, with peak expression 4 h after hCG followed by a sharp decline. Cyclooxygenase (COX)-2 and ADAMTS-1 mRNA also showed the expected expression profile after hCG (26, 27, 28).
FIG. 1. Validation of luteinization. Rats were hormonally stimulated with PMSG for 48 h, followed by an ovulatory bolus of hCG for up to 12 h. Serum progesterone () and granulosa cell expression of PR mRNA () increase after an ovulatory bolus of hCG. Serum progesterone was measured by RIA, and PR mRNA by real-time RT-PCR using rpl32 as an internal standard. Superscripts denote significant differences vs. 0 h (#, serum progesterone; *, PR mRNA). Levels of COX-2 and ADAMTS-1 mRNA are depicted on the right. Data are mean ± SEM (n = 4–8 per time point).
S-phase progression of granulosa cells after an ovulatory hCG bolus
Granulosa cells isolated before (0 h) and 2, 4, 8, and 12 h after the hCG bolus were stained with PI, and DNA content was analyzed using flow cytometry (Fig. 2). The percentage of granulosa cells in G0/G1 did not increase significantly until 12 h after hCG. Similarly, a significant decrease (45%; P < 0.05) in the percentage of cells in S phase relative to pre-hCG (0 h) levels was not observed until 12 h after hCG. Interestingly, the percentage of cells in G2/M compared with 0 h hCG increased significantly (177%; P < 0.05) 8 h after hCG before returning to pre-hCG levels at 12 h after hCG.
FIG. 2. Proliferation of periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Animals were hormonally stimulated as described in Materials and Methods, and granulosa cells were isolated before or after an ovulatory hCG bolus for use in FACS analysis. A, Representative histograms of the FACS analysis at different times after hCG. Note the lack of sub-G1 DNA, indicating the absence of apoptotic granulosa cells at any time point. B, Proportion of granulosa cells in G0/G1 (top), S phase (middle), and G2/M (bottom) before and after hCG. Different superscript letters indicate significant differences across time. Data are mean ± SEM (n = 4–8 per time point).
The granulosa cells used for FACS analysis were obtained by shredding the whole ovary before filtration. To ensure that post-hCG proliferation occurred in luteinizing rather than small follicles, BrdU (a thymidine analog) was injected into animals 2 h before killing. Figure 3 depicts localization of BrdU uptake before and after hCG and is representative of three animals per time point. Pre- and periovulatory follicles were determined on the basis of a large, well-developed antrum. Uptake of BrdU staining increased 24 h after PMSG (Fig. 3, B and C). An apparent decline in the number of BrdU-positive mural granulosa cells was observed starting 2 h after hCG, although proliferation of mural granulosa cells was variable until 6–8 h after hCG, with some follicles expressing very low levels of BrdU uptake soon after hCG and others with continued mural proliferation through 8 h (Fig. 3). BrdU uptake in mural granulosa cells was nearly absent in all follicles by 10 h after hCG. However, in some periovulatory follicles during early time points after hCG, and in nearly all of the latter time points through 10 h after hCG, BrdU-positive cells were localized to the cumulus and cumulus stalk region. Adjacent sections stained with hematoxylin and eosin did not contain substantial numbers of pyknotic nuclei, and thus there was not evidence of hCG-induced atresia. Also, to verify that large antral follicles after hCG luteinized, TIMP-1 expression was determined (29). TIMP-1 increased in all large, antral follicles after hCG, indicating the presence of luteinizing follicles (Fig. 3).
FIG. 3. BrdU uptake during the periovulatory interval. Hormonally stimulated animals were injected with the thymidine analog BrdU 2 h before killing before or after PMSG ± hCG, and the ovaries were fixed for Brdu immunohistochemistry (BrdU-positive cells are labeled green). The adjacent hematoxylin/eosin-stained section is depicted below each BrdU-stained section. A, Anti-BrdU-omitted negative control; B, control (no PMSG); C, PMSG 24 h; D, PMSG 48 h; E, hCG 2 h; F, hCG 4 h; G, hCG 6 h; H, hCG 8 h; I, hCG 10 h; J, hCG 24 h; K and L, TIMP-1-labeled follicles before hCG and 8 h after hCG, respectively (TIMP-1 is stained brown). Note that for hCG 2–10 h, two different sections from different animals are presented to demonstrate follicle to follicle variability. Data are representative of three animals per time point and reflect at least two follicles per animal.
Expression of G1/S-phase cyclins and Cdks during luteinization
Figure 4 depicts mRNA levels and Figure 5 shows Western blot analysis of key G1/S-phase genes. Although there was a slight downward trend in the expression of cyclin D2 mRNA beginning at 4 h after administration of hCG, it was not reduced significantly until 12 h after hCG (2.5-fold; P < 0.05). Protein expression of cyclin D2, however, did not change after hCG. Interestingly, mRNA levels of both of the cyclin D2 partners (Cdk4 and Cdk6) increased 6 h after hCG; Cdk4 mRNA remained elevated throughout the 12-h interval (1.7-fold; P < 0.05), whereas Cdk6 mRNA expression was transient, peaking at 6–8 h after hCG (5-fold; P < 0.05). The expression of Cdk4 protein did not change after hCG, whereas Cdk6 protein was nearly undetectable by Western blot before and after hCG (data not presented). Neither cyclin E mRNA nor Cdk2 mRNA levels were altered after hCG. The expression of cyclin E protein increased 6, 8, and 12 h after hCG (2-fold; P < 0.05), whereas values at 10 h were not significantly different from any other time point. There was a tendency for Cdk2 protein to be decreased 2–4 h after hCG (P = 0.11), followed by a trend toward a transient increase at 6 h (P = 0.06). Consistent with these observations of cyclin E and Cdk2 expression, the activity of Cdk2 was reduced 4 h after hCG, and a small increase in activity was observed 6–8 h after hCG (Fig. 6).
FIG. 4. Expression of cyclin D2, Cdk4, Cdk6, cyclin E, and Cdk2 mRNA in periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Animals were stimulated with PMSG for 48 h, followed by hCG for up to 12 h. Granulosa cells were isolated and RNA extracted for use in real-time RT-PCR analysis. Graphs are mRNA levels relative to the internal standard rpl32. Data are mean ± SEM (n = 4–8). Different superscript letters indicate significant differences across time.
FIG. 5. Expression of cyclin D2, Cdk4, cyclin E, and Cdk2 protein in periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Granulosa cells were isolated as in Fig. 4, and whole-cell protein was used in Western blot analysis. Graphs are protein levels normalized to total ERK2. Representative Western blots are shown below each graph. Data are mean ± SEM (n = 3). Different superscript letters indicate significant differences across time.
FIG. 6. Cdk2 activity in periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Granulosa cells were isolated as in Fig. 4, and Cdk2-containing complexes were immunoprecipitated and incubated with the substrate HIIIS in the presence of [32P]ATP. 32P-labeled HIIIS was detected using a phosphorimager, and total HIIIS was detected by Coomassie blue staining. The data in this figure were repeated three times with different samples.
Discussion
The current study was designed to test the hypothesis that granulosa cells from immature rats undergoing hormonal stimulation have additional hCG-stimulated proliferation before cell cycle arrest. After 48 h of PMSG, only approximately 10% of granulosa cells are in S phase, and this does not decline until nearly 12 h after an ovulatory stimulus, although a small but significant increase in the proportion of granulosa cells in G2/M phase was observed 8 h after hCG. Interestingly, the majority of post-hCG proliferation occurs in the region of the cumulus-oocyte complex rather than the mural granulosa layer. Although there is little evidence for dynamic regulation of cyclin D2, Cdk4, or Cdk6 proteins after hCG, the expression of cyclin E protein increases during the periovulatory interval, and there is a tendency, albeit statistically insignificant, for an initial decline followed by a transient increase in Cdk2 protein. Additionally, administration of hCG induces a rapid decline in Cdk2 activity, followed by a transient recovery of activity 4–8 h after hCG. These results suggest that additional proliferation in the cumulus region accompanies luteinization and may be regulated by Cdk2 activity.
Before hCG, approximately 10% of granulosa cells are in S phase. This is consistent with previous reports using normally cycling adult rats and immature animals undergoing hormonal stimulation (18) and indicates that a relatively small fraction of granulosa cells in preovulatory follicles transit S phase at any one time. Before an ovulatory stimulus, 86% of cells are in G0/G1. Although FACS analysis does not differentiate the two, we hypothesize that most, if not all, of these cells are in G1. First, these cells most likely express gonadotropin and growth factor receptors (30) and are therefore exposed to trophic stimulation. Second, in rhesus monkeys undergoing controlled ovarian stimulation, 50% of granulosa cells in pre-hCG follicles express Ki-67 (5). Ki-67 is expressed in G1/S/G2/M but not G0 cells (31), so this estimation of high levels of proliferation may be based primarily on G1 cells rather than S, G2, or M. Third, Chaffkin et al. (14, 15) were able to modulate progesterone levels in cultured human granulosa-lutein cells to reinitiate proliferation, suggesting that these cells are arrested in G1 rather than G0. It is currently unknown what prevents more than 10% of preovulatory granulosa cells from transiting the G1/S-phase boundary, but could reflect subsets of granulosa cells (e.g. basal vs. antral) with an inherently slow growth rate, or perhaps a mechanistic block before the G1 restriction point.
The proportion of granulosa cells in G0/G1 or S phase does not change until 12 h after hCG. Although it is possible that this represents poor hormonal stimulation, multiple markers of luteinization indicate a robust response to hCG. Oonk et al. (17) suggested that cellular reorganization leading to luteinization is complete 5–7 h after hCG; thus in rats, cell cycle arrest is not temporally coordinated with luteinization. The uncoupling of cell cycle arrest and terminal differentiation has been proposed in several models, including p27–/– mice, in which granulosa cells continue to proliferate but are otherwise normal (32). In contrast, completion of luteinization and cell cycle arrest in primate granulosa cells are more closely temporally linked, although no evidence exists for a causal relationship (5). It is possible that granulosa cells from all species require some minimum amount of time to arrest, i.e. more than 12 h, although this hypothesis is untested. However, Hirshfield et al. (18) demonstrated in rats an increase in the proportion of S-phase granulosa cells 5 h after an ovulatory stimulus. Furthermore, Agarwal et al. (19) found an increase in BrdU uptake 4 h after hCG in isolated rat granulosa cells. Although the current study did not detect a clear increase in S phase after hCG, a transient rise in the proportion of granulosa cells in G2/M occurs 8 h after hCG. It is possible that this reflects a final round of proliferation caused by hCG or, alternatively, that hCG can synchronize granulosa cells scattered near late G1/early S to finish the cell cycle together. It is clear, however, that hCG does not induce a rapid, synchronized arrest of granulosa cell proliferation, but rather a much slower, more subtle decline.
Because the FACS data used granulosa cells from all follicles in an ovary, including preantral follicles, it is impossible to assure that post-hCG S-phase cells stem from periovulatory follicles. In addition, it is not possible to reliably isolate granulosa cells for FACS around and after ovulation (i.e. >12 h after hCG); thus FACS is not useful for later time points. Uptake of the thymidine analog BrdU was used to localize proliferation during the periovulatory interval. Although an ovulatory stimulus elicits a general decline in proliferation of mural granulosa cells, substantial follicle to follicle variability exists. This variability may in part reconcile the fact that the proportion of S-phase cells measured by FACS analysis does not decline until 12 h after hCG. Given the existing data set, it is hypothesized that mural granulosa cells in most follicles exit the cell cycle by 4–6 h after hCG, although proliferation in some follicles continues up to 10 h. This variability of proliferation exists within individual ovaries, suggesting that the response of mural granulosa cells to an ovulatory stimulus is not predicated on the endocrine status of the animal, but rather can be attributed to the individual follicle, and may be due to several factors, notably the position of the follicle in the ovary and/or the relative level of follicular development. In contrast to mural granulosa cells, granulosa cells in the cumulus and cumulus stalk region of nearly all follicles continue to proliferate for 10 h after hCG, raising the possibility that cumulus proliferation during the periovulatory interval is very tightly regulated by the ovulatory gonadotropin stimulus.
The differences between mural and cumulus proliferation suggest that the mechanisms of cell cycle control are distinct between the two populations of cells. It seems likely that post-hCG proliferation of mural cells reflects the completion of a round of division initiated before the hCG rather than hCG-induced passage across the S-phase boundary. The continued proliferation of cumulus cells 10 h after hCG raises the possibility that these cells undergo an additional round of proliferation. Although the mitogenic stimulus for this proliferation remains unknown, it could be via the TGF-? family growth factors such as bone morphogenetic protein-15 and growth differentiation factor-9 (33, 34, 35), or epidermal growth factor family members (36). It has recently been reported that growth differentiation factor-9 is expressed in macaque periovulatory follicles (37), and thus it is possible that hCG induces the oocyte to initiate a final burst of proliferation in the region of the cumulus complex via the TGF? family. It is currently unclear why cumulus cells proliferate in response to an ovulatory stimulus, or even if this is necessary for ovulation and luteinization. However, two rounds of proliferation have been shown to be essential to terminal differentiation of 3T3-L1 preadipocytes into mature adipocytes (38). It thus seems likely that hCG-stimulated granulosa cell proliferation is essential for normal ovulation and luteinization to occur, and may be related to cumulus expansion, defects in which have been shown to decrease ovulation efficiency and fertilization (39, 40).
To better understand the temporal cell cycle dynamics of granulosa cells during the periovulatory interval, mRNA and protein levels of some positive cell cycle regulators responsible for G1- to S-phase transition were determined. Cyclin D2 mRNA levels are not reduced significantly until 12 h after hCG, although there is a downward trend beginning at 4 h after hCG. In contrast, there is no indication of changes in cyclin D2 protein levels after hCG. It has been reported that both cyclin D2 mRNA and protein decline precipitously 4 h after hCG (4), although more recent data in growing follicles suggest that changes in granulosa cell proliferation are not driven by changes in cyclin D2 expression (41). In contrast, both Cdk4 and Cdk6 mRNA increase after hCG. Although levels of Cdk6 protein are very low before and after hCG (current study and Ref. 41), Cdk4 may have an important non-cell-cycle role during luteinization, as granulosa cells from Cdk4–/– mice proliferate normally, but luteinization is impaired (42). Furthermore, it is possible that the continued expression of cyclin D2/Cdk4 acts to sequester CKIs increased in response to hCG [i.e. p21 (4)], thereby allowing cyclin E/Cdk2 activity to continue until the expression of specific CKIs exceeds cyclin D2/Cdk4.
The expression of cyclin E and Cdk2 mRNA does not change after hCG, although 6 h after an ovulatory hCG bolus, cyclin E protein increases, whereas Cdk2 protein has a tendency (P = 0.06) to increase. Because protein, but not mRNA, levels change after hCG, it is hypothesized that these gene products are regulated posttranslationally. The increasing expression of cyclin E protein is hypothesized to be a consequence of cells accumulating in the late G1 phase of the cell cycle. Cyclin E is synthesized and accumulates as cells progress through G1, peaking in late G1, and is degraded once cells enter S phase (43). Thus granulosa cells arrested in G1 are expected to maintain expression of cyclin E protein. The pattern of Cdk2 activity follows that of Cdk2 protein, suggesting that Cdk2, but not cyclin E, is a limiting factor in activity of the cyclin E/Cdk 2 complex. Thus, Cdk 2 activity may be suppressed in the latter stages of the periovulatory interval by decreasing Cdk2 protein levels as well as by increasing expression of p21 (4). Because cyclin E/Cdk2 complexes are specific to entry into S phase, it is possible that the observed Cdk2 activity is related to the BrdU uptake observed in cumulus granulosa cells after hCG, although specific data on this point are lacking. Alternatively, there is evidence that cyclin E/Cdk2 can use a variety of proteins as a substrate, including PR (44). The expression of PR mRNA and protein in luteinizing rat granulosa cells (26) coincides with the observed Cdk2 activity during the periovulatory interval, making it tempting to speculate that PR is phosphorylated/activated by Cdk2. Colocalization studies of cyclin E, Cdk2, PR, and proliferating granulosa cells after hCG are thus warranted.
In summary, granulosa cells continue to enter S phase for up to 10 h after an ovulatory stimulus, although the subset of proliferating granulosa cells may shift from mural to cumulus with hCG administration. Little evidence for regulation of cyclin D2 is observed, and although cyclin E/Cdk2 activity is initially suppressed, there is a small transient increase between 4 and 8 h after hCG. It is hypothesized that this transient increase in activity may be accounted for by Cdk2 protein levels and also by cyclin D2/Cdk4 complexes acting to sequester inhibitory proteins such as p21. Overall, these data are consistent with a model in which an ovulatory stimulus is predicted to induce additional proliferation in the luteinizing ovarian follicle.
Acknowledgments
We thank Dr. Tara Swan and Dr. Lynnette McCluskey for technical assistance, Jeanene Pihkala at the MCG Flow Cytometry Core Laboratory for assistance with FACS analyses, and Kimberly Smith at the Medical College of Georgia Auxiliary Research Lab for help with the tissue sectioning and TIMP-1 staining.
References
Richards JS, Russell DL, Ochsner S, Espey LL 2002 Ovulation: new dimensions and new regulators of the inflammatory-like response. Annu Rev Physiol 64:69–92
Murphy BD 2000 Models of luteinization. Biol Reprod 63:2–11
Rao MC, Midgley Jr AR, Richards JS 1978 Hormonal regulation of ovarian cellular proliferation. Cell 14:71–78
Robker RL, Richards JS 1998 Hormone-induced proliferation and differentiation of granulosa cells: a coordinated balance of the cell cycle regulators cyclin D2 and p27Kip1. Mol Endocrinol 12:924–940
Chaffin CL, Schwinof KM, Stouffer RL 2001 Gonadotropin and steroid control of granulosa cell proliferation during the periovulatory interval in rhesus monkeys. Biol Reprod 65:755–762
Johnson DG, Walker CL 1999 Cyclins and cell cycle checkpoints. Annu Rev Pharmacol Toxicol 39:295–312
Coqueret O 2002 Linking cyclins to transcriptional control. Gene 299:35–55
Ledda-Columbano GM, Pibiri M, Concas D, Cossu C, Tripodi M, Columbano A 2002 Loss of cyclin D1 does not inhibit the proliferative response of mouse liver to mitogenic stimuli. Hepatology 36:1098–1105
Sherr CJ, Roberts JM 1999 CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13:1501–1512
Hampl A, Pachernik J, Dvorak P 2000 Levels and interactions of p27, cyclin D3, and CDK4 during the formation and maintenance of the corpus luteum in mice. Biol Reprod 62:1393–1401
Christenson LK, Stouffer RL 1996 Proliferation of microvascular endothelial cells in the primate corpus luteum during the menstrual cycle and simulated early pregnancy. Endocrinology 137:367–374
Gaytan F, Morales C, Garcia-Pardo L, Reymundo C, Bellido C, Sanchez-Criado JE 1998 Macrophages, cell proliferation, and cell death in the human menstrual corpus luteum. Biol Reprod 59:417–425
Ben Rafael Z, Varello MA, Meloni F, Fateh M, Mastroianni Jr L, Flickinger GL 1987 Flow cytometric analysis of deoxyribonucleic acid in human granulosa cells from in vitro fertilization cycles: relationships to oocyte maturity and fertilizability and to follicular fluid steroids. J Clin Endocrinol Metabol 65:602–605
Chaffkin LM, Luciano AA, Peluso JJ 1993 The role of progesterone in regulating human granulosa cell proliferation and differentiation in vitro. J Clin Endocrinol Metab 76:696–700
Chaffkin LM, Luciano AA, Peluso JJ 1992 Progesterone as an autocrine/paracrine regulator of human granulosa cell proliferation. J Clin Endocrinol Metab 75:1404–1408
Chaffin CL, Stouffer RL 2002 Local role of progesterone in the ovary during the periovulatory interval. Rev Endocr Metab Disord 3:65–72
Oonk RB, Krasnow JS, Beattie WG, Richards JS 1989 Cyclic AMP-dependent and -independent regulation of cholesterol side chain cleavage cytochrome P-450 (P-450scc) in rat ovarian granulosa cells and corpora lutea: cDNA and deduced amino acid sequence of rat P-450scc. J Biol Chem 264:21934–21942
Hirshfield AN, Flickinger GL, Ben Rafael Z 1988 Flow cytofluorometric analysis of granulosa cell proliferation in rats. J Reprod Fertil 84:231–238
Agarwal P, Peluso JJ, White BA 1996 Steroidogenic factor-1 expression is transiently repressed and c-myc expression and deoxyribonucleic acid synthesis are induced in rat granulosa cells during the periovulatory period. Biol Reprod 55:1271–1275
Piontkewitz Y, Sundfeldt K, Hedin L 1997 The expression of c-myc during follicular growth and luteal formation in the rat ovary in vivo. J Endocrinol 152:395–406
Chaffin CL, Brogan RS, Stouffer RL, VandeVoort CA 2003 Dynamics of Myc/Max/Mad expression during luteinization of primate granulosa cells in vitro: association with periovulatory proliferation. Endocrinology 144:1249–1256
Campbell KL 1979 Ovarian granulosa cells isolated with EGTA and hypertonic sucrose: cellular integrity and function. Biol Reprod 21:773–786
Lovekamp TN, Davis BJ 2001 Mono-(2-ethylhexyl) phthalate suppresses aromatase transcript levels and estradiol production in cultured rat granulosa cells. Toxicol Appl Pharmacol 172:217–224
Sommer A, Bousset K, Kremmer E, Austen M, Luscher B 1998 Identification and characterization of specific DNA-binding complexes containing members of the Myc/Max/Mad network of transcriptional regulators. J Biol Chem 273:6632–6642
Seger R, Hanoch T, Rosenberg R, Dantes A, Merz WE, Strauss III JF, Amsterdam A 2001 The ERK signaling cascade inhibits gonadotropin-stimulated steroidogenesis. J Biol Chem 276:13957–13964
Park OK, Mayo KE 1991 Transient expression of progesterone receptor messenger RNA in ovarian granulosa cells after the preovulatory luteinizing hormone surge. Mol Endocrinol 5:967–978
Sirois J, Simmons DL, Richards JS 1992 Hormonal regulation of messenger ribonucleic acid encoding a novel isoform of prostaglandin endoperoxide H synthase in rat preovulatory follicles. Induction in vivo and in vitro. J Biol Chem 267:11586–11592
Espey LL, Yoshioka S, Russell DL, Robker RL, Fujii S, Richards JS 2000 Ovarian expression of a disintegrin and metalloproteinase with thrombospondin motifs during ovulation in the gonadotropin-primed immature rat. Biol Reprod 62:1090–1095
Hagglund AC, Ny A, Leonardsson G, Ny T 1999 Regulation and localization of matrix metalloproteinases and tissue inhibitors of metalloproteinases in the mouse ovary during gonadotropin-induced ovulation. Endocrinology 140:4351–4358
Hillier SG 2001 Gonadotropic control of ovarian follicular growth and development. Mol Cell Endocrinol 179:39–46
Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U, Stein H 1984 Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133:1710–1715
Jirawatnotai S, Moons DS, Stocco CO, Franks R, Hales DB, Gibori G, Kiyokawa H 2003 The cyclin-dependent kinase inhibitors p27Kip1 and p21Cip1 cooperate to restrict proliferative life span in differentiating ovarian cells. J Biol Chem 278:17021–17027
Nilsson EE, Skinner MK 2002 Growth and differentiation factor-9 stimulates progression of early primary but not primordial rat ovarian follicle development. Biol Reprod 67:1018–1024
Vitt UA, Hayashi M, Klein C, Hsueh AJ 2000 Growth differentiation factor-9 stimulates proliferation but suppresses the follicle-stimulating hormone-induced differentiation of cultured granulosa cells from small antral and preovulatory rat follicles. Biol Reprod 62:370–377
Otsuka F, Yao Z, Lee T, Yamamoto S, Erickson GF, Shimasaki S 2000 Bone morphogenetic protein-15: identification of target cells and biological functions. J Biol Chem 275:39523–39528
Park JY, Su YQ, Ariga M, Law E, Jin SL, Conti M 2004 EGF-like growth factors as mediators of LH action in the ovulatory follicle. Science 303:682–684
Duffy DM 2003 Growth differentiation factor-9 is expressed by the primate follicle throughout the periovulatory interval. Biol Reprod 69:725–732
Tang QQ, Otto TC, Lane MD 2003 CCAAT/enhancer-binding protein ? is required for mitotic clonal expansion during adipogenesis. Proc Natl Acad Sci USA 100:850–855
Chen L, Russell PT, Larsen WJ 1993 Functional significance of cumulus expansion in the mouse: roles for the preovulatory synthesis of hyaluronic acid within the cumulus mass. Mol Reprod Dev 34:87–93
Zhuo L, Kimata K 2001 Cumulus oophorus extracellular matrix: its construction and regulation. Cell Struct Funct 26:189–196
Zhang YM, Roy SK 2004 Downregulation of follicle-stimulating hormone (FSH)-receptor messenger RNA levels in the hamster ovary: effect of the endogenous and exogenous FSH. Biol Reprod 70:1580–1588
Moons DS, Jirawatnotai S, Tsutsui T, Franks R, Parlow AF, Hales DB, Gibori G, Fazleabas AT, Kiyokawa H 2002 Intact follicular maturation and defective luteal function in mice deficient for cyclin-dependent kinase-4. Endocrinology 143:647–654
Boonstra J 2003 Progression through the G1-phase of the on-going cell cycle. J Cell Biochem 90:244–252
Knotts TA, Orkiszewski RS, Cook RG, Edwards DP, Weigel NL 2001 Identification of a phosphorylation site in the hinge region of the human progesterone receptor and additional amino-terminal phosphorylation sites. J Biol Chem 276:8475–8483(Jennifer D. Cannon, Mary )
Address all correspondence and requests for reprints to: Charles L. Chaffin, Ph.D., Department of Physiology, Medical College of Georgia, CA2098, 1120 15th Street, Augusta, Georgia 30912. E-mail: cchaffin@mail.mcg.edu.
Abstract
Granulosa cell proliferation during luteinization and terminal differentiation has historically been assumed to decline rapidly after an ovulatory stimulus. In contrast, terminal differentiation in other cell types has recently been associated with a transient increase in proliferation, suggesting that this may occur in the ovarian follicle. The goal of the current study was to test the hypothesis that an ovulatory stimulus to rats results in additional granulosa cell proliferation before cell cycle arrest. Immature rats were given a single injection of pregnant mare serum gonadotropin (PMSG) followed by human chorionic gonadotropin (hCG) to initiate periovulatory events. The proportion of granulosa cells in S phase did not change until 12 h after hCG, although the majority of the post-hCG proliferation was localized to cumulus granulosa cells for up to 10 h after hCG. The expression of cyclin D2 mRNA did not decline until 12 h after hCG, although both cyclin-dependent kinase (Cdk)4 and Cdk6 mRNA increased at 6 h. Protein levels of cyclin D2 and Cdk4 did not change as a result of hCG, whereas cyclin E increased 6 h after hCG. Kinase activity of Cdk2 dropped markedly by 4 h after hCG, but a slight increase in activity was evident 6–8 h after hCG. These data suggest that cumulus granulosa cells continue to proliferate for up to 10 h after an ovulatory stimulus, possibly via cyclin E/Cdk2. It is concluded that proliferation is maintained in granulosa cells in the proximity of the oocyte during luteinization of the rat follicle.
Introduction
THE PERIOVULATORY INTERVAL is defined as the onset of an ovulatory gonadotropin stimulus until follicle rupture. In rats, this interval lasts approximately 12–16 h, during which time there are rapid changes in steroid synthesis, increased expression of proteolytic enzymes, as well as the production of angiogenic factors (1, 2). These and other functional cascades work in concert to ensure the extrusion of a fertilizable oocyte and formation of a corpus luteum (CL). Because the CL has characteristics that suggest terminal differentiation, e.g. growth arrest and hypertrophy, it has long been assumed that luteinization represents the terminal differentiation of granulosa into luteal cells (3). Part of this terminal differentiation process during the periovulatory interval has been hypothesized to consist of a rapid and irreversible arrest of granulosa cell proliferation immediately after the ovulatory surge (2, 4, 5). However, few studies have been undertaken to characterize the cell cycle during the periovulatory interval as a test of this hypothesis.
In essentially all cell types, cell cycle control involves the complex orchestration of a number of core stimulatory and inhibitory components (6). Extracellular mitogenic stimulation is mediated by the D-type cyclins (D1, D2, and D3), which associate and form complexes with the cyclin-dependent kinases (Cdks) Cdk4 or Cdk6. Thus, although the cyclin D/Cdk4,6 complexes control entry into G1 phase, regulators of entry and progression through S phase are cyclin E/Cdk2 complexes. The aforementioned cyclin/Cdk complexes, once activated, phosphorylate retinoblastoma (RB) and its family members, p107 and p130, thus releasing sequestered E2F transcription factors, allowing transcription of genes required for cell cycle progression (7). In contrast to cyclin D expression, which is induced by mitogenic signals, cyclin E is expressed periodically, with maximal cyclin E/Cdk2 activity occurring in late G1 and at the G1-S boundary. The central importance of cyclin E/Cdk2 has been established through the observation that overexpression of cyclin E obviates the need for cyclin D/Cdk4,6 and can shorten the length of G1 (8). Inhibitory components of the cell cycle include the Cdk inhibitors (CKIs), which bind and inhibit the activity of cyclin-Cdk, contributing to cell cycle arrest. Two families of inhibitors are currently known: the Cip/Kip family (p21, p27, and p57) and the INK4 family (p16, p15, p18, and p19) (9).
Luteal cells derived from antecedent granulosa cells have very low levels of proliferation. For example, Rao et al. (3) showed a marked decline in proliferative activity of granulosa-derived luteal cells from rats 24 h after administration of LH. Quiescence of luteal cells in mice is associated with decreased expression of Cdk2 and cyclin D1 and increased levels of p27 and cyclin D3 (10). Similarly, studies in the nonhuman primate ovary during the early luteal phase showed that 3?-hydroxysteroid dehydrogenase-positive cells are essentially nonproliferative (11). Examination of proliferation in the human CL indicates that the majority of cycling cells are stromal, primarily vascular, with only 5–15% of luteal cells staining positive for Ki-67 by the end of the early luteal phase (12). Interestingly, several studies in humans and primates suggest that periovulatory progesterone is a causative factor in granulosa cell cycle arrest (13, 14, 15), whereas others suggest that progesterone plays a later role by maintaining quiescence of granulosa-lutein cells (16).
Although the aforementioned studies have been primarily long-interval (>24 h after an ovulatory stimulus) studies, a few studies have examined granulosa cell proliferation at much earlier and shorter time intervals. Robker and Richards (4) suggested that proliferation of rat granulosa cells is arrested, and cyclin D2 expression decreased, within 4 h of an ovulatory human chorionic gonadotropin (hCG) stimulus, consistent with the hypothesis that rat granulosa cells are irreversibly programmed to become luteal cells within 5–7 h after an ovulatory stimulus (17). However, Hirshfield et al. (18) and Agarwal et al. (19) both found that hCG treatment induces a transient increase in the proportion of rat granulosa cells in S phase. Importantly, the expression of c-myc increases in rat granulosa cells after hCG treatment, suggesting that myc-mediated signaling may play an important role in periovulatory events (19, 20). In rhesus macaques undergoing controlled ovarian stimulation, granulosa cells express significantly less Ki-67 (a marker of proliferating cells) within 12 h of an ovulatory hCG bolus (5). However, this same study also reported an hCG-induced increase in cyclin D2 mRNA and unchanged cyclin E mRNA, whereas cyclin B1 mRNA levels were reduced 12 h after hCG. Furthermore, in vitro luteinization of macaque granulosa cells is associated with transient increases in DNA synthesis and myc expression, suggesting that hCG stimulates additional proliferation before cell cycle arrest occurs (21).
The goal of the present study was to establish the temporal dynamics of granulosa cell proliferation after an ovulatory stimulus to pregnant mare serum gonadotropin (PMSG)-primed immature rats. It is hypothesized that hCG stimulates a limited period of granulosa cell proliferation before the onset of cell cycle arrest.
Materials and Methods
Reagents
PMSG and hCG were obtained from Sigma Chemical Co. (St. Louis, MO). DMEM/F-12 culture medium was from Life Technologies, Inc. (Rockville, MD). Complete protease inhibitors were purchased in tablet form from Roche Molecular Biochemical (Mannheim, Germany). Bromodeoxyuridine (BrdU) was purchased from Sigma, and anti-BrdU from Fitzgerald Industries (Concord, MA). Antibodies against Cdk2, Cdk4, Cdk6, cyclin D2, and tissue inhibitors of metalloproteinases (TIMP)-1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), anti-cyclin E was from LabVision (Fremont, CA), antibody against ERK2 and protein A-agarose were obtained from Upstate (Charlottesville, VA), and anti-glyceradehyde-3-phosphate dehydrogenase was from Abcam (Cambridge, UK). Secondary antibodies were from Sigma or Jackson ImmunoResearch Laboratories (West Grove, PA). SuperSignal West Dura Extended Duration Substrate was from Pierce (Rockford, IL). Histone III S (HIIIS) was from Sigma.
Animals and tissue collection
All animal procedures were approved by the Medical College of Georgia Animal Care and Use Committee and were in accordance with the National Institutes of Health Guide to the Care and Use of Laboratory Animals. Immature (21-d-old) Sprague Dawley rats obtained from Harlan (Madison, WI) were kept in a 12-h light/dark regimen with food and water ad libitum. On postnatal d 26–27, rats were hormonally stimulated with 10 IU PMSG followed 48 h later by 10 IU hCG. Animals were killed at specific time points before (0 h) or up to 12 h after an ovulatory hCG bolus. At the time of killing, trunk blood was collected and serum obtained by centrifugation at 400 x g for 10 min. Ovaries were harvested into cold DMEM/F-12 medium, shed using 25-gauge needles, and then filtered through 150-μm-pore nylon mesh (Sefar America, Inc., Depew, NY) to collect granulosa cells (22). Enriched granulosa cells were pelleted and frozen for isolation of RNA and protein or prepared for flow cytometric analysis.
Flow cytometry
Isolated granulosa cells were washed (2000 rpm for 10 min at 4 C) two times in ice-cold fluorescence-activated cell-sorting (FACS) sample buffer (0.1% glucose/PBS) and resuspended in 100–200 μl FACS sample buffer to obtain a single-cell suspension. Cells were fixed by drop-wise addition of 1 ml ice-cold 70% ethanol while vortexing. Ethanol-fixed cells were stored at 4 C for at least 24 h before propidium iodide (PI) staining. Cells were centrifuged, all but 100–200 μl of ethanol removed, and then treated with 1 ml of PI staining solution (0.1 mg/ml PI and 0.5 mg/ml RNase A in FACS sample buffer). Stained cells were held at room temperature for at least 1 h before FACS analysis. Immediately before analysis, cells were passed through a Falcon 35-μm nylon mesh cell strainer cap (BD Biosciences, Bedford, MA) to remove aggregated cells. Flow cytofluorometric measurements of forward scatter, side scatter, and PI fluorescence were made using a three-color FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA) to determine DNA content. Single cells were chosen for DNA content analysis by gating PI fluorescence on a graph of PI pulse area vs. PI pulse width. Data acquisition was performed using CellQuest (version 3.3) software and data analysis with ModFit LT for Macintosh (version 2.0) software (Verity Software House, Inc., Topsham, ME).
BrdU uptake and immunohistochemistry
Immature rats were hormonally stimulated as above and injected (ip) with 5 mg/kg BrdU 2 h before killing, at which time they were perfused with 4% paraformaldehyde. Ovaries and spleens were removed and stored overnight in 4% paraformaldehyde, paraffin embedded, and sectioned at 3 μm. Sections were deparaffinized and rehydrated in toluene and sequential ethanol washes, respectively. For BrdU immunohistochemistry, slides were trypsinized and treated with HCl for denaturation of the DNA. Sections were then incubated in a 1:100 dilution of sheep polyclonal anti-BrdU for 2 h at room temperature, rinsed in 10 mM PBS, and then incubated in a 1:1000 dilution of Alexa Fluor 488 donkey antisheep (Molecular Probes, Eugene, OR) for 1 h at room temperature. Adjacent sections were stained for 10 min with hematoxylin (Vector Laboratories, Inc., Burlingame, CA) and for 5 min with Eosin (Sigma) and then dehydrated, cleared with Citrisolv (Fisher Scientific, Pittsburgh, PA), and coverslipped. For TIMP-1 immunohistochemistry, deparaffinized, rehydrated slides were pretreated with Target Retrieval Solution (pH 6.0; Dako Corp., Carpinteria, CA.) using a steamer (Black and Decker rice steamer), followed by a distilled water rinse. Endogenous peroxidase was quenched with 0.3% H2O2 in distilled water for 5 min followed by distilled water for 2 min. Slides were incubated in Power Block (Biogenex Laboratories Inc., San Ramon, CA.), rinsed in distilled water, and placed in 1x PBS for 5 min, followed by anti-TIMP-1 (Santa Cruz Biotechnology) at 1:100 for 1 h at room temperature and then peroxidase-conjugated AffinityPure secondary donkey antirabbit for 1 h, and rinsed in two changes of PBS. Detection was with diaminobenzidine substrate (Dako). Slides were counterstained with hematoxylin (Richard-Allan Scientific, Kalamazoo, MI) and visualized using the Olympus IX71 microscope and the MicroFire Imaging System (Olympus).
RNA isolation and RT-PCR
RNA was isolated using the RNAqueous-Micro Isolation Kit (Ambion, Austin, TX) and reverse transcribed, and 20 ng cDNA was used for multiplex real-time PCR (Cepheid, Sunnyvale, CA) using the 60 S ribosomal protein L32 (RPL32) as an internal standard (23). Primers and probes were designed using Primer Express Software (Applied Biosystems, Foster City, CA), and sequences are listed in Table 1.
TABLE 1. Primer/probe sequences
Protein isolation and Western blot analysis
Granulosa cell protein was isolated using F-buffer [10 mM Tris (pH 7.05), 50 mM NaCl, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, 5 mM zinc chloride, 100 mM sodium orthovanadate, 1% triton X-100, and mixed protease inhibitors (one tablet/25 ml complete protease inhibitors)] (24). Cell pellets were resuspended in F-buffer by rapid pipetting and held on ice for 10 min followed by vortexing for 30 sec. Lysates were briefly sonicated and centrifuged at 14,000 rpm at 4 C for 15 min. The supernatant was pipetted into a clean tube, and the protein concentration determined using the BCA protein assay kit (Pierce Chemical Co., Rockford, IL). Equal amounts of protein were separated by PAGE, transferred onto a polyvinylidene difluoride membrane (Millipore, Bedford, MA), blocked with 5% dry nonfat dairy milk, and then probed with respective primary and secondary antibodies. Immunoreactive bands were visualized using SuperSignal West Dura Extended Duration Substrate. Precision Plus Protein Standards (Bio-Rad, Hercules, CA) were run on each gel to identify the protein/band of interest based on size. Blots were reprobed with total ERK2 as a loading control. Total ERK2 was first tested as a loading control by comparing with glyceradehyde-3-phosphate dehydrogenase protein levels on the same blot. Both proteins showed similar profiles (data not presented). In addition, this protein is expressed at equivalent levels in rat granulosa cells before and after hCG (25), and the molecular mass is different from the cyclins/Cdks examined in the current study; thus blots did not require stripping before reprobing.
Cdk2 activity
Protein was isolated before or after hCG using a commercially available lysis buffer (Upstate). Lysates were immunoprecipitated with anti-Cdk2 (2 μg) for 2 h at 4 C, followed by protein A-agarose overnight at 4 C. Precipitates were washed twice in lysis buffer and once in kinase buffer (Upstate). The resulting Cdk2 immunocomplexes were incubated for 1 h at 37 C in kinase buffer, 10 μCi [32P]ATP (Perkin-Elmer Life Sciences, Inc., Boston, MA), and 10 μg HIIIS. 32P-labeled HIIIS was visualized using PAGE on a Typhoon 8600 phosphorimager (Amersham, Piscataway, NJ). The gel was then stained with Coomassie blue, and total HIIIS was used as a loading control.
RIA
Serum progesterone concentrations were determined using the Coat-A-Count progesterone RIA kit per the manufacturer’s specifications (Diagnostic Products Corp., Los Angeles, CA).
Statistical analysis
All data are presented as mean ± SEM. Bartlett’s 2 was used to test for heterogeneity of variance, and data were subsequently logarithmically transformed. Data were analyzed by one-way ANOVA followed by a Student Newman-Keuls’ means test. Differences were considered significant if P < 0.05. Time points with different superscript letters or symbols were significantly different.
Results
Validation of luteinization in the PMSG-primed rat model
Several markers of luteinization were used to ensure response to the hormonal treatment protocol. Serum progesterone concentrations increased significantly (P < 0.05) by 4 h and peaked 6 h after an ovulatory hCG bolus (Fig. 1). Progesterone receptor (PR) mRNA similarly increased (P < 0.05) within 2 h of hCG, with peak expression 4 h after hCG followed by a sharp decline. Cyclooxygenase (COX)-2 and ADAMTS-1 mRNA also showed the expected expression profile after hCG (26, 27, 28).
FIG. 1. Validation of luteinization. Rats were hormonally stimulated with PMSG for 48 h, followed by an ovulatory bolus of hCG for up to 12 h. Serum progesterone () and granulosa cell expression of PR mRNA () increase after an ovulatory bolus of hCG. Serum progesterone was measured by RIA, and PR mRNA by real-time RT-PCR using rpl32 as an internal standard. Superscripts denote significant differences vs. 0 h (#, serum progesterone; *, PR mRNA). Levels of COX-2 and ADAMTS-1 mRNA are depicted on the right. Data are mean ± SEM (n = 4–8 per time point).
S-phase progression of granulosa cells after an ovulatory hCG bolus
Granulosa cells isolated before (0 h) and 2, 4, 8, and 12 h after the hCG bolus were stained with PI, and DNA content was analyzed using flow cytometry (Fig. 2). The percentage of granulosa cells in G0/G1 did not increase significantly until 12 h after hCG. Similarly, a significant decrease (45%; P < 0.05) in the percentage of cells in S phase relative to pre-hCG (0 h) levels was not observed until 12 h after hCG. Interestingly, the percentage of cells in G2/M compared with 0 h hCG increased significantly (177%; P < 0.05) 8 h after hCG before returning to pre-hCG levels at 12 h after hCG.
FIG. 2. Proliferation of periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Animals were hormonally stimulated as described in Materials and Methods, and granulosa cells were isolated before or after an ovulatory hCG bolus for use in FACS analysis. A, Representative histograms of the FACS analysis at different times after hCG. Note the lack of sub-G1 DNA, indicating the absence of apoptotic granulosa cells at any time point. B, Proportion of granulosa cells in G0/G1 (top), S phase (middle), and G2/M (bottom) before and after hCG. Different superscript letters indicate significant differences across time. Data are mean ± SEM (n = 4–8 per time point).
The granulosa cells used for FACS analysis were obtained by shredding the whole ovary before filtration. To ensure that post-hCG proliferation occurred in luteinizing rather than small follicles, BrdU (a thymidine analog) was injected into animals 2 h before killing. Figure 3 depicts localization of BrdU uptake before and after hCG and is representative of three animals per time point. Pre- and periovulatory follicles were determined on the basis of a large, well-developed antrum. Uptake of BrdU staining increased 24 h after PMSG (Fig. 3, B and C). An apparent decline in the number of BrdU-positive mural granulosa cells was observed starting 2 h after hCG, although proliferation of mural granulosa cells was variable until 6–8 h after hCG, with some follicles expressing very low levels of BrdU uptake soon after hCG and others with continued mural proliferation through 8 h (Fig. 3). BrdU uptake in mural granulosa cells was nearly absent in all follicles by 10 h after hCG. However, in some periovulatory follicles during early time points after hCG, and in nearly all of the latter time points through 10 h after hCG, BrdU-positive cells were localized to the cumulus and cumulus stalk region. Adjacent sections stained with hematoxylin and eosin did not contain substantial numbers of pyknotic nuclei, and thus there was not evidence of hCG-induced atresia. Also, to verify that large antral follicles after hCG luteinized, TIMP-1 expression was determined (29). TIMP-1 increased in all large, antral follicles after hCG, indicating the presence of luteinizing follicles (Fig. 3).
FIG. 3. BrdU uptake during the periovulatory interval. Hormonally stimulated animals were injected with the thymidine analog BrdU 2 h before killing before or after PMSG ± hCG, and the ovaries were fixed for Brdu immunohistochemistry (BrdU-positive cells are labeled green). The adjacent hematoxylin/eosin-stained section is depicted below each BrdU-stained section. A, Anti-BrdU-omitted negative control; B, control (no PMSG); C, PMSG 24 h; D, PMSG 48 h; E, hCG 2 h; F, hCG 4 h; G, hCG 6 h; H, hCG 8 h; I, hCG 10 h; J, hCG 24 h; K and L, TIMP-1-labeled follicles before hCG and 8 h after hCG, respectively (TIMP-1 is stained brown). Note that for hCG 2–10 h, two different sections from different animals are presented to demonstrate follicle to follicle variability. Data are representative of three animals per time point and reflect at least two follicles per animal.
Expression of G1/S-phase cyclins and Cdks during luteinization
Figure 4 depicts mRNA levels and Figure 5 shows Western blot analysis of key G1/S-phase genes. Although there was a slight downward trend in the expression of cyclin D2 mRNA beginning at 4 h after administration of hCG, it was not reduced significantly until 12 h after hCG (2.5-fold; P < 0.05). Protein expression of cyclin D2, however, did not change after hCG. Interestingly, mRNA levels of both of the cyclin D2 partners (Cdk4 and Cdk6) increased 6 h after hCG; Cdk4 mRNA remained elevated throughout the 12-h interval (1.7-fold; P < 0.05), whereas Cdk6 mRNA expression was transient, peaking at 6–8 h after hCG (5-fold; P < 0.05). The expression of Cdk4 protein did not change after hCG, whereas Cdk6 protein was nearly undetectable by Western blot before and after hCG (data not presented). Neither cyclin E mRNA nor Cdk2 mRNA levels were altered after hCG. The expression of cyclin E protein increased 6, 8, and 12 h after hCG (2-fold; P < 0.05), whereas values at 10 h were not significantly different from any other time point. There was a tendency for Cdk2 protein to be decreased 2–4 h after hCG (P = 0.11), followed by a trend toward a transient increase at 6 h (P = 0.06). Consistent with these observations of cyclin E and Cdk2 expression, the activity of Cdk2 was reduced 4 h after hCG, and a small increase in activity was observed 6–8 h after hCG (Fig. 6).
FIG. 4. Expression of cyclin D2, Cdk4, Cdk6, cyclin E, and Cdk2 mRNA in periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Animals were stimulated with PMSG for 48 h, followed by hCG for up to 12 h. Granulosa cells were isolated and RNA extracted for use in real-time RT-PCR analysis. Graphs are mRNA levels relative to the internal standard rpl32. Data are mean ± SEM (n = 4–8). Different superscript letters indicate significant differences across time.
FIG. 5. Expression of cyclin D2, Cdk4, cyclin E, and Cdk2 protein in periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Granulosa cells were isolated as in Fig. 4, and whole-cell protein was used in Western blot analysis. Graphs are protein levels normalized to total ERK2. Representative Western blots are shown below each graph. Data are mean ± SEM (n = 3). Different superscript letters indicate significant differences across time.
FIG. 6. Cdk2 activity in periovulatory granulosa cells before (0 h) and up to 12 h after hCG. Granulosa cells were isolated as in Fig. 4, and Cdk2-containing complexes were immunoprecipitated and incubated with the substrate HIIIS in the presence of [32P]ATP. 32P-labeled HIIIS was detected using a phosphorimager, and total HIIIS was detected by Coomassie blue staining. The data in this figure were repeated three times with different samples.
Discussion
The current study was designed to test the hypothesis that granulosa cells from immature rats undergoing hormonal stimulation have additional hCG-stimulated proliferation before cell cycle arrest. After 48 h of PMSG, only approximately 10% of granulosa cells are in S phase, and this does not decline until nearly 12 h after an ovulatory stimulus, although a small but significant increase in the proportion of granulosa cells in G2/M phase was observed 8 h after hCG. Interestingly, the majority of post-hCG proliferation occurs in the region of the cumulus-oocyte complex rather than the mural granulosa layer. Although there is little evidence for dynamic regulation of cyclin D2, Cdk4, or Cdk6 proteins after hCG, the expression of cyclin E protein increases during the periovulatory interval, and there is a tendency, albeit statistically insignificant, for an initial decline followed by a transient increase in Cdk2 protein. Additionally, administration of hCG induces a rapid decline in Cdk2 activity, followed by a transient recovery of activity 4–8 h after hCG. These results suggest that additional proliferation in the cumulus region accompanies luteinization and may be regulated by Cdk2 activity.
Before hCG, approximately 10% of granulosa cells are in S phase. This is consistent with previous reports using normally cycling adult rats and immature animals undergoing hormonal stimulation (18) and indicates that a relatively small fraction of granulosa cells in preovulatory follicles transit S phase at any one time. Before an ovulatory stimulus, 86% of cells are in G0/G1. Although FACS analysis does not differentiate the two, we hypothesize that most, if not all, of these cells are in G1. First, these cells most likely express gonadotropin and growth factor receptors (30) and are therefore exposed to trophic stimulation. Second, in rhesus monkeys undergoing controlled ovarian stimulation, 50% of granulosa cells in pre-hCG follicles express Ki-67 (5). Ki-67 is expressed in G1/S/G2/M but not G0 cells (31), so this estimation of high levels of proliferation may be based primarily on G1 cells rather than S, G2, or M. Third, Chaffkin et al. (14, 15) were able to modulate progesterone levels in cultured human granulosa-lutein cells to reinitiate proliferation, suggesting that these cells are arrested in G1 rather than G0. It is currently unknown what prevents more than 10% of preovulatory granulosa cells from transiting the G1/S-phase boundary, but could reflect subsets of granulosa cells (e.g. basal vs. antral) with an inherently slow growth rate, or perhaps a mechanistic block before the G1 restriction point.
The proportion of granulosa cells in G0/G1 or S phase does not change until 12 h after hCG. Although it is possible that this represents poor hormonal stimulation, multiple markers of luteinization indicate a robust response to hCG. Oonk et al. (17) suggested that cellular reorganization leading to luteinization is complete 5–7 h after hCG; thus in rats, cell cycle arrest is not temporally coordinated with luteinization. The uncoupling of cell cycle arrest and terminal differentiation has been proposed in several models, including p27–/– mice, in which granulosa cells continue to proliferate but are otherwise normal (32). In contrast, completion of luteinization and cell cycle arrest in primate granulosa cells are more closely temporally linked, although no evidence exists for a causal relationship (5). It is possible that granulosa cells from all species require some minimum amount of time to arrest, i.e. more than 12 h, although this hypothesis is untested. However, Hirshfield et al. (18) demonstrated in rats an increase in the proportion of S-phase granulosa cells 5 h after an ovulatory stimulus. Furthermore, Agarwal et al. (19) found an increase in BrdU uptake 4 h after hCG in isolated rat granulosa cells. Although the current study did not detect a clear increase in S phase after hCG, a transient rise in the proportion of granulosa cells in G2/M occurs 8 h after hCG. It is possible that this reflects a final round of proliferation caused by hCG or, alternatively, that hCG can synchronize granulosa cells scattered near late G1/early S to finish the cell cycle together. It is clear, however, that hCG does not induce a rapid, synchronized arrest of granulosa cell proliferation, but rather a much slower, more subtle decline.
Because the FACS data used granulosa cells from all follicles in an ovary, including preantral follicles, it is impossible to assure that post-hCG S-phase cells stem from periovulatory follicles. In addition, it is not possible to reliably isolate granulosa cells for FACS around and after ovulation (i.e. >12 h after hCG); thus FACS is not useful for later time points. Uptake of the thymidine analog BrdU was used to localize proliferation during the periovulatory interval. Although an ovulatory stimulus elicits a general decline in proliferation of mural granulosa cells, substantial follicle to follicle variability exists. This variability may in part reconcile the fact that the proportion of S-phase cells measured by FACS analysis does not decline until 12 h after hCG. Given the existing data set, it is hypothesized that mural granulosa cells in most follicles exit the cell cycle by 4–6 h after hCG, although proliferation in some follicles continues up to 10 h. This variability of proliferation exists within individual ovaries, suggesting that the response of mural granulosa cells to an ovulatory stimulus is not predicated on the endocrine status of the animal, but rather can be attributed to the individual follicle, and may be due to several factors, notably the position of the follicle in the ovary and/or the relative level of follicular development. In contrast to mural granulosa cells, granulosa cells in the cumulus and cumulus stalk region of nearly all follicles continue to proliferate for 10 h after hCG, raising the possibility that cumulus proliferation during the periovulatory interval is very tightly regulated by the ovulatory gonadotropin stimulus.
The differences between mural and cumulus proliferation suggest that the mechanisms of cell cycle control are distinct between the two populations of cells. It seems likely that post-hCG proliferation of mural cells reflects the completion of a round of division initiated before the hCG rather than hCG-induced passage across the S-phase boundary. The continued proliferation of cumulus cells 10 h after hCG raises the possibility that these cells undergo an additional round of proliferation. Although the mitogenic stimulus for this proliferation remains unknown, it could be via the TGF-? family growth factors such as bone morphogenetic protein-15 and growth differentiation factor-9 (33, 34, 35), or epidermal growth factor family members (36). It has recently been reported that growth differentiation factor-9 is expressed in macaque periovulatory follicles (37), and thus it is possible that hCG induces the oocyte to initiate a final burst of proliferation in the region of the cumulus complex via the TGF? family. It is currently unclear why cumulus cells proliferate in response to an ovulatory stimulus, or even if this is necessary for ovulation and luteinization. However, two rounds of proliferation have been shown to be essential to terminal differentiation of 3T3-L1 preadipocytes into mature adipocytes (38). It thus seems likely that hCG-stimulated granulosa cell proliferation is essential for normal ovulation and luteinization to occur, and may be related to cumulus expansion, defects in which have been shown to decrease ovulation efficiency and fertilization (39, 40).
To better understand the temporal cell cycle dynamics of granulosa cells during the periovulatory interval, mRNA and protein levels of some positive cell cycle regulators responsible for G1- to S-phase transition were determined. Cyclin D2 mRNA levels are not reduced significantly until 12 h after hCG, although there is a downward trend beginning at 4 h after hCG. In contrast, there is no indication of changes in cyclin D2 protein levels after hCG. It has been reported that both cyclin D2 mRNA and protein decline precipitously 4 h after hCG (4), although more recent data in growing follicles suggest that changes in granulosa cell proliferation are not driven by changes in cyclin D2 expression (41). In contrast, both Cdk4 and Cdk6 mRNA increase after hCG. Although levels of Cdk6 protein are very low before and after hCG (current study and Ref. 41), Cdk4 may have an important non-cell-cycle role during luteinization, as granulosa cells from Cdk4–/– mice proliferate normally, but luteinization is impaired (42). Furthermore, it is possible that the continued expression of cyclin D2/Cdk4 acts to sequester CKIs increased in response to hCG [i.e. p21 (4)], thereby allowing cyclin E/Cdk2 activity to continue until the expression of specific CKIs exceeds cyclin D2/Cdk4.
The expression of cyclin E and Cdk2 mRNA does not change after hCG, although 6 h after an ovulatory hCG bolus, cyclin E protein increases, whereas Cdk2 protein has a tendency (P = 0.06) to increase. Because protein, but not mRNA, levels change after hCG, it is hypothesized that these gene products are regulated posttranslationally. The increasing expression of cyclin E protein is hypothesized to be a consequence of cells accumulating in the late G1 phase of the cell cycle. Cyclin E is synthesized and accumulates as cells progress through G1, peaking in late G1, and is degraded once cells enter S phase (43). Thus granulosa cells arrested in G1 are expected to maintain expression of cyclin E protein. The pattern of Cdk2 activity follows that of Cdk2 protein, suggesting that Cdk2, but not cyclin E, is a limiting factor in activity of the cyclin E/Cdk 2 complex. Thus, Cdk 2 activity may be suppressed in the latter stages of the periovulatory interval by decreasing Cdk2 protein levels as well as by increasing expression of p21 (4). Because cyclin E/Cdk2 complexes are specific to entry into S phase, it is possible that the observed Cdk2 activity is related to the BrdU uptake observed in cumulus granulosa cells after hCG, although specific data on this point are lacking. Alternatively, there is evidence that cyclin E/Cdk2 can use a variety of proteins as a substrate, including PR (44). The expression of PR mRNA and protein in luteinizing rat granulosa cells (26) coincides with the observed Cdk2 activity during the periovulatory interval, making it tempting to speculate that PR is phosphorylated/activated by Cdk2. Colocalization studies of cyclin E, Cdk2, PR, and proliferating granulosa cells after hCG are thus warranted.
In summary, granulosa cells continue to enter S phase for up to 10 h after an ovulatory stimulus, although the subset of proliferating granulosa cells may shift from mural to cumulus with hCG administration. Little evidence for regulation of cyclin D2 is observed, and although cyclin E/Cdk2 activity is initially suppressed, there is a small transient increase between 4 and 8 h after hCG. It is hypothesized that this transient increase in activity may be accounted for by Cdk2 protein levels and also by cyclin D2/Cdk4 complexes acting to sequester inhibitory proteins such as p21. Overall, these data are consistent with a model in which an ovulatory stimulus is predicted to induce additional proliferation in the luteinizing ovarian follicle.
Acknowledgments
We thank Dr. Tara Swan and Dr. Lynnette McCluskey for technical assistance, Jeanene Pihkala at the MCG Flow Cytometry Core Laboratory for assistance with FACS analyses, and Kimberly Smith at the Medical College of Georgia Auxiliary Research Lab for help with the tissue sectioning and TIMP-1 staining.
References
Richards JS, Russell DL, Ochsner S, Espey LL 2002 Ovulation: new dimensions and new regulators of the inflammatory-like response. Annu Rev Physiol 64:69–92
Murphy BD 2000 Models of luteinization. Biol Reprod 63:2–11
Rao MC, Midgley Jr AR, Richards JS 1978 Hormonal regulation of ovarian cellular proliferation. Cell 14:71–78
Robker RL, Richards JS 1998 Hormone-induced proliferation and differentiation of granulosa cells: a coordinated balance of the cell cycle regulators cyclin D2 and p27Kip1. Mol Endocrinol 12:924–940
Chaffin CL, Schwinof KM, Stouffer RL 2001 Gonadotropin and steroid control of granulosa cell proliferation during the periovulatory interval in rhesus monkeys. Biol Reprod 65:755–762
Johnson DG, Walker CL 1999 Cyclins and cell cycle checkpoints. Annu Rev Pharmacol Toxicol 39:295–312
Coqueret O 2002 Linking cyclins to transcriptional control. Gene 299:35–55
Ledda-Columbano GM, Pibiri M, Concas D, Cossu C, Tripodi M, Columbano A 2002 Loss of cyclin D1 does not inhibit the proliferative response of mouse liver to mitogenic stimuli. Hepatology 36:1098–1105
Sherr CJ, Roberts JM 1999 CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13:1501–1512
Hampl A, Pachernik J, Dvorak P 2000 Levels and interactions of p27, cyclin D3, and CDK4 during the formation and maintenance of the corpus luteum in mice. Biol Reprod 62:1393–1401
Christenson LK, Stouffer RL 1996 Proliferation of microvascular endothelial cells in the primate corpus luteum during the menstrual cycle and simulated early pregnancy. Endocrinology 137:367–374
Gaytan F, Morales C, Garcia-Pardo L, Reymundo C, Bellido C, Sanchez-Criado JE 1998 Macrophages, cell proliferation, and cell death in the human menstrual corpus luteum. Biol Reprod 59:417–425
Ben Rafael Z, Varello MA, Meloni F, Fateh M, Mastroianni Jr L, Flickinger GL 1987 Flow cytometric analysis of deoxyribonucleic acid in human granulosa cells from in vitro fertilization cycles: relationships to oocyte maturity and fertilizability and to follicular fluid steroids. J Clin Endocrinol Metabol 65:602–605
Chaffkin LM, Luciano AA, Peluso JJ 1993 The role of progesterone in regulating human granulosa cell proliferation and differentiation in vitro. J Clin Endocrinol Metab 76:696–700
Chaffkin LM, Luciano AA, Peluso JJ 1992 Progesterone as an autocrine/paracrine regulator of human granulosa cell proliferation. J Clin Endocrinol Metab 75:1404–1408
Chaffin CL, Stouffer RL 2002 Local role of progesterone in the ovary during the periovulatory interval. Rev Endocr Metab Disord 3:65–72
Oonk RB, Krasnow JS, Beattie WG, Richards JS 1989 Cyclic AMP-dependent and -independent regulation of cholesterol side chain cleavage cytochrome P-450 (P-450scc) in rat ovarian granulosa cells and corpora lutea: cDNA and deduced amino acid sequence of rat P-450scc. J Biol Chem 264:21934–21942
Hirshfield AN, Flickinger GL, Ben Rafael Z 1988 Flow cytofluorometric analysis of granulosa cell proliferation in rats. J Reprod Fertil 84:231–238
Agarwal P, Peluso JJ, White BA 1996 Steroidogenic factor-1 expression is transiently repressed and c-myc expression and deoxyribonucleic acid synthesis are induced in rat granulosa cells during the periovulatory period. Biol Reprod 55:1271–1275
Piontkewitz Y, Sundfeldt K, Hedin L 1997 The expression of c-myc during follicular growth and luteal formation in the rat ovary in vivo. J Endocrinol 152:395–406
Chaffin CL, Brogan RS, Stouffer RL, VandeVoort CA 2003 Dynamics of Myc/Max/Mad expression during luteinization of primate granulosa cells in vitro: association with periovulatory proliferation. Endocrinology 144:1249–1256
Campbell KL 1979 Ovarian granulosa cells isolated with EGTA and hypertonic sucrose: cellular integrity and function. Biol Reprod 21:773–786
Lovekamp TN, Davis BJ 2001 Mono-(2-ethylhexyl) phthalate suppresses aromatase transcript levels and estradiol production in cultured rat granulosa cells. Toxicol Appl Pharmacol 172:217–224
Sommer A, Bousset K, Kremmer E, Austen M, Luscher B 1998 Identification and characterization of specific DNA-binding complexes containing members of the Myc/Max/Mad network of transcriptional regulators. J Biol Chem 273:6632–6642
Seger R, Hanoch T, Rosenberg R, Dantes A, Merz WE, Strauss III JF, Amsterdam A 2001 The ERK signaling cascade inhibits gonadotropin-stimulated steroidogenesis. J Biol Chem 276:13957–13964
Park OK, Mayo KE 1991 Transient expression of progesterone receptor messenger RNA in ovarian granulosa cells after the preovulatory luteinizing hormone surge. Mol Endocrinol 5:967–978
Sirois J, Simmons DL, Richards JS 1992 Hormonal regulation of messenger ribonucleic acid encoding a novel isoform of prostaglandin endoperoxide H synthase in rat preovulatory follicles. Induction in vivo and in vitro. J Biol Chem 267:11586–11592
Espey LL, Yoshioka S, Russell DL, Robker RL, Fujii S, Richards JS 2000 Ovarian expression of a disintegrin and metalloproteinase with thrombospondin motifs during ovulation in the gonadotropin-primed immature rat. Biol Reprod 62:1090–1095
Hagglund AC, Ny A, Leonardsson G, Ny T 1999 Regulation and localization of matrix metalloproteinases and tissue inhibitors of metalloproteinases in the mouse ovary during gonadotropin-induced ovulation. Endocrinology 140:4351–4358
Hillier SG 2001 Gonadotropic control of ovarian follicular growth and development. Mol Cell Endocrinol 179:39–46
Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U, Stein H 1984 Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133:1710–1715
Jirawatnotai S, Moons DS, Stocco CO, Franks R, Hales DB, Gibori G, Kiyokawa H 2003 The cyclin-dependent kinase inhibitors p27Kip1 and p21Cip1 cooperate to restrict proliferative life span in differentiating ovarian cells. J Biol Chem 278:17021–17027
Nilsson EE, Skinner MK 2002 Growth and differentiation factor-9 stimulates progression of early primary but not primordial rat ovarian follicle development. Biol Reprod 67:1018–1024
Vitt UA, Hayashi M, Klein C, Hsueh AJ 2000 Growth differentiation factor-9 stimulates proliferation but suppresses the follicle-stimulating hormone-induced differentiation of cultured granulosa cells from small antral and preovulatory rat follicles. Biol Reprod 62:370–377
Otsuka F, Yao Z, Lee T, Yamamoto S, Erickson GF, Shimasaki S 2000 Bone morphogenetic protein-15: identification of target cells and biological functions. J Biol Chem 275:39523–39528
Park JY, Su YQ, Ariga M, Law E, Jin SL, Conti M 2004 EGF-like growth factors as mediators of LH action in the ovulatory follicle. Science 303:682–684
Duffy DM 2003 Growth differentiation factor-9 is expressed by the primate follicle throughout the periovulatory interval. Biol Reprod 69:725–732
Tang QQ, Otto TC, Lane MD 2003 CCAAT/enhancer-binding protein ? is required for mitotic clonal expansion during adipogenesis. Proc Natl Acad Sci USA 100:850–855
Chen L, Russell PT, Larsen WJ 1993 Functional significance of cumulus expansion in the mouse: roles for the preovulatory synthesis of hyaluronic acid within the cumulus mass. Mol Reprod Dev 34:87–93
Zhuo L, Kimata K 2001 Cumulus oophorus extracellular matrix: its construction and regulation. Cell Struct Funct 26:189–196
Zhang YM, Roy SK 2004 Downregulation of follicle-stimulating hormone (FSH)-receptor messenger RNA levels in the hamster ovary: effect of the endogenous and exogenous FSH. Biol Reprod 70:1580–1588
Moons DS, Jirawatnotai S, Tsutsui T, Franks R, Parlow AF, Hales DB, Gibori G, Fazleabas AT, Kiyokawa H 2002 Intact follicular maturation and defective luteal function in mice deficient for cyclin-dependent kinase-4. Endocrinology 143:647–654
Boonstra J 2003 Progression through the G1-phase of the on-going cell cycle. J Cell Biochem 90:244–252
Knotts TA, Orkiszewski RS, Cook RG, Edwards DP, Weigel NL 2001 Identification of a phosphorylation site in the hinge region of the human progesterone receptor and additional amino-terminal phosphorylation sites. J Biol Chem 276:8475–8483(Jennifer D. Cannon, Mary )