Destruction of the Germinal Disc Region of an Immature Preovulatory Chicken Follicle Induces Atresia and Apoptosis 1

Transcription

1 BIOLOGY OF REPRODUCTION 59, (1998) Destruction of the Germinal Disc Region of an Immature Preovulatory Chicken Follicle Induces Atresia and Apoptosis 1 Humphrey Hung-Chang Yao, Kendra K. Volentine, and Janice M. Bahr 2 Department of Animal Sciences, University of Illinois at Urbana-Champaign, Urbana, Illinois ABSTRACT The germinal disc region (GDR), which contains the germinal disc and overlying granulosa cells, is essential for completion of maturation of the preovulatory chicken follicle. The current study was conducted to test the hypothesis that destruction of the GDR (GDRX) of an immature preovulatory chicken follicle blocks ovulation, induces apoptosis, and causes atresia. The GDR of immature preovulatory follicles (F2) were destroyed by freezing with dry ice ( mm in diameter) h before ovulation. As a control for the effect of freezing, a nongdr portion (a portion of the follicular wall opposite to the GDR relative to the follicular stalk) of other F2 follicles were destroyed (non- GDRX). Treatment of F2 follicles by GDRX caused atresia and blocked ovulation of all treated follicles (6 of 6), whereas none of the nongdrx follicles (0 of 5) underwent atresia. Treatment of follicles by GDRX induced apoptotic DNA fragmentation (laddering) in theca and granulosa layers obtained from the frozen area and in the theca layer obtained from the follicular wall distal to the frozen area. In contrast, apoptosis was only present in theca and granulosa layers in the frozen area of the nongdrx follicle. Furthermore, the in situ DNA end-labeling technique demonstrated that in the GDRX follicle 24 h after treatment, cells in the theca interna, endothelial cells in blood vessels of the theca externa, and a few granulosa cells underwent apoptosis. These results indicate that destruction of the GDR of an immature preovulatory follicle causes atresia and apoptosis and blocks ovulation. These novel findings suggest that the GDR maintains development of the chicken preovulatory follicle by producing one or more survival factors. Without the GDR, chicken follicles cannot develop further and they eventually die. INTRODUCTION Oocytes have long been considered to have a passive role in the growth and development of follicles. However, evidence from several species suggests that the oocyte communicates with its surrounding cells and regulates functions and development of various follicular compartments. In the rabbit, removal of the oocyte induced spontaneous luteinization of granulosa cells in the antral follicle [1]. Inhibition of spontaneous luteinization of cultured rat granulosa cells by coculturing with the oocyte suggested that the oocytes secrete factors that prevent spontaneous luteinization [2]. In the rodent, factors secreted by oocytes stimulated cumulus expansion in response to FSH stimulation [], regulated granulosa steroidogenesis [4], stimulated granulosa proliferation [5], and inhibited plasminogen activator production by granulosa cells [6]. Murine oocytes also suppressed ex- Accepted March 9, Received December 15, Supported by National Science Foundation (NSF IBN and IBN ) and the Cooperative State Research, Education and Extension Service, U.S. Department of Agriculture (#5 24). This work was presented at the 0th annual meeting of the Society for the Study of Reproduction, Portland, Oregon, 1997; Abstract 17. H.H.-C.Y. and K.K.V. contributed equal amount of work on this project. 2 Correspondence: Janice M. Bahr, ASL 26, 1207 West Gregory Drive, Urbana, IL FAX: (217) 8286; j-bahr@uiuc.edu 516 pression of LH receptor mrna in cultured granulosa cells [7]. Furthermore, murine follicles that did not express the growth differentiation factor-9 (GDF-9) failed to develop beyond the primary stage [8]. These observations clearly indicate that mammalian oocytes are actively involved in regulating growth and differentiation of follicles. In the chicken, the germinal disc is the portion of the oocyte that contains the nucleus and the majority of the cytoplasmic organelles (Fig. 1). The germinal disc and its overlying granulosa cells are associated through gap junctions and interdigitations [9] and form a unit called the germinal disc region (GDR). The GDR is considered the growth center of the chicken follicle [10]. In the preovulatory chicken follicle, granulosa cells close to the GDR are highly proliferative, whereas granulosa cells distal to the GDR are more differentiated [10 12]. The GDR produces one or more heat- and protease-sensitive factors, which can stimulate proliferation of and decrease progesterone production by granulosa cells [12]. Furthermore, destruction of the GDR of the most mature follicle (F1) 24 h before its ovulation resulted in atresia and failure to ovulate [1]. These results demonstrate that the GDR, similar to the mammalian oocyte, actively participates in regulating the development of chicken follicles. The previous study demonstrating that destruction of the GDR of the most mature follicle (F1) caused follicular atresia [1] did not eliminate the possibility that a follicle might ovulate without the GDR if the follicle is allowed a longer recovery period after destruction of the GDR. This observation prompted us to examine whether an immature and fast-growing preovulatory follicle, such as the second largest follicle (F2), can survive without the GDR. The current experiments were conducted to investigate 1) whether destruction of the germinal disc region (GDRX) of an immature preovulatory follicle causes atresia and blocks ovulation and 2) whether GDRX induces apoptosis of treated follicles. Apoptosis is a programmed cell death that occurs during normal follicular atresia in many species [14] and has become a well-characterized indicator of dependency of tissues on growth factors. A characterization of the degenerating process of the follicle after GDRX will provide significant insight into the manner in which the GDR influences the development of the preovulatory chicken follicle. MATERIALS AND METHODS Animals Single-comb white Leghorn hens in their first year of reproductive age and with laying clutches of at least five eggs were used. Hens were housed individually and provided feed and water ad libitum. The lighting schedule was 17L:7D, with lights-on at 000 h. The times of oviposition (egg laying) were monitored daily at 1-h intervals between 0800 h and 1200 h and once at 1700 h for late oviposition.

2 GERMINAL DISC REGION AND FOLLICULAR APOPTOSIS IN CHICKENS 517 Experiment 1: Does GDRX of an Immature Preovulatory Follicle Cause Atresia and Block Ovulation? Eleven hens were randomly assigned to GDRX (n 6) and nongdrx (n 5) groups. In the GDRX group, the GDR of the F2 follicle was frozen whereas in the non- GDRX group, a nongdr portion of the follicular wall opposite to the GDR relative to the follicular stalk was frozen (see Fig. 1). At 72 h after treatments, ovaries were exposed to examine the presence of atretic follicles. To investigate whether treated follicles ovulated after treatments, the time of oviposition (an indicator of ovulation) for all treated hens was monitored for 72 h after surgery. The expected time of ovulation and oviposition of the F2 follicle were approximately 48 h and 72 h after surgery, respectively. Ovulation of treated F2 follicles was confirmed by the presence of a red mark on the postovulatory follicle caused by freezing. If ovulation failed to occur, the treated follicle would have become atretic and had a red mark on the follicular wall. FIG. 1. Cross section of the chicken preovulatory follicle. The follicle consists of an oocyte and its surrounding follicular wall. The oocyte contains a large amount of yolk and the germinal disc, which appears as a white plaque about mm in diameter on the surface of the oocyte. The germinal disc and its overlying granulosa cells are tightly associated and form a unit called the germinal disc region. A granulosa layer and a theca layer, which can be separated into theca interna and externa, enclose the oocyte. The nongdr is the follicular wall opposite to the GDR relative to the follicular stalk. (Modified from [12]). Surgery Hens were injected i.v. with 0.8 ml of Sudan Black B (0.5% in 1:1 ethanol-pbs) h before surgery to stain the yolk for better visualization of the germinal disc. One hour after oviposition, laparotomy was performed under sodium pentobarbital anesthesia (65 mg/ml, mg/kg BW; Sigma Chemical Company, St. Louis, MO). The second largest follicle (F2) was exposed, and the GDR or a nongdr (a portion of the follicular wall opposite to the GDR relative to the follicular stalk) was frozen by applying a piece of solid CO 2 ( mm in diameter) for 20 sec (Fig. 1). In the sham-treated group, surgeries were performed and F2 follicles were exposed but no freezing was applied. Follicles were rinsed with warm 0.9% saline to prevent dehydration. Birds were then sutured and bacitracin was applied to the incision. Birds regained full consciousness within 15 0 min after surgery. Tissue Collection At 24 or 72 h after surgery, F2 follicles were removed and put immediately into ice-cold PBS. Granulosa and theca layers were separated as previously described [15] and snap frozen in dry ice for DNA isolation. For histochemical analysis, whole follicles were fixed in 10% neutral buffered formalin for 24 h immediately after collection. Experiment 2: Does GDRX Induce Apoptosis of Treated Follicles? Twenty-eight birds were randomly assigned to four treatment groups: control, sham, nongdrx, and GDRX. The control group (n 4) received no surgery. In the sham group (n 4), surgery was performed and the F2 follicle was exposed, but no area of the follicular wall was frozen. The nongdrx (n 8) and GDRX (n 12) groups received the same treatments as in experiment 1. Treated F2 follicles were collected at either 0 h or 24 h after treatments. Granulosa and theca samples (15 mm in diameter) were obtained from the GDR and the nongdr of the four treatment groups for DNA isolation. For in situ apoptosis detection, entire follicles were collected. DNA Extraction and Electrophoresis The DNA extraction procedure was adapted from Tilly and Hsueh [16] with minor modifications. Briefly, tissues were homogenized with a Polytron (Kinematica, Luzern, Switzerland) at setting 1 in 0. ml of homogenization buffer (0.1 M sodium chloride, 0.01 M EDTA, 0.2 M sucrose, and 0. M Tris-HCl, ph 8). Homogenates were then processed as previously described [16]. After extraction and quantification of DNA, 10 to 15 g of DNA was loaded onto to a 2% agarose (Sigma) gel with ethidium bromide (0.5 g/ ml) and separated by electrophoresis for.5 4 h at 50 volts using TAE (40 mm Tris-acetate, 1 mm EDTA) as running buffer. Gels were destained overnight in double-distilled water and visualized and photographed on a UV transilluminator. In Situ Apoptosis Detection ApopTag Plus In Situ Apoptosis Detection Kit-Peroxidase (Oncor, Gaitherburg, MD) was used to recognize apoptotic cells. In brief, follicles were collected and fixed in 10% neutral buffered formalin for 0 60 min at room temperature. The yolk was removed by a pipette through an incision in the follicular wall, and follicles were fixed for 24 h more. A section of the follicular wall that did not contain the germinal disc was embedded in Paraplast (Brunswick Company, St. Louis, MO) after conventional dehydration with ethanol. Consecutive paraffin sections (5 7 m) were placed on poly-l-lysine coated glass slides. Deparaffinized sections were incubated with pepsin (0.5% in PBS, ph 2.0; Sigma) for 15 0 min at room temperature. Exposed -OH ends of DNA fragments in sections were labeled with digoxigenin-11-(d)-uridine triphosphate by terminal deoxynucleotidyl transferase (TdT). Incorporated nucleotides were localized with antidigoxigenin antibody-peroxidase conjugates. Slides were counterstained with methyl green for 25 sec, dehydrated in a graded ethanol series, cleared in xylene, and mounted. Sections of follicles without TdT reaction and sections of involuting mouse mammary glands were used as negative and positive controls for apoptosis, respectively. Statistical Analysis Chi-square analysis was performed for statistical comparison between groups. Differences were considered statistically significant when p 0.05.

3 518 YAO ET AL. TABLE 1. Ovulation records of F1, F2, and F follicles after F2 nongdrx and GDRX. Treatments a No. of birds Day 1 (F1) b Day 2 (F2) b Day (F) c F2 nongdrx F2 GDRX 6 6 a nongdrx, destruction of a nongdr portion of the follicular wall; GDRX, destruction of the GDR. b Ovulation of F1 and F2 follicles occurred at Day 1 and Day 2 after treatments, respectively, and was determined by oviposition (egg laying occurring approximately 24 h after ovulation) on Day 2 and Day after treatments. c Ovulation of F follicles was determined by the presence of eggs in the oviduct on Day after treatments. 0 6 FIG. 2. A) Percentage of atretic F2 follicles after destruction of the GDR (GDRX) or a nongdr of the follicular wall (nongdrx). B) Percentage of F2 follicles ovulating after destruction of the GDR (GDRX) or a nongdr of the follicular wall (nongdrx). The sample size for GDRX and non- GDRX groups is 6 and 5, respectively. Stars indicate statistical significance (p 0.05) between treatments. RESULTS Experiment 1: Does GDRX of an Immature Preovulatory Follicle Cause Atresia and Block Ovulation? In the GDRX group, all treated follicles (6 of 6) were atretic at 72 h after treatment. The follicular wall had holes and was hemorrhagic. Follicles decreased in size due to leakage of yolk. However, none of the F2 follicles (0 of 5) in the nongdrx group showed any sign of atresia at 72 h after treatment (Fig. 2A). Other preovulatory follicles (i.e., the largest [F1], third largest [F] follicles, etc.) in both GDRX and nongdrx groups were normal and showed no sign of atresia. Ovulation of F2 follicles was determined by monitoring oviposition (egg laying), which is an indicator that ovulation has occurred approximately 24 h earlier. No birds (0 of 6) in the GDRX group oviposited on the day that F2 follicles should have been laid. The presence of an atretic follicle with a red mark caused by freezing verified that treated F2 follicles underwent atresia and did not ovulate after GDRX. In the nongdrx group, 80% (4 of 5) of birds oviposited at the expected time (Fig. 2B). These birds had a red mark on the postovulatory follicle, the ovarian tissue remaining after ovulation. This observation indicated that F2 follicles in nongdrx-treated birds ovulated normally. The F1 and F follicles in GDRX-treated birds ovulated at their expected time (about 24 h and 72 h after treatments of the F2 follicle, respectively, Table 1). To determine whether destruction of the germinal disc or the overlying follicular wall, or a combination of these two (GDR), is responsible for atresia of treated follicles, we froze the follicular wall 2 mm distal to the germinal disc of F2 follicles. These follicles were normal and showed no signs of atresia (0 of 6, data not shown) 24 h after treatment. FIG.. Analysis of total cellular DNA isolated from granulosa and theca layers of control (Cont), nongdrx, and GDRX follicles 24 h after treatments. A) Diagram illustrates the sources of follicular wall samples (15 mm in diameter) that were removed from control, nongdrx, and GDRX follicles. The cross mark (X) indicates the area destroyed by freezing. Granulosa and theca samples were isolated from the GDR (G) and nongdr (N) of the follicular wall. B) Ethidium bromidestained gel of DNA isolated from granulosa samples. C) Ethidium bromide-stained gel of DNA isolated from theca samples. The molecular weight marker (M) is a 100-base-pair ladder. Experiments were repeated more than five times.

4 GERMINAL DISC REGION AND FOLLICULAR APOPTOSIS IN CHICKENS 519 FIG. 4. Detection of apoptotic cells by in situ DNA -end labeling. A D) Sections of follicular walls that do not contain the germinal disc. A) A section of the follicular wall of a control F2 follicle. B) A section of the follicular wall of the F2 follicle with a nongdr destroyed (nongdrx). C) A section of the follicular wall of the F2 follicle with GDR destroyed (GDRX). D) A consecutive section of C stained in the absence of the end-labeling enzyme (TdT) as the negative control. Staining for apoptotic cells (arrows) was detected in the theca interna (TI) of the GDRX follicle (C) only. Staining of apoptotic cells was abolished when the end-labeling enzyme was omitted (D). All sections were counterstained with methyl green Bar 10 m. Sections shown are representatives of four experiments. Experiment 2: Does GDRX Induce Apoptosis of Treated Follicles? To determine whether GDRX of F2 follicles induced apoptosis, the presence of internucleosomal DNA fragmentation (laddering) in follicles was determined at 0 and 24 h after treatments by gel electrophoresis. No DNA laddering was present in granulosa and theca layers of the control (Fig., B and C, Cont) or the sham (data not shown) groups. Apoptotic DNA laddering was detected only in granulosa samples that contained the frozen area (Fig. B, N in the nongdrx and G in the GDRX groups), not in samples collected from the nonfrozen area (Fig. B, G in the nongdrx and N in the GDRX groups). Identical results were found in theca samples of nongdrx follicles (Fig. C). However, apoptotic DNA laddering was detected in the theca layer of GDRX follicles that contained the frozen and the nonfrozen areas (Fig. C, G and N of the GDRXgroup). Tissues collected immediately after treatments (0 h) had no DNA fragmentation (data not shown). The presence of internucleosomal DNA fragmentation in F2 follicles 24 h after treatments was also examined in situ by a DNA -end-labeling technique. We examined the sections of follicular walls that did not contain the germinal disc. In agreement with the results of gel electrophoresis of extracted DNA, apoptotic cells were not present in granulosa and theca layers of control follicles (Fig. 4A) or in the follicular wall of nongdrx follicles (Fig. 4B). In the GDRX follicle, staining of apoptotic cells (Fig. 4C, arrows) was detected mainly in the theca interna and infrequently in the blood vessels of the theca externa. The absence of staining for apoptotic DNA when the DNA end-labeling enzyme (TdT) was omitted demonstrates the specificity of the labeling system for apoptosis (Fig. 4D). Dramatic morphological changes were observed in the follicular wall of

5 520 YAO ET AL. GDRX follicles. The granulosa layer (Fig. 4C, Gr) was detached from the theca interna (Fig. 4C, TI) due to accumulation of fluid between the two layers. The theca interna was disorganized and cells were loosely dispersed. The theca externa (Fig. 4C, TE) was increased in thickness. Most of the yolk had leaked out of the follicle through holes in the follicular wall and accumulated between the theca externa and loose connective tissue surrounding the theca externa. Staining of apoptotic cells in the positive control tissue (involuting mouse mammary glands) demonstrated the specificity of the in situ labeling system for apoptosis (data not shown). DISCUSSION The major findings of this research are that destruction of the GDR of an immature, fast-growing preovulatory chicken follicle (F2) causes follicular atresia, blocks ovulation, and induces apoptosis. In the chicken, normal follicular atresia occurs frequently in follicles smaller than 8 mm in diameter, whereas the incidence of atresia in preovulatory follicles is rare [17]. Atresia in preovulatory follicles can be induced by adenohypophysectomy [18], suggesting that gonadotropins are required for maintaining follicular growth. Yoshimura et. al. [1] showed that destruction of the GDR of the most mature follicle (F1) 24 h before its ovulation resulted in follicular atresia and failure of ovulation. In contrast, destruction of the GDR (GDRX) of an F1 follicle 12 h before ovulation did not cause atresia, and the follicle ovulated at the expected time. The lack of an effect of GDRX in the latter study may be attributed to the F1 follicle s being sufficiently mature to ovulate at 12 h before ovulation [19]. The GDR is apparently required for the completion of maturation of the F1 follicle, but a fully mature F1 follicle will ovulate without the GDR. However, the possibility remains that a follicle could ovulate without the GDR if it is allowed to recover for a longer time after destruction of the GDR. Our results indicate conclusively that destruction of the GDR of an immature preovulatory follicle (F2) caused atresia and blocked its ovulation. All birds in which the F2 follicles experienced GDRX missed oviposition on the day the treated F2 follicles should have been laid. Atresia induced by GDRX is irreversible even though the treated follicle is growing rapidly and is allowed a longer recovery period prior to its expected ovulation. The presence of an atretic follicle with a red mark caused by freezing verified that the atretic follicle was the treated F2 follicle. The GDRX treatment affected only the treated follicle and did not influence the development of the other preovulatory follicles. All F1 and F follicles in birds in which the F2 follicle had GDRX, ovulated at the expected times. The F follicles did not ovulate prematurely to replace the atretic GDRX F2. Destruction of a nongdr region of the follicular wall, even though causing localized apoptosis, had no detrimental effects on further follicular development or ovulation. The presence of internucleosomal DNA fragmentation in atretic GDRX follicles implies that apoptosis is the underlying mechanism for GDRX-induced atresia. Interestingly, apoptosis was most prominent in the entire theca layer but was limited only to the frozen area of the granulosa layer of GDRX follicles. Even at 24 h after GDRX when the follicle was visibly atretic, the granulosa layer, with the exception of the frozen area, had not undergone apoptosis. The theca layer underwent dramatic histological changes after GDRX and is apparently more susceptible to the deleterious effect of GDRX than the granulosa layer. Fragmented DNA was present only in the frozen area of nongdrx follicles, apparently caused by localized hypothermia [20]. A chicken follicle can survive localized apoptosis in a nongdr and still ovulate. However, the follicle dies without a viable GDR. This observation highlights the importance of the GDR to chicken follicles. Apoptosis occurs during normal follicular atresia in pigs, chickens, and rodents [14]. The mechanism for the induction of follicular apoptosis is still unknown. One explanation may be that follicular atresia is a process of selecting the most competent follicles for ovulation. Growth and development of follicles are under the control of trophic factors such as gonadotropins from the pituitary and growth factors from various ovarian compartments. Follicles incompetent to respond to these trophic factors during a critical stage of development will be eliminated through the process of apoptosis [21]. Our finding suggests that the germinal disc of chicken follicles is involved in orchestrating growth and development of the follicle along with the pituitary and other ovarian compartments. The germinal disc might communicate with other follicular compartments via paracrine factors. The identity of the germinal disc factor(s) is unknown; however, studies in mammals have provided several candidates. Cloning of the oocyte-specific factor, GDF-9, in the mouse [22] and in the human [2] indicated that the oocyte produces growth factor-like molecules. Female GDF-9 knockout mice were infertile, and all follicles in the ovary were arrested at the primordial and primary follicular stage [8]. In addition to GDF-9, mouse and rat oocytes also express and synthesize tumor necrosis factor alpha [24, 25]. Epidermal growth factor was found in human oocytes at different stages of follicular development [26]. It is of interest to examine whether the avian oocyte also produces these growth factors and how these growth factors regulate follicular functions in the chicken. Our results, in conjunction with previous studies, provide convincing evidence that the GDR actively participates in controlling the fate of chicken preovulatory follicles. Without the GDR, the follicle becomes atretic and will not ovulate. The presence of apoptosis in follicles induced by destruction of the GDR suggests that the GDR regulates follicular growth by producing a survival factor(s). Research is in progress to characterize the GDR factor(s). ACKNOWLEDGMENT The authors are grateful to Dr. Rex Hess (University of Illinois at Urbana-Champaign, IL) for his microscopic equipment and generous support of image processing. REFERENCES 1. El-Fouly MA, Cook B, Nekola M, Nalbandov AV. Role of the ovum in follicular luteinization. Endocrinology 1970; 87: Nekola MV, Nalbandov AV. Morphological changes of rat follicular cells as influenced by oocytes. Biol Reprod 1971; 4: Vanderhyden BC, Caron PJ, Buccione R, Eppig JJ. Developmental pattern of the secretion of cumulus expansion-enhancing factor by mouse oocytes and the role of oocytes in promoting granulosa cell differentiation. Dev Biol 1990; 140: Vanderhyden BC, Cohen JN, Morley P. Mouse oocytes regulate granulosa cell steroidogenesis. Endocrinology 199; 1: Vanderhyden BC, Telfer EE, Eppig JJ. Mouse oocytes promote proliferation of granulosa cells from preantral and antral follicles in vitro. Biol Reprod 1992; 46: Canipari R, Epifano O, Siracusa G, Salustri A. Mouse oocytes inhibit plasminogen activator production by ovarian cumulus and granulosa cells. Dev Biol 1995; 167:71 78.

6 GERMINAL DISC REGION AND FOLLICULAR APOPTOSIS IN CHICKENS Epigg JJ, Wigglesworth K, Pendola F, Hirao Y. Murine oocytes suppress expression of luteinizing hormone receptor messenger ribonucleic acid by granulosa cells. Biol Reprod 1997; 56: Dong J, Albertini DF, Nishimori K, Kumar TR, Lu N, Matzuk MM. Growth and differentiation factor 9 is required during early ovarian folliculogenesis. Nature 1996; 8: Yoshimura Y, Okamoto T, Tamura T. Electron microscope observation on LH-induced oocyte maturation in Japanese quail (Coturnix japonica). J Reprod Fertil 199; 98: Perry MM, Gilbert AB, Evans AJ. The structure of the germinal disc region of the hen s ovarian follicle during the rapid growth phase. J Anat 1978; 127: Marrone BL, Jamaluddin M, Hertelendy F. Regional pattern of cell maturation and progesterone biosynthesis in the avian granulosa cell layer. Biol Reprod 1990; 42: Tischkau SA, Bahr JM. Avian germinal disc region secretes factors that stimulate proliferation and inhibit progesterone production by granulosa cells. Biol Reprod 1996; 54: Yoshimura Y, Tischkau SA, Bahr JM. Destruction of the GDR of an immature preovulatory follicle suppresses follicular maturation and ovulation. Biol Reprod 1994; 51: Tilly JL, Kowalski KI, Johnson AL, Hsueh AJW. Involvement of apoptosis in ovarian follicular atresia and postovulatory regression. Endocrinology 1991; 129: Gilbert AB, Evans AJ, Perry MM, Davidson MH. A method for separating the granulosa cells, the basal lamina and the theca of the preovulatory ovarian follicle of the domestic fowl (Gallus domesticus). J Reprod Fertil 1977; 50: Tilly JL, Hsueh AJW. Microscale autoradiographic method for the qualitative and quantitative analysis of apoptotic DNA fragmentation. J Cell Physiol 199; 154: Gilbert AB, Perry MM, Waddington D, Hardie MA. Role of atresia in establishing the follicular hierarchy in the ovary of the domestic hen (Gallus domesticus). J Reprod Fertil 198; 69: Yoshimura Y, Tamura T. Histological and histochemical observations on the atretic follicles induced by adenohypophysectomy in the hen. Jpn Poult Sci 1985; 22: Etches RJ, MacGregor HC, Morris TF, Williams JB. Follicular growth and maturation in the domestic hens. J Reprod Fertil 198; 67: Nagle WA, Soloff BL, Moss AJM Jr, Henle KJ. Cultured Chinese hamster cells undergo apoptosis after exposure to cold but nonfreezing temperatures. Cryobiology 1990; 27: Hsueh AJW, Billig H, Tsafriri A. Ovarian follicle atresia: a hormonal controlled apoptotic process. Endocr Rev 1994; 15: McPherron AC, Lee SJ. GDF and GDF 9: two new members of the TGF-beta superfamily containing a novel pattern of cysteines. J Biol Chem 199; 68: McGrath SA, Esquela AF, Lee SJ. Oocyte-specific expression of growth/differentiation factor 9. Mol Endocrinol 1995; 9: Chen HL, Marcinkiewicz JL, Sancho-tello M, Hunt JS, Terranova PF. Tumor necrosis factor-alpha gene expression in mouse oocytes and follicular cells. Biol Reprod 199; 48: Marcinkiewicz JL, Krishna A, Cheung CMY, Terranova PF. Oocytic tumor necrosis factor alpha: localization in the neonatal ovary and throughout follicular development in the adult rat. Biol Reprod 1994; 50: Maruo T, Ladines-Llave CA, Samoto T, Matsuo H, Manalo AS, Ito H, Mochizuki M. Expression of epidermal growth factor and its receptor in the human ovary during follicular growth and regression. Endocrinology 199; 12: