| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Departments of Physiology,
b Microbiology and Immunology,
c Cell Biology and Anatomy, and
d Arizona Cancer Center, The University of Arizona, Tucson, Arizona 85724
ABSTRACT
Mammalian females are born with a finite number of ovarian oocytes, the vast majority of which ultimately undergo degeneration by atresia. The overall process of ovarian follicular atresia has been morphologically well described only in large antral follicles. Additionally, little attention has been focused on ultrastructural changes in the oocyte. Furthermore, most such morphological studies were performed prior to identification of apoptosis as a mechanism of physiological cell death. Therefore, the purpose of this study was to use electron microscopy to compare the process of atretic oocyte degradation in ovarian follicles of female Fischer 344 rats (38 days old) with ultrastructural characteristics of apoptosis. Examination of ovarian follicles revealed that nucleolar segregation, cytoplasmic or nuclear condensation, apoptotic body formation, and chromatin margination along the nuclear membrane are never observed in atretic oocytes during the degenerative process. Instead, early morphological changes in atretic oocytes include retraction of granulosa cell- and oocyte-derived microvilli and condensation of mitochondria and loss of cristae. These occurrences coincide with initiation of granulosa cell apoptosis. After most granulosa cells are lost, more severe changes occur, including segmentation of the oocyte and cytoplasmic vacuolization as atresia progresses. Thus, these results suggest that, during atresia, oocytes are removed by physiological oocyte cell death, a method that does not involve classically described apoptosis.
apoptosis, follicle, granulosa cells, ovary, ovum
INTRODUCTION
Females produce large numbers of oogonia during fetal ovarian development by mitotic division. Around the time of birth in most mammals, oogonia stop proliferating and become oocytes arrested in the diplotene stage of the first meiotic division [1, 2]. At this point, oocytes become surrounded by a layer of granulosa-like cells to form primordial follicles in which each individual follicle remains quiescent until a signal for its activation is received. Once an animal is reproductively mature, a few activated follicles are triggered to undergo further development toward ovulation during each reproductive cycle. Whereas, only a few follicles are selected to ovulate during each cycle, the vast majority degenerate and die [2]. Follicle degradation and loss, termed atresia, can occur at any stage of follicular development, from the smallest (primordial) to the most mature (antral). The process of atresia occurs continuously after birth until the follicle pool is completely depleted and reproductive senescence occurs [2–5]. In primates, this results in menopause.
It has been widely accepted that the underlying mechanism of ovarian atresia is apoptosis [6, 7]. The evidence for this has largely been acquired through the assessment of granulosa cells in large antral follicles. Morphological alterations corresponding to apoptosis have been described in granulosa cells of atretic rat follicles [8, 9]. Similar to most cell types [10, 11], apoptotic granulosa cells demonstrate chromatin margination, cytoplasmic vacuolization, cytoplasmic blebbing, both cytoplasmic and nuclear condensation and fragmentation, and nucleolar segregation. Biochemical assessments have supported the assignment of apoptosis in granulosa cells by demonstration of internucleosomal DNA fragmentation [8, 9, 12, 13] and TUNEL staining [14–16], although not all studies have supported these findings [17, 18]. Therefore, past studies have provided evidence that the death of granulosa cells during atresia occurs via apoptosis.
Whether the oocyte also undergoes degeneration by apoptosis during atresia has been a more controversial matter. For example, in a study using ovulated murine and human oocytes incubated in vitro, lack of DNA fragmentation and annexin-V membrane binding suggested that there was insufficient evidence to conclude that the mechanism of cell death was apoptosis [19]. However, this conclusion was challenged by another report in which ovulated murine oocytes incubated in vitro displayed increased caspase-3 activity, a widely accepted characteristic of apoptosis [20]. Neither of these studies involved morphological evaluation at the ultrastructural level, which is considered the definitive method for identifying apoptosis, as originally described by Wyllie et al. [10]. Additionally, these previous studies have characterized oocyte death in ovulated oocytes that are maintained in vitro. Thus, whether this is the process by which oocyte degeneration occurs during atresia in vivo remains unknown. Therefore, determination of the mechanism by which oocyte death occurs has still not been resolved.
Few ultrastructural observations of atretic oocytes have been described, and few of these reports have made comparisons with apoptosis. Vazquez-Nin and Sotello [21] described an altered distribution and appearance of organelles, loss of microvilli, and segmentation of the oocyte in degenerating oocytes of atretic preantral follicles from immature rats. Gondos [22] reported loss of microvilli, distorted cytoplasmic organelles, and altered appearance of the zona pellucida in grossly atretic medium-sized preantral follicles of immature rats. Interestingly, in mitotic development of oogonia, De Pol et al. [23, 24] reported condensation of chromatin and apoptotic body formation during human embryogenesis and shortly after parturition and described the process as similar to apoptosis. However, in describing atretic oogonia, Franchi and Mandl [3] suggested that there are differences in atresia of gametes between prenatal and postnatal ovaries. Therefore, studies describing ovarian follicular atresia have not compared ultrastructural features of atretic oocytes to those that are characteristics of apoptosis. Thus, the purpose of this report was to analyze critically in vivo oocyte degeneration in rats for ultrastructural features of either apoptosis or other modes of cell death during atresia at early stages of follicle development.
MATERIALS AND METHODS
Animals
Female Fisher 344 rats (21 days old) were purchased from Harlan (Indianapolis, IN). Animals were housed in plastic cages, given food and water ad libidum, and maintained on a 12L:12D cycle as previously reported [25–27]. Animals were allowed to acclimate for at least 1 wk before sacrifice. Experiments were approved by the University's Institutional Animal Care and Use Committee.
Tissue Collection and Processing for Electron Microscopy
Six rats aged 38 days were sacrificed by CO2 asphyxiation. The peritoneum was opened and ovaries were dissected out. Oviductal tissue and fat were removed and ovaries were fixed in one-half strength Karnovsky fixative for 24 h in 0.1 M cacodylate buffer (pH 7.2). After rinsing in the same buffer, tissues were postfixed in 2% osmium tetroxide (OsO4) in 0.1 M cacodylate buffer (pH 7.2) for 1 h. Tissues were then dehydrated through a graded series of ethanol and embedded in Spurr low viscosity epoxy resin as previously described [28]. One-micron sections were stained with toluidine blue, selected blocks were thin-sectioned, and grids were counterstained with uranyl acetate and lead citrate. Two to three blocks per ovary were thin-sectioned, and all follicles in each section were examined under the electron microscope. Electron micrographs were taken at 80 kV on a Philips CM12S electron microscope. Data represent ovaries from all six rats.
RESULTS
Rats used in this study were still immature at the time of sacrifice, though this time point is close to the time at which this strain becomes mature. Vaginas were still closed. Follicles of all stages were observed in collected ovaries, but no corpora lutea were present.
Oocytes in Primordial Follicles
Primordial follicles are composed of an oocyte surrounded by a small number of squamous granulosa cells (Fig. 1). Oocytes contained in healthy primordial follicles display lightly staining mitochondria with shelflike cristae, well-developed Golgi complexes, and are tightly apposed to the surrounding granulosa cells (Fig. 1, A and B). The nucleus contains small amounts of darkly staining heterochromatin (Fig. 1A), and large, intact nucleoli with well-developed nucleolonemata are present (not shown).
|
Atretic primordial follicles were identified by large amounts of cytoplasmic vacuolization and loss of identifiable organelles, though granulosa cells remain normal in appearance (Fig. 1C). Few of the primordial follicles observed were atretic (5 of 44 follicles examined); of those, all looked similar. Briefly, vacuoles and secondary lysosomes containing debris were the predominent structures, while cytoplasmic organelles became difficult to identify (Fig. 1D). Additionally, nuclei in atretic primordial follicles were never observed among the altered structures present.
Oocytes in Healthy Developing Follicles
Ultrastructural morphology of healthy oocytes in developing follicles appears similar regardless of the stage of development. Oocytes in follicles that have developed to or beyond the primary stage have acquired a zona pellucida. This is a layer completely composed of glycoproteins that encompasses and is tightly apposed to both the oocyte and the surrounding granulosa cell layer (Fig. 2, A and B) [2]. Microvilli from the oocyte and thicker prolongations from granulosa cells penetrate the zona pellucida (Fig. 2B). The oocyte cytoplasm is sparsely populated with loosely clustered organelles. The majority of organelles are normally located at the cortical and perinuclear regions of the oocyte (Fig. 2A). Mitochondria in oocytes appear round or oval, rather dark, and have small numbers of identifiable shelflike cristae (Fig. 2C). Very small amounts of rough endoplasmic reticulum (RER) are located adjacent to clusters of mitochondria and are uniformly filled with faintly electron-dense material (Fig. 2C). There are also small groupings of short, roughly parallel linear structures dispersed throughout the cytoplasm (Fig. 2D). These can be distinguished from ER, because they do not form fully enclosed structures. These have previously been called lamellae [21] or cytoplasmic rays [29], and are believed to be membranous components. In some oocytes, there are areas with localized concentrations of ribosomes, RER, and mitochondria (Fig. 2C) that may indicate highly active sites of protein synthesis. In contrast to primordial follicles (Fig. 1B), complex synthetic machinery is not observed in developing follicles (Fig. 2C). Vacuoles or lysosomes are rarely observed in oocytes of healthy follicles. The nucleus of developing follicles was never observed to display condensed chromatin outside of the nucleolus and had little electron-dense material (Fig. 2D). In follicles of primary to small growing preantral stages, oocytes contain nucleoli with well-developed nucleolonemata, suggesting they are actively synthesizing RNA. Oocytes in large growing preantral and antral follicles contain mostly round nucleoli with homogeneous substructures (not shown).
|
Granulosa Cells in Atretic Follicles
Atretic follicles were identified by pyknotic, very darkly staining and shrunken granulosa cells and, secondarily, by alteration of the shape of the oocyte from a normal uniformly staining and round appearance (Fig. 2A). Also, healthy follicles always maintained an intact layer of granulosa cells surrounding and in contact with the surface of the oocyte. Table 1 summarizes the comparison of ultrastructural characteristics of oocytes during follicular atresia with apoptosis. The majority (approximately 60% of 65 follicles examined) of growing and antral follicles appear unhealthy, as judged by oocyte and granulosa cell ultrastructure. The earliest signs of atresia in primary and larger follicles include well-established features of apoptosis in granulosa cells and loss of focal contacts between the oocyte and surrounding granulosa cells (Fig. 3). As atresia progresses, there is substantial loss of granulosa cells before the oocyte is grossly affected (Fig. 4A). Apoptosis in the granulosa cell layer, as evidenced by dark, pyknotic bodies, appears to be randomly distributed throughout an atretic follicle (Figs. 3A and 4A). The granulosa cell layers remain in contact with the membrana propria throughout the process of atresia but pull away from the oocyte, eventually leaving it apparently unattached within the follicular space (compare Figs. 3A, 4A, and 5A). Microvilli from granulosa cells become retracted or lost from the zona pellucida (Fig. 3B). This occurs concurrently with the first appearance of apoptotic granulosa cells (Fig. 3, A and B). At the same time, microvilli from the oocyte have also become fewer in number, with most terminating just inside the zona pellucida surface (Fig. 3B). At this stage, the oocyte appears morphologically normal (Fig. 3A), and no changes in cytoplasmic organelles can be observed (Fig. 3C).
|
|
|
Oocytes in Atretic Developing Follicles
As atresia progresses, ultrastructural alterations in oocytes are more readily apparent (Fig. 4). Organelles become more randomly distributed throughout the cytoplasm, and microvilli are almost completely lost (Fig. 4B). The RER either becomes lost or unidentifiable. Additionally, mitochondria stain more darkly and begin to lose their cristae (Fig. 4C). In a few degenerating oocytes, lamellae become much more abundant than in healthy-appearing oocytes (Fig. 5E). This was particularly observed in the smallest follicle types (not shown). Throughout the course of these changes, the nucleus of the oocyte remains normal, with no condensed chromatin (Fig. 5D). There was a general lack of nucleolonemata observed in nucleoli (not shown). Segregation of the nucleolus into components of variable density, a characteristic of apoptosis, could not be observed in oocytes of atretic follicles at any stage.
|
In follicles in advanced stages of atresia, most granulosa cells have been removed and the degenerating oocyte becomes segmented (Fig. 5). The oocyte divides into multiple round bodies remaining in contact with one another (Fig. 5). This was observed in 40% of 65 growing and antral follicles examined. The zona pellucida remains intact throughout the entire process, continuing to enclose the oocyte segments (Fig. 5A). Oocyte segmentation can be observed at both the electron and light microscope level (not shown). These segments are not similar to apoptotic bodies, because a budding from the surface is not seen and there is no increase in electron density (Fig. 5C). The number and size of vacuoles and secondary lysosomes increases; however, there is no apparent cytoplasmic condensation (Fig. 5C). The size of the space encompassed by the zona pellucida remained the same as in healthy follicles (60–80 µm). This suggests that there is no cellular shrinkage until the oocyte becomes lobed and occupies less space inside the zona pellucida. As the oocyte degenerates, multiple segments of the oocyte can sometimes be seen to contain parts of the nucleus (not shown); however, no observable condensation or clumping of DNA is observed in oocytes at any stage of degeneration (Fig. 5D). Oocyte segments usually remain in close proximity to one another (Fig. 5A), until very advanced stages of degeneration (not shown). Interestingly, microvilli are occasionally present on the surface and between some segments (Fig. 5F). Phagocytic cells of the immune system were not seen in association with oocytes at any stage of oocyte degeneration.
DISCUSSION
Comparison of Oocyte Cell Death with Apoptosis
Since Wyllie et al. [10] first described the morphological characteristics of physiological cell death, apoptosis, few studies have described the ultrastructure of the atretic oocyte and none have examined this in pubertal or adult animals. Most reviews of ovarian follicular atresia focus on changes in granulosa cells or equate the entire process with apoptosis [5–7]. Biochemical analyses of atretic follicles that have measured DNA integrity [8, 9] or increases in cell death-related mRNA levels, including bax and fas/fasL [4, 30], have confirmed that apoptosis is occurring in antral ovarian follicles. Because the oocyte is a very small component of these large follicles, such measurements most probably reflect the status of granulosa cells. Due to these restrictions, microscopic examination is required to study the process of atresia in oocytes in situ. Because ultrastructural characterization remains the most reliable method for the classification of cell death as apoptosis [11], the present study undertook a description of the sequence of ultrastructural changes that accompany atresia in rat ovarian follicles.
The observations reported here do not resemble classical descriptions of either necrosis or apoptosis. Cell swelling and organelle and membrane dissolution do not occur, demonstrating that necrosis is not occurring. Alterations in oocytes contained in rat atretic primary through antral follicles included: loss of both granulosa cell and oocyte microvilli from the zona pellucida, changes in cytoplasmic organelles, increased numbers of vacuoles and secondary lysosomes, and finally segmentation of the oocyte. Loss of microvilli and cytoplasmic vacuolization do resemble apoptosis, but other events differ from those associated with traditional apoptosis. For example, mitochondria do not maintain their characteristic appearance during early stages of atresia, as normally occurs in apoptosis. Also, nuclear and cytoplasmic condensation, which is reflected by an increase in electron density at the ultrastructural level, is not observed in degenerating oocytes. Segmentation of the oocyte appears to occur by a process distinct from apoptotic body formation. Traditional apoptotic bodies do not have microvilli present on their surfaces, such as those observed in oocyte fragments. Most significantly, condensed chromatin is never observed in oocytes of atretic follicles at any stage. These comparisons suggest that there are more differences than similarities between physiological oocyte cell death and apoptosis.
Other reports attempting to identify the mechanism of oocyte death have not discussed the possibility of alternative, nonapoptotic, types of physiological death [20]. Early ultrastructural studies occurred before apoptosis was characterized [3, 21]. More recent studies using ovulated oocytes failed to prove definitively that apoptosis was the mechanism of oocyte death [19, 20]. Moreover, because the observations made in those recent studies were in ovulated murine oocytes incubated in vitro, their conclusions and the ones reported here using follicles in whole ovaries might truly reflect different mechanisms of cell death.
Variations in Physiological Oocyte Cell Death
Follicles appear to undergo different processes of degradation depending on their stage of growth or development. In atretic primordial follicles observed here, few cytoplasmic organelles were identifiable, vacuoles and secondary lysosomes had accumulated in large numbers, and granulosa cells were unaffected. This appeared to be different from alterations in oocytes observed in developing follicles. Previous reports of fetal and neonatal oogonia and oocyte loss suggest that degradative processes also vary as oogonia become oocytes incorporated into follicles [3, 23]. Thus, differences in ultrastructural characteristics of physiological oocyte cell death suggest that the biochemical machinery involved also varies. In support of this, Boone and Tsang [31] have reported that the distribution of DNAse I expression, involved in oligonucleosomal cleavage of DNA, was variable among granulosa cells and oocytes in different follicle types of adult rats. Therefore, it seems likely that oocytes in postnatal rats have unique cell-death triggers, signal transduction pathways, and clearance mechanisms as compared with other cell types. Such flexibility has not been described for traditional apoptosis.
Unique Aspects of Oocytes
The unique nature of the oocyte relative to other cell types may be the cause for its unusual manner of cell death. Oocytes can remain arrested in meiosis for years, are surrounded by an acellular zona pellucida, are nonproliferating, and are known to rely on surrounding granulosa cells for survival [2]. Apoptosis is an active process thought to protect the rest of an organism from an aberrant cell. Meiotic oocytes may not be required to undergo apoptosis, because they pose no threat of excessive proliferation and tumor formation.
Conclusions
Overall, the results presented here support that oocyte loss in atretic follicles of postnatal rats can be morphologically distinguished from the two more widely described mechanisms of cell death, necrosis, and apoptosis. While it is generally accepted that granulosa cells are lost by apoptosis, the ability of the oocyte to undergo apoptosis is still in question. Based on ultrastructural criteria traditionally associated with apoptosis [10], oocyte death should be assigned to a different class of physiological cell death. Such variations in the mechanisms of cell death are becoming more widely accepted [4] and will be the subject of future investigations.
ACKNOWLEDGMENTS
We thank Patty Christian for her technical assistance in preparing the tissues.
FOOTNOTES
First decision: 26 April 2000.
1 This research was funded by the Arizona Disease Control Research Council, NIH grants ES98979 and Toxicology Center grant ES06694. 
2 Correspondence: Patrick J. Devine, 1501 North Campbell Ave., Department of Physiology, The University of Arizona, P.O. Box 245051, Tucson, AZ 85724-5051. FAX: 520 626 2382; pdevine{at}u.arizona.edu
Accepted: June 1, 2000.
Received: March 22, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  P. Reddy, D. Adhikari, W. Zheng, S. Liang, T. Hamalainen, V. Tohonen, W. Ogawa, T. Noda, S. Volarevic, I. Huhtaniemi, et al. PDK1 signaling in oocytes controls reproductive aging and lifespan by manipulating the survival of primordial follicles Hum. Mol. Genet., August 1, 2009; 18(15): 2813 - 2824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Tingen, S. K. Bristol-Gould, S. E. Kiesewetter, J. T. Wellington, L. Shea, and T. K. Woodruff Prepubertal Primordial Follicle Loss in Mice Is Not Due to Classical Apoptotic Pathways Biol Reprod, July 1, 2009; 81(1): 16 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. C. Zachos, M. G. Burch, R. B. Billiar, C. Li, E. D. Albrecht, and G. J. Pepe Regulation of Expression of Microvillus Membrane Proteins by Estrogen in Baboon Fetal Ovarian Oocytes Biol Reprod, December 1, 2008; 79(6): 1160 - 1168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A M Lobascio, F G Klinger, M L Scaldaferri, D Farini, and M De Felici Analysis of programmed cell death in mouse fetal oocytes Reproduction, August 1, 2007; 134(2): 241 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Nicholas, R. Alberio, A.A. Fouladi-Nashta, and R. Webb Relationship Between Low-Molecular-Weight Insulin-Like Growth Factor-Binding Proteins, Caspase-3 Activity, and Oocyte Quality Biol Reprod, April 1, 2005; 72(4): 796 - 804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Hussein Apoptosis in the ovary: molecular mechanisms Hum. Reprod. Update, March 1, 2005; 11(2): 162 - 178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Yang, J. Wang, Y. Shen, and S. K. Roy Developmental Expression of Estrogen Receptor (ER) {alpha} and ER{beta} in the Hamster Ovary: Regulation by Follicle-Stimulating Hormone Endocrinology, December 1, 2004; 145(12): 5757 - 5766. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. C. Zachos, R. B. Billiar, E. D. Albrecht, and G. J. Pepe Regulation of Oocyte Microvilli Development in the Baboon Fetal Ovary by Estrogen Endocrinology, February 1, 2004; 145(2): 959 - 966. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.T. Bridgham and A.L. Johnson Expression and Regulation of Fas Antigen and Tumor Necrosis Factor Receptor Type I in Hen Granulosa Cells Biol Reprod, September 1, 2001; 65(3): 733 - 739. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
