| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ovary |
Centre for Reproductive Health,3 Department of Obstetrics and Gynaecology, The University of Adelaide, Adelaide, South Australia 5005, Australia
School of Anatomy and Human Biology,4 The University of Western Australia, Crawley, Western Australia 6009, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
apoptosis, atresia, bovine, follicle, ovary, theca cells, theca interna
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Generally, there are three pathways through which cells can die; apoptosis, necrosis, or terminal differentiation [1], the last of which is exemplified in normoblasts differentiating into erythrocytes, epidermal cells into keratinocytes, or megakaryocytes into platelets. Apoptosis is an active process within the cell involving DNA cleavage into oligonucleasomal length fragments by an endogenous endonuclease. The chromatin condenses, followed by budding of the nucleus and finally the cell buds to create membrane-bound apoptotic bodies that eventually become phagocytosed by macrophages or, in some epithelia by adjacent cells. In contrast, necrosis occurs as a result of cell injury or trauma and is characterized by rupture of the cell, leading to swelling of the mitochondria and other organelles [1]. The nuclear DNA is then fragmented randomly as it becomes exposed to the external environment.
In the study of atresia, much attention has been paid to how follicular cells die. To date, it has been demonstrated in various species (rat [2–4], horse [5], bovine [6, 7], pig [8] and human [9]) that apoptosis is the underlying process of atresia and that it occurs throughout all stages of follicular development. However, in bovine antral follicles, Van Wezel et al. [7] showed that, whereas granulosa cells in the middle of the membrana granulosa undergo apoptosis (TUNEL positive, classic condensation pattern of chromatin, and phagocytosis by adjacent granulosa cells), those granulosa cells closest to the antrum did not display any of the normal characteristics associated with apoptosis [7]. Hence, whereas these cells had classic pyknotic nuclei, they were essentially nondetectable by TUNEL. Their DNA was randomly nicked as shown by the COMET assay and gel electrophoresis of the DNA aspirated from the follicular antrum, and yet these cells showed no evidence of necrosis by electron microscopic examination. This led the authors to conclude that the cell death observed had more in common with a terminal differentiation mechanism. This observation may only be relevant to bovine follicles, especially in light of the recent description of different types of follicular atresia in the bovine [10].
In bovine antral follicles, atresia can follow one of two processes, and the description of these resolves earlier conflicting descriptions of atresia in this species [10]. Antral atresia, as it is now called, occurs in antral follicles of all sizes and features the initial death of the antrally situated granulosa cells. These cells develop visible pyknotic nuclei. This is the classic form of atresia as observed in other species. The basal granulosa cells stay in close contact with the follicular basal lamina and remain healthy until atresia is well advanced. In contrast, follicles with cell death initially present within the basal layers of the membrana granulosa were described as undergoing basal atresia. Basal atresia is only observed in bovine follicles less than 5 mm in diameter and involves the initial death of the granulosa cells lining the follicular basal lamina. These basal cells undergo classic apoptosis with nuclear and cellular budding and phagocytosis by invading macrophages. Surprisingly, the antrally situated cells of basal atretic follicles undergo hypertrophy [10] and differentiate into progesterone-producing cells [11] before succumbing to death as atresia progresses. These two different types of atresia have been considered to arise by the different location of the younger cells either basally or antrally brought about by different rates of follicular antrum expansion during follicular growth [12].
A very important feature of atresia is the susceptibility of the oocyte and granulosa cells to death, which changes during folliculogenesis. In preantral follicles, but not antral follicles, cell death is often initially observed within the oocyte [13, 14]. Conversely, in antral follicles, it is the granulosa cells that die initially. The cumulus-oocyte complex often remains healthy and intact until the late stages of atresia [13, 14]. These observations probably merely reflect the differential rates of growth during folliculogenesis and hence vulnerability to death. Oocyte growth occurs mainly preantrally [15, 16], and the largest expansion of granulosa cells numbers occurs early in antrum formation [15].
The question logically arises of whether there is a stage at which the theca interna is the compartment most susceptible to cell death. Little research has been carried out focusing specifically on the theca interna. Cell death has been shown to occur in the theca interna of several species, including rat [4], pig [17], avian [8], and bovine [10], but is reported to occur at a lower rate than in the membrana granulosa [18, 19]. In an in-depth study of the ultrastructural changes of the theca interna during atresia in the sheep ovary, O'Shea et al. [20] observed cell death in the theca interna at all stages of atresia, but at a much lower frequency than in the membrana granulosa. More recently, Nakayama et al. [19], studying bovine follicles, reported that apoptotic cells were initially detected in the theca interna of bovine follicles in the early stages of atresia when apoptotic cells were visible in the middle layers of the membrana granulosa but concluded that apoptosis within the membrana granulosa is the initial symptom of atresia. Isobe and Yoshimura [21] reported high frequencies of TUNEL-positive cells within the theca interna, which were similar in both early and late atretic bovine follicles.
None of the previous studies on bovine follicles distinguished the two types of atresia. Current evidence would suggest that, at least in bovine basal atretic follicles, the theca interna undergoes significant alterations during atresia. The theca interna of basal atretic follicles is much less cellular and the cells are randomly situated rather than orientated circumferentially, as in healthy or antral atretic follicles [10]. There is also an increased amount of collagen within the theca interna of basal atretic follicles and cellular debris within the capillaries [10]. In addition, the follicular fluid of basal atretic follicles has significantly less of the thecal-derived steroids, testosterone, and androstenedione compared with healthy or antral atretic follicles [11]. Basal atretic follicles also have reduced levels of insulin-like factor 3 (INSL3; relaxin-like factor) expression in theca interna cells [22]. Mice null for INSL3 have increased levels of atresia [23].
Thus, in general, there is a paucity of literature on thecal cell death in atresia. Specifically in the bovine, theca cell death has not been investigated in the two types of follicular atresia. In addition, our studies of bovine atresia found reduced levels of thecal-derived hormones in basal atretic but not antral atretic follicles, suggesting that the theca of the two types of atretic follicles behaves differently. Therefore, our goal was to examine death of the different thecal cell types in both basal and antral atretic bovine follicles.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bovine ovaries (n = 29) were collected from a local abattoir from cows assessed visually as nonpregnant within 20 min postslaughter. Ovaries were sliced longitudinally into two or three slices approximately 5 to 8 mm thick, producing 66 slices. These were immediately immersed in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.2) and transported to the laboratory on ice. Tissue was transferred into fresh 4% paraformaldehyde and fixed overnight at 4°C before rinsing in 70% ethanol and embedding in paraffin by standard histological methods.
Classification of Follicular Health
One 5-µm-thick section from each paraffin-embedded ovarian slice was cut and stained with hematoxylin and eosin. Sections were viewed on an Olympus BX50 microscope (Olympus Australia Pty. Ltd.) and each antral follicle identified. Analysis was restricted to the examination of antral follicles 
5 mm in order to provide sized-matched controls for basal atretic follicles, which only occur in this size range. The cross-sectional diameter from the follicular basal lamina of follicles (1–5 mm) was measured using an occular micrometer, and the follicles were classified into one of three categories: healthy, antral atretic, or basal atretic, as previously described [10]. The condition of the membrana granulosa was used to ascertain follicular health. Cell death was identified as intensely stained round or crescent-shaped pyknotic nuclei [7] or as apoptotic nuclei. Follicles at the very late stages of atresia, which had no remaining membrana granulosa, were excluded because the type of atresia could not be determined.
Immunohistochemistry
Sections (5 µm thick) were cut from each portion of paraffin-embedded ovary and deparaffinized in xylene. Sections were rehydrated in decreasing concentrations of ethanol (100%, 95%, 70%, 50%) and immersed in H2O before treatment with blocking solution (10% normal donkey solution [Sigma Chemical Co., St. Louis, MO]) in antibody diluent containing 0.55 M sodium chloride and 10 mM sodium phosphate (pH 7.2), for 30 min at room temperature. Rabbit anti-bovine cholesterol side-chain cleavage cytochrome P450 (CPY11A; P450SCC) antiserum (OXYgene, Dallas, TX) was used to identify steroidogenic cells. Incubation with primary antibody (1:200) was conducted for 48 h at room temperature before incubating with biotin-donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories Inc., West Grove, PA) for 2 h. This was followed by incubation with a 1:100 dilution of Cy3-conjugated streptavidin (Jackson ImmunoResearch Laboratories Inc.) for 1 h at room temperature. Sections were washed 3 x 5 min in phosphate-buffered saline (PBS) containing 0.274 M sodium chloride, 5.4 mM potassium chloride, and 10 mM sodium phosphate between incubations, and mounted in fluorescence mounting medium (DAKO, Carpinteria, CA). Lectin from Bandeiraea simplicifolia (Sigma-Aldrich) was used as a marker for endothelial cells within the theca interna [24, 25]. Sections were incubated with 1:200 dilution of FITC-labeled lectin in antibody diluent for 1 h in a humidified chamber at room temperature.
TUNEL Labeling
A terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) method was used to identify degraded DNA. Following rehydration, sections (5 µm thick) were immersed in phosphate-buffered solution for TUNEL (PBST; 10 mM sodium/potassium phosphate in 0.317 M sodium chloride and 5 mM potassium chloride solution) for 10 min. Sections were treated with 5 µg/ml proteinase K (Boehringer Mannheim, Mannheim, Germany) in proteinase K buffer (50 mM Tris-hydrogen chloride, 5 mM EDTA) for 45 min at 37°C in a humidified chamber. Following washing in PBST (4 x 5 min), sections were treated with TUNEL reagents, 0.5 nM biotin-16-2'-deoxy-uridine-5'-triphosphate (biotin-dUTP; Roche Diagnostics GmbH, Mannheim, Germany), 50 U/ml terminal transferase (Roche Diagnostics GmbH) and 1.5 mM cobalt chloride in reaction buffer (30 mM Tris-chloride, pH 7.2, and 140 mM sodium cacodylate) for 1 h at 37°C. Sections were washed in PBST (3 x 5 min) before incubation with a 1:100 dilution of Cy3-conjugated streptavidin (Jackson ImmunoResearch Laboratories Inc.) for 1 h at room temperature in a humidified chamber.
Following immunohistochemistry or TUNEL labeling, sections were counterstained by incubation with the nuclear stain 4',6-diamidino-2-phenylindole dihydrochloride (DAPI) solution (Molecular Probes, Eugene, OR) for 30 min in a humidified chamber at room temperature.
Multiple Labeling
TUNEL labeling, lectin binding, and DAPI staining were combined to identify cell death in the vasculature of the theca interna. Immunohistochemical colocalization of P450SCC was combined with TUNEL labeling and DAPI staining in order to identify degraded DNA in steroidogenic cells. In this experiment, biotin-dUTP was replaced with a direct-labeled fluorescein-12-deoxy-uridine-5'-triphosphate (fluorescein-dUTP) (Molecular Probes). Terminal transferase was omitted in negative-control sections. When immunohistochemistry was combined with TUNEL labeling, sections were treated with proteinase K prior to incubating with primary antibody. The TUNEL reaction was conducted after completion of immunohistochemistry.
Image Analyses
Follicles for examination were chosen from those previously classified on hematoxylin and eosin-stained sections. On serial sections processed for immunohistochemistry, TUNEL labeling, or multiple labeling, selected follicles were identified. To ensure unbiased sampling of each follicle, the first field of view containing the membrana granulosa and the theca interna of stained sections was selected randomly and photographed using an ultraviolet filter and filters suitable for viewing FITC and Cy3 on an Olympus BX50 microscope with SPOT camera attachments (Diagnostic Instruments Inc., Sterling Heights, MI) and 40x and 60x magnification. Fields of view directly south, east, and west of the initial view were also photographed. Images were merged to identify colocalization of staining, and densitometry was performed on selected reference areas using analySIS software (Soft Imaging System GmbH, Münster, Germany).
Statistical Analyses
One-way analysis of variance (one-way ANOVA) and post hoc Student-Newman-Keuls and Duncan multiple range tests were performed using Statistical Package for the Social Sciences 11.5 for Windows (SPSS Inc., Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Composition of the Theca Interna
From 10 sections from 9 ovaries immunohistochemically stained for P450SCC to identify steroidogenic cells, 11 healthy, 14 early–midantral atretic follicles, 7 late antral atretic follicles, and 9 basal atretic follicles were examined. Positive immunostaining for P450SCC was localized exclusively to the cytoplasm of a population of cells within the theca interna. The area of the theca interna was measured, and all cells positively staining for P450SCC and identified by colocalization of DAPI were counted per cross-sectional area of theca interna. Healthy follicles and both early–midantral atretic and late antral atretic follicles did not differ in the numerical density of steroidogenic cells in the theca interna (Fig. 2A). In comparison, the number of steroidogenic cells in the theca interna of follicles undergoing basal atresia was significantly reduced (P < 0.001; Fig. 2A).
|
In sections in which lectin binding was used to identify the vasculature, healthy follicles and follicles in all stages of antral atresia showed uniform staining of cells (Fig. 3A). The vasculature within the theca interna of healthy and antral atretic follicles was situated in close proximity to the follicular basal lamina and vessels were predominantly oriented circumferentially around the follicle. In contrast, in basal atretic follicles, the lectin binding to the vasculature was uneven in appearance and randomly dispersed throughout the theca interna and, hence, the capillaries often appeared to be orientated toward the center of the follicle (Fig. 3B). Also, in basal atretic follicles, there was some additional labeling of individual cells. These did not appear to be associated with the vasculature, which was recognizable as closely associated cells both in basal atretic follicles and in healthy and antral atretic follicles (Fig. 3B).
|
From 10 sections from 8 ovaries, 7 healthy, 13 early– midantral atretic follicles, 6 late antral atretic follicles, and 18 basal atretic follicles were examined. Densitometric measures (Fig. 2B) showed the volume fraction of lectin binding per cross-sectional area of theca interna to be the same for healthy, early–midantral atretic follicles and late antral atretic follicles (Fig. 2B). Basal atretic follicles did not differ from healthy and late antral atretic follicles but did contain a significantly higher volume fraction of lectin binding within the theca interna in comparison with early– midantral atretic follicles (P < 0.05; Fig. 2B).
Cell Death in the Theca Interna
From 10 sections from 8 ovaries, 7 healthy, 13 early– midantral atretic follicles, 6 late antral atretic follicles, and 18 basal atretic follicles were examined. Labeling was observed within all sections treated by the TUNEL protocol and absent in negative-control sections when terminal transferase was excluded (results not shown). TUNEL labeling was rarely observed within the theca interna of healthy follicles (Fig. 4A). Those undergoing antral atresia had an increased amount of cell death in the theca interna compared with healthy follicles (Fig. 5A). Antral atretic follicles at the late stage had significantly more TUNEL-positive cells than those in the early to mid stages of antral atresia (P < 0.05; Fig. 5A). Follicles undergoing basal atresia had significantly more TUNEL-positive cells within the theca interna compared with all other follicles (P < 0.05; Fig. 5A). This TUNEL-positive staining was often seen in clusters of cells in close association with one another (Fig. 4, E and F).
|
|
TUNEL-Positive Steroidogenic Cells in the Theca Interna
From 10 sections from 9 ovaries immunohistochemically stained for P450SCC and labeled with TUNEL to identify cell death of steroidogenic cells, 11 healthy, 14 early–midantral atretic follicles, 7 late antral atretic follicles, and 9 basal atretic follicles were examined. Healthy and early– midantral atretic follicles did not differ significantly in numerical density of steroidogenic cell death per cross-sectional area of theca interna (P > 0.05; Fig. 5B). Both late antral and basal atretic follicles had significantly more steroidogenic cell death than healthy follicles (P < 0.001). Basal atretic follicles had significantly more steroidogenic cell death than all other follicle classifications (P < 0.01; Fig. 5B).
TUNEL-Positive Endothelial Cells in the Theca Interna
From 10 sections from 8 ovaries, 7 healthy, 13 early– midantral atretic follicles, 6 late antral atretic follicles, and 18 basal atretic follicles were examined by dual lectin binding and TUNEL labeling to identify death of endothelial cells (Fig. 4B, D and F). The number of TUNEL-positive endothelial cells per cross-sectional area of theca interna is shown in Figure 5C. Follicles undergoing atresia had significantly more TUNEL-positive endothelial cells than healthy follicles (P < 0.05; Fig. 5C). There was significantly more endothelial cell death occurring in follicles undergoing basal atresia than antral atresia (P < 0.001; Fig. 5C).
Cell Death in the Membrana Granulosa
From 10 sections from 8 ovaries, 8 healthy, 5 early– midantral atretic follicles, 6 late antral atretic follicles, and 14 basal atretic follicles were examined. Cell death in the membrana granulosa was calculated as a percentage of granulosa cells that were TUNEL-labeled (Fig. 6). All follicles undergoing atresia had significantly more cell death in the membrana granulosa than healthy follicles (P < 0.001). Those follicles in the late stages of antral atresia had a significantly greater proportion of granulosa cells that were TUNEL-positive than those in the early–midstages and those undergoing basal atresia (P < 0.01; Fig. 6). Follicles undergoing basal atresia and early–midantral atresia did not differ in their proportion of granulosa cells that were TUNEL-positive (P > 0.05; Fig. 6).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results confirm and extend observations on the basal atretic follicles. The reduced number of P450SCC-positive theca cells could have arisen by either enhanced death of these cells as observed here and/or reduced expression of P450SCC. Either way, it would be predicted that the synthesis and secretion of steroid hormones from the theca in these follicles would be reduced. In agreement with this, reduced levels of thecal-derived testosterone and androstenedione have been observed in the follicular fluid of basal atretic follicles [11]. The basal atretic follicles also have reduced levels of INSL3 [22] and INSL3 mRNA (unpublished results). INSL3 is expressed in steroidogenic cells of the theca interna [22]. Whereas not all roles of INSL3 have been discovered, INSL3 null mice have an increased rate of follicular atresia and luteal regression [23], suggesting that INSL3 in some way maintains cells or prevents apoptosis of cells. Consistent with these data is the suggestion that maybe INSL3 maintains the theca interna and that its reduced expression in basal atresia initiates death of cells in the theca interna. Of course an equally consistent hypothesis is that death of theca interna cells leads to a loss of expression of INSL3 or of cells expressing INSL3 in basal atretic follicles. Additional experiments will be required to address this issue.
The study by O'Shea et al. [20] on the theca interna of atretic follicles in the sheep ovary suggests that vasculature changes in the theca interna play a major role in the atretic process. In the current study, lectin from Bandeirea simplicifolia was used to identify endothelial cells [24, 25]. This lectin has been shown to colocalize with an antibody recognizing von Willebrand factor in endothelial cells of bovine follicles (unpublished results). Histological observations showed basal atretic follicles contained some binding, which was uneven in appearance compared with the intense and clear binding to the endothelial cells seen within healthy or antral atretic follicles. Confocal microscopy suggested that this binding occurred within individual cells not associated with vasculature structures (unpublished results). The cause of this is currently unknown. These cells could represent endothelial cells not associated with vasculature that are most likely to have arisen before the progression into atresia or the acquisition of lectin binding domains by nonendothelial cells as observed previously [26, 27]. Irrespective of this additional set of cells in basal atretic follicles that bind lectin, the vasculature in the theca interna clearly is oriented differently in these follicles. It appears to radiate toward the antrum, whereas in healthy and antral atretic follicles, it is orientated essentially parallel to the follicular basal lamina. Given the increased endothelial cell death observed here and the increased cell debris observed previously in the capillaries of the theca of basal atretic follicles [10], it can be speculated that the blood flow through these capillaries is reduced. Whether this is a cause or effect of basal atresia is not known at this stage.
The TUNEL method of labeling degraded DNA is a valid method for identifying cell death [21, 28–30], even if it does not definitively distinguish between apoptosis and necrosis. Apoptosis is reported to be the major cause of cell death in atresia [3, 8]; however, these reports are limited to studies of the membrana granulosa, and such studies do not account for other possible causes of cell death [7]. TUNEL labeling in this study showed a greater proportion of cell death within the theca interna of follicles undergoing basal atresia compared with healthy and antral atretic follicles. In addition, there was significantly more endothelial and steroidogenic cell death in basal atretic follicles compared with healthy follicles and even in the later stages of antral atresia. However, the amount of cell death observed in the membrana granulosa of basal atretic follicles was equal to that of follicles in the early–midstages and less that that of the late stage of antral atresia. Taken together, these results suggest that cell death is more prominent in the membrana granulosa in antral atretic follicles and more so in the theca interna in the basal atretic follicles.
We conclude that the theca interna can be very susceptible to cell death and may even be a site of initiation of atresia, at least in basal atretic follicles. This is a major finding because basal atresia involves a significant proportion (50%) of all atretic follicles <5 mm in diameter [10] in the bovine. Whereas antral atresia largely involves the membrana granulosa, it can occur at these and larger sizes of follicles. Thus, we suggest that, during follicle growth and development, the order of susceptibility to death is the oocyte, then both the theca interna and membrana granulosa, and finally the membrana granulosa. The logical corollary of this is that there will be many real and inducible causes of follicular atresia.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Raymond J. Rodgers, Department of Obstetrics and Gynaecology, The University of Adelaide, Adelaide, South Australia 5005, Australia. FAX: 61 8 8303 4099; ray.rodgers{at}adelaide.edu.au
Received: 11 March 2004.
First decision: 9 April 2004.
Accepted: 6 May 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H Cardenas, E Jimenez, and W F Pope Dihydrotestosterone influenced numbers of healthy follicles and follicular amounts of LH receptor mRNA during the follicular phase of the estrous cycle in gilts Reproduction, March 1, 2008; 135(3): 343 - 350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J Cannon, J. S Davis, and J. L Pate The class II major histocompatibility complex molecule BoLA-DR is expressed by endothelial cells of the bovine corpus luteum Reproduction, May 1, 2007; 133(5): 991 - 1003. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R Braw-Tal and Z Roth Gene expression for LH receptor, 17{alpha}-hydroxylase and StAR in the theca interna of preantral and early antral follicles in the bovine ovary Reproduction, April 1, 2005; 129(4): 453 - 461. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
