| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ovary |
Department of Farm Animal Health,4
Department of Biochemistry and Cell Biology,5 Faculty of Veterinary Medicine, Utrecht University, 3508 TD Utrecht, The Netherlands
Holland Genetics,6 Arnhem, The Netherlands
Institute for Animal Science and Health,7 Lelystad, The Netherlands
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3 mm were dissected. Follicular health was determined on the basis of four parameters: 1) judgment of the degree of atresia by stereomicroscope, 2) incidence of apoptotic nuclei among the granulosa cells, 3) estradiol and progesterone concentrations, and 4) insulin-like growth factor-I (IGF-I) binding proteins (IGFBPs)-2, -4, and -5 concentrations in the follicular fluid. In addition to the preovulatory follicle, 3.1 other follicles, mainly sized 3–4.5 mm, were found to be healthy based on the proportion of apoptotic nuclei, and concentrations of estradiol/progesterone, and IGFBPs. The ability of these follicles to respond with growth on the preovulatory and periovulatory FSH surges was supported by a comparison to the follicular population of four cows 31–68 h after the LH surge. The present results point to an alteration of the view on the follicular wave. The larger follicles during the first days of the follicular wave are, in general, derived from follicles that also joined the previous wave. A portion of these growing follicles are estradiol active and compete for dominance. Other growing follicles lack estradiol production and are probably derived from rather atretic follicles. The first newly recruited follicles do not reach the size of 3 mm before 31 h after the preovulatory FSH surge. At that time, the larger follicles are already competing for dominance.
follicle, follicular development, granulosa cells, ovary, ovulatory cycle
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3 mm are stimulated to grow. This is followed by a selection phase, in which the growth of all follicles except the dominant follicle is slowing down and will eventually stop while the dominant follicle continues to grow. Finally, during the dominance stage, the dominant follicles stays at its maximum size while the subordinate follicles regress in size. For the first follicular wave, the recruitment, selection, and dominance phases are in general described to occur at Days 1–3, Days 3–6, and Days 6–8 of the estrous cycle, respectively. The second follicular wave starts between Days 9 and 12 [4, 7]. Recently, more information has been provided on the mechanism of selection of the dominant follicle. Insulin-like growth factor-I (IGF-I) and its binding proteins (IGFBPs) have been found to play a crucial role in this. IGF-I synergizes with FSH in stimulating granulosa cell proliferation and steroidogenesis [8]. Follicular action of IGF-I is mediated by the IGFBPs. In particular, the lower molecular weight IGFBPs (BPs-2, -4, and -5) negatively influence the effects of IGF-I. Recently provided data point to a mechanism in which a decreased concentration of IGFBP-4 is an important determinant for a follicle to become dominant. This results in more IGF-I becoming available, which supports the growth of the future dominant follicle in spite of the decreasing FSH concentration [5, 9].
In contrast with the increase in knowledge about the process of dominant follicle selection, very little is known about the dynamics of follicles around the recruitment period of the new follicular wave. Our present knowledge is based mainly on ultrasound examinations that are known to have limitations in distinguishing small follicles. Current information about the first part of the recruitment phase is rather variable. The timing of initiation of the first follicular wave is defined as the first day that 4- to 5-mm follicles are detected in a cohort of growing follicles. The estimations of this onset of the first wave range from the day before ovulation [10], to the day of ovulation (Day 1 of the estrous cycle), or 1 day after that [4, 11]. Also, the estimations of the number of recruited follicles range considerably from less than 10 [5] to 24 [10].
One generally accepted view is that a follicular wave consists of fresh follicles that were recruited from the population smaller than 3 mm at the start of the wave [1, 10]. This view, however, has never been sustained with evidence. It is mainly based on the assumption that all subordinate follicles will be too atretic to join the next wave. Information about the health status of subordinate follicles shortly prior to the preovulatory FSH surge, in particular the follicles smaller than 5 mm, however, is lacking. In addition, nothing is known about the response of follicles with a moderate degree of atresia to the preovulatory and periovulatory FSH rises.
The present study focused mainly on acquiring more knowledge on the follicular population just before the start of follicular recruitment. For this, 10 normally cyclic cows were ovariectomized around the time of the LH surge (and preovulatory FSH peak). Per cow, 11–18 follicles 3 mm were dissected. The health status of the follicles was determined on the basis of four parameters: 1) stereomicroscopic evaluation of the degree of atresia of the dissected follicle, 2) flow cytometric estimation of the incidence of apoptotic nuclei among the granulosa cells, 3) concentrations of estradiol and progesterone in the follicular fluid, and 4) concentrations of IGFBPs in the follicular fluid. As a comparison, four cows were ovariectomized 31–68 h after the LH surge. This was done to check whether data published before by others [1, 5, 8, 9, 11–13] could be confirmed within our experimental setting and to acquire a better understanding of the intrafollicular processes taking place during the recruitment period.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Holstein-Friesian cows (n = 35) were selected on general clinical examination and normal ovarian cyclicity during at least 3 wk as established by the progesterone concentration in peripheral blood samples taken 3 times a week. The animals were housed in groups of six, fed silage and concentrate (to a maximum of 2 kg per cow per day), and supplied water ad libitum. The experiments were performed in four sessions.
We aimed to study the follicular population at the end of an undisturbed estrous cycle (shortly before the LH surge) or at the beginning of the subsequent cycle.
For logistical reasons, cows were synchronized for the start of the cycle using an ear implant for 9 days (3 mg Norgestomet, Crestar; Intervet International, Boxmeer, The Netherlands) accompanied by treatment with 3 mg Norgestomet with 5 mg estradiol-valerate (i.m.; Intervet International). Two days before removal of the implant, prostaglandin (15 mg Prosolvin i.m.; Intervet International) was administered to ensure complete regression of the corpus luteum. It is known from previous experiments of our lab and others [14] that the cows ovulate on average 3 days after removal of the Crestar ear implant, and this day was referred to as Day 1 of the new cycle. The animal handling during this cycle was limited to blood sampling: Days 0–14, once per day; Days 14–18, every 6 h; Day 19, every 3 h until ovariectomy. Ovariectomies were planned on Days 20, 21, and 22 in sessions 1–3, and also on Day 23 in session 4. Progesterone assays were performed on Days 19 and 20 to assess luteolysis; a rapid LH assay [15] was performed on Day 20 and/or 21 to determine the time of the LH surge. The time point of luteolysis was defined as the time after which the progesterone level decreased in at least three successive samples. In an earlier study [16], the interval between the time points of luteolysis and the preovulatory LH surge was found to be 61.1 ± 2.5 h for cows. Based on these data, cows that were 48–62 h after luteolysis were considered to be shortly before the LH surge. Dependent on their cycle characteristics, cows were either designated to the pre-LH group (ovariectomy 48–62 h after luteolysis) or to the post-LH group (ovariectomy 1.5–3 days after the LH surge). Retrospectively, the 3-h blood samples were tested for LH [2] to determine possible occurrence of the LH surge within the pre-LH group and to confirm the results of the rapid LH assay for the post-LH group.
Twenty-one of the 35 cows had luteolysis within the proper time window and were ovariectomized. The pre-LH group consisted of 14 cows that were prior to the LH surge and 3 cows that were either in the middle of the LH surge or approximately 7 and 10 h thereafter. The time period up to 10 h after the LH surge was considered to be too short to have major effects on the follicular population. The post-LH group consisted of four cows that were 31, 49, 65, and 68 h after the LH surge. Ovariectomy was performed by laparotomy through a flank incision under local infiltration anesthesia [2]. Ovaries were collected in saline (0.9% w/v NaCl; 37°C) and transported immediately to the laboratory.
Follicles were first roughly excised using scalpel knives and placed in PBS supplemented with 0.1% (w/v) polyvinylalcohol (PBS-PVA) at room temperature (RT). Subsequently, they were dissected free from extraneous tissue under a stereomicroscope using fine forceps. The diameter was measured and recorded with an accuracy of 0.3 mm. Follicles were then slit and the follicular fluids were collected. The follicular wall was turned inside out and the granulosa cells were scraped off in a small volume of PBS-PVA. The follicular fluid and the granulosa cell suspension were centrifuged for 3 min at 600 x g. The supernatant of follicular fluid was frozen at -30°C for analysis of IGFBPs and steroids. The two cell pellets were resuspended in PBS-PVA (100 µl each) and pooled. These granulosa cells were further processed for flow cytometry.
From the pre-LH cows, all follicles 5 mm and on average 11.4 follicles of 3–4.5 mm were dissected. The remaining 3- to 4.5-mm follicles (3.2 ± 2.0, range 0–7) were left to limit negative effects of a too long postmortem period that was kept at a maximum of 2 h. From the post-LH cows, all follicles 3 mm were dissected. In total, 14.4 ± 1.8 (average ± SD, range 12–18) and 12.8 ± 2.0 (range 11–16) follicles were dissected from the pre-LH and post-LH cows, respectively.
Stereomicroscopic Evaluation
Follicles were evaluated stereomicroscopically for sessions 2, 3, and 4 but not for session 1. Dissected follicles were examined under a stereomicroscope and their appearance was evaluated as described before [17]. Follicles were classified as nonatretic when having a uniformly bright appearance, extensive and very fine vascularization, a regular granulosa layer, and no free-floating particles in the follicular fluid. Follicles with some loss of translucency and a slightly grayish appearance or with some very small free-floating particles in the follicular cavity were classified as light atretic. Follicles were classified as atretic if they had a gray appearance, blood vessels either irregularly filled with clotted blood or empty, partial detachment of the membrana granulosa, and many large free-floating globules in the antral cavity. Follicles with a dark, often spotted appearance and a very dark cumulus were classified as heavy-atretic. For comparisons with other parameters, such as those shown in Figure 4, only nonatretic follicles were classified as healthy.
|
Flow Cytometric Determination of Atresia in Granulosa Cells
To determine the proportion of apoptotic granulosa cells, a flow cytometric method was used as described by Darzynkiewicz and Li [18]. The method has been applied previously for granulosa cells of the pig [19] and the cow [12, 20].
Granulosa cell suspensions (100 µl) were vortexed for 30 sec to disperse the cells and placed on ice. Carefully, 900 µl ethanol (70% in PBS; precooled in ice) was added on top of the water phase; the contents were mixed and stored at -30°C for a maximum of 3 wk.
The cells were centrifuged for 3 min at 600 x g and washed once with 1 ml PBS. The cells were then extracted for 10 min with a mixture of 500 µl PBS and 500 µl extraction buffer (0.192 M Na2HPO4, 0.004 M citric acid; pH 7.8) at RT, and centrifuged again. The pellet was taken up in 0.2–1 ml staining solution (50 µg/ml propidium iodide and 50 µg/ml RNase [DNase-free, Sigma, St Louis, MO] in PBS) and stained for at least 30 min at RT.
Samples were filtered through a nylon mesh to remove possible aggregates of cells. Flow cytometric analysis of the extracted and stained nuclei was performed on a FACScan flow cytometer equipped with a 100-mW argon laser exciting at 488 nm (Becton Dickenson, Franklin Lakes, NJ). The propidium iodide fluorescence of nonaggregated nuclear events was detected in FL-3 (630-nm long-pass filter) in linear mode.
Because it cannot be excluded that some apoptotic nuclei might fall apart and give more than one event in the subdiploid area, the term apoptotic events will be used rather than apoptotic nuclei.
To enable comparisons with the atresia stages as determined on other parameters, follicles were classified on the basis of their flow cytometric score in one of the following three groups: <5%, 5–15%, and >15% apoptotic events. These criteria were based on data of Blondin and Sirard [21] on proportion of pycnotic cells of follicles categorized by these authors as nonatretic, slightly atretic, and atretic. For comparisons as given in Figure 4, only follicles with <5% apoptotic events were considered to be healthy.
Classification of Atresia on the Basis of Estradiol Production
Concentrations of progesterone and estradiol in follicular fluids were estimated by validated solid-phase 125I RIA method (Coat-A-Count TKPG and TKE, respectively; Diagnostic Products Corporation, Los Angeles, CA) as validated for cow plasma by Dieleman and Bevers [15] with slight modifications. Briefly, aliquots of 1–2 µl follicular fluid were diluted into 50 µl 0.02 M borate buffer (pH 8.5) in 0.9% (w/v) NaCl and extracted with 2 ml diethylether. After evaporation of the diethylether, the residues were dissolved in 250 µl zero calibrator plasma (Diagnostic Products Corporation) or borate buffer for the RIA of progesterone and estradiol, respectively. Duplicate aliquots of 100 µl were used in the respective RIAs. Extraction efficiency was determined in parallel samples with tritiated steroid. The limits of quantitation, using 1 µl of follicular fluid, were 3 and 2 ng/ml for progesterone and estradiol, respectively. The interassay coefficients of variation were 11% and 9%, and the intraassay coefficients of variation were 8% and 9% for progesterone and estradiol, respectively.
The criteria to classify follicles as healthy on estradiol production differed between the follicular size categories and were based on measurements described before [22, 23]. Follicles of 3–4.5 mm were classified as healthy when the follicular fluid contained more than 2 ng/ml estradiol (which was the lower threshold of the assay) unless the estradiol concentration was lower than 5 ng/ml in combination with a progesterone concentration of >50 ng/ml. Follicles of 5–7.5 mm were classified as healthy when they contained more than 5 ng/ml estradiol, while follicles 8 mm were classified as healthy when the estradiol concentration was higher than 5 ng/ml and the estradiol/progesterone concentration ratio was >0.5.
Classification of Atresia on the Basis of IGFBPs
To estimate the presence of IGFBPs in follicular fluid, SDS-PAGE was performed on a Multiphor II (Amersham-Pharmacia, Uppsala, Sweden) using precast SDS-gels (ExcelGel SDS Homogeneous 15; Amersham-Pharmacia). Follicular fluid was diluted 24-fold in sample buffer (1% w/v SDS, 0.0625 mM Tris/HCl, pH 6.8; 0.05 mg/ml bromophenol blue), boiled for 2 min, and cooled on ice. A sample of 12 µl was run according to manufacturer's instructions. To enable normalization, a reference sample was run on each gel. This reference was a pool of follicular fluids taken from follicles of 2–5 mm from ovaries obtained at a slaughterhouse. The proteins were blotted on nitrocellulose using a semidry blotting apparatus (Hoefer; Amersham-Pharmacia). After iodination with Na125I (Amersham-Pharmacia), labeled IGF-II (Boehringer, Mannheim, Germany) was used for incubation of the blots. Digitized images were obtained by use of a Phosphor Imaging System (Molecular Imager; Bio-Rad, Hercules, CA). The IGF-II-bound bands corresponded to those previously reported by de la Sota et al. [24]. The bands sized 22 and 28 kDa (IGFBP4), 30 kDa (IGFBP-5), and 35 kDa (IGFBP-2) were quantified using Molecular Analyst software (Bio-Rad). The quantity was expressed as a ratio (in percentage) to the corresponding bands in the reference sample. A ratio of 50 means that the intensity of the IGFBP-2, -4, and -5 bands of the sample was 50% from that of the reference sample. As explained in the Results, follicles with IGFBP levels below 40 were considered to be IGFBP healthy.
Statistics
Variations in number of follicles per size class, follicular diameters, proportion of apoptotic events, or steroid and IGFBP concentrations concern standard deviations. P values were estimated by chi-square analysis for comparisons between two groups of follicles such as the pre-LH vs. the post-LH group or between two size classes of follicles. Unpaired t-tests were used for all other comparisons. Probabilities with P < 0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Based on stereomicroscopic evaluation, flow cytometry, and estradiol production, 16 of the 17 pre-LH cows had a clear preovulatory follicle (POF). In the other pre-LH cow, the largest follicle had lost its dominance and two other follicles sized 9 and 10 mm had become new dominant follicles. This cow was excluded from the data analysis. Ten of the 16 pre-LH cows had next to the dominant follicle at least one other follicle of 10 mm. The diameters of the preovulatory follicles were on average 15.8 ± 2.0 mm (n = 16).
The average number of follicles with a size of 5 mm was significantly higher for the post-LH cows (n = 4) than for the pre-LH cows (n = 16, POFs not included): 6.0 ± 2.0 vs. 3.5 ± 2.5 per cow, respectively (P < 0.05). This confirms that a new follicular wave has been initiated in the post-LH cows. The average size of these 5-mm follicles did not differ between the post-LH and pre-LH group (6.7 ± 1.7 mm vs. 7.1 ± 2.7 mm, respectively).
Due to a technical failure in session 3, flow cytometric evaluation of the proportion of apoptotic granulosa cells could not be achieved for 6 of the 16 pre-LH cows. The comparisons given in the sections below are, therefore, based on 10 pre-LH and 4 post-LH cows unless otherwise indicated.
Follicular Atresia According to Stereomicroscopic Evaluation
In the pre-LH cows, all POFs were judged as nonatretic. Taking all pre-LH follicles except the POFs together (n = 123; 9 cows), the proportions of follicles that were nonatretic, light-atretic, atretic, and heavy-atretic follicles was 11%, 21%, 30%, and 38%, respectively. Inclusion of the data for session 3 gave approximately the same results: 12%, 24%, 30%, and 34% of the follicles were nonatretic, light atretic, atretic, and heavy atretic, respectively (n = 220; 15 cows). For the post-LH follicles (n = 38; 3 cows), the proportions of nonatretic, light-atretic, atretic, and heavy-atretic follicles were 29%, 24%, 18%, and 29%, respectively. In line with the expectation that follicular health improves during the post-LH follicular wave, the proportion of nonatretic and light-atretic follicles combined was higher (P = 0.02) for the post-LH group (53%) than for the pre-LH group (33%).
Follicular Atresia According to the Incidence of Apoptotic Events of Granulosa Cells
In the pre-LH cows, all POFs had less than 5% apoptotic events (average: 1.9% ± 1.4%). Follicles with less than 5% apoptotic events were also present within all other size categories (Fig. 1). When all follicles except the POFs were combined, the post-LH group (n = 4) had a higher proportion of follicles with <5% apoptotic events than the pre-LH group (n = 10): 57% vs. 41% (from 49 and 127 follicles, respectively; P = 0.05), which agrees with the expected improvement of follicular health during the first follicular wave.
|
As shown in Figure 1A, most of the pre-LH follicles 4 mm were either relatively healthy (<5% apoptotic events) or very atretic (>15% apoptotic events), whereas a high proportion of follicles with 5–15% apoptotic events were found for the 3- to 3.5-mm follicles. This was in marked contrast with the cows 49, 65, and 68 h post-LH (Fig. 1B), in which the majority of the follicles 5 mm and most of the 3- to 3.5-mm follicles had <5% apoptotic events. Most of the post-LH follicles with >5% apoptotic events were present in the 4- to 4.5-mm size category (Fig. 1B).
There was a very high correlation between the evaluation of atresia by stereomicroscope and by flow cytometry. From the follicles classified by stereomicroscope as nonatretic, 81% had <5% apoptotic events. Likewise, 96% of the stereomicroscopically determined heavy-atretic follicles gave 5% apoptotic events in flow cytometry (Fig. 2). The flow cytometric method did not detect differences between follicles stereomicroscopically judged as light atretic or atretic (Fig. 2). A high proportion (51 of 82; 62%) of the follicles stereomicroscopically judged as light atretic and atretic were classified as healthy (<5% apoptotic events) according to their flow cytometric score (Fig. 2).
|
Follicular Atresia According to the Steroid Concentrations in Follicular Fluid
As expected, within the pre-LH group, all POFs (n = 10) were estradiol active. Even when the POFs are not included, the proportion of healthy follicles based on estradiol concentration was significantly higher (P < 0.05) for the pre-LH group (39%, n = 127) than for the post-LH group (27%, n = 49). This is in marked contrast with the classification based on apoptotic events for which the pre-LH group had a lower incidence of healthy follicles than the post-LH group (41% vs. 57%, previous section).
In general, the estradiol concentration in the estradiol-healthy follicles increased with increasing follicular size and decreased with increasing incidence of apoptotic events (results not shown). In the pre-LH group, all POFs contained high estradiol concentrations (range 645–3236 ng/ml). The estradiol concentrations of the estradiol-healthy 3- to 3.5-mm follicles were higher in the pre-LH group (13 ± 13 ng/ml; n = 20) than in the post-LH group (5.1 ± 1.4 ng/ml; n = 6; P = 0.02). All post-LH cows possessed one follicle with a clearly higher estradiol concentration than the other follicles (Table 1). Within the cows 31, 49, and 65 h after LH, the estradiol concentration of the most active and second most active large follicles was positively correlated with the time period after the LH surge (Table 1). The estradiol production of the largest follicles from the 68 h after LH cows, however, was much lower than that from the 49- and 65-h post-LH cows (Table 1). The incidences of apoptotic events for the largest estradiol-healthy follicles were below 1.5% for the four cows.
|
The progesterone concentrations were very variable, particularly for the estradiol-unhealthy follicles. Overall, the progesterone concentrations were higher in estradiol-unhealthy follicles than in estradiol-healthy follicles and rose with increasing follicular diameter. For example, in the pre-LH group, the progesterone concentrations ranged from 4 to 26 ng/ml (average 13 ± 7 ng/ml; n = 16) for the estradiol-healthy 3- to 3.5-mm follicles, from 4 to 128 ng/ml (average 29 ± 36; n = 17) for the estradiol-unhealthy 3- to 3.5-mm follicles, and from 3 to 516 ng/ml (average 66 ± 121 ng/ml; n = 18) for the estradiol-unhealthy 5- to 7.5-mm follicles.
Follicular Atresia According to the IGFBP-2, -4, and -5 Concentrations in Follicular Fluid
IGFBP levels have been described to be at a rather basal level in healthy small follicles, which either decrease during follicular dominance or increase during atresia [5, 24].
As shown in Figure 3, a relatively high proportion of the follicles had an IGFBP level below 40, in particular among the follicles with <5% apoptotic events, while the remaining follicles had IGFBP levels ranging from 40 to 759. Based on this unequal distribution, IGFBP levels below 40 were considered to be basal (healthy) levels, while follicles with IGFBP levels above 40 were considered to be IGFBP unhealthy.
|
In all POFs (n = 10), small IGFBPs were completely absent. Only two of all other follicles (both 4–4.5 mm, pre-LH group) completely lacked small IGFBPs as well, while eight other pre-LH follicles had an IGFBP quantity of 2–4. The estradiol production of these follicles was in general high and varied from 4 to 23 ng/ml for the 3- to 3.5-mm follicles (n = 3), from 40 to 128 ng/ml for the 4- to 4.5-mm follicles (n = 4), 62 and 95 ng/ml for the 5- to 7.5-mm follicles (n = 2), and 26 ng/ml for a 9-mm follicle. In the post-LH group, only two follicles had an IGFBP quantity below 5; these concerned the two largest follicles of the cow 65 h after the LH surge that had estradiol concentrations of 86 and 194 ng/ml. The amounts of IGFBPs in the largest estradiol-healthy follicles from the cows 31, 49, and 68 h after the LH surge were 19, 15, and 8, respectively. This was in the same range as the IGFBP levels in the 3- to 3.5-mm follicles of post-LH cows with <5% apoptotic events and that were estradiol healthy (16 ± 5; range 8–25; n = 8) that are expected to be newly recruited.
The judgment on IGFBPs was highly correlated with that on apoptotic events: 93% of the follicles with less than 5% apoptotic events (n = 80) were IGFBP healthy, while this was 40% and 8% for the follicles with 5–15% (n = 40) and >15% apoptotic events (n = 53), respectively. The pre-LH and post-LH groups had similar correlations.
Overview of the Health Status of the Follicular Populations
Figure 4 shows an overview of the follicular population of the pre-LH cows (average of 10 cows) and four individual post-LH cows classified as healthy or unhealthy on the basis of evaluation by stereomicroscope (Fig. 4A), incidence of apoptotic events of granulosa cells (Fig. 4B), estradiol and progesterone concentrations (Fig. 4C), and amounts of small IGFBPs in the follicular fluid (Fig. 4D). This figure illustrates several interesting observations. Most important, irrespective of which health parameter is taken, the pre-LH cows possessed next to the POFs also multiple other healthy follicles. In both the pre-LH and post-LH groups, the health classification on IGFBPs is very much like that on the incidence of apoptotic events. The post-LH cows lacked follicles of 4–4.5 mm that were healthy for estradiol, while also many follicles of 5 mm are healthy for apoptosis and IGFBPs but unhealthy for estradiol. Follicles of 3–3.5 mm healthy on apoptosis, estradiol, and IGFBPs were first observed at 65 h post-LH, suggesting that these represented newly recruited follicles.
Correlations Between Apoptosis, Estradiol, and IGFBP Levels
In the pre-LH group, 29% of all follicles were healthy for apoptosis, estradiol, and IGFBPs, while 38% were unhealthy for these three parameters. From the post-LH follicles, 26% and 28% were healthy or unhealthy for each of the three parameters. The remaining follicles, 33% for the pre-LH and 54% for the post-LH group, were healthy for one or two of the three parameters. The comparisons between these parameters give indications about the timing of their regulation during atresia and after growth stimulation by rises in FSH.
Next to the POFs, the pre-LH cows had on average 3 follicles 5 mm, which probably represented subordinate follicles that lost the struggle for dominance. On average, two of these follicles were unhealthy for all parameters. On average, one follicle was healthy on the basis of the incidence of apoptotic events and IGFBPs but unhealthy on the basis of estradiol production. This indicates the block of estradiol production to be the first atretic event in large follicles. From the 3- to 4.5-mm follicles, on average, 2.8 follicles were healthy on the basis of the incidence of apoptotic events, IGFBPs, and estradiol production while 3.9 follicles were healthy for none of these parameters. In addition, 1.1 follicles were healthy on apoptotic events but unhealthy for estradiol, while 2.0 follicles were unhealthy on the basis of apoptotic events and healthy for estradiol, which was not significantly different. In 3- to 4.5-mm follicles, apoptosis therefore appeared to occur concomitantly with the block of estradiol production.
In the post-LH cows, all follicles except one that were healthy for estradiol were also healthy for IGFBPs and apoptotic events. The reversed combination, apoptosis healthy and/or IGFBP healthy but estradiol unhealthy, occurred often in the follicles of 4–4.5 mm (on average 1.5 per cow; 100% from all apoptosis healthy follicles) and 5 mm (on average 3.25 per cow; 65% from all apoptosis healthy follicles) but much less frequently in the 3- to 3.5-mm follicles (on average 0.5 per cow; 29% from all apoptosis healthy follicles). Assuming that at least the follicles of 5 mm were already >3 mm during the previous wave, this indicates that a decrease of IGFBP levels and apoptotic events was the first response on the FSH surges without immediate restoration of estradiol production.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many studies have indicated that the proportion of apoptotic nuclei is a very reliable parameter for follicular health [12, 19, 20]. This is confirmed in our study, in which all preovulatory follicles had less than 5% apoptotic events while 96% of stereomicroscopically determined heavy-atretic follicles had more than 5% apoptotic events. It is a reasonable assumption that follicles with less than 5% apoptotic nuclei are still able to respond with growth on the preovulatory and periovulatory FSH rises. To our knowledge, this is the first study in which the follicular population shortly before the LH surge has been extensively studied. One unexpected finding is that the pre-LH ovaries contain a relatively high number of follicles with less than 5% apoptotic events, in particular in the 3- to 4.5-mm category. On average, the pre-LH cows contained 1.5 follicles of 3–3.5 mm, 2.6 follicles of 4–4.5 mm, 0.6 follicles of 5–7.5 mm, and 0.4 follicles of 8 mm with <5% apoptotic events. In the post-LH group, the number of follicles with less than 5% apoptotic events was much higher: on average 2.25 sized 3–3.5 mm, 0.75 sized 4–4.5 mm, and 4.75 sized 5 mm. When only the 49- to 68-h post-LH surge cows are taken, the proportion of 3- to 3.5-mm follicles with <5% apoptotic events increased during the 2–3 days following the LH surge from 33% in the pre-LH group to 75% in the post-LH group. These healthy 3- to 3.5-mm post-LH follicles, therefore, were very likely newly recruited from the population <3 mm. The absence of healthy 3- to 4.5-mm follicles in the 31-h post-LH cow suggests that it might take more than 31 h before newly recruited follicles reach the size of 3 mm. The dip in healthy follicles in the 4- to 4.5-mm post-LH group indicates that most newly recruited follicles did not reach that size yet. The average number of apoptosis-healthy 5-mm follicles rose from 1 in the pre-LH to 4.75 in the post-LH group. The most likely explanation for this finding is that the healthy 5-mm post-LH follicles developed from follicles that were already larger than 3 mm prior to the LH surge.
Both in the pre-LH and post-LH cows, the levels of small IGFBPs were highly correlated to the proportion of apoptotic events, indicating a relatively rapid effect of IGFBP concentration on the apoptosis of granulosa cells. This agrees with the reported crucial role of small IGFBPs in selection of the dominant follicle. The future dominant follicle attains FSH independence by keeping the levels of small IGFBPs, in particular IGFBP-2 and-4, low [5, 9]. A rather subtle difference in small IGFBP levels appears to be sufficient for this selection. Our finding that the IGFBP levels of the largest estradiol-active follicles of the post-LH cows were in the same range or just a little lower than that of presumably newly recruited 3- to 3.5-mm follicles is in agreement with this. A complete lack of IGFBPs was only found in the POFs of the pre-LH cows, which agrees with data from Funston et al. [24]. This indicates that the complete disappearance of the IGFBPs occurs rather late during follicular dominance. A large increase of IGFBP levels has been reported to occur at more advanced stages of atresia [25]. This has been confirmed by Austin et al. [5], who found that a large increase of IGFBPs in the subordinate follicles did not occur before 84 h after the periovulatory FSH peak (which is 108 h after the LH peak). Because our post-LH cows were all before 70 h after the LH peak, this implies that all post-LH follicles with high IGFBP levels were very likely derived from the previous wave.
Our comparisons on the health parameters for the pre-LH group indicate a block of estradiol production to be the first atresia event in large but not in small follicles. A main difference between these two populations is that large follicles have competed within the selection process for dominance. Convincing data have been published that a loss of capacity to produce estradiol is one of the first events distinguishing the future subordinate follicle from the future dominant follicle [5, 13, 26, 27]. The loss of estradiol production has been found to precede cessation of follicular growth and the increase of granulosa cell apoptosis [5, 26], which is in line with our observations on large follicles. No data have been reported before about the sequence of the induction of apoptosis and decrease of aromatase activity in small follicles.
Although the four post-LH cows were relatively soon after the LH surge (31, 49, 65, and 68 h post-LH, thus Day 1 or 2 of the new cycle), all of them had one follicle with a clearly higher estradiol concentration than the other follicles (Table 1). This suggests that the future dominant follicle is selected at this early stage already.
The 3- to 3.5-mm follicles of the 65- and 68-h post-LH cows that were healthy on apoptotic events and/or IGFBPs (Fig. 4) probably represent newly recruited follicles. All of these 3- to 3.5-mm follicles were estradiol active. Newly recruited follicles, therefore, become estradiol active very early after reaching the size of 3 mm. The involvement of attainment of aromatase activity in follicular recruitment has been reported before. Using in situ hybridization, mRNAs for the most steroidogenic enzymes have been detected in the theca cells from follicles from the preantral stage onward. Aromatase mRNA, however, was not detected in nonrecruited follicles <4 mm in diameter. Recruitment of follicles 4 mm was found to coincide with expression of mRNAs from aromatase and cytochrome P450 side-chain cleavage in the granulosa cells [1, 11, 26]. The start of transcription of aromatase mRNA in freshly recruited follicles of 4 mm is in agreement with the present results, with the difference that we detected estradiol production in follicles of 3 mm already. Probably, the amount of aromatase mRNA in 3-mm follicles is below the threshold for detection by in situ hybridization.
The situation in the 5-mm post-LH follicles is very different from that of the 3- to 4.5-mm follicles. The post-LH cows had on average 4.75 apoptosis healthy follicles 5 mm, but only 1.75 from these were estradiol active. The lack of estradiol activity indicates that these follicles were not newly recruited but likely developed from > 3-mm follicles that were rather atretic at the end of the previous wave. Apparently, these follicles still had a sufficient proportion of nonapoptotic granulosa cells that could respond on the FSH rises with growth. During this growth, both the IGFBP levels and proportion of apoptotic events decreased, while aromatase activity was not yet restored. Based on aromatase mRNA levels in granulosa cells, indications that part of the growing follicles 4 mm does not produce estradiol have been described before [11].
The post-LH cows, therefore, appear to have three groups of growing follicles: 1) the 3- to 3.5-mm follicles healthy on all parameters, probably representing newly recruited follicles, 2) the 5-mm follicles healthy on all parameters, probably developed from follicles that were healthy on all parameters prior to the LH surge already, and 3) follicles healthy for apoptosis and/or IGFBPs but unhealthy for estradiol, that likely developed from rather atretic pre-LH follicles. These findings point to some adaptations to the present view on the follicular wave, which are depicted in Figure 5. The pre-LH population can be roughly divided into three groups. One group is healthy on all parameters and responds on FSH with growth and increased estradiol production. The second group is light atretic and estradiol inactive. The rises of FSH stimulate these follicles to grow but estradiol production is not restored within the first 3 days. The third group is too atretic to respond with growth on FSH.
|
This concept conflicts with the generally accepted model that the follicular wave consists of three consecutive stages: 1) recruitment from new follicles, 2) selection of one or more of these recruited follicles to become dominant, and 3) the dominance period [1]. This model, however, is mainly based on ultrasound studies, which have limitations in distinguishing small follicles. The assumption that the largest follicles of the follicular wave are newly recruited from the <3-mm pool has never been sustained with evidence. In contrast, several studies sustain the presently proposed model.
The finding that the future dominant follicle can be distinguished at a rather early time point has been described before by several others. Sunderland et al. [7] reported the future dominant follicle to be sized 6.2 ± 0.3 mm at approximately 30 h after the LH surge. Austin et al. [5] described the largest follicles to have the highest estradiol concentration from 5 h after the periovulatory FSH peak (which is approximately 29 h after the LH surge) onward. Using ultrasound monitoring of follicular sizes, Kulick et al. [28] reported the future dominant follicle to be significantly larger than the future largest subordinate follicle at an even earlier time point: 5.8 ± 5.5 h after the LH surge. It is very unlikely for a newly recruited follicle to reach that stage at such an early time point.
Our finding that newly recruited follicles do not appear before Day 2 of the cycle is strongly supported by a study of Hagemann et al. [12]. In this study, cows were ovariectomized at defined stages during the cycle, after which all follicles 3 mm were dissected. The average number of 3- to 5-mm follicles per cow was 26 on Day 15 of the cycle, decreased to 10 on Day 2 of the next cycle, and increased to 27 on Day 7. This indicates that new follicles are mainly joining the follicular wave after Day 2, concomitantly rather than prior to the time point of selection of the dominant follicle. An additional finding of Hagemann et al. [12] was that the follicles of 3–4 mm were significantly less affected by atresia than the >5-mm follicles at Day 7, which is in line with our findings shortly prior to the LH surge. This might be a consequence of the later time of recruitment of these smaller follicles and therefore a shorter time period of exposure to a low FSH concentration. Alternatively, small follicles might be less sensitive for low FSH concentrations than larger follicles.
The concept that atresia can be reversed is certainly not new. The first signs of atresia are detected in large subordinate follicles rather soon after selection of the dominant follicle [5, 29]. When the dominant follicle is ablated within 2 days after deviation, the largest subordinate follicle, however, is still able to respond with growth on the rise of FSH concentration and becomes the new dominant follicle [30, 31]. Reversibility of atresia has also been reported as the mechanism by which exogenously administered gonadotrophins stimulate the growth of a large population of follicles [32, 33].
In the present study, 73% of the follicles with less than 5% apoptotic events (n = 49) were judged as at least light atretic by stereomicroscopic observation. In an earlier study using histology [17], the incidences of pycnotic nuclei of granulosa cells have been found to gradually increase in follicles judged as nonatretic, light atretic, atretic and high atretic. Most pycnotic cells were found in the part bordering the antrum and in the follicular cavity. The fact that follicles judged by stereomicroscope as light atretic and atretic did not differ in proportion of apoptotic nuclei suggests that not all pycnotic nuclei are detected as apoptotic events by flow cytometry. A possible explanation for this finding has been provided by van Wezel et al. [34]. These authors found that the pycnotic nuclei at the antrum side of the membrana granulosa stained negatively by TUNEL (terminal deoxy-UTP nick end labeling), indicating lack of extensive DNA breaks. The authors suggested that the granulosa cells that are sloughed of into the follicular fluid are not dying due to apoptosis but as a result of terminal differentiation [34]. The granulosa cells located near the outside of the follicle behaved differently: Nuclei with a pycnotic appearance also stained positively by TUNEL, indicating an apoptotic pathway of cell death for this cell population. Stereomicroscopic observation might, therefore, measure an earlier stage of atresia than flow cytometry.
In summary, our results put forward the following model. The largest follicles that respond with growth on the preovulatory and periovulatory FSH rises did join the previous wave as well. A part of these follicles are estradiol healthy and compete around Day 2 of the cycle for dominance. Another part of the growing follicles is estradiol unhealthy and is probably derived from rather atretic follicles that had lost aromatase activity. Recruitment of new follicles from the pool of <3-mm follicles starts at Day 2 of the cycle and proceeds, therefore, concomitantly with selection of the dominant follicle.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: P.J.M. Hendriksen, Department of Urology, Erasmus MC, JNI, Rm Be355a, Dr. Molewaterplein 50, 3015 GE Rotterdam, The Netherlands. FAX: 31 10 408 9386; p.hendriksen{at}erasmusmc.nl
3 Current address: Department of Urology, Erasmus MC, Rotterdam, The Netherlands
Received: 16 June 2003.
First decision: 20 July 2003.
Accepted: 14 August 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylase and cytochrome P450 aromatase in bovine follicles during the first follicular wave. Endocrinology 1995 136:981-989[Abstract]This article has been cited by other articles:
|
|
 |
|
  A. N. Fatehi, M. M. Bevers, E. Schoevers, B. A. J. Roelen, B. Colenbrander, and B. M. Gadella DNA Damage in Bovine Sperm Does Not Block Fertilization and Early Embryonic Development But Induces Apoptosis After the First Cleavages J Androl, March 1, 2006; 27(2): 176 - 188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Greenaway, P. A. Gentry, J.-J. Feige, J. LaMarre, and J. J. Petrik Thrombospondin and Vascular Endothelial Growth Factor Are Cyclically Expressed in an Inverse Pattern During Bovine Ovarian Follicle Development Biol Reprod, May 1, 2005; 72(5): 1071 - 1078. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E L Gastal, M O Gastal, M A Beg, and O J Ginther Interrelationships among follicles during the common-growth phase of a follicular wave and capacity of individual follicles for dominance in mares Reproduction, October 1, 2004; 128(4): 417 - 422. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Rubio Pomar, B. A.J. Roelen, K. A. Slot, H. T.A. van Tol, B. Colenbrander, and K. J. Teerds Role of Fas-Mediated Apoptosis and Follicle-Stimulating Hormone on the Developmental Capacity of Bovine Cumulus Oocyte Complexes In Vitro Biol Reprod, September 1, 2004; 71(3): 790 - 796. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
