| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ovary; |
Department of Biomedical Sciences, Cornell University, Ithaca, New York 14853
ABSTRACT
In cattle and other species in which the pool of resting, primordial follicles is formed during fetal life, little is known about the regulation of the early stages of ovarian follicular development. We used histological morphometry and a combination of observations in vivo and experiments in vitro to study the timing and regulation of follicle formation and the acquisition of the capacity of primordial follicles to initiate growth in cattle. In vivo, primordial, primary, and secondary follicles were first observed around Days 90, 140, and 210 of gestation, respectively. The long interval between the first appearance of primordial and primary follicles suggests that primordial follicles are not capable of activating when they are first formed, or they are inhibited from activating. This hypothesis was confirmed by the finding that most primordial follicles in pieces of ovarian cortex obtained from fetal ovaries older than 140 days activated (i.e., initiated growth) after 2 days in vitro, whereas follicles in cortical pieces from 90- to 140-day-old fetal ovaries did not. We tested the hypothesis that the oocytes of newly formed primordial follicles are not in meiotic arrest and found that before Day 141, most oocytes (
73%) were in prediplotene stages of prophase I, whereas after Day 140, the majority of oocytes (85%) had arrested at the diplotene stage. This observation was further confirmed by the finding that levels of mRNA for YBX2, a protein associated with meiotic arrest, were 2.3 times higher in ovarian cortical pieces isolated after versus before Day 141. Primordial follicles in cortical pieces from 90- to 140-day-old fetal ovaries did activate during a longer, 10-day culture, but activation could be inhibited by adding estradiol or progesterone, but not dihydrotestosterone (all at 10–6 M). Fetal ovaries secreted estradiol in vitro, and secretion by ovaries from 83 to 140-day-old fetuses declined precipitously (30-fold) with age, consistent with the hypothesis that estradiol inhibits activation of newly formed primordial follicles in vivo. In summary, the results show that newly formed primordial follicles do not activate in vivo or within 2 days in vitro and that capacity to activate is correlated with achievement of meiotic arrest by the oocyte and can be inhibited by estradiol, which fetal ovaries actively produce around the time of follicle formation.
cattle, estradiol, follicular development, gametogenesis, meiosis, ovarian follicle, ovary, primary follicles, primordial follicles, steroid hormones
Throughout reproductive life, the majority of follicles in mammalian ovaries are nongrowing, primordial follicles with an oocyte surrounded by a single layer of flattened granulosa cells. Individual primordial follicles remain quiescent for variable lengths of time. Follicle activation, the transition from quiescence to the growth phase, is characterized by a change in shape of the granulosa cells from flattened to cuboidal and the initiation of oocyte growth. Activation of primordial follicles is a nonreversible process, and therefore it is important in regulating the size of the resting primordial follicle pool, which in turn affects reproductive lifespan and fertility. Mechanisms regulating primordial follicle activation are still poorly understood, especially in large mammals.
In rodents, follicle formation occurs shortly after birth and is quickly followed by a synchronous, initial wave of follicle activation and growth [1, 2]. In contrast, in most domestic animals and in primates, including humans, follicle formation is initiated during fetal life and occurs asynchronously over a fairly long period of time [3–5]. During that time, some primordial follicles are activated and begin development at the same time that other follicles are still being formed. In the ovaries of fetal sheep, primordial, primary, and secondary follicles first appear at approximately Days 75, 100, and 120, respectively, of the 145-day gestation period [4]. In humans, primordial and primary follicles first appear around Days 130 and 170 of gestation, respectively [5]. In cattle, Russe [3] first observed primordial, primary, and secondary follicles at Days 90, 140, and 210, respectively, of the 280-day gestation period, whereas Tanaka et al. [6] reported these transitions at 74, 91, and 120 days. Erickson [7] observed that establishment of the first bovine primordial follicles was completed at approximately Day 130 of gestation, whereas Dominguez et al. [8] first detected primordial follicles in 90-day-old fetuses, in agreement with Russe [3]. Given these conflicting reports on the timing of follicular formation and development, the first aim of present study was to determine when primordial follicles are first formed, begin to activate, and develop to the secondary stage in bovine fetal ovaries in vivo.
Although how follicle activation is regulated in vivo is largely unknown, spontaneous activation of primordial follicles in vitro has been achieved for several species, including rodents [9, 10], cattle [11, 12], nonhuman primates [13], and humans [14]. To study regulation of follicle activation, our lab developed a serum-free culture system that supports activation in vitro of primordial follicles in small pieces of bovine and baboon ovarian cortex cultured in medium containing ITS+ (insulin/transferrin/selenous acid + BSA and linoleic acid) [11, 13]. Most primordial follicles in pieces of ovarian cortex from adult cattle or from fetal calves during the third trimester activate within 2 days of culture [11, 12]. However, primordial follicles first appear much earlier during bovine pregnancy (around Day 74, 90, or 130, depending on the study), and it is interesting that activated (i.e., primary) follicles are not observed until 17 to 50 days (depending on the study) after primordial follicles first appear [3, 6, 7]. Whether bovine primordial follicles have the capacity to activate as soon as they are formed, but are somehow prevented from activating in vivo, or whether they develop that capacity at a later time during gonadal development is not known. Therefore, the second aim of this study was to test the hypothesis that newly formed primordial follicles do not have the capacity to activate in vitro. There is evidence from earlier studies that oocytes in most primordial follicles in bovine fetal ovaries collected around Days 90–120 of gestation are still in the initial stages of prophase I of meiosis, whereas oocytes are arrested in the diplotene stage in all primordial and activated follicles in late fetal and postnatal bovine ovaries [15]. Thus, the third aim was to test the hypothesis that acquisition by primordial follicles of the ability to activate is correlated with the achievement of meiotic arrest by their oocytes. The questions addressed in this study are of both basic and practical interest, since the pool of resting primordial follicles may be used, eventually, as a resource for propagating valuable domestic animals or ameliorating human infertility.
Collection of Bovine Fetal Ovaries
Bovine female fetuses (primarily Holstein) were obtained at a local slaughterhouse (Cargill Regional Beef, Wyalusing, PA), and fetal age was estimated by crown-rump length [16]. Ovaries (donated by Cargill Regional Beef) were collected and transported to the laboratory in Leibovitz L-15 medium (Invitrogen, Carlsbad, CA) supplemented with 1% fetal bovine serum, 50 IU/ml penicillin, and 50 µg/ml streptomycin (Invitrogen) at ambient temperature (20°C–22°C), as previously described [11].
Ovarian cortical cultures.
The ovarian cortex of fetal ovaries at different gestational ages was dissected from the medullary tissue and cut into 0.5- to 1-mm3 pieces as described previously [11]. Four freshly isolated cortical pieces from each ovary were fixed immediately for histological analysis of numbers of primordial and primary follicles in Day 0 (uncultured) controls. The other cortical pieces were placed on uncoated culture well inserts (two pieces per well; two wells per treatment; Millicell-CM, 0.4-µm pore size; Millipore Corp., Bedford, MA) in the wells of 24-well Costar culture plates (Corning Inc., Corning, NY) with 300 µl serum-free medium, consisting of Waymouth medium MB 752/1 (Invitrogen) supplemented with 25 mg/l pyruvic acid (Sigma Chemical Co., St. Louis, MO), antibiotics (50 IU/ml penicillin, 50 µg/ml streptomycin; Invitrogen), and ITS+ (6.25 µg insulin, 6.25 µg transferrin, 6.25 ng selenous acid, 1.25 mg BSA, 5.35 µg linoleic acid per milliliter; Collaborative Biomedical Products; Becton Dickinson Labware, Bedford, MA). In one experiment, ovarian cortical pieces were cultured with estradiol, progesterone, or dihydrotestosterone (DHT; all at 10–6 M). The concentrations of steroids were chosen based on previous studies on the effects of steroids on early follicular development [17–19]. Cortical pieces were cultured at 38.5°C in a humidified incubator gassed with 5% CO2:95% air for 2 or 10 days (depending on the experiment) and then fixed for histological analysis of follicular activation and growth at the end of the culture period.
Gonadal cultures.
In one experiment, whole ovaries (n = 16) dissected from bovine fetuses at different stages of gestation (Days 83–255) were cut into pieces (0.5–1 mm3), and the pieces were cultured at 38.5°C in six-well Costar culture plates (Corning) with 2 ml serum-free medium. The medium and culture conditions were as described above, except that no culture well inserts were used. The medium was collected after 24 h and stored at –20°C for later assay of steroids.
Development of bovine fetal ovaries in vivo. One ovary from fetuses at different gestational ages (n = 19) was trimmed of any extra tissue, weighed, and then fixed in Bouin solution, followed by dehydration and embedding in paraffin. Ovaries then were sectioned at a thickness of 5 µm. A total of 3–10 consecutive sections taken from the middle of each ovary were stained with hematoxylin-eosin and examined for ovarian morphology.
Follicular activation and growth in cortical pieces in vitro. Follicular and oocyte survival and growth in cortical pieces were assessed by histological morphometry as previously described [11]. Briefly, on Days 0, 2, or 10 of culture, cortical pieces were fixed for 1 h in 2.5% glutaraldehyde, 2.5% formaldehyde in 0.075 M cacodylate buffer, pH 7.3. The pieces then were embedded in LR White plastic (EMS, Fort Washington, PA), and 2-µm serial sections were cut with a glass knife. Every other set of 10 consecutive sections from each piece of ovarian cortex was mounted on gelatin-coated slides and stained with toluidine blue. To avoid counting or measuring the same follicle twice, only one section in each set of 10 consecutive sections was examined, and only follicles with the germinal vesicle present in the section were counted and measured. Using criteria defined previously [11], follicles were first classified as primordial (an oocyte surrounded by one layer of flattened pregranulosa cells), primary (an oocyte surrounded by a single layer of cuboidal granulosa cells), or secondary (an oocyte surrounded by two or more layers of cuboidal granulosa cells). Because very few secondary follicles were observed in cortical pieces, the number of secondary follicles was combined with the number of primary follicles and expressed as "growing" or "activated" follicles in some experiments. Follicles were further classified as healthy (an oocyte with an intact germinal vesicle and nucleolus and with no more than three cytoplasmic vacuoles) or atretic (in early stages, an oocyte with more than three cytoplasmic vacuoles and slight condensation of chromatin; in later stages, fragmentation of the oocyte cytoplasm and/or nucleus and heavy chromatin condensation). The microscopic image was projected on a video monitor. The diameters of individual healthy follicles and their oocytes were measured by a computer-driven image analysis program (NIH Image; National Institutes of Health, Bethesda, MD). Each follicle and the enclosed oocyte were measured in two dimensions, and the two measurements were averaged.
Stages of meiotic prophase in oocytes of primordial follicles in bovine fetal ovaries in vivo. In histological sections of fetal ovaries obtained at different stages of gestation, the meiotic stage of oocytes in primordial follicles was examined. Using criteria defined previously [20], the numbers of oocytes in prediplotene stages (including leptotene, zygotene, and pachytene) and the diplotene stage of meiotic prophase I were determined by direct counts of oocytes at diplotene or prediplotene stages in sections prepared from at least three fetuses at each stage of gestation. Briefly, oocytes at the prediplotene stage can be recognized by the fine "speckled" appearance of their nuclear chromatin (lepotene), chromosomes showing marked polarization (zygotene), or a beaded appearance of the thickened chromosomes (pachytene). Oocytes at the diplotene stage are characterized by the appearance of well-defined chromosomes and faint lateral projections from some of the chromomeres [20].
Real-Time PCR Analysis of mRNA for YBX2 in Bovine Cortical Pieces
The abundance of mRNA for germ cell-specific Y box protein 2 (YBX2, also known as MSY2), a protein found only in diplotene and mature oocytes [21], in cortical pieces at different stages of gestation was quantified by real-time PCR. Cortical pieces were dissected from fetal ovaries (n = 16) obtained at different gestational stages. Total RNA in cortical pieces was isolated using VERSAGENE RNA purification kits (Gentra Systems Inc., Minneapolis, MN), according to the manufacturer's instructions. The amount and purity of RNA extracted were determined spectrophotometrically. Five micrograms of total RNA were used to generate cDNA in a 20-µl reaction mixture, using random primers and Superscript II reverse transcriptase (Invitrogen). Then, the reaction mixture was diluted with 40 µl distilled water and equal amounts (2.25 µl) from each sample of cDNA were used in the real-time PCR reaction.
Complementary DNA was subjected to real-time PCR quantification using an ABI 7000 series real-time machine (Applied Biosystems Inc., Foster City, CA) and SYBR Green Master Mix (Eurogenetec, Seraing, Belgium). Validated primer sequences for bovine YBX2 were used, and the sequences were as follows: forward, 5'-GTG CTG GCA ATC CAA GTC C-3'; reverse, 5'-CTT CTC TCC TTC CAC GAC ATC-3' [22]. Levels of RNA polymerase II mRNA were used as the reference mRNA for normalization, because it has been shown that it is the gene with the most constant expression in tissues [23]. Primer sequences specific for bovine RNA polymerase II have been validated previously in our lab. Amplification reactions were performed in duplicate for both YBX2 and RNA polymerase II for every cDNA sample for 40 cycles in 96-well real-time PCR plates. The threshold cycle (CT) was the cycle when the fluorescence surpassed an arbitrary level; the CTs from duplicate PCR reactions were averaged, and the average CT was used to calculate relative expression. The amount of YBX2 mRNA was calculated using the method of Pfaffl (relative expression = Etarget
CT(target)/EreferenceCT(ref), where E is the efficiency of the primer set and CT is the control CT minus the sample CT [24]). The identity of the product amplified was confirmed by sequencing at the Cornell University Life Sciences Core Laboratories Center after electrophoresis on an agarose gel.
Duplicate aliquots (25 or 50 µl) of unextracted medium collected from gonadal cultures were measured for estradiol, progesterone, testosterone, and androstenedione by radioimmunoassay (RIA). The characteristics of the antiserum and other aspects of the RIAs were as described previously [2, 25]. All samples were assayed in single RIAs for each hormone. The intraassay coefficients of variation (mean ± SEM) for the estradiol, progesterone, testosterone, and androstenedione RIAs were 5.2% ± 1.5%, 1.2% ± 0.7%, 4.2% ± 1.8%, and 4.3% ± 1.2%, respectively.
Mean numbers of primordial and activated (growing) follicles per section were calculated for freshly isolated and cultured cortical pieces from fetuses at different gestational stages (two to nine ovaries per group). Mean percentages of activated follicles in cultured cortical pieces and of oocytes in primordial follicles at the prediplotene or diplotene stage in histological sections from fetal ovaries at four different gestational stages were also calculated. Sets of data were log transformed if Hartley test indicated heterogeneity of variance among the means. Data are presented as mean ± SEM of nontransformed data. Statistical differences (P < 0.05) among the means were evaluated by one-way (percentage of oocytes at prediplotene/diplotene stages, levels of YBX2 mRNA, and levels of steroid hormones) or two-way (numbers of follicles per section and percentage of activated follicles per section; with gestational age or treatment and fetus as the two factors) ANOVA. When a significant P value was obtained with ANOVA, differences among individual means were tested using Duncan multiple range test. Nonlinear regression and Pearson correlation analyses were used to test the relationship between ovarian secretion of steroids and stage of gestation.
Experiment 1: The Timing of Follicular Development in the Bovine Fetal Ovary
To determine the timing of follicular development in bovine fetal ovaries, one ovary from each of 19 fetuses (Days 83–244 of gestation; length of gestation = 281 days) was examined. As expected, ovarian weight (based on one ovary per fetus) increased with the progression of gestation (Fig. 1). Like adult ovaries, fetal ovaries consist of an outer cortex and an inner medulla. In ovaries from 80- to 90-day-old fetuses (Fig. 2, A through C), most of the cortex was divided, by the penetration of medullary fibrovascular tissue, into many ovigerous cords (Fig. 2A) containing germ cells (oocytes and oogonia; Fig. 2B), whereas the outer parts of the cortex were not yet divided and were packed with mostly oogonia. The oocytes in the ovigerous cords near the interface of the cortex and medulla were in various stages of first meiotic prophase (Fig. 2C), but no primordial follicles were observed. In fetal ovaries obtained at Days 91–140 of gestation, ovigerous cords were still present in the outermost cortex. In addition, some primordial follicles were observed in the inner cortex (Fig. 2D). In ovaries obtained during the mid-second to early third trimester of gestation (Days 141–210), most ovigerous cords had regressed as more primordial follicles were formed and a few primary follicles were identified (Fig. 2E). Ovigerous cords were found only in the outermost layer of cortex and with age appeared to gradually become segregated from the surface epithelium of the ovary by connective tissue. From Day 211 on, the ovigerous cords had essentially disappeared (Fig. 2F). Layers of connective tissue developed progressively under the surface epithelium, and growing follicles at more advanced stages were evident in the innermost region of the cortex.
|
|
Based on the qualitative observations of fetal ovarian morphology detailed above, ovaries collected from fetuses at different ages of gestation were divided into four groups based on the following observed times of transition in follicular development: 1) Days 80–90, no follicles present; 2) Days 91–140, primordial follicles present; 3) Days 141–210, primary follicles appear; and 4) >210 days, secondary follicles appear. These groups were used in the designs of most of the experiments that follow.
Experiment 2: Effects of Gestational Age on Follicle Activation in Bovine Cortical Pieces In Vitro
Numbers of primordial and primary follicles in freshly isolated bovine cortical pieces. Ovarian cortical pieces were dissected from the second ovary of the 19 fetuses in experiment 1. Some pieces were cultured (see below), and others were fixed immediately and analyzed by histological morphometry. In the freshly isolated cortical pieces, no follicles were present until after Day 90 of gestation, when primordial follicles were first observed (Fig. 3), consistent with the results of morphological analysis of the intact ovaries from the same fetuses (described above). Numbers of primordial follicles increased with the progression of gestation and plateaued after Day 140. Primary follicles were rare until after Day 140 of gestation. Secondary follicles (Fig. 2F) were first observed in cortical pieces isolated after Day 210 of gestation, but the cohort of growing follicles comprised mostly primary follicles.
|
Effect of stage of gestation on follicle activation in bovine cortical pieces in vitro. Cortical pieces from the fetal ovaries at the four gestational stages described above were cultured for 2 days to examine when during gestation bovine primordial follicles first acquire the capacity to activate in vitro. After 2 days in culture, no follicles were observed in cortical pieces from fetuses obtained before Day 91 of gestation (Fig. 4A). In cortical pieces from 91- to 140-day-old fetuses, primordial follicles and an occasional primary follicle were observed on Day 0; the number of follicles in the two developmental categories had not changed after 2 days of culture (i.e., follicle activation did not occur; Fig. 4A). In contrast, in cortical pieces isolated after Day 140 of gestation, there was a marked decrease in the number of primordial follicles and a concomitant increase in growing follicles (primary follicles plus the occasional secondary follicle) after the 2-day culture (P < 0.05; Fig. 4A), similar to our previous results for fetal ovaries obtained during the last trimester of pregnancy [11]. Similarly, the percentage of follicles that were activated (growing) increased during culture only when cortical pieces were obtained after Day 140 of gestation (P < 0.05; Fig. 4B), suggesting that primordial follicles in fetal bovine ovarian cortex cannot activate within 2 days in vitro until after 140 days of gestation.
|
Experiment 3: Numbers of Oocytes in Prediplotene and Diplotene Stages of Meiotic Prophase in Fetal Ovaries at Different Stages of Gestation
To test the hypothesis that primordial follicles acquire the capacity to activate when their oocytes reach meiotic arrest, we examined the meiotic stages of oocytes in primordial follicles in bovine fetal ovaries obtained at different gestational ages. The percentage of primordial follicles with oocytes that had reached the diplotene stage, the resting stage of prophase I of meiosis, increased as cortical pieces were isolated progressively later in gestation (Fig. 5). The ovarian cortex of fetuses less than 90 days of gestation was filled with ovigerous cords containing oogonia in mitosis and oocytes in all stages of meiotic prophase, with the great majority in prediplotene stages; no follicles were observed. Primordial follicles appeared after Day 90 of gestation, and most were located in the ovarian cortex, as in the experiment described above. Most oocytes in primordial follicles (73%) were at prediplotene stages in ovaries from Days 91–140 of gestation. In contrast, in ovaries examined after Day 140 of gestation, most oocytes in primordial follicles (>80%) had reached the resting meiotic state (diplotene stage; Fig. 5).
|
Experiment 4: Levels of YBX2 mRNA in Cortical Pieces from Fetal Ovaries at Different Stages of Gestation
As a complement to the morphological assessment of meiotic stage of the oocyte in experiment 3, we used real-time PCR to measure the abundance of mRNA for the germ cell-specific protein YBX2, since it is not present in oocytes until the diplotene stage [21]. Consistent with the results of morphological analysis of meiotic stages of oocytes in primordial follicles (Fig. 5), measurement of levels of mRNA for YBX2 protein in cortical pieces from fetal ovaries at four different stages of gestation showed that steady-state levels of YBX2 mRNA increased with the progression of gestation. Levels of mRNA for YBX2 were 2.3-fold greater in ovarian cortex obtained from fetuses after Day 140 of gestation compared with fetuses before Day 90 or between Days 91 and 140 (P < 0.05; Fig. 6), consistent with the dramatic shift in oocytes from the prediplotene to the diplotene stage of meiosis I after Day 140 (Fig. 5).
|
Experiment 5: Effects of Length of Culture on Follicle Activation in Bovine Cortical Pieces In Vitro
The finding that activation of primordial follicles in fetal bovine ovarian cortex did not occur within 2 days in vitro unless the fetus was older than 140 days of gestation (experiment 2) is interesting, because it suggests that when primordial follicles are first formed, they are not capable of activating. To determine whether primordial follicles can acquire the capacity to activate in vitro, we examined the effect of length of culture on follicle activation by culturing cortical pieces from fetuses obtained before Day 141 of gestation for 2 or 10 days. After 2 or 10 days in culture, no follicles were observed in cortical pieces from fetuses younger than Day 91 of gestation (data not shown). Interestingly, when cortical pieces from 91- to 140-day-old fetal ovaries were cultured for 10 days, there were dramatic 94- and 3.9-fold increases in the number of primary follicles compared with Day 0 and Day 2, respectively (P < 0.05), without any change in the number of primordial follicles (P > 0.05; Fig. 7A). This suggests that some primordial follicles acquired the capacity to activate during the longer (10-day) culture period, and also that new primordial follicles were formed during culture. Compared with Day 0, the percentage of follicles that were activated had increased after 10 days in culture (P < 0.01; Fig. 7B).
|
Experiment 6: Effects of Steroids on Follicle Activation In Vitro in Ovarian Cortical Pieces from 91- to 140-Day-Old Fetal Calves
Shemesh et al. [26, 27] reported that bovine fetal ovaries transiently produce estradiol during the period of gestation before and during follicle formation. This led us to hypothesize that the steroid milieu in vivo may affect the capacity of follicles to activate. To determine the effects of steroids on the activation of newly formed primordial follicles, cortical pieces from fetuses obtained between Days 91 and 140 of gestation were cultured with estradiol, progesterone, or DHT (all at 10–6 M) for 10 days. After 10 days in culture, there were no differences in the numbers of primordial follicles in cortical pieces cultured without or with estradiol, progesterone, or DHT (P > 0.05; Fig. 8A). However, estradiol and progesterone decreased the number of primary follicles by around 90% and 85%, respectively, compared with DHT or control cultures (P < 0.05; Fig. 8A). Similarly, the percentage of follicles that were activated was also lower in cortical pieces cultured with estradiol or progesterone (P < 0.05; Fig. 8B). It is interesting that after 10 days of culture there were significantly fewer total follicles in cultures with estradiol compared with control medium or DHT (P < 0.05), whereas in the presence of progesterone, follicle numbers were intermediate (i.e., not different from any of the other treatments, P > 0.05).
|
Experiment 7: Secretion of Steroid Hormones by Bovine Fetal Ovaries In Vitro
To determine the effect of stage of gestation on the production of estradiol, progesterone, testosterone, and androstenedione by fetal ovaries, we cultured ovarian pieces obtained from 83- to 255-day-old fetuses for 24 h. The accumulation of estradiol in the medium in organ cultures of pieces of bovine fetal ovaries was highest around Days 80–100 of gestation, after which the levels declined gradually with age to levels almost undetectable between Day 141 and Day 193 (Fig. 9A). After Day 210, secretion of estradiol in vitro increased again. Estradiol secreted by ovarian pieces from fetuses younger than 140 days of gestation was significantly higher than that by ovarian pieces from fetuses 141 to 210 days old (1.08 ± 0.27 vs. 0.19 ± 0.13 ng/ovary per 24 h; P < 0.05, Fig. 9A). In addition, nonlinear regression analysis and Pearson correlation revealed a significant negative association between estradiol secretion and fetal age (P < 0.004 for both analyses).
|
Progesterone secreted by ovarian pieces from fetuses 83 to 165 days old appeared to decrease with age, and secretion increased to higher levels again after Day 190 of gestation (Fig. 9B). However, ANOVA, nonlinear regression analysis, and Pearson correlation testing showed no significant association between fetal age and progesterone secretion. The secretion of testosterone and androstenedione by the same fetal ovaries was very low (0–0.15 ng/ovary per 24 h and 0–0.27 ng/ovary per 24 h, respectively) and showed no consistent pattern over time (data not shown).
Cattle are an important model for studies of follicular development because of their economic importance and the similarities between follicular development in cattle and women. The current studies have resolved the controversy about the timing of follicular formation in cattle by showing that primordial, primary, and secondary follicles first appear around 90, 140, and 210 days of gestation, respectively. These results suggest that when follicles are first formed they cannot initiate growth in vivo, since there is a 50-day gap between the first appearance of primordial versus primary follicles. Likewise, newly formed primordial follicles in ovarian cortical pieces from Day 90–140 fetal ovaries did not have the capacity to activate within 2 days in vitro. However, the results show that they can develop this capacity during a longer culture period and that it can be suppressed by estradiol or progesterone, but not by DHT. Furthermore, the results show that the capacity to activate is correlated with the achievement of meiotic arrest in the oocyte and with an increase in steady-state levels of mRNA for a protein that is a marker for meiotic arrest, YBX2.
To study the mechanisms that regulate follicle formation and the initiation of follicular growth in fetal bovine ovaries, we first needed to know when these developmental transitions occur, but previous reports on the timing of follicular development in fetal bovine ovaries were surprisingly contradictory. Our results showing that the first primordial follicles are formed around 90 days of gestation, that the first primary follicles appear at around 140 days, and that the first secondary follicles develop around 210 days are in close accord with the earlier studies by Russe [3] and Dominguez et al. [8]. In contrast, they do not support a more recent report by Tanaka et al. [6] of earlier initiation of follicle formation at 74 days of gestation, nor Erickson's report of first follicle formation at the later time of 130 days of gestation [7]. Additionally, our results are consistent with the approximately 50-day gap between first appearance of primordial versus primary follicles that Russe [3] observed and not with the much shorter 17-day interval from follicle formation to activation reported by Tanaka et al. [6]. The reasons for these discrepancies among studies are not clear, but they do not appear to be due to differences between breeds.
The almost 2-mo delay that we and Russe [3] observed between the first appearance of primordial versus primary follicles in fetal bovine ovaries is quite different from the timing of these events in rats and mice, where a subset of primordial follicles activates soon after follicle formation (just after birth), and early secondary and early antral follicles are present by Postnatal Days 7 and 21, respectively [1, 2, 28]. However, the delay in cattle is consistent with the 25- and 40-day gaps between follicle formation and activation reported for sheep [4] and humans [5], respectively, species in which follicle formation also occurs during fetal life. Newly formed primordial follicles may remain in the resting stage because they are not yet capable of activating, because an activation signal is missing, and/or because activation is inhibited by a circulating or local factor until a certain stage of gestation.
To begin to distinguish among these possibilities, we next examined the relationship between gestational age and the ability of primordial follicles in cortical pieces from bovine ovaries to activate in vitro (i.e., isolated from the normal milieu in vivo). The finding that very few primordial follicles had activated after 2 days in culture when cortical pieces were obtained at Days 91–140 of gestation, whereas when ovaries were older than 140 days, most follicles activated within 2 days in vitro, suggests that follicles are not capable of activating until after Day 140 of gestation. A suggestion in the earlier literature [15] that bovine oocytes are not in meiotic arrest when primordial follicles first form led us to test the hypothesis that newly formed follicles do not activate because their oocytes have not reached meiotic arrest and are still in early stages of prophase. The dramatic shift between Days 91–140 and after Day 140 from predominantly prediplotene oocytes to mostly diplotene oocytes, coincident with an increase in the abundance of mRNA for YBX2 protein, a marker for diplotene oocytes [21], strongly supports the idea that few activated follicles are observed in vivo or within 2 days in vitro before Day 140 of gestation, because prophase I is still in progress in most oocytes. YBX2, a member of the Y box multigene family of proteins, is found only in germ cells, and in females it is specific to diplotene and mature oocytes [21]. Authors of early studies of follicular formation and development in fetal bovine ovaries reported that bovine oocytes "rest" at the pachytene stage of prophase I [7, 29]. Baker and Franchi [15] reexamined this question, albeit with a small number of animals, and observed that oocytes in primordial follicles of 3- to 4-mo-old fetuses were at the pachytene stage, whereas oocytes in late fetal and early postnatal calves were at the diplotene stage. Our observations, based on a larger number of ovaries spanning the last two trimesters of pregnancy, confirm and extend those of Baker and Franchi [15]. The correlation between the sharp increase in the number of oocytes observed in diplotene and the increase in ovarian mRNA for the marker protein YBX2 adds credence to the assessments of meiotic stage based on morphology of the chromosomes.
Since primordial follicles did not usually activate in vivo or within 2 days in vitro when the fetuses were younger than 141 days, we tested the hypothesis that they could acquire the capacity to activate during a longer culture period. The increase in primary follicles during a longer, 10-day culture supported that hypothesis, but it raised the question of why these follicles would not have activated had they remained in vivo during that 10-day period. Apparently, something about the in vitro situation enabled them to gain the capacity to activate between Days 2 and 10 of culture. One possibility is the absence in vitro of an inhibitor(s) that is present in vivo. Since steroid hormones are known to exert inhibitory effects on follicle formation and activation in rodents [18, 19, 30], and steroids are produced by fetal ruminant ovaries [6, 26, 27, 31], we next tested the hypothesis that one or more steroid hormones can inhibit activation in vitro. The finding that addition to the culture medium of estradiol or progesterone, but not the nonaromatizable androgen DHT, suppressed the follicle activation that was observed in control cultures suggests that a decrease in intraovarian estradiol and/or progesterone could be the mechanism that allows follicle activation to commence around Day 140 in vivo. Similarly, estradiol and/or progesterone inhibited the breakdown of oocyte nests and follicle assembly in ovaries from newborn mice and rats [18, 19], and estradiol and progesterone inhibited the initial wave of follicle activation in rat ovaries in vitro [18]. In contrast, low doses of estradiol increased follicle formation and activation in ovaries of embryonic and neonatal hamsters [32]. Consistent with the effects of estradiol on follicle activation in the current studies, mRNA for estradiol receptors was detected in bovine ovaries as early as Day 45 of pregnancy [33, 34]. The mechanism(s) of steroid inhibition of activation and the potential role of steroid receptors remain to be explored in future experiments.
As an initial test of the hypothesis that progesterone and estradiol suppress follicle activation in vivo until around Day 140 of gestation, we measured the accumulation of progesterone, androgens, and estradiol in cultures of ovarian pieces from whole ovaries of various ages (Days 83–255 of gestation). Those results pointed to estradiol as the (or a) potential inhibitor of follicle activation in vivo, since the secretion of estradiol by fetal ovaries was high around Day 90, when the first primordial follicles appear, and decreased precipitously between Days 90 and 140, when the first primary follicles are observed. There was a significant negative correlation between estradiol and gestational age. Shemesh et al. [26] examined estradiol secretion by bovine ovaries from about 35 to 85 days of age and found that the negligible secretion at 35–45 days of gestation increased dramatically from about 45 to 65 days and then decreased again, which is consistent with a later report by Dominguez et al. [8]. When Shemesh [27] treated ovaries with testosterone in vitro, estradiol secretion was three times higher when ovaries were obtained at about 50–68 days of age versus 75–105 days. In the study by Tanaka et al. [6], ovarian content of estradiol was very low around the time that bovine primordial follicles appeared, but the number of ovaries examined at that time was very small. Quirke et al. [31] reported 3- to 4-fold increases in ovarian progesterone and estradiol from Days 55 to 75 of ovine pregnancy, when the first primordial follicles appear, but they did not measure the steroids after Day 75. These reports of estradiol production by, or presence in, ruminant ovaries around the time of follicle formation are consistent with the detection of aromatase and its mRNA in fetal ovine and bovine ovaries [31, 33, 34].
In the current studies, progesterone secretion by cultured ovarian pieces also decreased between Days 83 and 140 of gestation, during follicle formation and before the appearance of primary follicles, but the decline was not as consistent or as dramatic as the decrease in estradiol, and no significant correlation between progesterone and gestational age was detected. However, whether progesterone directly inhibits follicle activation in vitro or exerts its effects by acting as a precursor for estradiol synthesis is not known at this point and should be resolved in future experiments. If the latter, then changes in the capacity of fetal ovaries to aromatize androgens to estradiol may be the critical factor, rather than the level of progesterone. Secretion of both testosterone and androstenedione was uniformly very low, but whether this is due to rapid conversion of androgens to estradiol and/or to changes in the production of androgens is a question for further study. Later in gestation, small antral follicles begin to develop, and these are likely responsible for the elevated secretion of estradiol and progesterone by cultured ovarian pieces obtained during the last trimester of pregnancy.
In summary, bovine primordial follicles do not activate to become primary follicles in vivo or in vitro (2-day culture) until after Day 140 of gestation. In vivo, the appearance of the first primary follicles is correlated with the achievement of the resting, diplotene stage of meiotic prophase I in primordial follicles and with an increase in ovarian concentrations of mRNA for YBX2, a protein associated with the diplotene stage. The capacity to activate can be acquired during a longer culture period, and that activation can be inhibited by estradiol or progesterone, but not DHT. The temporal pattern of decline in the capacity of bovine fetal ovaries to secrete estradiol suggests that this steroid is involved in regulating follicle activation, and perhaps follicle formation, in bovine fetal ovaries. Understanding the mechanisms that regulate the formation of the pool of primordial follicles and their transition to the growing phase is essential for the manipulation of this pool for practical purposes.
ACKNOWLEDGMENTS
The authors thank Cargill Regional Beef for the donation of bovine ovaries. The cooperation of Mr. John Couture at Cargill is gratefully acknowledged. We thank Dr. K. Hinrichs for training us to distinguish diplotene from prediplotene oocytes, Dr. J. Folger for help with real-time PCR analysis, and Dr. S.M. Quirk for helpful comments on the manuscript. We are grateful to Drs. D.T. Armstrong and T.G. Kennedy for the androstenedione antiserum, Dr. G.D. Niswender for the estradiol antiserum, Dr. W. Hansel for the progesterone antiserum, and Dr. V. Gay for the testosterone antiserum.
FOOTNOTES
1Supported by National Research Initiative Competitive Grant 2003-35203-13532 from the U.S. Department of Agriculture Cooperative State Research, Education, and Extension Service. 
Correspondence: 2FAX: 607 253 3476; e-mail: JF11{at}cornell.edu
Received: 26 November 2007.
First decision: 17 December 2007.
Accepted: 22 February 2008.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  F. Ghafari, S. Pelengaris, E. Walters, and G.M. Hartshorne Influence of p53 and genetic background on prenatal oogenesis and oocyte attrition in mice Hum. Reprod., June 1, 2009; 24(6): 1460 - 1472. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Fowler, S. Flannigan, A. Mathers, K. Gillanders, R. G. Lea, M. J. Wood, A. Maheshwari, S. Bhattacharya, E. S. R. Collie-Duguid, P. J. Baker, et al. Gene Expression Analysis of Human Fetal Ovarian Primordial Follicle Formation J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1427 - 1435. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-dihydrotestosterone (DHT), and estradiol (E2; all at 10–6 M) on follicle activation in vitro in ovarian cortical pieces from 91- to 140-day-old fetal calves cultured for 10 days. A) Number of primordial and primary follicles in fetal bovine ovarian cortex after 0 or 10 days in culture. B) Percentage of activated (primary) follicles. Means with different letters (a, b) differ significantly (P < 0.05; n = 4 fetuses).