| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
` Department of Anatomy, Physiological Sciences and Radiology, North Carolina State University, Raleigh, North Carolina 27606
ABSTRACT
The hypothesis was that long-term treatment of cattle with estradiol (E2) and bovine somatotropin (bST) would alter the earliest stages of folliculogenesis. Nonlactating Holstein cows (n = 26) were treated in a 2 x 2 arrangement with E2 (2 x 24 mg implants, 67.1 ± 1.4 days) and bST (Posilac, 63.6 ± 1.5 days). At Day 67 ± 1.3, one ovary was removed for morphometric and immunohistochemical analysis. For each ovary, 388 ± 38 microscopic fields (2 x 2 mm) were examined and follicles within each field were classified by histological stage. Fields that contained no follicles were classified as empty. Empty fields (n = 100 per ovary) were further classified as containing no evidence of follicles or containing atretic remnants of follicles. Approximately 30 4-µm sections per ovary were stained for proliferating cell nuclear antigen (PCNA), and 150 fields per ovary were evaluated. Additional sections (n = 10 per ovary) were assessed immunohistochemically for apoptosis, and fluorescence intensity was determined for each follicle. Treatment with bST significantly decreased percentage of empty fields containing atretic remnants. Treatment with E2 induced activation of follicles as shown by a decrease in percentage of primordial follicles and an increase in percentage of primary follicles as determined by PCNA staining. At the primary follicle stage the combination of bST + E2 decreased apoptosis as shown by decreased fluorescence intensity. Thus, E2 induced activation of follicles, bST enhanced survival, and the combination lowered atresia.
apoptosis, estradiol, follicle, growth hormone, ovary
INTRODUCTION
The neonatal bovine ovary contains a large pool of primordial follicles; however, less than 0.1% of these grow to maturity and ovulate during the reproductive lifetime of a cow [1]. Most of the follicles that enter the growing pool are destined to undergo atresia. Superovulation with gonadotropins allows more of these growing follicles to ovulate; however, there is a large amount of variation in response to superovulatory treatments.
Erickson [1] demonstrated that the number of follicles per cow varied from 0 to more than 700 000, but that within a cow, the follicle population of one ovary was a good predictor of the follicle population in the other ovary. Recently, we demonstrated that the superovulatory response of one ovary to FSH is related positively to the number of primordial and tertiary microscopic follicles and the number of medium (4–7-mm) surface follicles present on the contralateral ovary immediately before superovulation [2]. We concluded that if hormonal treatments could be used to increase rates of activation and survival of follicles between the primordial and antral stage, then a cow that would normally be a low responder because of a low number of growing follicles might have a greater superovulatory response.
Although most attempts at enhancing superovulatory response have focused on altering the number of emerging follicles about 1 wk before superovulation, it has been demonstrated that growth of the bovine follicle from the earliest antral stage to the ovulatory stage requires about 42 days [3]. Extrapolating backward from these data, the growth of a follicle from activation into the growing pool to ovulation would require between 80 and 100 days [4]. Therefore, any long-term attempt to alter ovulatory response would require treatment for 60–80 days.
The objective of the present study was to examine the effects of long-term treatment with estradiol (E2) and bovine somatotropin (bST) on folliculogenesis in cattle. We chose to evaluate E2 because it has been demonstrated to bind to granulosa cells in rats [5], and the mRNA and protein for the E2 receptor has been localized in granulosa cells from all follicular stages in sheep [6]. Estradiol in combination with insulin increased the number of primary follicles (50–100 µm; two to five granulosa cell layers) in bovine cortical slices after 48 h in culture compared to treatment with insulin or E2 alone [7]. Furthermore, isolated bovine preantral follicles cultured with E2 for 7 days grew to a larger diameter than did control follicles [8]. Finally, Price and Webb [9] implanted cattle with estradiol for 23 days and observed a shift in the population of follicles; there were more follicles <6 mm and fewer follicles >6 mm in E2-treated cows, even though there were no differences in serum gonadotropin concentrations.
We chose to evaluate bST because it stimulates production of insulin-like growth factor I (IGF-I), a potent mitogen that stimulates granulosa cell growth in the cow [10]. Recently, it has also been demonstrated that bST prevents apoptosis in preovulatory rat follicles through stimulation of IGF-I [11]. Finally, treatment with bST for 42 days increased the number of small (2–5-mm) antral follicles present in the bovine ovary [12].
Our hypothesis was that long-term treatment with E2 and bST would alter folliculogenesis. If this hypothesis was true, then it should result in a larger pool of FSH-responsive follicles present on the ovary when superovulation was initiated.
MATERIALS AND METHODS
Animals and Tissue Preparation
Nonlactating Holstein and Jersey cows (n = 26) 3–5 yr of age were maintained on a mixed pasture of fescue, white clover, and Bermuda grass between September and December. These cows were healthy, had produced 1–3 calves each, and had lactated normally before being culled from the dairy herd for reasons unrelated to fertility. Animals were treated in a 2 x 2 arrangement with E2 (two Compudose ear implants containing 24 mg E2 each; Elanco Products Co., Indianapolis, IN) and bST (500 mg, Posilac sq; Monsanto Corp, St. Louis, MO). Treatments consisted of: 1) Control—saline every 2 wk and sham implants (n = 6); 2) bST—bST every 2 wk and sham implants (n = 6); 3) E2—saline every 2 wk and 2 Compudose ear implants (n = 7); and 4) bST + E2—bST every 2 wk and 2 Compudose ear implants (n = 7). Interval from insertion of E2 implants to their removal averaged 67 ± 1.4 days and interval from first injection of the slow-release bST until 14 days after the last injection (end of normal payout period) averaged 63.6 ± 1.5 days. Serum hormone (E2, GH, and IGF-I) profiles for the treatment periods have been published [13]. The ovary contralateral to the corpus luteum was removed by colpotomy [14] on Day 8 following an observed estrus that occurred within 14 days after an injection of bST. The E2 implants were removed at the time of ovariectomy. All experimental procedures were approved by the North Carolina State University Institutional Animal Care and Use Committee.
Morphometrics
Immediately after ovariectomy, surface follicles of the excised ovary were counted and classified as small (1–3 mm), medium (3–7 mm), or large (>7 mm). Fluid was aspirated from follicles >3 mm in diameter. The ovary was divided into quarters and two quarters were fixed in Bouin's solution. The tissue was dehydrated using a graded series of ethanol, cleared with Clearite (Richard-Allan Medical, Richland, MI), and embedded in paraffin (Paraplast Plus; Baxter). Thirty to fifty consecutive sections (6 µm) from one quarter were mounted onto glass slides and stained with a periodic acid-Schiff reaction and hematoxylin counterstain.
Histological sections were examined as described [2] using a superimposed counting grid. Briefly, a 1 cm x 1 cm grid divided into 25 2 mm x 2 mm counting fields was printed onto acetate film (3M Corporation, Austin, TX). Grids were glued (Krazy Glue, Borden, Columbus, OH) onto the underside of each slide below each section. For each section, every completely filled field that contained cortical tissue was counted. Follicles were classified as: 0) primordial—an oocyte surrounded by a single layer of flattened pregranulosa cells, 1) primary—an oocyte surrounded by a single layer of one or more cuboidal granulosa cells, 2) secondary—an oocyte surrounded by two or more layers of cuboidal granulosa cells, and 3) tertiary—an oocyte surrounded by two or more layers of granulosa cells but no larger than 1 mm in diameter with a distinct antrum. To avoid duplicate counting, primordial and primary follicles were counted on the first section in which the nucleus of the oocyte appeared. Secondary and tertiary follicles were easier to track and were counted in the first field in which they were encountered. If a 2 mm x 2 mm field contained no follicles it was classified as empty.
Activation
The initial morphometric examination indicated that estradiol stimulated activation of primordial follicles into the growing pool based on an increase in size of the granulosa cells in the single layer surrounding the oocyte. Nevertheless, we were concerned that this might be an artifact, because Hulshof et al. [8] had shown that E2 increased the size of granulosa cells of isolated preantral follicles without increasing mitotic activity. Therefore, immunohistochemical detection of proliferating cell nuclear antigen (PCNA) was performed to determine if E2 was truly increasing the rate of activation of primordial follicles. PCNA is expressed when granulosa cells divide and its expression has been observed as follicles move from the primordial to primary stage [15, 16]. Thirty additional paraffin-embedded sections (4 µm) were cut from each ovary and used for detection of PCNA using a PCNA detection kit (Zymed, San Francisco, CA) according to the manufacturer's instructions. Briefly, sections were deparaffinized, rehydrated in ethanol, and quenched in 3% hydrogen peroxide in methanol to block endogenous peroxidase activity. The sections were placed in boiling 0.01 M citrate for 10 min to improve antigen recovery. Sections were incubated for 45 min at room temperature with a biotinylated mouse anti-PCNA. Streptavidin-horseradish peroxidase was added, and sections were incubated for 10 min at room temperature. Diaminobenzadine was added for 3.5 min at room temperature. Positive controls were sections of mouse small intestine, and negative controls were generated by excluding the antibody. For each ovary, 150 fields were counted and follicles were classified as primordial (no PCNA-positive cells) or primary (one or more PCNA-positive cells).
Atresia
During the initial morphometric examination, numerous follicular remnants were observed but were not counted because an oocyte nucleus was not present in any cross section. To assess potential effects of treatments on atresia we re-examined 100 empty fields for each cow and classified each field as either an empty field that contained no remnants or a field that contained atretic follicular remnants. Atretic remnants had degenerate basement membranes, pyknotic nuclei, and did not contain an oocyte.
To investigate atresia further, we quantified apoptosis in 10 additional sections (6 µm) per ovary using the ApopTag Plus kit (Oncor, Gaithersburg, MD) according to the manufacturer's instructions. Briefly, the 3' hydroxy ends of DNA fragments were labeled with UTP-11-digoxygenin using terminal deoxytransferase (TdT) and visualized by fluorescein-labeled antibody. Involuting mouse mammary tissue was used as a positive control, and negative controls were generated by excluding the TdT. Follicles were examined under darkfield microscopy using an Olympus BH-2 microscope equipped with a reflected-light fluorescence attachment (494 nm). For each primary and secondary follicle, a computer image of the entire follicle section was acquired using the Optimas Image Analysis software package (Bioscan Inc., Edmonds, WA), and the optical density for each follicle was determined as described by Singh et al. [17]. For each tertiary follicle, four images of the mural granulosa were acquired at 90° intervals around the circumference. A random sample (5 x 5 pixels) was taken for each image, and fluorescence intensity was analyzed using the Optimas Image Analysis software package. For tertiary follicles, an average fluorescence intensity was calculated from the four measurements.
Statistical Analysis
The proportion of follicles within various stages of development was analyzed using the CATMOD procedure of Statistical Analysis Systems [18] with stage of follicle as the dependent variable and E2, bST, and their interaction as the independent variables. The percentage of empty fields and the percentage of fields containing remnants was also analyzed by CATMOD. The percentage of follicles positively stained for PCNA and average fluorescence intensity were analyzed by least-squares ANOVA, and the residual error term was used to test the main effects of E2 and bST as well as their interaction.
RESULTS
Morphometrics
Follicles of different morphological classes are illustrated in Figure 1A. There was no difference in the number of fields per ovary (388 ± 38) counted among the four treatment groups (P > 0.1). Altogether, 6,693 follicles were counted and classified among all fields of all ovaries (Table 1). Proportion of fields that were classified as empty (control, 59%; E2, 65%; bST, 60%; and E2 + bST, 62%) was influenced by the interaction of E2 and bST (P < 0.05), because E2 increased the percentage of empty fields, but bST administered with E2 prevented the E2-stimulated increase in empty fields.
|
|
Distribution of follicles among the different stages was influenced by E2 and bST (Table 1). When the percentage of primordial follicles was compared, the interaction (P < 0.05; Table 1) between E2 and bST was associated with a lower proportion of primordial follicles in E2-treated cows compared to cows that received bST + E2. The inverse was observed in the interaction (P < 0.05, Table 1) between E2 and bST for percentage of primary follicles, where E2 was associated with a greater proportion of primary follicles compared to cows that received bST + E2. There was a strong tendency (P < 0.1) for E2 to decrease percentage of follicles at the secondary stage (Table 1). There was also a strong tendency (P < 0.1) for bST to increase the percentage of follicles at the tertiary stage (Table 1).
Activation
A follicle that reacted positively for PCNA is illustrated in Figure 1B. Based on classification using PCNA staining, the percentage of follicles in the primordial class was decreased by long-term treatment with E2 (P < 0.05; Table 2). Percentage of follicles in the primary stage was increased in cows treated with E2 (P < 0.05), but bST tended to inhibit this E2-stimulated increase (bST + E2, P < 0.1).
|
Atresia
When fields classified as empty by morphological criteria were reanalyzed for remnants of follicles, bST significantly increased the percentage of truly empty fields (bST, 87% versus no bST, 81%, P < 0.05) and decreased the frequency of fields with atretic remnants (bST, 13% versus no bST, 19%, P < 0.05). Follicles assessed for degree of apoptosis using the ApopTag kit are illustrated in Figure 1, C and D. Treatment with bST + E2, compared to either hormone alone, significantly decreased degree of apoptosis as measured by the mean fluorescence intensity in primary follicles (interaction, P < 0.05; Fig. 2). There was no effect of E2, bST, or their interaction on mean fluorescence intensity of secondary or tertiary follicles.
|
DISCUSSION
The results of this study demonstrate that long-term treatment of cattle with E2 and bST altered distribution of preantral follicles among primordial, primary, secondary, and tertiary classes. Estradiol stimulated activation of primordial follicles, and E2 in combination with bST decreased the rate of atresia in primary follicles. The combined treatment with E2 and bST nearly doubled the number of ovulations in response to superovulation with bST when the cows used in the current study were superovulated after unilateral ovariectomy [13]. Thus, treatments that activate and sustain development of preantral follicles may lead to greater responses to superovulation when those follicles reach the Graafian follicle stage.
Earlier morphometric studies suggested that E2 may stimulate activation of follicles in cattle, but an important question that remained was whether E2 did this through induction of hypertrophy [8] or hyperplasia [7] of granulosa cells. When we stained sections for PCNA, the results showed clearly that E2 induced hyperplasia, causing a reduction in primordial follicles and an increase in primary follicles.
The exact mechanisms by which this shift from primordial to primary follicles is occurring remain to be elucidated. It may be a direct effect of E2 on the granulosa cells, because Stumpf [5] demonstrated that tritiated E2 bound to granulosa cells in rat ovaries, and the protein and mRNA for the E2 receptor has been localized in ovine granulosa cells at all stages of development [6]. The fact that Hulshof et al. [8] observed an increase in size but no increase in proliferation of granulosa cells when isolated bovine preantral follicles were cultured with E2 for 7 days might be explained by direct effects of E2 on the granulosa cell and indirect effects of E2 mediated through the stroma. In the study by Peluso and Hirschel [7], cortical slices instead of isolated follicles were cultured and there was an increase in tritiated thymidine incorporation into the cortical slices. Thus, granulosa cells may require E2-induced mitogenic factors from ovarian stroma to emerge from the arrested state. This concept is supported by the observation of Revelli et al. [19], who localized the protein and mRNA for the E2 receptor in fibroblast-like cells surrounding follicles in the human ovary. Possibly E2 works through the stromal tissue to stimulate production of local growth factors that in turn stimulate activation of the primordial follicle.
Alternatively, in a recent study, Durlinger et al. [20] observed a rapid decrease in the size of the primordial pool in mice deficient in antimullerian hormone (AMH). This led them to hypothesize that AMH may be an inhibitor of activation. In the rat, AMH mRNA expression is decreased in some preantral follicles during estrus [21]. Therefore, it is possible that long-term treatment of cattle with estradiol resulted in a down-regulation of the ovarian inhibitor of primordial follicle activation, AMH, resulting in an increase in the percentage of activated follicles. Further investigation will be necessary to determine the effects of long-term treatment with estradiol on ovarian AMH expression in cattle.
Long-term treatment with bST decreased the occurrence of atretic remnants, defined as follicular structures with degenerate basement membranes, pyknotic nuclei, and no oocyte. It was not possible to determine the stage that these follicles had reached before undergoing atresia. The basement membrane was degraded and the nucleus of the oocyte had disappeared, but based on the diameter, these remnants appeared to be from preantral and early antral follicles. This raises a question of how bST might be affecting atresia in preantral follicles. Treatment with bST has been demonstrated to increase serum growth hormone, IGF, and insulin concentrations [12, 13, 22, 23], and peripheral growth hormone and IGF concentrations were increased in our cows [13]. One or more of these hormones may be influencing preantral follicle growth.
Lucy et al. [24] could not detect GH receptors in granulosa cells of bovine antral follicles using immunohistochemistry or Northern blot analysis; however, recent in situ studies have localized mRNA for the GH receptor in the oocyte and granulosa cells of preantral follicles of sheep [25] and cattle [26]. Therefore, it remains possible that some of the effects observed in the present study were due to a direct effect of bST on the preantral follicles.
It is generally accepted that the effect of bST on antral follicle development is mediated through increases in serum IGF and/or insulin concentrations. Gong et al. [23] injected groups of cows with different doses of bST and noted that all doses of bST increased serum growth hormone concentrations, but only doses that increased IGF and insulin concentrations increased the number of small antral follicles (<5 mm). This would argue that increases in systemic IGF and insulin may be involved in the increase in small antral follicles. Further evidence for a systemic rather than local effect of IGF was provided by Kirby et al. [27], who treated cattle with bST daily for 16 days and observed that serum IGF concentrations increased but ovarian IGF expression was not altered. Taken together, these studies indicate that systemic IGF may have a greater role in maintaining the growth of antral follicles than locally produced IGF.
Additional evidence regarding how bST decreases atresia in preovulatory follicles comes from studies in the mouse. Bovine growth hormone decreased the rate of apoptosis in isolated preovulatory mouse follicles, apparently through an IGF-mediated action [11]. Treatment of follicles with bST increased IGF mRNA, as determined by Northern blot, but when follicles were cotreated with bST and IGF binding protein-3 (to bind IGF), the bST-induced decrease in apoptosis was ablated. These results contrast with those of Kirby et al. [27] who did not detect changes in IGF mRNA in ovaries of bST-treated cows, but in their study, whole ovarian homogenates were used, and this may have reduced sensitivity in detecting changes in individual follicles.
Wandji et al. [28] demonstrated that there is some binding of IGF to preantral follicles and that IGF binding increases at the antral stage, but they did not differentiate between binding by receptors or binding proteins. If there are specific IGF receptors, preantral follicles may only need a minimal amount of IGF stimulation to maintain growth in the preantral stage.
Histological examination of ovaries of mice that are null mutants for the IGF-I gene revealed the presence of primordial, primary, and early antral follicles, but only rarely preovulatory Graafian follicles [29]. The ovaries of these mice had reduced FSH receptor mRNA content, and replacement treatment with IGF increased ovarian FSH receptor mRNA content to equivalence with wild-type mice [30]. These results led the authors to conclude that IGF is not necessary for preantral follicle development but is required for growth of antral follicles to ovulatory size; however, there remains the possibility that insulin is working through the IGF receptor to stimulate growth of preantral follicles in these IGF-deficient mice. For example, in feed-restricted Brangus heifers that would be expected to have reduced concentrations of IGF, treatment with insulin increased ovulation rate [31], and the authors proposed that in these heifers insulin might be working through the IGF receptor to stimulate follicular growth. Therefore, follicles of IGF-I knockout mice may be growing to later stages by using insulin to compensate for a lack of IGF.
Apoptotic cell death of granulosa cells is a hallmark of atresia, and evidence for a role of bST in decreasing atresia in preantral follicles comes from our observation that E2 and bST in combination decreased the relative amount of apoptosis in primary follicles. While there were no alterations in the numbers of antral follicles in these cows following the treatment period, the combination of long-term bST and E2 did improve the superovulatory response to FSH [13]. We are aware of only one other account of long-term effects of bST treatment on follicle growth, Kirby et al. [22] observed that bST treatment increased the number of 6- to 9-mm follicles for up to 3 wk after treatment with bST was discontinued. This would indicate that bST was promoting continued growth of follicles at the early antral stages, leading to more antral follicles present in the ovary 21 days later. Furthermore, it was recently reported that when apoptosis was suppressed in mouse preantral follicles by treatment with 8-bromo-cGMP in vitro, the follicles grew more rapidly in response to FSH [32]. Therefore, bST + E2 may be increasing growth rates of preantral follicles by decreasing the relative amount of apoptosis in the primary stage, and resulting in follicles that have more granulosa cells and grow faster.
In the present study, long-term treatment with bST in combination with E2 reduced the amount of follicle activation observed in cows treated with E2 alone. One possible explanation for this is that bST maintained the growing pool of follicles, and the growing pool suppressed activation of follicles through local feedback. A large percentage of the primordial follicles activate spontaneously in bovine cortical slices in vitro [16, 33]; however, it is likely that any inhibitory influence of the growing pool of follicles is removed because cortical slices containing mostly primordial follicles and a few primary follicles are cultured. Because these cortical slices contain few growing follicles, this may allow spontaneous activation of the primordial follicles. Furthermore, it has been demonstrated that in mice secondary follicles can inhibit the growth of primary follicles [34], providing support for the concept that follicles at later stages of development can influence the growth of follicles at the earlier stages of development.
In conclusion, long-term treatment of cows with E2 increased activation of primordial follicles and bST + E2 decreased apoptosis at the primary stage. We believe that this allowed a greater number of follicles to grow through the secondary stage to the antral stage. Further studies will be needed to elucidate the mechanisms through which E2 stimulates activation and bST + E2 decrease apoptosis in primary follicles.
ACKNOWLEDGMENTS
We thank Karl Hedrick for his assistance managing the animals, Douglas Shaw for his assistance with surgeries, Pamela Schoppee and Karen Swanchera for their assistance in processing tissue samples, Angela Rogers and Patience Davis for their assistance with histological processing, and Charlotte Farin for assistance with fluoresence microscopy.
FOOTNOTES
First decision: 13 March 2001.
1 Correspondence: Jack Britt, University of Tennessee, Institute of Agriculture,
P.O. Box 1071, Knoxville, TN 37901-1071; FAX: 865-974-8781; jack-britt{at}utk.edu 
2 Current address: VRT6-0016B, Biomedical Sciences, Cornell University,
Ithaca, NY 14853.
3 Current address: Departamento de Zootecnia, Universidade Federal de
Lavras, C P 37, 37200-000 Lavras MG, Brazil.
Accepted: April 4, 2001.
Received: February 9, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. Wang, E. R. Prossnitz, and S. K. Roy G Protein-Coupled Receptor 30 Expression Is Required for Estrogen Stimulation of Primordial Follicle Formation in the Hamster Ovary Endocrinology, September 1, 2008; 149(9): 4452 - 4461. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Wang and S. K. Roy Development of Primordial Follicles in the Hamster: Role of Estradiol-17{beta} Endocrinology, April 1, 2007; 148(4): 1707 - 1716. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A Slot, J. Kastelijn, A. Bachelot, P. A Kelly, N. Binart, and K. J Teerds Reduced recruitment and survival of primordial and growing follicles in GH receptor-deficient mice. Reproduction, March 1, 2006; 131(3): 525 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Spicer Effects of Estradiol on Bovine Thecal Cell Function In Vitro: Dependence on Insulin and Gonadotropins J Dairy Sci, July 1, 2005; 88(7): 2412 - 2421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Kezele and M. K. Skinner Regulation of Ovarian Primordial Follicle Assembly and Development by Estrogen and Progesterone: Endocrine Model of Follicle Assembly Endocrinology, August 1, 2003; 144(8): 3329 - 3337. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
