| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a School of Animal and Microbial Sciences, University of Reading, Whiteknights, Reading RG6 6AJ, United Kingdom
b School of Biological and Molecular Sciences, Oxford Brookes University, Oxford OX3 0BP, United Kingdom
ABSTRACT
The relationship between follicle growth and plasma inhibin A, FSH, LH, estradiol (E), and progesterone was investigated during the normal bovine estrous cycle and after treatment with steroid-free bovine follicular fluid (bFF) to arrest follicle development. In the first study, four heifers were monitored over three prostaglandin (PG)-synchronized cycles. Blood was collected every 2–8 h, and ovaries were examined daily by ultrasonography. Inhibin A was measured using a modified enzyme-linked immunosorbent assay that employed a new monoclonal antibody against the
subunit of bovine inhibin. Plasma inhibin A (
50 pg/ml before luteolysis) rose steadily during the induced follicular phase (P < 0.05) to a peak (125 pg/ml) coincident with the preovulatory E/LH/FSH surge. After ovulation, inhibin A fell sharply (P < 0.05) to a nadir (55 pg/ml) coincident with the secondary FSH rise. During the next 3 days, inhibin A increased to approximately 90 pg/ml in association with growth of the new dominant follicle (DF). Plasma E also rose twofold during this period, whereas FSH fell by approximately 50%. Inhibin A was negatively correlated with FSH (r = -0.37, P < 0.001) and positively correlated with E (r = 0.49, P < 0.0001). Observations on eight cycles (two cycles/heifer), in which growth of the ovulatory DF was monitored from emergence to ovulation, showed that the first-wave DF (DF1) ovulated in three cycles and the second-wave DF (DF2) in five cycles. After PG, plasma inhibin A and E increased similarly in both groups, with concomitant falls in FSH. In the former group, the restricted ability of DF1 to secrete both inhibin A and E was restored after luteolysis. Results indicate that dynamic changes in the secretion of both E and inhibin A from the DF contribute to the fall in FSH during the follicular phase and to the generation and termination of the secondary FSH surge, both of which play a key role in follicle selection. In the second study, bFF (two dose levels) was administered to heifers (n = 3–4) for 60 h starting from the time of DF1 emergence. Both doses suppressed FSH (P < 0.05) and blocked DF1 growth to the same extent (P < 0.01), although inhibin A levels were only marginally raised by the lower dose (not significant compared to controls). The high bFF dose raised (P < 0.001) inhibin A to supraphysiological levels (1 ng/ml). A large "rebound" rise in FSH occurred within 1 day of stopping both treatments, even though the inhibin A level in the high-dose bFF group was still approximately threefold higher than that in controls. This indicates that desensitization of gonadotropes to inhibin negative feedback is a contributory factor, together with reduced ovarian output of E, in generation of the post-bFF rebound in FSH.
follicle, FSH, inhibin, ovary, ovulatory cycle
INTRODUCTION
Follicle turnover continues throughout the bovine estrous cycle in a wave-like manner [1–3], with two or three waves developing during a normal 18- to 24-day cycle. The emergence of each wave is preceded by a transient increase in circulating FSH levels [4, 5], which fall again around the time of selection of the dominant follicle (DF) from its cohort. This decline in pituitary FSH secretion has been attributed, at least in part, to a negative feedback effect of inhibins produced by developing follicles. A role for inhibins in regulating FSH release in cattle is supported by evidence from passive and active immunization studies, in which immunoneutralization of inhibin has been associated with elevated circulating FSH [6–9] and increases in follicle development and ovulation [7–10]. Further support comes from studies in which inhibin was administered either in the crude form of steroid-free bovine follicular fluid (bFF) [11–16] or in a highly purified form [14] to intact or ovariectomized heifers. In these studies, plasma FSH levels were greatly suppressed during the period of treatment with steroid-free bFF, and in intact animals, the growth of follicles 5 mm or greater in diameter was inhibited [17, 18]. Following cessation of bFF treatment, a characteristic "rebound" hypersecretion of FSH occurs in intact animals, although the endocrine mechanism for this response requires clarification. Several studies have questioned the extent to which these effects of steroid-free bFF on FSH secretion and follicle development result from its content of inhibin [18, 19]
The endocrine role of inhibins cannot be fully evaluated without reliable means for determining circulating levels. The fully processed form of bioactive inhibin is a 32-kDa dimer consisting of dissimilar and ß subunits. Two isoforms of the ß subunit (ßA and ßB) are expressed in the bovine ovary [20], potentially giving rise to inhibin A and inhibin B, but to date, only inhibin A has been isolated from bFF [21]. In addition to 32-kDa inhibin, larger-molecular-weight inhibin forms have been isolated from bFF, and these have also been shown to have FSH-suppressing activity in vitro [22]. Initial attempts to measure circulating levels of inhibins by RIA relied on antibodies raised against the subunit [23, 24]. Because large quantities of nonbioactive, free subunit are produced and found in the peripheral circulation [25], these results likely are confounded by the inability of these RIAs to discriminate between the free subunit and dimeric inhibins. Subsequently, two-site immunoradiometric assays that specifically detect dimeric inhibins have been developed. These assays are useful in determining levels of dimeric inhibins in bFF [26, 27], but they lack the sensitivity to determine peripheral levels. More recently, two-site enzyme-linked immunosorbent assays (ELISAs) that use monoclonal antibodies (mAbs) specific for , ßA, and ßB subunits have been developed and used to measure the levels of circulating inhibin A and B in humans [28–30]. The ELISA for inhibin A was subsequently modified for use in sheep [31], but it still proved incapable of quantifying the lower levels present in the peripheral blood of normal (i.e., non-FSH treated) cattle.
With the aid of a new mAb raised against the bovine c1–32 peptide, we recently developed an ELISA with improved sensitivity for bovine inhibin A. The objectives of the present study were to validate this modified two-site ELISA for measuring inhibin A levels in bovine plasma and to use the assay to relate plasma inhibin A levels to ovarian follicle turnover and gonadotropin and steroid profiles during the normal bovine estrous cycle and in heifers with FSH secretion and follicle development perturbed by treatment using exogenous inhibin in the form of steroid-free bFF.
MATERIALS AND METHODS
Animals
Study 1: normal estrous cycle
Four British Friesian heifers (2 yr old, 400 kg live weight) were housed indoors and loosely tethered by a head collar and chain in individual stalls. They were fed a maintenance ration of concentrates and straw and had free access to water. Estrous cycles were synchronized by giving three i.m. injections (0.5 mg) of a prostaglandin (PG) F2 analogue, cloprostenol (Estrumate; Coopers Animal Health Ltd., Crewe, Cheshire, UK) at 14-day intervals. The day of the third PG injection was designated as Day 0. Two further PG injections were given 14 and 28 days later to induce luteolysis on Day 10–11 of the synchronized cycles under investigation. Blood samples (20 ml) were taken via an indwelling jugular vein catheter every 8 h for an 80-h period before PG-induced luteolysis (Time 0), every 2 h for the period 0–120 h after PG-induced luteolysis, and then at 8-h intervals for the remaining 9 days until the next PG injection. The same sampling regime of two hourly samples for 120 h after PG injection, followed by eight hourly samples for the next 9 days, was repeated during the next two cycles. Catheter patency was maintained by flushing with sterile 0.9% (w/v) saline containing sodium heparin (100 IU/ml) after withdrawal of each sample. Blood samples were collected into polystyrene tubes containing 50 µl of sodium heparin (5000 IU/ml), and after centrifugation (2000 x g for 30 min at 4°C), the plasma was separated and stored at -20°C until assay. Plasma LH and FSH concentrations were measured in all samples, inhibin A concentrations in samples taken at 8-h intervals (0800, 1600, and 2400 h), and steroid hormones in a single daily sample (0800 h). Follicle and luteal development was monitored daily from Day 0 by transrectal ultrasonography [32] using a scanner fitted with a 7.5-MHz, linear-array transducer (Concept 2000; Dynamic Imaging Ltd., Livingstone, Lothian, UK). Ultrasound observations were recorded on videotape for subsequent sequential analysis.
Study 2: response to bFF treatment
Ten Simmental cross Friesian heifers (3 yr old, 600 kg live weight) were housed as described above and fed a maintenance ration of grass silage. Cycles were synchronized by two injections of PG given 14 days apart. A progesterone-releasing intravaginal device (PRID; Sanofi Animal Health, Watford, Herts, UK) was inserted 14 days after the second PG injection. The device was removed 6 days later, 1 day after a third PG injection.
Heifers were assigned to one of three groups: "high-bFF" (4 ml of bFF every 4 h, n = 3), "low-bFF" (1.3 ml of bFF + 2.7 ml of bovine serum every 4 h, n = 4), or "control" (4 ml of bovine serum every 4 h, n = 3). Steroid-free bFF was prepared using the charcoal extraction method as described by Beard et al. [14]. Bovine serum for use as the control vehicle and for dilution of the low-bFF dose was charcoal extracted in the same manner. Charcoal treatment removed 99.8% of endogenous steroids from bFF (data not shown). Steroid-free bFF and bovine serum preparations were sterilized by passage through 0.2-µm membrane filters before use. The treatment period lasted for 60 h, starting at 0800 h on the day of ovulation (i.e., 4 days after PRID removal). Treatments were administered via indwelling jugular vein catheters and were flushed from the catheters with sterile 0.9% (w/v) saline containing sodium heparin (100 IU/ml). Plasma inhibin A, FSH, estradiol-17ß (E), and progesterone (P) were measured in daily blood samples (20 ml) collected as described for study 1 for 12 days from the time of PRID removal. Ovarian follicle and luteal development were monitored daily by transrectal ultrasonography as described for study 1, starting at the time of PRID removal, for 12 days.
Inhibin A Assays
Preparation and biotinylation of mAb to bovine subunit
Hypogonadal (hpg strain) mice were immunized with a synthetic peptide corresponding to residues 1–32 of the bovine C subunit coupled to tuberculin-purified protein derivative as described previously [33]. Hybridoma supernatants were screened initially using ELISA plates coated with purified, 32-kDa bovine inhibin using a standard procedure [34]. For secondary screening, positive clones showing the highest responses were tested in a competitive RIA procedure (employing 125I-labeled 32-kDa bovine inhibin with different concentrations of unlabeled 32-kDa bovine inhibin as competitor) to identify the mAb having the highest affinity for bovine 32-kDa inhibin. This hybridoma was recloned in methyl cellulose, and the resultant clone (PPG14/6) was identified as an immunoglobulin (Ig) G1 secretor. This hybridoma was expanded in culture, IgG purified from the supernatant by standard protein G chromatography, and biotinylated as described previously [34].
Assay procedure
Serial dilutions of purified 32-kDa bovine inhibin A (assay standard: 1.56–100 pg per 100 µl), pooled bFF (1:5000–1:40 000 v:v), and bovine plasma samples (neat to 1:8 v:v) were prepared using ovariectomized cow plasma as diluent to compensate for any matrix effect of bovine plasma in the assay. Aliquots (100 µl) of diluted standards and test samples were mixed with 50 µl of SDS solution (6% [w/v]) and heated for 10 min at 90°C. After cooling, 100 µl of ELISA buffer (10% [w/v] BSA, 5% [v;clv] Triton X-100, 2% [v/v] normal mouse serum, 0.1% [w;clv] sodium azide, and 0.15 M sodium chloride in 0.1 M Tris-HCl buffer [pH 7.5]) and 50 µl of distilled water containing 10% (v/v) hydrogen peroxide (Sigma UK Ltd., Poole, Dorset, UK) were added. Tubes were mixed and incubated for 30 min at room temperature before transferring duplicate, 100-µl aliquots to microplates coated with E4 mAb against ßA [34], which were mixed again and placed in a humid box overnight. Plates were washed (15 cycles) with wash buffer (0.05 M Tris-buffered saline [pH 7.5] containing 0.05% [v/v] Tween 20) before adding to each well 50 µl of ELISA buffer containing 1 µg/ml of biotinylated subunit-specific mAb (PPG14/6). After incubating for 2 h at room temperature, plates were washed (15 cycles), and 50 µl of ELISA buffer containing extravidin-alkaline phosphate conjugate (1:20 000; Sigma) were added to each well. After a further 1 h of incubation, plates were washed thoroughly (15 cycles), and bound alkaline phosphatase was quantified using a commercially available ELISA amplification kit (Immuno Select ELISA Amplification System; Gibco BRL, Uxbridge, Middlesex, UK) according to the supplier's instructions. Absorbance at 492 nm was read on a microplate reader (Emax; Molecular Devices, Winnersh Triangle, Berks, UK), and data were processed by immunoassay curve-fitting software (Riacalc; Pharmacia, Milton Keynes, Bucks, UK).
Validation of the inhibin A ELISA for bovine plasma
Pooled bFF and plasma from both normal cycling heifers and superovulated heifers treated with FSH gave assay dilution curves parallel to the purified 32-kDa bovine inhibin standard (data not shown). Recovery of bovine inhibin A standard added to individual bovine plasma samples before assay was 109% ± 6% (n = 16). The assay detection limit was approximately 15 pg/ml, and mean within- and between-plate coefficients of variation (CVs) were less than 10%. Neither recombinant human (rh) activin A, rh-activin B, or rh-inhibin B (gifts from Genentech, Inc., South San Francisco, CA) at 25 ng/ml; bovine inhibin subunit (pro-C [25]) at 100 ng/ml; nor rh-follistatin-288 (gift from NIDDK) at 100 ng/ml showed any detectable cross-reaction in the ELISA (all <0.1% relative to the bovine inhibin A standard).
Fractionation of bFF using SDS-PAGE
To determine whether different molecular mass forms of inhibin A cross-react in the ELISA, 1-µl aliquots of pooled bFF were fractionated by SDS-PAGE under both reducing and nonreducing conditions as described previously [30, 35]. Prestained molecular weight markers were used to calibrate the gel (RainbowMarkers; Amersham PLC, Milton Keynes, Bucks, UK). Gel strips were cut into 38 slices (1.5 mm in width) using a scalpel blade, and gel slices were eluted in 0.5 ml of ELISA buffer for 3 days at 4°C. Assay of the resultant extracts showed that the inhibin A ELISA detects inhibin A forms in bFF with molecular masses of approximately 30, 55, 65, and 75 kDa (Fig. 1). No inhibin A immunoreactivity was detected when bFF was fractionated under reducing conditions, confirming the specificity of the ELISA for the -ßA dimer.
|
Gonadotropin Assays
Plasma FSH concentrations were measured by RIA as described previously [9, 36] using U.S. Department of Agriculture bFSH-BP-1 as the standard. The detection limit was 1.2 ng/tube, and mean within- and between-assay CVs were both less than 10%. Plasma LH concentrations were measured by RIA as described previously [13, 14]. The detection limit of the assay was 0.15 ng/tube, and within- and between-assay CVs were both less than 10%.
Steroid Assays
In study 1, plasma E concentrations were measured in daily plasma samples using the RIA described by Glencross and Pope [37]. The detection limit of the assay was 1.7 pg/ml. Inter- and intraassay CVs were 10.5% and 15.0%, respectively. In study 2, plasma E concentrations were measured using a more sensitive RIA procedure as described previously [38, 39]. The detection limit was 0.5 pg/ml, and mean intra- and interassay CVs were 6.3% and 14.9%, respectively. Concentrations of P in daily plasma samples collected in both studies were measured by the competitive ELISA described by Sauer et al. [40] as modified by Glencross et al. [8, 36]. The detection limit was 0.2 ng/ml, and mean inter- and intraassay CVs were both less than 10%.
Statistical Analysis and Data Presentation
Initially, plasma hormone profiles for all 12 PG-synchronized cycles monitored in study 1 were combined for statistical analysis and results presentation. Before amalgamation, individual hormone profiles were realigned around the time of each preovulatory LH surge, because the interval from PG injection to the LH surge varied both within and between heifers. On the basis of ovarian ultrasound observations, hormone profiles for eight cycles were subsequently reanalyzed from the day of emergence (i.e., the day that the follicle was first 
5 mm in diameter) of the first-wave DF (DF1) through to the subsequent ovulation to cover the entire period of growth and regression or ovulation of this follicle. Because sampling did not start until Day 0, the period of development of DF1 associated with the first follicular phase was not monitored before PG injection. Complete data, therefore, were only available for two cycles per heifer (n = 8 cycles in total). Examination of individual profiles showed that in three of eight cycles, DF1 ovulated, whereas in five of eight cycles, the second-wave DF (DF2) ovulated. Therefore, cycles in which DF1 ovulated were plotted separately from those in which DF2 ovulated. Plasma hormone profiles were not realigned in study 2. Statistical analysis of plasma hormone concentrations was performed by repeated measures ANOVA, with Fisher's protected least significant differences post-hoc test used for individual comparisons between different time points (study 1 and 2) and between treatment groups (study 2). A P value of less than 0.05 was regarded as statistically significant.
RESULTS
Study 1: Observations During PG-Synchronized Cycles
Plasma inhibin A, gonadotropin, and steroid profiles
Figure 2 shows plasma hormone profiles based on amalgamated data from all 12 cycles studied. The mean interval between PG injection and the preovulatory LH surge was 83 ± 3 h. Plasma inhibin A (50 pg/ml before PG injection) rose after PG-induced luteolysis (P < 0.05) to maximal levels (125 pg/ml) coincident with the preovulatory peak in gonadotropins and E. After ovulation, inhibin A fell sharply (P < 0.05) to a nadir (55 pg/ml) coincident with the secondary FSH rise. During the next 3 days, inhibin A increased gradually to approximately 90 pg/ml. Plasma FSH levels fell approximately threefold after PG injection, followed by a preovulatory surge coincident with the LH surge. A smaller, secondary rise in FSH occurred approximately 24 h later. Plasma E concentrations increased immediately after PG-induced luteolysis to reach a peak within 3 days, shortly before the preovulatory LH surge. Levels then fell sharply before increasing again to a much smaller peak (P < 0.05) 4–5 days after the preovulatory LH surge. Plasma P levels decreased by more than 80% within 1 day of PG injection. Levels of P remained low until Day 6–7 after PG-injection, when they started to rise again, confirming that ovulation and new corpus luteum formation had occurred.
|
Follicle development Analysis of sequential ovarian ultrasound images showed that in five of eight cycles for which full data were obtained (i.e., DF growth observed from emergence to ovulation), the ovulatory DF emerged within 1 day of PG-induced luteolysis and grew during the induced follicular phase (Fig. 3a). By contrast, in three of eight cycles, the ovulatory DF was already identifiable as a morphologically dominant follicle (>10 mm in diameter) at the time of PG-induced luteolysis (Fig. 3b). The interval from induced luteolysis to ovulation was shorter (P < 0.05) in the latter group (4.3 ± 0.3 vs. 5.3 ± 0.2 days). Ovulatory follicles larger than 10 mm on the day of induced luteolysis were DF1s (Fig. 3a), whereas those emerging at approximately the time of PG injection were DF2s (Fig. 3b). The interval from the previous ovulation to PG injection was shorter (P < 0.05) for cycles in which DF1 ovulated than for those in which DF2 ovulated (8.3 ± 0.3 vs. 9.2 ± 0.2 days, respectively).
|
Relationship between follicle development and plasma hormone levels A significant effect of time after ovulation was observed with inhibin A levels (P < 0.0001), and a significant time x ovulatory DF group interaction was also found (P < 0.0001). Plasma inhibin A levels were low on the day of ovulation (Day 0; Fig. 3) but increased in association with the growth of DF1, reaching maximum levels at approximately Day 3 regardless of the wave origin of the ovulatory follicle. Mean levels then declined progressively, reaching a nadir on Day 8. Following PG-induced luteolysis, inhibin A levels increased regardless of whether DF1 or DF2 ovulated. However, inhibin A levels increased more quickly during cycles in which DF1 ovulated, reaching a maximum 11 days after PG, compared with 12 days to maximum levels during cycles in which DF2 ovulated.
Plasma FSH levels were similar throughout the study period for cycles in which the first- or second-wave DF ovulated (Fig. 3). The FSH levels increased following ovulation but decreased again following emergence of the first follicle wave, reaching a nadir 3 days after ovulation. Mean levels then progressively increased to a maximum on Day 8, then fell again during the subsequent induced follicular phase. Peak levels occurred on the day of PG injection during cycles in which DF1 ovulated and on the day after PG injection during cycles in which DF2 ovulated. The interval from PG injection to the preovulatory LH surge was shorter when the ovulatory follicle was derived from the first rather than the second follicle wave, because the LH surge occurred later (90.0 ± 4.0 vs. 68.7 ± 4.1 h, respectively; P < 0.05).
Plasma E levels were similar between the two follicle groups between Day 0 and PG-induced luteolysis (Fig. 3). However, E levels increased more rapidly (P < 0.0005) during the cycles in which DF1 ovulated compared to those in which DF2 ovulated. Mean plasma E levels increased within 1 day of PG injection when DF1 ovulated rather than on Day 11 during cycles in which DF2 ovulated. Levels of E returned to basal by Day 13 when DF1 ovulated, compared with Day 14 when DF2 ovulated. Plasma P levels were similar between the two groups.
A negative linear relationship was found between plasma FSH and both inhibin A (r = -0.37, P < 0.0001) and E (r = -0.36, P < 0.0001). Conversely, a positive linear relationship was found between E and inhibin A (r = 0.49, P < 0.0001).
Study 2: Effects of bFF on Plasma Hormones and Follicle Turnover
Plasma hormone concentrations were similar among the three treatment groups from the time of PRID removal until the start of the treatment period.
FSH
Plasma FSH levels were suppressed to the same extent (fourfold lower than control levels) during treatment with either high or low doses of bFF (Fig. 4). Within 1 day of the end of treatment, a large rebound rise in FSH levels occurred that lasted for approximately 2 days in both bFF-treated groups. Levels of FSH were higher (P < 0.001) on the day after the end of treatment (Day 7) in heifers treated with the low dose of bFF than in those receiving the high dose, but they were similar in the two groups of bFF-treated heifers by Day 8. Levels of FSH increased in control heifers on Days 11 and 12 after PRID removal, which is consistent with the first-wave dominant follicle losing functional dominance.
|
Inhibin A
Within 4 h of the start of bFF treatment, inhibin levels in the high-dose group had increased (P < 0.001) to supraphysiological levels (1 ng/ml). Inhibin A levels remained elevated during bFF treatment but fell by 60% within approximately 12 h of stopping treatment, although levels were still some threefold higher than those in control heifers at that time. During the treatment period, circulating inhibin A levels were only marginally (not significant compared to controls) raised in those heifers treated with the low dose of bFF, despite the marked effect on plasma FSH levels.
Steroids Plasma E levels were similar among bFF-treated and control heifers during the treatment period. A small increase in the E concentration was found in control heifers 8 days after PRID removal, but this increase was absent in both bFF-treated groups (P < 0.05 compared with controls). Plasma P levels were similar among the three treatment groups throughout the study period.
Follicle and luteal development In control heifers, the first wave of follicles emerged on the day of ovulation, and one of these follicles had become dominant within 3 days. This follicle retained morphological and functional dominance throughout the study period (i.e., emergence of the second wave was not observed). Emergence of the first follicle wave was also observed in all four heifers treated with the low dose of bFF but in only two of three heifers treated with the high dose. When follicle wave emergence was observed, the largest follicle grew to only approximately 6 mm in diameter and then regressed, being no longer visible by ultrasonography 3 days after the end of bFF treatment. A second wave of follicles emerged 3–4 days after the end of bFF treatment in both the low- and high-dose groups of heifers, and one follicle from this wave had become dominant within 6 days of the end of treatment. Development of the corpus luteum was not affected by either dose of bFF.
DISCUSSION
We previously reported use of a two-site ELISA for measurement of peripheral plasma concentrations of inhibin A in ewes [31]. However, this assay was insufficiently sensitive to measure circulating inhibin A levels during the bovine estrous cycle and only useful for measuring raised inhibin levels in gonadotropin-stimulated cattle. By developing a new mAb (clone PPG14/6) and incorporating it into this system, the sensitivity of the assay has been greatly improved (detection limit: 15 vs. 50 pg/ml [31]). This has enabled us to determine profiles of inhibin A during the bovine estrous cycle for the first time and to explore the relationship between inhibin A, E, and FSH secretion and the pattern of follicle development. The ELISA reported is specific for dimeric inhibin A (i.e., -ßA dimer), as demonstrated by negligible cross-reactivity with both the free subunit of inhibin (pro-C), rh-inhibin B, activin A, activin B, and follistatin-228, which were all less than 0.1% relative to the bovine inhibin A standard. This specificity arises from use of two mAbs: one against the ßA subunit, and the other recognizing an epitope on the subunit. As expected, this system recognizes larger-molecular-weight forms of inhibin A [22] as well as the fully processed, 32-kDa form. Following fractionation of bFF by SDS-PAGE, the assay detected inhibin A forms corresponding to 75, 65, 55, and 32 kDa. Under reducing conditions, no immunoreactivity was detected, confirming that the assay only detects intact, disulfide-linked -ßA dimers. The precision and accuracy of the assay are both within acceptable limits.
Following PG-induced luteolysis, inhibin A levels increased during the follicular phase, but as reported by others [8, 12, 41], FSH levels fell. This finding supports an endocrine role for inhibin A as a negative feedback suppressor of pituitary FSH at this stage of the cycle, and it is consistent with the observation that passive immunization against inhibin during the follicular phase raises plasma FSH levels in cows [6]. Because E levels also increase during the follicular phase [42, 43] coincident with final maturation of the ovulatory follicle, pituitary FSH release probably is coregulated by both E and inhibin A acting in concert [44, 45]. The increased inhibin A levels observed during the bovine follicular phase in study 1 contrasts with our recent observations in the cycling ewe, in which no consistent rise in plasma inhibin A was evident after induction of luteolysis [31]. The explanation for this difference is not known, but it might reflect a difference between sheep and cattle in the relative contribution that the DFs make to the total inhibin A levels in the peripheral circulation. Thus, a greater proportion of circulating inhibin in the ewe might be derived from the small follicle population, which would tend to lessen the impact of the DFs on circulating levels.
In study 1, inhibin A levels fell sharply following the LH surge, reaching a nadir coincident with the postovulatory rise in FSH that is considered to provide the stimulus for emergence of the first follicular wave [4, 5]. During emergence of the new follicle wave and selection of DF1, plasma inhibin A levels increased coincident with the postovulatory rise in E. The combined effects of raised inhibin A and E account for the associated decline in FSH at this time, which forms the basis of the follicle-selection mechanism [46]. Approximately 5 days later, inhibin A levels declined again, allowing a second increase in FSH. Evidence from passive immunoneutralization studies supports a functional role for inhibin in regulating FSH secretion and follicle selection during the early luteal phase [7]. During cycles in which PG injection coincided with maximum concentrations of FSH, DF1 ovulated following luteolysis (three of eight cycles). However, when PG was injected after maximum FSH, coincident with the emergence of the second follicle wave, DF2 ovulated. This variation in origin of the ovulatory DF was associated with differences in the interval from the previous ovulation to the PG injection. The protocol used in study 1 involved successive PG injections at a fixed time relative to one another rather than relative to ovulatory events. The interval from ovulation to PG injection was approximately 8 days for cycles in which the first-wave DF ovulated, compared with approximately 9 days for cycles in which the second-wave DF ovulated. This allowed for subtle differences in endocrine profile and follicle status at the time of PG injection, which appear to be sufficient to significantly alter ovulatory follicle dynamics. Other studies have shown that the interval from estrus/ovulation to PG administration influences the origin of the ovulatory follicle. Kastelic et al. [47] found that DF1 ovulated in heifers injected with PG as much as 8 days after ovulation. Also, in this study, the interval from PG-induced luteolysis to the subsequent estrus/ovulation was increased in heifers ovulating DF2, which is consistent with the increase in the interval from PG injection to peak E and inhibin A associated with second-wave ovulatory DFs in the present study. It is noteworthy that, despite inhibin A and E having reached basal levels before PG-induced luteolysis, the ovulatory DF of wave 1 evidently can resume production of inhibin A and E after P levels have fallen, presumably because of increased pulsatile LH secretion. Indeed, levels of inhibin A and E increased more rapidly during these cycles (i.e., reaching peak levels 1 day earlier) than during cycles in which DF2 ovulated.
The relationship between the pattern of inhibin A release and that of E, FSH, and follicle growth presented in study 1 was similar to that reported by Taya et al. [24] for immunoreactive (ir) inhibin levels in cycling cows and by Hasegawa et al. [23] in superovulated cows. However, the relative profiles of ir-inhibin differ from those of cycling cows reported by Hasegawa et al. [23]. In their study, ir-inhibin levels fell as E levels increased during the follicular phase, but then increased following the LH surge. The results of these earlier studies may be confounded by the fact that nonbioactive, free subunit is present at high concentrations in the peripheral circulation of cattle [21, 25], and this subunit cross-reacts in the RIA system used by these authors.
The results of study 1 appear to contradict reports of raised intrafollicular inhibin concentrations during DF1 atresia [26, 27, 48], and they highlight the uncertainties of trying to relate intrafollicular hormone contents with systemic levels. The present study has revealed that circulating dimeric inhibin A levels are lowest at the time of DF1 atresia, but that they increase during the growth phase of the first-wave DF and during the final maturation of the ovulatory DF arising from either wave 1 or 2. This finding is more consistent with the observation that levels of , ßA, and ßB subunit mRNA are higher in E-active than in atretic bovine follicles [20]. The increase in 32-kDa inhibin found in atretic follicles has been suggested to result from continued proteolytic processing of larger molecular weight forms of dimeric inhibin rather than from de novo synthesis [48]. However, this would not readily explain the apparent inverse relationship between intrafollicular and circulating levels of inhibin A during the growth and plateau/atresia stages of DF1 development. An alternative explanation is that the exit rate of inhibin from the follicle is, somehow, restricted during the plateau/atresia of DF1 development, leading to accumulation of inhibin in follicular fluid. It is tentatively suggested that this could involve a reduced blood flow through the thecal capillaries at this time.
During the period of bFF treatment in study 2, FSH levels were suppressed to approximately 40% of control levels. This is consistent with other studies in which heifers were treated with steroid-free bFF during the follicular phase [12, 18] or the postovulatory period [4, 15]. Interestingly, in the present study, FSH levels were suppressed to the same extent in heifers treated with the high or low dose of bFF, despite only a marginal (nonsignificant) increase in circulating inhibin A levels in the latter group above those of serum-treated controls. This observation may reflect the high potency of inhibins as suppressors of pituitary FSH release, because an apparently small increase in inhibin is associated with a dramatic fall in FSH secretion. Alternatively, as suggested elsewhere [18, 19], other nonsteroidal factors in bFF (follistatin?) may be partially responsible for suppressing FSH release and for arresting follicle development. On the basis of the present results, the possibility of direct ovarian effects of inhibin or other nonsteroidal factors in bFF cannot be excluded.
A rebound hypersecretion of FSH following withdrawal of bFF treatment has been reported in other studies of intact heifers [12, 18] but does not occur in ovariectomized animals [11, 14]. This suggests that the response is ovary mediated and, probably, depends on a transient reduction in the ovarian output of E and/or inhibin. The present study has demonstrated, to our knowledge for the first time, that circulating E levels are reduced at the time of the FSH rebound, whereas plasma inhibin A levels are not. Indeed, plasma inhibin A levels in the high-dose bFF group were actually approximately threefold higher than those in controls at the time of the FSH rebound. This suggests that pituitary gonadotropes become desensitized to the suppressive action of inhibin. The FSH rebound was of similar magnitude in both the low- and high-dose groups but was delayed by approximately 1 day in the latter group, which may be the result of a slower clearance rate of inhibin following administration of the high dose of bFF. Emergence of the first follicle wave was observed in all heifers treated with the low dose of bFF and in two of three heifers treated with the high dose, but in an environment of subbasal FSH levels, continued development of the DF was suppressed. Other studies have shown that suppressing FSH release at various stages during development of the dominant [16, 17] or induced-persistent follicles [49] brings about their premature regression. Collectively, these data confirm that basal levels of FSH (as observed during follicle dominance in normal cycles) play a key role in maintaining follicle development.
In summary, we have validated a specific and sensitive two-site ELISA capable of quantifying inhibin A levels in peripheral plasma during the normal bovine estrous cycle. An inverse relationship between the release of pituitary FSH from the pituitary and inhibin was demonstrated. These results support a key role for inhibin A produced by the DF in the generation and termination of transient peaks of FSH secretion, which are intimately involved in the recruitment and selection of follicles. A role for E as a coregulator of FSH is also likely in this regard, and the possible involvement of other nonsteroidal follicular products cannot be discounted. The characteristic FSH rebound that follows cessation of bFF treatment can be accounted for by a transient reduction of circulating E, but not of inhibin A, combined with a desensitization of pituitary gonadotropes to inhibin.
ACKNOWLEDGMENTS
We are grateful to K. Edwards for technical assistance and to D.J. Bolt (U.S. Department of Agriculture, Beltsville, MD) and S. Raiti (NHPP, Baltimore, MD) for providing gonadotropin assay reagents.
FOOTNOTES
First decision: 7 August 2000.
1 Supported by Biotechnology and Biological Sciences Research Council and Ministry of Agriculture, Fisheries and Food. 
2 Correspondence. FAX: 44 118 931 0180; p.g.knight{at}reading.ac.uk
Accepted: October 2, 2000.
Received: June 28, 2000.
REFERENCES
C-subunit on plasma anti-inhibin titers, plasma FSH concentrations and the incidence of multiple ovulation in heifers. Anim Reprod Sci 1996; 41:1–12-subunit on gonadotrophin concentrations in heifers. Anim Reprod Sci 1995; 38:63–71-subunit and its isolation from bovine follicular fluid. J Mol Endocrinol 1989; 2:189–200 ßA dimer) during the estrous cycle, after manipulation of ovarian activity and during pregnancy in ewes. J Reprod Fertil 1998; 113:159–166 and ß subunit. J Endocrinol 1991; 129:R9–R12 treatment on selection and development of the ovulatory follicle in heifers. Anim Reprod Sci 1990; 23:169–180[CrossRef] subunit and -ß dimer) in morphologically dominant follicles during their growing and regressing phases of development in cattle. Biol Reprod 1993; 48:268–276[Abstract]This article has been cited by other articles:
|
|
 |
|
  J.J. Ireland, A.E. Zielak-Steciwko, F. Jimenez-Krassel, J. Folger, A. Bettegowda, D. Scheetz, S. Walsh, F. Mossa, P.G. Knight, G.W. Smith, et al. Variation in the Ovarian Reserve Is Linked to Alterations in Intrafollicular Estradiol Production and Ovarian Biomarkers of Follicular Differentiation and Oocyte Quality in Cattle Biol Reprod, May 1, 2009; 80(5): 954 - 964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R Kayani, C. Glister, and P. G Knight Evidence for an inhibitory role of bone morphogenetic protein(s) in the follicular-luteal transition in cattle Reproduction, January 1, 2009; 137(1): 67 - 78. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Veiga-Lopez, W. Ye, D.J. Phillips, C. Herkimer, P.G. Knight, and V. Padmanabhan Developmental Programming: Deficits in Reproductive Hormone Dynamics and Ovulatory Outcomes in Prenatal, Testosterone-Treated Sheep Biol Reprod, April 1, 2008; 78(4): 636 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Burns, F. Jimenez-Krassel, J. L.H. Ireland, P. G. Knight, and J. J. Ireland Numbers of Antral Follicles During Follicular Waves in Cattle: Evidence for High Variation Among Animals, Very High Repeatability in Individuals, and an Inverse Association with Serum Follicle-Stimulating Hormone Concentrations Biol Reprod, July 1, 2005; 73(1): 54 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J M Haughian, O J Ginther, K Kot, and M C Wiltbank Relationships between FSH patterns and follicular dynamics and the temporal associations among hormones in natural and GnRH-induced gonadotropin surges in heifers Reproduction, January 1, 2004; 127(1): 23 - 33. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Webb, P. C. Garnsworthy, J.-G. Gong, and D. G. Armstrong Control of follicular growth: Local interactions and nutritional influences J Anim Sci, January 1, 2004; 82(13_suppl): E63 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.J.M. Hendriksen, B.M. Gadella, P.L.A.M. Vos, E. Mullaart, T.A.M. Kruip, and S.J. Dieleman Follicular Dynamics Around the Recruitment of the First Follicular Wave in the Cow Biol Reprod, December 1, 2003; 69(6): 2036 - 2044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Sisco, L. J. Hagemann, A. N. Shelling, and P. L. Pfeffer Isolation of Genes Differentially Expressed in Dominant and Subordinate Bovine Follicles Endocrinology, September 1, 2003; 144(9): 3904 - 3913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Jimenez-Krassel, M. E. Winn, D. Burns, J. L. H. Ireland, and J. J. Ireland Evidence for a Negative Intrafollicular Role for Inhibin in Regulation of Estradiol Production by Granulosa Cells Endocrinology, May 1, 2003; 144(5): 1876 - 1886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.I. Parker, D.M. Robertson, N.P. Groome, and K.L. Macmillan Plasma Concentrations of Inhibin A and Follicle-Stimulating Hormone Differ Between Cows with Two or Three Waves of Ovarian Follicular Development in a Single Estrous Cycle Biol Reprod, March 1, 2003; 68(3): 822 - 828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Tanaka, H. Taniyama, N. Tsunoda, C. B. Herath, R. Nakai, H. Shinbo, N. Nagamine, Y. Nambo, S.-i. Nagata, G. Watanabe, et al. Localization and Secretion of Inhibins in the Equine Fetal Ovaries Biol Reprod, January 1, 2003; 68(1): 328 - 335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kaneko, J. Noguchi, K. Kikuchi, J. Todoroki, and Y. Hasegawa Alterations in Peripheral Concentrations of Inhibin A in Cattle Studied Using a Time-Resolved Immunofluorometric Assay: Relationship with Estradiol and Follicle-Stimulating Hormone in Various Reproductive Conditions Biol Reprod, July 1, 2002; 67(1): 38 - 45. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.A. Beg, D.R. Bergfelt, K. Kot, and O.J. Ginther Follicle Selection in Cattle: Dynamics of Follicular Fluid Factors During Development of Follicle Dominance Biol Reprod, January 1, 2002; 66(1): 120 - 126. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
