| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
School of Human Development,3 University of Nottingham, Queen's Medical Centre, Nottingham NG7 2UH, United Kingdom
Department of Reproductive and Developmental Sciences,4 Obstetrics and Gynaecology, University of Edinburgh, Edinburgh EH3 9EW, United Kingdom
ABSTRACT
The aim of this study was to examine the role of LH on the growth of the large preovulatory follicle and its secretion of hormones in sheep. Ewes with ovarian autotransplants were treated with GnRH-antagonist at the time of luteal regression and different LH regimes applied for 60–66 h before administration of an ovulatory stimulus (hCG). In Experiment 1 (N = 24; n = 8), ewes received either no LH or constant or pulsatile infusion of LH at the same dose (1.25 µg/h). In Experiment 2 (N = 12, n = 6), LH was constantly infused at a rate of 1.25 µg or 2.5 µg oLH/h. In Experiment 1, animals receiving either pulsatile or constant LH exhibited increases in estradiol and inhibin A secretion (P < 0.001) and a depression in FSH (P < 0.001) that resembled the normal follicular phase. Similarly in Experiment 2, doubling the dose of LH resulted in a two-fold increase in ovarian estradiol secretion (P < 0.05) but no other changes. All animals receiving LH, regardless of the pattern of stimulation, ovulated and established a normal luteal phase. In contrast, no LH treatment resulted in constant immuno-active LH without pulses, unchanged FSH and inhibin A concentrations (P < 0.05), and basal estradiol secretion (P < 0.001). Morphologically normal large antral follicles were observed in this group and although corpora lutea formed in response to hCG, progesterone profiles were abnormal. In conclusion, these results suggest that LH is an essential requirement for normal ovulatory follicle development and subsequent luteal function and show that a pulsatile mode of LH stimulation is not required by ovulatory follicles.
estradiol, follicle-stimulating hormone, inhibin, inhibin A, luteinizing hormone, ovary, ovulatory follicle, progesterone
The growth and development of ovarian follicles from the late preantral [1] to the preovulatory stage [2] is regulated by the pituitary gonadotropins, LH and FSH. These gonadotropins exert their effects on ovarian somatic cells via specific membrane bound receptors that exhibit alternate patterns of expression. FSH receptors occur exclusively on the granulosa cells of ovarian follicles from primary through to the preovulatory stages of follicle development [3], whereas LH receptors first develop on the cells of the theca interna at the tertiary stage of development and this pattern of expression is maintained through to the preovulatory stage [4]. In addition, it is well established that the granulosa cells of large estrogenic antral follicles also develop LH receptors [2, 5]. In common with most species, pituitary LH secretion is pulsatile in domestic ruminants and the amplitude and frequency of this pulsatile pattern of secretion varies across the estrous cycle in response to changes in the level and type of ovarian steroidal feedback [6]. In contrast, peripheral FSH is not detectably pulsatile in sheep and cattle and pituitary FSH release is controlled through the synergistic action of estradiol and inhibin [7].
Through the utilization of a variety of gonadotropin suppression model systems, it has been convincingly demonstrated that in sheep [8–10], cattle [11, 12], and humans [13], antral follicle development beyond a diameter of 2–4 mm is totally dependent on gonadotropic support and that FSH and LH act in concert to control antral follicle development and selection in these species. However, due to the fact that both FSH and LH release are controlled by the same releasing factor, GnRH, it has proved difficult to define concisely the individual actions of each gonadotropin.
As its name implies, FSH plays a pivotal role in the stimulation of both follicle growth and differentiation [2, 14–16]. FSH alone [9, 17], but not LH [18], can stimulate the growth of follicles to a preovulatory size in ewes made hypogonadotropic by long-term treatment with GnRH analogues. Similarly, in cattle short-term FSH infusion of GnRH-agonist suppressed heifers induced changes in the mRNA expression of major developmental markers that were identical to those observed in recruited follicles during the first follicular wave in normal animals [5]. Further, specific depression of FSH by treatment with follicular fluid [19–22] or inhibin [23], or withdrawal of FSH support in GnRH-antagonist depressed ewes [17] results in a rapid decline and atresia of ovulatory-sized follicles. The pivotal role of FSH in antral follicle development has been confirmed by gene knockout in mice [24] and naturally occurring mutations in humans [25] in FSH or its receptor with a block in folliculogenesis at the pre-antral stage of developmental. Recent studies in sheep, however, have shown that while FSH is not required for pre-antral follicle development, it can influence the rate of growth and health of preantral follicles [1].
While LH is primarily thought of as a steroidogenic hormone, evidence is accumulating that it can directly and indirectly influence the development and maintenance of antral follicles. A direct role of LH in antral follicle development is suggested by the recent results from LH and LH-R knockout mice, which show normal preantral follicle development but an absence of healthy antral preovulatory follicles and many abnormal antral follicles containing degenerating oocytes [26–28]. In most species, it is well established that the final maturation and development of antral follicles to ovulation following luteolysis is dependent on an increase in the pulsatile secretion of LH [13, 29–31] as FSH is depressed by large amounts of estradiol and inhibin A secreted by the ovulatory follicle. Similarly, ovulation can be induced in anestrus ewes by exogenous administration of hCG, GnRH, or LH [8, 32–34], or the increase in the release of endogenous LH, which results from exposure to rams [35]. In addition, LH administered as a series of relatively high-amplitude, low-frequency pulses is able to inhibit FSH-stimulated follicle development in GnRH agonist-suppressed ewes [18] and immunoneutralization of basal LH by co-administration of LH antiserum also inhibits FSH-stimulated follicle development [36]. Finally, utilizing GnRH-antagonist suppressed ewes, we have been able to demonstrate that ovulatory follicles can be maintained by pulsatile LH alone when FSH support is withdrawn [17] and that subsequent withdrawal of pulsatile LH resulted in acute atresia of these LH-dependent ovulatory sized follicles [37]. Taken together, these observations suggest that both FSH and LH play a role in the control of selection of the ovulatory follicle(s) and its development to a preovulatory size.
In addition to direct effects on follicle development, LH can also indirectly affect follicle development through its steroidogenic role. The main action of LH on steroid secretion is the stimulation of the conversion of cholesterol to pregnenolone [38], although 3ß-hydroxysteroid dehydrogenase is stimulated as well [39]. As the cholesterol-pregnenolone conversion is an early step in steroid synthesis, the synthesis of all ovarian steroids is increased following administration of LH [40], and as LH secretion is pulsatile [41], each pulse of LH is followed by an increase in the secretion of androstenedione and estradiol from the ovary (rat: [42] sheep: [29, 43]; cow: [44, 45]). Thus, it appears likely that the level of ovarian estrogen secretion is dependent not only on the presence of an estrogenic follicle in the ovary but also the pattern of pulsatile LH that this follicle is exposed to [17]. Theoretically, LH is therefore able to influence the amount and pattern of pituitary FSH release through its action in controlling the level of ovarian estradiol secretion.
From this review, it is clear that LH plays an important role in ovulatory follicle development but key questions remain unanswered in terms of its folliculogenic and steroidogenic actions. The first of these is whether LH is necessary for normal ovulatory follicle development when FSH concentrations remain in the normal physiological range (as compared to super-physiological in most ovarian stimulation regimes). The second question is whether the pulsatile pattern of LH stimulation that the ovary is normally exposed to is essential for its folliculogenic and steroidogenic actions or is merely a consequence of the pulsatile secretion of GnRH by the pituitary. We have previously used long-term GnRH-antagonist treatment to chronically suppress endogenous gonadotropin secretion in ewes with an ovarian autotransplant to examine the relative roles of LH and FSH in ovulatory follicle selection [17, 46]. The antagonist has the advantage of allowing acute suppression of endogenous LH and the autotransplant facilitates the determination of morphological and functional follicular responses to exogenous gonadotropic stimuli. In the current experiment, we use a modification of this model to examine these questions in that GnRH-antagonist treatment results in an acute suppression in endogenous pulsatile LH secretion but has no effect on endogenous FSH for several days after treatment [47]. The aim of the present study was therefore to examine the effect of dose of LH and pattern of stimulation with LH (pulsatile or constant) on ovarian ovulatory follicle development and ovarian hormone secretion.
Procedures were conducted in accordance with the Animals (Scientific Procedures) act of 1986 (UK) and following local ethical review. The experimental animals were Merino cross ewes in which the left ovary and its vascular pedicle had been autotransplanted to a site in the neck [48] at least 12 months previously in order to facilitate collection of ovarian venous blood and monitoring of ovarian follicular development by transdermal ultrasound. The animals used for Experiment 1 were Scottish Blackface Merino cross (n = 24) and those used for Experiment 2 were Border Leicester or Finn Merino cross (n = 12). Both experiments were conducted during the mid-breeding season (November–December) and prior to the start of intensive blood sampling the animals were penned indoors under natural lighting and fed a maintenance diet consisting of concentrates and hay ad libitum.
On the day prior to the start of frequent blood sampling, all animals had a silastic cannula (0.8 mm inside diameter x 1.7 mm outside diameter; Dow Corning, Midland, MI) inserted into the jugular vein cranial to the ovarian jugular venous anastomosis for sampling ovarian venous blood. At the same time, a similar-sized Silastic cannula was inserted into the contralateral jugular vein for infusion of LH as previously described [17]. After cannulation, the animals were placed in metabolism crates in ventilated rooms under natural lighting and treated prophylactically with antibiotics (3 ml i.m./3 days; Clamoxil, SmithKline Beecham, Surrey, UK) and heparin (5000 IU i.v./12 h; Leo Laboratories, Aylesbury, Bucks, UK). The animals had been habituated to the housing conditions and frequent handling prior to the start of the experiment.
The LH used for was NIADDK-oLH-26, which has a biological potency of 2.3 U/mg with one unit having an activity of 1 mg NIH-LH-S1 and FSH contamination of less than 0.5% by weight. The hCG used for induction of ovulation was obtained from Intervet (Chorulon, Milton Keynes, Buckinghamshire, UK). All gonadotropin preparations were dissolved in 0.9% (w/v) sterile saline with 1% (v/v) normal sheep plasma. The LH was administered intravenously via the second jugular cannula using Harvard infusion pumps either continuously (2 ml/h) or in a pulsatile fashion through connection to a timer (2-minute infusion at a rate of 1 ml/min). The GnRH-antagonist Cetrorelix (GnRHa; donated by Asta Medical AG, Frankfurt, Germany) was dissolved in sterile water to a concentration of 1 mg/ml and administered subcutaneously at the time of induction of luteal regression at a dose rate of 50 µg/kg. Pilot experiments had shown that this preparation had similar potency and duration of action as that utilized previously [17, 46, 49]. The prostaglandin F2
analogue, cloprostenol (Estrumate, Schering-Plough Animal Health, Uxbridge, Middlesex, UK) was used according to the manufacturer's instructions to synchronize the animals' estrous cycles by inducing luteal regression.
Experiment 1. Luteal regression was induced on Day 12 of the estrous cycle by intramuscular injection of 125 µg cloprostenol (PG) and at the same time each animal received a subcutaneous injection of GnRHa (Fig. 1). To ensure down-regulation, the GnRHa injection at the same dose rate was repeated 48 h later. There were three treatment groups, with animals being randomized according to breed and weight between groups: 1) No LH, which received no exogenous LH for the following 66 h after PG (n = 8); 2) Constant LH, which received a constant infusion of LH at a rate of 1.25 µg/h for 66 h after PG (n = 8); and 3) Pulsed LH, which received pulsatile LH at a rate of 1.25 µg/h for 66 h after PG (n = 8). For the first 12 h after PG, this LH was administered as 2.5 µg boluses every 2 h and thereafter as 1.25 µg boluses every hour. Thus both treatment groups were given the same amount of LH (i.e., 1.25 µg/h). This regime was designed to mimic the increase in LH pulse frequency observed during the normal follicular phase in this species [29, 50]. Sixty-six hours after PG, the normal length of the follicular phase in these animals [51], all ewes received 100 µg of LH combined with 1000 IU of hCG administered as an intravenous injection in order to induce ovulation.
|
Experiment 2. This experiment utilized the same basic design as Experiment 1 in terms of synchronization, GnRHa treatment, and administration of an ovulatory dose of LH/hCG at 60 h after luteal regression. Experimental treatment involved constant infusion of LH at two dose rates; 1) 1.25 µg oLH/h for 60 h after PG (n = 6) and 2) 2.5 µg oLH/h for 60 h after PG (n = 6).
From the time of PG/GnRHa administration, ovarian venous blood samples were taken at 6-hour intervals in Experiment 1 and 3-hour intervals in Experiment 2, until 24 h after the end of administration of the ovulatory stimulus (84 h after PG; Fig. 1). Thereafter, ovarian venous blood samples were collected at 6–12 hour intervals for a further 12–24 h. In addition, daily jugular venous blood samples were collected by venipuncture every day for 7 days after administration of the ovulatory stimulus for determination of progesterone concentrations as a confirmation of ovulation. Further, in both experiments, more frequent ovarian venous blood samples were collected at 10-minute intervals for 4 h from 35.5–39.5 h after PG.
The diameter of the antral cavity and position of all follicles greater than 2 mm in diameter in the medial/lateral, dorsal/ventral, and cranial/caudal planes were determined every 12 h, as previously described [51, 52] using a combined real time Aloka 500 ultrasound scanner with a linear 7.5 MHz transducer probe (Dynamic Imaging, Livingston, UK). To estimate ovulation rate, animals were scanned 6 days after the artificial LH surge for the presence of corpora lutea [51].
Plasma concentrations of LH, FSH, estradiol [53], progesterone [52], and inhibin A [54] in venous plasma were determined using previously described radioimmunoassays. The sensitivity of the assays for LH, FSH, androstenedione, estradiol, and progesterone and inhibin A were 0.2 µg/l (NIDDK, oLH, S23), 0.3 µg/l (USDA, oFSH, SIAFP-RP2), 175 pmol/l, 50 pmol/l, 380 pmol/l, and 30 ng/l respectively. The intra- and inter-coefficients of variation for all the immunoassays were less than 12% in the 20%–80% effective dose range.
Statistical analysis of hormonal profile data was performed by repeated sample analysis of variance, with data being partitioned on the basis of treatment and time (ANOVA) after testing for normality and homogeneity. For periods of intensive blood sampling, overall mean level of secretion of LH and estradiol were calculated and compared by one-way analysis of variance. In Experiment 2, significant differences in FSH values were noted between treatment groups at the zero time point and data were therefore normalized relative to the time of PG and expressed as a percentage change from the initial blood sample. Analysis for the effect of treatment on follicle number and ovulation rate was performed following 
(x+0.5) transformation.
Gonadotropins in peripheral plasma. The pattern of LH stimulation achieved, determined during the period of intensive blood sampling 35.5–39.5 h after PG, is shown in Figure 2. No LH treatment resulted in complete absence of pulsatile LH secretion and a constant low concentration of 1.32 ± 0.89 µg/l. Constant infusion of LH resulted in a marked increase in peripheral LH concentrations (P < 0.05) and a constant pattern of secretion with an overall value of 1.66 ± 0.12 µg/l. Pulsatile mode of infusion resulted in regular LH pulses with a frequency of 60 min, an amplitude of 0.54 ± 0.09 µg/l, and basal level of 1.40 ± 0.04 µg/l. The overall mean LH value for the pulsatile group was 1.62 ± 0.09 µg/l and this did not differ from the constant infusion group but was significantly higher than the no LH group (P < 0.05). It was notable in all animals that the first sample of the period of intensive sampling was elevated. This is a common observation in these animals and is attributed, despite their habituation to handling, to the activity around the animal that accompanies the start of a pulse bleed. Injection of an LH/hCG bolus to induce ovulation 60 h after PG resulted in a marked increase in circulating LH values in all animals, with an overall mean of 17.33 ± 3.46 µg/l.
|
The pattern of LH stimulation applied had a marked effect on the pattern of FSH secretion during the infusion period. In animals receiving no LH, FSH concentrations remained constant and unchanged over the entire experimental period (Fig. 3) with an overall mean value of 2.05 ± 0.15 µg/l between 0–60 h after PG. In contrast, FSH values in both groups that received LH decreased significantly (P < 0.001) over the same period of LH treatment and returned to pre-treatment values following hCG (Fig. 3). Overall FSH concentrations for this period (0–66 h) did not differ between modes of LH stimulation (Constant 1.58 ± 0.09 µg/l ; Pulsed 1.54 ± 0.09 µg/l) but were significantly (P < 0.05) lower than in animals receiving no LH. In both groups receiving LH, FSH concentrations increased (P < 0.05) following administration of the ovulatory stimulus, and this increase was temporally related to the fall in ovarian estradiol and inhibin A secretion.
|
Estradiol secretion. The pattern of LH stimulation applied had a marked effect on the pattern of ovarian estradiol secretion during the infusion period. Estradiol concentrations in the ovarian venous plasma of animals receiving no LH remained low and constant at an overall concentration of 0.21 ± 0.02 nmol/l in the period 66 h after PG (Fig. 3). In contrast, both groups receiving LH exhibited a highly significant increase (P < 0.001) in ovarian estradiol secretion in the period of LH treatment to a peak that occurred at the time of hCG/LH injection, 66 h post-PG. Over this period, estradiol concentrations determined in the 6-h samples tended to be higher in the group receiving constant LH (2.29 ± 0.25 nmol/l) but this difference was not significantly different from the pulsed LH group (1.65 ± 0.17 nmol/l; P = 0.08; Fig. 3).
More intensive sampling however, in the period 35.5–39.5 h after PG, revealed marked differences in the pattern and level of estradiol secretion in the different treatment groups. As expected, pulsatile LH stimulation resulted in a pulsatile mode of estradiol secretion with regular estradiol pulses with a frequency of 60 min, an amplitude of 2.32 ± 0.08 nmol/l, a basal level of 1.53 ± 0.02 nmol/l, and an overall mean of 2.71 ± 0.14 nmol/l (Fig. 2). Constant LH infusion resulted in a high constant non-pulsatile mode of estradiol secretion that had an overall mean of 4.28 ± 0.18 nmol/l over the period of intensive bleeding and this rate of secretion was significantly higher (P < 0.01) than the pulsatile mode of stimulation. In the absence of LH, estradiol secretion was very low over the intensive bleeding window at an overall rate of 0.19 ± 0.02 nmol/l, and this concentration was significantly lower than for both groups of ewes receiving LH (P < 0.001).
Administration of an ovulatory dose of LH/hCG at 66 h following PG resulted in an acute fall in ovarian estradiol secretion in ewes receiving both modes of LH stimulation, and estradiol remained at basal levels for the remainder of the experimental period (Fig. 3). This pattern of secretion was directly analogous to that observed following a normal pre-ovulatory LH surge [53]. In contrast, in animals that had not formally received LH, estradiol secretion increased from basal levels prior to LH/hCG to 0.55 ± 0.12 nmol/l within 12 h of treatment and continued to increase to a level of 1.65 ± 0.43 nmol/l by the end of the experimental period (Fig. 3; P < 0.05).
Inhibin A secretion. The pattern of LH stimulation applied had a marked effect on the pattern of ovarian inhibin A secretion during the infusion period (Fig. 3). Prior to treatment, inhibin A concentrations did not vary between groups but increased significantly (P < 0.001) during the period of LH treatment, regardless of mode of administration, to a peak that occurred at the time of ovulatory stimulus and declined thereafter (Fig. 3). Again, this pattern of secretion was directly analogous to that previously observed during the normal follicular phase [52]. In contrast, inhibin A secretion in ewes receiving no LH remained constant throughout the experimental period and concentrations were significantly lower than LH treated groups in the 66-h period between PG and LH/hCG administration (P < 0.05).
Ovulatory follicle development and formation of a corpus luteum. At the time of PG, the ovaries of all animals contained on average 2.5 ± 0.5 large antral follicles greater than 3.5 mm in diameter. Ultrasound revealed that the mean diameter of ovulatory follicles at the time of luteal regression was 4.5 ± 0.4, 4.3 ± 0.3, and 4.1 ± 0.3 mm for no LH, pulsed LH, and infused LH, respectively. At the time of administration of the ovulatory stimulus at 66 h after PG, ovulatory follicle diameter had increased to 4.9 ± 0.3, 5.1 ± 0.4, and 5.2 ± 0.3 mm for no LH, pulsed LH, and infused LH, respectively, but this increase was not statistically significant (and there was no effect of mode of LH stimulation). In response to the ovulatory stimulus, all animals formed a morphologically normal corpus luteum. A feature of the ovarian autotransplant is that its position under the skin precludes rupture of ovulatory follicles and these will typically appear on ultrasound during the early luteal phase as large fluid-filled bodies whose central cavity gradually becomes filled with echogenic tissue as the follicular cells luteinize, finally forming a characteristic CL that consists of a small central cavity surrounded by luteal tissue that is easily distinguishable from the surrounding stromal tissue. This increase was observed in animals from all groups and did not vary according to mode of LH stimulation (No LH, 11.7 ± 0.6; Pulsed LH, 12.4 ± 0.6; Infused LH, 12.3 ± 0.7). Ovulation rates in the three experimental treatment groups did not differ (No LH, 2.4 ± 0.3; Pulsed LH, 2.4 ± 0.2; Infused LH, 2.5 ± 0.3).
The formation of corpora lutea, visualized by ultrasound scans, was accompanied by initially normal increases in peripheral progesterone concentrations but there were differences in the profiles obtained according to the pattern of LH stimulation applied to the ovulatory follicle (Fig. 4). Progesterone profiles in animals that received LH were similar but in ewes that received no LH progesterone profiles deviated on Day 5 of the luteal phase (Fig. 4) so that by the end of the sampling period, progesterone concentrations differed (P < 0.05) between the no LH (5.01 ± 1.05 nmol/l) and LH treated groups (Pulsatile LH 12.45 ± 2.57 nmol/l; Constant LH 14.78 ± 2.56 nmol/l).
|
Gonadotropins in peripheral plasma. The two different rates of constant LH infusion resulted in constant levels of LH in venous plasma, with the high dose resulting in an overall mean level of 1.44 ± 0.12 µg/l, which was significantly (P < 0.05) higher than the low-dose level (1.02 ± 0.10). The level of LH stimulation applied had no effect on the pattern of FSH secretion during the infusion period, with a marked decline in FSH values (P < 0.001) in both treatment groups during the infusion period that was identical to that observed in Experiment 1 (in groups receiving LH) to a value that was 42 ± 9% and 38 ± 7% of the concentration at the time of luteal regression in the low and high groups respectively (data not shown).
Ovarian estradiol and inhibin A secretion. The level of LH stimulation applied had a marked effect on the level of ovarian estradiol secretion during the infusion period (Fig. 5) with the high dose group having significantly higher (P < 0.05) estradiol concentrations during the infusion period. In both groups there was a highly significant increase in ovarian estradiol secretion over the period of the LH infusion from a common baseline that peaked at the time of hCG/LH administration (P < 0.001). At this time estradiol concentrations were approximately 2-fold higher in the high LH dose group (Low 0.84 ± 0.26 vs. High 1.46 ± 0.31 nmol/l; P < 0.05). This difference was also evident from the period of intensive blood sampling, in the period 35.5–39.5 h after PG, which revealed constant patterns of secretion in both groups with overall mean levels of secretion of 0.37 ± 0.10 and 0.83 ± 0.10 in low- and high-dose groups, respectively (Fig. 5; P < 0.05). Administration of an ovulatory dose of LH/hCG at 60 h following PG resulted in an acute fall in ovarian estradiol secretion in ewes receiving both levels of LH stimulation and remained at basal levels for the remainder of the experimental period (Fig. 5).
|
The dose of LH infused in Experiment 2 had no effect on the pattern or level of inhibin A secretion which was similar to that observed in Experiment 1 with a 3-fold increase (P < 0.001) in inhibin A secretion across the period of infusion to a peak that occurred at the time of ovulatory stimulus and declined thereafter (data not shown). There was no significant difference in the level of inhibin A secretion in ewes stimulated with different doses of LH.
Ovulatory follicle development and formation of a corpus luteum. At the time of PG, the ovaries of all animals contained on average 3.1 ± 0.5 large antral follicles greater than 3.5 mm in diameter. Ultrasound revealed that the mean diameter of ovulatory follicles at the time of luteal regression was 4.0 ± 0.4 and 3.9 ± 0.5 mm for the low and high LH groups respectively. At the time of administration of the ovulatory stimulus at 60 h after PG, ovulatory follicle diameter had increased (P < 0.05) to 5.1 ± 0.3 and 5.0 ± 0.4 mm for the low and high LH groups respectively. In response to the ovulatory stimulus, all animals formed a morphologically normal corpus luteum, with diameters of 11.2 ± 0.4 and 11.4 ± 0.5 mm for the low and high LH groups, respectively, and there was no difference in progesterone profiles (data not shown). Ovulation rates in the two experimental treatment groups did not differ (Low LH 2.5 ± 0.4 vs. High LH 2.4 ± 0.3).
The results of the present experiments suggest that LH is an essential requirement for normal ovulatory follicle development and subsequent luteal function and show that a pulsatile mode of LH stimulation is not required by the ovulatory follicle. Further, the results illustrate clearly how LH, by modulating both estradiol and inhibin A secretion by the ovulatory follicle, can indirectly control the level of pituitary FSH release and hence the fate of FSH-dependent follicles. Finally, the results of Experiment 2 indicate, within the range tested, that the amount of estradiol secreted by the ovulatory follicle in response to a constant level of stimulation is directly related to the level of LH stimulation so that doubling the amount of LH infused results in a 2-fold increase in ovarian estradiol secretion in the absence of any change in follicle number.
The endocrine and ovarian responses observed to GnRH-antagonist treatment administered at the time of luteal regression observed in the present experiment agreed with those determined previously [47], with an acute depression in pulsatile LH stimulation but with little short-term effect on circulating FSH concentrations. Our previous work with several different types of GnRH-antagonist have shown that FSH concentrations do not decline until 7–10 days after treatment [47; B.K. Campbell unpublished], and this differential response can be attributed to marked differences in the mechanisms controlling LH and FSH synthesis, storage, and release in this species [55]. In the present experiment, we have exploited this window of differential gonadotropin secretion to show that absence of LH stimulation for a 60-h period, when compared to ewes who received (pulsatile or constant) LH, resulted in unchanged FSH concentrations, a profound suppression in ovarian estradiol secretion, an abnormal pattern of ovarian inhibin A secretion, and subsequent inadequate luteal function after ovulation. As indicated in the introduction to this paper, the essential role of LH in stimulating estradiol secretion is well established but its effect on inhibin A secretion was unexpected, as previous findings have indicated that FSH stimulates ovarian inhibin secretion in vivo [56] and that inhibin A production by cultured granulose cells is FSH dependent [57]. Although we [52] and others [58] have previously shown that inhibin A concentrations increase during the follicular phase of the cycle and during the development of the first follicular wave [52], this pattern of secretion had previously been regarded as a consequence of FSH-primed pre-ovulatory follicle development rather than being directly influenced by LH. However, this finding is in line with that of LaPolt et al. [59], who showed that inhibin subunit mRNA expression was increased by LH in FSH-primed rat granulosa cells [59].
In ewes receiving LH, regardless of mode of delivery, the marked increase in ovarian estradiol and inhibin A secretion is the most likely explanation for the decline in jugular venous FSH concentrations in the period after luteal regression, as none of these changes were observed in ewes that received no LH (Fig. 3). Further, FSH concentrations increased in LH-treated ewes when estradiol and inhibin A secretion fell following delivery of an ovulatory stimulus. The continued development and normal ovulation of large antral follicles present at the time of luteal regression in LH-treated ewes, despite profoundly suppressed FSH levels, is consistent with our previous findings that ovulatory follicles can transfer their gonadotropic requirements from FSH to LH [17]. In contrast, in ewes that received no LH, FSH concentrations were maintained in the normal physiological range and it is likely that this FSH was able to support the large antral follicles present in the ovary at the time of luteal regression over the following 60 h when no pulsatile LH support was provided. However, although these follicles appeared morphologically normal on ultrasound and were able to form a morphologically normal corpus luteum, their endocrine response to an ovulatory stimulus (Fig. 3) and ability to secrete progesterone in the early luteal phase (Fig. 4) was abnormal. Inadequate luteal function has been associated previously with the follicular phase pattern of LH stimulation in sheep [33], but in this instance high-frequency, high-amplitude pulses of GnRH were observed to induce this phenomenon that was subsequently attributed to perturbations in the release of luteolytic signals by the uterus [60]. In the present instance, lack of LH has been observed to induce inadequate luteal function and as the ovarian autotransplant involves isolation of the ovary from any luteolytic signals generated by the uterus [61], these effects can only be attributed to the normality of the ovulatory follicle. This conclusion is also supported by recent data showing that physiological pulsatile GnRH ovarian stimulation regimes induce normal luteal function [62]. Therefore, the results of the present experiment indicate that LH stimulation is an essential requirement for normal ovulatory follicle development and subsequent luteal function in sheep.
The profound nature of the effects observed in the GnRH-antagonist treated ewes that received no LH were surprising given the fact that immuno-active LH levels remained above 1 ng/ml throughout the treatment window and that constant LH stimulation resulted in an increase of only 24% in circulating LH concentrations. Although it is possible that a proportion of this immuno-active LH is not bioactive, it is also possible that these effects of LH are dependent on achieving certain threshold concentrations in the peripheral circulation. The phenomenon of a threshold concentration has been well established in terms of the ovarian response to circulating FSH levels [63] and significantly, changes of a similar magnitude to those observed for LH have long been known to represent biologically significant changes in the stimulation of ovarian follicle development by FSH [13, 63]. The fact that the effects of LH observed in the present experiment were determined under conditions of normal FSH that were around threshold levels is an important distinction in terms of the results we have obtained. We [10, 17] have previously shown little effect of LH on FSH-stimulated ovulatory follicle development and similar findings have been made when comparing the efficacy of pure FSH and preparations containing both FSH and LH activity in ovarian stimulation regimes [64] although some workers have suggested that the variation in response is reduced with pure FSH-alone in cattle [65]. In all these instances, FSH levels have been maintained at values well above the threshold concentrations required to induce gonadotropin-dependant follicle development, and under these circumstances it appears likely that FSH can mask and essentially replace the ovulatory follicles' normal requirement for LH. However, it is pertinent to add that there is a growing body of evidence in human-assisted reproductive technology—where pure recombinant FSH has been in use for a number of years—that the developmental capacity of oocytes is improved in some types of patients in which ovulatory-sized follicle development has been stimulated by gonadotropin preparations containing both FSH and LH activity [64].
The second major objective of the present study was to examine the ovulatory follicles' requirement for a pulsatile mode of LH stimulation. In a normal estrous cycle, the pulsatile nature of LH stimulation is transformed from a low-frequency, high-amplitude mode of secretion under the influence of progesterone during the luteal phase, to a high-frequency, low-amplitude mode of secretion during the follicular phase in response to the fall in progesterone following luteal regression and the increase in estradiol secretion as the ovulatory follicle responds to this increase in LH drive [29]. In the present experiment, this transition was mimicked in the pulsatile group as previously reported [10] and compared with the same dose of LH delivered as a constant infusion. The results indicate unequivocally that the ovary is able to respond to a constant level of stimulation and ewes so treated exhibited a pattern of preovulatory estradiol and inhibin A secretion, follicle development, and ovulation that did not differ from that determined to a pulsatile mode of LH stimulation (Figs. 3 and 4). More frequent blood sampling revealed that unlike the normal pattern of pulsatile estradiol secretion [43] constant infusion resulted in a constant high level of estradiol secretion that was significantly higher than the overall level observed in response to pulsatile LH. This observation was significant, as it had been hypothesized that constant LH may result in a loss in ovarian sensitivity through down-regulation of LH receptors in a manner that is analogous to the down-regulation of GnRH receptors in gonadotrophs exposed to constant patterns of GnRH stimulation [66]. In Experiment 1, there was no evidence of such an effect and when this hypothesis was tested further in Experiment 2, by doubling the dose of LH infused, we observed a further doubling in the level of estradiol secretion.
Although the results of Experiments 1 and 2 are not directly comparable due to breed differences in the experimental animals, the results of both experiments show clearly that constant LH infusion over the range tested had no deleterious effects on ovarian estradiol secretion and that level of LH stimulation is directly related to the level of estradiol secretion. As detailed in the introduction, the main action of LH on steroid secretion is the stimulation of the conversion of cholesterol to pregnenolone through activation of the enzyme P450 side chain cleavage [38] although effects of LH on granulosa cell aromatase activity in the presence of excess androgen substrate have also been observed [67; B.K. Campbell unpublished]. Thus it is likely that the effects of LH observed on ovarian estradiol production are mediated by effects on both supply of androgen substrate for aromatization and direct effects on granulosa cell aromatase activity.
The results of the present experiment have also confirmed and extended our understanding of the mechanisms whereby LH can indirectly influence antral follicle development by controlling ovarian estradiol and inhibin A secretion (Fig. 3). Both these ovarian hormones have been shown to act synergistically in controlling pituitary FSH release [7] and accordingly high levels of LH-stimulation in the present experiment resulted in an increase in estradiol and inhibin A secretion, a depression in FSH, and ovulation of a normal number of follicles. These endocrine manipulations mimicked closely those observed during the normal follicular phase (Fig. 3; [51, 53]) and, as detailed above, are consistent with a mechanism whereby LH can support the continued development of FSH-primed ovulatory follicles with LH receptor on their granulosa cells whilst at the same time inducing regression of competing FSH-dependent follicles by reducing peripheral FSH to below threshold values.
While this scenario appears plausible for stages of the estrous cycle when LH pulse frequency is high, such as during the follicular phase and for the first follicular wave during the early luteal phase [50], the mechanisms controlling dominant follicle development during the mid and late luteal phases, when LH pulse frequency is low, have remained unclear. With the autotransplant model, we have previously shown that ovarian estradiol and inhibin A secretion during emergence of the second follicular wave of the luteal phase remains unchanged [51] and evidence for such a relationship in normal ewes [68, 69] and cattle [70], in which peripheral values for both hormones are low, remains equivocal. However, the results of the present experiment suggest that a threshold level of LH stimulation may exist and that the amount rather than the pattern of LH stimulation applied to the ovary is the critical determinant. However, aside from the experimental model developed in the present MS, in a normal animal the overall mean amount of LH supplied to the ovary does not depend on the basal LH level but rather is controlled by the hypo-pituitary axis through modulation of a pulsatile mode of secretion to increase pulse frequency during the follicular phase and by increasing the amount of LH released per pulse during the luteal phase. It is therefore possible that the low-frequency, high-amplitude pattern of LH secretion observed during the mid and late luteal phase of the estrous cycle provides sufficient LH support to modulate emergence of waves of dominant follicle development at these stages of the estrous cycle. This question is one which could be appropriately addressed using the short-term antagonist model employed in the current experiment.
In conclusion, the results of these experiments support both direct and indirect roles for LH in controlling ovulatory follicle development in sheep. The results of Experiment 1 suggest that, as with FSH, a threshold concentration of LH exists that is an essential requirement for normal ovulatory follicle development and subsequent luteal function. Further, the results of both Experiments 1 and 2 show that normal patterns of estradiol and inhibin A secretion, ovulatory follicle development, and ovulation can be induced by constant infusion of LH and that, therefore, a pulsatile mode of LH stimulation is not required. Rather, the amount of estradiol secreted depends on the total amount of LH secreted rather than its mode of administration. Finally, the results illustrate clearly how LH, by modulating both estradiol and inhibin A secretion by the ovulatory follicle, can indirectly control the level of pituitary FSH release and hence the fate of FSH-dependent follicles.
ACKNOWLEDGMENTS
We gratefully acknowledge the technical assistance of Ms. Linda Harkness, Mrs. Joan Docherty, Miss Marjorie Thomson, Mr. John Hogg and Mrs. Catherine Pincott-Allen. We thank the NIAMDD for the ovine gonadotropins and Asta Medical AG for the GnRH antagonist.
FOOTNOTES
1Supported by EEC Grant AIR3 CT 92-0232, MRC programme grant G8929853, and NRK was supported by BBSRC S11749. 
Correspondence: 2School of Human Development, University of Nottingham, Floor D East Block, Queen's Medical Centre, Nottingham NG7 2UH, United Kingdom. FAX: 44 115 823 0704; e-mail: bruce.campbell{at}nottingham.ac.uk
Received: 2 May 2006.
First decision: 15 June 2006.
Accepted: 8 December 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  R. Duggavathi, P.M. Bartlewski, D.M.W. Barrett, E. Bagu, and N.C. Rawlings Short- and Long-Term Effects of Unilateral Ovariectomy in Sheep: Causative Mechanisms Biol Reprod, March 1, 2008; 78(3): 490 - 496. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
