| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Physiology (C.A.S., A.I.T., A.J.T.), Monash University, Victoria 3800, Australia; Prince Henrys Institute of Medical Research (I.J.C.), Clayton 3187, Australia; and Reproductive Sciences Program (K.M.B., F.J.K.), University of Michigan, Ann Arbor, Michigan 48109
Address all correspondence and requests for reprints to: Dr. Alan J. Tilbrook, Department of Physiology, P.O. Box 13F, Monash University, Victoria 3800, Australia. E-mail: alan.tilbrook{at}med.monash.edu.au.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The mechanisms and mediators by which stress inhibits LH secretion are not known, but recent evidence in female sheep suggests that cortisol plays a significant role. Specifically, stress-like elevation in plasma cortisol levels inhibited LH pulses in ewes that were either ovariectomized (20, 21) or in the follicular phase of the estrous cycle (22, 23). In the former, the predominant effect of cortisol appeared to be a suppression of pituitary responsiveness to GnRH, rather than hypothalamic release of GnRH. Administration of cortisol to ovariectomized anestrous ewes to achieve plasma concentrations similar to those seen during psychosocial or immune/inflammatory stress caused reduction in LH pulse amplitude with no effect on the frequency of pulses or on any aspect of GnRH secretion (20). Moreover, in ovariectomized anestrous ewes in which endogenous GnRH secretion was blocked by estradiol, cortisol caused a reduced LH response to GnRH, suggesting a direct pituitary action (20). We now have provided evidence to suggest that the actions of cortisol to reduce pituitary responsiveness to GnRH in ovariectomized ewes are mediated via the type II glucocorticoid receptor (24). Similar studies have not been conducted in rams and there have been no formal sex comparisons of the effects of cortisol on the secretion of LH in sheep. Because there are sex differences in the effects of isolation and restraint stress on pulsatile LH secretion, we sought to determine whether there are similar sex differences in regard to the inhibitory effects of cortisol. We conducted two experiments to test the hypothesis that there are sex differences in the inhibitory effect of cortisol on pulsatile LH secretion and pituitary responsiveness to GnRH in gonadectomized sheep. First, we examined LH pulse parameters in gonadectomized ewes and rams treated with cortisol. Second, we used the hypothalamo-pituitary disconnected sheep model, where hypothalamic input to the pituitary is surgically removed (25). With appropriate pulsatile GnRH replacement, we were then able to examine direct pituitary effects of cortisol to suppress responsiveness to GnRH in both gonadectomized ewes and rams. We have used this model previously to demonstrate changes in pituitary responsiveness to GnRH in gonadectomized ewes and rams during isolation and restraint stress (26).
| Materials and Methods |
|---|
|
|
|---|
Experimental procedure
Experiment 1: effect of infusion with cortisol on LH secretion in gonadectomized ewes and rams.
Gonadectomized ewes (n = 5) and rams (n = 6) were infused (iv) with either saline, a low dose (250 µg/kg·h), or a high dose (500 µg/kg·h) of cortisol (Solucortef, HSA, Sydney, Australia) for 30 h using Graseby MS 16A syringe drivers (Smiths Medical Australia Pty Ltd, Gold Coast, Australia). The experimental procedure was conducted three times so that every animal received every treatment in a randomized order. Blood samples were collected at 10-min intervals during a 6-h pretreatment period (h 6 to 0) and throughout the first 6 h (h 06) and the final 6 h (h 2430) of the infusions. Plasma concentrations of LH were measured in all samples, and plasma concentrations of cortisol were measured in samples collected at half-hourly intervals.
Experiment 2: effect of infusion with cortisol on LH secretion in hypothalamo-pituitary disconnected gonadectomized ewes and rams.
Hypothalamo-pituitary disconnected gonadectomized ewes (n = 5) and rams (n = 6) were infused (iv) with either saline or cortisol (250 µg/kg·h) for 30 h using a cross-over design. Two replicates were conducted 7 d apart with two ewes and three rams in the first replicate and three ewes and three rams in the second replicate. The experiment was divided into three sampling periods (Fig. 1
) with an initial pretreatment period of 5 h when neither vehicle nor cortisol were administered. The second sampling period occurred during the first 6 h of the infusion of cortisol, and the third sampling period occurred during the last 5 h of this infusion.
|
RIAs
LH.
RIAs for LH were conducted according to a procedure previously described (28). In experiment 1, the sensitivity of the LH assay was 0.5 ng/ml (n = 13). The intraassay coefficient of variation was 12% at 3.5 ng/ml, 10% at 7.4 ng/ml, and 9.3% at 12 ng/ml. The interassay coefficient of variation was 8.8% at 3.9 ng/ml, 8.7% at 8.0 ng/ml, 7.6% at 12 ng/ml, and 14% at 26 ng/ml. In experiment 2, the sensitivity of the assay was 0.50 ng/ml (n = 6). The intraassay coefficient of variation was 7.0% at 3.3 ng/ml, 11% at 5.3 ng/ml, 7.5% at 6.5 ng/ml, and 9.9% at 17.7 ng/ml. The interassay coefficient of variation was 10.9% at 3.5 ng/ml, 11.3% at 6.0 ng/ml, 12.5% at 7.9 ng/ml, and 6.7% at 9.1 ng/ml.
Cortisol.
Cortisol RIAs were conducted according to a previously described protocol (29). In experiment 1, the sensitivity of the cortisol RIA was 2.0 ng/ml (n = 6). The intraassay coefficient of variation was 9.4% at 9.7 ng/ml and 7.3% at 117 ng/ml. The interassay coefficient of variation was 16% at 14 ng/ml and 1.8% at 125 ng/ml. In the second experiment, the mean sensitivity of the assay was 1.8 ± 0.8 ng/ml (n = 3). The intraassay coefficient of variation was 11% at 17.5 ng/ml and 12% at 122 ng/ml. The interassay coefficient of variation was 3.2% at 8.3 ng/ml and 2.1% at 65 ng/ml.
Statistical analyses
All data were statistically analyzed using repeated measures ANOVA. The within-subjects factors were the treatment given during the infusion (saline, low dose of cortisol or high dose of cortisol in experiment 1, and saline or cortisol in experiment 2) and the period of sampling (h 6 to 0, 16, or 2430 of the infusion in experiment 1 and h 6 to 0, 06, or 2430 during experiment 2), whereas the between-subjects factor was sex. The parameters of LH secretion analyzed in experiment 1 were the mean plasma concentrations of LH (nanograms per milliliter); the amplitude of LH pulses (nanograms per milliliter), calculated as the difference between the peak and preceding nadir of a LH pulse; the frequency of LH pulses, expressed as the number of LH pulses per hour; and total LH output, which was the product of the amplitude multiplied by the number of pulses per hour and expressed as nanograms per milliliter per hour. Pulses of LH were identified according to an accepted definition (30) as abrupt increases that were greater than the assay sensitivity, that exceeded the previous value by at least three times the SD of the previous value, and that were followed by a progressive decline at a rate consistent with the reported half-life of LH of 29 min (31). In experiment 2, the mean plasma concentrations of LH and the amplitude of LH pulses were statistically analyzed. The amplitude of LH pulses (nanograms per milliliter) were defined as the difference between the maximal concentration of LH that occurred after the delivery of an injection (iv) of GnRH and the concentration of the sample immediately before the commencement of the GnRH pulse (19).
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
Mean (±SEM) LH pulse amplitude
In ewes, the mean (±SEM) amplitude of LH pulses did not differ significantly between the pretreatment and infusion periods for the low dose of cortisol, but was significantly (P < 0.05) lower during the first 6 h of infusion of the high dose of cortisol than during the pretreatment period or last 6 h of infusion (Fig. 5
). Furthermore, for both doses of cortisol, the mean LH pulse amplitude was significantly (P < 0.05) lower during the first 6 h of cortisol infusion than the first 6 h of saline infusion. There was no significant difference between the mean LH pulse amplitude during the last 6 h of infusion of saline or cortisol in ewes. In rams, the mean (±SEM) amplitude of LH pulses was significantly (P < 0.05) reduced during the first 6 h of infusion of both doses of cortisol and there was a further reduction (P < 0.05) during the last 6 h of infusion of the low dose of cortisol (Fig. 5
). There was a trend for a further reduction in LH pulse amplitude during the last 6 h of infusion of the high dose of cortisol (Fig. 5
), but this was not statistically significant (P = 0.076). The mean LH pulse amplitude in rams during the first 6 h of infusion of the low dose of cortisol, and the first and last 6 h of infusion of the high dose of cortisol was significantly (P < 0.05) lower than for the corresponding periods of saline infusion. The mean (±SEM) amplitude of LH in rams was significantly (P < 0.05) lower in rams than in ewes during the last 6 h of infusion of both doses of cortisol (Fig. 5
).
Mean (±SEM) number of LH pulses per hour (LH pulse frequency)
In ewes, the mean (±SEM) number of LH pulses per hour was not reduced significantly during infusion of the low dose of cortisol, but was significantly (P < 0.05) reduced during the first and last 6 h of infusion of the high dose of cortisol (Fig. 6
). Moreover, the mean number of LH pulses per hour during the last 6 h of infusion of both doses of cortisol was significantly (P < 0.05) lower than during the last 6 h of saline infusion. In rams, there was a significant (P < 0.05) reduction in the mean number of LH pulses per hour during the last 6 h of infusion of both doses of cortisol (Fig. 6
). The mean number of LH pulses per hour in rams was significantly (P < 0.05) lower during the last 6 h of infusion of the high dose of cortisol than during the last 6 h of infusion of saline. The mean (±SEM) number of LH pulses per hour was significantly (P < 0.05) less in rams than in ewes during the first and last 6 h of infusion of both doses of cortisol (Fig. 6
).
Mean (±SEM) total output of LH (LH pulse amplitude multiplied by LH pulses per hour)
In ewes, there was a significant (P < 0.05) reduction in the mean (±SEM) total LH output (nanograms per milliliter per hour) during the first 6 h of infusion of the low dose of cortisol although, by the last 6 h of infusion, there was no difference to pretreatment (Fig. 7
). When ewes were infused with the high dose of cortisol the total LH output decreased significantly (P < 0.05) during the first 6 h of the infusion and remained lower than pretreatment throughout the infusion period (Fig. 7
). There was no difference between the first and last 6 h of infusion in total LH output in ewes at either dose. In rams, the total LH output did not differ significantly (P = 0.078) between pretreatment and the first 6 h of the infusion with the low dose of cortisol but the total LH output during the last 6 h was significantly (P < 0.05) reduced compared with pretreatment (Fig. 7
). The difference in total LH output in rams between the first and last 6 h of infusion of the low dose of cortisol was just significant (P = 0.05). For the high dose of cortisol, the total LH output in rams was significantly (P < 0.05) decreased after 6 h of infusion and there was no further significant reduction during the infusion (Fig. 7
). Mean (±SEM) total LH output was significantly (P < 0.05) lower in rams than in ewes during the last 6 h of infusion of both doses of cortisol (Fig. 7
).
Experiment 2: effect of infusion with cortisol on LH secretion in hypothalamo-pituitary disconnected gonadectomized ewes and rams
Cortisol.
The mean (±SEM) plasma concentrations of cortisol during the control periods on both experimental days were not significantly different to those during the infusion of saline (5.6 ± 0.5 ng/ml; Fig. 8
). During the cortisol infusion, the mean (±SEM) plasma concentration of cortisol achieved was 154 ± 19 ng/ml. There were no significant differences between the plasma concentrations of cortisol achieved in the first 6 h and the final 5 h of the infusion. The plasma concentrations of cortisol did not differ between ewes and rams.
|
|
| Discussion |
|---|
|
|
|---|
In experiment 2, we used the hypothalamo-pituitary disconnected sheep model to investigate pituitary actions of cortisol, and found no effect of infusion with cortisol on the amplitude of GnRH-induced LH pulses in gonadectomized ewes and rams. Because this model allows assessment of direct pituitary actions of exogenous treatments (19, 25), it seems reasonable to conclude that cortisol does not directly affect the function of the pituitary gonadotropes in either sex. Nevertheless, these data contrast to those showing that cortisol suppressed the amplitude of LH pulses induced by exogenous GnRH in ovariectomized ewes in which endogenous GnRH secretion was blocked by estradiol during the nonbreeding season (20) and our recent observation that cortisol can inhibit pituitary responsiveness to GnRH in cultures of dispersed ovine anterior pituitary cells (41). The reasons for these contrasting findings are not known, but there are a number of fundamental differences between these experimental approaches including treatment with estradiol, the stage of the circannual breeding cycle, the degree of stimulation of the pituitary by GnRH, and the presence or absence of neural input from the hypothalamus to the median eminence. The gonadectomized sheep in our experiments were not treated with steroids, raising the possibility that estradiol may affect the actions of cortisol. It is well accepted that gonadal steroids can influence the activity of the hypothalamo-pituitary adrenal axis (42), and we have found that sex steroids can influence the manner in which isolation and restraint stress impacts on the hypothalamo-pituitary-gonadal axis in sheep (14). Nevertheless, when ovariectomized ewes, untreated with ovarian steroids, were treated with cortisol during the nonbreeding season, there was a reduction in the amplitude of LH pulses without effect on the frequency of LH pulses or any aspect of the secretion of GnRH (20), which questions the likelihood that the presence of estradiol can account for the differences observed in the two model systems. It remains possible that estrogen treatment of ovariectomized hypothalamo-pituitary disconnected sheep (the model used in experiment 2) influences the effect of cortisol on GnRH-induced LH secretion. The previous experiments with ovariectomized ewes were conducted during the nonbreeding season (20), whereas the current experiments were conducted during the breeding season. We have found that there is a seasonal effect on the stress-induced inhibition of LH secretion in hypothalamo-pituitary disconnected sheep (26). It was recently shown that the cortisol-induced suppression of LH secretion in ovariectomized ewes is not influenced significantly by season (21); however, this has not been examined in hypothalamo-pituitary disconnected sheep. With respect to the degree of stimulatory input to the pituitary, the study in ovariectomized anestrous ewe used hourly injections of GnRH (21), whereas the GnRH injections were given every 2 h in experiment 2. We have shown that the degree of stimulation of the pituitary by GnRH can influence a different endocrine system in rams, that being the pituitary actions of inhibin to influence the secretion of LH (43). Nonetheless, this has not been explored in terms of the actions of cortisol. Finally, it is possible that neural input from the brain to the median eminence is necessary for cortisol to inhibit pituitary responsiveness to GnRH, which has been suggested previously (20, 24). Our use of hypothalamo-pituitary disconnection would have prevented this mechanism from operating because neural inputs from the hypothalamus to the external neurosecretory zone of the median eminence are removed with this procedure. In birds, a gonadotropin inhibitory peptide of central origin, named gonadoptropin-inhibitory hormone (GnIH), acts at the pituitary to reduce responsiveness to GnRH (44, 45). The central distribution and functional roles of GnIH have been identified in hamsters, rats, and mice (46), but it is not known whether GnIH, or a similar central factor, exists in sheep. Furthermore, work is clearly needed to elucidate sites and mechanisms by which cortisol can affect pituitary responsiveness to GnRH in sheep.
Decreases in the amplitude of LH pulses in ovariectomized ewes treated with cortisol in experiment 1 were expected based on previous studies (20, 21), and we now have extended this to show that cortisol also inhibits LH pulse amplitude in rams. Furthermore, LH pulse amplitude was inhibited to a greater extent in the rams than in the ewes, and this could be due to effects within the brain and/or pituitary gonadotropes. Previous evidence suggests that, in ovariectomized ewes, cortisol acts predominantly at the pituitary (20) via the type II glucorticoid receptor (24), but it is not clear that this is the case in rams. The results of experiment 2, where there was no effect of cortisol on LH concentration or pulse amplitude in response to injections of GnRH in hypothalamo-pituitary disconnected ewes and rams, would suggest that the key mechanism of action of cortisol to inhibit LH pulse amplitude in sheep is at the level of the brain to decrease the amplitude of GnRH pulses rather than at the level of the pituitary to inhibit responsiveness to GnRH. Nevertheless, this interpretation is not necessarily correct if neural inputs from the brain to the median eminence are required for the pituitary actions of cortisol, as discussed above, and this warrants further research. It is especially difficult to draw firm conclusions about the sites of action of cortisol in ewes because LH pulse amplitude was inhibited only at the high dose of cortisol, and the dose of cortisol used in experiment 2 was equivalent to the low dose in experiment 1. Although a higher dose of cortisol may have resulted in a decrease in the LH response to GnRH injections in experiment 2, we believe this to be unlikely because we have administered a much higher dose (1.4- to 2.5-fold higher plasma concentrations) than that in the current study to hypothalamo-pituitary disconnected ovariectomized ewes and rams without effect on LH response to GnRH injections (47). The low dose of cortisol used in these experiments gave plasma concentrations of cortisol in the range seen in response to immune/inflammatory stress (32) and observed in cortisol-treated ovariectomized ewes in which endogenous GnRH secretion was blocked by estradiol during the nonbreeding season (20). To assess fully whether there are sex differences in the actions of cortisol to suppress GnRH pulse amplitude, it will be necessary to measure GnRH in the hypophyseal portal blood in ewes and rams treated with cortisol. We also found a decrease in LH pulse frequency in both sexes infused with cortisol, indicating that GnRH secretion was reduced. Our results in ewes agree with the recent finding that cortisol treatment caused a minimal decrease in LH pulse frequency in gonadectomized ewes (21) and suggest that at least some of the inhibitory actions of cortisol on LH pulses in ovariectomized ewes and rams are due to hypothalamic actions on GnRH neurons. Nonetheless, our data suggest that there is a sex difference in the hypothalamic action of cortisol to suppress LH secretion with a greater effect in rams.
Because the animals were gonadectomized, it is feasible that the sex differences in the actions of cortisol to suppress pulsatile LH secretion in experiment 1 were due, in part at least, to organizational effects of the gonadal steroids to influence the sensitivity of the hypothalmo-pituitary gonadal axis to the actions of cortisol. The animals in this study were gonadectomized postpubertally and thus had normal exposure to gonadal steroids until at least 2 months before the experiment. Whether there are also activational actions of gonadal steroids on the responsiveness of the hypothalamo-pituitary unit to the actions of cortisol has not been determined in sheep although the mechanisms by which LH secretion is suppressed during isolation and restraint stress are clearly influenced by gonadal steroids (14). It is readily accepted that gonadal steroids influence the basal and stress-induced activity of the hypothalamo-pituitary adrenal axis in rodents (42) and we have shown that some of the differences in the hypothalamo-pituitary adrenal axes of rams and ewes are due to gonadal factors (48). Nonetheless, this research has not been extended to investigate the importance of gonadal steroids in influencing the sensitivity to the inhibitory actions of cortisol on LH secretion.
The results of experiment 1, where the frequency of LH pulses was decreased in castrated rams treated with cortisol in the absence of sex steroids differ from a study where administration of cortisol resulted in a reduction in the frequency of LH pulses in rams that had been castrated prepubertally and treated with estradiol but not in rams that were not treated with estradiol (33). If, as discussed above, prior exposure to gonadal steroids influences the sensitivity of the hypothalamo-pituitary gonadal axis to cortisol, the differences between these studies may have been due to the different times of castration as our animals were castrated postpubertally. It has been shown that the time of castration can influence the sensitivity of the hypothalamo-pituitary unit to the feedback effects of estradiol (49) and the regulation of the secretion of GnRH and LH (17) in rams but the influence of time of castration on the effects of cortisol to inhibit LH secretion are not known. Furthermore, the plasma concentrations of cortisol achieved in experiment 1 were almost double the concentrations in the study of Daley et al. (33), which may also explain the differences in findings between studies.
In conclusion, our data show that gonadectomized rams are more sensitive than gonadectomized ewes to the inhibitory effects of cortisol on LH secretion. Furthermore, there are sex differences in the specific effects of cortisol on LH pulses that may indicate sex differences in the mechanisms of action of cortisol to suppress LH in sheep. Nevertheless, cortisol can clearly inhibit LH pulse amplitude and frequency in both sexes, provided the dose is high enough. Although inhibition of LH pulse frequency is indicative of central actions of cortisol to inhibit GnRH secretion, it is not possible to determine whether reductions in LH pulse amplitude in this study were due to central actions to inhibit GnRH pulse amplitude and/or pituitary responsiveness to GnRH. Finally, the results of experiment 2 suggest that neural inputs from the hypothalamus to the median eminence may be necessary for cortisol to reduce pituitary responsiveness to GnRH, involving an as yet unidentified hypothalamic factor such as GnIH, although further research is necessary to explore this possibility.
| Acknowledgments |
|---|
| Footnotes |
|---|
Present address for I.J.C.: Department of Physiology, P.O. Box 13F, Monash University, Victoria 3800, Australia.
C.A.S., I.J.C., K.M.B., A.I.T., F.J.K., and A.J.T. have nothing to declare.
First Published Online September 7, 2006
Abbreviation: GnIH, Gonadoptropin-inhibitory hormone.
Received May 18, 2006.
Accepted for publication August 28, 2006.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. E. Oakley, K. M. Breen, A. J. Tilbrook, E. R. Wagenmaker, and F. J. Karsch Role of Estradiol in Cortisol-Induced Reduction of Luteinizing Hormone Pulse Frequency Endocrinology, June 1, 2009; 150(6): 2775 - 2782. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. R. Wagenmaker, K. M. Breen, A. E. Oakley, A. J. Tilbrook, and F. J. Karsch Psychosocial Stress Inhibits Amplitude of Gonadotropin-Releasing Hormone Pulses Independent of Cortisol Action on the Type II Glucocorticoid Receptor Endocrinology, February 1, 2009; 150(2): 762 - 769. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. E. Oakley, K. M. Breen, I. J. Clarke, F. J. Karsch, E. R. Wagenmaker, and A. J. Tilbrook Cortisol Reduces Gonadotropin-Releasing Hormone Pulse Frequency in Follicular Phase Ewes: Influence of Ovarian Steroids Endocrinology, January 1, 2009; 150(1): 341 - 349. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. M. Breen, T. L. Davis, L. C. Doro, T. M. Nett, A. E. Oakley, V. Padmanabhan, L. A. Rispoli, E. R. Wagenmaker, and F. J. Karsch Insight into the Neuroendocrine Site and Cellular Mechanism by which Cortisol Suppresses Pituitary Responsiveness to Gonadotropin-Releasing Hormone Endocrinology, February 1, 2008; 149(2): 767 - 773. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |