This study tested the hypothesis that central administration of corticotrophin-releasing hormone (CRH) and/or arginine vasopressin (AVP) will affect the secretion of LH in rams and that testosterone is necessary for these actions to occur. Plasma LH levels were measured in castrated rams during 1 h infusion of either 100 microliter vehicle/mock cerebrospinal fluid (CSF) or mock CSF containing 25 microgram CRH, 25 microgram AVP or 25 microgram of each peptide through guide cannulae into the third cerebral ventricle. These intracerebroventricular (i.c.v.) infusions were given to the castrated rams following injections (i.m.) each 12 h of oil or 8 mg testosterone propionate for 7 days. Blood samples were collected every 10 min for 4 h before i.c.v. infusion, during infusion and for 4 h following the infusion. Infusion of vehicle did not affect any endocrine parameters. In contrast, the plasma concentrations of LH and the amplitude of LH pulses were increased significantly during and following infusion of CRH, and this effect was not influenced by whether the castrated rams were treated with testosterone propionate or whether the CRH was administered in combination with AVP. Infusion of AVP alone did not affect LH secretion. The frequency of LH pulses and the plasma concentrations of FSH did not change with any of the i.c.v. treatments. The plasma concentrations of cortisol were significantly increased by CRH and AVP infusions. The plasma concentrations of cortisol achieved during and following i.c.v. infusion of CRH and AVP combined were greater than the concentrations achieved as a result of treatment with AVP alone but were similar to those with CRH. There was no effect of testosterone propionate on cortisol levels. These results show that CRH, but not AVP, is capable of acting either centrally or at the pituitary level to increase the secretion of LH in rams and these actions are not affected by testosterone. The stimulatory effects of CRH on LH secretion are to increase the amplitude of GnRH pulses and/or the responsiveness of the pituitary to the actions of GnRH with no effect on the frequency of GnRH pulses. The secretion of FSH in rams is not influenced by either CRH or AVP. The effect of CRH to increase LH pulse amplitude occurs in the face of increased cortisol levels, further reinforcing our belief that this adrenal steroid does not affect the reproductive axis in this species.

Prolonged stress is known to impair reproduction. It has been proposed that reproduction will also be impaired when a severe acute stress occurs during a period of elevated plasma concentrations of oestradiol, such as during the follicular phase of the oestrous cycle. In this experiment, we hypothesised that repeated acute and sustained elevation of cortisol would suppress the secretion of LH in ovariectomised pigs and that these effects would be enhanced in the presence of oestradiol negative feedback. Cortisol (or vehicle) was administered 12 hourly to ovariectomised pigs (n=6/treatment) for 8 days in the absence of oestradiol treatment and for a further 8 days during treatment with oestradiol. Vehicle was administered to 'control' pigs, 10 or 20 mg cortisol was administered i.v. to pigs to produce 'repeated acute' elevation of cortisol and 250 mg cortisol was administered i.m. to pigs to give a 'sustained' elevation of cortisol. Both before and during treatment with oestradiol, plasma concentrations of LH were monitored on the day before treatment, on the 4th and 8th days of treatment and following an i.v. injection of GnRH at the end of the 8th day of treatment. The repeated acute elevation of cortisol did not impair any parameters of LH secretion (i.e. mean plasma concentrations of LH, pulse amplitude or frequency, pre-LH pulse nadir or the LH response to GnRH) in the absence or in the presence of oestradiol. In contrast, when the elevation of cortisol was sustained, the mean plasma concentrations of LH and the pre-LH pulse nadir were significantly (P<0.05) lower on the 8th day of treatment than on the day before treatment and on the 4th day of treatment. Nevertheless, no other parameters of LH secretion were affected and these effects only occurred in the absence (not in the presence) of oestradiol. In conclusion, cortisol needed to be elevated for more than 4 days to impair the secretion of LH, and oestradiol did not enhance the impact of cortisol on LH secretion in ovariectomised pigs.

A dramatic late-gestation increase in fetal plasma cortisol concentrations is critical for the timing of parturition in the sheep. This increase appears to depend upon an intact hypothalamo-pituitary unit and is characterised by increasing responsiveness of the fetal adrenal gland to ACTH. ACTH has been postulated as the critical determinant of the late-gestation cortisol increase; however, recent evidence has suggested that other factors, including the ACTH precursor, pro-opiomelanocortin, may also be involved. To further define the role of ACTH in determining the timing of parturition and the responsiveness of the fetal adrenal gland, intact (INT/ACTH) and hypophysectomised (HX/ACTH) fetuses received a continuous infusion of ACTH(1-24) from the time of surgery (approximately 115 days gestational age (GA)) at a rate we have previously shown to generate normal fetal cortisol concentrations and term parturition in HX fetuses. A third group of saline-infused intact fetuses (INT/SAL) served as the control group. Adrenal responsiveness was assessed by cortisol responses to ACTH(1-24) challenges at 120, 130 and 140 days GA. There were no differences between the three groups of fetuses in the timing of parturition, the late-gestation increase in cortisol concentrations or the size of the adrenal cortex. In both INT/SAL and INT/ACTH fetuses, there were significant increases in basal immunoreactive-ACTH concentrations with advancing GA, although no such increase was observed in HX/ACTH fetuses. The proportion of total ACTH immunoreactivity present in low molecular weight (LMW) forms in INT/ACTH fetuses was greater than that in INT/SAL fetuses, while the level of LMW ACTH in HX/ACTH fetuses was intermediate. Both ACTH(1-24)-infused groups of fetuses had dramatically enhanced adrenal responsiveness to ACTH(1-24) at all GAs tested when compared with INT/SAL fetuses and there was a correlation (in rank order) between the proportion of LMW ACTH immunoreactivity and adrenal responsiveness. From these observations it appears that there is a separate regulation of adrenal responsiveness from basal cortisol concentrations and that an increase in basal cortisol concentrations can occur in the absence of an increase in basal ACTH concentrations. Furthermore, an increase in adrenal responsiveness does not appear to predict the timing of parturition nor basal cortisol concentrations. Taken together with previous studies it appears that ACTH plays an essential role in maintaining the growth of the fetal adrenal and enhancing its responsiveness, but a late-gestation increase in ACTH concentrations is not required to regulate basal cortisol concentrations or the timing of parturition.

There is a sex difference in the hypothalamo-pituitary-adrenal (HPA) axis of many species, although there are sparse data on the sheep. In the present study we have compared the HPA axes of intact and gonadectomised adult male and female sheep at the level of the median eminence, pituitary and adrenal glands using a variety of in vitro approaches. The concentration of arginine vasopressin (AVP) was higher (P<0.01) in the median eminence of male than female sheep, and was also elevated by gonadectomy of either sex (P<0.01). The concentration of corticotrophin-releasing factor (CRF) in the median eminence did not differ between the sexes, but was also elevated in both sexes following gonadectomy (P<0.01). Anterior pituitary pro-opiomelanocortin mRNA concentrations were higher (P<0.05) in intact male sheep than in intact females, with the levels in gonadectomised animals of both sexes being intermediate. In contrast to this finding, basal ACTH secretion from anterior pituitary cells was higher (P<0.05) in cultures derived from female sheep than those from males, but gonadectomy was without effect. There was no effect of sex or gonadectomy on in vitro ACTH secretion in response to AVP, CRF or the combination of AVP and CRF, and in all cases the combination of AVP and CRF generated greater (P<0.0001) ACTH secretion than AVP alone. AVP alone was more effective (P<0.01) than CRF alone as an ACTH secretagogue. The adrenal glands were larger (P<0.05) in female than male sheep, with no effect of gonadectomy. Basal cortisol production was greatest (P<0.05) in cultures of adrenal cells from intact male sheep, though ACTH- and 8BrcAMP-induced cortisol production was greater in the cultures of cells from females (P=0.05); there were no effects of gonadectomy. Cultures of adrenocortical cells from male sheep had greater (P<0.05) basal cAMP production, but ACTH-stimulated cAMP production did not differ between any of the groups of animals. These findings show a range of differences in the HPA axis of male and female sheep. Furthermore, they suggest that the heightened activity of the axis in the female occurs primarily due to differences at the level of the adrenal gland, and that greater adrenal responsiveness of female animals is due to differences in the latter stages of steroidogenesis, rather than an effect on ACTH signal transduction at its receptor.

A single intraperitoneal injection of lipopolysaccharide (LPS) causes a biphasic suppression of testicular steroidogenesis in adult rats, with inhibition at 6 h and 18-24 h after injection. The inhibition of steroidogenesis is independent of the reduction in circulating LH that also occurs after LPS treatment, indicating a direct effect of inflammation at the Leydig cell level. The relative contributions to this inhibition by intratesticular versus systemic responses to inflammation, including the adrenal glucocorticoids, was investigated in this study. Adult male Wistar rats (eight/group) received injections of LPS (0.1 mg/kg i.p.), dexamethasone (DEX; 50 microg/kg i.p.), LPS and DEX, or saline only (controls), and were killed 6 h, 18 h and 72 h later. Treatment with LPS stimulated body temperature and serum corticosterone levels measured 6 h later. Administration of DEX had no effect on body temperature, but suppressed serum corticosterone levels. At the dose used in this study, DEX alone had no effect on serum LH or testosterone at any time-point. Expression of mRNA for interleukin-1beta (IL-1beta), the principal inflammatory cytokine, was increased in both testis and liver of LPS-treated rats. Serum LH and testosterone levels were considerably reduced at 6 h and 18 h after LPS treatment, and had not completely recovered by 72 h. At 6 h after injection, DEX inhibited basal IL-1beta expression and the LPS-induced increase of IL-1beta mRNA levels in the liver, but had no effect on IL-1beta in the testis. The effects of DEX on IL-1beta levels in the liver were no longer evident by 18 h. In LPS-treated rats, DEX caused a significant reversal of the inhibition of serum LH and testosterone at 18 h, although not at 6 h or 72 h. Accordingly, DEX inhibited the systemic inflammatory response, but had no direct effect on either testicular steroidogenesis or intra-testicular inflammation, at the dose employed. These data suggest that the inhibition of Leydig cell steroidogenesis at 6 h after LPS injection, which was not prevented by co-administration of DEX, is most likely due to direct actions of LPS at the testicular level. In contrast, the later Leydig cell inhibition (at 18 h) may be attributable to extra-testicular effects of LPS, such as increased circulating inflammatory mediators or the release of endogenous glucocorticoids, that were inhibited by DEX treatment. These data indicate that the early and late phases of Leydig cell inhibition following LPS administration are due to separate mechanisms.

In this study we used an isolation/restraint stress to test the hypothesis that stress will affect the secretion of LH differently in gonadectomised rams and ewes treated with different combinations of sex steroids. Romney Marsh sheep were gonadectomised two weeks prior to these experiments. In the first experiment male and female sheep were treated with vehicle or different sex steroids for 7 days prior to the application of the isolation/restraint stress. Male sheep received either i.m. oil (control rams) or 6 mg testosterone propionate injections every 12 h. Female sheep were given empty s.c. implants (control ewes), or 2x1 cm s.c. implants containing oestradiol, or an intravaginal controlled internal drug release device containing 0.3 g progesterone, or the combination of oestradiol and progesterone. There were four animals in each group. On the day of application of the isolation/restraint stress, blood samples were collected every 10 min for 16 h for the subsequent measurement of plasma LH and cortisol concentrations. After 8 h the stress was applied for 4 h. Two weeks later, blood samples were collected for a further 16 h from the control rams and ewes, but on this day no stress was imposed. In the second experiment, separate control gonadectomised rams and ewes (n=4/group) were studied for 7 h on 3 consecutive days, when separate treatments were applied. On day 1, the animals received no treatment; on day 2, isolation/restraint stress was applied after 3 h; and on day 3, an i. v. injection of 2 microg/kg ACTH1-24 was given after 3 h. On each day, blood samples were collected every 10 min and the LH response to the i.v. injection of 500 ng GnRH administered after 5 h of sampling was measured. In Experiment 1, the secretion of LH was suppressed during isolation/restraint in all groups but the parameters of LH secretion (LH pulse frequency and amplitude) that were affected varied between groups. In control rams, LH pulse amplitude, and not frequency, was decreased during isolation/restraint whereas in rams treated with testosterone propionate the stressor reduced pulse frequency and not amplitude. In control ewes, isolation/restraint decreased LH pulse frequency but not amplitude. Isolation/restraint reduced both LH pulse frequency and amplitude in ewes treated with oestradiol, LH pulse frequency in ewes treated with progesterone and only LH pulse amplitude in ewes treated with both oestradiol and progesterone. There was no change in LH secretion during the day of no stress. Plasma concentrations of cortisol were higher during isolation/restraint than on the day of no stress. On the day of isolation/restraint maximal concentrations of cortisol were observed during the application of the stressor but there were no differences between groups in the magnitude of this response. In Experiment 2, isolation/restraint reduced the LH response to GnRH in rams but not ewes and ACTH reduced the LH response to GnRH both in rams and ewes. Our results show that the mechanism(s) by which isolation/restraint stress suppresses LH secretion in sheep is influenced by sex steroids. The predominance of particular sex steroids in the circulation may affect the extent to which stress inhibits the secretion of GnRH from the hypothalamus and/or the responsiveness of the pituitary gland to the actions of GnRH. There are also differences between the sexes in the effects of stress on LH secretion that are independent of the sex steroids.

To further understand the relative roles of the pituitary gland and ACTH in the regulation of mRNAs encoding proteins that are essential for adrenal development, we investigated the effects of, first, an ACTH infusion and labour in intact fetuses and, secondly, the effect of an ACTH infusion to fetuses with and without a pituitary gland, on the relative abundance of the mRNA encoding for the ACTH receptor (MC2R), steroidogenic factor 1 (SF-1), cholesterol side-chain cleavage enzyme (P450(scc)), 3beta-hydroxysteroid dehydrogenase (3betaHSD) and 17alpha-hydroxylase (P450(C17)) in the fetal adrenal gland. ACTH(1-24) infusion (14.7 pmol/kg per h) to intact fetuses was without effect on the abundance of mRNA encoding MC2R and SF-1, irrespective of whether the infusion was given for 18 (115-132 days of gestation) or 32 days (115 days to term (147 days of gestation)). Hypophysectomy (HX) did not alter the expression of MC2R mRNA; however, the abundance of SF-1 mRNA fell by approximately 50% following the removal of the pituitary gland. ACTH(1-24) infusion to HX fetuses failed to restore levels of SF-1 mRNA to that seen in intact animals. P450(scc) and 3betaHSD mRNAs were increased by ACTH(1-24) infusion for 18 days in intact animals, although no effects of the infusion were seen on P450(C17) mRNA levels. For all three of these mRNAs, there was a significant increase in their abundance between 132 days of gestation and term in intact fetuses. By term, ACTH(1-24) infusion was without any additional effect on their abundance. HX decreased the expression of P450(scc), 3betaHSD and P450(C17) mRNAs, while ACTH(1-24) infusion to HX fetuses increased the expression of these mRNAs to levels seen in intact animals. There were significant correlations between the abundance of the mRNA for P450(scc), 3betaHSD and P450(C17), but not MC2R and SF-1, and premortem plasma cortisol concentrations. These results emphasise the importance of the pituitary gland and ACTH in the regulation of the enzymes involved in adrenal steroidogenesis. Factors in addition to ACTH may also play some role, as the infusion was not always effective in increasing the abundance of the mRNAs. Surprisingly, the mRNA for MC2R and SF-1 did not appear to be regulated by ACTH in the late-gestation ovine fetus, though a pituitary-dependent factor may be involved in the regulation of SF-1 mRNA abundance.

There are sex differences in the response to stress and in the influence of stress on reproduction which may be due to gonadal steroids but the nature of these differences and the role of the gonads are not understood. We tested the hypotheses that sex and the presence/absence of gonads (gonadal status) will influence the cortisol response to injection of ACTH, insulin-induced hypoglycaemia and isolation/restraint stress, and that sex and gonadal status will influence the secretion of LH in response to isolation/restraint stress. Four groups of sheep were used in each of three experiments: gonad-intact rams, gonadectomised rams, gonad-intact ewes in the mid-luteal phase of the oestrous cycle and gonadectomised ewes. In Experiment 1 (n=4/group), jugular blood samples were collected every 10 min for 6 h; after 3 h, two animals in each group were injected (i.v.) with ACTH and the remaining two animals were injected (i.v.) with saline. Treatments were reversed 5 days later so that every animal received both treatments. Experiment 2 (n=4/group) used a similar schedule except that insulin was injected (i.v.) instead of ACTH. In Experiment 3 (n=5/group), blood samples were collected every 10 min for 16 h on a control day and again 2 weeks later when, after 8 h of sampling, all sheep were isolated and restrained for 8 h. Plasma cortisol was significantly (P<0.05) elevated following injection of ACTH or insulin and during isolation/restraint stress. There were no significant differences between the sexes in the cortisol response to ACTH. Rams had a greater (P<0.05) cortisol response to insulin-induced hypoglycaemia than ewes while ewes had a greater (P<0.05) cortisol response to isolation/restraint stress than rams. There was no effect of gonadal status on these parameters. Plasma LH was suppressed (P<0.05) in gonadectomised animals during isolation/restraint stress but was not affected in gonad-intact animals, and there were no differences between the sexes. Our results show that the sex that has the greater cortisol response to a stressor depends on the stressor imposed and that these sex differences are likely to be at the level of the hypothalamo-pituitary unit rather than at the adrenal gland. Since there was a sex difference in the cortisol response to isolation/restraint, the lack of a sex difference in the response of LH to this stress suggests that glucocorticoids are unlikely to be a major mediator of the stress-induced suppression of LH secretion.