Stress has long been known to inhibit the reproductive axis, and the work of Christian (1971) demonstrating infertility secondary to high population density is often cited as a seminal report. Shortly after the isolation and sequencing of CRH, it was demonstrated in rats that CRH inhibited LH secretion (Rivier and Vale 1984) and GnRH secretion (Petraglia et al. 1987), and further primate studies showed inhibition of LH secretion by injection of CRH (Olster and Ferin 1987).
While early studies used peripheral administration of high doses of CRH, subsequent studies demonstrated that intracerebrovascular administration of CRH demonstrated much greater potency and confirmed a central site of action of the inhibition, pointing to direct inhibition of GnRH by CRH (Gambacciani et al. 1986; Nikolarakis et al. 1986a, 1986b; Olster and Ferin 1987; Petraglia et al. 1987). However, the peripheral administration of CRH also demonstrated an opioid-mediated inhibition by CRH that could be abolished by dexamethasone pretreatment, suggesting a role for pituitary-derived opioids, most probably [^-endorphin from anterior pituitary corticotropes. Anatomical studies demonstrate that CRH neurons synapse with GnRH neurons (MacLusky et al. 1988); in vitro studies demonstrate that CRH can function as a secretagogue for (^-endorphin secretion from the arcuate ^-endorphin system (Nikolarakis et al. 1986a). Studies in primates by the Knobil laboratory (Williams et al. 1990) recording multiunit activity from the arcuate nucleus (i.e., the GnRH pulse generator) demonstrated that CRH administration induced inhibition of the rhythmic firing of the arcuate nucleus accompanying LH secretory pulses, as well as abolishing LH pulses. Studies with a CRH antagonist, R-helical CRHg-41, demonstrated the antagonist's ability to reverse stress-induced LH suppression in rats, confirming a central CRH-based mechanism by which stress inhibits LH secretion (Rivier et al. 1986). While the primate and rat studies have clearly pointed to CRH as the primary mechanism by which stress inhibits GnRH release, this is not true in all species (e.g., central CRH has no effect on GnRH or LH secretion in sheep [Tilbrook et al. 1999]), and some stressors act through cortisol (Debus et al. 2002). The demonstration of a central CRH effect on GnRH release does not preclude an effect of cortisol in both rats and primates, including humans.
So is there evidence that cortisol may also be involved in the inhibition of reproductive function? Several studies have demonstrated that ACTH administration reduces the increase in serum LH concentrations following ovariectomy or orchidectomy in rats (Mann et al. 1982; Schwartz and Justo 1977). This effect is dependent on the presence of the adrenal but could also involve adrenal production of gonadal steroids, which is regulated by ACTH (Putnam et al. 1991). Glucocorticoids also exert inhibitory effects on GnRH secretion or LH responsiveness to GnRH, including direct effects of cortisol on the gonadotrope (Suter and Schwartz 1985). Radovick et al. (1990) demonstrated a glucocorticoid-responsive element (GRE) on the GnRH gene, providing the potential for glucocorticoids to modulate GnRH gene expression. Diminished LH response to GnRH following long-term prednisolone treatment has been found in women (Sakakura et al. 1975). Patients with Cushing's disease, in which cortisol is increased but central CRH is likely to be low because of excessive glucocorticoid feedback on paraventricular nucleus of the hypothalamus CRH, show inhibition of LH secretion. Recent studies in ewes have found that 1) LH secretory amplitude is clearly inhibited by stress; 2) the effects of stress or endotoxin are reversed by metyrapone inhibition of cortisol synthesis; and 3) infusion of stress levels of cortisol can produce inhibition of LH pulse amplitude but not frequency, which is blocked by RU486, a glucocorticoid receptor antagonist (Breen et al. 2004; Debus et al. 2002). Finally, a recent study of exercise-induced reproductive abnormalities in adolescent girls concluded that "in active adolescents, increased cortisol concentration may. . . precede gonadotropin changes seen with higher levels of fitness" (Kasa-Vubu et al. 2004, p. 1). These data suggest that cortisol, in addition to central CRH, may also play a role in LH disruption.
Other studies in humans have linked hypothalamic-pituitary-gonadal (HPG) axis abnormalities to HPA axis activation. These include exercise-induced amenorrhea, anorexia nervosa, and hypothalamic amenorrhea. In all three syndromes, hypercortisolemia has been observed, indicating overactivity of the HPA axis (Berga et al. 1989; Casanueva et al. 1987; Hohtari et al. 1988; Loucks et al. 1989; Suh et al. 1988; Villanueva et al. 1986). In all three syndromes, CRH has been used as a challenge to evaluate pituitary and adrenal function. The response to exogenous CRH challenge demonstrates diminished ACTH or cortisol responses, suggesting that high baseline cortisol exerts negativefeedback effects on the hormonal responses to CRH (Berger et al. 1983; Biller et al. 1990; Gold et al. 1986; Hohtari et al. 1991). In anorexia nervosa, the hormonal abnormalities in both HPA and HPG axes are secondary to weight loss. Weight restriction and low body weight are also observed in exercise-induced amenorrhea, and low body weight has been reported in hypothalamic amenorrhea. Even relatively mild degrees of weight loss in normal-weight or obese subjects can lead to disturbances in both axes, as manifested by resistance to dexamethasone and by disturbances in menstrual regularity or amenorrhea (Berger et al. 1983; Edelstein et al. 1983; Pirke et al. 1985). Consequently, these three syndromes present with evidence of increased HPA axis activation and disrupted HPG functioning and amenorrhea. The disturbances in LH secretion in anorexia nervosa and hypothalamic amenorrhea have been evaluated primarily by examining the characteristics of LH pulsatile activity. In anorexia nervosa, LH secretory patterns may revert to prepubertal levels of low nonpulsatile secretion or to a pubertal pattern of entrainment of LH secretion to the sleep cycle. Studies by Reame et al. (1985) in women with hypothalamic amenorrhea demonstrated that LH secretion in the follicular phase is slowed to the rate normally observed during the luteal phase. In these individuals, LH and FSH responses to GnRH appear normal, indicating that the reduced pulse frequency is not secondary to pituitary changes but presumably due to changes in the GnRH pulse generator. Figure 7-2 summarizes the various levels at which hormones of the HPA axis may impinge on the reproductive axis. Despite suggestions that reproductive hormones may play a role in mood disorders, the HPG axis has received little examination in depression.
FIGURE 7-2. Effects of the hypothalamic-pituitary-adrenal axis on the hypothalamic-pituitary-gonadal axis.
Was this article helpful?