Answers are grounded in this paper and cite their sources.
THE IDEA that by activating the hypothalamic-pituitary-adrenal (HPA) axis, stress and overt or latent psychological disturbances have the potential of inhibiting the hypothalamic-pituitary-gonadal (HPG) axis has been circulated for a long time (1). It is generally thought that, if severe enough, this condition may lead to the suppression of the normal menstrual cycle in a syndrome referred to as functional hypothalamic amenorrhea or functional hypothalamic chronic anovulation (2). When fully established, the syndrome is characterized by ovarian quiescence, amenorrhea, and infertility. Although there appears to be a relationship between the type and severity of the stress and the proportion of women who develop amenorrhea (3), in practice it is difficult if not impossible to identify a threshold at which stress will interfere with the normal cycle. The probable reason for this is that in circumstances that are not life threatening the actual stressor may be less critical in determining the outcome than the individual’s response to it because psychobiological characteristics may heighten the responsiveness to the stressor (2). This particular problem may also in part reflect our ignorance of the initial steps by which stress interferes with normal cyclic events. One may assume that the degree of hypoestrogenism varies according to the severity of the stress challenge and that intermittent ovarian function may persist in less severe stress conditions. Potential differences in the individual responses to a particular stress paradigm make it difficult in clinical practice to confirm the independent association of a specific challenge with the initiation of the functional hypothalamic chronic anovulation syndrome (2). Furthermore, in many women more than one behavior may, in fact, be involved. For instance, this clinical syndrome has also been associated with other life style variables, such as weight loss or eating disorders, and certain types of excessive exercise (jogging, athletics) (4–6). Although it has been speculated that exercise- and weight-related amenorrhea is more probably caused by disturbances in the metabolic balance (7), mediators of the HPA axis may also be activated in these situations. Most investigators agree that the final neuroendocrine event responsible for the functional chronic anovulation syndrome is a decrease in the activity of the hypothalamic GnRH pulse generator. Reports in women with the established syndrome have usually demonstrated a significant slowing of LH pulse frequency, probably reflecting a decreased GnRH pulse activity (8). As proper folliculogenesis requires an optimal gonadotropin pulse regimen (9), it would be expected that a decrease in the GnRH-LH pulse frequency would lead to deficiencies in this process and to an abnormal menstrual cycle. Notwithstanding the universal recognition of a potential link between stress, and by implication an activated HPA axis, and an inhibition of HPG, data confirming this association in the human are scant and at best indirect. The most revealing data derive from a careful comparison of circadian cortisol secretion in patients with the functional chronic anovulation syndrome to that in eumenorrheic women or in women with other causes of anovulation. Only women with the functional chronic anovulation syndrome were found to be characterized by increased cortisol secretion, usually in the form of an amplified, but phase-intact, circadian excursion (8, 10). Studies in these women also reveal a blunted cortisol response to exogenous administration of CRH, suggesting that the increase in cortisol secretion may well reflect increased endogenous CRH activity (4, 11). Increased cortisol secretion has also been noticed in eating disorders or after exercise in many women. Furthermore, elevated levels of CRH in cerebrospinal fluid have been reported in amenorrheic women with anorexia nervosa (12), although the precise source of this neurohormone in lumbar CSF remains to be clarified. The preceding data suggest that in at least some amenorrheic patients, the endocrine HPA axis has been activated. Yet, there is presently no direct confirmation of a causal relationship between this phenomenon, and the suppression of the GnRH pulse generator and the HPG axis and the induction of the clinical syndrome. Furthermore, the independent association of stress, exercise, or eating disorder with the syndrome has up until now been impossible to confirm. However, it has been reported that although many of the graver stress situations where high rates of secondary amenorrhea occur are associated with malnutrition, in general the amenorrhea antedates the malnutrition (3). Demonstrating an independent causal association between stress and the HPG axis is a difficult task ahead. In the established syndrome, it may be impossible to trace back to the original stress challenge and to analyze the pathways involved; furthermore, in a chronic situation the response to stress may vary, and different neuroendocrine elements may become involved over time. Thus, initial studies should focus on the identification of relevant and reliable stress paradigms in the human so that prospective investigations on an HPA-HPG link can be instigated. The concept that physical and emotional stress activates central and peripheral responses that will preserve homeostasis has been well studied in animals. Classically, the main actors in this general adaptational response are believed to be the HPA endocrine axis and the autonomic nervous system (13). Activation of HPA involves the release of the neurohormone CRH. Although CRH neurons are scattered throughout several brain regions, one of the main area in the response to stress is thought to be located in the parvicellular aspects of the hypothalamic paraventricular nucleus (14). Central CRH release results both in the activation of the peripheral components of the HPA axis, leading to an increase in ACTH and cortisol, and in the activation of the sympathetic nervous system with increases in glucose release, heart rate, and blood pressure. Activation by stress of the locus ceruleus/autonomic system in the brain stem results in the stimulation of norepinephrine release from several central networks and enhanced arousal and anxiety. CRH may not be the only HPA neurohormone involved in the stress response; there is good evidence in the human and in animals that vasopressin of paraventricular origin is colocalized with CRH in perikarya and secretory granules and coreleased in stress (15, 16). Vasopressin is known to act synergistically with CRH as an ACTH secretagogue (17). The mechanisms that control the activity of the GnRH pulse generator are still chiefly under study, and only recently has evidence of a link and cross-talk between the neuroendocrine HPA axis and the GnRH pulse generator been obtained in the rodent and the nonhuman primate. The first line of evidence derives from the observation that the administration of CRH results in an immediate decrease in pulsatile GnRH and LH release (18, 19). Similar acute inhibitory effects on gonadotropin secretion after central administration of vasopressin, the second HPA neurohormone, have also been observed in the ovariectomized monkey (20). Although CRH administration obviously also activates the peripheral endocrine HPA axis resulting in ACTH and cortisol release, the acute inhibition of pulsatile GnRH activity by CRH in the ovariectomized monkey is clearly the result of a central action of CRH. Indeed, inhibition of gonadotropin release cannot be reproduced by a short term ACTH infusion and is still readily observed after CRH administration to the adrenalectomized monkey (21, 22). Unfortunately, efforts to demonstrate an inhibitory action of CRH on gonadotropin release in the human have not been successful (23, 24), except by one group of investigators (25, 26). The controversy perhaps reflects the facts that cerebroventricular injection of the neurohormone is not possible in the human and that at acceptable levels for treatment insufficient amounts of the compound may reach the proper central area after peripheral administration. This topic may have to be reassessed, however, taking into consideration the possibility that the gonadotropin response to CRH may vary according to the ovarian endocrine status. The relevance of CRH or vasopressin injection data in animals is difficult to interpret, because it is not possible presently to assess whether injected amounts of these compounds are physiological or pharmacological. A more direct line of evidence for an HPA-HPG link would require a demonstration that the neutralization of endogenous HPA neurohormone activity results in the disruption of the hypothesized HPA-HPG link. Such evidence has been obtained in the rat, in which the inhibitory effects of a stressful physical stimulus, such as that resulting from electrical foot shocks, on GnRH and gonadotropin secretion is prevented by cotreatment with a CRH antagonist or antibody (27). As mentioned above, the difficulty in studies in the primate is to identify a stress challenge (a stimulus that activates the HPA axis) that is relevant, quantifiable, and reproducible (i.e. in which individuality of the response is minimized), so that comparative experimental protocols can be developed. In initial studies in the nonhuman primate, our group has investigated the effects of an infectious-like stress challenge, such as that which follows the administration of the inflammatory cytokine interleukin-1 (IL-1) or of endotoxin, which is known to release endogenous cytokines (28). In the monkey, this challenge reliably produces an acute infectious-like syndrome and an activation of HPA, as judged by the increases in central CRH, ACTH, and cortisol release; it also results in the immediate suppression of pulsatile LH and FSH release (29, 30). A similar effect of endotoxin has been shown in the sheep (31). Significantly, in the monkey, but not in the rat, the inhibitory effect of IL-1 on LH is prevented by the central injection of a CRH antagonist (30, 32). In accord with the postulated role of paraventricular vasopressin, an antagonist to this neurohormone is equally effective in this regard (33). Thus, these data in the primate provide direct support to the concept that one of the mechanisms by which stress is inhibitory to the GnRH pulse generator is through the activation and the central release of neuroendocrine components of the HPA axis. Whether this conclusion is generally applicable or is limited to this particular stress challenge remains to be elucidated. A word of caution is in order here. There are several instances in the nonhuman primate indicating that an apparent activation of HPA, as determined by an increase in cortisol, does not necessarily translate into a suppression of gonadotropin secretion. For example, experimental data indicate that the increased central CRH activity that follows the reduced cortisol negative feedback induced by metyrapone does not result in an acute inhibition of LH secretion (34). Other data show that although there is frequently a correlation between activation of HPA and suppression of reproductive hormone secretion in instances of social stress, this link is not present in all individual animals (35). In a series of studies, no evidence could be provided that fasting-induced LH suppression in the male monkey is the result of the mild HPA activation observed in this instance (36). Clearly, further research is warranted in both the human and relevant animal species to fully elucidate what central or peripheral mechanisms are required to activate the postulated HPA-HPG link. It is also important to note here that, as expanded upon later in this review, the acute activation of HPA may, under certain circumstances, elicit an increase rather than a decrease in LH release, a result that, because unexpected, may have escaped notice in previous protocols. Unfortunately, a direct demonstration of the role of central CRH on LH secretion has not been possible to date in the human because CRH antagonists are rapidly degraded when given parentally and probably require central administration for effectiveness. However, experiments using μ-opiate receptor antagonists indicate that an increased endogenous opioid activity may somehow also account for the decreased pulsatility of the GnRH pulse generator in patients with functional hypothalamic chronic anovulation. Indeed, studies have shown that the administration of naloxone or naltrexone acutely restores normal LH pulse frequency, at least in a subgroup of these patients (37). The hypothalamic β-endorphin center, in fact, resides within the arcuate nucleus of the hypothalamus, which is also the location for the GnRH pulse generator in the primate. In animals, the acute inhibitory action of CRH on pulsatile LH release is also clearly prevented by naloxone or by an antiserum to β-endorphin (38–40). [Interestingly, in the monkey, αMSH, a derivative of the same prohormone giving rise to β-endorphin, can also antagonize the inhibitory effects of CRH or of IL-1 on pulsatile LH secretion (41). This suggests that the posttranslational processing of POMC may represent a step in the processes that control the response to stress.] As the animal studies suggest that increased endogenous opioid activity reflects, among many possible other causes, enhanced central CRH release and mediates the endocrine actions of CRH on the HPG axis, the above observation in the human may be viewed in support, if only indirectly, of the existence of a HPA-HPG link. After relevant and standardized stress paradigms are devised and tested for their effect on HPG, one potential complication is that the varying ovarian endocrine status throughout the menstrual cycle may modulate the response differentially. Thus, stress challenges should be tested at various stages of the cycle, and the responses compared. In this endeavor, investigators may have to be careful not to generalize data, because each particular stress challenge may well activate HPA through a different central pathway with its own particular sensitivity to the ovarian steroids. Experiments in the nonhuman primate serve to illustrate this aspect. As an example, although small amounts of estradiol appear to enhance the inhibitory effect of an immobilization or of a hypoglycemic stress on gonadotropin secretion (42, 43), they exert a protective action against the effect of an inflammatory-like stress challenge (44). In the rodent, there is some evidence that sex steroids may interact with both central and peripheral substrates of stress, thereby possibly modifying the HPA response. Such effects may reflect the presence of estrogen-responsive elements in the CRH gene promoter area and modulatory actions of estrogen on CRH gene expression (45). Thus, it may be that reported observations of slight gender differences in cortisol levels or of a sexual dimorphism in the immune/inflammatory reaction or in autoimmune disease may reflect direct stimulatory effects of estrogen on the CRH gene (46, 47). The latter may also explain the greater activity of CRH at proestrous in the rat (48) or the enhancement of cortisol release in estrogen-treated ovariectomized monkeys (49). Yet, it is important to emphasize at this that not all studies demonstrate a correlation between estrogen and HPA activity and in some instances a relationship has been The conclusion from a of the on this is the for comparative studies of the acute effects of well stress challenges on the HPG axis at various stages of the menstrual cycle. In this it is impossible to general for this Although the HPA-HPG link that activation of HPA will a in gonadotropin secretion, studies in our suggest that a outcome is possible under a endocrine Indeed, activation of HPA by IL-1 or endotoxin in the monkey the not the or the results in an acute release of LH This observation with the inhibitory effect of the cytokine on pulsatile LH secretion in the of estradiol a result more in with the A release of LH after HPA activation can also be in the ovariectomized monkey with to estradiol levels such a result is not to the nonhuman primate is by observations in women estrogen in a similar increase in LH in response to endotoxin administration between species may occur because in the rodent IL-1 an inhibitory effect on gonadotropin release at all stages of the cycle In the however, an increase in LH also observed after central CRH but only in animals In the monkey, our studies indicate that the that is probably responsible for this acute stimulatory effect of HPA on LH release may be as this effect is readily prevented by the administration of a antagonist have speculated that in stress the small but significant increase in that in response to HPA with estradiol to enhance LH secretion. In support of this is the observation that the increase in LH is prevented by the administration of a CRH that HPA activation is required for this effect to occur such an increase in LH is not observed when the cytokine is given in the probably reflects the that at estradiol cannot with the that an increase in LH after activation of HPA is an of one type of stress, it should be that this also reported in response to a of exercise in women the not the and in eumenorrheic after a these observations of a stimulatory effect of the HPA-HPG link a or could these be of relevance to the reproductive It is obviously to but one could that this response may a cycle or recognition of an acute stress reaction and perhaps represent a by which an acute stress stimulus, insufficient to the reproductive cycle, may interfere with normal cyclic Although there are no data in the to the effects of a increase in LH in the to there are several indicating that elevated LH at that of the cycle may the the the from a interfere with Although the long term of chronic stress amenorrhea, the initial effects of an activation of HPA on cyclic have not been initial studies in the monkey have with the effects of a infectious-like stress the of the cycle, a stress that produces mild The if by other studies, are of potential first because they show that a short term stress may exert effects on the cycle well its and second because it could be that this stress challenge, although endocrine effects to be readily in a clinical and insufficient to the menstrual cycle, may, in fact, be to interfere with cyclic disturbances occur in response to this stress challenge HPA activation at this of the cycle results in a of the that in a subgroup of monkeys well over the stress challenge also results in a reduced function in the form of secretion the in the cycle. cyclic may be of a more universal initial response to HPA activation and as such be part of the the initial in this syndrome. long and abnormal should be part of the of in the functional chronic anovulation syndrome (2). Although our own data in the monkey are presently limited to a type of stress, there have been observations of a of in women after the initiation of an exercise Although have this on neuroendocrine and endocrine HPA-HPG and responses may also have to be in observations have shown that CRH is present in several peripheral In the CRH and CRH are in and In some studies, CRH has been shown to by human in through mechanisms that the CRH and IL-1 and rat have been shown to the CRH and it has been speculated that CRH may also in physiological in that The treatment of patients with functional hypothalamic chronic anovulation is and the of this It may taking of the variables, of the hypoestrogenism to the HPG axis, and of infertility. regard to the pulsatile GnRH or gonadotropin are usually successful (2). as mentioned above, the injection of an antagonist may acutely pulsatile gonadotropin secretion results have been obtained in regard to a long term of this studies have reported successful rates in with but to demonstrate over treatment in women with hypothalamic amenorrhea for a different may be found in a of CRH antagonists that have been shown to exert a and to CRH antagonists may to be more than presently because of their to be of a specific condition in which a CRH antagonist may be effective would be to the increase in LH in the to that in a of if this increase were to reflect an acute stress response. In this the CRH antagonist may the small rise in that is probably responsible for the LH increase and for the normal of A CRH antagonist may also be in patients with the established functional hypothalamic amenorrhea syndrome to GnRH estrogen secretion, and a normal menstrual cycle. Whether with this antagonist would be successful in a well established chronic condition is difficult to aspects in chronic stress have not been investigated in the primate, and in the rodent suggest that HPA may become in the chronic stress For example, in the rat it is thought that vasopressin a more specific role in chronic stress by HPA responsiveness at a time at least in some stress the CRH response rapidly For instance, in of stress the proportion of CRH neurons increases and data suggest that there may be of an vasopressin response Vasopressin is effective in LH secretion in the primate, as above, but the of both chronic stress remains It may well be that a successful treatment may require the of as Although there is general that the functional hypothalamic amenorrhea syndrome is to stress and the suppression of the normal activity of the GnRH pulse the independent association between stress and the inhibition of the GnRH pulse generator remains to be demonstrated in the The challenge for the remains to identify relevant and reliable stress paradigms so that prospective investigations of an HPA-HPG link can be and endocrine responses can be activated research will be by the that different stress challenges may activate each to varying and that each may be by the presence of each ovarian In regard to the neuroendocrine response to stress, can from animal studies that both HPA CRH and vasopressin, and the endogenous opioid will a role in the inhibition by stress of the axis. the of CRH and vasopressin to GnRH and gonadotropin secretion and their of the effects of several types of stress challenges have been studies in the nonhuman primate of the effects of a short term stress on the menstrual cycle are of potential to the because they indicate that although a stress may be insufficient to amenorrhea, it may interfere with the normal cycle in and thereby normal studies have also a gonadotropin response to a stress challenge in the presence of such as the to resulting in an acute release of The most responsible for this stimulatory effect of HPA on LH release, at least in the acute may be by the in response to HPA activation Whether this an by which an acute stress stimulus, insufficient to the reproductive cycle, may interfere with the normal of folliculogenesis and with remains to be A of the central response to stress and of its inhibition of the HPG axis. A of the peripheral for the increase in gonadotropin in response to stress in the presence of The and for their critical of the and their
No takes yet. Share an insight, caveat, or question.
Michel Ferin (1999) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: