| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Physiology University of California San Francisco San Francisco, California 94143-0444
Address all correspondence and requests for reprints to: Mary F. Dallman, Department of Physiology, University of California San Francisco, San Francisco, California 94143-0444. E-mail: Dallman{at}itsa.ucsf.edu
 |
Introduction |
|---|
 Top Introduction References |
|---|
Between 1950 and the 1970s, when determination of the activity in the HPA axis depended on bioassays for ACTH and fairly insensitive assays for corticosterone, the neonatal HPA axis was thought to undergo a stress nonresponsive period (SNRP) between days 3–14 of life (1, 2). Although, at the end of gestation and during the first day of life, fetal and very young neonatal rats were shown to respond to stress with increased corticosterone, this responsivity vanished in subsequent days, not to return until after 2 weeks postpartum. Based on the lack of plasma and adrenal corticosterone responses to many different stressors after day 1 of life, the results suggested that the axis was absolutely unresponsive, and that, if there was any response to stress at all, it was very slow compared with that of adults (3).
In retrospect, this period of study pointed strongly to both insensitivity of the adrenal gland to stimulation by ACTH, and to the lack of amplification of corticosterone levels in plasma because of very low levels of the corticosteroid binding protein, transcortin (4). Subsequently, study of the adrenal capacity to synthesize and secrete glucocorticoids in response to ACTH has shown that in sheep, the fetal adrenal undergoes a period of unresponsiveness like that of the neonatal rat adrenal (5). The precise mechanism(s) of this damped adrenal responsivity to ACTH is still unknown, although in fetal sheep it appears to involve immaturity at the membrane receptor system (5). In infant rats, although insensitive to ACTH, there is a relatively high level of constitutive corticosterone secretion (6) that provides basal levels of nonrhythmic corticosterone in plasma allowing maturation, without excess, of developing corticosteroid-responsive systems. Although the mechanism of adrenal insensitivity to ACTH in neonatal rats is not solved, a very tight, positive correlation between the corticosterone response to ACTH and the peripheral-type benzodiazepine receptor has been demonstrated, suggesting that this cholesterol transporter may be a key to the stress-hyporesponsive adrenal (7). The periods of hyporesponsiveness appear to be highly protective for the organisms. Because elevated fetal corticosteroids appear to trigger parturition in sheep (5), and elevated corticosteroids result in harmful developmental advances in neonatal rats (8), banking of adrenal fires is a physiologically meaningful protective mechanism.
In the 1980s and 1990s, cracks appeared in the monolithic notion of a fetal or neonatal period of HPA unresponsiveness, and during this time, the phenomenon acquired the sobriquet of the "stress-hyporesponsive period" (SHRP) (9). During this period, there was a report that the capacity of CRF to respond to adrenalectomy or stress was highly limited ontogenetically, and that this could explain the SHRP (10). However, there was also evidence that this might not be the case. Adrenalectomy was shown to produce increases in ACTH that were of both normal amplitude and time-course, compared with adults (11); the characteristics of the adult HPA axis of corticosteroid feedback sensitivity (12) and facilitated responsivity to repeated stressors (13) were shown to be similar in neonates during the SHRP. A stunning example of full responsivity of ACTH secretion to stress during the SHRP of rats was provided by showing remarkably normal responses to injected endotoxin throughout this period (14).
Responsivity in the usual neural components of the HPA axis during the SHRP was far less certain. Although stimuli provoked evidence for increased neural activity in the hypothalamic paraventricular nuclei (PVN) (15), most reported that many of the stimuli shown to provoke elevated ACTH secretion in neonatal rats did not stimulate normal increases in CRF gene transcription as they do in adult rats. The paper by Dent and Levine (16) in this issue of Endocrinology now clarifies this problem and suggests strongly that not only do neonatal rats work in their own time domain, but, as in any good paper, also suggests that there are other mysteries about the rich neonatal HPA axis left to be plumbed.
After the relatively minor stimulus of ip injection of isotonic saline, Dent and Levine (16) show that significant responses occur in expression of both heteronuclear RNA (hnRNA) and messenger RNA (mRNA) for CRF in the PVN, as well as in plasma ACTH and corticosterone in rat pups exposed at 6, 12, or 18 days of life. At all ages, the authors show significant responses in all variables in pups that heteronuclear RNA (hnRNA) remained with their mothers until the time of injection stress. The results show clearly that, although the adrenal undergoes an SHRP, the rest of the HPA axis is entirely competent to respond during this time. This study of stress applied to rat pups taken from their mothers just before the "injection" finally puts the notion of a hyporesponsive period in the central components of the HPA axis in its final resting place.
Why does this study, and not others, find stress-responsiveness in hypothalamic CRF expression? In all likelihood this is because the authors, with their usual care and fine experimental design, performed a time course to measure the response that ranged from 15 min to 4 h after injection stress. Because changes in CRF mRNA content in adults usually do not occur at times less than 60 min after stress is applied, studies before this one had not looked at the earlier times. As Dent and Levine show, the responses in both CRF hnRNA and mRNA are greatest at 15 min, the earliest time measured in pups aged 6–12 days, and the levels are back to basal values by 30 min (16).
Although rapid, this time course still raises the question about what the changes in both hnRNA and mRNA of CRF after stress mean in the economy of the HPA axis. In both neonates and adults, plasma ACTH and corticosterone concentrations are elevated above 0-time within 5 min after injection of isotonic saline (e.g. 12). It appears that in both neonates and adults, responses in CRF hnRNA occur at about the same time as the hormonal responses; the responses in CRF mRNA to the stimulus are either synchronous with, or follow the hormonal secretory responses.
It stands to reason that when called upon acutely to secrete preformed CRF peptide, the neurons responsible must also adjust for this activity by replenishing peptide stores. It is this activity, perhaps stimulated by the same signal that depolarizes the neuron to cause secretion, that is measured as changes in synthetic activity, or CRF hnRNA. Because the synthetic and storage capacity of the CRF neuron appears to increase with ontogeny (10), temporal responses in CRF mRNA to peptide depletion would probably be faster in pups than adults because there is a smaller initial pool of CRF mRNA into which newly processed mRNA is added. It is also possible that the half-life of CRF mRNA increases as development proceeds, thus helping to account for brisker responses in pups.
Refilling CRF peptide stores in axon-terminals of the PVN neurons in the median eminence is essential after stress-induced depletion, to maintain adequate future responsivity in the HPA axis. It seems clear that the changes in CRF hnRNA and mRNA occurring after the application of stress must define subsequent reactivity of the HPA axis, rather than reflect a temporal sequence proceeding from excitation of CRF synthesis through release and excitation of successive components of the axis. An early estimate of the time it takes to replenish CRF peptide content in the median eminence after strong stimuli was approximately 80 min (17).
A counterintuitive result also emerges from the studies by Dent and Levine, that shows the plasticity of the HPA axis in pups (16). This is the diminished CRF responses of pups that had been removed from their mothers for 24 h before the stress of saline-injection, despite their augmented hormonal responses. Levine and his co-workers (18) had shown previously that depriving 6- and 12-day old pups of their mothers for up to 24 h results in marked amplification of ACTH and corticosterone responses to a new stress. These responses were also amplified after the stress of saline injection in the present study. However, unlike the results in pups that remained with their mothers, the maternally deprived pups had either absent or significantly smaller or delayed hypothalamic CRF responses, providing an apparent paradoxical dichotomy between central and peripheral responses of the HPA axis.
However, maternal deprivation for 24 h has been shown to persistently stimulate corticosterone secretion for the first 16 h of deprivation (19) and reduced CRF mRNA at 24 h compared with nondeprived controls (20). Therefore, this duration of deprivation appears to serve as a chronic stressor. In adult rats, a frequent response in the plastic HPA axis to sustained stimulation is induction of arginine vasopressin (AVP) expression and synthesis in CRF neurons (21). At the corticotrope cell, AVP potentiates the action of CRF (22), and the combination of CRF and AVP at the pituitary greatly amplifies ACTH secretion, compared with the effects of either peptide alone. Because regulation of the neonatal HPA axis in the Y2K now so strongly resembles that of the adult axis, it would not be surprising if the apparently paradoxical results in the deprived infants turned out to be a consequence of addition of AVP to the secretory mix from CRF neurons, as suggested by Dent and Levine (16).
Although forward regulation of the HPA axis in stressed neonates now appears to be fairly well understood, the presence and interactions of the mother with the neonate present very intriguing and important problems still to be solved. Providing some components of maternal behavior, such as stroking by the investigator of the maternally deprived pups restores to normal PVN c-Fos, CRF mRNA, and plasma ACTH (20) and CRF-R2 expression in the ventromedial hypothalamus (23). Compared with adults that were nondeprived as pups, adults that were deprived as pups exhibit marked alterations in HPA activity, neurochemistry, and behavior (24, 25). It seems that the next issue to solve in studies of this sensitive neonatal period, is the mechanism(s) by which maternal behaviors, or stroking and feeding of deprived pups by the investigator, both diverge and converge on regulation of the HPA axis in pups to program them into the adults, with the distinct behavioral, autonomic, and neuroendocrine repertoires, that result.
Received March 7, 2000.
|
References |
|---|
|
TopIntroductionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. C. Tang, K. G. Akers, B. C. Reeb, R. D. Romeo, and B. S. McEwen Programming social, cognitive, and neuroendocrine development by early exposure to novelty PNAS, October 17, 2006; 103(42): 15716 - 15721. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Boisse, A. Mouihate, S. Ellis, and Q. J. Pittman Long-Term Alterations in Neuroimmune Responses after Neonatal Exposure to Lipopolysaccharide J. Neurosci., May 26, 2004; 24(21): 4928 - 4934. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Tang, B. C. Reeb, R. D. Romeo, and B. S. McEwen Modification of Social Memory, Hypothalamic-Pituitary-Adrenal Axis, and Brain Asymmetry by Neonatal Novelty Exposure J. Neurosci., September 10, 2003; 23(23): 8254 - 8260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Tang Neonatal Exposure to Novel Environment Enhances Hippocampal-Dependent Memory Function During Infancy and Adulthood Learn. Mem., September 1, 2001; 8(5): 257 - 264. [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 |
