Elsevier

Physiology & Behavior

Volume 82, Issue 4, 30 September 2004, Pages 751-762
Physiology & Behavior

Stress induces CRF release in the paraventricular nucleus, and both CRF and GABA release in the amygdala

https://doi.org/10.1016/j.physbeh.2004.06.013Get rights and content

Abstract

In the hypothalamus, corticotropin-releasing factor (CRF) initiates the hypothalamic–pituitary–adrenal (HPA) axis response to stress, resulting in the release of glucocorticoids, including cortisol. Extrahypothalamic CRF, particularly in the limbic system, also appears to play a role in the stress response. To further define brain CRF response to stress, immunosensor-based microdialysis probes were used to measure the extracellular levels of CRF in the paraventricular nucleus of the hypothalamus (PVN) and in the amygdala of sheep during a predator (dog) exposure stress. In addition, gamma amino butyric acid (GABA) was measured in the amygdala and cortisol was measured in venous blood. Exposure to the predator stress increased CRF in the PVN and both CRF and GABA in the amygdala. These were followed in time by a rise in venous cortisol. Application of a CRF antagonist to the amygdala, immediately prior to stress, had a small effect on the subsequent observed stress responses. This treatment, however, significantly reduced the responses to a repeat stress administered 2 days later, compared to nontreated animals. Application of a GABA antagonist to the amygdala prior to stress had no effect on the subsequent observed stress response but increased the response to the stress repeated 2 days later. Perfusion with 4-aminopyridine, a neuronal depolarising agent, into the PVN induced a release of CRF accompanied shortly thereafter by a small increase in CRF in the amygdala, and 5–10 min later by an increase in venous cortisol. Perfusion into the amygdala increased the levels of both CRF and GABA but had no effect on either PVN CRF or venous cortisol. These data support roles for both the PVN and amygdala in stress responsiveness. It suggests further that actions at the amygdala can strongly influence subsequent responsiveness to a further stress, mediated in part by both CRF and GABA actions.

Introduction

Corticotropin-releasing factor (CRF), a 41 amino acid peptide, has a long-established functional role in stress reactivity, primarily at the hypothalamic level, controlling pituitary release of adrenocorticotropin hormone (ACTH), which in turn results in adrenal release of glucocorticoids, including cortisol [20], [42], [43], [44], [45]. This hypothalamic–pituitary–adrenal (HPA) axis is regarded as the fundamental and traditionally documented system of stress response [3]. CRF in the HPA arises from neurones with cell bodies in the paraventricular nucleus of the hypothalamus (PVN) and axon terminals in the median eminence [13], [14], [17]. Within the PVN itself, there are also short loop axons releasing CRF.

Beyond HPA regulation, CRF also appears important in other brain areas with widespread distribution of CRF immunoreactive terminals and binding sites through the brain. This includes areas such as the amygdala [25], [26], [27], hippocampus and the locus ceruleus [12]. These areas appear to contribute to various behavioural aspects of the stress response and may also be involved in feedback to the HPA. The amygdala has extensive connections with areas of the hypothalamus, including the PVN, and interactions seem highly likely in coordinating stress responses [28]. Similarly, there is a high extent of feedback from the hippocampus to the PVN via the medial hippocampal tract and lateral septum [18]. Work utilising a mouse gene knockout model [28] suggests that CRF receptors in the limbic system drive aspects of the stress response independently of the HPA axis and also strongly modulate HPA responsiveness.

New probe techniques have allowed rapid and direct measurements of both cortisol and CRF in different brain areas of animals behaving in both stressful and nonstressful environments [8], [9]. Similar measurement techniques for systemic hormones have also been used with some success [5], [6]. These probes are solid-state immunosensors in which current produced by the activation of a bound ligand conjugated to a redox species is measured at a set voltage. The bound ligand competes with the analyte of interest for binding sites, resulting in a current change inversely proportional to the presence of the analyte of interest. Rapid and frequent measurements offer the opportunity to longitudinally track CRF changes in the brain and their relationship to hormonal variation. This same method of approach has been applied successfully to monitoring amino acid neurotransmitters, including gamma amino butyric acid (GABA) [7].

Using these methods, amygdala CRF was demonstrated, in sheep, to increase with stress. It was also shown that circulating cortisol contributed feedback to the responsiveness of the amygdala to further stressors tested some time later [9]. The neurohormone actions of CRF at the level of the brain appear regulated by glucocorticoids in a complex and often indirect fashion [14], [37], [38]. Fast-rate sensitive feedback and delayed concentration level feedback have both been repeatedly demonstrated [4], [15], [39]. Stress exposure appears able to result in either a feedback inhibition or a feed-forward sensitisation towards a further stress [4], [15], [22], [38]. Corticosteroid feedback has been demonstrated at both the PVN and amygdala level and other brain areas also show CRF regulation by glucocorticoids [31], [34]. Adrenalectomy produces tonic hypersecretion of CRF in the locus cereleus, presumably due to a loss of inhibitory glucocorticoid tone [12] while prior stress exposure can sensitise locus cereleus neurones to CRF. Glucocorticoid feedback thus appears a critical variable in predicting subsequent stress responsiveness and may be involved in dysfunctional stress response in conditions, such as depression and anxiety [29], [30], [40]. Regulation is extremely complex and is still relatively poorly understood. As well as a circadian rhythm that can interact with stress responsiveness, cortisol secretion also shows pulsatile ultradian characteristics across time in both rodents [46], [47] and in sheep [16]. In rodents, these ultradian rhythms have a considerable influence on the magnitude of stress response [46], [47]. Simultaneous monitoring of CRF in a number of brain areas would assist in developing an overall model of stress control and feedback.

Numerous other transmitter systems also contribute to stress, including serotonin and catecholamines. The GABA system has been recognised as important for some time and traditionally has provided a model for many anxiolytic pharmacological developments [39]. Corticosterones have been demonstrated to regulate characteristics of GABA receptors in the hippocampus [31] and pharmacological antagonism of GABA in the PVN increases CRF secretion [1], suggesting GABA is involved in CRF control. GABA may also be important in such control at the amygdala level as changes in extracellular GABA in this brain area have been noted with conditioned fear stressors [42] and in cardiovascular responses to stress [41]. Understanding this overlay with CRF and glucocorticoids into a system of control would also be useful in constructing a model of brain–stress response.

The aims of this study were to measure extracellular CRF changes in both the PVN and amygdala simultaneously during exposure to a stress, and to also measure GABA activity in the amygdala. The purpose was to establish temporal relationships between these systems and to a systemic change in cortisol. A second purpose was to explore changes in these systems relative to repeat stress situations. To evaluate the potential for causal involvement, both GABA and CRF in the amygdala were manipulated pharmacologically immediately prior to stress impositions and any effect on both the immediate following stressor and a stressor 2 days later was examined.

Section snippets

Animals and surgery

Thirty-six female sheep (ewes) were used in the study. At the time of study, they were 18- to 20-month-old Romney crossbreds, with live weights between 37 and 42 kg. They were separated from a main flock 3 weeks prior to experimentation and run together until experimental completion, with ad libitum access to pasture and water.

One week prior to commencing experimentation, animals were anaesthetised with a combination of 2% (W/V) xylazine hydrochloride (Rompun, Bayer, Germany) and zolazepam

Results

Data are presented as mean±S.E. unless otherwise stated.

Discussion

This study demonstrates for the first time that, in sheep, PVN CRF levels change with a stressor and relate in time to other aspects of the stress response, including venous cortisol changes, and amygdala CRF responses. With imposition of a stressor, clear changes were seen in both amygdala and PVN CRF prior to changes in venous cortisol. Two discernible peaks were seen in the amygdala, one rapid and of short duration followed by a second peak that mirrored in time the rise in venous cortisol.

Acknowledgements

I acknowledge support from the Foundation for Research, Science and Technology, New Zealand and the New Zealand Royal Society Marsden Fund. This work was performed with animal ethics approval.

References (47)

  • S. Makino et al.

    Corticosterone effects on corticotropin-releasing hormone mRNA in the central nucleus of the amygdala and the parvocellular region of the paraventricular nucleus of the hypothalamus

    Brain Res.

    (1994)
  • Z. Merali et al.

    Differential impact of predator or immobilisation stessors or central corticotropin-releasing hormone and bombesin-like peptides in fast and slow seizing rats

    Brain Res.

    (2001)
  • T.P. Obrenovitch et al.

    Combined intracerebral microdialysis and electrophysiological recording: methodology and applications

    J. Neurosci. Methods

    (1993)
  • E.M. Pich et al.

    Corticotropin-releasing factor release from the mediobasal hypothalamus

    Neuroscience

    (1993)
  • R.M. Richter et al.

    Sensitization of cocaine-stimulated: increase in extracellular levels of corticotropin-releasing factor from the rat amygdala after repeated administrations as determined by intracranial microdialysis

    Neurosci. Lett.

    (1995)
  • J.B. Rosen et al.

    Glucocorticoid treatment increases the ability of CRH to induce seizures

    Neurosci. Lett.

    (1994)
  • J.J. Sandford et al.

    The psychobiology of anxiolytic drugs: Part 1. Basic neurobiology

    Pharmacol. Ther.

    (2000)
  • J. Schulkin et al.

    Induction of corticotropin-releasing hormone gene expression by glucocorticoids: implications for understanding the states of fear and anxiety and allostatic load

    Psychoneuroendocrinology

    (1998)
  • O. Stork et al.

    Reduction of extracellular GABA in the mouse amygdala during and following confrontation with a conditioned fear stimulus

    Neurosci. Lett.

    (2002)
  • W. Vale et al.

    Assay of corticotropin releasing factor

    Methods Enzymol.

    (1983)
  • V. Bartuanusz et al.

    Local gamma aminobutyric acid and glutamate control of hypophyseotropic corticotropin releasing factor neuron activity in the PVN of the hypothalamus

    Eur. J. Neurosci.

    (2004)
  • G.P. Chrousos

    The HPA axis and the stress response

    Endocr. Res.

    (2000)
  • C.J. Cook

    Basal and stress response cortisol levels and stress avoidance learning in sheep

    NZ Vet. J.

    (1996)
  • Cited by (102)

    • Measurement of neurotransmitters excreted in the urine of behaviorally healthy dogs in home and boarding kennel conditions

      2022, Journal of Veterinary Behavior
      Citation Excerpt :

      Although we are unaware of any studies analyzing urinary GABA and Glu as biomarkers of stress, we anticipated a difference in GABA and Glu based on previous rodent studies reporting a link between stress and these amino acids NTs. In several areas of the brain, GABA is triggered during a stressful event to provide inhibitory feedback on the neuroendocrine (e.g., glucocorticoid) effects of the biologic stress response (Krystal et al., 2002; Cook, 2004). Glucocorticoids trigger Glu release in the human brain, which likely influences cognitive and behavioral responses to stress (Popoli et al., 2011).

    • Glutamatergic lateral hypothalamus promotes defensive behaviors

      2020, Neuropharmacology
      Citation Excerpt :

      In this study we showed that glutamatergic LH projects to PVN, and activating terminals in PVN is sufficient to induce defense-related behaviors. PVN is one of major brain areas harboring CRF neurons which plays an important role in regulating hypothalamic–pituitary–adrenocortical axis in response to fear and stress (Cook, 2004). TMT exposure elevates corticosterone and adrenocorticotropic releasing hormone levels in blood plasma (Day et al., 2004) and increases the c-Fos expression in PVN CRF neurons (Myers and Rinaman, 2005).

    View all citing articles on Scopus
    View full text