Review
Brain development under stress: Hypotheses of glucocorticoid actions revisited

https://doi.org/10.1016/j.neubiorev.2009.07.006Get rights and content

Abstract

One of the conundrums in today's stress research is why some individuals flourish and others perish under similar stressful conditions. It is recognized that this individual variability in adaptation to stress depends on the outcome of the interaction of genetic and cognitive/emotional inputs in which glucocorticoid hormones and receptors play a crucial role. Hence one approach towards understanding individual variation in stress coping is how glucocorticoid actions can change from protective to harmful. To address this question we focus on four hypotheses that are connected and not mutual exclusive. First, the classical Glucocorticoid Cascade Hypothesis, in which the inability to cope with chronic stress causes a vicious cycle of excess glucocorticoid and downregulation of glucocorticoid receptors (GR) in the hippocampus triggering a feed-forward cascade of degeneration and disease. Second, the Balance Hypothesis, which takes also the limbic mineralocorticoid receptors (MR) into account and proposes that an integral limbic MR:GR imbalance is causal to altered processing of information in circuits underlying fear, reward, social behaviour and resilience, dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis and impairment of behavioural adaptation. The MR:GR balance is altered by gene variants of these receptor complexes and experience-related factors, which can induce lasting epigenetic changes in the expression of these receptors. A particular potent epigenetic stimulus is the maternal environment which is fundamental for the Maternal Mediation Hypothesis. The outcome of perinatal gene × environment interaction, and thus of MR:GR-mediated functions depends however, on the degree of ‘matching’ with environmental demands in later life. The Predictive Adaptation Hypothesis therefore presents a conceptual framework to examine the role of glucocorticoids in understanding individual phenotypic differences in stress-related behaviours over the lifespan.

Introduction

Why are some individuals vulnerable to affective disorders like depression, anxiety and, post-traumatic stress disorder while others are resilient under similar stressful experiences? In response, it is thought that the mechanisms underlying these inter-individual differences in coping with stress depend on the secretion and action of stress hormones (e.g., corticotrophin releasing hormone (CRH), vasopressin, endorphins, adrenocorticotrophic hormone (ACTH), glucocorticoids and also adrenaline), which are largely shaped by gene–environment interactions throughout life. We will focus on the role of the hypothalamus–pituitary–adrenal (HPA) axis, notably its end products cortisol (human) and corticosterone (human and rodents), collectively termed CORT here. CORT does not act alone but operates in the context of the humoral and nervous signals that mediate the stress response (de Kloet et al., 2005, McEwen, 2007). CORT dampens the initial stress (defence) reactions to a given stressor, and prevents them from overshooting (Munck et al., 1984, Sapolsky et al., 2000). The actions exerted by CORT are as diverse as the variety of stressors and depends on the functionality of the mineralocorticoid (MR) and glucocorticoid receptors (GR) in the brain (de Kloet et al., 1998).

The two CORT receptor types, MR and GR, differ in ligand affinity and distribution throughout the brain. Both receptor types are abundantly expressed and colocalized in neurons from several areas of the limbic system including the hippocampus, amygdala and prefrontal cortex. MR has a 10-fold higher affinity (Kd: 0.5 nM) for CORT than GR (Kd: 5.0 nM) and is occupied to a large extent already at basal circulating levels of CORT (Reul and de Kloet, 1985). GR is additionally recruited with MR when CORT levels rise, such as during stress, at the circadian elevation prior to the activity period, and the ultradian secretory bursts occurring every hour (Windle et al., 1998). Together, MR and GR orchestrate neuroendocrine and behavioural responses to promote adaptation to stress from the initial appraisal of a novel experience up to memory storage and retrieval processes (Fig. 1) (de Kloet et al., 1998, Herman et al., 2003, McEwen, 2007). MR- and GR-mediated actions occur in concert with other stress mediators such as the corticotrophin, urocortin, sympathetic and parasympathetic systems.

MR and GR belong to the family of ligand-inducible transcription factors. Modification of the transcription of target genes, the so-called genomic action of CORT, is therefore most likely one of the main mechanisms underlying CORT actions in the brain. These genomic events are slow in nature as they can occur within a time-frame of approximately 15–30 min after receptor activation, and may last for less than 1 h up to days, depending on whether there is continuous exposure to the hormone or a single CORT pulse that is washed out after 20 min in experimental conditions (Dong et al., 1988, Jun et al., 1999, Morsink et al., 2006a, Morsink et al., 2006b). Classically, the two CORT receptor types (i.e., MR and GR) were viewed upon as being nuclear receptors, regulating genomic events underlying plasticity and energy metabolism. Their action mechanisms in transactivation and transrepression show profound differences: both receptors interact as dimers with the transcription machinery, but GR rather than MR monomers can modulate the action of other transcription factors that are activated by the initial stressor. Transrepression therefore underlies the dampening of the stress reaction by CORT. In addition, MR and GR recruit distinctly different co-regulator patterns, and their properties are differently affected by sumoylation and proteasomal activity (Meijer et al., 2006, Tirard et al., 2007).

The slow nuclear MR- and GR-mediated actions exert complementary effects in coping with the stressor. MR is important for the maintenance of neuronal integrity and stability in the face of the stress response. This can be easily demonstrated after adrenalectomy, which leads to an enhanced apoptosis in the hippocampal dentate gyrus, which can be restored by substitution with low amounts of CORT activating the MR (20 μg/ml; Nair et al., 2004, Stienstra et al., 1998). In mouse mutants with a site-specific knockout of MR, viability of neurons is also compromised (Gass et al., 2000). In the brain, GR prevents the initial stress reaction from overshooting, brings the target cell back to baseline, and facilitates recovery through promotion of energy metabolism. GR in cooperation with amine and neuropeptide signals facilitates encoding of the experience (see recent reviews de Quervain et al., 2009, Roozendaal et al., 2009).

Recently, a striking discovery has extended the ‘slow/nuclear’ mode of action of MR and GR to a ‘fast/membrane’ mode of action. On the cellular level, a rapid CORT-dependent enhancement of glutamate release in CA1 hippocampal neurons is accomplished through a non-genomic pathway involving membrane-located MR and GR. These membrane receptor variants of the nuclear receptors have a lower ligand affinity than those in the nucleus (Joels et al., 2008, Karst et al., 2005, Tasker et al., 2006). Rapid non-genomic behavioural effects of CORT have been reported on novelty-induced locomotor activity and aggressiveness in rats (Mikics et al., 2004, Sandi et al., 1996). These effects occur within minutes after CORT application and cannot be blocked by protein synthesis inhibitors, excluding a genomic mode of action.

The phases of the stress response (stress, recovery, adaptation; Fig. 1) are crucial for the processes underlying memory storage. Experience, problem solving, coping style, and response patterns are stored in memory for future use. It is well-established that CORT promotes memory consolidation through GR-mediated actions. Strong emotions that are also promoted by CORT are remembered best. At the same time behavioural responses that are of no more relevance are extinguished by CORT (Bohus and de Kloet, 1981, Oitzl and de Kloet, 1992). Moreover, incoming information that is not relevant to the current situation is suppressed (Pu et al., 2007). These actions are predominantly GR-dependent but also require the concomitant action of MR and other stress signals such as CRH, vasopressin, and the catecholamines. Whether MR-mediated actions in memory retrieval processes are accomplished by membrane and/or nuclear receptors remains to be established.

Stressors activate MR and GR, which result in up- or downregulation of gene expression patterns in the brain. MR and GR operate in balance and its disruption may compromise resilience and thus enhance vulnerability to diseases. Therefore, it has been suggested that the balanced activation of MR and GR is crucial for mental health (de Kloet et al., 2005).

CORT exerts feedback on targeted brain circuits which were initially activated by a given stressor (either real or imagined), that is usually novel in nature. It is well-known that if a novel situation is perceived as stressful, limbic-midbrain circuits are activated for immediate coping with the threat. Prominent in the brain's stress circuitry are the amygdala for regulation of emotional responses and the formation of strong emotional memories (McGaugh, 2004, Phelps and LeDoux, 2005), the hippocampus (which defines context in term of time and place) for learning and memory processes (Sanders et al., 2003), and frontal cortical regions for planning and control of actions. These brain structures interact to control emotions and cognitive processes and govern neuroendocrine responses (Quirk and Beer, 2006).

We propose that the MR:GR balance operates in time through activation of the membrane MR by rising levels of CORT (Fig. 1), which act to amplify initial stress reactions in a feed-forward fashion (Joels et al., 2006). At the same time containment of the initial stress reactions occurs via membrane and genomic GR-mediated feedback (Morsink et al., 2006b, Sapolsky et al., 2000) while the processing and storage (Morsink et al., 2006b, Oitzl et al., 2001, Reichardt et al., 1998) of the stressful experience in the memory is facilitated. If the control of this initial reaction as well as the management of the later adaptive phase becomes imbalanced, critical neurotransmitter and neuropeptide systems including serotonin, noradrenalin, dopamine, CRH, vasopressin and oxytocin may change their action patterns, which might in turn enhance the vulnerability to affective disorders.

With this timely knowledge on the action mechanism of CORT at hand we will review from a historical perspective theories that attempt to explain why some individuals flourish and others perish under similar stressful conditions. For this purpose we will first focus on the Glucocorticoid Cascade Hypothesis formulated at a time when GR was thought the sole mediator of CORT actions in the brain. The cascade hypothesis was quite influential until the existence of the high affinity “MR” was appreciated. Hence the Balance Hypothesis proposes that MR:GR balance is crucial for homeostasis and health. MR:GR balance depends on genetic and experience-related factors. The latter environmental influence is accounted for by the lasting epigenetic changes in the expression of MR:GR induced by early-life experience. Such epigenetic modulations are fundamental for the Maternal Mediation Hypothesis which takes the nature of mother–pup interaction into account. At last, the individual phenotypic differences in emotional and HPA reactivity as well as cognitive performance not only depend on the outcome of gene × experience-related interactions, but also depend on the actual conditions the individual is facing in later life, hence an integrated process as is articulated in the Predictive Adaptation or Mismatch Hypothesis.

Section snippets

The glucocorticoid cascade hypothesis of stress and aging

The glucocorticoid cascade hypothesis mostly has been discussed in the context of the aging process. The HPA axis promotes an organisms’ ability to cope with environmental stressors and facilitates behavioural adaptation (de Kloet et al., 2005; also called “allostasis” McEwen, 2003). If the stress response is inadequate or excessive and prolonged, the cost of reinstating homeostasis might become too high, which is a condition that is termed “allostatic load” (McEwen, 2003). At senescence,

MR and GR in behavioural adaptation

CORT actions are time- and context- dependent and therefore display an enormous diversity depending on the type of stressors and brain areas involved. This implies that the choice of experimental design and model system in which the effects of CORT are studied has profound consequences for the results. The characterization of the effects of CORT under different conditions such as, for example, exposure to psychosocial, restraint or mild novelty stress, will ultimately result in a more refined

Predictability of early-life experiences for later life

The perinatal period represents a unique period in ontogeny which is particularly sensitive to stress and where the fine-tuning of the stress-regulating systems and emotions can be permanently altered. Early-life adversities like maternal distress, maternal influenza during pregnancy, birth complications, loss of a parent, a depressed or addicted parent, exposure to family conflict and violence, and neglect or physical maltreatment (De Bellis, 2002, Heim et al., 2004, King et al., 2005) are

Stress and drugs: individual differences

Although in the centre of interest, little is known about the relation between early-life experiences, specifically maternal care and the development of drug addiction. Animals readily self-administer almost every drug abused by humans and drug self-administration in animals is highly predictive of abuse liability in humans. Like humans, animals express individual differences in drug intake (Piazza et al., 1989). In adult animals, stressors like food restriction, social defeat, electric foot

Predictive adaptation hypothesis

A survey of the literature provides strong support to the observation that developmental programming effects are associated with obvious detrimental outcomes later in life. This seems particularly valid for metabolic disorders programmed by an adverse intra-uterine environment, e.g., prenatal stress, nutrition deficiency (Barker, 2006, Barker, 2007). Persistent epigenetic differences have been recently shown to be associated with prenatal exposure to famine in humans (Heijmans et al., 2008).

Concluding remarks

In this essay we have linked brain development, stress and stress hormones with the biology of resilience and health. To that effect the balance hypothesis is fundamental. During stress and in the context of other signals, CORT promotes (via MR) the defence against the stressor, while GR-mediated effects facilitate processing of the stressor and storage of stressful events into memory. Enhanced vulnerability to disease is proposed to occur in case of imbalance of MR-activating and

Acknowledgements

The support by the Royal Netherlands Academy of Arts and Sciences, the Top Institute of Pharmaceutical Sciences T5 #209, NWO-DN95420 IRTG, Smartmix Senter Novem SSM06019 is gratefully acknowledged. We thank Mrs. Wendy Rodger for editorial assistance.

References (169)

  • J.M. Deminiere et al.

    Increased locomotor response to novelty and propensity to intravenous amphetamine self-administration in adult offspring of stressed mothers

    Brain Res.

    (1992)
  • E. Ferrari et al.

    Age-related changes of the adrenal secretory pattern: possible role in pathological brain aging

    Brain Res. Brain Res. Rev.

    (2001)
  • P.D. Gluckman et al.

    The fetal, neonatal, and infant environments-the long-term consequences for disease risk

    Early Hum. Dev.

    (2005)
  • J. Grootendorst et al.

    Reversal of cognitive deficit of apolipoprotein E knockout mice after repeated exposure to a common environmental experience

    Neuroscience

    (2001)
  • J. Grootendorst et al.

    Repeated exposure to rats has persistent genotype-dependent effects on learning and locomotor activity of apolipoprotein E knockout and C57Bl/6 mice

    Behav. Brain Res.

    (2001)
  • C. Henry et al.

    Prenatal stress in rats facilitates amphetamine-induced sensitization and induces long-lasting changes in dopamine receptors in the nucleus accumbens

    Brain Res.

    (1995)
  • J.P. Herman et al.

    Central mechanisms of stress integration: hierarchical circuitry controlling hypothalamo–pituitary–adrenocortical responsiveness

    Front. Neuroendocrinol.

    (2003)
  • I. Heuser et al.

    The role of mineralocorticoid receptors in the circadian activity of the human hypothalamus–pituitary–adrenal system: effect of age

    Neurobiol. Aging

    (2000)
  • M.C. Holmes et al.

    The role of 11beta-hydroxysteroid dehydrogenases in the brain

    Mol. Cell. Endocrinol.

    (2006)
  • M. Joels et al.

    The coming out of the brain mineralocorticoid receptor

    Trends Neurosci.

    (2008)
  • M. Joels et al.

    Chronic stress: Implications for neuronal morphology, function and neurogenesis

    Front. Neuroendocrinol.

    (2007)
  • M. Joels et al.

    Learning under stress: how does it work?

    Trends Cogn Sci.

    (2006)
  • S.E. Kawakami et al.

    Long maternal separation accelerates behavioural sensitization to ethanol in female, but not in male mice

    Behav. Brain Res.

    (2007)
  • T.A. Kosten et al.

    Enhanced acquisition of cocaine self-administration in adult rats with neonatal isolation stress experience

    Brain Res.

    (2000)
  • S. Levine

    The ontogeny of the hypothalamic–pituitary–adrenal axis. The influence of maternal factors

    Ann. N. Y. Acad. Sci.

    (1994)
  • S. Levine

    Influence of psychological variables on the activity of the hypothalamic–pituitary–adrenal axis

    Eur. J. Pharmacol.

    (2000)
  • S. Levine

    Primary social relationships influence the development of the hypothalamic–pituitary–adrenal axis in the rat

    Physiol. Behav.

    (2001)
  • S. Levine

    Developmental determinants of sensitivity and resistance to stress

    Psychoneuroendocrinology

    (2005)
  • L. Lu et al.

    Effect of environmental stressors on opiate and psychostimulant reinforcement, reinstatement and discrimination in rats: a review

    Neurosci. Biobehav. Rev.

    (2003)
  • S.J. Lupien et al.

    Stress hormones and human memory function across the lifespan

    Psychoneuroendocrinology

    (2005)
  • N. Marissal-Arvy et al.

    Gain of function mutation in the mineralocorticoid receptor of the Brown Norway rat

    J. Biol. Chem.

    (2004)
  • B.S. McEwen

    Mood disorders and allostatic load

    Biol. Psychiatry

    (2003)
  • K.J. McLaughlin et al.

    The effects of chronic stress on hippocampal morphology and function: an evaluation of chronic restraint paradigms

    Brain Res.

    (2007)
  • M.J. Meaney et al.

    Environmental regulation of the development of mesolimbic dopamine systems: a neurobiological mechanism for vulnerability to drug abuse?

    Psychoneuroendocrinology

    (2002)
  • M.J. Meaney et al.

    Epigenetic mechanisms of perinatal programming of hypothalamic–pituitary–adrenal function and health

    Trends Mol. Med.

    (2007)
  • O.C. Meijer et al.

    Steroid receptor coregulator diversity: what can it mean for the stressed brain?

    Neuroscience

    (2006)
  • E. Mikics et al.

    Genomic and non-genomic effects of glucocorticoids on aggressive behavior in male rats

    Psychoneuroendocrinology

    (2004)
  • M.C. Moffett et al.

    Maternal separation alters drug intake patterns in adulthood in rats

    Biochem. Pharmacol.

    (2007)
  • M.I. Morano et al.

    The role of the hippocampal mineralocorticoid and glucocorticoid receptors in the hypothalamo–pituitary–adrenal axis of the aged Fisher rat

    Mol. Cell. Neurosci.

    (1994)
  • S. Moriceau et al.

    Corticosterone controls the developmental emergence of fear and amygdala function to predator odors in infant rat pups

    Int. J. Dev. Neurosci.

    (2004)
  • F. Ambroggi et al.

    Stress and addiction: glucocorticoid receptor in dopaminoceptive neurons facilitates cocaine seeking

    Nat. Neurosci.

    (2009)
  • D.J. Barker

    Adult consequences of fetal growth restriction

    Clin. Obstet. Gynecol.

    (2006)
  • D.J. Barker

    The origins of the developmental origins theory

    J. Intern. Med.

    (2007)
  • G.A. Bonanno et al.

    The human capacity to thrive in the face of potential trauma

    Pediatrics

    (2008)
  • E. Borcel et al.

    Chronic stress in adulthood followed by intermittent stress impairs spatial memory and the survival of newborn hippocampal cells in aging animals: prevention by FGL, a peptide mimetic of neural cell adhesion molecule

    Behav. Pharmacol.

    (2008)
  • K.T. Brady et al.

    Treatment of patients comorbid for addiction and other psychiatric disorders

    Curr. Psychiatry Rep.

    (2007)
  • V. Brinks et al.

    Strain specific fear behaviour and glucocorticoid response to aversive events: modelling PTSD in mice

    Prog. Brain Res.

    (2008)
  • V. Brinks et al.

    Differential MR/GR activation in mice results in emotional states beneficial or impairing for cognition

    Neural Plast.

    (2007)
  • V. Brinks et al.

    Emotion and cognition in high and low stress sensitive mouse strains: a combined neuroendocrine and behavioral study in BALB/c and C57BL/6J mice

    Front Behav. Neurosci.

    (2007)
  • S. Cabib et al.

    Abolition and reversal of strain differences in behavioral responses to drugs of abuse after a brief experience

    Science

    (2000)
  • Cited by (288)

    View all citing articles on Scopus
    View full text