Pathophysiology of affective disorders: functional interaction of stress hormones and hippocampal excitation

Abstract

An important new study by Kvarta, Bradbrook, Dantrassy, Bailey, and Thompson (J Neurophysiol 114: 1713–1724, 2015) examined the effects of persistent stress and excessive glucocorticoid levels on hippocampal function and emotional behavior in rodents. The authors specifically implicate the temporoammonic pathway as being susceptible to reductions in excitatory function in the context of chronic stress. We discuss the importance of this new finding in the broader context of medication development for major depressive disorder.

major depressive disorder (MDD), one of the most common affective (or mood) disorders, is a complex neuropsychiatric disease characterized by severe anhedonia (a marked inability to enjoy everyday pleasures), metabolic disturbances, and cognitive impairment. Exposure to chronic, uncontrollable stress over a lifetime has been identified as one primary risk factor for the development of MDD. The pathophysiology of persistent and unresolved stress may initially result from prolonged overactivation of the hypothalamic-pituitary-adrenal (HPA) axis, resulting in elevated levels of the glucocorticoid cortisol (or corticosterone in rodents). HPA axis activation represents an adaptive response intended to be acute, whereby cortisol increases energy substrate availability in the periphery and stimulates alertness and arousal in the central nervous system to sufficiently meet the challenges of a stressor. In contrast, chronic stress and depression, along with several other psychiatric disorders, are associated with dysregulation of normal HPA axis function, elevated levels of cortisol, and morphological changes in vulnerable brain regions that mediate emotional regulation and cognition, such as the hippocampus.

Animal models of depression attempt to recapitulate MDD phenotypes through schedules of unrelenting stress exposure, including the chronic unpredictable stress (CUS) and chronic mild stress (CMS) models, in which rodents are subjected to a variety of stressors at random times per day for several weeks. Such models have high face validity in that they mimic the random encounters with mildly stressful events that an unfortunate individual might experience on a day-to-day basis. Over time, these procedures reliably produce depression-associated, anhedonia-like behaviors such as decreased preference for natural rewards (e.g., sucrose) and increased circulating corticosterone. Morphologically, CUS produces dendritic atrophy of CA3 pyramidal neurons of the hippocampus (Magariños and McEwen 1995), and these morphological changes have been demonstrated following excess glucocorticoid administration alone (Woolley et al. 1990). Importantly, this structural damage can be effectively countered via antidepressant medication (Ota and Duman 2013).

As an anatomical correlate of episodic learning and memory formation, sensory information from the entorhinal cortex enters the hippocampus through either the heavily researched trisynaptic pathway or the temporoammonic (TA) pathway, with the TA pathway comprising direct projections from the entorhinal cortex to CA1 pyramidal neurons in the hippocampus (Remondes and Schuman 2004). In addition to impacting CA3 neurons, chronic stress also impairs spatial memory via reduced long-term potentiation (LTP) in the hippocampal CA1 area of rodents (Kim et al. 2001). A recent study (Kvarta et al. 2015) now suggests that elevated corticosterone is sufficient for the development of both the behavioral profile and neuroadaptational manifestation of a chronic stress-induced, depression-like state, specifically due to altered synaptic function of the hippocampal temporoammonic pathway. The new investigation by Kvarta et al. lends support to previous findings of hippocampal dysfunction due to chronic stress and specifically implicates dysfunction of TA-CA1 synapses as a primary cause of stress-induced cognitive deficits associated with depression (anhedonia and memory impairment).

In their study, corticosterone-exposed animals exhibited an increased aversion to feeding in a novel environment and a decrease in sucrose preference (indicative of an anhedonia-like state) compared with controls. After validating their model of anhedonia, the authors then evaluated the effect of metyrapone, a corticosterone synthesis inhibitor, on CUS-induced anhedonia-like behavior. Concomitant metyrapone treatment with 3 wk of CUS exposure prevented the previously observed measures of anhedonia-like behavior, further supporting the hypothesis that elevated corticosterone is a predominant mediator of motivational deficits within MDD. A common but perhaps less appreciated dimension of MDD is the resulting cognitive deficits that exacerbate this disease. To model this condition, animals were trained on a hippocampus-dependent spatial memory task. CUS animals were impaired on this task compared with unstressed animals, because they took significantly more time to locate a hidden platform in a water maze. In contrast, metyrapone treatment in CUS animals alleviated the stress-induced impairment of spatial memory.

Selective serotonin reuptake inhibitors (SSRIs) are well known to attenuate depression-like symptoms caused by chronic stress. However, the effectiveness of conventional SSRIs to treat cognitive deficits and memory impairment is uncertain (Orzechowska et al. 2015). It is therefore necessary to investigate alternative mechanisms for the development of new and innovative treatment approaches. Altered glutamatergic plasticity and dendritic morphology in the hippocampus are also proposed to promote MDD symptomatology following a history of chronic stress exposure. An important emerging area of research focuses on understanding the influence of glucocorticoids on glutamatergic system dysregulation linked with affective disorders, including depression. Kvarta et al. (2015) suspected that corticosterone might facilitate depressive behavioral profiles through altered glutamatergic signaling at vulnerable synapses within the hippocampus. Indeed, recordings from TA-CA1 field excitatory postsynaptic potentials indicated that chronic corticosterone treatment significantly reduces AMPA-mediated excitation. The authors suggest this directly results from the downregulation of AMPA glutamate receptor 1 (GluR1) subunits to consequently decrease the ratio of AMPA to NMDA receptor-mediated current (AMPA/NMDA ratio; Fig. 1). Importantly, metyrapone-treated animals were resistant to AMPA-mediated signaling alterations, displaying normal levels of excitation and AMPA/NMDA ratios. This research, in addition to previous work published by the Thompson laboratory (Cai et al. 2013; Kallarackal et al. 2013), provides very convincing evidence in support of the excitatory synapse hypothesis of depression, which implicates compromised excitatory neurotransmission in the form of glutamate receptor deficiency as the primary source of depression symptoms. Both antidepressants and the corticosterone synthesis inhibitor are able to recover excitatory neurotransmission in the CUS model. This is particularly interesting, illustrating that distinct neurotransmitter systems (amino acid, monoamine, and endocrine) profoundly influence one another to foster psychiatric pathologies.

Fig. 1.

Chronic glucocorticoid exposure impairs AMPA receptor-mediated signaling in the hippocampus. Electrophysiological recordings measuring temporoammonic (TA)-CA1 field excitatory postsynaptic potentials indicate that chronic corticosterone treatment significantly reduces AMPA function. Reduced AMPA-mediated excitation in TA-CA1 synapses is likely the direct result of the downregulation of GluR1 subunit expression and decreased AMPA/NMDA ratio. Further exploration of novel interventions for major depressive disorder aimed at targeting these neuroadaptations (including selective glucocorticoid receptor antagonists and the fast-acting antidepressant ketamine) are warranted.

At the molecular level, changes in total AMPA subunit content are potentially indicative of altered function, but the transient activity of individual subunits is also tightly regulated via phosphorylation by specific kinases anchored at the synapse. Kvarta and colleagues previously demonstrated that increases in GluR1 phosphorylation at serine 831 are a mechanism by which serotonin enhances TA-CA1 excitatory transmission (Cai et al. 2013), representing one possible explanation for the efficacy of SSRIs. Again, however, such conventional antidepressants have mixed effects in alleviating MDD-associated cognitive deficits, whereas the current study has shown that a reduction in corticosterone (via administration of metyrapone) is able to do so. In future investigations, it would be worthwhile to confirm the ability of corticosterone to induce posttranslational modifications, to determine if there are differential changes compared with antidepressant treatment, and to understand which of these phosphorylation events regulates cognitive behaviors. Determination of more specific information about individual subunits could then generate more hypotheses and experiments where novel therapeutic compounds might prove to be useful in this depression model and beyond.

The authors also discovered stress-induced reductions in glucocorticoid receptor (GR) expression within TA-CA1 synapses, an effect that was absent in metyrapone-treated animals, potentially reflecting a compensatory response to excessive corticosterone levels. The GR is a classic example of a steroid receptor functioning as a transcription factor, with GR nuclear localization and transcriptional activity being tightly regulated by GR phosphorylation. The significance of GR phosphorylation in affective disorders has been demonstrated in the context of alcohol dependence, which is associated with depression-like symptoms in some patients. Specifically, our colleagues Vendruscolo et al. (2015) have shown that GR phosphorylation is significantly increased in the central amygdala of alcohol-dependent rats, whereas systemic administration of the GR antagonist mifepristone reduces excessive alcohol drinking in both rats and human alcoholics. Mifepristone also has demonstrated efficacy in reducing symptoms of mild depression and cognitive disruption in patients suffering from Cushing's syndrome (Fleseriu et al. 2012), and newer, more selective GR antagonists are now available. Increased GR phosphorylation is associated with increased transcriptional activity; however, the ultimate downstream consequences of this are unknown. Thus GR-regulated gene expression in vulnerable brain areas is an additional factor to consider in future work on MDD. Additionally, it would be interesting to determine the potential contribution of genomic vs. nongenomic GR signaling on modulation of excitatory neurotransmission.

It is worth noting additional up and coming therapeutic strategies to target excitatory signaling in MDD. Changes in the AMPA/NMDA ratio commonly occur in response to chronic stress and can have dramatic effects on excitatory potential of brain regions and circuits. For example, evidence suggests that chronic restraint stress enhances NMDA receptor-mediated synaptic currents in the hippocampus (Kole et al. 2002). Unfortunately, AMPA and NMDA channels mediate the majority of excitatory brain signaling, potentially limiting these receptors as targets for therapeutic intervention. However, NMDA receptor antagonism via ketamine produces a fast-acting antidepressant effect in depressed individuals, including those who are classified as resistant to treatment with conventional antidepressants (Sanacora and Schatzberg 2015). In their preclinical model, Kvarta et al. found that neither CUS nor corticosterone administration altered NMDAR-mediated signaling at TA-CA1 synapses. Further exploration of this discrepancy in chronic stress-induced changes in NMDAR function is important for understanding the interaction between chronic stress and depression, and may also provide new treatments for MDD that bolster or replace SSRI therapy. One possible mechanism may relate to ketamine's ability to increase glutamate release, which may rapidly restore glutamatergic deficits in MDD patients.

In summary, the recent study by Kvarta et al. (2015) strongly supports the hypothesis that chronic elevations in glucocorticoids, as induced by either natural stressors or direct hormonal treatment, underlies pathophysiological alterations in glutamatergic signaling to produce a behavioral profile closely associated with affective disorders such as MDD. Blocking corticosterone synthesis via metyrapone effectively prevents these neuroadaptations and associated behavioral alterations, although the ability of metyrapone to attenuate or reverse these changes has yet to be demonstrated. Moreover, MDD-associated cognitive deficits are not always alleviated in conjunction with other primary symptoms such as anhedonia, and it is unknown which glutamatergic neuroadaptations are specific to a particular symptom. Lastly, there is evidence that chronic stress and SSRI treatment induces gender-specific differences in depressive symptoms and excitatory signaling in the hippocampus (Mitic et al. 2013). Given that MDD is twice as likely to affect women compared with men, the effect of sex hormones on hippocampal circuitry in the context of chronic stress warrants further investigation. Despite an abundance of remaining questions, exciting new research is beginning to uncover just how the adaptive physiological response to stress can become dysregulated to alter brain structure and function. A greater understanding of how hippocampal circuitry is altered following chronic stress will no doubt prove beneficial in the discovery of new therapeutic strategies to combat MDD and closely associated affective disorders.

GRANTS

We acknowledge the generous support of the Biomedical Research Training Program at LSU Health New Orleans, which is supported by National Institute on Alcohol Abuse and Alcoholism Grant T32AA007577.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.A.M. and A.R.P. prepared figures; M.A.M. and A.R.P. drafted manuscript; M.A.M. and A.R.P. edited and revised manuscript; M.A.M. and A.R.P. approved final version of manuscript.