The relationship between GABA and stress: ‘it's complicated’

GABA is well known to be a critical regulator of the body's physiological response to stress through tight regulation of hypothalamic–pituitary–adrenal (HPA) axis function (Cullinan et al. 2008). Nearly half of the synapses onto corticotropin-releasing hormone (CRH) neurons in the paraventricular nucleus of the hypothalamus (PVN), which govern the activity of the HPA axis, are GABAergic. CRH neurons have been shown to be regulated by both phasic and tonic GABAergic inhibition, which are distinct types of GABAergic inhibition mediated by specific subtypes of GABAA receptors (GABAARs) with unique pharmacological properties and subcellular distribution. Despite the well-accepted role for GABA in the regulation of the HPA axis, few laboratories have investigated the functional contribution of GABAergic signalling and plasticity on CRH neuronal activity and HPA axis function. In fact, many of the synaptic mechanisms regulating the HPA axis have only recently been elucidated (Levy & Tasker, 2012), and the authors of the article that this Perspective highlights have contributed considerably to our knowledge on this topic (Bains et al. 2015). The negative feedback of glucocorticoids, including corticosterone (CORT), is well established in the regulation of the HPA axis involving actions on glucocorticoid receptors on CRH neurons. The article in this issue of The Journal of Physiology by Colmers & Bains (2018) further examines the effects of CORT on diverse GABAergic mechanisms regulating CRH neurons. The authors demonstrate that CORT alters GABAergic signalling through changes in the expression of postsynaptic GABAARs and tonic GABAergic inhibition. The authors also expose that GABAergic control of CRH neurons is constitutively under suppression through actions on presynaptic GABAB receptors, thereby minimizing GABA release, which can be enhanced by blockade of GABA transporters (GATs). Further, this study also demonstrates that GATs suppress the activation of extrasynaptic GABAARs, limiting tonic GABAergic control of CRH neurons, which can be unmasked by blocking GATs. These data demonstrate that CRH neurons are regulated by both tonic and phasic GABAergic inhibition, although the magnitude of these different types of inhibition may be modified by the extent of GAT function and/or the presence of CORT. Thus, this study demonstrates three separate GABAergic mechanisms controlling the activity of CRH neurons in the PVN: GAT-mediated influence on presynaptic GABAB receptors, GAT regulation of tonic GABAergic inhibition, and CORT-mediated postsynaptic GABAAR regulation. Although the mechanisms of corticosterone-mediated regulation of GABAARs in this context are unknown, a neuroactive derivative of corticosterone, THDOC, has been shown to regulate the expression of GABAARs via phosphorylation of specific GABAAR subunits (Abramian et al. 2014). Interestingly, similar mechanisms have been also shown for other steroid hormones, including ovarian hormones and their neurosteroid derivatives, such as allopregnanolone (Modgil et al. 2017). Thus, there is the potential for ovarian hormone-mediated changes in GABAARs in CRH neurons, which may also impact the regulation of CRH neurons and, thus, the HPA axis. Further studies are required to investigate this potential novel mechanism linking ovarian and stress hormones. Consistent with an interaction between ovarian hormones and stress, altered stress reactivity and vulnerability to mood disorders have been demonstrated over the oestrous cycle (MacKenzie & Maguire, 2014). It is tempting to speculate that the oestrous cycle-linked changes in stress reactivity may involve changes in the GABAergic control of CRH neurons, similar to the CORT observations described here, given the evidence for alterations in GABAAR subunit expression over the oestrous cycle in other brain regions (MacKenzie & Maguire, 2014). The evidence that GATs limit presynaptic GABAB-mediated reduction in GABA release, thereby facilitating GABAergic signalling, whereas, they prevent extrasynaptic GABAAR-mediated tonic inhibition postsynaptically, by limiting the concentration of GABA reaching these receptors, suggests that GATs perform a unique function in balancing tonic and phasic GABAergic inhibition. Further, data presented in Colmers & Bains (2018) demonstrate that presynaptic GABAB receptors are constitutively active on presynaptic terminals impinging on CRH neurons in the PVN. These findings suggest that at baseline GAT function is incapable of completely limiting GABAB-mediated suppression of GABA release. However, the authors also demonstrate a lack of tonic current at baseline, which is only unmasked after GAT inhibition. Therefore, it appears that at baseline GAT function allows some degree of presynaptic GABAB signalling but not actions on extrasynaptic GABAARs which mediate tonic inhibition. This may be due to the localization of GATs in relation to presynaptic GABAB receptors and extrasynaptic GABAARs. Either way, these findings reveal a novel mechanism and unique tuning of the GABAergic control of CRH neurons, which may be further influenced by stress. The physiological relevance of these CORT-mediated regulatory mechanisms controlling CRH neurons remains to be fully explored. For example, how does this mechanism influence stress reactivity? It is reasonable to assume that these changes, such as the CORT-mediated increase in tonic GABAergic inhibition of CRH neurons, would provide a novel negative feedback onto the HPA axis. However, evidence of the stress-induced collapse of the chloride gradient and compromised GABAergic inhibition (Hewitt et al. 2009; Sarkar et al. 2011) add further complexity to the role of GABAergic control of the HPA axis and stress reactivity. In addition to the potential effects of stress-induced CORT on GABAAR regulation, the effects of stress on GAT expression and/or function remains to be determined. Very few studies have investigated changes in GABA uptake associated with stress, which this study suggests may have profound effects on HPA axis function. Previous studies have demonstrated stress-induced changes in the GABAergic regulation of CRH neurons, including altered GABAAR subunit expression, a reduction in presynaptic GABA release, and overall impairments in GABAergic signalling onto CRH neurons (Herman & Tasker, 2016). The CORT-mediated effects on the GABAergic regulation of CRH neurons demonstrated in Colmers & Bains (2018) may contribute to these stress-induced changes in GABAergic regulation of CRH neurons. Future studies are required to fully appreciate the roles of stress and CORT in the GABAergic regulation of CRH neurons and the HPA axis, but this highlighted study takes a huge step in furthering our knowledge regarding the complicated relationship between GABA and stress. The author serves on the Scientific Advisory Board for SAGE Therapeutics. The author is funded by NIH National Institute of Neurological Disorders and Stroke (NS102937).