The lateral habenula (LHb) is a brain structure within the epithalamus which is now well established to have a causative role in the pathogenesis of major depressive disorder (MDD; Sartorius et al., 2010; Li et al., 2011; Yang et al., 2018a; Hu et al., 2020). Long-range excitatory neurons in the LHb project to and exert net inhibitory control over the ventral tegmental area (VTA; Ji and Shepard, 2007; Jhou et al., 2009), and the dorsal raphe nucleus (DRN; Wang and Aghajanian, 1977; Ferraro et al., 1996), thus acting as the “off switch” of the midbrain reward circuitry. From an evolutionary standpoint, this serves the important function of ensuring behaviours with negative outcomes are not reinforced (Hikosaka, 2010). However, this system becomes dysregulated in MDD and the LHb becomes hyperactive (Li et al., 2011; Lecca et al., 2016; Tchenio et al., 2017; Cui et al., 2018; Yang et al., 2018a), thus likely potentiating inhibitory modulation of the downstream reward circuitry and curtailing the associated positive emotions.

In addition to the wealth of information that now exists relating to how the LHb controls its’ efferent targets, it is also known that the LHb receives afferent input from a variety of brain regions pivotal in emotional processing. These include the internal segment of the globus pallidus (Hong and Hikosaka, 2008), analogous to the rodent entopeduncular nucleus (Shabel et al., 2014; Meye et al., 2016; Wallace et al., 2017), the lateral hypothalamus (Stamatakis et al., 2016; Lecca et al., 2017; Lazaridis et al., 2019; Trusel et al., 2019) and the ventral pallidum (Knowland et al., 2017; Faget et al., 2018; Stephenson-Jones et al., 2020; Pribiag et al., 2021), as well as receiving reciprocal input from the VTA (Stamatakis et al., 2013; Root et al., 2014). These structures provide a combination of excitatory, inhibitory and GABA/glutamate co-releasing input to the LHb, in a very fine balance which has been shown to be altered in depressive states (Shabel et al., 2014; Knowland et al., 2017). In general, convergent evidence suggests that excitatory LHb afferents promote aversive states and depressive behaviour (Barker et al., 2017; Knowland et al., 2017; Lecca et al., 2017; Lazaridis et al., 2019), while conversely inhibitory afferents promote behavioural reinforcement (Faget et al., 2018; Stephenson-Jones et al., 2020) and reduce depressive behaviour (Winter et al., 2011; Huang et al., 2019). Taking this into account, we here review the current state of knowledge regarding the processing of inhibitory signalling within the LHb. We will discuss what is currently known about inhibitory input to the LHb, how this is altered in depressive states, and how this may eventually be exploited to develop novel therapies for MDD.

It is well-accepted that the majority of LHb neurons form a physiologically homogenous, but morphologically diverse population (Weiss and Veh, 2011), of which most neurons are glutamatergic (Omelchenko et al., 2009; Brinschwitz et al., 2010). Yet mounting evidence now indicates the existence of locally-targeting inhibitory neurons (Zhang et al., 2018; Flanigan et al., 2020; Webster et al., 2020; Nakamura et al., 2021), which may form multiple distinct sub-classes. One such population express the inhibitory marker glutamic acid decarboxylase 2 (GAD-2) and are confined to the medial LHb (Flanigan et al., 2020). These neurons have been shown to inhibit other LHb neurons to promote aggressive behaviour in mice (Flanigan et al., 2020) which, as mentioned earlier, can be interpreted as a reward-seeking behaviour (Golden et al., 2016). We and others have recently shown that some LHb neurons which express the well-known cortical interneuron marker parvalbumin (Tremblay et al., 2016) are GABAergic and inhibit local LHb neurons (Webster et al., 2020; Nakamura et al., 2021). These neurons are likely distinct from the aforementioned population of GAD-2 expressing neurons in that they are confined to the lateral LHb and also express the vesicular GABA transporter (VGAT; Webster et al., 2020), another well-known marker of inhibitory neurons, but largely do not express GAD-2 (Nakamura et al., 2021). Note however that the expression patterns of GABAergic markers within the LHb appears to differ between mice and rats (Quina et al., 2020), with VGAT expression in rats observed in the medial LHb and GAD-2 expression largely absent (Zhang et al., 2018; Quina et al., 2020). Furthermore, early electrophysiological and morphological studies have proposed that a population of neurons akin to the very distinctive cortical neurogliaform interneuron exists within the LHb in rats (Weiss and Veh, 2011; Wagner et al., 2016), although these neurons appear to be different to their cortical counterparts at least in that they do not express similar molecular markers (Webster et al., 2021).

Interestingly, however, some studies have also shown that LHb neurons which express inhibitory markers also express the excitatory marker vesicular glutamate transporter 2 (VGLUT-2), and project to the midbrain (Zhang et al., 2018; Quina et al., 2020). Furthermore, these neurons appear to release exclusively glutamate in the tegmentum of the midbrain (Quina et al., 2020). Thus it may be the case that within the LHb, inhibitory neurons do not exist exclusively as the well-described populations of locally-targeting interneurons that they are known to form in other brain structures (Tremblay et al., 2016), but rather may be dual-functioning neurons which serve the additional purpose of exciting downstream regions (Figure 2).

While the majority of work studying the role of the LHb in MDD and other psychiatric disorders has focused on alterations in direct excitatory and inhibitory inputs, there is an emerging body of evidence indicating that various other neurotransmitters and neuropeptides can influence the activity of the LHb, both directly and indirectly. These are discussed below.

To date, the majority of work focusing on the LHb as a potential therapeutic target for MDD has aimed to reduce the excitability of LHb neurons (for reviews, see Nuno-Perez et al., 2018; Yang et al., 2018b; Hu et al., 2020). Referring back to the original hypothesis that the LHb is hyperactive in depression (Li et al., 2011), there is now substantial evidence that reducing the excitability of LHb neurons via a variety of means has an antidepressant effect. Much work has focussed on normalising LHb activity by the use of deep brain stimulation (DBS), a technique whereby electrodes are implanted within the brain to modulate neuronal activity (Sartorius et al., 2010; Meng et al., 2011; Tchenio et al., 2017; Jakobs et al., 2019). This technique appears to hold great promise as a therapeutic intervention in that LHb-targeted DBS has been shown to induce full remission in a human patient (Sartorius et al., 2010). Interestingly, the antidepressant effect of DBS appears to be dependent on the stimulation frequency, in that high frequency stimulation (>100 Hz) has an antidepressant effect in rodents (Meng et al., 2011; Tchenio et al., 2017; Jakobs et al., 2019), while lower frequencies stimulation (5–20 Hz) appears to promote depressive behaviour (Elmer et al., 2019; Jakobs et al., 2019).

But what are the mechanisms by which high frequency DBS within the LHb elicits an antidepressant effect? Interestingly, in a recent investigation in rodents, Tchenio et al. (2017) showed that firing activity of LHb neurons recorded in vivo decreases during local LHb DBS. The authors, in line with a previous study (Li et al., 2011) also demonstrated that such high frequency stimulation dampens AMPA transmission onto LHb cells. It has also been proposed that DBS activates presynaptic GABAergic terminals to promote GABA release (Li et al., 2004), which may be a possible explanation given the extensive GABAergic innervation the LHb receives (Brinschwitz et al., 2010). However, whether DBS modulates GABAergic transmission in the LHb remains to be tested.

Modulating the electrical firing pattern and synaptic activity of LHb neurons via various pharmacological interventions has also been shown to have therapeutic benefits. For instance, blockade of NMDA receptors reduce bursting activity, and is at least partially responsible for the antidepressant actions of ketamine (Cui et al., 2018; Yang et al., 2018a), while AMPA receptor blockade has also been demonstrated to have antidepressant efficacy (Li et al., 2017; Zhang et al., 2019). Additionally, both local LHb chemogenetic inhibition (Nair et al., 2013; Tchenio et al., 2017), and direct pharmacological inhibition (Winter et al., 2011) and input-specific synaptic inhibition (Huang et al., 2019) have displayed efficacy as potential therapeutic options in rodent models of depression. The therapeutic benefits of direct inhibition of the LHb can be linked to both GABA_Aand GABA_B signalling. Direct injection of the GABA_A agonist muscimol has been shown to alleviate depressive symptoms in both the classical learned helplessness rat model of depression (Winter et al., 2011) and interestingly in depression induced by a rat model of Parkinson’s disease (Wang et al., 2017). Moreover, foot shock exposure in mice triggers a diminished functionality in GABA_B transmission and has been observed in various rodent models of depression, such as acute learned helplessness (Lecca et al., 2016), maternal separation (Tchenio et al., 2017), and chronic social defeat (Li Z.-L. et al., 2021). Altogether the latter studies highlight the GABA_B receptor in the LHb as a potential target to treat depressive symptoms. Mechanistically, a stressful event (i.e., foot shock exposure) in mice triggers internalisation of both GABA_B receptors and G protein-gated inwardly rectifying potassium (GIRK) channels, along with a concurrent increase in protein phosphatase 2A activity, which is known to regulate the expression of these channels. Consistently, pharmacological inhibition of protein phosphatase 2A normalises GABA_Band GIRK expression and alleviates depressive symptoms (Lecca et al., 2016). It is also notable that glycinergic signalling within the LHb has been linked to depressive behaviour, in that direct injection of glycine has an anxiolytic effect in a rat model of alcohol withdrawal, while intra-LHb injection of the glycinergic antagonist strychnine is anxiogenic (Li W. et al., 2019).

In summary, as the mechanisms by which inhibitory signalling within the LHb are processed become clearer, so too do the potential strategies by which this can be exploited as a therapeutic strategy in MDD. The consistency between pre- and postsynaptic modulation of LHb neurons in relation to depressive behaviour is striking- inhibitory afferents promote reward-like states, while excitatory afferents promote aversion, and consistently activation of postsynaptic inhibitory GABA and glycine receptors appears to have an antidepressant effect, as does blockade of excitatory AMPA and NMDA receptors. As such, novel therapies for MDD may attempt to capitalise on this knowledge by aiming to selectively modulate such pathways.

All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.