The involvement of K⁺ channels in depression and pharmacological effects of antidepressants on these channels

Abstract

Depression is a common and complex psychiatric illness with multiple clinical symptoms, even leading to the disability and suicide. Owing to the partial understanding of the pathogenesis of depressive-like disorders, available pharmacotherapeutic strategies are developed mainly based on the “monoamine hypothesis”, resulting in a limited effectiveness and a number of adverse effects in the clinical practice. The concept of multiple pathogenic factors be helpful for clarifying the etiology of depression and developing the antidepressants. It is well documented that K⁺ channels serve crucial roles in modulating the neuronal excitability and neurotransmitter release in the brain, and abnormality of these channels participated in the pathogenic process of diverse central nervous system (CNS) pathologies, such as seizure and Alzheimer’s disease (AD). The clinical and preclinical evidence also delineates that the involvement of several types of K⁺ channels in depressive-like behaviors appear to be evident, suggesting these channels being one of the multiple factors in the etiology of this debilitating disorder. Emerging data manifest that diverse antidepressants impact distinct K⁺ channels, such as Kv, Kir and K_2P, meaning the functioning of these drug via a “multi-target” manner. On the other hand, the scenario of antidepressants impinging K⁺ channels could render an alternative interpretation for the pharmacological effectiveness and numerous side effects in clinical trials. Furthermore, these channels serve to be considered as a “druggable target” to develop novel therapeutic compound to antagonize this psychiatry.

Introduction

Depression is a highly prevalent mental disorder, with multiple clinical symptoms such as anhedonia, negative mood and suicidal thoughts, leading to a heavy burden on economy and society. Especially, major depressive disorder (MDD), a chronic and recurring neuropsychiatric illness, impinges on millions of people worldwide and induces the disability [1], even life threatening. Nowadays, although aggressive research efforts are conducted over several decades, the virtual mechanism underlying depression is still largely uncharacterized, with a resultant consequence of poor pharmacological treatments for patients with depression. For instance, one-third of patients with MDD lack an effective response to antidepressant medications [2]. Initially, the monoaminergic deficiency hypothesis of depression is considered as an acceptable concept underlying the etiology of this disease, and in turn serotonin reuptake inhibitors (SSRIs) known as monoaminergic-based drugs, such as fluoxetine, paroxetine, citalopram, fluovoxanine and sertraline, are available for clinical medications for depression with chronic antidepressant benefits, actions of which occur after weeks to months, actually being regarded as a potential drawback [3]. The other chemicals for depression treatment, such as tricyclic antidepressants (TCA), serotonin-norepinephrine reuptake inhibitor (SNRIs) and monoamine oxidase inhibitor (MAOI), are also available in the clinic, with similar antidepressant efficacy [2, 4]. These classes of antidepressants exhibit varying side effect profiles or drawbacks, which is one of the important determinants for choosing the appropriate medication for an individual patient in the clinical practice when efficacy of antidepressants being identical (see the review by Bauer et al. [3]).

Recently, clinical and preclinical studies have provided considerable evidence supporting the hypothesis of an imbalance of excitation–inhibition (E:I) within the prefrontal cortex (PFC) in which stress results in the dysfunction of GABAergic system with the related alteration of glutamatergic excitatory counterpart [5]. Accordingly, ketamine, a glutamatergic-based drug, generates a surge of glutamate that mediates the occurrence of synaptogenesis, subsequently restoring both glutamatergic and GABAergic synapses [6] and mediating a rapid and long-lasting antidepressant action, especially being effective for treatment-resistant depression (TRD) with anxiety [7]. Clearly, antidepressant effects conferred by ketamine display a distinct mechanism from conventional antidepressants. Furthermore, the involvement of aberrant burst firing generated by the lateral habenula (LHb) neurons with hyperactivity in animal models of depression is evident, and the reduction of bursting activity after exposure to ketamine is associated with the attenuation of depressive-like symptoms in related animal models [8], which is, at least in part, contributed to the antidepressant response of the N-methyl-D-aspartate receptor (NMDAR) inhibitor ketamine. Nevertheless, the precise interpretation for the action of ketamine on treating depressive-like behaviors is not yet fully clarified.

The accompanying increase in understanding of depression, several novel threads for overcoming this neuropsychiatric disorder are emerging. For example, the antagonists of T-type Ca²⁺ channels (Ca_V3), which initiate NMDA-dependent bursting in the brain’s LHb neurons and opposite the depressive symptoms [9], are regarded as the promising antidepressants. On the other hand, the KCNQ (Kv7) channels, such as KCNQ2-5 that are expressed predominantly in the neural system, are also thought to be a potential target of treatment for MDD [10]. Interestingly, accumulating evidence from clinical and preclinical studies also displays that antidepressant drugs serve to impact multiple targets, which possibly facilitates the development of novel therapeutic drugs or strategies for the patients with depression. Of these targets, K⁺ channels enable substantial modulation of neuronal excitability and activity, and undergo intensely studies by the investigators.

The membrane-spanning protein K⁺ channels with numerous family members exist in both excitable and nonexcitable cells, allowing fast diffusion of K⁺ ions across cell membranes [11]. Generally functional K⁺ channels are multiple complexes composed of pore-forming α subunits and cytoplasmic accessory β subunits [11, 12]. According to the primary amino acid sequence of α subunits, K⁺ channels are divided into three main groups: voltage-gated K⁺ (Kv) channels containing six transmembrane (TM) domains and a single pore-forming (P) domain, inwardly rectifying K⁺ (Kir) channels with two TM domains and a single P domain, and two-pore-domain K⁺ (K_2P) channels possessing four TM domains and two P domains [11–13]. Various primary cell properties, such as membrane excitability, growth, cell volume, are modulated by K⁺ channels in different biological cells. Especially, K⁺ channels manipulating the electrical activity of excitable cells, such as neurons, cardiomyocytes and pancreatic β-cells, plays fundamental roles in tuning neurotransmitter release, heart rate, and insulin secretion [11, 14]. Dysfunctions of K⁺ channels can participate into the pathogenesis of many diseases in various organ systems, such as LQT syndromes, episodic ataxia, and epilepsy [11, 12]. As such, K⁺ channels are taken as therapeutic drug targets for several common diseases in the clinic practice. For example, the oral hypoglycemic drug sulfonylureas that block the K_ATP channels in pancreatic β-cells enable glycemic control for patients with type 2 diabetes mellitus (T2DM) [15].

In neural systems, the K⁺ efflux conferred by K⁺ channels, such as voltage-gated K⁺ (Kv) channels, induces the membrane hyperpolarization, usually conducting an inhibitory effect on neuronal excitability. The inwardly rectifying K⁺ (Kir) channels and background two-pore-domain K⁺ (K_2P) channels have a fundamental role in setting resting membrane potentials of neurons or glial cells [16]. Several types of K⁺ channels, including Kv, K_ATP and K_2P channels, participate into the neuroprotection against ischemic injury and apoptotic stimuli [17–19]. The opening of K⁺ channels, such as KCNQ, is able to lower the neuronal and behavioral hyperexcitability, a feature of several neurological disorders, including epilepsy, addiction and anxiety [20]. Furthermore, the heteromeric channels formed by KCNQ2 and KCNQ3 that are responsible for native M-type currents in most neurons [20] are activated by ezogabine (retigabine), with significant ameliorations in depressive symptoms and anhedonia in patients with MDD [10]. Interestingly, a growing body of evidence reveals that antidepressant drugs are capable of impinging on the diverse K⁺ channels, such as Kv channels [21], K_2P channels [22] and K_ATP channels [23], indicative of more or less contribution for antidepressant effects. Presently, we summarize these data as to the action of antidepressant chemicals on K⁺ channels and hopefully find more hints underlying the pharmacological efficacy and/or adverse effects of these drugs in the clinic.

Kv channels

Kv channels composed of four pore-forming α-subunits and regulatory β-subunits exhibit critical abilities in regulating resting membrane potentials, action potentials and firing rates in diverse neurons, subsequently governing the neuronal excitability and neurotransmitter release. For example, M-currents encoded by KCNQ2 and KCNQ3 channels with slow activation and non-inactivation kinetics can generate a substantial physiological action to limit repetitive or burst-firing, and accordingly decline the excitability in distinct brain regions [24]. The neuronal hyperexcitability due to Kv channel dysfunction, especially for Kv7 (KCNQ) channels, is associated with neurological disorders, such as epilepsy and pain [25, 26]. At present, the activity of antidepressant targeting Kv channels is well established by lots of published reports, suggestive of the resultant alteration of neuronal excitability and more or less linkage with the therapeutic efficiency.

Kv1 channels

The subcellular localization of Kv1.1-Kv1.7 is uncovered in the axons/terminals [27], with the features of low voltage-activated opening on small depolarizations to potentials ranging from resting potentials (−70 mV) to around −40 to −30 mV, accordingly regulating the number of action potentials (AP) on the depolarization [16]. The Kv currents carried by Kv1.4 and Kv1.2 channels exhibit the important ability in control of neuronal excitability and dopamine release in midbrain dopamine neurons (DANs) [28], which are considered as primary neural cells for the midbrain reward center, and are involved in the pathogenesis of depression. Fluoxetine, a first-line antidepressant drug that inhibits the serotonin reuptake, is able to conduct a blocking effect on Kv1.1, Kv1.3, Kv1.4 and Kv1.5 channels, as well as native Kv currents in hippocampal and cerebellar granule neurons [21, 29–32]. Interestingly, several other selective serotonin reuptake inhibitors, dapoxetine, citalopram, sertraline, paroxetine and fluvoxamine, can induce inhibitory effects on Kv1.5 channels [33–37], which is known as the molecular correlate of ultrarapid K⁺ current IKur in human cardiomyocytes [38]. The effects of these antidepressants on Kv1.5 channels could be clinically relevant in the upper range of the therapeutic plasma concentrations [33–37]. As such, citalopram-induced cardiovascular risk within the clinical range could be attributed to its block on Kv1.5 channels, which allows for the prolongation of the action potential in atrial cardiomyocytes [34].

In learned helplessness rats, dendrotoxin-I, a blocker of Kv1.1, Kv1.2 and Kv1.6 channels, results in a significant enhancement of the spike firing of serotonin neurons, and conversely the upregulation of dendrotoxin-I-sensitive Kv1 channels is capable of lowering the firing of serotonin neurons [39]. Additionally, the inhibition of antidepressant ketamine on dendrotoxin-I-sensitive Kv1 channels favors the spike firing of serotonin neurons [39], thereby triggering the release of 5-HT. Treatment with ketamine also makes a direct blockage of Kv1.5 channels [40], which could be associated with its cardiovascular effects due to depolarization-induced elevation of vascular tone [41]. Herein, inhibition of Kv1 channels conferred by SSRIs and ketamine in related tissues, such as neurons and cardiovascular myocytes, may underlie the enhanced release of monoaminergic neurotransmitters and the occurrence of adverse cardiovascular effects.

Kv2 channels

The expression of Kv2.1 that is high-voltage activated channels with slow kinetics of inactivation and deactivation is evident throughout the mammalian brain [42], especially being high level in both hippocampal and cortical neurons where this channel is a major contributor to the delayed rectifier K⁺ currents [43, 44]. The neuronal and behavioral hyperexcitability serves to occur after genetic deletion of Kv2.1 channels [45], and enhanced spontaneous firing activity after blocking this channel is also detected in dopamine neurons [46]. The increased level of Kv2.1 channels is observed in the brain of chronic mild stress rats, which is abolished by application of fluoxetine [47], and fluoxetine also directly inhibits the Kv2.1 channels at clinical concentrations [48]. It is likely that enhanced neuronal excitability mediated by the SSRIs block on Kv2.1 channels facilitates antidepressant efficiency. Interestingly, fluoxetine can impact pancreatic β-cells to boost the insulin secretion [49], a benign side effect that could be beneficial for patients with depression and diabetes. Nowadays, the exact mechanism underlying the action of fluoxetine on insulin secretion remains obscure. Notably, Kv2.1 channels residing in pancreatic β-cells substantially modulate the insulin release via governing the glucose-induced islet electrical activity [50, 51], and inhibition of fluoxetine on Kv2.1, therefore, may provide an explanation for the potentiated insulin secretion.

Besides fluoxetine [48], other antidepressants, including amoxapine and citalopram, also exert a blocking action on Kv2.1 channels in cortical neurons and possibly generate an alternative antidepressant way via increasing presynaptic efficiency, which arises from the prolongation of action potential duration following the Kv2.1 inhibition [52, 53]. More intriguingly, retigabine, an opener of Kv7 channels underlying the neuronal M-currents, could produce an inhibitory action on Kv2.1 channels at clinically relevant concentrations (0.3–3 µM) [54]. It is well established that retigabine-enhanced M-currents encoded by neuronal Kv7.2–Kv7.5 channels account for the mechanism of counteracting hyperexcitability-related disorders such as epilepsy and tinnitus [55]. In addition, activation of KCNQ (Kv7) channels by treatment with retigabine normalizes neuronal hyperactivity and depressive behaviors [56], and these channels are thought to be a promising target for exploring novel treatment for depression and specifically for MDD [56]. During administration of retigabine, apparently, reduced hyperexcitability by the activation of KCNQ (Kv7) channels conflicts with the fact that enhanced excitability by the blocking of Kv2.1 channels. It is likely that this scenario is a concerted action in that both lowered excitotoxicity by Kv7 channel activation and inhibited pro-apoptotic Kv2.1 currents generate the neuroprotective effects in hyperexcitability-related disorders [54], inasmuch as neuronal hyperexcitability is closely associated with the depressive-like behaviors in some of brain regions, such as the dopaminergic (DA) neurons in the ventral tegmental area (VTA) [56]. It is no doubt that more precise experiment is needed to address the contradictory observation.

Kv3 channels

Kv3 channels are ubiquitously expressed in diverse brain regions such as the cerebellar cortex [57] and hippocampus [58], with the features of high thresholds of activation and fast kinetics, including very rapid activation and deactivation [59], accordingly contributing to shaping the repolarization of the AP and fast firing [57]. One example is that a key role of Kv3.1-Kv3.2 channels in producing fast-spiking properties is evident in cortical GABAergic interneurons, allowing high-frequency synaptic transmission [60]. Emerging evidence reveals that the aberrant activity of Kv3.1 is also associated with several neuropsychiatric disorders, such as schizophrenia [61] and depression [62]. Reduced Kv3.1 activity manifests vulnerability to depressive behavior, and accordingly upregulation of Kv3.1 or acute activation of Kv3.1 generates resilience to depression [63].

Two SSRI antidepressants, fluoxetine and paroxetine, display an inhibitory effect on open pore of Kv3.1 channels in a use-dependent manner, which could be clinical relevant [64, 65]. Treatment with the SSRI fluoxetine serves to modulate the Kv3.1b transcripts in a serotonergic cell line [66], and also enables enhanced phosphorylation and inhibition of Kv3.1β channels [67]; the blocking action on Kv3.1 channels conferred by paroxetine may interfere with repetitive firing of inhibitory neurons, providing a mechanistic explanation for pro-convulsive effects of SSRIs [65]. Application of vortioxetine, a novel oral SSRI drug, exerts a blocking action on the delayed-rectifier K⁺ currents (I_K(DR)) in neuronal cells, which could be carried by Kv3 or Kv2 channels [68]. Thus, inhibition of Kv3 channels exhibits an important potential to remit the depressive-like symptoms.

Kv4 channels

Kv4.3 channels underlying the rapidly activating and inactivating K⁺ currents (transient outward current, I_to) in the heart are widely expressed in neuronal cells in the brain, which are also known as the molecular basis of somatodendritic A-type Kv currents [69]. This channel serves to be activated after prior hyperpolarization and inactivated at the resting potential, regulating the action potential number and neuronal excitability [16]. Previous evidence showed that the suppression of 4-aminopyrine on A-type Kv currents can enhance the spontaneous basal release of [3H]5-HT in hippocampal neurons [70]; as such, inhibiting Kv4.3 channels after applying fluoxetine at therapeutic dosages could relieve depression due to the enhancement of 5-HT levels in neuronal cells [21, 70]. Other antidepressants, such as dapoxetine [71], duloxetine [72], also exert blocking action on Kv4.3 channels, and five widely used antidepressant drugs, imipramine, amitriptyline, mianserine, maprotiline, and trazodone, reduce the transient outward current (I_to) carried by Kv4.2 and Kv4.3 channels in ventricular myocytes [73].

Kv7 channels

The low-voltage-activated Kv7 (KCNQ) family consists of five members, Kv7.1 to Kv7.5, and is associated with several inherited diseases, including long QT syndrome, benign familial neonatal seizures (KCNQ2 and KCNQ3) and progressive deafness (KCNQ4) [74]. Besides the Kv7.1 that is located predominantly in peripheral tissue, for example, the heart and kidney, other Kv7 channels are abundantly distributed in the central nervous system, such as prefrontal cortex [75], ventral tegmental area (VTA) [56], lateral habenula [76], two of which (KCNQ2 and KCNQ3) form homodimers or heterodimers responsible for neuronal M-type currents, which are critical for regulating neural excitability [75]. The activation of Kv7 channels generates the neuronal hyperpolarization, and in turn limit repetitive or burst firing in the diverse neural tissues, such as the prefrontal cortex [55, 75], ventral tegmental area (VTA) [56, 77], lateral habenula [76], each of which is known to hold the important role in the pathogenesis and treatment of depression. The activation or overexpression of Kv7 channels can counteract the depressive-like behaviors arising from social defeat stress in VTA-NAc DA neurons, with the suppression of VTA DA neuronal firing rate [78]. Opening of Kv7.4 channels, which are predominantly expressed in VTA dopaminergic neurons, results in the attenuation of depressive behaviors in the social defeat mouse model of depression [79]. Preclinical studies point to Kv7.2/Kv7.3 channel agonists as novel therapeutic drugs for depressive disorders [10], and retigabine (ezogabine), which is approved by FDA as an anticonvulsant, can reverse depressive symptoms due to the activation of Kv7 channels [10, 56]. Recently, a novel result in which adjunctive treatment with Kv7 opener retigabine mediates enhanced antidepressant-like effects of ketamine indicates that Kv7 channel is able to work as a downstream regulator of ketamine action [80].

Kir channels

The inwardly rectifying K⁺ (Kir) channels have an important feature in which the inward K⁺ current is bigger than the outward K⁺ current across the cell membrane [81], with a subunit structure consisting of two transmembrane domains (TM1 and TM2) and one conserved pore-loop [82]. The basic physiological functions of Kir channels are to set the resting membrane potential, tune the excitability and discharge K⁺ ions [83]. The superfamily of Kir channels containing 15 membranes in mammals is divided into four subgroups: K⁺ transport channels (Kir1.x, Kir4.x, Kir5.x, and Kir7.x), classical Kir channels (Kir2.x), G protein-gated Kir channels (GIRK or Kir3.x) and ATP-sensitive K⁺ channels (Kir6.x) [84]. Collected evidence from human and animal studies indicates that Kir channels, such as Kir4.1, GIRK and K_ATP, are involved in the pathogenesis of depression, and antidepressants can impinge on these Kir channels to exhibit pharmacological effects and/or side actions.

Kir4.1 channels

Spatial potassium buffering is one of the most important functions of astrocytes, which removes excessive extracellular K⁺ in synapses through Kir4.1 and/or Kir4.1/Kir5.1 channels [84–86], contributing to the modulation of the excitability, synaptic plasticity, and glutamate metabolism in neuronal cells [84, 86, 87]. The clinical and preclinical data are indicative of an important role of Kir4.1 channels in the etiology and medicine of multiple neural disorders, such as epilepsy and depression [86–88]. The growing body of evidence indicates that the upregulation of astrocytic Kir4.1 channels appears to play a potential pathogenic role for depression [86, 89, 90]. The elevated expression of Kir4.1 in LHb astrocytes causes the hyperpolarization of membrane potentials of LHb neurons, and in turn boosts the burst firing of these neurons and generates depressive-like symptoms [9]. The local infusion of ketamine, an NMDAR inhibitor, is capable of lowering the number of bursting LHb neurons and attenuating depressive-like behaviors [8]. Furthermore, ketamine can dampen the vesicular delivery of Kir4.1 to the plasma membrane, accordingly inhibiting neuronal firing in LHb neurons [91–93]. As such, the data analysis from many studies has shown that Kir4.1 channels have a potential as a therapeutic target when treating depression. In fact, multiple antidepressants, such as SSRIs and TCAs, could inhibit Kir4.1 channels in a reversible manner (see the review by Ohno et al. [87]), providing a novel insight for analyzing pharmacologic action conferred by antidepressants. The IC₅₀ value of fluoxetine inhibition on Kir4.1 falls within the range of its brain concentrations, and thus this blocking action could be clinically relevant [94].

Notably, after extensive efforts, fluorescence-based thallium flux assays, a molecular target-based approach using high-throughput screen (HTS) of small-molecule libraries, are generated to develop next-generation Kir channel modulators (see the review by Weaver et al. [3]). Thallium flux assay is conducted to screen 76,575 compounds from the VICB library for small-molecule modulators of homomeric Kir4.1 channels, and the most potent inhibitor is 2-(2-bromo-4-isopropylphenoxy)-N-(2,2,6,6-tetramethylpiperidin-4-yl) acetamide (VU0134992), with the functional binding site located in the ion-conduction pathway [95]. Although the beneficial effects of VU0134992 on renal functions, such as diuresis and natriuresis, is evident [95], the therapeutic potential of VU0134992 for depression remains elusive. Recently, Lys05, a dimeric chloroquine with effective inhibitory ability on native Kir4.1 channels, has manifested rapid-onset antidepressant actions in multiple canonical depression rodent models [88], exhibitive of the potential of small-molecule medication for depression and Kir4.1 being a druggable target for translational studies. To date, the data regarding depression treatment using small-molecule agents targeting K⁺ channels are still limited, although lots of small-molecule antagonists or agonists have been developed and the preclinical and clinical pharmacological studies have been performed.

GIRK channels

The G protein-activated inwardly rectifying K⁺ (GIRK) channels located in distinct regions of the brain, including frontal cortex, hippocampus, amygdala, thalamus, and ventral tegmental area, are identified as a heterotetramer of GIRK1 and GIRK2 subunits [92], two members of a large family of Kir channels, which can function in tuning the neuronal excitability and the inhibitory neurotransmission [84, 96]. It is believed that the dysfunction of neuronal GIRK channels is implicated in the etiology of several disorders, such as epilepsy, schizophrenia, autism, mood disorders and Parkinson’s disease [97].

Experimental data validate that various antidepressants, such as fluoxetine, paroxetine, imipramine, desipramine, amitriptyline, nortriptyline, clomipramine, maprotiline, citalopram, atomoxetine, reboxetine, sertraline, duloxetine, and amoxapine suppress GIRK channels in different potency [98–102], providing an underlying mechanism of therapeutic effects and adverse effects of these clinical antidepressants. It is worth noting that the concentrations of some antidepressants like fluoxetine in brain tissues are apparently higher than these in the plasma [103, 104]. Moreover, high levels of antidepressants could be available as a consequence of accumulation in brain tissues, which could be attributed to drug being high lipophilicity and affinity for adipose tissues. These factors may determine whether there exists the clinical relevance for some antidepressants, although these drugs virtually exhibit pharmacological actions in relatively high concentrations in some observations.

A large body of evidence points to that the blockade of NMDARs is responsible for the fast antidepressant efficacy of ketamine [105], and blocking NMDAR using AP5 and Ro-25-6981 is capable of decoupling of GABAB receptors from GIRK channels, with a resulting reduction of the protein levels of GIRK [106], which may contribute to the rapid attenuation of depressive-like behaviors conferred by ketamine. On the other hand, the incidence of seizures, a serious side effect, is associated with the clinic use of SSRIs, including fluoxetine [107, 108]. The blockage of SSRIs on GIRK can induce a resultant enhancement of neuronal excitability, and GIRK2 knockout mice develop spontaneous seizures [109], both of which could be causal clues for mechanisms underlying seizures after application of SSRIs.

K_ATP channels

K_ATP channels are abundantly expressed in several excitable tissues, such as heart, smooth and skeletal muscles, pancreas, and brain, and link membrane electrical activity to energy metabolism, manifesting the substantial modulation of distinct physiological function. It is well documented that there exist high levels of K_ATP channels composed of SURx and Kir6.x subunits in the varying brain regions, including prefrontal cortex, amygdala, hypothalamus and hippocampus [110], each of which is critically relevant to the depressive disorder. Unfolded data display that the blocker of K_ATP channels, glibenclamide, produce an antidepressant action [111], and conversely the K_ATP channel agonists, such as cromakalim, promote the depressive-like behaviors in the forced swimming test (FST) [111], suggestive of the involvement of this channel in etiology of depression. K_ATP channel also participates into antidepressant-like effects of the anticonvulsant drug carbamazepine in mice, which is a part of drug interacting with NO/cGMP/K_ATP channel pathway [112]. Likewise, the marked antidepressant-like effects of simvastatin also involve NO-cGMP-K_ATP channel pathway [113]. Interestingly, another anticonvulsant drug gabapentin and muscle relaxant baclofen are capable of manifesting the antidepressant-like activity through suppressing K_ATP channels [114, 115].

The prolonged usage of antidepressants can enhance the risk of diabetes, but three antidepressants, aroxetine, clomipramine and fluoxetine, fail to reverse the glucose-mediated inhibition of K_ATP channels in pancreatic β-cells [116]. Nevertheless, emerging evidence exhibits that modulatory actions of fluoxetine on neuronal excitability are undertaken via inhibiting K_ATP channels [23]. Noticeably, fluoxetine can directly block the mitochondrial K_ATP channels, which consist of pore-forming Kir6.1 and sulphonylurea-binding SUR1 subunits [117]. It is well established that hetero-octameric complexes formed by SURx and Kir6.x constitute the neuronal K_ATP channels, and therefore, above evidence points to that fluoxetine possibly can influence K_ATP channels located in depression-related brain regions to exert the pharmacological action. More experiments are clearly necessary to clarify the effects of antidepressants on K_ATP channels.

BK channels

The widespread distribution of large conductance Ca²⁺- and voltage-activated K⁺ (BK) channels in central nervous system, including cortex, basal ganglia, hippocampus, thalamus, habenulo-interpeduncular tract and cerebellum, is validated [118], and opening of this channel generates the repolarization or hyperpolarization of neuronal cells against Ca²⁺ influx and hyperexcitability, suggesting a neuroprotective role for the injury stimuli [119]. The declined activity and mRNA expression of BK channels in lateral amygdala (LA) neurons in fear conditioned mice are associated with the depressive-like behaviors, owing to resulting hyperexcitability in response to the depolarization induced by BK inhibition [120]. Herein, it is possible that BK channels could be considered as a “druggable target” to accept the intense study, although the evidence of present antidepressants impacting this channel is not available.

K_2P channels

The two-pore-domain background K⁺ (K_2P) channel is a dimer composed of two pore-forming domains and four transmembrane segments, which is ubiquitously distributed in central nervous system [121], with physiological functions of setting the membrane potential and modulating the excitability in the neuronal cells. The high expression of TASK-3 channels is evident in paraventricular nuclei of thalamus, locus coeruleus and the dorsal raphe [122], and abundant distribution of TREK-1 is also detectable in brain regions correlated with depression, including the prefrontal cortex (PFC) and midbrain dorsal raphe nucleus (DRN) [122, 123]. Of 15 members of the K_2P family, noticeably, TASK-3 and TREK-1 are thought to be associated with the pathophysiology of depression with the supporting evidence arising from knockout mouse models, even being considered as therapeutic targets [124].

Several members of SSRIs, including fluoxetine, norfluoxetine, paroxetine and citalopram, exert directly inhibitory actions on TREK-1 and/or TREK-2 with the varying affinity [22], indicative of a faster antidepressive effect compared with the classical antidepressant drugs. The inhibition of fluoxetine on TREK-1 occurs at clinical brain concentrations, probably producing the clinical consequences [125]. The related content could be seen in the published review paper [22, 124]. Previous data displayed that only one antidepressant belonging to SSRIs, fluoxetine, has a blocking action on TASK-3 channels [125], and circumstantial evidence whereby TASK-3 mutants dampen the fluoxetine-mediated suppression of REM sleep supports the concept of antidepressants targeting this channel [126].

Summary

In the present review, we summarize the involvement of K⁺ channels, including Kv, Kir, BK and K_2P, in the pathogenic process of depression, and the impact of several kinds of antidepressants on above channels (Table 1). Considerable evidence exhibits a scenario in which K⁺ channels appear to be one of the multiple factors responsible for the etiology of depression, and be regarded as one of the multiple targets influenced by antidepressants, an alternative explanation of therapeutic actions and adverse effects in the clinical trials. To date, intense studies from numerous investigators are undertaken to overcome the drawback of classical antidepressants, such as SSRIs, with a slow onset of therapeutic action and the ineffectiveness for many patients [2]. It is no doubt that clarifying the exact pathogenic mechanisms and exploring the novel therapeutic targets is very crucial to the medication of this complex brain deficit. Hence, the accumulating observation, despite limited data, as to K⁺ channels and depression, may render a new clue to address these issues.

Table 1.

Effects of antidepressants on K⁺ channels.

K+ channels	Subtypes	Direct modulation	Modulation of expression and vesicular delivery
Kv	Kv1	Fluoxetine inhibition on Kv1.1, Kv1.3, Kv1.4 and Kv1.5 [29, 30, 32]; Inhibition of fluoxetine, dapoxetine, citalopram, sertraline, paroxetine and fluvoxamine on Kv1.5 [31, 33–37]; Ketamine inhibition on Kv1 [39] and Kv1.5 [40]	–
	Kv2	Inhibition of fluoxetine, amoxapine, citalopram and retigabine on Kv2.1 [48, 52–54]	Fluoxetine downregulation on Kv2.1 [47]
	Kv4	Inhibition of imipramine, amitriptyline, mianserine, maprotiline, and trazodone on Kv4.2 and Kv4.3 [73] Inhibition of fluoxetine, dapoxetine and duloxetine on Kv4.3 [70–72]	–
	Kv7	Retigabine activation on Kv7.2/7.3 [10, 56]	–
Kir	Kir4.1	Inhibition of SSRIs and TCAs on Kir4.1 (see the review by Ohno et al. [87]) Inhibition of Lys05 on Kir4.1 [88]	Ketamine inhibition of Kir4.1 vesicular delivery [91]
	GIRK	Inhibition of fluoxetine, paroxetine, imipramine, desipramine, amitriptyline, nortriptyline, clomipramine, maprotiline, citalopram, atomoxetine, reboxetine, sertraline, duloxetine, and amoxapine on GIRK [98–102]	–
	KATP	Fluoxetine inhibition on KATP [23, 117]	–
K2P	TASK	Fluoxetine inhibition on TASK-3 [125]	–
	TREK	Inhibition of fluoxetine, norfluoxetine, paroxetine and citalopram on TREK-1 and TREK-2 (see the review by Djillani et al. [22])	–

The observed alterations in behavioral paradigms after application of drugs with the antidepressant-like effect to K⁺ channels in animal models of depression is exhibited in Table 2. The beneficial effects on depressive behaviors resulting from actions of these chemicals on some K⁺ channels are available, and additional validation is required in animal models of depression and depressed patients.

Table 2.

Behavioral changes after application of drugs with the antidepressant-like effect to K⁺ channels in animal models of depression.

K+ channels	Subtypes	Behavioral changes
Kv	Kv1	Ketamine reverses LH by inhibiting Kv1 [39] Blockade of Kv1.4 by UK-78,282 increases the motivation for sucrose reward [134]
	Kv7	Upregulation of Ketamine on Kv7.2 lowers the immobility in FST, and improves the social and anxiolytic behaviors assessed by social box [80]; Kv7 opener ketamine induces an increase in SPT and reduces the immobility in FST [56]; activation of Kv7.4 by fasudil improve social interaction time and SPT [79]; opening of Kv7 by Lu AA41178 reduces the immobility in FST [78]
Kir	Kir4.1	Declined surface expression of Kir4.1 by ketamine in astrocytes improves the depressive behaviors in FST, LH and SPT [9, 130] Kir4.1 inhibitor Lys05 exerts rapid-onset antidepressant actions and releases the depressive-like behaviors in NST, FST and SPT [88]
	KATP	Inhibition of memantine on KATP lowers the immobility in FST and TST [135]; blockage of glibenclamide on KATP channels reduces the immobility in FST [111]; inhibition of gabapentin and baclofen on KATP channels declines the immobility in FST [114, 115]
K2P	TREK	Blockage of spadin and SID1900 on TREK-1 reverses the depressive-like behaviors assessed by FST, SPT, and open field tests [136] (or see the review by Borsotto et al. [124])

LH learned helplessness, FST the forced swim test, TST tail suspension test, SPT sucrose preference test, NSF novelty-suppressed feeding test.

As mentioned previously, 5-HT deficiency is considered as a main cause for the pathogenesis of depression. As such, maintaining the 5-HT levels in synaptic clefts by some ways, such as the blockage of 5-HT reuptake, is a potential strategy to treat this illness. Compared with wild-type mice, enhanced activity of serotonergic neurons of the midbrain dorsal raphe nucleus (DRN) in TREK-1 knockout mice leads to more 5-HT release to counteract the depressive symptom [127], meaning a regulatory role of K⁺ channels in 5-HT secretion. Moreover, TASK-3 channels, a potential antidepressant target [126], also function as a regulator in the releasing of 5-HT in serotonergic raphe neurons [128]. On the other hand, 5-HT serves to promote the GABA release in the entorhinal cortex (EC) through inhibiting TASK-3 channels in GABAergic interneurons [129], exhibitive of converse modulation of serotonin on K⁺ channels. Exploring the relationship between serotonin and K⁺ channels is essential for depression studies, providing valuable information for the involvement of K⁺ channels in depression and pharmacological actions of antidepressants.

Intriguingly, the downregulation of some K⁺ channels, such as Kv1, Kv2, Kv4, Kir and K_2P, is able to improve the depressive-like symptoms; however, the upregulation of Kv7 channels could generate a similar effect in the animal models of depression (Fig. 1). The exact mechanism underlying this discrepancy is still open, although corresponding authors in the related research present diverse possible explanations. After carefully screening those published articles, we propose that elevating neuronal excitability after the downregulation of related K⁺ channels, with the exception of Kv7 channels, facilitate the release of monoamine neurotransmitters, such as serotonin, from monoaminergic neurons, thereby increasing the levels of monoamines in the synaptic cleft (Fig. 1A). On the other hand, opening Kv7 channels by some chemicals like retigabine, however, suppresses the aberrant hyperexcitability in some brain tissues, such as ventral tegmental area (VTA) dopaminergic neurons [56, 77], with the significant remission of depressive-like symptoms (Fig. 1B). Additionally, enhanced activity of Kv7 channels by ketamine and retigabine in glutamatergic neurons of the ventral hippocampus generates sustained antidepressant effects [80], suggesting an involvement of glutamatergic neurons in the effects of antidepressants on K⁺ channels (Fig. 1C). As mentioned above, ketamine reduces the surface density of the astroglial Kir4.1 channel and accordingly presents the rapid and long-lasting beneficial effects in patients with a major depressive disorder [130]. The following process could be that reduced surface density of the astroglial Kir4.1 by ketamine ameliorates the hyperpolarization of nearby LHb neurons, and subsequently inhibits the burst firing of these neurons and opposite depressive-like symptoms [9], meaning that antidepressants impacting K⁺ channels regulate the glia-neuron interaction to opposite depression.

Fig. 1. Schematic of the possible explanations underlying the therapeutic effects of antidepressants acting on diverse K⁺ channels.

A During neuronal hypoactivity in the depression state, the opening of K⁺ channels, such as Kir, K_2P and Kv (without Kv7) channels, contributes to the efflux of K⁺ and a reduction of the release of excitatory or monoamine neurotransmitters. After applying the antidepressants like SSRIs, the onset of the blockage of these channels leads to the depolarization and neuronal hyperactivity, with the release of release of excitatory or monoamine neurotransmitters. B In some neural tissues like VTA, the elevated level of dopaminergic neurons in the synaptic cleft is attributed to the hyperactivity-mediated vesicle discharge in the depression state; however, antidepressants like retigabine reverse this process via activating the Kv7 channels, manifesting the therapeutic effects. C In glutamatergic neurons, application of antidepressants, such as retigabine and ketamine, also induce an enhanced activity of Kv7 channels to counteract depression.

Apparently, there exist the irreconcilable contradictions for some observations obtained from different investigators. As mentioned previously, retigabine possesses the pharmacological activity to inhibit Kv2.1 [54] and activate Kv7 [10], both of which are distributed in the overlapping brain area, such as the hippocampus [54, 131] and frontal cortex [47, 132], and the resultant effects secondary to treatment with this drug could be combinational and complex. Likewise, the M-currents carried by Kv7 channels in the lateral habenula (LHb) can be boosted by application of retigabine, which produces a hyperpolarization and accordingly reduces the firing rate and the excitability of LHb neurons [76]. Nevertheless, the activation of T-type Ca²⁺ channels by the hyperpolarization of LHb neurons arising from enhanced expression of Kir4.1 channels in nearby astrocytes serves to generate the burst firing that increases depressive-like behaviors [8, 9]. Clearly, the paradoxical role of the hyperpolarization in LHb neurons is evident, suggestive of the complex mechanisms underlying the antidepressant action of retigabine.

Interesting questions arising from experimental data are whether the alterations of expression and activity of K⁺ channels are the cause of depression or the accompanying symptom of this psychiatric illness. The antidepressant ketamine ameliorates the depressive behaviors by inhibiting Kv1 channels, and the upregulation of Kv1 channels is responsible for the firing attenuation of 5-hydroxytryptamine neurons in learned helplessness rats [39], suggestive of the alteration of these channels contributing to the pathogenesis of depression. As another example, the upregulation of astrocytic Kir4.1 expression in the animal model of depression indicates a causal role of Kir4.1 in depression [9]. On the other hand, stress exposure leads to the change of expression of Kv7.2 and Kv7.3 in the medial prefrontal cortex [133], suggesting an accompanying symptom. It appears that these issues are important for the mechanism underlying depression, while remaining to be determined.

Some K⁺ channels, such as Kir4.1 [9] and TASK-3 [126], are considered as therapeutic targets for depression. Nevertheless, alterations of other K⁺ channels may be responsible for the side effects of antidepressant drugs. For example, citalopram block on Kv1.5 channels could result in arrhythmogenic risk due to the prolongation of the action potential in atrial cardiomyocytes [34]. Notably, the neuronal hyperexcitability arising from the inhibition of antidepressants on various K⁺ channels, such as Kv2.1 [45, 48], is capable of elevating the incidence of seizure, another potential adverse effect occurring in depression-epilepsy co-morbidity [108]. Based on current data, it seems reasonable to assume that these exist polyhedral roles of K⁺ channels in antidepressant effects and depression. Apparently, there are still many unanswered questions in the field concerning K⁺ channels and their role in the treatment of depression. Therefore, more studies are definitely necessary to clarify the potential role of the K⁺ channels in pathogenesis of depression and the possible influences of antidepressants targeting these channels in the remission of depressive-like symptoms.

Author contributions

XTL wrote the manuscript.

Funding

This work was supported by the grants from research project of JCUT.

Competing interests

The author declares no competing interests.