Transient Potassium Channels: Therapeutic Targets for Brain Disorders

Transient potassium current channels (IA channels), which are expressed in most brain areas, have a central role in modulating feedforward and feedback inhibition along the dendroaxonic axis. Loss of the modulatory channels is tightly associated with a number of brain diseases such as Alzheimer’s disease, epilepsy, fragile X syndrome (FXS), Parkinson’s disease, chronic pain, tinnitus, and ataxia. However, the functional significance of IA channels in these diseases has so far been underestimated. In this review, we discuss the distribution and function of IA channels. Particularly, we posit that downregulation of IA channels results in neuronal (mostly dendritic) hyperexcitability accompanied by the imbalanced excitation and inhibition ratio in the brain’s networks, eventually causing the brain diseases. Finally, we propose a potential therapeutic target: the enhanced action of IA channels to counteract Ca2+-permeable channels including NMDA receptors could be harnessed to restore dendritic excitability, leading to a balanced neuronal state.


INTRODUCTION
Sequential synaptic inputs often cause dendritic spikes. Fast-acting membrane depolarization in the dendritic tree is converted into axo-somatic action potentials (APs) that do not only propagate down the axon but also back into the dendritic tree (Golding et al., 2001;Bender and Trussell, 2012). With distance from the soma, the amplitude of back-propagation action potentials (bAPs) is gradually attenuated by the inhibitory action of A-type transient potassium channels, I A for short (Hoffman et al., 1997;Frick et al., 2003). It is because the density of these channels gradually increases from the soma to distal dendrites. In addition to the role of I A on feedback propagation, the channels play a critical role in feedforward, pre-to-post-synaptic transmission and integration by suppressing dendritic spikes (Cai et al., 2004;Kim et al., 2007;Jung et al., 2008;Kim and Hoffman, 2008;Yang et al., 2015). Therefore, a dysfunction of I A channels can lead to loss of control over neuronal excitability, the hallmark of several diseases in the brain. In this review, we discuss (1) the distribution (mostly in human) and function (in animals and human) of I A channels in the brain, (2) the common down-regulation of I A channels in several brain diseases and (3) a potential I A channel-based therapeutic target.

SUBTYPES OF I A CHANNELS
The potassium channels form a large and diverse family of ion channels that are involved in establishing the resting membrane potential, determining the action potential waveform, regulating neurotransmitter release, and modulating rhythmic firing patterns (Llinas, 1988;Joho and Hurlock, 2009). There are four major classes of potassium channels in the brain: Calcium-activated, inwardly rectifying, leak, and voltage-gated potassium channel. Particularly, I A channels, one type of voltage-gated potassium channels, are characterized by closed, opened, and inactivated channel states depending on the membrane voltage. I A channels form a large macromolecular complex comprising four ion-conducting alpha-subunits and beta-subunits (either being cytoplasmic or transmembrane) with auxiliary regulatory and supporting proteins (Leicher et al., 1998;Grizel et al., 2014). This voltage-gated potassium channels usually have a homotetrameric structure (with all alphasubunits being identical), but some of them are heterotetrameric (with two or more non-identical alpha-subunits) (Sokolova, 2004;Grizel et al., 2014). Kv3.3,Kv3.4,Kv4.1,Kv4.2,and Kv4.3) that are divided into discrete families on the basis of sequence similarity form an ion pore and infrastructure of the channel. The alpha-subunits determine the fast kinetic property of I A channels, rapidly activating and inactivating (Coetzee et al., 1999;Covarrubias et al., 2008). The alpha-subunits are major components on which pharmacological agents target. Most generally used 4aminopyridine (4-AP) not only blocks the subtypes of I A channels such as Kv1.4 and Kv4.x, but also other subtypes of voltage-gated potassium channels (Gutman et al., 2005). Alternatively, biotoxins isolated from venoms of the tarantula spider (PaTx1 and 2, HmTx1, HpTx1 and 2), sea anemone (BDS I and II) and scorpion (alpha-KTx15 subfamily) have relatively higher selectivity on the specific subtypes of I A channels (Sanguinetti et al., 1997;Diochot et al., 1999;Escoubas and Rash, 2004;Yeung et al., 2005;Prestipino et al., 2009). Plus, some components (Diclofenac and BmP02) have been identified to activate I A channels (Liu et al., 2005;Wu et al., 2016a,b).
The activity and expression of I A channels can be modulated by certain post-translational modifications of phosphorylation and palmitoylation (Wulff et al., 2009) with various protein kinases such as Ca 2+ /calmodulin-dependent protein kinase II (CaMKII), cAMP-dependent protein kinase (PKA), protein kinase C (PKC), and mitogen-activated protein kinase (MAPK) (Jerng et al., 2004). CaMKII phosphorylation of Kv1.4 is reported to regulate the inactivation gate of I A channels arising from Kv1.4 (Roeper et al., 1997;Varga et al., 2004). Similarly, CaMKII phosphorylated Kv4.2 upregulates both the expression level of Kv4.2 proteins and the peak current of Kv4.2 (Roeper et al., 1997;Varga et al., 2004). Meanwhile, either PKA or PKC activation decreases the opening of I A channels encoded mostly by Kv4.2; accordingly, their activation increases the amplitude of bAPs in distal dendrites (Hoffman and Johnston, 1998). In other studies, PKC reduces inactivation of I A channels encoded by Kv3.3 and Kv3.4 (Covarrubias et al., 1994). Thus, the complex structure of I A channels likely determines their class and functions.

CLOSE ASSOCIATION OF I A WITH SEVERAL BRAIN DISEASES
Subtypes of I A channels are differentially distributed in the brain. The macroscopic expression pattern of I A channels is now investigated with human brain where clinical value can be highlighted. Kv1.4 is predominantly expressed in prefrontal cortex, cerebellum peduncles, pituitary and pineal gland in a ranking order; kv3.3 in prefrontal cortex, pineal gland, pituitary and cerebellum peduncles; kv3.4 in cerebellum peduncles, pituitary, prefrontal cortex and pineal gland; kv4.1 in pineal gland, prefrontal cortex, pituitary and cerebellum peduncles; kv4.2 in cerebellum peduncles, cerebellum, prefrontal cortex and hypothalamus; kv4.3 in subthalamic nucleus, pineal gland, cerebellum peduncles and prefrontal cortex (Figure 1 and Table 1). Furthermore, there are significant functional consequences of I A on brain disorders ( Table 2). Interestingly, the mRNA expression of I A channels is largely downregulated in the condition of brain dysfunction ( Table 3). In the following sections, we discuss the tight involvement of I A channels in brain diseases, such as Alzheimer's disease, epilepsy, fragile X syndrome (FXS), Parkinson's disease, chronic pain, tinnitus and ataxia.

ALZHEIMER'S DISEASE
Alzheimer disease (AD) is a neurodegenerative disease characterized by progressive deterioration of cognitive function. The early symptoms of AD are found in language, perception, movement, and memory. In the final phase, it causes widespread neuronal death, leading to severe memory loss, emotional disturbance and language dysfunction. Thus, patients' independent living is impossible and relies on caregivers (Chen, 2005). One potential cause of AD appears to be the abnormal folding of the proteins Tau and/or Amyloid beta (A β ) (Angulo et al., 2004). Interestingly, I A channels are downregulated with the plaque of Tau and/or Amyloid beta (A β ), as reported in a study of cultured hippocampal neurons from an AD animal model (Hokama et al., 2014). This is further evidenced by other studies that A β -induced dendritic hyperexcitability is known  (Su et al., 2004). Note that expression of Kv1.4 is elevated in prefrontal cortex, cerebellum peduncles, pituitary and pineal gland; Kv3.3 in prefrontal cortex, pineal gland, pituitary, cerebellum peduncles; Kv3.4 in cerebellum peduncles, pituitary, prefrontal cortex and pineal gland; Kv4.1 in pineal gland, prefrontal cortex, pituitary and cerebellum peduncles; Kv4.2 in cerebellum peduncles, cerebellum, and prefrontal cortex and hypothalamus; Kv4.3 in subthalamic nucleus, pineal gland, cerebellum peduncles, pituitary and prefrontal cortex (when only top 25% of expression levels is considered).
to arise from a sustained increase in intracellular Ca 2+ and dysfunction of I A channels in dendrites on hippocampal CA1 neurons of rodents (Chen, 2005;Morse et al., 2010;Hall et al., 2015). The loss of Ca 2+ homeostasis in dendrites is attributed to abnormally enhanced back-propagation action potentials (bAPs) resulting from dysfunction of I A channels. In other words, the dysfunction of I A channels induced by A β misfolding and aggregation results in excessive dendritic Ca 2+ influx, leading to the subsequent destruction of synapses and ultimately cell death. In contrast, a study reported that up-regulation of the kv3.4 channel (but not caused by A β deposition) causes alterations of the neuron activity in early AD (Angulo et al., 2004). In any cases, it is likely that the alteration of I A channels disrupts synaptic function of neurons, being involved in the clinical manifestation of AD.

EPILEPSY
Epilepsy is characterized by repeated convulsive seizures. Seizures are known to be caused by excessive excitation or suppressed inhibition in neural networks Cox, 2008, 2011;Park et al., 2018). One-third of epilepsy is a generalized seizure often accompanied by loss of consciousness, affecting the entire brain. Two-thirds of epilepsy occurs as focal seizures. Then, it often proceeds into the generalized seizures (Appleton et al., 2012). Epileptic seizure occurs as the result of environmental factors such as brain injury, stroke, brain tumors and infections, and genetic factors. Also, many of the patients are often accompanied by psychological symptoms such as anxiety and depression (Engel, 2001;Vadlamudi et al., 2003;Berkovic et al., 2006). In any case, epileptic seizures are strongly associated with the downregulation of I A channels (Tables 2, 3). In fact, the drugs that downregulate the activity of I A channels have been recognized as proconvulsants and, therefore, inhibiting the activity and expression of I A channels is the well-established models of epilepsy (Bernard et al., 2004;Su et al., 2008). Indeed, in adult Wistar rats with spontaneous behavioral seizures, the expression of Kv4.2 is decreased in the hippocampus (Lei et al., 2016). Also, in a seizure induction study, pregnant rats were infused with a neurotoxin, MAM (methylazoxymethanol). This neurotoxin reduced DNA synthesis and caused neuronal heterotopia in the hippocampus of the rat fetuses. Postnatally, their hippocampal neurons showed I A channels (encoded by Kv4.2) dysfunctional and exhibited hyperexcitability. As a result, their seizure thresholds were lower than those in normal rats (Castro et al., 2001). With status epilepticus induced by kainic acid, the rat model is associated with the activation of extracellular signal-regulated kinase (ERK), an enzyme that causes the structural and functional impairment of dendritic Kv4.2-I A channels (Lugo et al., 2008). Apart from the expression level change of Kv4.2 in animal seizure model, the hippocampus of Kv4.2 knockout (KO) mice increases the sensitivity to convulsion-inducing stimulation compared to that in wild type mice, indicating the depletion of Kv4.2 is associated with deviant network excitability and increased seizure susceptibility (Barnwell et al., 2009). Consistent with these findings in animals, a study with surgical tissue from patients with hippocampal sclerosis-induced epilepsy showed that the dendritic regions of the hippocampus exhibit the decreased expression of Kv4.2. Similarly, the patient with one temporal lobe epilepsy carried a gene mutation of Kv4.2 channels (Singh et al., 2006;Aronica et al., 2009). Thus, these results suggest that the functional and/or structural reduction of I A channels is associated with abnormal neuronal excitability and epilepsy in the temporal lobe in both animals and humans.

FRAGILE X SYNDROME
Fragile X syndrome (FXS) is the most common heritable mental disability. In general, the person with FXS has a long, narrow face, big ears and flexible fingers, often having autistic behaviors. FXS is associated with the dysfunction of the protein fragile X mental retardation protein (FMRP) encoded by a gene FMR1 on the X chromosome, a regulator of protein synthesis in axons and dendrites (Deng et al., 2011). FMRP possessing multiple RNA binding domains binds to polyribosome complexes and regulates protein synthesis (Ashley et al., 1993;Siomi et al., 1993;Brown et al., 2001;Greenough et al., 2001;Li et al., 2001;Todd et al., 2003;Darnell et al., 2005). Loss of FMRP causes a variety of symptoms such as sensory hypersensitivity and repetitive/excessive behavior which are hallmarks of FXS (Musumeci et al., 1999;Bureau et al., 2008;Bhakar et al., 2012;Yang et al., 2014b;Zhang et al., 2014). Recent findings show that FMRP is associated with the expression and activation of Kv4.2 channels in the mouse hippocampus ( Table 2). There are two different conclusions on the interactional relation between FMRP and Kv4.2 channels. On one hand, in an fmr1 KO mouse model, the dendritic protein level of Kv4.2 in the hippocampus was found to be reduced, possibly causing neural hyperexcitability which underlies a plausible mechanism of FXS (and its associated epilepsy) (Gross et al., 2011). Also, the finding is supported by a following study with fmr1 KO mice that Kv4.x-mediated currents are reduced in dendrites, but not soma, of hippocampal CA1 pyramidal neurons, and accordingly the amplitude of back-propagation action potentials (bAPs) is increased in distal dendrites (Routh et al., 2013). On the sharp contrary, the local translation of Kv4.2 in CA1 dendrites increases in fmr1 KO mice, indicating that FMRP is a negative controller of dendritic Kv4.2 (Lee et al., 2011). In any cases, dendritic I A channels have been found important in FXS. Further studies are required to explore the dynamic role of dendritic I A channels in FXS.

PARKINSON'S DISEASE
Parkinson's disease (PD) is one of the most common neurodegenerative disorders characterized by an impairment of dopaminergic systems. Patients who suffer from PD have problems with voluntary behavior and thinking. Dementia and emotional problems such as depression and anxiety can be gradually developed over time. The cause of PD has several candidates: (1) environmental factors such as exposure to pesticides and a history of head injury and (2) genetic factors related to certain genes including SNCA, LRRK2, GNA, PRKN, PINK1, PARK7, VPS35, EIF4G1, DNAJC13, and CHCHD2 (Kalia and Lang, 2015). The pathological hallmark of PD is known to be the loss of dopaminergic neurons in the substantia nigra (SN). Another pathological feature of PD is the formation of protein aggregates called Lewy bodies (Davie, 2008). Furthermore, PD is also associated with dysfunction of I A channels (Lawson, 2000). For instance, the inhibition of Kv4.3 channels increases the spontaneous activity of nigral dopamine (DA) neurons and disrupts DA release, ultimately inducing the onset of PD (Subramaniam et al., 2014). In the research, the enhanced neuronal firing was observed only in DA neurons of the SN, but not in DA neurons of the ventral tegmental area. These animal studies are consistent with findings that I A expression is altered in PD patients, as shown in Table 3.

CHRONIC PAIN
Chronic pain severely affects most daily activities, disrupting social life with financial burden (Breivik et al., 2006;Bouhassira et al., 2008;Henschke et al., 2015). Two symptoms dominate the disease: (1) inflammatory pain caused by a tissue reaction to injury or infection and (2) neuropathic pain from nerve injury, which prevails after the primary damage has healed (Woolf and Salter, 2000). At the root of chronic pain lies the activity of sensory neurons in the dorsal root ganglion (DRG) and spinal dorsal horn (DH). These neurons are responsible for transmitting nociceptive information from the periphery to the brain. Chronic pain is largely associated with neuronal hyperexcitability of the DRGs and DH neurons. Interestingly, several lines of evidence indicate that these neurons' hyperexcitability is caused by reduced inhibitory currents and/or expression of I A channels in the DRGs and DH neurons (see Table 2). For example, in a rat diabetic neuropathy model, the reduction of 4-aminopyridine (4-AP)-sensitive I A currents and the decreased mRNA expression level of I A alpha-subunits; Kv1.4, Kv4.2, and Kv4.3 in the medium and large-diameter DRG neurons leads to the hyperexcitability of the neurons through BDNF activity on TrkB receptor (Cao et al., 2010). In addition, by peripheral nerve injury resulting in neuropathic pain, hypoacetylated histone H4 at the location of Kv4.2-NRSE is associated with the decreased mRNA expression of Kv4.3 in mouse DRG neurons, and the reduction of Kv4.3 is implicated in neuronal hyperexcitability (Kim et al., 2002;Uchida et al., 2010). Finally, DRGs in a neuropathic rat model show reduced Kv4.3 protein expression after spinal nerve ligation to DRGs, which drives mechanical hypersensitivity (Chien et al., 2007). Apart from DRGs, neurons in the DH of animals with formalin-induced inflammatory chronic pain are Kv4.2-deficient and exhibit increased excitability, resulting in increased sensitivity to tactile and thermal stimuli   (Hu et al., 2006). Moreover, in a rat model of cystitis with visceral hypersensitivity, the decrease in 4-aminopyridine (4-AP)-sensitive I A current as well as the decreased expression of Kv1.4 channels are observed in DH neurons, leading to neuronal hyperexcitability (Hayashi et al., 2009). In fact, the role of I A channels in chronic pain has so far been studied mostly in the DRGs and DH neurons. It would be of great interest to investigate the role of I A channels along the somatosensory pathway.

TINNITUS
Tinnitus is a ringing in the ear that can be perceived even in the absence of external acoustic stimuli. Tinnitus is induced mainly by environmental factors such as traumatic noise, aging and ear infection. Environmental damage to cochlear hair cells alters the activity of the neural networks along the ascending auditory pathway, increasing the ratio of excitation to inhibition in central auditory neurons. Our group previously reported that auditory neurons in animals with noise-induced tinnitus exhibit intrinsic hyperexcitability. Also, in our preliminary data, noiseinduced tinnitus is probably associated with the downregulation of Kv1.4 channels in rat auditory neurons (Yang et al., 2012;Tetteh et al., 2017). Thus, we believe this hyperexcitability to be due to the abnormal expression and/or function of I A channels ( Table 2).

ATAXIA
Ataxia is a neurodegenerative disorder caused by neural atrophy along the cerebellum-spine axis. Patients suffer from movement discoordination such as progressive ataxic gait and limb movements as well as difficulties with speech and eye movements (Duarri et al., 2013). Ataxia has three types depending on the parts of the dysfunction: (1) cerebellar ataxia due to dysfunction of the cerebellum; (2) sensory ataxia due to dysfunction of the sensory system including dorsal columns of the spinal cord, thalamus and parietal lobes; (3) vestibular ataxia due to dysfunction of the vestibular system. The main causes of ataxia are the focal lesion of corresponding CNS region, the exogenous substance like ethanol and vitamin b12 deficiency. Besides, genetic factors affect the degeneration of the cerebellum and/or the spine; 22 genes related to spinocerebellar ataxia (SCA) have been identified so far (Durr, 2010;Sailer and Houlden, 2012). The mRNA of Kv4.3 channels is highly expressed in the cerebellum of humans (Figure 1). Indeed, in animals and human patients with ataxia (  (Duarri et al., 2012(Duarri et al., , 2015Lee et al., 2012; Table 2). In addition to Kv4.2 channels, Kv3.3 is the causative gene of SCA type 13, an autosomal dominant neurological disorder (Zhao et al., 2013). There are lines of evidence that SCA type 13 is caused by point mutations in the coding region of the Kv3.3 gene on chromosome  We analyzed human and animal gene expression data from GEO (http://www.ncbi.nlm.nih.gov/geo). Data were analyzed by GEO2R (http://www.ncbi.nlm.nih.gov/geo/ geo2r/). GSE6834 and GSE36980 were analyzed for Alzheimer's disease. GSE20168 and GSE20291 were analyzed for Parkinson's disease. GSE6834 and GSE28674 were analyzed for epilepsy, GSE57034 and GSE71034 were analyzed for fragile X syndrome. GSE69619 was analyzed for Chronic pain. GSE61019, GSE55177, and GSE61908 were analyzed for Ataxia. Tinnitus data was analyzed by Affymetrix. Statistical significance, * p < 0.05; * * p < 0.01.
19q13 (Herman-Bert et al., 2000;Waters et al., 2005Waters et al., , 2006Zhao et al., 2013). Moreover, the expression and function of Kv3.3 channels are altered in CHO cells transfected with mutated SCA13 as well as mRNA expression of Kv3.3 in animal with cerebellar ataxia (Herman-Bert et al., 2000;Waters et al., 2005Waters et al., , 2006Zhao et al., 2013); Table 3). Taken all together, the alteration of I A channels is deeply involved in several types of ataxia.

CELLULAR MECHANISMS AND THERAPEUTIC TARGETS OF I A
I A channels and Ca 2+ -permeable channels including NMDA receptors (NMDARs) and voltage-gated calcium channels (VGCCs) are abundant in dendrites, and their interaction determines the level of dendritic excitability and synaptic plasticity (Frick et al., 2003;Cai et al., 2004; FIGURE 2 | The contribution of I A channels on neuron excitability and brain disease. (Ai) A schematic of two photoactivated locations (distal and apical trunk) of a hippocampal neuron. (Aii) Examplar gluEPSPs (or glutamate-evoked EPSPs) in response to photoactivation in the distal and apical trunk. Note that there is a sudden increase in the response, being referred to as a NMDA-dependent dendritic spike and Ca 2+ -dependent dendritic spike in the distal and apical trunk, respectively. (Aiii) The blockage of 4-AP (3 mM) sensitive-I A channels potentiates the responses of NMDARs but not AMPARs. The left traces illustrate NMDAR (and VGCC)-mediated EPSPs, respectively, in a thin oblique dendrite or in an apical dendrite over an input strength; weak (averaged 6.3 µJ) and strong (averaged 11.2 µJ) energy. The right traces illustrate AMPAR-mediated EPSPs in a thin oblique dendrite over an input strength; weak (averaged 2.4 µJ) and strong (averaged 4.3 µJ) energy. (B) A putative model regarding the contribution of I A channels on neuron excitability and brain disease. It shows a molecular mechanism that dendritic impairment of I A channels can augment AP generation. The data was reused with permission from Yang et al. (2015). Losonczy and Magee, 2006;Kim et al., 2007;Jung et al., 2008;Losonczy et al., 2008;Wang et al., 2014;Yang et al., 2015). Ca 2+ channels are kinetically slow to activate and inactivate (Figure 2A; Yang et al., 2015). Ca 2+ dendritic spikes are regenerative and, thus, difficult to regulate once initiated.
To efficiently suppress Ca 2+ -mediated hyperexcitability, an active filter is required just before it occurs. Rapid action of I A channels may play such a role as the active filter. For instance, NMDA spikes are mostly induced in distal dendrites of cortical neurons where most synaptic events occur (Larkum et al., 2009;Yang et al., , 2014aYang et al., , 2016. In turn, VGCCs magnify the NMDA-mediated dendritic spikes in apical dendrites, propagating the spikes to the axo-somatic sodium spikes zone for AP generation. Here, I A channels could suppress the anterograde propagation of local Na + -and Ca 2+ -mediated signals from dendrites as well as the backward spread of APs into dendrites (Losonczy and Magee, 2006;Losonczy et al., 2008). When I A channels are not active, the conductance of Ca 2+ -permeable channels can easily cause the membrane to slip toward depolarization, leading to synaptic enhancement in the axon terminals. In fact, internalization-induced reduction of I A channels enhances synaptic plasticity while increased I A channels do not display synaptic plasticity, suggesting a synaptic role of Ca 2+ -permeable channels heavily relying on the activity of I A channels (Kim et al., 2007;Jung et al., 2008). The activity of Ca 2+permeable channels which are inflated by inactive I A channels can often cause excessive neuronal excitability, resulting in an imbalance of excitation and inhibition in the whole network. We view this mechanism as a common cause for brain diseases such as Alzheimer's disease, epilepsy, FXS, Parkinson's disease, chronic pain, tinnitus and ataxia ( Figure 2B). Most drugs that are currently used to treat these diseases directly modulate ion channels and receptors that affect intrinsic and synaptic properties. However, as most of these ion channel-related drugs act across the whole brain, they produce a wide range of non-specific side effects, leading to the discontinuation of many potential treatments in the past. For instance, antagonists of NMDA receptors and voltagesensitive sodium channels could in principle be used to suppress hyperexcitability. However, they also disrupt the physiological function of unintended brain areas. Furthermore, the drugs that activate GABA receptors also suppress other brain areas, often resulting in sedating and over-suppression effects that significantly reduce the quality of life. One way to circumvent these limitations of classical excitatory or inhibitory drugs would be to develop chemical compounds that "indirectly" modulate Na + channels, NMDA or GABA receptors. A recent study reports that the application of 4-AP restores low-and highthreshold dendritic spikes in the distal and apical dendrites, respectively, which can be driven by Ca 2+ -permeable channels (NMDARs and/or Ni 2+ -sensitive voltage-gated Ca 2+ channels) in rat hippocampal CA1 neurons (Yang et al., 2015). A similar study demonstrates that glutamate-mediated dendritic spikes are enhanced by 4-AP treatment and normalized by D-AP5 (an NMDA receptor antagonist) in mouse L3 pyramidal neuron (Biro et al., 2018). Thus, Ca 2+ -permeable channels and I A channels play a counterbalancing role in regulating neuronal excitability (Figure 2). In this respect, I A channels as an indirect modulator can be predominantly activated in a hyperexcitable condition caused by excessive activity of Ca 2+ -permeable channels, yet having minimal impact on the resting condition of neurons. Furthermore, given homeostatic regulation of neural networks, the lack of sensory or modulatory inputs likely causes the increment of excitation and/or decrement of inhibition in the affected neurons. These changes seem to be required for maintaining network excitability for neuron survival. In the hyperexcitable condition of the affected neurons often causing brain disorders, for example, an agonist of I A channels can function to modulate neuronal excitability by counteracting against Ca 2+ -permeable channels/receptors. Thus, I A channels as a therapeutic target may open an avenue for more sustainable treatments of brain disease with minimal interference in normal physiological function, leading to longterm clinical benefits. However, it is notable that there are limitations incurred when attempting to link the animal disease model to human neuropathology in the context of the expression and function of I A channels. It is currently unclear whether the suggested therapeutic mechanism of I A channels with animal disease models can apply to human diseases having a different neural network. Further research is required to delineate the precise contributions of I A channels to brain diseases in human.