Neurodegenerative disorders (NDs) are one of the most devastating types of chronic diseases and lead to a significant social and medical burden on society. With the growing elderly population, the number of patients with NDs is also increasing. Although many pharmaceutical companies are struggling to develop novel therapeutics for neurological diseases, some of the world’s leading pharmaceutical companies have declared their abandonment of the development of therapeutics for NDs. Alzheimer’s disease (AD) is the most common ND, which accounts for 60–70% of all dementia (Association, 2016). Nevertheless, no new treatments for AD have been developed in over a decade. Some of the reasons for the difficulty in treating NDs are the combination of complex causative factors and irreversible structural and functional damage of neurons.

Neurodegenerative disorders are typically progressive, late-onset disorders, and aging is the greatest risk factor (Wakabayashi et al., 2014). In addition, genetic and environmental factors not only contribute to their pathogenesis independently but also interact with each other to increase their effects. The pathogenesis of NDs involves an initial toxic insult and consequences of the secondary pathological damage. The most primary causative hypothesis of AD is the intraneuronal accumulation of amyloid-beta (Aβ) and hyperphosphorylated tau protein (Iqbal et al., 2010; Murphy and LeVine, 2010). Parkinson’s disease (PD) is caused by the degeneration of dopaminergic neurons in the substantia nigra par compacta (SNpc), with subsequent dopamine deficiency (Michel et al., 2013; Kalia and Lang, 2015). A type of hereditary ND, Huntington’s disease (HD) is an autosomal dominant disorder caused by CAG repeat expansion within the Huntington (HTT) gene (Kremer et al., 1994). The third common ND after AD and PD is amyotrophic lateral sclerosis (ALS) that is characterized by the deterioration of motor neurons.

Neurodegenerative disorders such as AD, PD, HD, and ALS are distinguished by clinical symptoms and specific neuronal sites with distinct pathology. However, apparent clinical symptoms are manifested only after extensive pathological damage, with significant neuronal and synaptic loss. Eventually, the contribution of individual insults reaches a common end state, which causes severe impairments in the function and plasticity of neuronal and glial cells (Rasband, 2016). Over the past few decades, there has been considerable effort to understand the pathogenesis of NDs. To date, a number of studies have reported that oxidative stress, ER (endoplasmic reticulum) stress, abnormal Ca²⁺ homeostasis, protein misfolding, aggregation, neuroinflammation, and mitochondrial dysfunction are highly related to neuronal damage. The relationship between them, however, has not been completely elucidated. Besides, even though Aβ is still a compelling candidate in the pathogenesis of AD, the latest experiments raise doubts about the Aβ hypothesis and Aβ -based drug development for AD (Du et al., 2018). Therefore, it is necessary to find the missing links amongst the risk factors of NDs and to discover new therapeutic targets based on novel mechanisms.

Ion channels are key determinants of brain function, since the physiological function of neurons is to carry information or impulses via electrical signals (action potentials) to communicate with each other at synapses. Thus, neurological channelopathies have been identified mainly in voltage-gated and ligand-gated channels or receptors that result from genetically determined defects in their function. However, based on patients with progressive NDs with adulthood manifestations, we have focused on the age-related susceptibility to environmental toxins and chemicals. Recently, emerging evidence has indicated that transient receptor potential (TRP) channels, ubiquitously expressed throughout the brain (Sawamura et al., 2017), play a significant role in the regulation of physiological functions, as well as in reactive oxygen species (RO)-related human diseases. Based on the polymodal activation of TRP channels acting as cellular sensors, many researchers are investigating their activation mechanisms (Takada et al., 2013). Here, we review the potential involvement of TRP channels in NDs, where oxidative stress and disrupted Ca²⁺ homeostasis have been characterized with respect to pathological consequences in neuronal apoptosis. Second, we discuss the latest findings in the field of TRP to provide insight into the research and quest for alternative therapeutic candidates for the treatment of NDs.

Ca²⁺ homeostasis is crucial to the normal physiological functions of neurons, such as neuronal survival, growth, and differentiation. Hence, long-lasting Ca²⁺ dyshomeostasis can eventually lead to neuronal loss. Accumulated evidence strongly implicates that abnormal Ca²⁺ levels stimulate dysregulation of intracellular signaling, which consequently induces neuronal cell death (Barnham et al., 2004). Therefore, disruption of Ca²⁺ homeostasis in neuronal cells leads to ROS generation and ATP depletion, following the mitochondrial dysfunction in NDs such as AD, PD, HD, and ALS. Interestingly, a close correlation between the increase in [Ca²⁺]_i and other pathogenic mechanisms has been reported, such as Aβ deposition (Demuro et al., 2010), imbalance between ROS and antioxidant function (Gorlach et al., 2015), and mitochondrial dysfunction (Contreras et al., 2010; Pivovarova and Andrews, 2010).

Some reports have shown bidirectional crosstalk between amyloid pathology and the Ca²⁺ pathway. Most studies reported that Aβ increases intracellular Ca²⁺ levels by inducing ER Ca²⁺ depletion (Suen et al., 2003; Abramov et al., 2004b; Ferreiro et al., 2008). Abramov et al. (2004a) identified that Aβ causes Ca²⁺-dependent oxidative stress by the activation of NADPH oxidase in astrocytes and that the reduced antioxidant activity induces neuronal death. Recently, Calvo-Rodriguez et al. (2019) suggested that Aβ oligomers exacerbate Ca²⁺ remodeling from ER to mitochondria in aged neurons but not in young neurons. Conversely, Itkin et al. (2011) argued that Ca²⁺ stimulates the formation of Aβ oligomers, leading to neuronal toxicity in AD.

The imbalance between ROS production and antioxidant defenses results in the excessive accumulation of ROS and oxidative stress. Since aged neurons with low antioxidant capacity are more vulnerable to oxidative insults (Chen et al., 2012), ROS overproduction can chronically lead to irreversible oxidation (Ivanova et al., 2016). Oxidative stress also causes mitochondrial dysfunction, which itself aggravates ROS generation. Moreover, the opening of the mitochondrial permeability transition pore (mPTP) and the release of cytochrome c into cytoplasm activate pro-apoptotic caspases (Guo et al., 2013). Oxidative stress and mitochondrial dysfunction are well known to be related to an increase in cytosolic Ca²⁺ levels that underlies the pathogenesis of AD (Cenini and Voos, 2019), PD (Blesa et al., 2015), HD (Zheng et al., 2018), and ALS (Carri et al., 2015). Interestingly, the mitochondrial metabolic state can affect the Mg²⁺ concentration of both the matrix and the cytoplasm, where Mg²⁺ interferes with mitochondrial Ca²⁺ transport and mitochondrial ATP generation (Llorente-Folch et al., 2015). Since the efficient removal of [Ca²⁺]_i requires ATP, impairment of mitochondrial ATP generation prevents Ca^{2 +} pumps from operating both in the plasma membrane and in the ER (Ott et al., 2007). Thus, dysregulation of Ca²⁺ signaling is one of the key processes in early stage neuronal loss. In spite of its significance, the mediator of such aberrant Ca²⁺ increase and its source are not fully understood. To test the effectiveness of preventing Ca²⁺ overload for ND therapy, a variety of appropriate channel candidates should be further examined by developing channel-specific drugs for new channel targets. Last but not least, experimental data on existing drugs also need to be reevaluated.

Oxidative stress can directly modulate the gating properties of ion channel proteins. Pathological mechanisms underlying the dysregulation of ion channels by oxidation have been previously proposed in a variety of diseases, especially cancer (Reczek and Chandel, 2018) and NDs (Gorlach et al., 2015). Under normal conditions, defensive antioxidants can protect or repair the damage caused by oxidation. However, the target proteins are retained in their oxidized forms and are activated as long as the antioxidant activity is reduced. To date, several studies have shown that oxidative stress is involved in the modulation of activities of voltage-gated Ca²⁺ (Gorlach et al., 2015; Ramirez et al., 2016), Na⁺, and K⁺ (Sesti et al., 2010) channels and ligand-gated receptors such as NMDA (Kamat et al., 2016), AMPA (Joshi et al., 2015), GABA (Bradley and Steinert, 2016), and RyR (Zima and Mazurek, 2016). However, there is limited evidence that directly identifies the oxidative modification of a channel protein based on molecular mechanisms.

Since TRP channels are non-selective, Ca²⁺-permeable channels that can be opened at resting membrane potential in response to various stimuli, we focused on TRP channels. The activation of TRP channels consequently changes membrane depolarization toward the action potential threshold. When TRP channels open, they allow sodium and calcium into the cytoplasm, which subsequently triggers the opening of voltage-dependent Ca²⁺ channels. This is why TRP channels are upstream risk factors, ahead of voltage-dependent channels (Numata et al., 2011). Therefore, the hyperactivation of TRP channels is responsible for neuronal excitotoxicity, which is closely associated with NDs. In the following section, we will address the physiological and pathological roles of TRP channels in neurons through the recent studies related to TRP channels and our study of the TRPC5 channel.

As mentioned above, TRP channels are widely expressed in almost every mammalian cell, predominantly in the brain. TRP channels can be activated by diverse stimuli ranging from temperature, mechanical or osmotic stress, chemical compounds, and redox modification (Sawamura et al., 2017; Samanta et al., 2018). Based on sequence homology, the TRP superfamily is divided into six subfamilies in mammals: TRPC (classical or canonical; seven sub-members), TRPM (melastatin; eight sub-members), TRPV (vanilloid; six sub-members), TRPA (ankyrin; one sub-member), TRPP (polycystin; three sub-members), and TRPML (mucolipin; three sub-members) (Figure 1). Notably, most TRP channels (except TRPM4 and TRPM5) are non-selective channels with consistent Ca²⁺ permeability (Guinamard et al., 2011). TRP channels are tetrameric protein complexes that can be assembled into homomeric or heteromeric channels, either with the same subfamily members or with the other subfamily members. Thus, when TRP channels assemble with different subunits, further heterogeneity diversifies their functions. In addition to the physiological roles of TRP channels in neurons, a number of studies regarding the pathological functions have been reported. Intracellular Ca²⁺ influx through TRP channels is involved either in neuronal survival or death and is discussed with respect to the different TRP channel families in the following sections (Bollimuntha et al., 2011).

Various causes of NDs have been reported recently. However, most of the brain lesions in ND present alongside several pathological environments, such as the presence of ROS, impaired antioxidant systems, and disrupted Ca²⁺ homeostasis (Di Meo et al., 2016). In consideration of the results discussed above, the regulation of TRP channels plays a key role in the Ca²⁺-dependent neuronal death in NDs. The involvement of TRP channels in NDs is summarized in Figure 1. To investigate the role of TRP channels in NDs, TRP channel antagonists and TRP-KO mice have been generated and utilized. TRP KO mice exhibit behavioral and neurological phenotypes.

TRPC3 is required for slow excitatory postsynaptic potential in cerebellar Purkinje cell synapses, and consequently, severe ataxic phenotypes have been shown in TRPC3 KO mice. In contrast to TRPC1/4 double-KO or TRPC1/4/6 triple-KO mice, TRPC3 KO mice show movement deficits of the hind-paws (Hartmann et al., 2008). In addition, TRPC4 plays a role in fear and anxiety-related behaviors. TRPC4 KO mice show innate fear responses in elevated plus maze and open-field tests. These fear responses result from mGluR-mediated EPSC in the lateral nucleus of the amygdala neurons (Riccio et al., 2014). TRPC5 also plays an essential role in fear-related behavior. In addition, disruption of burst firing in the potent muscarinic antagonist pilocarpine-induced seizure in TRPC5 KO mice was reduced. Seizure-induced neuronal loss in the hippocampal region was also reduced in TRPC5 KO mice (Phelan et al., 2013). TRPM2 KO mice exhibited disturbed EEG rhythms and bipolar disease-related behavior, including impairment of social behavior and increased anxiety (Jang et al., 2015). TRPM7 KO mice exhibited clasping, tremors, and slow movement associated with Mg²⁺ deficiency (Ryazanova et al., 2010).

TRPC KO mice combined with ND models have also been generated. In models of AD, TRPC6 modulates cleavage of APP by gamma secretase and APP (C99) interaction with PS1 (Wang et al., 2015). TRPC6 overexpression in APP/PS1 mice results in a reduction of Aβ accumulation in the hippocampus (Table 3). Therefore, TRPPC6 overexpression improves spatial learning and memory in APP/PS1 mice. In addition, the expression of the inflammatory factors TNF-α, IL-1β, COX-2, and IL-6 is regulated by levels of TRPC6 via Aβ, and levels of TRPC6 are increased by Aβ via NF-κB in BV-2 microglia cells (Lu et al., 2018). TRPM2 expression is involved in synapse loss, microglial activation, and spatial memory deficits in APP/PS1 mice (Ostapchenko et al., 2015). Activation of TRPV1 channels is required to trigger long-term depression at interneuronal synapses (Gibson et al., 2008) and prevents Aβ-involved impairment of functional networks in the hippocampus (Balleza-Tapia et al., 2018). Astrocytic Ca²⁺ hyperactivity is induced by Aβ oligomers via TRPA1 in the hippocampus. Moreover, astrocyte hyper-excitability is replaced by CA1 neuronal activity in APP/PS1 mice (Lee et al., 2016). Moreover, TRPA1 regulates astrocyte-derived inflammation in APP/PS1 mice. TRP channel antagonists regulate the production of ROS, APP processing, and Aβ accumulation. The TRPV4 antagonist HC-067047 attenuates the H₂O₂-induced Ca²⁺ influx (Suresh et al., 2015). Aβ-mediated cell damage was attenuated by treatment with TRPV4 blockers ruthenium red and gadolinium chloride (Bai and Lipski, 2014).

TRPC1, TRPC3, TRPM2, TRPM7, and TRPV1 have been shown to be involved in PD. TRPC1 activation reduces dopaminergic neuronal death (Selvaraj et al., 2009). TRPC3-mediated Ca²⁺ influx contributes to the survival of neurons in the SN (Zhou et al., 2008). Additionally, MPP⁺-induced oxidative stress increases intracellular Ca²⁺ via TRPM2 activation (Sun et al., 2018). TRPM7 channels regulate magnesium homeostasis in cells, and the presence of Mg ameliorates MPP⁺ toxicity (Hashimoto et al., 2008; Paravicini et al., 2012). Ca²⁺ influx via TRPV1 in dopaminergic neurons mediates mitochondrial dysfunction, microglial activation, ROS generation, and cell death (Kim et al., 2005; Nam et al., 2015). TRPV1 or TRPN-like channel-dependent dopamine release is mediated by CB1 stimulation (Table 3; Oakes et al., 2019). In addition, regulation of TRPV1 activity is closely related to the survival of dopaminergic neurons. TRPM2 is controlled by oxidative stress (Belrose et al., 2012; Huang et al., 2018). Therefore, the regulation of TRP channels contributes to overcoming dopamine depletion and the loss of dopaminergic neurons in PD patients. Likewise, oxidizing modulation of posttranslational modification (glutathionylation) of TRPC5 leads to apoptosis in an HD model (Figure 2D; Hong et al., 2015). Attenuation of TRPC5 activity by KD, blocker, or depalmitoylation shows therapeutic effects against oxidative stress by lowering TRPC5 toxicity (Hong et al., 2019). Additionally, regulation of TRPML1, TRPM2, and TRPM7 activity might also be a therapeutic strategy for ALS (Hermosura and Garruto, 2007; Hermosura et al., 2008; Tedeschi et al., 2019). The TRPV1 antagonist capsazepine has antihyperkinetic effects in a model of HD (Lastres-Becker et al., 2003).

As previously reported, TRP channels can be assembled as homo- or heteromeric complexes in nature. However, individual TRP channel KO models that lack phenotypes are limited in their ability to determine the cause of the functional compensation of each TRP channel. Moreover, in vivo studies regarding chronological changes in TRP channel expression patterns in the brain are needed for NDs. The regulation of TRP channels can be a novel therapeutic target for NDs. Nevertheless, a limitation to the development of TRP channel-specific antagonists is that their structures remain unknown. Therefore, structural analyses must precede pathological and clinical studies.

Transient receptor potential channels may not be the only, or main, pathogenic factors contributing to the pathogenesis of ND. Research in the field of ND is challenging; it is necessary to either conclusively prove a relationship between pathogenic factors or identify new therapeutic targets. In achieving one of these two possibilities, it is important not to underestimate the potential of TRP channels, based on the physiological and pathological functions of TRP channels discovered so far. Through the interrelationship between disruption of Ca²⁺ homeostasis and the development of NDs, lowering the activity of TRP channels is sufficient to enable expectations for new therapeutic strategies. As we have discussed, increased TRP channel activity has been widely observed in NDs, and model studies have shown that abnormal function due to upregulation of TRP channels can be controlled by drug treatments. Since many drugs have been reported to be TRP channel inhibitors, understanding binding modes will provide deep insight for pharmacological application in NDs. However, direct evidence for drug binding to TRP channels is unavailable. To address this issue, the most effective approach for understanding drug binding would be a structural study.

For several decades, the detailed structures and topologies of TRP channels were not understood, although the first high-resolution structure of TRPV1 was recently resolved by a single particle cryo-EM (Liao et al., 2013). Since TRPV1 was the first membrane protein to be characterized from single particle cryo-EM and the biochemical methods were relatively similar for other family members, most follow-up studies concentrated on TRP channel structure. However, though high-resolution structures have been resolved for most ND-related TRP channels, the structures of the drug-bound forms are mostly unknown. Here we discuss the technical possibilities for determining the drug-bound structure of the TRP channel family.

For analysis of drug-channel binding, which would improve our understanding of the inhibition mechanisms and would provide clues for developing drug design, the determination of the high-resolution structure is absolutely necessary. In the TRP family, the resolutions of most ND-related TRP structures are above 3.5 Å, with a few exceptions such as TRPC6, TRPA1, and TRPV4 (3.8, 4.24, and 3.8 Å, respectively) (Paulsen et al., 2015; Deng et al., 2018; Tang et al., 2018). TRPC5 broke the 3 Å barrier (Duan et al., 2019), and TRPM2 also came close to hitting the barrier (3.07 Å) (Zhang et al., 2018). The structures of other members of the TRP family that are associated with NDs, such as TRPC3, TRPC4, TRPM7, and TRPV1, were determined at ∼3.3 Å (Liao et al., 2013; Duan et al., 2018; Fan et al., 2018; Liu et al., 2018). Therefore, there is still room for improvement in their resolutions, possibly by biochemical techniques such as nanodisc reconstruction. The resolution of the drug-bound form does not always guarantee higher resolution than that of the apo structure, but generally, the occupation of an inhibitor in the binding cavity facilitates conformational stabilization, resulting in higher resolution. Therefore, since there is optimism regarding drug-binding studies, we suggest that ongoing attention and efforts be focused on this area of therapeutic target research.

CH, BJ, HP, and IS wrote and discussed the review at almost all stages. JL professionally edited the manuscript for English language. JC, JK, and Y-CS analyzed representative tables presenting TRP channel expression data and also provided technical support. IS and CH supervised the entire writing process.

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.