T-type Calcium Channel Blockers as Neuroprotective Agents

Abstract

T-type calcium channels are expressed in many diverse tissues, including neuronal, cardiovascular, and endocrine. T-type calcium channels are known to play roles in the development, maintenance, and repair of these tissues but have also been implicated in disease when not properly regulated. Calcium channel blockers have been developed to treat various diseases and their use clinically is widespread due to both their efficacy as well as their safety. Aside from their established clinical applications, recent studies have suggested neuroprotective effects of T-type calcium channels blockers. Many of the current T-type calcium channel blockers could act on other molecular targets besides T-type calcium channels making it uncertain whether their neuroprotective effects are solely due to blocking of T-type calcium channels. In this review, we discuss these drugs as well as newly developed chemical compounds that are designed to be more selective for T-type calcium channels. We review in vitro and in vivo evidence of neuroprotective effects by these T-type calcium channel blockers. We conclude by discussing possible molecular mechanisms underlying neuroprotective effects by T-type calcium channel blockers.

Introduction

Calcium signaling plays a vital role in the survival of neurons. After acute injury or with increasing age, calcium homeostasis can be disrupted, leading to neuronal dysfunction. Theoretically, pharmacologic interventions modulating calcium may yield neuroprotective effects. Recent studies have successfully shown the use of T-type calcium channel blockers (CCBs) in treating a number of neurological diseases in animal models.

Calcium carriers and calcium channels located on the plasma membrane of many cell types are responsible for calcium homeostasis within the cell. Calcium channels can be further divided into ligand bound or voltage gated calcium channels (VGCC). VGCCs can be divided into two groups: high-voltage activated calcium channels (L, N, P/Q, and R-types) and low-voltage activated calcium channels (T-types) (Figure 1). The VGCCs are defined by their alpha (α) subunits sub-categorized as the L-types CaV1.1 (α1S), 1.2 (α1C), 1.3 (α1D), 1.4 (α1F), the P/Q-type CaV2.1 (α1A), the N-type CaV2.2 (α1B), the R-type (α1E) and the T-types as CaV3.1 (α1G), 3.2 (α1H), 3.3 (α1I) [83,85,28,31]. L-type calcium channels possess at least two additional subunits that may help differentiate them from the T-type calcium channel [3,31]. T-type calcium channels are predominantly found in neurons but have been found in other cells including cardiac myocytes, pacemaker cells, glial cells, fibroblasts, osteoblasts, retinal cells, and adrenocortical cells [16,40]. At the systemic level, inhibition of T-type calcium channels may result in long-term organ protection due to improvement of local microcirculation and reduction of adverse hormonal effects [67].

Figure 1. Calcium Homeostasis through T-type Calcium Channels.

Calcium can enter a cell through either calcium carriers or calcium channels. Calcium channels can be further subdivided into ligand gated or voltage gated. Voltage gated calcium channels can be characterized by high voltage (CaV1 L-type and CaV2 N, P/Q, and R-types) or low voltage (T-type). L, N, P/Q, R-types and T-type calcium channels are further characterized by their alpha subunits into CaV1.1 (α1S), 1.2 (α1C), 1.3 (α1D), 1.4 (α1F), CaV 2.1 (α1A), 2.2 (α1B), 2.3 (α1E), and CaV3.1 (α1G), 3.2 (α1H), 3.3 (α1I). The degrees of homology between classes are noted.

At the cellular level, T-type (“T” is for transient) calcium channels open at approximately −70mV whereas L-type (“L” is for large or long-lasting) open at a more depolarized potential of approximately −20mV. Specific T-type calcium channel properties include having a low-open channel conductance, selective regulation by GTPases, and playing roles in pacemaking [65,18,74,78]. It is important to note that there are differences of biophysical properties, regional expression, functionality, pharmacological sensitivity, potential for activation, kinetics of inactivation and deactivation, and permeability among the various T-type calcium channels [30,66,23,46,47]; however, given that current drugs are not yet channel subtype specific, we will not discuss this further. When the membrane potential is between −80mV and −40mV, T-type calcium channels can cycle from open to closed and back to open, such that at all times, some T-type calcium channels are open producing a “window” current. The calcium influx from this “window” current is counter-balanced by an energy-consuming pumping mechanism. A disruption of this balance during aging or after injury could contribute to neuronal malfunction. As we will now discuss, blockers for T-type calcium channels have been developed to treat various diseases. These blockers show neuroprotective effects both in vitro and in vivo [59,80,42,5]. However, molecular mechanisms underlying their neuroprotective effects are still unclear.

Overview of T-type Calcium Channel Blockers

One major class of CCBs is a family of antiepileptic drugs which includes ethosuximide, trimethadione, and zonisamide (Table 1). Ethosuximide has a succinimide structure whereas trimethadione is an oxazolidinedione. Both ethosuximide and trimethadione are used for the treatment of absence seizures. Absence seizures have a generalized, non-convulsive pattern with a characteristic 3-Hz spike and wave electrical pattern on electroencephalography that is due to the T-type calcium channels providing the neurons with an oscillatory capacity [26,7,12,13]. Of the three T-type calcium channel subtypes, CaV3.1 is expressed in the thalamocortical relay nucleus. CaV3.1 knockout mice provide protection from absence seizures [35,63]. It is believed that ethosuximide and trimethadione effectively block this channel. While absence seizures involve a generalized cortical involvement, partial seizures affect only a small region of the brain. Zonisamide is chemically classified as a sulfonamide and is unrelated to other antiepileptic agents. Zonisamide is effective in the treatment partial seizures [11,24], childhood epilepsy [17], West syndrome (infantile spasms), and juvenile myoclonic epilepsy. General side effects of these three antiepileptic medications include nausea, vomiting, headache, mental status changes, neuropathy, and change in weight. All three of these antiepileptic drugs reach therapeutic concentration to effectively block the T-type calcium channels and show neuroprotective functions (Table 1).

Table 1.

Clinically Used Drugs with Possible Neuroprotective Effects

Drug	Pathways Inhibited	MW (g/mole)	Clinical Dose/mM	IC50 (mM)-T channel	Neuroprotective?
Anti-Epileptics
Ethosuximide	T-Type (CaV3.1, CaV3.2); Na	141.17	250-750mg PO BID/3.5-10	0.3-1.0	Y
Trimethadione	T-Type (CaV3.2)	143.14	300-600mg TID-QID/6-16.7	?	Y
Zonisamide	T-Type; Oxidative; Carbonic Anhydrase Inhibitor	212.23	400-600mg qD/1.89-2.83	0.05-0.5	Y
Anti-Hypertensives/Angina
Amlodipine	T-Type (CaV3.2>CaV3.1 or CaV3.3) and L-Type; Na; K	408.88	5-10mg qD/0.12-0.24	0.031	Y
Aranidipine*J	T-Type and L-Type	388.37	5-20mg qD/0.012-0.051	0.03-0.04	?
Azelnidipine*CJ	T-Type and L-Type	582.65	8-16mg qD/0.014-0.028	0.04-0.07	Y
Barnidipine*SJO	T-Type and L-Type	528	10-15mg qD/0.019-0.028	0.005-0.02	?
Benidipine*IJCO	T-Type and L-Type	542.02	2-4mg qD/0.0037-0.0074	0.003-0.2	?
Efonidipine*J	T-Type and L-Type	631.66	40mg qD/0.63	0.0029	?
Mibefradil*O	T-Type and L-Type; Na; K	568.55	400mg qD/0.70	0.00017-0.00029	Y
Nicardipine	T-Type and L-Type	515.99	20mg TID-120mg qD/0.11-0.23	0.0028	Y
Nimodipine	T-Type and L-Type	418.44	60mg q4h/0.86	0.0056	Y
Other
Lomerizine*J	T-Type and L-Type	541.46	5-20mg qD/0.0009-0.0036	0.00000046	Y
Pimozide	T-Type; Dopamine	461.55	2-10mg BID/0.0089-0.043	0.000036-0.000054	Y

^*Not currently FDA approved; J= Japan, C= China, I= India, S= Spain, O= Other; MW = Molecular Weight

Traditionally, many antihypertensive/antianginal CCBs were thought to function through their blockade of L-type calcium channels. However, many common CCBs are now known to block both L-type and T-type calcium channels with similar potencies [19,20,52]. Among clinically relevant CCBs to treat hypertension, three separate classes exist: dihydropyridines (amlodipine), phenylakylamines (verapamil), and benzothiazepines (diltiazem). Among the dihydropyridine class, one subclass blocks only L-type calcium channels and the other subclass blocks both L-type and T-type calcium channels. The latter subclass includes amlodipine, aranidipine, azelnidipine, barnidipine, benidipine, efonidipine, nicardipine, and nimodipine [2,21,69,67,6,19,20,45,52]. Of these listed, amlodipine, nicardipine, and nimodipine are FDA approved for use in the United States (Table 1). The other drugs are used elsewhere or are undergoing clinical trials to establish their safety profiles. Verapamil, a phenylakylamine, blocks both L-type and T-type channels with higher affinity for depolarized channels than for resting channels [8]. Mibefradil, another phenylalkylamine, was withdrawn from the market due to its interactions with other drugs metabolized through the cytochrome P450 pathway although it is highly effective to block T-type calcium channels. In general, side effects of these drugs include headache, fatigue, flushing, chest pain, shortness of breath, limb edema, and in rare cases, can cause arrhythmias, syncope, erectile dysfunction, and depression. Many of these CCBs have demonstrated neuroprotective effects (Table 1).

There are several other drugs able to block T-type calcium channels. T-type calcium channels are potently blocked by a subset of neuroleptic drugs such as pimozide and penfluridol from the diphenylbutylpiperidine family [4,57]. The diphenyldiperazine flunarizine can also block T-type calcium channels, preferentially for α1G and α1I. Fluoxetine and trazodone, primarily for treating depression, block T-type calcium channels [72,73,38]. Lomerizine is an antimigraine medication used in Japan and has been shown to have neuroprotective effects [32]. Many anesthetic agents such as isoflurane, propofol, and nitrous oxide can effectively block T-type calcium channels [70,69,50], although they also block many other ion channels as well.

Understanding through which mechanisms these drugs work is important to providing safe and efficacious interventions to neurological diseases. This requires the exploration of specific T-type CCBs with the understanding that each T-type calcium channel is unique. Unlike L-type CCBs, there are no specific blockers for T-type calcium channels on the market, providing an active field to develop specific T-type CCBs (Table 2). Table 2 presents a partial list of current drug candidates that have been designed to be more specific for T-type calcium channels with some of them already having demonstrated neuroprotective effects. A number of T-type specific CCBs have been engineered and are at various phases of testing such as neuroactive steroids [71], tetramethylcyclopropanecarboxamide derivatives [60], and pyridyl amides [53]. In addition, structure-based approaches have been applied to explore new blockers [51], and T-type calcium channels can also be blocked by activation of other molecular pathways including GTPases [68,27,29]. Some specific examples of more channel selective blockers include TTA-A2 and KYS05047 which have been shown to have effects on neuronal circuits [75,37,55]. A1048400 blocks both N-type and T-type calcium channels and has been shown to reduce tactile pain [9]. 202-W92 is a weak T-type calcium channel antagonist and is a similar compound to lamotrigine and sipatrigine. T-type calcium channels can be inhibited by sipatrigine but not lamotrigine [1,39]. 202-W92 inhibits neuronal calcium channels in a dose-dependent manner. In a stroke model, 202-W92 reduced infarct volume by 75-80% (Table 2). Given the novelty of these chemicals, not much is known about their mechanisms of neuroprotection or their compatibility with human use.

Table 2.

Drug Candidates with Possible Neuroprotective Effects

Drug Candidate	Pathways Inhibited	IC50 -T channel	Neuroprotection
A1048400	T-Type and N-Type	1.2-4.6μM	?
KYS05044	T-Type (CaV3.1, CaV3.2)	1μM	?
ML218	T-Type (CaV3.1-CaV3.3)	150-310nM	Y
NNC 55-0396	T-Type	7μM	Y
RQ-00311610	T-Type	110-170nM	?
TTA-A2	T-Type (CaV3.1-3.3)	50-100nM	Y
TTA-P2	T-Type	22nM	?
VH04	T-Type (CaV3.1)	100nM	?
Z941/944	T-Type	50-160nM	Y

Neuroprotection evidence from in vitro and in vivo models

Ischemia Model: Oxygen Glucose Deprivation (OGD)

The early evidence for neuroprotection of T-type CCBs was demonstrated in vitro [44,43,54,48,56]. In an OGD model commonly used to mimic ischemic insult in vitro, significant neuroprotection was observed when extracellular calcium levels were artificially lowered. In the same OGD culture with normal calcium level but with the addition of CCBs, there was nearly identical neuroprotection as in the calcium depleted culture, suggesting that the modification of calcium levels can protect neurons [48]. Why is modification of calcium neuroprotective in the OGD model? One theory is that T-type calcium channels remain open at membrane potentials near threshold providing a small “window” current. This “window” current is necessary for normal development but may play a role in calcium toxicity. Small increases (provided by the “window” current) in intracellular calcium trigger the mitochondrial calcium uniporter to intake calcium and activates the TCA cycle resulting in physiologically normal increases of ATP. However, excess calcium triggers increased pore formation and the release of multiple pro-apoptotic enzymes into the cell, resulting in activation of the cell death cascade. During OGD, ATP stores are rapidly depleted as ATP is used by plasma ion channels in an attempt to restore homeostasis. In cells where either the mitochondrial calcium uniporter was effectively blocked (regardless of intracellular calcium levels), the extracellular calcium was removed, or the calcium channels were blocked, there was a significant delay in the depletion of ATP during OGD resulting in prolonged neuronal protection [22]. Also of note, neuroprotection in ischemia models has been noted for other CCBs including L and N-type channels [10].

Neuronal Cell Line Model

Overload of intracellular calcium due to the use of local anesthetics such as bupivicaine is neurotoxic. Cell viability is dose dependent with bupivicaine. In the SH-SY5Y cells treated with bupivicaine alone, intracellular calcium was dramatically increased leading to cell death; however, when pretreated with dihydrochloride (a T-type CCB), intracellular calcium was reduced leading to neuroprotection [79].

Peripheral Neuropathy Model

T-type calcium channels are upregulated in patients with peripheral neuropathy. A week after sciatic nerve injury, rats had hyperalgesia. This hyperalgesia was significantly reduced with the use of mibefrabil and ethosuximide [33]. In mice with an L5-L6 spinal nerve injury, the density of T-type calcium channels in the small dorsal root ganglion neurons in the injured set increased whereas the density did not change in the medium and large injured neurons. Small neurons convey pain and temperature whereas neurons with a large diameter convey touch and mechanoreceptive information. An increase in density of calcium channels in the small nerve fibers decreases the resting potential of the pain fibers making them hyperexcitable resulting in hyperalgesia. Mibefradil and ethosuximide alleviate the decrease in resting potential that was induced by the nerve injury, thus reducing the pain response [82].

Hearing Loss Model

Cisplatin is a widely used chemotherapeutic drug that results in irreversible hearing loss by damaging outer hair cells in the cochlea. L-type CCBs diltiazem, nicardipine, nifedipine, and dantrolene do not protect hair cells from cisplatin damage while T-type CCBs flunarizine and pimozide protect hair cells after exposure [62]. However, this protection is not completely through T-type calcium channels [61]. Flunarazine acts through calcium independent mechanisms to activate the antioxidant response element to drive Nrf2 into the nucleus to drive the transcription of phase II antioxidants by binding to the promoter regions. Among the phase II elements, heme oxidase1 forms both carbon monoxide and bilirubin, both of which are known to protect against cisplatin. Flunarizine protects hair cells against cisplatin by inhibiting the translocation of NF-κB (which can be pro-apoptotic) to the nucleus (which subsequently degrades intracellularly) and by increasing the levels of Nrf2 [61].

Noise-induced hearing loss is the second most common form of sensorineural hearing loss. Previously, only the L-type CCBs appeared to reduce the auditory loss [76]. Our recent studies found that trimethadione and ethosuximide were protective of auditory hair cells and spiral ganglion neurons (SGNs) [59,5]. However, similar to in vitro studies, it was unclear whether this neuroprotection in vivo was due to blocking of T-type calcium channels or regulating other signaling pathways. It is possible that these blockers act on oxidative, neurotrophic, or both pathways to protect auditory neurons against acoustic trauma [14].

Recently, we provided compelling data that blockers for T-type calcium channels protect against age-related hearing loss of SGNs via the α1H subunit of CaV3.2 T-type calcium channels [42]. Age-related hearing loss or presbycusis is the most common age-related neurodegenerative disease. α1H (CaV3.2) is strongly expressed in SGNs and increases in expression with age [42], suggesting that with increasing age, more calcium can be in the SGNs. α1H is not expressed in inner ear hair cells whereas α1G and α1I are expressed weakly in hair cells [59]. C57BL/6 mice null for α1H had preserved hearing with aging compared to control mice consistent with preserved number of SGNs. This indicates that T-type calcium channels may play a role in the progressive hearing loss in this strain. Interestingly, the administration of trimethadione or ethosuximide to α1H null mice provided no additional hearing protection, indicating that the T-type CCBs protect hearing function through CaV3.2 channel [42] in the C57 mouse.

Neuroprotective Mechanisms of T-type Calcium Channels

We have illustrated a number of examples highlighting the neuroprotective effects of T-type CCBs. Except in one case [42], where we showed that the neuroprotective effect was due to direct inhibition of the T-type calcium channel alone, the neuronal protection in the examples did not provide direct evidence that the current blockers act on T-type calcium channels for their neuroprotective effects due to the fact that T-type CCBs do not specifically act solely on T-type calcium channels. We can only conclude that the neuroprotective effects of the class of drugs specified as T-type CCBs provide neuronal protection by blocking T-type calcium channels and/or other signaling cascades.

Because no current blockers for T-type calcium channels are highly specific, every blocker may act through a unique but partially overlapping mechanism for its neuroprotective effect. With that in mind, we focus on one T-type CCB, zonisamide, and discuss its possible neuroprotective mechanisms. Zonisamide is an FDA-approved antiepileptic drug developed in the 1970's that has recently gained interest in treating neuropathic pain, headaches, eating disorders, and Parkinson's disease. Zonisamide blocks T-type calcium channels, blocks other channels, and binds to other proteins common for this type of blocker.

Besides zonisamide's blocking T-type calcium channels such as CaV3.2 (α1H) to control the excess calcium influx seen in disease states (Figure 2 Pathway 1), it may be neuroprotective due to scavenging oxygen free radicals. Zonisamide has been shown to reduce brain ischemia through inhibition of nitric oxide synthase. Zonisamide blocks the induction of post-ischemic long-term potentiation (iLTP). While zonisamide reduces iLTP, the protective effects can be antagonized by a cGMP analog. Nitric oxide activates soluble guanylate cyclase which increases cGMP. Zaprinast, an inhibitor of phosphodiesterases, will elevate cGMP levels and even in the presence of zonisamide, will restore the damage from iLTP. Therefore, nitric oxide and cGMP are important in the pathogenesis of iLTP and are modulated in part by zonisamide [15] (Figure 2 Pathway 2). Zonisamide can also increase neuronal viability through upregulation of superoxide dismutase (SOD2). Treatment with zonisamide reduces the number of pro-apoptotic molecules such as caspase 3, caspase 9, p-JNK [34] which are important molecules in the oxidative cascade (Figure 2 Pathway 3). Zonisamide increases glutathione (an antioxidant) by increasing the cysteine/glutamate exchange transporter (Figure 2 Pathway 4). Cisplatin causes damage by generation of reactive oxygen species. Ebselen is a glutathione peroxide mimic with antioxidant capacity has been found to be protective against cisplatin toxicity. It acts by increasing the levels of Nrf2 in the nucleus which activates phase II antioxidant genes (heme oxygenase, NADPH, quinine oidoreductase, and gamma glutamylcysteine synthetase). In Nrf2 knockout mice, hearing rapidly deteriorates, corresponding to loss of hair cells and SGNs [25] (Figure 2 Pathway 5). Zonisamide preserves neurons by reducing free radical stresses [36,41].

Figure 2. Possible Pathways of T-type Calcium Channel Blocker Zonisamide.

Zonisamide functions through multiple mechanisms. Due to its multifactorial mechanisms of action, it is not clear by which set of pathways provides its neuroprotective effects. 1) Zonisamide functions as a T-type calcium channel blocker. 2) Zonisamide inhibits NOS1. 3) Zonisamide upregulates SOD2. 4) Zonisamide increases GSH. 5) Zonisamide increases NRF2. 6) Zonisamide inhibits MAO-B. 7) Zonisamide increases VEGF. 8) Zonisamide inhibits ryanodine receptors resulting in a decrease of excess calcium. Arrows indicate increase or activation. Ended lines indicate inhibition. References in text.

Zonisamide also ameliorated the dopamine depletion seen in mice treated with MPTP, whose metabolite MPP is a toxin that affects the nigrostriatal neurons. Parkinson's disease appears to be a result of mitochondrial dysfunction with a significant reduction in mitochondrial respiratory chain complex I activity. MPP inhibits complex I activity and is often used as a model for Parkinson's disease. In the MPP model, levels of intracellular calcium were increased, suggesting a possible role of toxicity to the neurons. Zonisamide's inhibition of MPTP metabolism is by inhibiting MAO-B, an activator of MPTP. In zonisamide treated groups, levels of calcium were significantly lower. Likewise, in the MPP group, levels of caspase-3, a marker for apoptosis were dramatically increased compared to the levels in the zonisamide group [84]. Other drugs that inhibit MAO-B may be neuroprotective as well [64] (Figure 2 Pathway 6).

Zonisamide can act on other pathways to protect neurons. It can upregulate mRNA for astrocyte-derived neurotrophic factor VEGF providing neuroprotection [14] (Figure 2 Pathway 7). During a hyperexcitable state, zonisamide inhibits both GABA and glutamate and also prevents calcium overload by suppressing ryanodine [81] (Figure 2 Pathway 8). The calcium induced calcium release system is comprised of the ryanodine receptor contributing to an elevation of intraneuronal calcium. Zonisamide prevents the overload response of induced ryanodine activation.

Conclusion

Drug development is time consuming and costly. It is estimated that it takes ten years and $1.8 billion dollars to take a drug from discovery to market approval [77]. Not only is this system burdened with cost and time, there are many hurdles that suppress innovative designs [49,58]. This makes the novel applications of already proven safe and efficacious medications within tested dosage ranges for various therapeutic applications more appealing.

The therapeutic role of CCBs is an exciting frontier that exists at the intersection of personalized medicine and pharmacogenomics. Numerous CCBs with T-type calcium channel activity are being used by millions of people in a safe and efficacious manner. Combining the broad array of epidemiologic data and translational science, we can assess which commonly used drugs can be repurposed for neuroprotection. Which drugs are useful? Are certain populations better protected? Is there a genetic basis to this protection? The importance of such identifications of currently available drugs cannot be understated. As the disease prevalence of many neurological diseases continues to increase, the immediate availability and protocolled use of ameliorating agents is required. The data we have shown illustrates the immense potential that T-type CCBs have in a number of neurological diseases. Our next steps are to continue to unveil currently approved agents and to prove their efficacy, safety, and at the same time, to characterize underlying molecular mechanisms as they provide neuronal protection.