N-Type Ca Channel in Epileptic Syndromes and Epilepsy: A Systematic Review of Its Genetic Variants

N-type voltage-gated calcium channel controls the release of neurotransmitters from neurons. The association of other voltage-gated calcium channels with epilepsy is well-known. The association of N-type voltage-gated calcium channels and pain has also been established. However, the relationship between this type of calcium channel and epilepsy has not been specifically reviewed. Therefore, the present review systematically summarizes existing publications regarding the genetic associations between N-type voltage-dependent calcium channel and epilepsy.


Voltage-Dependent Calcium Channels
Voltage-gated calcium channels (VGCCs) mediate Ca 2+ influx in response to action potentials and subthreshold depolarizing signals. They participate in the signal transduction of many physiological processes, such as contraction, synaptic transmission, hormone secretion, enzyme activity, or gene expression [1]. At the molecular level, VGCCs are comprised of α1, α2/δ, β, and γ subunits ( Figure 1). The α1 subunit is the ion-conducting element and contains the gating and voltage sensor machinery of the channel [2]. Ten different α1 subunits have been characterized and defined the types of these channels, which have been classified into two groups: high-voltage activated (HVA), when open at relatively depolarized membrane potentials, and low-voltage activated (LVA) Ca 2+ channels when opening near the resting membrane potential (Table 1) [2][3][4]. The biophysical and pharmacological properties of these channels are included in Table 2. The other subunits modulate and regulate the α1 pore-forming subunit: The β-subunit from the cytoplasm, the α2δ subunit from the extracellular matrix, and the γ subunit, not always present, that seems to downregulate the channel activity [1,5,6] (Table 3).
So far, the association of some voltage-gated calcium channel components with epilepsy has been widely reported. For instance, some pathogenic variants in CACNA1A and CACNA1E genes cause developmental and epileptic encephalopathies [MIM# 617106 and MIM# 618285]. In other cases, this association remains controversial. For instance, although different variants in CACNA1H have been reported in patients with congenital forms of idiopathic generalized epilepsies (IGEs) or absence epilepsy [7][8][9], their classification as pathogenic variants and the association of CACNA1H with epilepsy has been questioned [10]. The relationship between N-type voltage-gated calcium channels and epilepsy has not been specifically reviewed. forms of idiopathic generalized epilepsies (IGEs) or absence epilepsy [7][8][9], their classification as pathogenic variants and the association of CACNA1H with epilepsy has been questioned [10]. The relationship between N-type voltage-gated calcium channels and epilepsy has not been specifically reviewed.   Adapted from Gurkoff et al. (2013) [11].

N-Type Ca Channel
N-type calcium channels are characterized by their exclusively neuronal tissue distribution, prominent inhibition by G-proteins, and their participation in nociception [12]. They are sensitive to ω-conotoxins, small neurotoxic peptides isolated from the venom of sea snails of the genus Conus [13], some of which have been used in functional assays and for pharmacological purposes [14].
The CACNA1B gene [MIM*601012] encodes for the pore-forming subunit of this channel (α1B), which is expressed throughout the central nervous system and in various brain regions, controlling neurotransmitter release from neurons. α1B can form different complexes with the auxiliary α2δ and β subunits to form the N-type Ca channels.
The α2δ subunits are membrane-bound extracellular glycoproteins abundantly expressed in the central and peripheral nervous system. The α2δ pre-proteins are posttranslationally cleaved into two polypeptides, α2 and δ [15], which remain attached by disulfide bridges and extra-cellularly linked to the plasma membrane through a glycosylphosphatidylinositol (GPI) anchor attached to delta [16]. Four different α2δ subunits have been described in humans, coded by CACNA2D1 through CACNA2D4 genes. Different studies have pointed out the association of α2δ1 (coded by CACNA2D1) and α2δ2 (coded by CACNA2D2) with α1B at the plasma membrane [17]. However, an increasing number of studies suggest that the α2δ subunits may have additional functions independent of the calcium channel function. Thus, a model proposed that the uncleaved α2δ subunits maintain immature calcium channels in an inhibited state, while the proteolytic processing of α2δ leads to an activation of the channels, acting as a checkpoint for trafficking of mature calcium channel complexes into neuronal processes [18]. In support of this model, the α2δ1 subunit was proved to be of great relevance for membrane trafficking of the α1B subunit in some types of neurons [19].
The β subunits regulate the channel by modulating its gating and its interaction with other proteins and molecules, such as G proteins [20]. According to the quantitative analysis of Cav2 channel proteins in the mammalian brain, from the four β subunits, β4 is the most abundant auxiliary subunit of all Cav2 channel subtypes, including N-type [21]. This subunit is encoded by the CACNB4 gene [MIM*601949]. In addition, the β3 is the second most frequent subunit assembled to α1B [21], and may be implicated in the association of the N-type channel with pain processing [22].
The auxiliary subunits α2δ1, α2δ2, and β4 may have different effects in the α1B subunit and coffered different properties to the N-Type channels [12].

Methods
Electronic research was conducted to identify pertinent English articles examining the genetic associations between the N-type voltage-dependent calcium channel and epilepsy. Two types of databases were consulted: PubMed for publications and ClinVar and HGMD professional for specific case reports ( Figure 2). Therefore, studies with both clinical and preclinical information were included in this review. most frequent subunit assembled to α1B [21], and may be implicated in the association of the N-type channel with pain processing [22].
The auxiliary subunits α2δ1, α2δ2, and β4 may have different effects in the α1B subunit and coffered different properties to the N-Type channels [12].

Methods
Electronic research was conducted to identify pertinent English articles examining the genetic associations between the N-type voltage-dependent calcium channel and epilepsy. Two types of databases were consulted: PubMed for publications and ClinVar and HGMD professional for specific case reports ( Figure 2). Therefore, studies with both clinical and preclinical information were included in this review. A systematic literature research of PubMed was performed to identify eligible articles until 4 December 2022 using the following search terms: (epilepsy OR epileptic OR seizure) AND (N-type voltage-dependent calcium channel OR Cav2.2 OR CACNA1B). The search identified 166 potential articles.
Reviews, editorials, commentaries, and articles in a different language than English were excluded. We screened the titles and abstracts to check if they fell within the scope of this review. In some cases, when abstracts were not available or more information was required to decide, a quick review of the whole article was carried out. At this stage, a total of 34 original articles were selected.
For clinical cases also ClinVar and HGMD professional databases were revised. From 75 pathogenic variants for CACNA1B reported in ClinVar, 71 were copy number variants (CNVs) affecting more genes (35 deletions and 36 duplications), the other 4 variants were described by Gorman et al. [24] and included in the systematic review. A systematic literature research of PubMed was performed to identify eligible articles until 4 December 2022 using the following search terms: (epilepsy OR epileptic OR seizure) AND (N-type voltage-dependent calcium channel OR Cav2.2 OR CACNA1B). The search identified 166 potential articles.
Reviews, editorials, commentaries, and articles in a different language than English were excluded. We screened the titles and abstracts to check if they fell within the scope of this review. In some cases, when abstracts were not available or more information was required to decide, a quick review of the whole article was carried out. At this stage, a total of 34 original articles were selected.
For clinical cases also ClinVar and HGMD professional databases were revised. From 75 pathogenic variants for CACNA1B reported in ClinVar, 71 were copy number variants (CNVs) affecting more genes (35 deletions and 36 duplications), the other 4 variants were described by Gorman et al. [24] and included in the systematic review.
Regarding the auxiliary subunits, from 17 pathogenic variants for CACNA2D1 reported in ClinVar, two were chromosomal rearrangements (a ring chr7 and an inversion); 12 were CNVs affecting more genes (6 deletions and 6 duplications), the other three variants were described by Dahimene et al. [25] and also included in the systematic review.
From 32 pathogenic variants for CACNA2D2, 9 were CNVs affecting more genes (3 deletions and 6 duplications), one was an intragenic deletion in a patient with epileptic encephalopathy and 22 were SNPs or frameshift variants (3 missense, 6 nonsense, 2 splicing variants, 11 frameshift) of which 20 were associated with epilepsy or seizures and 2 were not associated to a specific phenotype.
From 17 pathogenic variants for CACNB4, 16 were CNVs affecting more genes (7 deletions and 9 duplications), and one was a frameshift variant detected in a patient with idiopathic epilepsy.

N-Type Ca Channel and Epilepsy: Clinical Evidence
More than 25 epileptic patients with pathogenic variants in CACNA1B or other genes encoding auxiliary subunits of VGCC have been reported in the last eight years, pointing to, and indeed confirming, the association of N-type voltage-gated calcium channel with epilepsy (Table 4).

α1 Subunit: CACNA1B
Bi-allelic loss-of-function CACNA1B pathogenic variants have been associated to neurodevelopmental disorder with seizures and nonepileptic hyperkinetic movements [MIM# 618497]. Until now, only six children from three unrelated families have been reported [24]. These patients presented a complex and progressive neurological syndrome, with epileptic encephalopathy between 9 and 30 months, severe neurodevelopmental delay, and with regression, among other features. Three sibs from a consanguineous family were homozygous for the variant c.3665del, (p.Leu1222Argfs*29); two sibs were compound heterozygous for the variants c.3573_3574del, (p.Gly1192Cysfs*5), and c.4857+1G>C; and an adopted child homozygous for the variant c.1147 C>T (p.Arg383Ter) in CACNA1B. Currently, it is necessary to further investigate the involvement of pathogenic variants in the CACNA1B gene in neurodevelopment and epileptic disorders, as well as to fully understand the etiopathogenic mechanisms of the disease.

α2δ Subunits: CACNA2D1 and CACNA2D2 CACNA2D1
So far, only two unrelated patients have been reported with early-onset developmental epileptic encephalopathy due to biallelic variants causing a loss-of-function of the CACNA2D1 gene [25]. Two frameshift variants were found, either homozygous or in compound heterozygosity with the missense variant p.(Gly209Asp), which abolishes the proteolytic maturation of α2δ and severely affects Ca V 2 calcium channel function. Generalized seizures developed at age 11-19 months. Both patients showed severe hypotonia, as an initial manifestation before three months of age, microcephaly, absent speech, spasticity, choreiform movements, orofacial dyskinesia, and cortical visual impairment. Brain imaging revealed corpus callosum hypoplasia and progressive volume loss in both.
In addition, four patients with intellectual disability (ID) and generalized epileptic seizures were previously reported as carriers of different variants, which allowed the authors to propose this gene as a candidate gene for epilepsy and ID. A de novo balanced translocation that disrupted the CACNA2D1 gene, a de novo 7.5 Mb deletion harboring 14 genes, and a 2.72Mb deletion detected in a patient and in her mother, who presented mild ID but not epilepsy, were initially published [28]. It is worth noting that the deletions contain other genes (PCLO, SEMA3E, SEMA3A), which could conceivably contribute to the phenotype, as they are involved in neuronal synapsis, axonal guidance, or development of the nervous system. Two of these patients additionally showed different inconsistent dysmorphic features; two of them associated short stature, and one patient revealed MRI findings (bilateral frontotemporal polymicrogyria, mega cisterna magna, and a cyst in the cavum veli interpositi). Intriguingly, the larger deletion was associated with the mildest phenotype in a girl with an IQ of around 80 who initially had an apparently normal neuromotor development but underwent a slight regression after several seizures. On the other hand, a de novo splicing variant was posteriorly reported in a fourth patient with epilepsy and ID [26], and a de novo frameshift variant was detected in a female with normal intelligence and infantile spasms that responded to ACTH with good outcome [27]. Currently, it is not established whether these loss-of-function variants in a heterozygous state cause intellectual disability and epilepsy. Taking into consideration the recessive pattern seen in the above patients and the fact that this gene shows some tolerance to lossof-function variants [41], as well as some deletions observed in healthy individuals [42,43], one should be cautious about the relevance of these genetic variants. Maybe some heterozygous loss-of-function variants in this gene indeed predispose to epilepsy and/or ID with incomplete penetrance. However, they may also be casual findings or, in some cases, be present in combination with other unnoticed variants in trans in the CACNA2D1 gene, which altogether contribute to the clinical manifestations.

CACNA2D2
Initial evidence of the phenotype associated with CACNA2D2 pathogenic variants came from the mouse mutant strain ducky, which carried null alleles for Cacna2d2. These animals are a model for absence epilepsy characterized by behavioral arrest synchronous with spike-wave discharges and severe cerebellar ataxia. This reflects the strong expression of α2δ2 in particular neurons, specifically cerebellar Purkinje cells [45]. In humans, biallelic CACNA2D2 variants cause a phenotypic spectrum ranging from congenital ataxia with cerebellar vermian atrophy on brain imaging [44] to cerebellar atrophy and developmental and epileptic encephalopathies [30,31]. The homozygous variant p.L1040P in the CACNA2D2 gene was initially related to a dysfunction of α2δ2, resulting in reduced current density and slow inactivation in both N-type (Cav2.2) and L-type (Cav1.2) calcium channels. The resulting prolonged calcium entry during depolarisation and changes in the surface density of calcium channels was proposed to underly the epileptic phenotype [30]. Later, abolished expression of the CACNA2D2 gene, due to a homozygous frameshift loss-of-function variant, in a patient was significantly associated with severe delays of psychomotor development, epilepsy with absence seizures, dyskinetic movements, and cerebellar atrophy [31]. Some dysmorphic features and microcephaly were also present. The identification of novel pathogenic or likely pathogenic variants in the homozygous state leads further support to the association of the CACNA2D2 gene with developmental and epileptic encephalopathy with cerebellar atrophy and ataxia [29,33]. On the other hand, Valence et al. describe an individual with congenital ataxia and cerebellar atrophy, where the milder neurologic phenotype was attributed to the partial splicing defect of a rare homozygous splice variant [44]. These authors demonstrated a diminished expression of wild-type mRNA transcript, which would explain the milder neurologic conditions without ID and with a single febrile seizure episode.
Considerable clinical overlap between individuals with homozygous loss-of-function variants in CACNA2D1 and in the CACNA2D2 gene has been highlighted [25]. However, atrophy of the brain affects the cerebrum in individuals with CACNA2D1 variants, whereas cerebellar atrophy was consistently reported in subjects with CACNA2D2 variants, a difference most probably due to a higher expression of the CACNA2D2 gene in cerebellar cells [33]. Similarly, the predominant expression of CACNA2D4 in the retina leads to retinal dysfunctions in affected patients, which indicates that any of the other α2δ subunits cannot compensate for the loss of α2δ1 or α2δ2 during development [25].

β Subunit: CACNB4
Mutations in the CACNB4 gene might confer susceptibility to idiopathic generalized epilepsy-9 (IGE9) and juvenile myoclonic epilepsy-6 (JME6) [MIM# 607682]. Taking into account that the lethargic (lh) mouse, who has inactivated the Cchb4 gene, present a complex neurological disorder that includes absence epilepsy and ataxia [46]; Escayg et al. screened in 90 familial with IGE and 71 families with episodic ataxia for mutations in CACNB4 genes [35]. Two coding changes that were considered probably pathogenic were identified in three different families. A female patient with juvenile myoclonic epilepsy (JME) was heterozygous for the premature-termination mutation c.1444C>T (p.Arg482Ter). In addition, functional analysis in Xenopus laevis oocytes co-expressing α1A and β4 subunits indicated that p.Arg482Ter might have altered the kinetic of inactivation [35]. Besides, this variant may alter other functions of CACNB4 as a repressor of neuronal gene expression [47]. The other variant, c.311G>T (p.Cys104Phe), was identified in nine individuals from two unrelated families. Two affected individuals with an atypical but similar clinical syndrome of IGE with rare juvenile atypical prolonged absences and occasional GTCS from one family; and from the other family, two asymptomatic individuals, four relatives with a more heterogeneous phenotype of ataxia and/or vertigo, and one case with febrile seizures as a child [35]. However, according to ClinVar (Variation ID:7608) and based on different submitters, the pathogenicity of this variant is unclear.
In a patient heterozygous for the pathogenic variant c.1702C>T (p.Arg568Ter) in SCN1A, de novo, with severe myoclonic epilepsy in infancy (SMEI), a second variant in CACNB4, was reported. The c.1403G>A (p.Arg468Gln) in CACNB4, inherited from his father, who had a febrile seizure associated with measles at the age of seven, was considered as a possible phenotype modifier [36].
In a cohort of 280 unrelated patients with early-onset developmental and epileptic encephalopathy (DEE) screened for alteration in 172 genes, the pathogenic variant c.21delC (p.Lys8Argfs*26) in CACNB4 was identified, although specific clinical data from this case was not provided [37].
Finally, from a cohort of 390 subjects with neurodevelopment disorder, a homozygous variant in CACNB4 was detected in two siblings from first-degree consanguineous healthy parents. Both patients present severe developmental delay and cerebellar atrophy with epilepsy, among other features. The younger brother presented focal seizures and responded to treatment, while the older sister, with a more severe phenotype, presented focal and tonic seizures refractory to treatment. In silico tools, animal models, and clinical and genetic data suggest the c.377T>C (p.Lys126Pro) CACNB4 variant is likely pathogenic and impairs the formation of the calcium channel complexes. However, both siblings also carried 11 other rare homozygous variants, which could contribute to the severe phenotype [38].
Although some cases with epilepsy and alterations in CACNB4 have been reported, the variability in the phenotype, the weak evidence of pathogenicity in some cases, and the presence of other genetic variants in others make it necessary to get further evidence to establish a stronger association between CACNB4 with epilepsy.
Only a patient with atypical Rolandic epilepsy has been reported with a de novo variant in CACNG2 gene so far [40]. Without further evidence, the c.295+1G>C variant of CACNG2 (NM_006078.4) was predicted to affect splicing and classified as a likely pathogen. Previously, a provisional association with intellectual disability was proposed based on a single case without epilepsy [MIM #614256] [49]. However, the bi-allelic disruption of cacng2 is well characterized in an animal model such as the stargazer mouse, which presents spike-wave discharges (SWDs), representative of absence seizures in humans [50,51]. Therefore, the association between CACNG2 alteration and epilepsy is still weak, and its implication in N-type (Cav2.2) is not yet well established.
Summarizing all the above, bi-allelic loss-of-function pathogenic variants in CACNA1B have been associated with a complex and progressive neurological syndrome, with epileptic encephalopathy, severe neurodevelopmental delay, and regression, among other features. However, more cases should be collected and reported to fully understand the associated phenotype. As mentioned before, the auxiliary subunits are not specific for N-type calcium channels. In fact, an increasing number of recent studies suggest that, for instance, individual α2δ isoforms exert specific neuronal functions beyond their classical role as calcium channel subunits. Even so, the main phenotypes associated with bi-allelic loss-function variants in CACNA2D1 and CACNA2D2 genes are epileptic and developmental encephalopathies, no doubt due to its participation in N-and/or P/Q-type calcium channels. However, the relative importance of one or another channel has, so far, not been disclosed. Finally, the association of other auxiliary subunits, such as CACNB4 and CACNG2, with epilepsy requires more evidence to be definitively established.

N-Type Ca Channel and Epilepsy: Evidences in Animal Models
Using gene knockout mice of particular genes is one of the most effective methods in conducting a successful study on clinical phenotypes, drug discovery, or in preclinical studies. In the case of N-type Ca 2+ channel studies, and more specifically the CACNA1B gene, knockout murine models were applied to examine various clinical consequences (learning and memory, cardiovascular, nociceptive, neurodegenerative); however, its role in epilepsy has never been published [52,53]. On the other hand, disruption of the Cacng2 gene in the stargazer mouse is associated with recurrent epileptic seizures and ataxia. Although this gene codifies a neuronal-specific gamma subunit that may participate in the formation of several types of calcium channels, not only N-type, an in vitro study in Xenopus oocytes demonstrated that this subunit interferes with G protein modulation of N-type calcium channels expressed in Xenopus oocytes [54]. References to other naturally occurring mouse mutant strains for other components of N-type VGCCs, such as ducky, which carried null alleles for Cacna2d2, or lethargic, inactivated for the Cacnb4 gene, have been mentioned above in relation to their epileptic phenotype. However, the subunits coded for these mutated genes are not specific to the N-type VGCCs, so other types of channels might also be implicated.
Other genetic, pharmacological, and electrophysiological data suggest a significant, although poorly understood, role for N-type voltage-dependent calcium channels in epilepsy. Different animal models, mainly in rat and mouse, have allowed analyzing different brain tissues in order to glimpse the implication of VGCCs in epilepsy (Table 5). For instance, partially isolated cortex ("undercut") is a well-established animal model of posttraumatic epileptogenesis, where dysfunction of Ca 2+ channels in presynaptic GABAergic terminals via N-type channels plays a prominent role [55]. The Wistar Albino Glaxo Rijswijk (WAG/Rij) rat is a strain that spontaneously develops a clinical phenomenon resembling human absence epilepsy, where excessive calcium entry into the cell, have been proposed to trigger the epileptic activity. The blockage of the N-type calcium channel with conotoxins on this strain decreased the frequency and duration of spike-wave discharges, which let us conclude that N-type calcium channels play an activator role together with T-type VGCCs [56]. By analysis of the evoked epileptiform activity in neocortical slices of rats pointed to the importance of calcium influx mainly via the N-type calcium channel in the epileptiform activity induced in vitro [57]. Amygdala kindling is an experimental model for temporal lobe epilepsy obtained by reiterative electrical stimulations. This stimulation induces a significant increase of N-type VGCCs in the hippocampus, twice in the contra-lateral to the stimulating electrode. The authors argued for the contribution of these channels to seizure-induced synaptic plasticity [58,59]. The increased expression of α1B in both the strata granulosum and the molecular layer of the dentate gyrus after pilocarpine-induced status epilepticus in mice have been suggested to be involved in the occurrence of spontaneously recurrent seizures [60]. In other animal models, the seizure-sensitive gerbil, the elevated expressions of VGCC subtypes (including N-type VGCC) may increase Ca 2+ -dependent excitatory transmission in the hippocampus [61]. Finally, DBA/2J is a mouse inbred strain with high susceptibility to audiogenic seizures. Development of N-type calcium channels in the brain is different in epileptic mice (exposed to auditory stimulation) from nonepileptic mice (not exposed), and this difference was argued to represent a change in synaptic activity that plays a role in epileptogenesis [62].
On the other hand, different studies on physiological conditions related to epilepsy further support the implication of this type of calcium channel. Ethanol withdrawal is one of these conditions, which in rats leads to a significant decrease of the protein levels of α1B subunit, specific for the N-Type VGCC, and this decrease was proposed to secondarily enhance the susceptibility to seizures [63]. Similarly, the loss of the somatostatin function postseizure could result in abnormal synaptic potentiation that contributes to epileptogenesis through N-type Ca 2+ channels [64]. It is known that crotoxin, a snake venom, causes chronic seizure effects by inducing calcium-dependent glutamate release via N and P/Q calcium channels [65]. Finally, in a study on the potential relationships among aging, calcium channels, and late-onset epilepsy, no overall age-related change in the number of L-type and N-type VGCC in brain areas frequently involved in seizure activity was found. Consequently, age-related changes in brain Ca 2+ physiology was proposed to be associated with changes in VGCC function rather than the channel number [66]. Table 5. Contribution of different animal models to clarify the association of N-type VGCC with epilepsy.

Animal
Tissue/Seizure Type Effect Ref.
rat neocortex Epileptiform activity in the rat neocortex may occur, at least partially, via the activation of the N-type VGCC. [57] mouse dentate gyrus/CA1 hippocampus Somatostatin (SST) reduction of dendritic Ca 2+ through N-type VGCC may contribute to the modulation of synaptic plasticity at long-term potentiation (LTP) synapses. Loss of SST postseizure could contribute to epileptogenesis. [64] rat pyramidal (Pyr) neurons of the sensorimotor cortical Decreased calcium influx via N-type VGCC in presynaptic GABAergic terminals is a mechanism contributing to decreased inhibitory input onto layer V Pyr cells in this model of cortical posttraumatic epileptogenesis. [55] rat frontal and occipital cortical/absence epilepsy T and N-type VGCCs play activator roles in spike-wave discharges (SWDs) and have positive effects on the frequency and duration of these discharges.
[56] N-type VGCC translocation occurs at acute stages during and after pilocarpine-induced status epilepticus (PISE). The increased expression of this channel in the strata granulosum and dentate gyrus at the chronic stage may be involved in the occurrence of spontaneously recurrent seizures. [60] gerbil hippocampus The elevated expressions of VGCC subtypes, including N-type VGCC, may increase Ca 2+ -dependent excitatory transmission in the hippocampus of the seizure-sensitive gerbil. [61] rat hippocampus, entorhinal cortex, and neocortex No overall age-related change in the number of L-type and N-type VGCC in brain areas frequently involved in seizure activity suggests that age-related changes in brain Ca 2+ physiology may be associated with VGCC function rather than a channel number.

Anti-Epileptic-Drug Response and Therapeutic Target
The crucial role of calcium ions and glutamate-induced calcium influx via N-methyl-D-aspartate receptors (NMDARs) in neuropathological conditions is described in various studies. The subsequent abnormal NMDAR-dependent calcium accumulation in postsynaptic neurons seems to contribute to a cascade of cellular events that lead to neuronal cell death [67].
Excessive calcium entry into depolarized neurons contributes significantly to neuronal injury. Taking into account that: (i) VGCCs regulate intracellular calcium concentration, essential for several neuronal functions such as cellular excitability, neurotransmitter release, hormone secretion, intracellular metabolism, neurosecretory activity and gene expression, and (ii) N-type Ca channel expression is exclusive for neuronal tissue; it has been proposed that a blockade of N-type VGCCs may be useful for treating neurological diseases, including ischemia, pain, and epilepsy.
Different antiseizure drugs (ASD) may act as N-type channel blockers (Table 6). However, they don't show the same efficacy against different types of seizures, probably due to their additional mechanisms of action. For many decades carbamazepine was the standard treatment for focal-onset seizures [68], but its use has declined with the marketing of new antiseizure medications that have pharmacokinetic advantages. It may exacerbate absence, myoclonic, and atonic seizures. Gabapentin can be used as an adjunctive treatment for focal seizures. As carbamazepine, it may worsen myoclonus [69]. Both lamotrigine and levetiracetam are broad-spectrum ASD, and they are widely used as first-line ASD for focal seizures and generalized tonic-clonic seizures [70]. Besides, levetiracetam has shown efficacy against myoclonic seizures [71]. Finally, lacosamide is effective against focal-onset seizures as well as generalized onset tonic-clonic seizures [72]. It has not demonstrated efficacy against absences or myoclonic seizures, but it is unlikely to worsen them.

Conclusions
The evidence so far gathered allows us to be confident about the causality of biallelic loss-of-function variants in several components of the N-type voltage-gated calcium channels, coded by CACNA1B, CACNA2D1, and CACNA2D2 genes, on epileptic and developmental encephalopathies. On the other hand, further evidence is necessary to define the implication of other components of these channels, such as CACNB4 or CACNG2. Similarly, the role of heterozygous variants in CACNA2D1 as a predisposing factor to epilepsy and/or intellectual disability remains to be confirmed.
Besides, different animal models, mainly in rat and mouse, have allowed analysis of different brain tissues in order to try to clarify the implication of N-type VGCCs in epilepsy. Even more, distinct ASD may act as N-type channel blockers, and blockade of N-type VGCC has been proposed as a feasible treatment for epilepsy. Therefore, although there is currently strong evidence, future investigations will allow us to finally establish the association of N-type Ca channels with epilepsy syndromes and epilepsy.