Mutations in voltage-gated L-type calcium channel: implications in cardiac arrhythmia

ABSTRACT The voltage-gated L-type calcium channel (LTCC) is essential for multiple cellular processes. In the heart, calcium influx through LTCC plays an important role in cardiac electrical excitation. Mutations in LTCC genes, including CACNA1C, CACNA1D, CACNB2 and CACNA2D, will induce the dysfunctions of calcium channels, which result in the abnormal excitations of cardiomyocytes, and finally lead to cardiac arrhythmias. Nevertheless, the newly found mutations in LTCC and their functions are continuously being elucidated. This review summarizes recent findings on the mutations of LTCC, which are associated with long QT syndromes, Timothy syndromes, Brugada syndromes, short QT syndromes, and some other cardiac arrhythmias. Indeed, we describe the gain/loss-of-functions of these mutations in LTCC, which can give an explanation for the phenotypes of cardiac arrhythmias. Moreover, we present several challenges in the field at present, and propose some diagnostic or therapeutic approaches to these mutation-associated cardiac diseases in the future.


Introduction
Cardiac arrhythmia is one of the major causes for sudden cardiac death (SCD) [1,2]. Over the past decades, some arrhythmic susceptible genes, including voltage-gated L-type calcium channel (LTCC), have been identified many arrhythmiaassociated mutations [3][4][5]. LTCC is the main pathway for calcium ion influx into excitable cells in response to the membrane depolarization [6,7], which forms one part of cardiomyocyte action potentials (APs). In the heart, LTCC is a multisubunit protein complex composed by four subunits: α 1 subunit and auxiliary β, α 2 δ and γ subunits, encoded by CACNA1C or CACNA1D, CACNB2, CACNA2D and CACNG, respectively [8,9]. It has been known that the mutations in these LTCC genes induce the dysfunctions of calcium channels, which result in the abnormal excitations of cardiomyocytes, and finally lead to cardiac arrhythmias.
In this work, we summarize recent findings on the mutations in the different subunits of LTCC, which are associated with cardiac arrhythmias (Figure 1), and describe the functional roles of these mutations in channel properties (Table 1), which can shed a light in understanding of cardiac electrophysiological characteristics in these diseases.
The extracellular α 2 and transmembrane δ subunit, linked with each other via disulfide bonds, are encoded by the gene CACNA2D [32]. The α 2 δ-1 subunit, encoded by CACNA2D1, is abundant in skeletal and cardiac muscles [33,34]. The α 2 δ-2 and α 2 δ-3 subunits, encoded by CACNA2D2 and CACNA2D3 respectively, express in neurons and some other tissues [33,35]. In addition, the expression of α 2 δ-4 subunit, encoded by CACNA2D4, is mostly copious in non-neuronal cells apart from retinal neurons [36,37]. By bioinformatics, the proteins encoded by several other genes have been identified, and they share the similar structure with α 2 δ subunits [38], but have not been proved to function as calcium α 2 δ subunits. The α 2 δ-1 subunit plays a vital role in hypertensive vascular Ca V 1.2 channel properties [39,40]. In the heart, the atrium is characterized by increased α 2 δ-1 protein expression as compared with the ventricle, which might lead to the Ca V 1.2 electrophysiological differentiation between atrial and ventricular cardiomyocytes [41]. Furthermore, Figure 1. Predicted topology of Ca V α 1 subunit with associated β 2 and α 2 δ subunits shows the location of functional mutations. All mutations in α 1 subunit are derived from α 1C (Ca V 1.2), except one mutation G403_404ins is come from α 1D (Ca V 1.3). AID = α-subunit interaction domain; BID = β-subunit interaction domain; BrS = Brugada syndrome; CCD = cardiac conduction disease; ERS = early repolarization syndrome; GK = guanylate kinase; LQTS = long QT syndrome; SH3 = Src homology 3; SQTS = short QT syndrome; SNP = single nucleotide polymorphism; TS = Timothy syndrome.  deletion of α 2 δ-1 in mice induced the decreased Ca V 1.2 Ca 2+ currents of cardiomyocytes, and also decreased the basal myocardial contractility and relaxation [42]. These observations imply the potential roles of α 2 δ subunits in cardiac arrhythmias. The γ subunit, which is encoded by CACNG, has 8 isoforms and consists of 4 transmembrane domains [43]. The first γ 1 subunit was cloned from skeletal muscle [44,45], whereas it was not expressed in cardiac muscle [46]. The γ subunit distinguishingly modulates the functions by altering both activation and inactivation properties of LTCC [46]. For example, targeted disruption of γ 1 subunit increased the amplitude of peak calcium current in isolated myotubes [47]; whereas coexpression with γ 2 subunit could depolarize the activation and inactivation curve of Ca V 1.2 channels [48]. In the heart tissue, four different γ subunits, γ 4 , γ 6 , γ 7 and γ 8 , are expressed, and they all physically interact with the Ca V 1.2 complex, which diversify the functions of Ca V 1.2 channels [46]. To date, no evidences indicate the association of γ subunits and cardiac arrhythmias, thus it is of value to investigate their possible links.

Long QT and Timothy syndromes
Long QT syndrome (LQTS) is one of the most common cardiac electrophysiological diseases with an estimated prevalence of 1 in every 2,534 persons [49]. It is a typical cardiac repolarization abnormality defined by heart ratecorrected QT interval (QTc) prolongation on resting electrocardiogram (ECG) [4,50], and characterized by an increased trend for ventricular tachycardia with torsade de pointes (TdP). LQTS exists as a congenital genetic disease (cLQTS) with mutations described in different genes, which cause different types of LQTS. LQTS also can be acquired (aLQTS) by drugintake [51] or structural heart disease [52], which may be more prevalent than cLQTS [53]. To date, 16 genes were identified to responsible for LQTS, of which, mutations in CACNA1C is accounted for long QT syndrome 8 (LQT8) [54]. Two canonical mutations in CACNA1C (p.G406R and p.G402S) are associated with a severe LQT8, namely Timothy syndrome (TS).

Timothy syndrome 1
In the 1990s, a severe syndrome which was called as the heart-hand syndrome, now known as TS or TS1, was observed in young children with significant clinical phenotypes: syndactyly, severe cardiac arrhythmia, congenital heart disease, developmental abnormalities, autism, and neurological dysfunction [55][56][57]. Most patients were diagnosed during neonatal period or rarely in late infancy, only two cases reported the fetal hydrops as antenatal expression of TS1 [58,59].
Splawski et al. discovered a de novo G-to-A mutation, which caused a p.G406R substitution in translational sequence at position 1216 (c. G1216A) of CACNA1C exon 8A (also known as exon 8 [60]) in all 13 individuals with TS. However, the p.G406R mutation was not identified in 180 ethnically matched controls [61], indicating that p.G406R mutation is associated with TS. This amino acid p.G406 is completely conserved in Ca V α 1C subunit of other mammalian species, and is located at the C-terminal portion of the S6 of domain I [61,62].
The inactivation of the voltage-gated calcium channel has two mechanisms: voltage-dependent inactivation (VDI) and calcium-dependent inactivation (CDI), and they play an essential role in controlling excitation-contraction coupling in cardiomyocytes [6,[63][64][65]. Functional analysis revealed that p.G406R mutation caused the maintained inward calcium currents by impairing the two types of channel inactivation [61]. By measuring whole-cell ionic currents in transfected human embryonic kidney-293 (HEK293) cells, the p. G406R mutation slowed the VDI of Ca V 1.2 calcium channels, while it possibly accelerated the kinetics of CDI [66]. It is known that the small glycine residues offer flexibility to allow for the I-II loop to interact with the intracellular pore of the channel [67][68][69], and G406 is located wiαthin a short stretch of nonhelical motif at the start of the α 1C I-II loop [70]. Therefore, p.G406R mutation would produce a more bulky arginine residue to impede the movement of the I-II loop, leading to the slower channel inactivation. As a result of abnormally huge calcium ion influx, the action potential duration (APD) can be prolonged. Thus, the patients' ECG showed a prolonged QT interval leading to death from cardiac arrhythmia [66]. Interestingly, this mutant Ca V 1.2 channel function could be modulated by one anchoring protein AKAP150, elimination of which could abrogate the prolonged QT interval [71], implying AKAP150 may be a promising target for TS.
To explore the effects of the p.G406R mutation on the electrical activity and contraction of living cardiomyocytes, human skin cells from TS patients were reprogrammed to generate induced pluripotent stem cells (iPSC), and differentiated into cardiomyocytes [72]. The electrophysiological recording and calcium imaging studies revealed excessive calcium ion influx, irregular electrical activity and contraction, prolonged APs, and abnormal calcium transients in these ventricularlike cells [72,73]. Roscovitine, both an atypical LTCC blocker and a cyclin-dependent kinase inhibitor, which can enhance the VDI of Ca V 1.2, restored the electrical and calcium signaling properties of cardiomyocyte from TS patients [72][73][74][75].
Recently, TS mouse model had been generated and proved to have more arrhythmogenic events, which might be due to the increase of cytosol calcium concentration ([Ca 2+ ] i ) by sarcolemmal Ca 2+ leak and the impairment of VDI [76]. Syndactyly or craniofacial abnormalities may also be caused by irregular Ca V 1.2 TS channels, because Ca V 1.2 TS channels affected the development of jaw and mandibular in the mouse and zebrafish models [77].

Timothy syndrome 2
Splawski et al. also identified two mutations, c. G1204A and c.G1216A, in exon 8 (also known as exon 8a [60]) of the CACNA1C gene, resulting in the p.G402S and p.G406R substitution, respectively [62]. Owing to that the exon 8 splice variant was found to highly express in human heart and brain, the patients who carried this mutation had a longer QT interval and more severe arrhythmia than patients with a mutation in exon 8A. Since the symptoms caused by the mutations in exon 8 were different from previous TS, this disease was named as TS2. The patients suffered from multiple arrhythmias and SCD due to the extreme prolongation of QT interval, but syndactyly might not be manifested [62,78], and these patients were more likely died of TdP and ventricular fibrillation [62]. Structurally, the glycine residue at 402 position of IS6, together with the same positions in the IIS6, IIIS6 and IVS6 segments, form the G/A/G/A motif, which is near the inner channel mouth of Ca V 1.2. These four residues interact with larger bulky residue from neighboring S6 helices to stabilize the inactivation gate [68]. In TS2, in addition to p.G406R mutation, p.G402S will also introduce a more bulky serine residue to disrupt the tightly sealing of the S6 helices, resulting in the slower deactivation [68]. Although neither heterozygous nor homozygous p.G402S transgenic mice can survive to weaning because of the high expression of exon 8 and a fatal extremely high level of calcium channels mutation, TS2-like mouse model with p.G406R had been generated by heterozygously keeping an inverted neomycin cassette in exon 8A [79], these mice could survive through adulthood due to the lowered expression of p. G406R LTCCs. The survived mice displayed the behavioral abnormalities, corresponding to the core aspects of autism spectrum disorder [79].

Timothy syndrome with novel CACNA1C mutations
A newborn, with healthy parents, had the similar symptoms with TS, he presented with prolonged QT interval and associated polymorphic ventricular tachycardia, dysmorphic facial features, syndactyly of the hands and feet, and joint contractures. By full gene sequencing, the patient had no p. G402S and/or p.G406R mutations, but a novel CACNA1C mutation p.A1473G had been detected in exon 38 [80]. This case expanded the molecular basis of TS, however the electrophysiological properties of this mutant channel need further investigations. Recently, another novel CACNA1C mutation p.I1166T in exon 28 was identified in a young male with diagnosed TS, and this mutation induced an overall loss of current density and a shift of activation of Ca V 1.2 channels, which lead to an increased window current [81]. However, Wemhoner et al. found this p.I1166T mutation could induce a leftward shift of activation curve, but an increased peak current density [82]. These different results maybe attribute to different clones of Ca V 1.2 subunits, or even different ratios of subunits expressed in HEK293 cells.
Long QT syndromes without extracardiac signs Some LQT8, with typical long QT interval, have no extracardiac phenotypes. A novel mutation p. R518C/H in CACNA1C was just identified in patients with a complex phenotype including LQTS, hypertrophic cardiomyopathy and congenital heart defects, which was annotated as "cardiac only TS" [83]. This mutation in Ca V 1.2 also revealed a complex channel phenotypes, including loss of current density and inactivation in combination with increased window and late currents [83]. By whole-exome sequencing and bioinformatics/systemic biology, Boczek et al. further identified 4 novel mutations of CACNA1C in LQTS, including p.K834E, p.P857R, p.P857L, and p. R1906Q [84], 3 of which were located at the conserved proline, glutamic acid, serine and threonine rich (PEST) domain of Ca V α 1 II-III loop, this domain acts to proteolytic signaling through the cellular quality control system [85]. Functionally, p.P857R mutation, cosegregated with the disease within the pedigree, significantly increased calcium currents and surface membrane expression of the channel as compared with wild type (WT) Ca V 1.2 channel [84]. Interestingly, p.R1906Q locates one amino acid away from α 1C C-terminal STIM1 (stromal-interacting molecule 1, a calcium store sensor) binding domain, which may affect STIM1-mediated Ca V 1.2 channel gating inhibition and channel internalization [86,87]. The identification of these CACNA1C mutations co-segregating with disease indicated that CACNA1C genetic perturbations might underscore autosomal dominant LQT8 in the absence of TS.
Additionally, five novel CACNA1C mutations (p.P381S, p.M456I, p.A582D, p.R858H, and p. G1783C) were identified in the patients with LQT8, but without typical TS phenotypes [88]. Of significance, p.R858H mutant channel had a larger I Ca , and p.A582D mutant channel displayed a slower inactivation [88], which could partially explain the phenotype of long QT interval. By screening 540 patients with LQTS, 6 more CACNA1C mutations were identified recently, including p.A28T, p.R860G, p.I1166T, p.I1166V, p.I1475M and p.E1496K [82]. These mutations affected the different properties of channel. Briefly, p.A28T increased calcium current of L-type channel (I Ca,L ) and positively shifted steady-state inactivation (SSI). p.R860G positively shifted SSI and increased steady-state current (SSC). Both p.I1166T and p.I1166V mutations showed an increased I Ca,L , but p.I1166T negatively shifted the steady-state activation (SSA) and increased SSC. p.I1475M and p.E1496K negatively shifted the SSA, moreover p.E1496K increased SSC and slowed the current decay [82]. Using computational cardiac AP model, the researchers found these gain-of-function mutations delayed repolarization of the cardiac myocytes, which induced prolongation of APD [82]. Recently, a novel CACNA1C mutation p.L762F was identified to associate with the development of LQTS. This mutation slowed the channel inactivation and increased persistent and window current, which attributed to the gain-of-function of Ca V 1.2 channels [89]. Furthermore, a novel mutation of CACNA1C (p.N2091S) is also found in autopsynegative sudden unexplained death of a 24-yearold white female. No premortem ECGs were available and neither decedent had any documented family history of arrhythmia-related cardiac events. However, this p.N2091S mutation induced a dramatic increase of I Ca and a minor hyperpolarization of V 0.5 of channel activation, indicating a LQTS-like gain-of-function electrophysiological phenotype [90].

Brugada syndromes
Brugada syndrome (BrS), first described in 1992, is an inherited cardiac arrhythmic syndrome associated with a high risk of ventricular fibrillation without structural heart disease [91]. The ECG of the Brugada patient is characterized by right bundle branch block and ST segment elevation in precordial leads V1-V3.
The symptom of BrS typically manifests during the adulthood, but the youngest individual diagnosed with BrS is only 2 days old, while the oldest is 84 years old [92,93]. Generally, BrS is more prevalent among male patients owing to the gender differences in the expression of cardiac transient outward potassium channel (I to ). Because of the presence of a more predominant I to in males compared with females, male patients are more likely to induce the loss of the AP dome and the development of phase 2 reentry and polymorphic ventricular tachycardia [94,95]. Mutations in CACNA1C, CACNB2 and CACNA2D1 can cause BrS3, BrS4 and BrS9, respectively [96,97].

Brugada syndrome 3
A male European descent with BrS and a shortened QTc interval showed a heterozygous mutation in exon 2 of the CACNA1C gene, which made a p.A39V substitution, near the N-terminus within a highly conserved region of the Ca V α 1C [98]. In another case, a male Turkish patient with BrS and a shortened QT interval, was detected a novel heterozygous mutation in exon 10 of the CACNA1C gene was predicted to result in a p. G490R substitution in I-II loop [98]. Patch-clamp analysis in Chinese hamster ovary (CHO) cells, which transfected with p.A39V or p.G490R mutant Ca V 1.2 channels, showed the current amplitude was dramatically reduced as compared with WT channels, although voltage at peak current remained unchanged [98]. These results gave an sound explanation for the shortened QT interval. Moreover, p.A39V mutant channel had a defect in trafficking of mature LTCCs, however, confocal imaging revealed the normal trafficking of p.G490R mutant Ca V 1.2 channels from ER to cell membrane [98]. Therefore, it is reasonable to conclude that p.G490R mutation may affect the channel open probability (P o ), but not cell surface expression. Interestingly, this p.A39V mutation in N-terminus also affected the calmodulin-dependent activation of Ca V 1.2 channel [99], but didn't affect the surface expression of neuronal Ca V 1.2 channels [100], which might be involved the mechanisms of BrS. Recently, a p.R632R in Ca V 1.2 α 1C due to c.G1869A mutation in exon 14 of CACNA1C gene was identified in a patient with BrS, this mutation caused an aberrant splicing, which in turn made a premature stop codon in the downstream. This type of mutation might lead to nonsense-mediated mRNA decay, which induced the loss-of-function of Ca V 1.2 calcium channel [101].
By screening 162 probands with BrS or BrS combined with short QT syndrome (SQTS) (BrS + SQTS), there were 7 newly identified mutations in CACNA1C, including p.E1115K, p.R1880Q, p. V2014I, p.D2130N, p.E1829_Q1833dup, p.C1873Y and p.E850del [102]. Functionally, Ca V 1.2 channel with p.E1115K had a reduced calcium influx. Mechanistically, this mutation strictly reduced single channel conductance, but did not change the voltage or calcium-dependent gating [103]. p. V2014I mutation in Ca V 1.2 channel reduced peak current density and shifted half-inactivation voltage to more negative potentials, whereas voltage at the maximum peak current remained unchanged. In another mutation, the duplication of five amino acids in exon 43 of CACNA1C (p. E1829_Q1833dup) resulted in nearly compete suppression of calcium current (I Ca ) [102]. These lossof-function mutations may help to explain the shortening of QT interval and other features of ECG.

Brugada syndrome 4
In two right precordial ECG leads, the patient displayed a type I ST-segment elevation, and a novel p.T11I mutation in CACNB2b was identified. This mutant Ca V β 2b had a faster fast and slow decay I Ca and a reduced total charge compared to WT Ca V β 2b , but had almost same peak calcium current density, SSI and recovery from inactivation [104]. In 6 affected family members with BrS and a shortened QTc interval, a c.C1442T heterozygous mutation in exon 13 of the CACNB2b gene resulted in p.S481L substitution downstream of the BID of β 2b subunit [98]. This mutation was not found in either 4 phenotype-negative family members or in 400 ethnically matched control alleles. The calcium current in the cells coexpressing with p.S481L Ca V β 2b was dramatically smaller than WT Ca V β 2b , but the surface expression was almost the same, which indicated that the p.S481L channel traffics normally [98]. Due to the fact that the mutation is located at the position nearby the BID domain binding with α 1C subunits, the mechanism of dysfunctions of calcium channel could be interfered by the stimulatory role of β 2 subunit on LTCC [105].
A patient with BrS, who has a p.V2014I mutation in CACNA1C, also had a p.D601E polymorphism in CACNB2b, and this polymorphism dramatically augmented the late calcium current of LTCC, which prolonged the QT interval, the modulatory effect of this single nucleotide polymorphism probably interprets the fact that QTc in this proband is not associated with SQTS [102]. Other mutations were also identified [102,106], including p.S143F and p.T450I, which were reported in the prevailing BrS-associated variants of NHLBI GO Exome Sequencing Project population [107]. However, the functions of these mutations on the Ca V β 2 subunit remain unknown.

Brugada syndrome 9
Human mutations in CACNA2D1 have been identified to be associated with several kinds of cardiac dysfunctions, including BrS [97]. It has been found 3 different missense mutations in CACNA2D1 (p. S709N, p.D550Y, and p.Q917H) in 3 BrS patients from a cohort made up of 205 patients with BrS [102]. Among these mutations, p.S709N and p. Q917H were also identified in Exome Sequencing Project population by Risgaard et al. [107]. Nevertheless, more works are needed to explore the possible functions of these mutations on Ca V channels.

Short QT syndromes
The short QT syndrome (SQTS), first described in 2000 [108], was defined by an abnormally shortened QT intervals and a propensity for cardiac arrest (CA) [109,110]. SQTS has been associated with the gain-of-function mutations in 3 distinct potassium channels, KCNH2, KCNQ1 and KCNJ2, which cause SQT1, SQT2 and SQT3, respectively [111][112][113][114]. Nowadays, more evidences indicate the mutations in LTCC are also linked to SQTS. Some cardiac arrhythmias have the combined phenotypes of SQTS and BrS, since the shortened QT interval is also one of the manifestations of BrS. Thus, BrS3, caused by the mutation in CACNA1C, such as p.A39V, p.G490R, p.E1829_Q1833 duplication, and p.E850 deletion, also known as SQT4; and BrS4, caused by CACNB2 p.S481L mutation, is also known as SQT5 [98].

Short QT syndrome 4
One proband's ECG showed that the QTc interval of the patient was shorter than that of healthy controls and characterized with aborted CA, and a novel CACNA1C mutation (p.R1977Q) was first identified in this SQTS patient, but the parents didn't carry this mutation [115]. However, the roles of p.R1977Q mutation in SQTS are still unknown. Recently, a novel CACNA1C mutation p.R1937P was reported in a Chinese family of hypertrophic cardiomyopathy with early repolarization and SQTS, p.R1937P induced the loss-offunction of Ca V 1.2 channels, which dramatically decreased the I Ca and hyperpolarized the SSI [116].

Short QT syndrome 6
A novel type of SQTS (SQT6) caused by a mutation p.S755T in the CACNA2D1 was reported in 2011 [117]. The ECG of the patient revealed a shortened QT interval. Templin et al. took advantage to analyze the functional roles of the p.S755T mutation in transfected HEK293 cells. Compared with the WT variant, p.S755T α 2 δ-1 subunit extremely reduced the barium currents under the whole-cell patch clamp recording [117]. Western blotting showed that the membrane expression of α 1C subunit with mutant α 2 δ-1 was similar to WT α 2 δ-1 subunit, thus this mutation might only alter the open probability (P o ) of the Ca V 1.2 channels without modulating the membrane expression of the α 1C subunit of LTCC.

Cardiac conduction disease
Combined mutations in CACNB2b (p.K170N/ L399F) in one infant displaying BrS phenotype had a severe intraventricular conduction delay, and another case with conduction delay had identified p.D538E mutation in CACNB2b [118]. A novel polymorphism (p.D601E) was also identified in CACNB2b in the family displaying first-degree atrioventricular block, which induced the slowed inactivation of Ca V 1.2 channels by strongly increasing the total charges [119]. Together with the mutation of SCN5A (Na V 1.5 sodium channel) in this case, the sodium current reduction can cause loss of the right ventricular epicardial AP dome absent from the slowed inactivation of calcium current and slowed conduction, whereas p. D601E polymorphism could restore the dome of APs [119]. These results suggest the gain-of-function mutation in calcium channel can rescue the loss-of-function of sodium channel mutation in the process of cardiac conduction disease without BrS. It is notable that the p.D601E polymorphism in CACNB2b was also found in the BrS, and this polymorphism could increase the late calcium current [102], indicating the multiply functions of p. D601E polymorphism.

Idiopathic ventricular fibrillation and early repolarization syndrome
Early repolarization has been found in patients with idiopathic ventricular fibrillation (IVF) [120,121], these IVF associated with early repolarization was also called early repolarization syndrome (ERS). Mutations in the LTCC genes, including CACNA1C and CACNB2b, have been found to associate with both diseases. Burashnikov et al. reported CACNB2b p.A73V mutation was associated with IVF; and CACNB2b p.S160T or p.R571C, CACNA2D1 p.S956T, and CACNA1C p.E850 deletion had possible roles in ERS [102]. The glutamic acid at 850 position is located within the PEST domain of Ca V α 1 II-III loop. The mutations in the PEST domain are proved to affect the surface membrane expression of Ca V 1.2 channels [84]. Functionally, p.E850 deletion in CACNA1C is found to lead to an almost complete loss-of-function in I Ca [90]. Thus, the p. E850 deletion might decrease the surface expression of LTCC presumably because of an aberrant protein degradation, which induces a dramatic reduction of I Ca .

Bradycardia and sinoatrial arrhythmia
Autorhythmicity of sinoatrial nodal pacemaker cells results from the slow auto-depolarization as a result of Ca V 1.3 calcium influx [122,123]. The inactivated Ca V 1.3 channels in mice induced strong reduction of calcium current in pacemaker cells and profoundly affected its pacemaking effect, which showed predominant sinoatrial nodal dysfunction [124]. With the inactivated Ca V 1.3 calcium currents, both sinoatrial arrhythmia and bradycardia have been observed [23,125]. A homozygous 3-bp insertion in CACNA1D, inducing p. G403_404ins, was first screened from a Pakistani family with pronounced bradycardia. This p. G403_404ins, located at alternative spliced exon 8b of Ca V α 1D , resulted in nonconducting Ca V 1.3 channels [126], which may explain the phenotype of bradycardia.

Challenges and future diagnostic or therapeutic approaches
Mosaicism implies that in an organism from a single zygote that is the presence of genetically distinct cell line [127]. Taken TS as an example, the p.G406R mutation in exon 8 could be detected in the oral mucosa sample but not in the blood sample of the patients' mother [61], whereas in the case of a Chinese girl with a typical TS phenotype, the G406R mutation also expressed in her father's oral mucosa, sperm and white blood cell [128]. Additionally, cardiac arrhythmias also have phenotypic heterogeneity. The genotype-negative or even phenotype-negative LQT8 patients had been identified several novel mutations in CACNA1C [82,88]. Hence, careful screening of parental tissue in family with incidence of congenital cardiac arrhythmias, prenatal DNA screening, and monitoring of the fetal echocardiogram are extremely significant for the congenital cardiac arrhythmia. Remarkably, whole-genome or whole-exome sequencing [129] may be necessary for the identification of potential mutations of LTCC in cardiac arrhythmias.
LTCC has many variants because of transcriptional and post-transcriptional modification, which is complicate and may be the future challenges of the field. For example, alternative splicing, one of most important mechanisms of posttranscriptional modification [130,131], generates more than 20 alternative spliced exons in α 1C mRNA, which forms many variants of α 1C [14,60,132]. It has been indicated the mutations in these alternative spliced exons of LTCC can also attribute to the abnormal cardiac excitation [61,62,126]. Therefore, it is noteworthy to screen the potential mutations in these alternative spliced exons of LTCC in the patients with cardiac arrhythmias.
To date, most functional studies utilized heterologous systems expressing WT or mutant calcium channels in the cultured cells, which do not fully recapitulate the phenotypes of cardiomyocytes. With the advent of iPSC, the concept of cellular reprogramming has been revolutionized, which makes producing unique human iPSC from somatic cells possible [133], then induced to specific cells. TS patient-specific iPSC-derived cardiomyocyte (iPSC-CM) carrying a missense mutation p.G406R in the Ca V 1.2 had been generated by traditional method or genetically encoded fluorescent indicators, which increases the inward calcium currents leading to prolonged QT interval [72,134]. Recently, genetic mutations could be directly introduced into generic embryonic stem cell or non-diseased iPSCs by genome editing techniques, to produce the phenotypes of cardiac arrhythmia [135], providing an exciting approach to drug screening in the iPSC-CM-based disease model. However, iPSC techniques have several limitations when being used as a treatment strategy for the diseased heart, e.g. the stem cellderived cardiomyocytes being cocultured with native cardiomyocytes showed an slow conduction velocity [136] and inadequate excitation contraction coupling [137]. Nevertheless, the iPSC approach could potentially be a powerful tool for diagnosis and prognosis of cardiac arrhythmias and provide a robust assay for developing new drugs to treat these diseases [138][139][140]. Moreover, isogenetic control of iPSCs provides a method to reveal the pathogenic mechanism underlying the specific disease phenotype. In brief, the iPSC-based technology will play a central role in specific aspects of translational medicine [141].
It has been understood that major LTCC mutation-associated cardiac arrhythmias are attributed to the over-activities of calcium channel functions, which are due to the gain-of-function mutations in LTCC. However, the treatment of cardiac arrhythmias by calcium channel blockers meets a lot of challenges, e.g. limited range of available choices, and insufficient efficacy [2,142]. Moreover, the LTCC channel blockers, which have been used in clinic for decades, have none or very little selectivity for the different LTCC variants to some extent. At the same time, these drugs induce a lot of adverse effect risks because of off-target effects. Therefore, developing highly selective calcium channel blockers targeting different variants of LTCC, or even mutant LTCC will be much attractive in the management of these mutation-associated cardiac arrhythmias.

Conclusions
Taken together, the mutations of the LTCC genes can cause different cardiac arrhythmias, including LQTS, TS, BrS, SQTS and other cardiac arrhythmias, which are critically associated with increased risks of cardiovascular diseases, syncope and even SCD. The mechanisms how these mutations cause the distinct cardiac arrhythmias are dependent on their different roles in the channel functions (structure-function mechanisms), thus more efforts on the crystal structure analysis of the LTCC are appreciated to understand these mutation-associated cardiac arrhythmias. Nevertheless, there are some challenges in the detection, diagnosis and treatment of cardiac arrhythmias at present, the future studies will be necessary to get a better understanding of the roles of LTCC in the context of cardiac arrhythmias, which will be much valuable for the diagnosis and even management of these diseases.

Disclosure statement
No potential conflict of interest was reported by the authors.

Author Contributions
Q.Z., J.C., Y.Q. and J.W. drafted the manuscript. J.W. prepared the figures. J.W. edited the manuscript. L.Z. read and gave the comments to this work.