Epilepsy in hemiplegic migraine: Genetic mutations and clinical implications

Abstract

Objective

We performed a systematic review on the comorbidities of familial/sporadic hemiplegic migraine (F/SHM) with seizure/epilepsy in patients with CACNA1A, ATP1A2 or SCN1A mutations, to identify the genotypes associated and investigate for the presence of mutational hot spots.

Methods

We performed a search in MEDLINE and in the Human Gene Mutation and Leiden Open Variation Databases for mutations in the CACNA1A, ATP1A2 and SCN1A genes. After having examined the clinical characteristics of the patients, we selected those having HM and seizures, febrile seizures or epilepsy. For each gene, we determined both the frequency and the positions at protein levels of these mutations, as well as the penetrance of epilepsy within families.

Results

Concerning F/SHM-Epilepsy1 (F/SHME1) and F/SHME2 endophenotypes, we observed a prevalent involvement of the transmembrane domains, and a strong correlation in F/SHME1 when the positively charged amino acids were involved. The penetrance of epilepsy within the families was highest for patients carrying mutation in the CACNA1A gene (60%), and lower in those having SCN1A (33.3%) and ATP1A2 (30.9%) mutations.

Conclusion

Among the HM cases with seizure/epilepsy, we observed mutational hot spots in the transmembrane domains of CACNA1A and ATP1A2 proteins. These findings could lead to a better understanding of the pathological mechanisms underlying migraine and epilepsy, therein guaranteeing the most appropriate therapeutic approach.

Keywords

FHM epilepsy genetics migraine

Introduction

More than a century of investigation has found that seizures and migraine can co-occur in affected individuals and families, and there is much evidence that these diseases also share genetic susceptibility (1). In both epilepsy and migraine with aura (MwA), susceptible brain regions are hyperexcitable and attacks are believed to begin with hypersynchronous neuronal firing. The involvement of this mechanism in migraine without aura (MwoA) is less clear (1). Moreover, the mechanisms responsible for neuronal hypersynchronous activity are also not well understood. However, genetically determined dysfunction of ion transporters seems to indicate a common underlying mechanism for both paroxysmal disorders. In fact, over the last two decades, the mutations in the ion transportation genes CACNA1A, ATP1A2 and SCN1A have been identified in familial hemiplegic migraine (FHM) (2 –4). The prevalence of epilepsy in FHM has been estimated to be around 7% (5), while among patients with sporadic HM (SHM), in which recently an increasing number of CACNA1A and ATP1A2 mutations have been reported, its value is unknown. Though the underlying pathophysiological mechanisms of these associations are not well understood, it is evident that in these patients DNA variants are able to confer susceptibility to both HM and seizure/epilepsy (6). Different questions are still unanswered regarding the presence of epilepsy in HM patients, including the possible presence of a specific genotype related to this endophenotype, its penetrance within families, and the clinical characteristics of these patients. Since these items have not yet been investigated, we have here performed a systematic review on these comorbidities. The results drawn from our analysis add insight that can improve our capacity to predict, to diagnose, and manage patients with HM and epilepsy.

Materials and methods

Search strategy and study selection

We followed PRISMA 2009 guidelines for systematic review and meta-analysis (7). We sought to identify all published studies on the association between HM and seizure/epilepsy in patients or families carrying mutations in one of the HM genes. A search on MEDLINE (Ovid, New York, NY) from September 1993 to January 2016 was performed covering the time from the identification of the first FHM gene (4). No language restrictions were applied. The MEDLINE search strategy included both subject headings (MeSH terms) and text words for the target condition, including: FHM AND epilepsy; FHM AND seizure; HM AND epilepsy; HM and seizures; CACNA1A AND seizure/epilepsy; ATP1A2 AND seizure/epilepsy; and SCN1A AND seizure/epilepsy. We also searched reference lists and relevant journals not indexed in MEDLINE or EMBASE. Additionally, we searched in the Human Gene Mutation Database (HGMD, http://www.hgmd.cf.ac.uk/ac/index.php) and Leiden Open Variation Database (LOVD, http://www.lovd.nl/3.0/home) for mutations in the CACNA1A, ATP1A2 and SCN1A genes and their related phenotypes. We considered DNA variants as “mutations” when 1) functionally characterized and/or 2) segregated with the disease and/or 3) de novo variants. We also reviewed three publicly accessible variant resources to exclude polymorphisms (considering a Minor Allele Frequency <0.002): db SNP at http://www.ncbi.nlm.nih.gov/SNP/; ExAC Browser at http://exac.broadinstitute.org/ and Exome Variant Server at http://evs.gs.washington.edu/EVS/.

Inclusion and exclusion criteria

Published studies and cases from the HGMD/LOVD were included if they reported an association between HM and seizure/epilepsy and their mutations were identified. We excluded review articles, editorials, commentaries, hypothesis papers, letters with no new data, as well as abstracts.

Data extraction

Two review authors independently extracted data; disagreements were resolved by discussion or referring to a third author. We collected data on genotype (DNA mutation), protein modification and patient information related to HM, headache, seizure, epilepsy and other neurological phenotypes, such as intellectual disability (ID), ataxia, and coma.

Data synthesis

For each gene, we selected the mutations associated with HM and seizure/epilepsy, so as to determine the relative frequencies of these mutations among all known mutations for that specific gene. Moreover, we calculated the penetrance (within a family, the percentage of individuals having a gene mutation and exhibiting epilepsy) of HM, seizure/epilepsy and other neurological phenotypes (ID, ataxia, coma). We also plotted the mutations in each protein structure regarding their locations in the transmembrane, extracellular or cytoplasmic domains (Figure 1). We performed a Chi-square test to investigate for associations between mutation location and phenotype.

Figure 1.

The figure illustrates the distribution of all the mutations leading to HM in the three protein channels (CACNA1A, ATP1A2, SCN1A). The numbers inside the circles correspond to mutations, which are listed in a rising numerical order in the legend. Mutations associated with HM and epilepsy are shown as black circles, while those associated with HM are shown in white. The reference transcripts used for the CACNA1A gene is NM_001127221.1, for ATP1A2 is NM_000702.3, and for SCN1A is NM_001165963.1.

Results

Our search found 22 family reports, three family series studies, 13 case reports, one case series study, one case series and a family series study (Table 1 –3), for a total of 252 patients belonging to 56 families, which were all included in this study (2,3,5,8 –44).

Table 1.

List of the CACNA1A gene (NM_001127221.1) mutations associated with HM/epilepsy, and clinical characteristics of patients.

Mutation	Protein change	Domain	Number of patients/ families	Patients with migraine/total	Patients with epilepsy – seizure/total	Patients with other phenotype/total	References
c.653C > T	S218L	IS4–IS5	1/1	SHM1 (1/1)	Post-traumatic complex partial (1/1)	Ataxia/ID (1/1)	11,28,39
			1/1	SHM1 (1/1)	Post-traumatic generalized (1/1)	Ataxia (1/1)
			3/1	FHM1 (1/3)	Generalized, GEFS+ (2/3)	Ataxia (1/3)
			1/1	SHM1 (1/1)	Partial (1/1)	Ataxia (1/1)
1/1	SHM1 (1/1)	Post-traumatic generalized (1/1)	Ataxia (1/1)
c.653C > T	S218L	IS4–IS5	1/1	SHM1 (1/1)	Hemiconvulsion- hemiplegia-epilepsy syndrome (1/1)	–	37
c.653C > T	S218L	IS4–IS5	1/1	SHM1 (1/1)	Generalized (1/1)	Ataxia (1/1)	38
c.653C > T	S218L	IS4–IS5	1/1	SHM1 (1/1)	Complex partial, generalized (1/1)	–	29
c.1088T > C	F363S	IS6–IIS1	1/1	SHM1 (1/1)	Partial (1/1)	Ataxia/ID (1/1)	40
c.4037G > A	R1346Q	IIIS4	5/1	FHM1(5/5)	Partial (2/5)	Ataxia (4/5)	21
c.4037G > A	R1346Q	IIIS4	1/1	SHM1 (1/1)	Partial (1/1)	Ataxia/ID (1/1)	40
c.4037G > A	R1346Q	IIIS4	1/1	SHM1 (1/1)	No data (1/1)	Ataxia/ID (1/1)	34
c.4037G > A	R1346Q	IIIS4	1/1	SHM1 (1/1)	Absence (1/1)	Ataxia/ID (1/1)	32
c.4037G > A	R1346Q	IIIS4	1/1	SHM1 (1/1)	No data (1/1)	Ataxia/ID (1/1)	33
c.4151A > G	Y1385C	IIIS5	1/1	SHM1 (1/1)	Partial (1/1)	Ataxia/ID (1/1)	36
c.4151A > G	Y1385C	IIIS5	1/1	SHM1 (1/1)	Partial, generalized (1/1)	Ataxia (1/1)	5
c.4517T > C	F1506S	IIIS6	1/1	SHM1 (1/1)	Partial (1/1)	Ataxia/ID (1/1)	40
c.4999C > T	R1667W	IVS4	1/1	SHM1 (1/1)	Generalized (1/1)	Ataxia (1/1)	5
c.5009A > G	K1670R	IVS4	1/1	SHM1 (1/1)	Generalized (1/1)	–	40
c.5047T > A	W1684R	IVS4–S5 junction	2/1	FHM1 (2/2)	Partial, generalized (1/2)	Ataxia (1/2)	5
c.5085G > A	V1696F	IVS5	2/1	FHM1 (2/2)	Generalized (2/2)	Ataxia/ID (2/2)	13
c.5126T > C	I1709T	IVS5	1/1	SHM1 (1/1)	Status Epilepticus (1/1)	–	40
c.5126T > C	I1709T	IVS5	3/1	FHM1 (3/3)	Complex partial, generalized (2/3)	Ataxia (3/3)	17
Total			33/23	31/33	27/33	25/33

Abbreviations: FHM, familial hemiplegic migraine; SHM, sporadic hemiplegic migraine; GEFS+, Generalized epilepsy with febrile seizures plus; ID, intellectual disability. “No data” refers to the absence of a detailed clinical description, but the patient had epilepsy.

Table 2.

List of the ATP1A2 gene (NM_000702.3) mutations associated with HM/epilepsy, and clinical characteristics of patients.

Mutation	Protein change	Domain	Number of patients/ families	Patients with migraine/ total	Patients with epilepsy- seizure/total	Patients with other phenotype/ total	References
c.103_105delAAG	K35del	N-term	2/1	FHM (1/2)	No data (1/2)	–	40
c.901G > A	G301R	M3	11/1	FHM (8/11)	Partial, generalized (5/11)	Ataxia (3/11)	21
c.1091C > T	T364M	M4–M5	1/1	SHM (1/1)	Febrile (1/1)	–	35
c.1133C > A	T378N	M4–M5	4/1	FHM (4/4)	Generalized (4/4)	–	23
c.1642C > T	R548C	M4–M5	7/1	FHM (7/7)	Absence, Generalized (1/7)	–	18
c.1766T > C	I589T	M4–M5	1/1	SHM (1/1)	No data (1/1)		27
c.1844G > A	G615E	M4–M5	1/1	SHM (1/1)	Generalized (1/1)	–	40
c.1843G > A	G615R	M4–M5	4/1	FHM (4/4)	Partial (1/4)	ID (1/4)	25
c.2170G > A	R689Q	M4–M5	18/1	FHM (14/18)	Febrile (6/18)	–	24
c.2105G > A	C702Y	M4–M5	7/1	FHM (1/7)	Complex partial, febrile, status epilepticus (3/7)	–	15
c.2152G > A	D718N	M4–M5	7/1	FHM (6/7)	Generalized (1/7)	ID (1/7)	26
c.2395T > C	L764P	M4–M5	33/1	FHM (28/33)	No data (2/33)	–	2
c.2324A > G	Y775C	M5	1/1	SHM (1/1)	Febrile (1/1)	ID (1/1)	40
c.2564G > T	G855V	M7	1/1	SHM (1/1)	Absence (1/1)	ID (1/1)	40
c.2563G > A	G855R	M7	3/1	FHM (3/3)	Febrile (3/3)	–	14
c.2620G > A	G874S	M7–M8	6/1	FHM (5/6)	GEFS+, complex partial (3/6)	–	12
c.2763T > C	W887R	M7–M8	7/1	FHM (6/7)	No data (2/7)	–	2
c.2698G > C	G900R	M7–M8	14/1	FHM (11/14)	Status epilepticus, complex partial, generalized, partial, febrile (5/14)	–	25
c.2723G > A	R908Q	M7–M8	3/1	FHM (2/3)	Migralepsy, generalized (2/3)	ID (1/3)	20
c.2780A > G	Q927P	M8	2/1	FHM (1/2)	Generalized (1/2)	–	40
c.2936C > T	P979L	M9–M10	5/1	FHM (5/5)	Generalized (1/5)	ID (1/5)	26
c.2980_2982delCTC	L994del	M10	1/1	SHM (1/1)	Febrile (1/1)	ID (1/1)	40
c.2995G > C	D999H	M10	31/1	FHM (30/31)	Generalized, partial, absence, febrile (10/31)	ID (5/31)	16
c.3005G > A	R1002Q	M10	1/1	SHM (1/1)	No data (1/1)	–	31
c.3007A > G	K1003E	M10	1/1	SHM (1/1)	Partial (1/1)	–	40
c.3019C > T	R1007W	M10	7/1	FHM (4/7)	Partial, generalized (2/7)	–	19
c.3027T > A	T1009X	M10	1/1	SHM (1/1)	Generalized (1/1)	–	30
Total			180/27	149/180	62/180	15/64

Table 3.

List of the SCN1A gene (NM_001165963.1) mutations associated with HM/epilepsy, and clinical characteristics of patients.

Mutation	Protein change	Domain	Number of patients/ families	Patients with migraine/total	Patients with epilepsy-seizure/total	Patients with other phenotype/total	References
c.787C > G	L263V	IS5	6/1	FHM (6/6)	Generalized, complex partial (4/6)	–	9
c.787C > G	L263V	IS5	5/1	FHM (5/5)	Generalized, complex partial (3/5)		8
c.3521C > G	T1174S	IIS6–IIIS1	6/1	FHM (2/6)	Febrile, partial (3/6)	–	10
c.4465C > A	Q1489K	IIIS6–IVS1	22/3	FHM (22/22)	Febrile, partial (3/22)	–	3
c.4467G > C	Q1489H	IIIS6–IVS1	6/1	FHM (6/6)	Partial with secondary generalization (2/6)	Elicited repetitive daily blindness (4/6)	42
Total			45/7	41/45	15/45	4/45

Abbreviations: FHM, familial hemiplegic migraine; SHM, sporadic hemiplegic migraine; GEFS+, generalized epilepsy with febrile seizures plus; ID, intellectual disability.

F/SHM1 and CACNA1A

Families and patients

Our search found 33 patients belonging to 23 families. In particular, 15 patients belonged to five families and 18 were sporadic cases. Of the former 15 patients, only nine had epileptic comorbidity, so the total number of patients having HM and epilepsy (here defined as F/SHME1, Familial/Sporadic Hemiplegic Migraine and Epilepsy type 1) was 27, nine familial (9/27, 33.3%) and 18 sporadic (18/27, 66.6%) (Table 1).

Mutation type and its distribution

We found 10 different CACNA1A variants associated with HM and seizures/epilepsy (Table 1). The 10 variants were all missense, the most frequent being S218L, which lies in the 3rd cytoplasmic domain, and was encountered in 10/27 patients (37%). The other nine variants produced a random distribution at the genomic level whereas, at the protein level, two were localized in the 4th and 12th cytoplasmic domains, while the remaining seven variants were all localized in the transmembrane domains III and IV (7/10, 70%); in fact, none of the mutations in the I and II transmembrane domains led to either seizure or epilepsy (p < 0.01). Nine out of 10 patients having S4 segment involvement showed a substitution of either an Arginine (R) or Lysine (K) (positively charged amino acid) with another neutral amino acid (Q, Glutamine; W, Tryptophan).

Mutation frequency and penetrance of seizures/epilepsy

The overall frequency of mutations associated with HM and seizure/epilepsy was 10/154 (6.5%), where 154 was the total number of mutations reported in the HGMD/LOVD for the CACNA1A gene related to three different phenotypes: Episodic Ataxia type 2 (EA2), Spinocerebellar Ataxia type 6 (SCA6), and FHM1. Of these 154 mutations, 30 were associated with HM (sporadic or familial) (legend of Figure 1), resulting in a frequency of HM mutations associated with seizure/epilepsy equal to 33.3% (10/30). This frequency was unexpectedly high considering the low number of patients, as these mutations are generally associated with severe phenotypes (HM and ataxia or HM, ataxia, and ID) in SHM cases. Regarding the penetrance of epilepsy, we found it to be 60%, particularly high, even though it was calculated on the basis of few familial cases.

Clinical description

The S218L variant was associated with the development of “post-traumatic seizures” and, in only one case, with generalized epilepsy with febrile seizures plus (GEFS+). While the seizures associated with the other mutations were classified as either generalized or focal, in only one case was absence-like seizure diagnosed. Moreover, the presence of ataxia alone (15/27 patients, 55.5%) or in association with ID (10/27 patients, 37%) was very frequent. Eight of the 10 more complex and severe cases (having HM, ataxia and ID) were sporadic (Table 4).

Table 4.

Summary of the results from this review. Abbreviations: GEFS+, generalized epilepsy febrile seizures plus; ID, intellectual disability; HM, hemiplegic migraine; FHME, familial hemiplegic migraine and epilepsy; SHME, sporadic hemiplegic migraine and epilepsy.

Features	CACNA1A	ATP1A2	SCN1A
Number of patients with HM and epilepsy	27	62	15
FHME	9 (33.3%)	53 (85.5%)	15 (100%)
SHME	18 (66.6%)	9 (14.5%)	0 (0%)
Number of mutations associated with epilepsy/total number of HM mutations	10/30 (33.3%)	27/66 (40.9%)	4/9 (44.4%)
Penetrance of epilepsy	60%	30.9%	33.3%
Protein domain affected	Transmembrane (70%) Cytoplasm (30%) –	Transmembrane (41%) Cytoplasm (41%) Extracellular (18%)	Transmembrane (25%) Cytoplasm (75%) –
Principal clinical presentation of epileptic phenotype	Post traumatic Generalized Focal GEFS+ –	– Generalized – – Febrile	– Generalized Focal – Febrile
Percentage of patients with severe phenotype (HM, epilepsy, ataxia, ID)	37%	11.3%	0%

F/SHM2 and ATP1A2

Families and patients

We found 180 patients belonging to 27 families. In particular, 171 patients belonged to 18 families, while nine patients were sporadic cases. Of the 171 patients, 53 had epileptic comorbidity, so that the total number of patients having HM and epilepsy (here defined as F/SHME2, Familial/Sporadic Hemiplegic Migraine and Epilepsy type 2) was 62, 53 familial (53/62, 85.5%) and nine sporadic (9/62, 14.5%) (Table 2).

Mutation type and its distribution

We found 27 different ATP1A2 variants associated with HM and seizures/epilepsy (Table 2). Of these variants, two were small deletions, and the remaining 25 single nucleotide substitutions led to missense mutations. The mutations associated with seizure/epilepsy were located in the transmembrane (11/27, 41%), cytoplasmic (11/27, 41%) or extracellular domains (5/27, 18%). This finding was in contrast to the distribution of mutations observed in FHM2. In fact, of the 66 mutations responsible for FHM2 (legend of Figure 1), 43 were located in the cytoplasmic domains (43/66, 65.2%), while only 15/66 (22.7%) and 8/66 (12.1%), were situated in one of the 10 transmembrane domains or in the extracellular domains, respectively. Moreover, 11/15 (73,3%) mutations affecting the transmembrane domains were associated with epilepsy, compared with only 16/51 (31%) of those outside the transmembrane domains (p < 0.05) (Figure 1).

Mutation frequency and penetrance of seizures/epilepsy

The overall frequency of mutations associated with seizure/epilepsy was 27/73 (36.9%), where 73 was the total number of mutations reported in the HGMD/LOVD for the ATP1A2 gene related to migraine or HM phenotypes. Of these 73 mutations, 66 were associated with HM (sporadic or familial) (legend of Figure 1), resulting in a frequency of ATP1A2 mutations associated with HM and seizure/epilepsy equal to 40.9% (27/66), while the median penetrance of epilepsy was calculated to be 30.9% within the 18 families.

Clinical description

The epileptic phenotype appeared to be mostly of a generalized type, and in eight families it was reported to be triggered by fever. Ataxia was present in three patients belonging to the same family and was associated with the G301R mutation, whereas ID was observed in both the sporadic (three patients) and familial cases (four patients belonging to four families), with an overall frequency of 11.3% (7/62 patients) (Table 4).

FHM 3 and SCN1A

Families and patients

We found 45 patients belonging to seven families (100% of familial cases). A total of 15 patients had the epileptic comorbidity endophenotype here defined as FHME3 (Familial/Sporadic Hemiplegic Migraine and Epilepsy type 3) (Table 3).

Mutation type and its distribution

We found four different SCN1A mutations associated with HM and seizures/epilepsy (Table 3). All four DNA variants were single nucleotide substitutions leading to missense mutations. The L263V mutation was located in a transmembrane domain (IS5), whereas the T1174S, Q1489K and Q1489H mutations were located in the IIS6-IIIS1 and IIIS6-IVS1 cytoplasmic domains, respectively. However, the T1174S variant needs to be considered with caution, since there isn’t a definitive conclusion about its clinical significance.

Mutation frequency and penetrance of seizures/epilepsy

The overall frequency of SCN1A mutations associated with HM and epilepsy was 4/852 (0.47%), with 852 being the total number of mutations reported in the HGMD/LOVD for the SCN1A gene associated with different phenotypes (FHM3, Dravet syndrome, generalized epilepsy with febrile seizures plus (GEFS+), febrile seizures). To date, only nine SCN1A mutations associated with FHM3 have been described (Figure 1 and legend) (3, 8 –10, 41 –44); therefore the relative frequency of SCN1A mutations associated with HM and seizure/epilepsy was 44.4% (4/9), with a median penetrance of the epilepsy within these families of 33.3%.

Clinical description

L263V was more often associated with generalized epilepsy, while patients with T1174S or Q1489K mutations had focal seizures plus febrile seizures. Finally, the two patients with Q1489H mutation had partial epilepsy with occasional secondary generalization (childhood benign epilepsy), complicated by elicited repetitive daily blindness. There were no reported cases of ataxia or ataxia/ID in these families (Table 4).

Discussion

After having reviewed the cases of patients with HM and seizures/epilepsy carrying mutations in the CACNA1A or ATP1A2 or SCN1A genes, we present here a detailed picture of the frequency and location of each mutation, as well as of the penetrance of epilepsy within the families (Table 4).

Patients with F/SHME1 were mostly sporadic cases (66.6%), while the penetrance of epilepsy in the few familial cases was very high (60%), thus suggesting a strong correlation between specific mutations and the onset of epileptic discharge. Specifically, with the exceptions of S218L, F363S and W1684R, the other seven CACNA1A variants were associated with seizure/epilepsy (7/10, 70%) and were localized in the transmembrane domains (S1–S6). Moreover, patients with F/SHM1 were more likely to develop epilepsy when the mutation modified positively charged amino acids in the III or the IV S4/S5 domains, whereas no missense mutation affecting the first two S4 segments of the Ca_v2.1 channel (encoded by the CACNA1A gene) resulted in being associated with seizures/epilepsy. Accordingly, Beyl et al. (45) reported that neutralization of gating charges in the I and II S4 transmembrane domains of the Ca_v1.2 calcium channel (encoded by the CACNA1C gene) only slightly affect the equilibrium constant of voltage sensor transition. Therefore, F/SHM1 patients harbouring a missense mutation lying in the III or the IV S4/S5 domains appear to be more susceptible to developing epilepsy and therein need to be closely followed up and managed for this higher risk. The mechanisms underlying this genetic susceptibility can be hypothesized on the basis of the functional and animal studies conducted to date. To this regard, studies in heterologous expression systems have reported that the mutations associated with FHM1 lead to an enhanced single channel Ca²⁺ influx also for mild depolarizations, indicating an increased channel open probability due to a shift to lower voltages of channel activation (6). Notably, compared with the other FHM1 mutations, S218L (the most frequent mutation among FHME1 patients) produces the largest shift of activation and the largest gain-of-function of human Ca_v2.1 channels, especially in the case of small depolarization (1,6,46). Furthermore, S218L knockin mice have shown a unique increased susceptibility to repetitive successive cortical spreading depression (CSD) events, following a single CSD-inducing stimulus (6,47,48). This particularly low CSD threshold and the strong tendency to respond with multiple CSD events may in fact cause the severe neurological phenotype, and could also be involved in an increased susceptibility to seizures (49). In support of this hypothesis, recent research suggests that repeated CSD activity is able to lower the seizure threshold or induce increased excitability, therein promoting a change into epileptic status in isolated normal or injured (traumatic or perinfarctual) brains (49,50). This also might explain the underlying mechanisms of the other mutations associated with FHME1.

The association of CACNA1A mutations with ataxia is explained by the relevant role of Cav2.1 channels in the cerebellum for the maintenance of the highly regular, intrinsic, pacemaking of Purkinjie cells (PCs). The disruption of the pacemaking of PCs can be the result of either gain or loss-of-function mutations (51). Episodic Ataxia type 2 (EA2) is a disease that starts in childhood or early adolescence (age range 2–32 years) and is characterized by paroxysmal attacks of ataxia, vertigo, and nausea, typically lasting hours to days. EA2 patients usually have truncating CACNA1A mutations, but some of these can have missense mutations such as G1483R, F1491S, V1494I, R1662H, and R1665Q. Interestingly, these missense variants lie mostly in the III and IV S4, S5 or S6 transmembrane domains (51), where the FHME1 mutations are also mainly localized. The functional consequences of EA2 missense mutations have been investigated by expressing the channels in either mammalian cells or oocytes, and a clear dominant negative effect of these mutants was revealed (51). Therefore, it is not possible to exclude a dominant negative effect also for the missense mutations leading to FHME1. Overall, it appears that FHME1 mutations generate an intermediate phenotype through a gain-of-function and/or dominant negative effect, in a continuum clinical spectrum between HM and EA2 where the clinical connecting point is the ataxia. In fact, ataxia is consistent in EA2, present in 51.7% of FHME1 patients, while rare or completely absent in FHM1. Mutations associated with F/SHME1 had a worse prognostic value, as the percentage of severe cases (HM, epilepsy, ataxia, ID) was high (37%) (Table 4). Clinically, the epilepsy was refereed as generalized, focal or post-traumatic (in patients with the S218L mutation) and only in one case was a GEFS+ phenotype reported.

Patients with F/SHME2 were mostly familial cases (85.5%), where the penetrance of epilepsy was quite high (30.9%), suggesting a rather strong effect of some specific mutations in the generation of epileptic discharge. Moreover, regarding the genotype-phenotype correlation, our review evidenced that 73.3% of mutations affecting the transmembrane domains were associated with epilepsy, thus patients carrying such mutations could be much more susceptible to developing epilepsy compared to those having mutations outside the transmembrane domains, for whom this risk appears to be lower (31%). Missense mutations affecting the transmembrane domains can lead to both the perturbation of catalytic site functions and the modification of pump kinetic characteristics (1,2,15,19). Specifically, malfunctioning of the Na⁺, K⁺-ATPase pump may lead to neuronal hyperexcitability, which facilitates paroxysmal depolarizing shift (PDS) or CSD, causing seizures or migraine, respectively (1,2,19). This malfunctioning can occur due to three synergistic events: An elevation of extracellular K⁺ levels, an accumulation of glutamate in the synaptic cleft, or an increase in intracellular Ca²⁺ concentrations (15). Clinically speaking, epilepsy has been described mainly as generalized and in some cases triggered by fever. The severe cases, which were very rare among FHM2 patients, were instead more frequent (11.3%) among patients with F/SHME2 (Table 4).

Finally, regarding 852 SCN1A variants, these have been described in association with different phenotypes (GEFS+, Dravet syndrome, FHM3, febrile seizures), which can affect the entire protein structures in the intracellular, extracellular and transmembrane domains. Patients with FHME3 were all familial cases, and the penetrance of epilepsy was 33.3%, a value that is lower than expected for a gene whose mutations are strongly associated with epilepsy. The genotype-phenotype correlation in these cases seemed to be highly dependent on the functional effect of the mutation, rather than their positions in the canal proteins. In fact, mutations leading to loss of function were more commonly associated with severe phenotype (Dravet syndrome), whereas gain-of-function mutations were usually associated with either GEFS+ or FHM3, as they were milder phenotypes (3,5,9). In the case of FHM3 experimental evidence, it suggested that a single mutation can have double functional effects. In fact, Cestele et al. (10), applying a genetic and functional approach, studied a family having the T1174S (c.3521 C > G) SCN1A mutation associated with seizure and/or HM, reporting that this variant induced divergent functional effects, loss of function or gain of function, a finding consistent with both epilepsy and HM phenotypes, respectively. However, to date, the T1174S variant should not be considered a proven mutation causing disease. In fact, its clinical effect is still uncertain: Its frequency in polymorphism databases is 0.27% (European population), and it was not predicted to be damaging by the majority of prediction tools. On the other hand, the T1174S variant appears to be enriched in patients with epilepsy and/or FHM, compared to its frequency in the general population, and therefore might have a reduced penetrance, as one recent study suggested (52). Clinically, epilepsy was described as generalized, focal, partial with secondary generalization, and in some families febrile seizures were reported. Compared with FHME1 and FHME2 individually, the prognosis for FHME3 was not worse, as there were no severe cases or patients with other relevant complications among this subgroup.

Therapeutic implications

The efficacies of antiepileptic drugs (AEDs) in migraine and epilepsy have been reported to be mediated by interactions involved at multiple sites. In fact, these AED targets have more molecular sites in the brain, which are able to alter either voltage-operated ion channels or ligand-gated channels (53 –55). Among the drugs widely used for epilepsy, topiramate is the most effective for migraine when compared with valproate and gabapentin, whereas lamotrigine has been reported to be effective for MwA in open studies, but has not been demonstrated to be effective for MwoA (56). The effectiveness of some antiepileptic drugs (topiramate, valproate and lamotrigine) and other migraine prophylactic drugs for CSD threshold have been evaluated only in different animal studies (57,58), so the therapeutic effect of these drugs in migraine patients is not clear. Other relevant targets for migraine pathophysiology could also be effective in migraine but, even in this case, evidence has derived only from animal models, and therein these findings cannot be extrapolated to humans (59 –61).

Sodium valproate has been reported to reduce HM attacks, epileptic seizures and severe FHM attacks in patients suffering from both conditions (18), while evidence regarding the efficacy of topiramate is limited in HM (62) whereas lamotrigine has been reported to improve MwA in open-label studies, but not for MwoA in controlled trials vs a placebo or vs an active drug (63). This drug has a proven suppressive effect on CSD, which could explain its selective action on MwA (58). To date, the choice of preventive therapeutic strategies for HM has been based upon recommendations for common migraine, employing empirical strategies formulated on a limited number of FHM case reports. Conversely, AEDs including phenytoin, oxcarbazepine and vigabatrin block voltage-gated sodium channels and GABAergic transmissions, thereby reducing hyperexcitability mechanisms, but they are inefficacious in migraine (64). Despite these limitations with AEDs, the use of some anticonvulsivants, considered efficacious for both migraine and epilepsy, could be reasonable choices for treatment for HM and complex phenotypes, particularly HM and epilepsy. The evidence regarding the use of AEDs in literature is scarce. Therefore, clinicians should consider prescribing AEDs to HM patients having mutations associated with a marked susceptibility to developing epilepsy, instead of other anti-migraine drugs.

Conclusion

This review found that, in HM, mutations associated with seizure/epilepsy are not only frequent in S/FHM3 (44.4%) and S/FHM2 (40.9%) but also in S/FHM1 (33.3%), where in the latter the penetrance of epilepsy had the highest value (60%). Moreover, regarding the genotype-phenotype correlation, F/SHME1 and F/SHME2 had mutational hot spots at the protein levels. In fact, we observed prevalent and statistically significant involvement of the transmembrane domains for S/FHME1 (70%) and S/FHME2 (41%). In the former condition, the strongest genotype-phenotype correlation was observed when positively charged amino acids were involved in the S4 segments, while in the latter condition about 73% of mutations affecting the transmembrane domains were associated with seizure/epilepsy. These findings provide a better understanding of the pathological mechanisms underlying migraine and epilepsy and, if confirmed, these results would necessitate the recognition of an F/SHME subgroup.

Footnotes

Clinical implications

In hemiplegic migraine, the relative number of mutations associated with seizure/epilepsy was elevated not only for ATP1A2 and SCN1A genes, but also for the CACNA1A gene, where the penetrance of epilepsy for the latter had the highest value (60%).

Concerning F/SHM-Epilepsy1 (F/SHME1) and F/SHME2 endophenotypes, we observed a prevalent involvement of the transmembrane domains, and a strong genotype-phenotype correlation in F/SHME1 when the positively charged amino acids were involved.

The few SCN1A mutations associated with HM and epilepsy (FHME3) all belonged to familial cases. These genotype-phenotype correlations seem to depend on the functional effects of the mutations rather than their positions in the canal proteins.

Our results suggest the presence of mutational hot spots at the protein levels rather than a genotype-phenotype correlation at least for FHME1 and FHME2.

Author contributions

Conception/design: PP, CC, CP, SP. Collection and/or assembly of data: CS, PP, BC. Data analysis and interpretation: PP, CS, CC, SP. Manuscript writing: PP, CC, CML, CP, SP; Figures and tables: CC, CS, PP, CP. Revisions of manuscript: PP, CC, BC, CML, CP, CS, and SP. All authors provided important intellectual content and critically revised the final version of the manuscript.

Acknowledgement

We thank Dr Saviz Yaghmai for his precious support in preparing the illustration.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: PC receives/received research support from Biogen, Lundbeck, Merck-Serono, Sanofi-Aventis and UCB Pharma. He also receives/received support from Ricerca Corrente and Ricerca Finalizzata IRCCS Fondazione Santa Lucia. All other authors report no disclosures. This study is not industry-sponsored.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Rogawski MA. Migraine and epilepsy—Shared mechanisms within the family of episodic disorders. In: Noebels JL, Avoli M, Rogawski MA, et al. (eds) Jasper's Basic Mechanisms of the Epilepsies [Internet]. 4th edition. Bethesda (MD): National Center for Biotechnology Information (US), 2012, pp.1377–1398. Available from: https://www.ncbi.nlm.nih.gov/books/NBK98193/.

De Fusco

Marconi

Silvestri

et al.

Haploinsufficiency of ATP1A2 encoding the Na+/K+ pump alpha2 subunit associated with familial hemiplegic migraine type 2. Nat Genet 2003; 33: 192–196.

Dichgans

Freilinger

Eckstein

et al.

Mutation in the neuronal voltage-gated sodium channel SCN1A in familial hemiplegic migraine. Lancet 2005; 366: 371–377.

Ophoff

Terwindt

Vergouwe

et al.

Familial hemiplegic migraine and episodic ataxia type-2 are caused by mutations in the Ca2+ channel gene CACNL1A4. Cell 1996; 87: 543–552.

Ducros

Denier

Joutel

et al.

The clinical spectrum of familial hemiplegic migraine associated with mutations in a neuronal calcium channel. N Engl J Med 2001; 345: 17–24.

Pietrobon

. Familial hemiplegic migraine. Neurotherapeutics 2007; 4: 274–284.

Moher

Liberati

Tetzlaff

et al.

PRISMA Group). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. J Clin Epidemiol 2009; 62: 1006–1012.

Barros

Ferreira

Brandão

et al.

Familial hemiplegic migraine due to L263V SCN1A mutation: Discordance for epilepsy between two kindreds from Douro Valley. Cephalalgia 2014; 34: 1015–1020.

Castro

Stam

Lemos

et al.

First mutation in the voltage-gated Nav1.1 subunit gene SCN1A with co-occurring familial hemiplegic migraine and epilepsy. Cephalalgia 2009; 29: 308–313.

10.

Cestèle

Labate

Rusconi

et al.

Divergent effects of the T1174S SCN1A mutation associated with seizures and hemiplegic migraine. Epilepsia 2013; 54: 927–935.

11.

Chan

Burgunder

Wilder-Smith

et al.

Electroencephalographic changes and seizures in familial hemiplegic migraine patients with the CACNA1A gene S218L mutation. J Clin Neurosci 2008; 15: 891–894.

12.

Costa

Prontera

Sarchielli

et al.

A novel ATP1A2 gene mutation in familial hemiplegic migraine and epilepsy. Cephalalgia 2014; 34: 68–72.

13.

de Vries

Stam

Beker

et al.

CACNA1A mutation linking hemiplegic migraine and alternating hemiplegia of childhood. Cephalalgia 2008; 28: 887–891.

14.

de Vries

Stam

Kirkpatrick

et al.

Familial hemiplegic migraine is associated with febrile seizures in an FHM2 family with a novel de novo ATP1A2 mutation. Epilepsia 2009; 50: 2503–2504.

15.

Deprez

Weckhuysen

Peeters

et al.

Epilepsy as part of the phenotype associated with ATP1A2 mutations. Epilepsia 2008; 49: 500–508.

16.

Fernandez

Hand

Sweeney

et al.

A novel ATP1A2 gene mutation in an Irish familial hemiplegic migraine kindred. Headache 2008; 48: 101–108.

17.

Kors

Melberg

Vanmolkot

et al.

Childhood epilepsy, familial hemiplegic migraine, cerebellar ataxia, and a new CACNA1A mutation. Neurology 2004; 63: 1136–1137.

18.

Lebas

Guyant-Maréchal

Hannequin

et al.

Severe attacks of familial hemiplegic migraine, childhood epilepsy and ATP1A2 mutation. Cephalalgia 2008; 28: 774–777.

19.

Pisano

Spiller

Mei

et al.

Functional characterization of a novel C-terminal ATP1A2 mutation causing hemiplegic migraine and epilepsy. Cephalalgia 2013; 33: 1302–1310.

20.

Roth

Freilinger

Kirovski

et al.

Clinical spectrum in three families with familial hemiplegic migraine type 2 including a novel mutation in the ATP1A2 gene. Cephalalgia 2014; 34: 183–190.

21.

Spadaro

Ursu

Lehmann-Horn

et al.

A G301R Na+/K+ -ATPase mutation causes familial hemiplegic migraine type 2 with cerebellar signs. Neurogenetics 2004; 5: 177–185.

22.

Stam

Vanmolkot

KRJ

Kremer

HPH

et al.

CACNA1A R1347Q: a frequent recurrent mutation in hemiplegic migraine. Clin Genet 2008; 74: 481–485.

23.

Swoboda

Kanavakis

Xaidara

et al.

Alternating hemiplegia of childhood or familial hemiplegic migraine? A novel ATP1A2 mutation. Ann Neurol 2004; 55: 884–887.

24.

Vanmolkot

Kors

Hottenga

et al.

Novel mutations in the Na+, K+-ATPase pump gene ATP1A2 associated with familial hemiplegic migraine and benign familial infantile convulsions. Ann Neurol 2003; 54: 360–366.

25.

Vanmolkot

Kors

Turk

et al.

Two de novo mutations in the Na,K-ATPase gene ATP1A2 associated with pure familial hemiplegic migraine. Eur J Hum Genet 2006; 14: 555–560.

26.

Jurkat-Rott

Freilinger

Dreier

et al.

Variability of familial hemiplegic migraine with novel A1A2 Na+/K+-ATPase variants. Neurology 2004; 62: 1857–1861.

27.

Al-Bulushi

Al-Hashem

Tabarki

. A wide clinical phenotype spectrum in patients with ATP1A2 mutations. J Child Neurol 2014; 29: 265–268.

28.

Curtain

Smith

Ovcaric

et al.

Minor head trauma-induced sporadic hemiplegic migraine coma. Pediatr Neurol 2006; 34: 329–332.

29.

Debiais

Hommet

Bonnaud

et al.

The FHM1 mutation S218L: A severe clinical phenotype? A case report and review of the literature. Cephalalgia 2009; 29: 1337–1339.

30.

Gallanti

Tonelli

Cardin

et al.

A novel de novo nonsense mutation in ATP1A2 associated with sporadic hemiplegic migraine and epileptic seizures. J Neurol Sci 2008; 273: 123–126.

31.

Jen

Klein

Boltshauser

et al.

Prolonged hemiplegic episodes in children due to mutations in ATP1A2. J Neurol Neurosurg Psych 2007; 78: 523–526.

32.

Knierim

Leisle

Wagner

et al.

Recurrent stroke due to a novel voltage sensor mutation in Cav2.1 responds to verapamil. Stroke 2011; 42: e14–e17.

33.

Malpas

Riant

Tournier-Lasserve

et al.

Sporadic hemiplegic migraine and delayed cerebral oedema after minor head trauma: A novel de novo CACNA1A gene mutation. Dev Med Child Neurol 2010; 52: 103–104.

34.

Sánchez-Albisua

Schöning

Jurkat-Rott

et al.

Possible effect of corticoids on hemiplegic attacks in severe hemiplegic migraine. Pediatr Neurol 2013; 49: 286–288.

35.

Toldo

Cecchin

Sartori

et al.

Multimodal neuroimaging in a child with sporadic hemiplegic migraine: A contribution to understanding pathogenesis. Cephalalgia 2011; 31: 751–756.

36.

Vahedi

Denier

Ducros

et al.

CACNA1A gene de novo mutation causing hemiplegic migraine, coma, and cerebellar atrophy. Neurology 2000; 55: 1040–1042.

37.

Yamazaki

Ikeno

Abe

et al.

Hemiconvulsion-hemiplegia-epilepsy syndrome associated with CACNA1A S218L mutation. Pediatr Neurol 2011; 45: 193–196.

38.

Zangaladze

Asadi-Pooya

Ashkenazi

et al.

Sporadic hemiplegic migraine and epilepsy associated with CACNA1A gene mutation. Epilep & Behav 2010; 17: 293–295.

39.

Stam

Luijckx

Poll-Thé

et al.

Early seizures and cerebral oedema after trivial head trauma associated with the CACNA1A S218L mutation. J Neurol Neurosurg Psych 2009; 80: 1125–1129.

40.

Riant

Ducros

Ploton

et al.

De novo mutations in ATP1A2 and CACNA1A are frequent in early-onset sporadic hemiplegic migraine. Neurology 2010; 75: 967–997.

41.

Vanmolkot

Babini

de Vries

et al.

The novel p.L1649Q mutation in the SCN1A epilepsy gene is associated with familial hemiplegic migraine: Genetic and functional studies. Hum Mutat 2007; 28: 522–530–522–530.

42.

Vahedi

Depienne

Le Fort

et al.

Elicited repetitive daily blindness: A new phenotype associated with hemiplegic migraine and SCN1A mutations. Neurology 2009; 72: 1178–1183.

43.

Weller

Pelzer

de Vries

et al.

Two novel SCN1A mutations identified in families with familial hemiplegic migraine. Cephalalgia 2014; 34: 1062–1069.

44.

Fan

Wolking

Lehmann-Horn

et al.

Early-onset familial hemiplegic migraine due to a novel SCN1A mutation. Cephalalgia 2016; 36: 1238–1247.

45.

Beyl

Kügler

Hohaus

Depil

et al.

Methods for quantification of pore-voltage sensor interaction in Ca(V)1.2. Pflugers Arch 2014; 466: 265–274.

46.

Tottene

Pivotto

Fellin

et al.

Specific kinetic alterations of human CaV2.1 calcium channels produced by mutation S218L causing familial hemiplegic migraine and delayed cerebral edema and coma after minor head trauma. J Biol Chem 2005; 280: 17678–17686.

47.

van den Maagdenberg

Pizzorusso

Kaja

et al.

High cortical spreading depression susceptibility and migraine-associated symptoms in Ca(v)2.1 S218L mice. Ann Neurol 2010; 67: 85–98.

48.

Kaja

Van de Ven

Broos

et al.

Severe and progressive neurotransmitter release aberrations in familial hemiplegic migraine type 1 Cacna1a S218L knock-in mice. J Neurophysiol 2010; 104: 1445–1455.

49.

Gorji

Speckmann

. Spreading depression enhances the spontaneous epileptiform activity in human neocortical tissues. Eur J Neurosci 2004; 19: 3371–3374.

50.

Williams

Hartings

et al.

Penetrating ballistic-like brain injury in the rat: Differential time courses of hemorrhage, cell death, inflammation, and remote degeneration. J Neurotrauma 2006; 23: 1828–1846.

51.

Pietrobon

. CaV2.1 channelopathies. Eur J Physiol 2010; 460: 375–393.

52.

Lal

Reinthaler

Dejanovic

et al.

Evaluation of presumably disease causing SCN1A variants in a cohort of common epilepsy syndromes. PLoS One 2016; 11: e0150426 1–14–e0150426 1–14.

53.

Calabresi

Cupini

Centonze

et al.

Antiepileptic drugs as a possible neuroprotective strategy in brain ischemia. Ann Neurol 2003; 53: 693–702.

54.

Costa

Leone

Saulle

et al.

Coactivation of GABA(A) and GABA(B) receptor results in neuroprotection during in vitro ischemia. Stroke 2004; 35: 596–600.

55.

Costa

Martella

Picconi

et al.

Multiple mechanisms underlying the neuroprotective effects of antiepileptic drugs against in vitro ischemia. Stroke 2006; 37: 1319–1326.

56.

Mulleners

McCrory

Linde

. Antiepileptics in migraine prophylaxis: An updated Cochrane review. Cephalalgia 2015; 35: 51–62.

57.

Ayata

Jin

Kudo

et al.

Suppression of cortical spreading depression in migraine prophylaxis. Ann Neurol 2006; 59: 652–661.

58.

Bogdanov

Multon

Chauvel

et al.

Migraine preventive drugs differentially affect cortical spreading depression in rat. Neurobiol Dis 2011; 41: 430–435.

59.

Calabresi

Galletti

Rossi

et al.

Antiepileptic drugs in migraine: From clinical aspects to cellular mechanisms. Trends Pharmacol Sci 2007; 28: 188–195.

60.

Martella

Costa

Pisani

et al.

Antiepileptic drugs on calcium currents recorded from cortical and PAG neurons: Therapeutic implications for migraine. Cephalalgia 2008; 28: 1315–1326.

61.

Galletti

Cupini

Corbelli

et al.

Pathophysiological basis of migraine prophylaxis. Prog Neurobiol 2009; 89: 176–192.

62.

Striano

Zara

Santorelli

et al.

Topiramate-associated worsening symptoms in a patient with familial hemiplegic migraine. J Neurol Sci 2008; 272: 194–195.

63.

Lampl

Katsarava

Diener

et al.

Lamotrigine reduces migraine aura and migraine attacks in patients with migraine with aura. J Neurol Neurosurg Psych 2005; 76: 1730–1732.

64.

Linde

Mulleners

Chronicle

et al.

Antiepileptics other than gabapentin, pregabalin, topiramate, and valproate for the prophylaxis of episodic migraine in adults. Cochrane Dat Syst Rev 2013; 6: CD010608 1–17–CD010608 1–17.