WHOLE EXOME SEQUENCING OF HEMIPLEGIC MIGRAINE PATIENTS SHOWS AN INCREASED BURDEN OF MISSENSE VARIANTS IN CACNA1H AND CACNA1I GENES 201 9 In the general population, females are overrepresented in most forms of migraine including hemiplegic migraine. The overall female:male sex ratio in our HM cohort was ~1.97:1. This observed difference in prevalence will in part be explained by the fact that females are more inclined to consult a physician and thus are diagnosed earlier and more often than males.40 We cannot rule out that there is any sexual dimorphic effect at any of the CACNA1x genes (i.e. a sex bias in gene function), but we consider this a minor factor compared to the ascertainment bias. We have hypothesised that HM may not be autosomal dominant in a substantial proportion of cases, but rather is genetically a more complex trait. The difficulty in confirming this hypothesis lies in how to identify such variants, as they are neither identified by gene association approaches, nor in genome-wide association studies (GWAS). In order to identify such variants, we have used a methodology adapted from a TRAPD analysis. The methodology has proven itself by the identification of functional genetic variants in idiopathic hypogonadotropic hypogonadism.39 By slightly adapting the method, we have been able to investigate all missense variance and thereby determine the variant and subject burden. Similarly, our results show that the accumulation of missense variants in CACNA1H and CACNA1I plays a role in HM. CACNA1H and CACNA1I encode the α1 subunits of CaV3.2 and CaV3.3 LVA T-type calcium channels, respectively (Figure 1).22, 23 CACNA1H is expressed ubiquitously, whereas CACNA1I is predominantly expressed in the brain.25 Ca V3.2 and CaV3.3 LVA T-type calcium channels, 41 which open by only a small membrane depolarization, coupled with their tonic inactivation near resting membrane potential, underlie the spike/rebound bursting phenomenon seen with many types of neurons 42, 43. These channels are localized at presynaptic nerve terminals44 where they control synaptic transmission by directly triggering the release of neurotransmitters.45-47 Inactivation of Cacna1h in mice led to decreased nociceptive signalling48, 49 and several neurological symptoms,50, 51 whereas Cacna1i knock-out mice, and also Cacna1i/Cacna1h double knockout mutants, show implications for sleep rhythmogenesis.52 T-type channels are important for human physiology, so mutations in these channels may lead, at least in theory, to channelopathies with clinical manifestations resulting from aberrant biophysical characteristics and/or cell surface trafficking issues of channels due to a gain or loss of channel function. Indeed, specific missense variants in CACNA1H have been implicated in a range of human conditions,50 including autism spectrum disorders53 and amyotrophic lateral sclerosis.54 Many missense variants in the human CACNA1H gene have been reported in patients presenting with a range of epilepsy syndromes,50 so the gene was labelled a risk gene for idiopathic generalised epilepsies.55 Functional analyses in embryonic kidney cells, however, revealed that the variants in CACNA1H generally produce mild biophysical changes and in some cases do not alter the gating of the channel and variants do not segregate with the phenotype.50 Hence their contribution to human epilepsies should be debated, as was recently suggested.28 In line with this suggestion, it is not unexpected that CACNA1H variants identified in HM patients also not solely cause disease, although a burden of variants in this gene can still contribute to HM risk. Similarly, CACNA1I loss-of-function variants were identified that alter the
RkJQdWJsaXNoZXIy MTk4NDMw