Investigation of CACNA1I Cav3.3 Dysfunction in Hemiplegic Migraine

Abstract

Familial hemiplegic migraine (FHM) is a severe neurogenetic disorder for which three causal genes, CACNA1A, SCN1A, and ATP1A2, have been implicated. However, more than 80% of referred diagnostic cases of hemiplegic migraine (HM) are negative for exonic mutations in these known FHM genes, suggesting the involvement of other genes. Using whole-exome sequencing data from 187 mutation-negative HM cases, we identified rare variants in the CACNA1I gene encoding the T-type calcium channel Cav3.3. Burden testing of CACNA1I variants showed a statistically significant increase in allelic burden in the HM case group compared to gnomAD (OR = 2.30, P = 0.00005) and the UK Biobank (OR = 2.32, P = 0.0004) databases. Dysfunction in T-type calcium channels, including Cav3.3, has been implicated in a range of neurological conditions, suggesting a potential role in HM. Using patch-clamp electrophysiology, we compared the biophysical properties of five Cav3.3 variants (p.R111G, p.M128L, p.D302G, p.R307H, and p.Q1158H) to wild-type (WT) channels expressed in HEK293T cells. We observed numerous functional alterations across the channels with Cav3.3-Q1158H showing the greatest differences compared to WT channels, including reduced current density, right-shifted voltage dependence of activation and inactivation, and slower current kinetics. Interestingly, we also found significant differences in the conductance properties exhibited by the Cav3.3-R307H and -Q1158H variants compared to WT channels under conditions of acidosis and alkalosis. In light of these data, we suggest that rare variants in CACNA1I may contribute to HM etiology.

Keywords: CACNA1I; Cav3.3; T-type calcium channels; familial hemiplegic migraine; hemiplegic migraine; ion channel; migraine genetics; voltage gated calcium channels.

FIGURE 1

Location and functional expression of Cav3.3 HM-associated variants. (A) Schematic representation of the Cav3.3 channel protein and location of the HM-associated variants functionally characterized. Modular architecture of Cav3.3 showing the four homologous domains (DI–DIV). Each domain is composed of a voltage sensor module (transmembrane segments S1–S4) and a pore module (S5, a re-entrant pore loop and S6). The locations of the amino acids changed by HM-associated rare variants are marked with circles. Color scheme maintained hereafter: R111G (), M128L (), D302G (), R307H (), and Q1158H (). (B) Top: representative whole-cell currents from Cav3.3 WT and HM-associated variants R111G (fuchsia), M128L (orange), D302G (blue), R307H (green), and Q1158H (purple) expressed in HEK293T cells. Voltage-gated Ca²⁺ currents were elicited by a 100-ms pulse to −10 mV (V_h: −90 mV, inset). Scale bars: 1 nA. Bottom: summary of current density calculated from peak current amplitude (I_peak/cell capacitance in pA/pF). In this and subsequent SuperPlots (Lord et al., 2020), individual values and mean ± SEM (calculated from all data points in each group) are shown. Biological variabilities from independent transfections and recording time points are conveyed by different symbols. Within biological replicates, empty symbols correspond to determinations made 24 h post-transfection (□: mean 24 h), and solid symbols correspond to 48 h post-transfection recordings (■: mean 48 h). The statistical significance was determined through a paired Student’s t-test against Cav3.3 WT. p ≤ 0.001 (***), 0.001 < p ≤ 0.01 (**), or 0.01 < p ≤ 0.05 (*). (C) Confocal micrograph from immunostained Cav3.3 WT and variants (indicated by the corner labels) showing distinct expression patterns. Paired images of Alexa 488 fluorescence of center stack (top) and the depth-encoded full stack sets are shown for each construct. The color scale (bottom right) indicates relative distance from the coverslip. DAPI nuclear stain is not shown here for clarity, see Supplementary Figure 1A. Scale bar 10 microns.

FIGURE 2

Voltage dependence of activation of Cav3.3 WT and HM-associated variants. (A) Normalized conductance (G/G_max) vs. voltage relationships for Cav3.3 and HM variants. Representative current families are included as insets. Dotted lines are drawn at G/Gmax = 0.5 to highlight the voltage of half-maximal activation (activation V_0.5). (B) SuperPlots of activation V_0.5 and slope [d(x)]. Symbols represent determination made from cells transfected on the same day (i.e., biological replicates). 0.001 < p < 0.01 (**), 0.01 < p 0.05 (*).

FIGURE 3

Steady-state inactivation (SSI) of HM-associated Cav3.3 variants. (A) Availability plots were constructed from normalized peak currents (I/I_max) vs. the inactivating pre-pulse potential. Representative SSI current traces for each channel variant are shown in the insets. Gray dotted lines indicate the potential at which 50% of the channels are inactivated (inact V_0.5). (B) SuperPlots summarizing SSI V_0.5 and slope [d(x)] for all variants. 0.001 < p < 0.01 (**), 0.01 < p 0.05 (*).

FIGURE 4

Cav3.3-WT and HM-associated variant window currents. The window current (I_W) was determined from the area under the overlapping normalized activation and inactivation curves (AUC) of WT and HM-associated Cav3.3 variants (average traces from Figures 2, 3). (A) Total AUCs were divided by Cav3.3-WT AUC (I_WVar/I_WWT) for comparison. In (B,C) window currents are highlighted by the shading color corresponding to each HM-associated variant and plotted along Cav3.3 WT activation and SSI curves (in black).

FIGURE 5

Cav3.3-WT and HM-associated variant current kinetics. (A) The speed of Ca²⁺ current activation (τ_act) and inactivation (τ_inact) during 100-ms stimuli (inset) was evaluated by fitting (red) the elicited currents with an exponential product equation (bottom), where (C) is the y-intercept. Scale bars: 10 ms, 500 pA. (B) Representative currents from Cav3.3 variants studied. (C) Summary of time constants of current activation (τ_act). (D) Summary of time constants of current inactivation (τ_inact). p < 0.001 (***), 0.001 < p < 0.01 (**).

FIGURE 6

pH_o modulation of Cav3.3- WT-, - R307H-, and -Q1158H-mediated currents. (A) Representative Ca²⁺ current traces in response to 200-ms depolarizing pulses to −20 mV (V_h = −90 mV, 0.1 Hz, inset) recorded in control (black, pH 7.4), acidic (red, pH 6.5), or alkaline (blue, pH 8.0) pH_o conditions. Scale bars: 10 ms, 200 pA. Dashed line indicates zero level. (B) Relative change in peak current amplitude (I pH X/I pH7.4) for WT and HM-associated Cav3.3 variants (pH 7.4: ○; pH 6.5: ; pH 8.0: ; difference between pH 6.5: , or pH 8.0: ). (C) Relative changes in inactivation time constant (τ_inact pH X/τ_inact pH7.4) for WT and HM-associated Cav3.3 variants (pH 7.4: ○;, △; pH 6.5: ; pH 8.0: ; difference between pH 6.5: , or pH 8.0: ). In (B,C) paired t-test for pH 7.4 vs. pH X: *p ≤ 0.05; **p ≤ 0.05; ***p < 0.005; ****p < 0.0005. One-way ANOVA with Tukey’s multiple comparisons test for WT vs. variant: ^†p ≤ 0.05; ^††p ≤ 0.05; ^†††p ≤ 0.005; ^††††p ≤ 0.0001.