Complex effects on CaV2.1 channel gating caused by a CACNA1A variant associated with a severe neurodevelopmental disorder

Abstract

P/Q-type Ca²⁺ currents mediated by Ca_V2.1 channels are essential for active neurotransmitter release at neuromuscular junctions and many central synapses. Mutations in CACNA1A, the gene encoding the principal Ca_V2.1 α_1A subunit, cause a broad spectrum of neurological disorders. Typically, gain-of-function (GOF) mutations are associated with migraine and epilepsy while loss-of-function (LOF) mutations are causative for episodic and congenital ataxias. However, a cluster of severe Ca_V2.1 channelopathies have overlapping presentations which suggests that channel dysfunction in these disorders cannot always be defined bimodally as GOF or LOF. In particular, the R1667P mutation causes focal seizures, generalized hypotonia, dysarthria, congenital ataxia and, in one case, cerebral edema leading ultimately to death. Here, we demonstrate that the R1667P mutation causes both channel GOF (hyperpolarizing voltage-dependence of activation, slowed deactivation) and LOF (slowed activation kinetics) when expressed heterologously in tsA-201 cells. We also observed a substantial reduction in Ca²⁺ current density in this heterologous system. These changes in channel gating and availability/expression manifested in diminished Ca²⁺ flux during action potential-like stimuli. However, the integrated Ca²⁺ fluxes were no different when normalized to tail current amplitude measured upon repolarization from the reversal potential. In summary, our findings indicate a complex functional effect of R1667P and support the idea that pathological missense mutations in Ca_V2.1 may not represent exclusively GOF or LOF.

Figure 1

The R1667P mutation occurs at the R4 position of the Repeat IV S4 voltage-sensing α-helix. (a) Schematic representation of Ca_V2.1 with Green Fluorescent Protein (GFP) fused to the amino-terminus (GFP-Ca_V2.1). The R to P substitution at residue 1667 is indicated by the red star. (b) Sequence comparison of the Repeat IV S4 helices of all known human Ca_V2.1 variants (cf. accession no. NP_001120693.1), human Ca_V2.1 with the R to P substitution at the R4 position and human Ca_V2.2 (accession no. NM_001243812). Basic residues in positions R2-R6, as defined by Ca_V2.2 Cryo-EM structure are shown in green and the R4 R to P substitution is shown in red. (c, d) AlphaFold2 modeling of the voltage-sensing module of Repeat IV of human Ca_V2.1. The S1–S4 helices are viewed from the lateral aspect (c) and from an extracellular (d) vantage points. Potential hydrogen bonds between R1667 and N1579 in S1, T1606 in S2, and S1641 in S3 are indicated by the blue dashed lines. F1609 (i.e., the gating charge transfer center) is also labelled. Panels (c) and (d) were published with permission https://creativecommons.org/licenses/by/4.0/ (e) Missense 3D model showing the impact of the arginine to proline substitution at position 1667. The mutant stick structure (blue) is overlaid on the wild-type stick structure (white) with R1667 shown in green and the R1667P substitution shown in red. In both cases, F1609 is shown; this residue is colored green and red in the in the wild-type and mutant structures, respectively.

Figure 2

The R1667P mutation causes a profound reduction in Ca²⁺ current density and a hyperpolarizing shift in Ca_V2.1 activation. (a) Ca²⁺ current families recorded from tsA-201 cells expressing GFP-Ca_V2.1 (left) or GFP-Ca_V2.1 R1667P (right) with auxiliary β₄ and α² δ-1 subunits. Currents were elicited by a 25 ms step depolarizations from − 80 mV to indicated test potentials; the repolarization voltage was − 40 mV. Confocal images confirming successful heterologous expression of GFP-Ca_V2.1 and GFP-Ca_V2.1 R1667P are shown in the insets. Scale bars = 10 µm. (b) Comparison of GFP-Ca_V2.1 (filled circle; n = 17) and GFP-Ca_V2.1 R1667P (open circle; n = 19) average peak I–V relationships. Currents were evoked at 0.1 Hz by test potentials ranging from − 50 mV through + 80 mV in 10 mV increments. Amplitudes were normalized by capacitance (pA/pF). (c) Normalized G–V curves were fit by Eq. (1) with the following respective parameters for GFP-Ca_V2.1 and GFP-Ca_V2.1 R1667P: V_G = 0.9 ± 0.7 and − 10.0 ± 0.9 mV; k = 5.4 ± 0.2 and 6.6 ± 0.8 mV, respectively. Throughout, data are presented as mean ± SEM; the total number of cells in a data set is indicated in parentheses.

Figure 3

The R1667P mutation slows activation and deactivation. Ca²⁺ currents were recorded from tsA-201 cells expressing either GFP-Ca_V2.1 (a) or GFP-Ca_V2.1 R1667P (b). Currents were elicited by 25 ms step depolarizations from − 80 mV to test potentials ranging from − 20 mV through + 30 mV (a-top). Activation was fit by Eq. (2); the time constants of activation (τ_act) for representative cells are indicated. (c) Comparison of τ_act for GFP-Ca_V2.1 (filled circle; n = 17) or GFP-Ca_V2.1 R1667P (open circle; n = 15) measured at the indicated test potentials. Representative tail currents were recorded from tsA-201 cells expressing either GFP-Ca_V2.1 (d) or GFP-Ca_V2.1 R1667P (e) upon repolarization from + 30 mV to the indicated potentials (d-top). Deactivation was fit by Eq. (2). (f) Comparison of τ_deact for GFP-Ca_V2.1 (filled circle; n = 13) and GFP-Ca_V2.1 R1667P (open circle; n = 9) measured at the indicated repolarization potentials. Significant differences by two-tailed, unpaired t-test are indicated (*Denotes P < 0.05; ***Denotes P < 0.001).

Figure 4

The R1667P mutation has little effect on closed-state inactivation. (a) A 5 s conditioning step from the steady holding potential (− 80 mV) to increasing potentials ranging from − 100 to + 30 mV (in 10 mV increments) was applied before repolarizing the membrane to − 80 mV for 5 ms. Test currents were then evoked by a 25 ms step depolarization to + 20 mV. The protocol is not drawn to scale. Representative Ca²⁺ currents recorded from tsA-201 cells expressing GFP-Ca_V2.1 (b) or GFP-Ca_V2.1 R1667P (c) after pre-pulses to − 100, − 80, − 60, − 40, − 20, 0 and + 20 mV. Normalized steady-state inactivation curves for GFP-Ca_V2.1 (filled circle; n = 13) and GFP-Ca_V2.1 R1667P (open circle; n = 6) are shown in (d); amplitudes were normalized by the maximal Ca²⁺ current in each cell. The normalized inactivation relationships were fit with Eq. (3) with the following fit parameters for GFP-Ca_V2.1 and GFP-Ca_V2.1 R1667P: V_1/2inact = − 39.3 ± 15 and − 42.1 ± 1.8 mV; k = − 12.6 ± 0.6 and − 10.7 ± 0.7 mV, respectively. (e) Overlay of the smooth conductance and closed-state inactivation curves (from Figs. 2c and 4d, respectively) for cells expressing GFP-Ca_V2.1 (black lines) and GFP-Ca_V2.1 R1667P (red lines).

Figure 5

The R1667P mutation reduces total Ca²⁺ flux in response to a single action potential-like stimulus. Ca²⁺ currents attributable to GFP-Ca_V2.1 (a) and GFP-Ca_V2.1 R1667P (b) were evoked by an action potential-like waveform consisting of a 1 ms ramp from − 80 mV to + 30 mV followed immediately by a 1 ms ramp back to − 80 mV. Traces shown are the average of 10 recordings. (c) Comparison of total charge flux normalized to cell membrane capacitance (fC/pF) for cells expressing GFP-Ca_V2.1 (filled circle; n = 13) or GFP-Ca_V2.1 R1667P (open circle; n = 12). (d) Absolute charge flux normalized to tail current amplitude at the reversal potential (i.e., maximal conductance) (fC/pA). Means are indicated by the dashed lines of the boxes. Boxes represent the 25th/75th percentiles. Bars represent the 5th/95th percentiles. A significant difference is indicated (***) in (c).