Conservation of Ca2+/calmodulin regulation across Na and Ca2+ channels

Abstract

Voltage-gated Na and Ca2+ channels comprise distinct ion channel superfamilies, yet the carboxy tails of these channels exhibit high homology, hinting at a long-shared and purposeful module. For different Ca2+ channels, carboxyl-tail interactions with calmodulin do elaborate robust and similar forms of Ca2+ regulation. However, Na channels have only shown subtler Ca2+ modulation that differs among reports, challenging attempts at unified understanding. Here, by rapid Ca2+ photorelease onto Na channels, we reset this view of Na channel regulation. For cardiac-muscle channels (NaV1.5), reported effects from which most mechanistic proposals derive, we observe no Ca2+ modulation. Conversely, for skeletal-muscle channels (NaV1.4), we uncover fast Ca2+ regulation eerily similar to that of Ca2+ channels. Channelopathic myotonia mutations halve NaV1.4 Ca2+ regulation, and transplanting the NaV1.4 carboxy tail onto Ca2+ channels recapitulates Ca2+ regulation. Thus, we argue for the persistence and physiological relevance of an ancient Ca2+ regulatory module across Na and Ca2+ channels.

Figure 1. Homology but divergent function for Ca²⁺versus Na channels

(A) CI region of Ca²⁺ (Ca_V1.3) and Na channels (Na_V1.5, Na_V1.4). Dual vestigial EF-hands shaded in rose and green. IQ domain, blue. (B) Ca²⁺ channel regulation inducible by channel Ca²⁺ influx. Ba²⁺ influx as negative control. (C) Left, Ca_V1.3 current traces carried by Ca²⁺(red) or Ba²⁺(black). Vertical bar for Ca²⁺ trace. Ba²⁺ trace scaled ~3× downward for kinetic comparison. Right, r₃₀₀ (fraction of peak current remaining after 300-ms depolarization) versus V_test potential, plotted as mean ± SEM (8 cells). (D) Na channels characterized under pipet dialysis with 0 or 10 µM Ca²⁺. (E) Schematic of reported Ca²⁺ effects on inactivation. Left, h_∞, fractional current remaining after prepulses (V_hold). Right, purported Ca²⁺-induced voltage shifts of h_∞. (F) Na_V1.5 currents under protocol in panel E (black, 0 Ca²⁺; red, 10 µMCa²⁺buffered with HEDTA). See Figure S1. (G, H) Normalized form of h_∞ unaffected by Ca²⁺. Potential Ca²⁺-induced reduction in Na_V1.4 h_∞ (rose dashed line). Error bars, SEM throughout. Fit function: h_∞ = 1 / (1 + exp ((V_hold – V_1/2) / SF), where SF = 6.2 (Na_V1.4) and 7.5 (Na_V1.5).

Figure 2. CDI of Na channels under Ca²⁺photouncaging

(A) Na_V1.5 currents unaffected by 10 µM Ca²⁺. Gray dots, peak currents before uncaging. Bottom, mean data for CDI versus Ca²⁺-step amplitude.CDI = 1 – average peak I_Na of last 3–4 responses after Ca²⁺ uncaging / peak I_Na before uncaging. Symbols, mean ± SEM of ~5 uncaging events compiled from 23 cells. See Figure S2. (B) Na_V1.4 peak currents decline during 2 µM Ca²⁺ step (rose fit). Format as in panel A. Bottom, mean CDI plotted versus Ca²⁺. Each symbol, mean ± SEM of ~5 uncaging events compiled from 35 cells. See Figures S2–S4. (C) Na_V1.4 currents specifying h_∞ at ~100 nM Ca²⁺. Bottom, h_∞ curve (mean ± SEM, 5 cells). (D)~3 µM Ca²⁺ step uniformly suppresses Na currents. Bottom, corresponding mean h_∞ curve (red symbols and fit), where symbols plot mean ± SEM (5 cells). Upwardly scaled h_∞ curve (rose) same as before uncaging (black).

Figure 3. Na channel regulated by Ca²⁺spill over from Ca²⁺ channels

(A) Schematic, Ca²⁺ spill over from Ca_V2.1 inhibiting Na channels. (B) Dual voltage-pulses selectively evoke identical Na_V1.4 currents. See Figure S5. (C) Intervening +30 mV pulse (V_inter) activates Ca_V2.1, diminishing ensuing Na current. r_Na, fraction of Na current remaining after Ca_V2.1 Ca²⁺ influx. (D) V_inter to +90 mV rescues the second Na_V1.4 current. (E, F) Na current amplitude unperturbed by Ba²⁺ influx through Ca_V2.1 channels. (G) Mean relation for r_Na versus V_inter shows U-shape with Ca²⁺ (red) but not Ba²⁺ (black). Symbols, mean ± SEM (6 cells). (H–K) Restricting Ca²⁺ to Ca_V2.1nanodomain prevents Na channel CDI. Panel K, symbols, mean ± SEM (5 cells), format as in panel G. See Figure S5.

Figure 4. Multi-channel stochastic records of Na_V1.4 CDI

(A) Multi-channel records from HEK293 cells coexpressing Na_V1.4 and Ca_V2.1 channels. On-cell patch configuration. Voltage protocol (top), multi-channel record (middle), and ensemble average current (bottom). Red shading, Ca²⁺ entry. Ensemble average shows reduced Na channel activity after Ca²⁺ entry (pulse ii) versus before (pulse i). (B) Amplitude histogram analysis of patch from panel A shows no change in unitary current following Ca²⁺ entry (top before interpulse; bottom after interpulse). Amplitude histogram analysis of events occuring 0.5–17 m sec after pulse onset during −30 mV test pulses. Fits (black) to data (gray) derived from amplitude analysis of low-pass filtered stochastic channel simulations with added gaussian noise. Dashed-red lines, unitary current i used to generate fits. (C) Expanded time base display of ensemble average currents from panel A, before (upper) and after (below) Ca²⁺. Fast inactivation essentially identical; same time constant for both mono exponential fits (black curves). (D) Multi-channel stochastic records of separate patch with only Na_V1.4 channels. No difference in channel activity before and after intervening pulse (mean decrement ~ 0.2 ± 3% (mean ± SEM, n = 5 patches). Second multi-channel record chosen to illustrate rare occurrence of persistent gating mode. Format as in panel A.

Figure 5. Calmodulin as Ca²⁺ sensor for Na_V1.4CDI

(A) Mutating putative Ca²⁺-coordinating residues in Na_V1.4 EF-hand did not alter CDI. Format as in Figure 2A. Symbols, mean ± SEM of ~3 uncaging events from 12 cells. (B) CaM₁₂₃₄ abolishes CDI. Symbols, mean ± SEM (~6 uncaging events from 27 cells). (C) Eliminating N-lobe Ca²⁺-binding (CaM₁₂) abolishes CDI. Symbols, mean ± SEM of 4–5 uncaging events from 12 cells. (D) Eliminating C-lobe Ca²⁺ binding (CaM₃₄) spares CDI. Symbols, mean ± SEM of ~5 uncaging events from 12 cells. (E) Mutating Na_V1.4 III–IV loop spares CDI. Format as in Figure 2A. Symbols, mean ± SEM of 4–5 uncaging events from 18 cells. (F) No CDI in Na_V1.4–1.5CT chimera. Symbols, mean ± SEM of 4–5 uncaging events (13 cells). (G) Substituting dual alanines for key isoleucine-glutamine residues in Na_V1.4 IQ domain yields facilitating Na currents. Bottom, mean data confirms facilitation, shown as negative CDI. Symbols, mean ± SEM of ~13 uncaging events (20 cells).

Figure 6. Physiology of Na channel Ca²⁺ regulation

(A) No Ca²⁺-regulation of native Na_V1.5 in ventricular myocytes. Format as in Figure 2A. Symbols, mean ± SEM from 5–6 uncaging events (13 cells). (B) Endogenous Na_V1.4 channels in GLT cells exhibit CDI. Minimal contamination by Ca²⁺-activated Cl current (<5% of I_Na) subtracted. Each symbol, mean ± SEM from 4–5 uncaging events (12 cells). (C, D) Recombinant Na_V1.4 channels with mutations for K- and cold-aggravated myotonias show weaker CDI. Symbols, mean ± SEM from ~9 uncaging events (indicated number of cells).

Figure 7. Persistence of CaM/CI module across Na and Ca²⁺channel superfamilies

(A) Transferring Na_V1.4 carboxy tail to Na_V1.5 backbone (Na_V1.5–1.4CT) confers latent Ca²⁺regulation (wild-type Na_V1.4, gray fit in bottom subpanel). Format as in Figure 2A. Symbols, mean ± SEM from 4–5 uncaging events (10 cells). (B) Phylogenetic tree of the Na and Ca²⁺ channel superfamilies. (C) Transplanting Na_V1.4 carboxy tail onto Ca_V1.3 backbone (Ca_V1.3–Na_V1.4CT) yields chimeric channel that retains Ca²⁺ regulation. Format as in Figure 1C. Symbols, mean ± SEM, 7 cells. CDI measured under low Ca²⁺ buffering (see Extended Experimental Procedures). (D) Coexpressimg CaM₁₂₃₄with Ca_V1.3–Na_V1.4 CT abolishes CDI. Format as in Figure 1C. Symbols, mean ± SEM, 7 cells. CDI measured as in panel C. (E) Maximum likelihood phylogenetic tree shows conservation among Ca²⁺ and Na channel CI regions, across major eukaryotic phyla. Format as in Figure 1A. Consensus sequence patterns for motifs on top. Sequence alignment starts at the center with the Paramecium Ca²⁺ channel. Ca²⁺ channels from progressively more advanced organisms branch to the top (pale colors), and those for Na channels branch to the bottom (darker colors).