Ca2+-dependent regulation of sodium channels NaV1.4 and NaV1.5 is controlled by the post-IQ motif

Abstract

Skeletal muscle voltage-gated Na⁺ channel (Na_V1.4) activity is subject to calmodulin (CaM) mediated Ca²⁺-dependent inactivation; no such inactivation is observed in the cardiac Na⁺ channel (Na_V1.5). Taken together, the crystal structures of the Na_V1.4 C-terminal domain relevant complexes and thermodynamic binding data presented here provide a rationale for this isoform difference. A Ca²⁺-dependent CaM N-lobe binding site previously identified in Na_V1.5 is not present in Na_V1.4 allowing the N-lobe to signal other regions of the Na_V1.4 channel. Consistent with this mechanism, removing this binding site in Na_V1.5 unveils robust Ca²⁺-dependent inactivation in the previously insensitive isoform. These findings suggest that Ca²⁺-dependent inactivation is effected by CaM's N-lobe binding outside the Na_V C-terminal while CaM's C-lobe remains bound to the Na_V C-terminal. As the N-lobe binding motif of Na_V1.5 is a mutational hotspot for inherited arrhythmias, the contributions of mutation-induced changes in CDI to arrhythmia generation is an intriguing possibility.

Fig. 1

C-terminal tail and CaM control of Na_V CDI. a Schematic of Na_V1.5, Na_V1.4 and Na_V1.5-CTail1.4 (Na_V1.5 residues 1–1773 with Na_V1.4 1599–1836) used in experiments below. b Pulse protocol for Na⁺ channel current recordings and assessment of CDI. The pulse protocol with a 150 msec step to +20 mV (red) activates co-expressed Ca_V2.1 while a step to −40 mV (black) does not. c Current elicited by the pulse protocols. The Na⁺ current is probed before (P1) and after (P2) Ca²⁺ entry into the cells due to the intermediate depolarization at −40 mV (where the Ca²⁺ channels are closed) and +20 mV (where the Ca²⁺ channels are open). The pre-pulse P1 and the test pulse P2 are used to probe the Na⁺ current in the absence of Ca²⁺ due to the intermediate depolarization at −40 mV or in the presence of Ca²⁺ due to the activated Ca²⁺ channels during the further depolarization at +20 mV. d Na⁺ currents measured during P1 and P2. The P2 current after Ca²⁺ influx (red trace) compared to P2 current with no Ca²⁺ influx (black). e Data points of CDI measurement (open circles, filled circle as mean with bars showing ±1 standard deviation) with CaM WT, CaM₁₂ or CaM₃₄ showing only the N-lobe of CaM is required to bind Ca²⁺ for CDI to occur. Statistical significance was determined by an unpaired t-test. Supplementary Table 2 lists values of the individual data points shown. f Alignment of the sequences of Na_V1.4 and Na_V1.5 in IQ and post-IQ regions

Fig. 2

Structures of Na_V1.4 CTerm in complex with CaM, ± Ca²⁺. a 1.8 Å resolution structure of the complex of the Na_V1.4 CTerm and apo CaM. CTerm colored gray and CaM light teal (N-lobe) and light orange (C-lobe). b Close-up of the apo C-lobe and helix α VI interactions. c Close-up of N-lobe and EFL interaction. d 3.3 Å resolution structure of the complex of the Na_V1.4 CTerm and (Ca²⁺)₄-CaM. CTerm is colored black and CaM dark teal (N-lobe) and dark orange (C-lobe). e Close-up of Ca²⁺-C-lobe and helix αVI interactions. f Close-up of Ca²⁺-N-lobe and EFL interaction. Residues shown have more than 10 Ų BSA. g–i Schematics of CaM N-lobe control in Na_V regulation. g Activated conformation of CaM and Na_V1.4 (PDB ID: 6MBA). h Ca²⁺-inactivated conformation of CaM and Na_V1.4 (PDB ID: 6MC9). i Ca²⁺-insensitive conformation of CaM and Na_V1.5 (PDB ID: 4JQ0, FHF molecule not displayed)

Fig. 3

Structural overlap of the N-lobe and C-lobe of CaM bound to Na_V1.4 CTerm. a CaM N-lobes aligned by helices B and C (residues 32–55) show the N-lobe displacement from closed to open upon Ca²⁺ binding (70° of opening). b CaM C-lobes aligned by helices F and G (residues 105–128) show the small conformational change (8° of opening) experienced by the C-lobe upon Ca²⁺ binding; both C-lobes are semi-open (Supplementary Fig. 2). c Close-up of the C-lobe of (Ca²⁺)₄-CaM bound to helix α VI, showing the C-lobe’s EF-hand loop residues (CaM residues 93–104 and 129–140) and the anomalous scattering (F₊ − F₋)e^iφ-calc map at a 4σ contour within 10 Å of either Ca²⁺ ion. d, Close-up of the N-lobe of (Ca²⁺)₄-CaM showing the N-lobe’s EF-hand loop residues (CaM residues 20–31 and 56–67) and the anomalous scattering (F₊ − F₋)^eiφ-calc map at 3.5σ contour within 10 Å of either Ca²⁺ ion

Fig. 4

Na_V1.4 CTerm and Na_V1.5 CTerm populations with bound CaM. a Reaction scheme of Ca²⁺ ions binding to CaM lobes. Thermodynamic data of these binding reactions were reported previously^, . b Reaction scheme of CaM in four Ca²⁺-saturation states binding to Na_V CTerm. Thermodynamic data of these binding reactions are reported here. c–f Panels showing the relative population (Z-axis) of four CaM species bound to Na_V CTerm, modeled as a function of [Ca²⁺] and [CaM] using the two schemes above. Na_V1.4 CTerm-bound species on the left, Na_V1.5 CTerm-bound CaM-species on the right. c Population of apo CaM bound to Na_V CTerm. d Population of (Ca²⁺)_2-N–CaM bound to Na_V CTerm. e Population of (Ca²⁺)_2-C–CaM bound to Na_V CTerm. f Population of (Ca²⁺)₄–CaM bound to Na_V CTerm. Only Na_V1.4 CTerm shows a significant population of bound (Ca²⁺)_2-C–CaM (orange surface)

Fig. 5

Na_V CTerm-CaM populations as a function of Ca²⁺. a The relative populations of four CaM-bound CTerm species are shown at a fixed CaM concentration of 10 µM (fixed CaM at 1 µM shows similar trends, Supplementary Fig. 8). The isoform CTerm species add to 100%; free CTerm is included in calculations but not shown in figure. At ~10 µM Ca²⁺ Na_V1.4 CTerm shows a dominant species of (Ca²⁺)_2-C–CaM bound to CTerm (solid orange line) while Na_V1.5 CTerm shows a dominant species of (Ca²⁺)₄–CaM bound to CTerm (dashed purple line). b The relative populations of CTerm-bound CaM, showing Ca²⁺-saturation of the CaM N- or C-lobe. At ~10 M Ca²⁺ CaM bound to Na_V1.4 CTerm has little (Ca²⁺)₂-N-lobe (solid teal) while CaM bound to Na_V1.5 CTerm has primarily (Ca²⁺)₂-N-lobe (dashed teal line). This is consistent with the hypothesis that Na_V1.5 CTerm contains an NLBM while Na_V.14 CTerm does not

Fig. 6

Deletion of Na_V1.5 post-IQ NLBM reveals CDI. a Schematic of the sodium channel Na_V1.5 displaying CTerm and post-IQ NLBM region; currents unaffected by 10 µM Ca²⁺. Gray dots, peak currents before uncaging. Bottom, mean ± ecm for CDI versus Ca²⁺-step amplitude. CDI = 1−average peak I_Na of last three to four responses after Ca²⁺ uncaging/peak I_Na before uncaging. b Schematic of Na_V1.5 displaying CTerm and the post-IQ deletion; Na⁺ currents reduced strongly at 10 µM Ca²⁺, the envelope of peak currents is outlined in red and the difference in peak current before and after uncaging is shaded. Current measurements of Na_V1.5ΔP-IQ, shown as in (a)