The distinct role of the four voltage sensors of the skeletal CaV1.1 channel in voltage-dependent activation

Figures

• 

  Figure 1.. Time course of the activation of CaV1.1 and RYR1 channels. (A) Membrane current from CaV1.1 channels expressed in Xenopus oocytes displaying gating (gray) and ionic currents (red). (A and B) The rearrangement of CaV1.1 voltage sensors upon depolarization (gating currents, gray) activates two processes with very distinct kinetics: CaV1.1 ionic current (slow; red in A) and SR Ca2+ release (fast; in B) from dissociated mouse muscle fibers (flexor digitorum brevis) loaded with the Ca2+ dye OGB-5N. Recordings are shown on the same time scale to illustrate their drastically different kinetics.

• 

  Figure S1.. Topology of CaV1.1 channels, S4 segments sequence alignment, and voltage dependence of CaV1.1 channel opening for WT and Cys mutants. (A) The topology of the pore-forming α1S subunit of skeletal CaV1.1 channels is shown: four concatenated homologous but nonidentical repeats (I–IV) form the α1S subunit. Each repeat includes six transmembrane segments (S1–S6): S1–S4 make the VSD (colored cylinders), while S5 and S6 contribute to one quarter of the pore (gray cylinders). (B) A sequence alignment of the four S4 segments is based on the most recent structure of CaV1.1 channels (bold font corresponds to residues included in S4 helices; Protein Data Bank accession no. 6JP5; Zhao et al., 2019). Part of S3 and S4 loops are also reported to illustrate the location of the amino acids substituted into cysteines for fluorophore conjugation (pink, with a rhodamine molecule above them). Putative voltage-sensing, charged amino acids are in blue. (C) Mean voltage dependence of channel opening (G(V)) for WT and Cys mutants is shown (n = 3–7). G(V)s for WT channels have been measured from oocytes with (pink circle) or without (black circles) fluorophore labeling. G(V)s from Cys mutants are all from…

• 

  Figure 2.. WT CaV1.1 does not elicit voltage-dependent fluorescence change when labeled with a thiol-reactive fluorophore. (A) Top view of a cryo-EM structure of the pore-forming α1S subunit of CaV1.1 channels (Protein Data Bank accession no. 5GJV; Wu et al., 2016), illustrating four VSDs that surround and control the central pore. (B) Simultaneously recorded Ba2+ currents (black) and fluorescence signals (gray traces) at the indicated membrane potentials from human CaV1.1 channels (α1S+β1a+Stac3) without an engineered Cys, expressed in Xenopus oocytes and labeled with thiol-reactive fluorophores, are shown.

• 

  Figure 3.. Depolarization-driven structural rearrangements of each VSD in the human CaV1.1 channel exhibit distinct time- and voltage-dependent properties. (A) Simultaneously recorded Ba2+ currents (black) and fluorescence signals (colored traces) reporting local protein structural changes at the indicated membrane potentials from human CaV1.1 channels (α1S+β1a+Stac3) expressed in Xenopus oocytes and labeled with thiol-reactive fluorophores are shown. Fluorophore conjugated at the extracellular flank of S4 helices of each VSD reported time- and voltage-dependent structural rearrangements upon depolarization. (B) Ionic current (black) elicited by 20 mV depolarization is shown superimposed to the fluorescence traces from each VSD (holding potential was −90 mV). Note that only VSD-I fluorescence signal exhibited kinetics compatible with pore opening (see also Fig. S2). (C) Mean voltage dependence of VSD activation constructed at 200 ms from experiments as in A. Lines represents fits to single Boltzmann distributions (SEM bars are within the symbols, when not visible; n = 5–7). Note that each VSD exhibited distinct voltage-dependent properties. Table 3 reports fitting parameters of Boltzmann…

• 

  Figure S2.. VSD-I time course of activation follows ionic current kinetics. (A and B) The activation kinetics of ionic current (black symbols) and of each VSD (colored symbols) were well described by the sum of two exponential functions (τfast, empty symbols; τslow, filled symbols). In the plots, fast and slow time constants (A) and their relative amplitudes (B) are compared for each individual VSD with those of the ionic current. Note that all domains share with the pore a kinetic component with τslow ∼100 ms. However, only VSD-I exhibits a fast component compatible with that of the ionic current; the other VSDs display a τfast approximately fivefold faster, likely more suitable for RYR1 activation.

• 

  Figure 4.. Charge movement and fluorescence signals arise from the same conformational changes. (A and B) Gating current and fluorescence recordings are shown in the same time scale (enlarged view in B). Note that VSD-I barely activates during the time course of gating current measurements (10 ms). (C) Mean voltage dependence of charge movement (Q(V)), channel activation (G(V)), and individual VSD activations (F(V)) are shown. For clarity, only Boltzmann distributions describing G(V) and F(V) datasets are plotted (data points were already presented in Fig. S1 C and Fig. 3 C, respectively). The sum of F(V) curves weighed by their respective valence (pink diamonds) resembles the voltage dependence of the mean Q(V) curve (black circles; SEM bars are within the symbols, when not visible, n = 5).

• 

  Figure S3.. The R174W mutation disfavors CaV1.1 channel activation. (A and B) Ba2+ currents from Xenopus oocytes expressing human CaV1.1 WT (A) or R174W (B) channels are shown. Holding potential was −90 mV. Membrane potentials steps are 10 mV apart.

• 

  Figure 5.. A naturally occurring, charge-neutralizing mutation (R174W) impairs VSD-I activation and voltage-dependent channel opening, leaving VSDs II, III, and IV unaffected. (A) Voltage dependence of pore opening, as measured by normalized Ba2+ current tails for WT or R174W channels. Lines are fits to single Boltzmann distributions. Note that the R174W mutation caused a rightward-shift of the current voltage dependence G(V). (B) Mean voltage dependence of VSD activation constructed from experiments as in Fig. S4. Lines are fits to single Boltzmann distributions. Note that the R174W mutation impaired VSD-I activation, as revealed by a more depolarized voltage dependence, leaving VSDs II, III, and IV unaffected. Tables 1 and 3 report fitting parameters of Boltzmann functions describing G(V) and F(V) curves for R174W CaV1.1 channels, respectively.

• 

  Figure S4.. Optically tracked voltage-dependent rearrangements of the individual VSD in human CaV1.1 R174W mutant. (A–D) Simultaneously recorded Ba2+ currents (black) and fluorescence signals reporting local protein structural changes (colored traces) from human R174W CaV1.1 channels (α1S+β1a+Stac3) expressed in Xenopus oocytes are shown. The topology above the recorded traces highlights the position of the R174W mutation (innermost gating charge of S4 in VSD-I; yellow star) and the location of the conjugated thiol-reactive fluorophore (pink rhodamine molecule). Note that, for VSD-I (A), larger depolarizations are needed to achieve voltage sensor activation as compared with WT VSD-I (Fig. S4 A versus Fig. 3 A, blue traces), since the neutralization of one of its gating charges (R174W) impaired its voltage-dependent activation (Fig. 5 B). Note that the current scale in A is 10-fold bigger than in B, C, and D.

• 

  Figure 6.. Scheme of a structurally relevant, allosteric model of CaV activation, accounting for both time- and voltage-dependent properties of VSD activation and pore opening. See Pantazis et al. (2014), Savalli et al. (2016), and Allosteric model of CaV... Briefly, each VSD and the pore (equilibrium constant J and L) are modeled as two-state particles, undergoing resting (R)–active (A), or closed (C)–open (O), voltage-dependent transitions, respectively. VSD activation allosterically stabilizes the pore open state by coupling energy W.

• 

  Figure 7.. VSD-I is the main contributor to CaV1.1 pore opening. A CaV allosteric model (scheme in Fig. 6) was used to simultaneously fit steady-state and kinetics data in both WT and R174W channels. An error function was minimized using a Markov chain Monte Carlo sampling as in Savalli et al. (2016). Only parameters characterizing VSD-I activity could vary independently for WT and R174W channels. (A and B) Experimental G(V) and F(V) data points are shown superimposed to the model predictions (black lines) for WT (A) and R174W (B) channels. (C–F) Ionic currents (C and E) and fluorescence traces (D and F) from each VSD (normalized to the steady-state probability of activation) in both WT (C and D) and R174W (E and F) channels are shown superimposed with model predictions (black lines, best solutions out of 1.5 ∙ 106 trials from three replicas). Shaded areas depict all solutions within the 95% CI. Solution uniqueness and CIs are described in Fig. 8 and Table S1.

• 

  Figure S5.. Model prediction of each VSD contribution to pore opening. The intrinsic voltage dependence of the pore is extremely rightward-shifted (no VSDs coupled; gray curves). The activation of each voltage sensor provides an energetic contribution to the stabilization of the open state of the channel, resulting in an overall facilitation of channel opening (leftward shift of the G(V); dotted colored lines). Please note that VSD-I activation caused the largest leftward shift of the G(V) (A; blue dotted line), with a smaller contribution of VSD-IV (D; orange dotted line) and negligible roles of VSD-II (B) and VSD-III (C).