Mechanism of pore opening in the calcium-activated chloride channel TMEM16A

Figures

• 

  Fig. 1. Mechanistic analysis of power spectra for wild-type TMEM16A. a Representative section of the steady-state currents recorded at the indicated intracellular Ca2+ concentrations at +80 mV. The data correspond to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_O\left( t \right) = \overline {P_O} \frac{{I(t)}}{{\bar I}} = \overline {P_O} \frac{{NiP_o\left( t \right)}}{{Ni\overline {P_o} }}$$\end{document}POt=PO¯I(t)Ī=PO¯NiPotNiPo¯, where I is the macroscopic current, Po the open probability, N the number of channels, i the unitary current, and the bar notation indicates the mean of. For display, the current traces were filtered at 2.5 kHz using a digital 4-pole Bessel lowpass filter. b Power spectra of the respective steady-state currents shown in a. Data are averages of the indicated number of patches (80 nM, n = 8; 200 nM, n = 8; 400 nM, n = 8; 800 nM, n = 8; 2 µM, n = 8; 4 µM, n = 7). c Concentration-response relation. Data are averages of eight patches, errors are SEM. b, c Solid lines are the…

• 

  Fig. 2. Electrostatic profiles of the Ca2+ binding site and their effect on the activation sequence. a Pore region of a TMEM16A subunit in the Ca2+-free conformation (PDB: 5OYG) in Cα representation. Ca2+-binding residues are shown as sticks. Yellow spheres indicate the points at which the electrostatic potential is plotted. b Electrostatic potential (Φ) along the path for the indicated channel configurations (insets). Left, α6-loose conformation in the absence of Ca2+. Center, α6-loose conformation with the upper Ca2+ binding site occupied. Right, α6-tight conformation with the upper Ca2+ binding site occupied. Green spheres correspond to bound Ca2+ ions and small spheres display the trajectory at which the electrostatic potential was plotted (colors indicate the local electrostatic potential). The data point at zero (*) on the x-axis corresponds to the position at the vacant lower Ca2+ binding site. c Relationship between the calculated electrostatic potential and the estimated kon for Ca2+ binding (displayed in Fig. 1e). Errors are 95% confidence intervals. d Relative probability of the two possible activation sequences (top, highlighted in brown and green) as a function of Ca2+…

• 

  Fig. 3. Mechanistic double-mutant cycle analysis of residues of the gate region. a Conformational rearrangement of the inner pore entrance upon Ca2+ binding. Cα-trace of indicated transmembrane helices of the Ca2+-free (PDB: 5OYG) and the Ca2+-bound conformation of TMEM16A (PDB: 5OYB) are shown. Sidechains of residues of the gate are displayed as sticks, sphere indicates position of the gating hinge. b Reduced mechanism depicting gating transitions at saturating Ca2+ concentrations. c Single-channel current (i) and open probability (Po) estimated from non-stationary noise analysis at saturating Ca2+ concentrations. Circles correspond to individual data points. Bars refer to median and interquartile range of data. d–f Equilibrium constants for mutants in the gate region for transitions d, L32, e, L21, and f, L10. Bars indicate the best-fit values of the averaged data shown in Supplementary Fig. 7d (WT, n = 7; I550A, n = 6; I551A, n = 7; I641A, n = 7; I550A/I641A, n = 7; I551A/I641A, n = 6). Errors are 95% confidence intervals. g Coupling energy (Gcoupling) and h independent energetic contribution (Gindep) of the indicated residues. Bars indicate quantities calculated using g, Eqs. 31–32,…

• 

  Fig. 4. Mechanistic double-mutant cycle at the α4–α6 interface. a Relation between Ile 551 and Gln 649 in Ca2+-free and -bound conformations. Representation is as in Fig. 3a. b–d Equilibrium constants for mutants at the interface for transitions b, L32, c, L21, and d, L10. Bars indicate the best-fit values of the averaged data shown in Supplementary Fig. 7d (Q649A, n = 7; Q649L, n = 7; I551A/Q649A, n = 8; I551A/Q649L, n = 7). Errors are 95% confidence intervals. e Coupling energy (Gcoupling) and f independent energetic contribution (Gindep) of the indicated residues. Bars indicate quantities calculated using e, Eqs. 31–32, 35 and f, Eq. 34 from the best-fit values shown in b–d. Errors correspond to standard errors. Asterisks indicate significant deviation from zero in a two-sided one-sample t-test (e I551/Q649: ***p = 2e−7 and ***p = 3e−5; I551/L649: ***p = 5e−6 and *p = 0.023; f ***p = 1e−7, ***p = 6e−4, and ***p = 0.001).

• 

  Fig. 5. Functional coupling in the open state. a Instantaneous I-V relations for the indicated mutants recorded at saturating Ca2+ concentrations. Solid lines are fits to a phenomenological ion permeation model described in Eq. 1. Dashed line indicates the relation of WT. Data are averages of the indicated number of patches (WT, n = 6; I550A, n = 6; I551A, n = 8; I641A, n = 7; Q649A, n = 6; I550A/I641A, n = 10; I551A/I641A, n = 10; I551A/Q649A, n = 16), errors are SEM. b Minimal ion permeation model to account for the I-V relations. c, d Conduction parameters c, ΔEa(σβ) and d, ΔEa(σh) corresponding to the barrier heights for ion conduction at the inner pore entrance and the narrow pore relative to that of the outer barrier of the indicated mutants. Bars indicate quantities calculated using Eq. 2 from the best-fit values of the averaged data shown in a. Errors are 95% confidence intervals. e Coupling energy (Ecoupling) for residue pairs. Bars indicate quantities calculated using Eqs. 31–32, 35 from the values shown in c and d. Errors correspond to standard errors. Asterisks indicate significant deviation from zero in a two-sided one-sample t-test (I550/I641: ***p = 0.002; I551/I641: ***p…

• 

  Fig. 6. Rate-equilibrium free-energy relations of the gating steps. Plots show the experimental relation between the forward (blue) and backward (red) rate constants and the corresponding equilibrium constants for the transitions relating the gating intermediates and end points (from C3 to O0). Each data point corresponds to the parameter(s) obtained for a mutant in the gate region shown in Supplementary Fig. 7d. Errors correspond to 95% confidence intervals (respective error bars are not visible when they are smaller than the symbols). Solid lines are fits to the pair of rate-equilibrium free energy relations for each transition (Eq. 36). The estimated value of phi is presented as best-fit ± standard error.

• 

  Fig. 7. Activation mechanism. a Cartoon depiction of the activation mechanism. The states are as in Fig. 1d. Inset (top) shows a schematic representation of the energy maxima in the gate region between connected states. The relative location of the transition states was obtained from phi value analysis. The two alternative pathways for activation at physiological and at high Ca2+ concentrations are indicated. b Schematic depiction of the relationship between residues of the gate region in different steps during activation. The view is from the extracellular side. Gate residues located on α4 (Ile 550 and Ile 551) retain their interaction with the gate residue Ile 641 on α6 in the transition between the three closed conformations although the interaction between Ile 551 and Ile 641 successively weakens until it is replaced by an interaction with Gln 649 upon transition into the open state where the gate has dissociated and the pore has dilated to permit ion conduction.