Molecular bases for the asynchronous activation of sodium and potassium channels required for nerve impulse generation

Abstract

Most action potentials are produced by the sequential activation of voltage-gated sodium (Nav) and potassium (Kv) channels. This is mainly achieved by the rapid conformational rearrangement of voltage-sensor (VS) modules in Nav channels, with activation kinetics up to 6-fold faster than Shaker-type Kv channels. Here, using mutagenesis and gating current measurements, we show that a 3-fold acceleration of the VS kinetics in Nav versus Shaker Kv channels is produced by the hydrophilicity of two "speed-control" residues located in the S2 and S4 segments in Nav domains I-III. An additional 2-fold acceleration of the Nav VS kinetics is provided by the coexpression of the β1 subunit, ubiquitously found in mammal tissues. This study uncovers the molecular bases responsible for the differential activation of Nav versus Kv channels, a fundamental prerequisite for the genesis of action potentials.

Figure 1. Acceleration of VS Movement in Mammalian Nav Channels by the β1 Subunit

(A) Activation gating current traces for Nav1.4 and Nav1.2 in presence or absence of the β1 subunit. (B) Graph showing gating current activation time constant as a function of the pulse voltage for Nav1.2 (circles) and Nav1.4 (squares) in presence (open symbols, n = 3 for Nav1.2, n = 3 for Nav1.4) or absence (full symbols, n = 4 for Nav1.2, n = 5 for Nav1.4) of the β1 subunit. (C) Graph showing the gating current activation time constant as function of the pulse voltage for Shaker (full squares) and Kv1.2 (open circles). All error bars indicate ±SEM.

Figure 2. Two Speed-Control Residues in Voltage Sensors

(A) Portion of the S2 and S4 segments in several voltage sensors (Nav1.4, GI:116453; Shaker, GI:24642916; Kv1.2, GI:4826782; NavAB, GI:339961377; and NaChBac, GI:38489212) were aligned with respect to conserved charged residues (bold black). Rapid VSs contain hydrophilic residues at conserved speed-control sites (bold red), while slow VSs contain hydrophobic residues (bold blue) at homologous positions. (B) Sequence of the NaSlo1, Naslo2, NaSlo3, and Shaker-I287T/V363T mutants. (C) Normalized ionic current traces for WT Nav1.4 (red) and the NaSlo3 mutant (blue). (D) The time constant of the activating rise of ionic current is plotted as a function of voltage for WT (open squares, n = 4), NaSlo1 (full up triangles, n = 5), Naslo2 (full down triangles, n = 6), and Naslo3 (full circles, n = 6). (E) Gating current recordings for Shaker (top, left), Nav1.4 (top, right), NaSlo1 (bottom, left), and Shaker-I287T/V363T (bottom, right). (F) Voltage dependence of the activation gating time constant for Nav1.4 (red circles, n = 6), Shaker (blue squares, n = 6), Shaker I287T/V363T (red triangles, n = 6), and NaSlo1 (blue triangles, n = 6). (G) Mean τ_max values for the indicated WT or mutant Shaker (full bars) and Nav1.4 (open bars) channels. See also Figure S2 for a larger sequence alignment of eukaryotic Nav channels. All error bars indicate ±SEM.

Figure 3. Hydrophilic Conversion of Speed-Control Residues in Nav1.4 DIV Accelerates Fast Inactivation

(A) Partial amino acid sequence of the NaFas mutant showing the insertion of hydroxylated Thr residues at the S2 and S4 speed-control sites in DIV. (B) Ionic current recordings for Nav1.4 WT (blue) and the NaFas mutant (red). (C) Normalized sodium current recordings for NaFas (red) and WT Nav1.4 (blue). (D) Graph showing the weighted time constant for the fast inactivation in WT Nav1.4 (blue squares, n = 5) and in the NaFas mutant (red squares, n = 7). All error bars indicate ±SEM.

Figure 4. A Mechanism for the Speed-Control Residues in Voltage Sensors

(A and B) The V_½ of the Q–V curve (right) and the τ_max values for activation (left) and deactivation (middle) are plotted as a function of the hydrophobicity (Kyte and Doolittle, 1982) of the amino acid present at position I287 (A) and V363 (B) in the Shaker VS. The native residue is indicated by an asterisk. The data points and trend lines are gradually colored from red (more hydrophilic) to blue (more hydrophobic). The data are representative of at least five independent experiments. (C) A hypothetical two-state VS energy landscape at 0 mV helps to interpret the phenotypes for the I287 mutants: decreasing the side-chain hydrophobicity at the position of I287 (blue trace to red trace) mainly decreases the energy barrier with a small relative stabilization of the resting state. (D) The residue homologous to I287 forms a hydrophobic barrier that insulates two water-accessible pockets in a resting state model (Vargas et al., 2011). (E) Interpretation of the phenotypes for the V363 mutants using a two-state VS energy landscape at 0 mV: decreasing the side-chain hydrophobicity at the position of V363 destabilizes the resting conformation. (F) The residue homologous to V363 is surrounded by membrane lipids in a resting state model (Vargas et al., 2011). See also Figure S4 for a detailed gating current analysis of Shaker mutants. All error bars indicate ±SEM.

Figure 5. Gating Current Recordings for Ci-VSP L224 Mutants

(A) Activating gating current traces for the indicated L224 substitutions and WT Ci-VSP. (B–D) The plots show the relation between the V_½ of the Q–V curve (B), the slowest time constant (τ_max) recorded during activation (C) or deactivation (D), and the hydrophobicity of the side chain present at the position of L224 (Kyte-Doolittle). The trend lines and experimental points are gradually colored from red to blue according to the hydrophobicity value. The native residue L224 is indicated by an asterisk. The data in (B)–(D) are representative of four to seven independent experiments. All error bars indicate ±SEM.