Kinetic relationship between the voltage sensor and the activation gate in spHCN channels

Abstract

Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are activated by membrane hyperpolarizations that cause an inward movement of the positive charges in the fourth transmembrane domain (S4), which triggers channel opening. The mechanism of how the motion of S4 charges triggers channel opening is unknown. Here, we used voltage clamp fluorometry (VCF) to detect S4 conformational changes and to correlate these to the different activation steps in spHCN channels. We show that S4 undergoes two distinct conformational changes during voltage activation. Analysis of the fluorescence signals suggests that the N-terminal region of S4 undergoes conformational changes during a previously characterized mode shift in HCN channel voltage dependence, while a more C-terminal region undergoes an additional conformational change during gating charge movements. We fit our fluorescence and ionic current data to a previously proposed 10-state allosteric model for HCN channels. Our results are not compatible with a fast S4 motion and rate-limiting channel opening. Instead, our data and modeling suggest that spHCN channels open after only two S4s have moved and that S4 motion is rate limiting during voltage activation of spHCN channels.

Figure 1.

Predicted kinetic relationship between S4 movement and ionic currents at −120 mV for a published HCN channel model (Altomare et al., 2001). (A) State diagram for the Altomare model. There are four independent fast S4 movements (horizontal transitions) and slow rate-limiting activation-gate (vertical) transition. Both the S4 movement and the gate opening transitions are highly voltage dependent. The arrow on the state diagram denotes the most likely transition pathway. The continuous arrow denotes fast transitions and the dashed line slow transitions. At −120 mV, α = 289.7 s⁻¹, β = 0.59 s⁻¹, γ = 30.2 s⁻¹, δ = 0.35 s⁻¹, and f = 2.23 (Altomare et al., 2001). (B) Predicted S4 movement (dashed line; movement² and movement⁴ denoted with dotted lines), and onset of ionic currents (continuous line).

Figure 2.

Fluorescence from 332C reports on gating charge movement. (A) Sequence alignment of spHCN and HCN1 S4 domains. The red box highlights residues studied with VCF. Marked residues exhibit state-dependent modification by intracellular (*), extracellular (#), or both intra- and extracellular reagents (&), or state-independent modification by either intracellular or extracellular reagents (o) (Mannikko et al., 2002; Bell et al., 2004; Vemana et al., 2004). (B) Current (bottom, black) and fluorescence (top, red) from Alexa-488–labeled 332C/435Y channels in response to voltage steps from 0 to −160 mV (middle). The P435Y mutation has earlier been shown to render spHCN channels nonconducting (Mannikko et al., 2005). Currents at very negative potentials are from endogenous hyperpolarization-activated chloride channels. (C) Integrated gating charge (bottom, black) from tail currents in B and fluorescence (top, red), showing similar time courses of Q and F. Integration started 1 ms after the voltage pulse to allow complete settling of the capacitive transient. (D) Q(V) and F(V) (measured at arrows in C), showing similar voltage dependence of the gating charge and the fluorescence. Q: V_1/2 = −90.7 ± 8.4 mV, z = 1.44 ± 0.35; F: V_1/2 = −85.3 ± 9.4 mV, z = 1.32 ± 0.17 (P > 0.1; n = 7). (E) Charge movement (bottom, black) and fluorescence (top, red) from uninjected oocytes in response to an identical voltage protocol as in A. (F) Q_off gating currents measured at −15 mV in response to longer and longer (Δt = 12.5 ms) prepulses to −120 mV. Inset shows voltage protocol and currents during the protocol. (G) Integrated gating charge Q_off (black) as a function of prepulse length (from F) and normalized fluorescence change (red) measured simultaneously during the prepulse fitted with single exponential curves.

Figure 3.

Fluorescence from 332C precedes channel opening. (A) Representative current (A₁) and fluorescence (A₂) records from Alexa-488-maleimide–labeled 332C channels. Channels were held at 0 mV and then stepped to negative potentials (0 to −140 mV, ΔV = 20 mV), followed by a step to +50 mV. (B and C) Steady-state (B) and kinetic analysis (C) of the voltage dependence of the conductance (black symbols) and fluorescence (red symbols) from A. The current and the fluorescence signals were fitted with double exponentials. (D) Amplitudes of the fast (▪) and slow (•) fluorescent components from A₂. (E) Normalized currents and fluorescence from A during a step to −120 mV. Note that the fluorescence signal precedes the currents, as expected if S4 movement causes channel opening.

Figure 4.

Time course of fluorescence relative to the ionic current. Relative comparison of the fluorescence and the ionic current for steps to indicated voltages. The fluorescence signal is also shown to powers of 2 (F²) and 4 (F⁴). The fluorescence signals to a power of 2 overlap very well with the currents at −160 to −120 mV.

Figure 5.

Fluorescence from N-terminal S4 residues. Representative current (top, black) and fluorescence (bottom, red) records for Alexa-488-maleimide–labeled residues 327C–332C. Channels were held at 0 mV and then stepped to negative potentials (0 to −140 mV, ΔV = 20 mV), followed by a step to +50 mV. Note that residues (328–332C) all have fast fluorescence signals that preceded channel opening, presumably because they report on S4 movement during the gating charge translocation.

Figure 6.

A modified 10-state model reproduces the experimental data. (A) Representative current (black lines) and fluorescence (red lines) records from Alexa-488-maleimide–labeled 332C channels (from Fig. 3 A), overlaid with the best fit (χ² = 16.6) from our 10-state model (dashed line; see parameter values in Table II). In this model, the fluorescence changes only during the rate-limiting S4 gating charge movement (see Fig. 8). (B and C) Normalized currents and fluorescence from B Alexa-488–labeled 332C channels (data from Fig. 4 C) and (C) the 10-state model during a step to −120 mV. Note that the fluorescence signal precedes the currents, as expected if S4 movement causes channel opening. The fluorescence to a power of 2 (F²) and to a power of 4 (F⁴) is also shown for each case. The currents are similar to the F² curve for both the experimental data and for the 10-state model, whereas the F⁴ curve follows after the current in the both the experimental data and the 10-state model.

Figure 7.

Cooperative models did not improve the fits. (A–C) Different possible models for HCN channel opening generating F² = I relationship. The arrows through the state diagrams denote the most likely transition pathway. Continuous arrows denote fast transitions and dashed lines slow transitions. (A) An allosteric model with four independent S4 movements with a non rate-limiting activation-gate transition. This is the model used in Fig. 6. (B) An allosteric model with cooperative S4 movements, in which the movement of each S4 affects the movement of the other S4s by a constant factor c. A similar model was suggested for negative cooperativity in S4 movement in Shaker K channels (Zagotta et al., 1994a). (C) A dimer-of-dimers model for S4 activation of HCN channels as earlier proposed (DiFrancesco, 1999; Ulens and Siegelbaum, 2003). (D) Representative current (black lines) and fluorescence (red lines) records from Alexa-488-maleimide–labeled 332C channels (from Fig. 3 A), overlaid with the best fit (χ² = 16.9) from the dimer-of-dimers model (dashed lines; see parameter values in Table II). In this model, the fluorescence changes only during the rate-limiting S4 gating charge movement. (E) Normalized currents and fluorescence from the dimer-of-dimers model from D during a step to −120 mV. The fluorescence to a power of 2 (F²) and to a power of 4 (F⁴) is also shown. The currents clearly precede the F² curve in the dimer-of-dimers model. In contrast, the current overlays the F² curve in both the experimental data and the 10-state model (Fig. 6, B and C).

Figure 8.

Model for fluorescence from S4 during HCN channel gating. Proposed model for how the S4 movement in HCN channels generates the fluorescence data from 332C channels. The S4 movement (1) in the different subunits is assumed to be independent. The proposed model for the transmembrane movement of S4 charges, with both a vertical S4 movement and a simultaneous opening of an intracellular crevice, is a combination of two models earlier proposed for HCN channels (Mannikko et al., 2002; Bell et al., 2004; Vemana et al., 2004). Channel opening (2) is assumed to be by concerted, allosteric conformational changes in all four subunits. For simplicity, only two, out of the four, subunits are shown. S4 is shown in green, the intracellular activation gate is black, and the fluorophore is shown in red (more fluorescence intensity is shown with more and thicker radiating lines). Fluorescence from the gating charge-carrying region of the voltage sensor (i.e., from 332C), increases when the gating charge-carrying portion of S4 moves down in response to hyperpolarization (1), but does not change during as a direct result of channel opening (2). Fluorescence decreases when S4 moves up in response to depolarization (3), but does not change as a direct result of channel closing (4).