Fig 1.
Multiple time courses of Po covering a wide voltage range for HCN2 channel gating.
Currents were activated by hyperpolarizing pulses to the voltage Va with the duration ta followed by depolarizing pulses to the voltage Vd to generate deactivation. (a) Representative current trace. Va = - 140 mV, ta = 4 s, Vd = -40 mV. (b) Steady-state activation in the absence of cAMP and in the presence of 10 μM cAMP. The Boltzmann curves, obtained with pulses of 4 s duration, were computed according to a previous report by equation S1 [47]. The values were: no cAMP: z = 5.15, Vh = -118.4 mV; 10 μM cAMP: z = 4.88, Vh = -97.1 mV. The blue and red arrows indicate the used values of Va and Vd, respectively. (c) Set of Po traces following 27 different double pulse protocols in the absence of cAMP. Va = -140 mV: ta = 0.3 s, 1 s, 4 s; Va = -125 mV: ta = 0.5 s, 2 s, 11 s; Va = -110 mV: ta = 1.5 s, 5 s, 15 s (n = 5–18). (d) As c but in the presence of cAMP. Va = -130 mV: ta = 0.15 s, 0.3 s, 3 s; Va = -100 mV: ta = 0.5 s, 2 s, 11 s; Va = -90 mV: ta = 1.5 s, 5 s, 15 s. In c and d Vd was -40 mV, 20 mV, 80 mV (n = 7–10).
Fig 2.
Global fit of activation and deactivation time courses in either the absence or presence of cAMP.
(a) Scheme of model 1n. (b) Fit of the 27 time courses of activation and deactivation in the absence of cAMP. (c) Scheme of model 1a. (d) Fit of the 27 time courses of activation and deactivation in the presence of 10 μM cAMP. The experimental traces and fitted curves are given in black and red color, respectively. Shades of gray indicate s.e.m. The parameters are provided by Table 1.
Table 1.
Parameters for the fits with models 1n (no cAMP) and 1a (10 μM cAMP).
Fig 3.
Effect of cAMP on the time-dependent population of states.
The time courses were computed with model 1n (no cAMP) or 1a (10 μM cAMP). (a,b) Full activation and deactivation without and with cAMP. (c,d) Moderate voltage-dependent activation and deactivation without and with cAMP.
Fig 4.
Effect of cAMP on the probability flux densities along the main pathways of activation from 0 mV to -130 mV and subsequent deactivation to -40 mV.
The time courses were computed with model 1n (no cAMP) or 1a (10 μM cAMP). The colors of the arrows in the top schemes correspond to the colors of the probability flux density time courses. The predominant effect of cAMP is to strongly accelerate the probability flux density of the activation pathway.
Fig 5.
Effect of cAMP on computed gating currents per channel and moved gating charges.
The gating currents per channel were computed with model 1n (no cAMP) or 1a (10 μM cAMP). The holding potential was set to 0 mV. (a) ON-gating currents at the voltages between -140 and -70 mV. The main effect of cAMP is to significantly accelerate both the rising and the decay phase of the gating current, resulting in an increase of the peak amplitude. (b) OFF-gating currents at -40 mV following voltage pulses indicated in (a) containing a fast and a slow component. Both components are decreased by cAMP. (c) Effect of cAMP on the total gating charge computed by integrating the gating currents for a voltage pulse of 1.5 s duration in the absence and presence of cAMP. The holding potential was set to 0 mV. The effect of the voltage-dependent transition C1→C1* is only minor.
Fig 6.
Cartoons illustrating the mutual interactions between mutual stimuli.
The graphs show schematically two S6-helices forming the inner gate at their inner end, two S4-helices being the voltage sensor and the CNBD with two of the four binding sites (circles). Blue color symbolizes a cause, green and red color an activating and inhibiting effect, respectively. (a) Effects of voltage-evoked activation on binding affinity and Po. (b) Effects of cAMP binding on gating charge and ON-gating current. (c) Effects of pore opening on OFF-gating current. For further explanation see text.