|
Figure 1. Chemical structure of NBD-cAMP and the modulation of WT spHCN channels. (A) Chemical structure of the NBD-cAMP (Axxora). (B) Top: Voltage protocol. Bottom: Current traces of the WT spHCN channel recorded in the absence of cAMP (black), 5 μM NBD-cAMP (green), 10 μM NBD-cAMP (red), or 10 μM regular cAMP (blue). (C) Competitive binding assay for the specificity of NBD cAMP binding to spHCN channels. Raw fluorescent images of membrane patch showing a decrease in the 0.5-μM NBD-cAMP fluorescence signal upon addition of saturating concentration of 50 μM nonfluorescent cAMP. (D) Normalized fluorescent intensities from competitive binding experiments. The fluorescence intensity of NBD-cAMP (black, control, n = 5) decreased upon adding regular cAMP (red, 10 μM, n = 6; blue, 50 μM, n = 6). Error bars here and after represent SEM (see Statistics in Materials and methods).
|
|
Figure 2. Inactivated spHCN channels show decreased binding to NBD-cAMP. (A) In response to a hyperpolarizing voltage step from +40 mV to −80 mV, the WT spHCN channel transiently opens and then inactivates (black trace; no cAMP). Adding cAMP (10 μM) to the bath solution abolishes the inactivation so that the channel shows typical voltage-dependent activation and deactivation (red). (B) Current traces of the WT spHCN channel in response to a series of voltage steps from −40 to −130 mV at a −10-mV interval. Top: Current traces in the absence of cAMP. Inset shows a zoomed view over the transient phase of activation–inactivation. Bottom: Current traces with cAMP. (C) Raw PCF results of WT spHCN channels. 2 mM Trolox was added to the bath solution to reduce photobleaching and PDM of the channel (Idikuda et al., 2018). The membrane patch was held at 0 mV and then at +80 mV for 6 s to stabilize the optical signal and to let the channel reach a steady state, preceding the hyperpolarizing voltage step from +80 to −100 mV. From top to bottom: Voltage protocol, current trace, laser pulses, and CCD camera exposure protocol, normalized fluorescence intensity. Three representative images (a, b, and c) are shown in D. (D) Raw fluorescence images of the membrane patch along the hyperpolarization voltage step (a, b, and c in C). (E) Averaged results showing the significant reduction in fluorescence intensity (31.4 ± 0.02%; paired t test; P < 0.0001; n = 12) upon the voltage step from +80 to −80 mV. The fluorescence intensity was normalized to the averaged value of the last three images before the voltage step. Green trace shows the single-exponential fit of the time-dependent decrease in cAMP binding. (F) Averaged results showing a significant recovery in fluorescence intensity (51.1 ± 0.03%; paired t test; P < 0.0001; n = 12) upon the depolarization voltage step from −80 to +80 mV. Green trace shows the single-exponential fit of the time-dependent increase in fluorescence intensity. Asterisks here and after represent levels of statistical significance (see Statistics in Materials and methods).
|
|
Figure 3. A single point mutation in S6, F459L, abolishes the channel inactivation and reverses the decrease in NBD-cAMP binding upon membrane hyperpolarization. (A) Current traces of the spHCN/F459L mutant channel recorded in response to a series of hyperpolarization voltage steps (no cAMP). (B) Current traces of the spHCN/F459L channel recorded in the presence of 10 μM cAMP. Notice the changes in channel kinetics (faster activation and slower deactivation) compared with the traces shown in A. (C) Top: Current traces of spHCN/F459L in response to the hyperpolarization voltage step from +80 to −80 mV. Bottom: Averaged results corresponding to the dashed blue box shown in C showing a significant increase in ΔΔF upon the voltage step from +80 to −80 mV (58.4 ± 0.05%; paired t test; P < 0.0001; n = 12). (D) Top: Current traces of spHCN/F459L in response to the depolarization voltage step from −80 to +80 mV. Bottom: Averaged results corresponding to the dashed blue box shown in D showing a significant decrease in ΔΔF upon the voltage step from −80 to +80 mV (34.78 ± 0.03%; paired t test; P < 0.0001, n = 17).
|
|
Figure 4. Comparing the decrease (WT spHCN) or increase (spHCN/F459L) in NBD-cAMP binding in response to the same set of hyperpolarization voltage steps. (A) The membrane patches were held at +80 mV for 5 s before the application of hyperpolarization voltage steps from −60 to −120 mV at a −20-mV interval. Fluorescence intensities for each patch were normalized to the value of the last three images just before the voltage steps. At −60 mV, paired t test: P < 0.0001, n = 12. (B) Percentage of the decreases in fluorescence intensity versus hyperpolarization voltage steps. Averaged results of the WT spHCN channel shown in A are used in the calculation. (C) Time constant of the decrease in fluorescence intensity of the WT spHCN channel versus hyperpolarization voltage steps. The profiles of the decrease in fluorescence intensity after the hyperpolarization voltage steps shown in A were fitted by a single-exponential function. (D) Normalized fluorescence intensity for the spHCN/F459L mutant channel. At −60 mV, paired t test: P < 0.0001, n = 13. (E) Percentage of the increases in fluorescence intensity of the spHCN/H462A mutant channel versus hyperpolarization voltage steps. Averaged results shown in D are used in the calculation. (F) Time constant of the increase in fluorescence intensity of the spHCN/H462A channel versus hyperpolarization voltage steps. Profiles of the increase in fluorescence intensity after hyperpolarization voltage steps shown in D were fitted by a single-exponential function.
|
|
Figure 5. Both ZD7288 and Cs+ can block the spHCN current, but only ZD7288 affects the binding of NBD-cAMP. (A) ZD7288 blocks the currents of the WT spHCN channel. Top: Voltage step, laser pulse, and image collection protocol. Bottom: Current traces before (black) and after (red) adding 60 µM ZD7288. (B) Normalized fluorescence intensity before (black) and after (red) ZD7288 application. Voltage step, +80 to −80 mV (n = 17). (C) Normalized fluorescence intensity before (black) and after (red) ZD7288 application. Voltage step, +80 to −100 mV. (D) 2 mM Cs+ was added to the pipette solution to block the WT spHCN current from the extracellular side. Top: Voltage step, laser pulse, and image collection protocol. Bottom: Current traces collected with Cs+ added to the pipette solution. (E) Normalized fluorescence intensity without (black) or with Cs+ (red). Voltage step, +80 to −80 mV. Because the pipette solution was not exchanged during the experiments, control results and Cs+ results were collected from different patches (n = 13). (F) Normalized fluorescence intensity without (black) or with Cs+ (red). Voltage step, +80 to −80 mV.
|
|
Figure 6. Strategy to specifically lock the spHCN channel in either the open or closed state. (A) Alignment of primary protein sequences of representative HCN and other potassium channels in the region encompassing the selectivity filter and the last transmembrane segment (S2 in KcsA or S6 in other channels). Relevant residues are shown in bold and marked with a different color. (B) Mutations introduced to the S6 of the spHCN channel to make the locked-open (H462C-L466C) or locked-closed (H462Y-Q468C) channel. (C) Current traces of the WT spHCN channel. Black, control in the absence of cAMP and Cd2+. Red, 10 μM cAMP. Blue, 10 μM cAMP and 1µM Cd2+. (D) Current traces of the locked-open spHCN channel. Black, control in the absence of cAMP and Cd2+. Red, 10 μM cAMP. Blue, 1 µM Cd2+ without cAMP. The locked-open effect was persistent after the washing off of cAMP (bath solution containing Cd2+). (E) Current traces of the locked-closed spHCN channel. Black, control in the absence of cAMP and Cd2+. Red, 10 μM cAMP. Blue, 2 μM Cd+2 and 10 µM cAMP.
|
|
Figure 7. The inactivated WT spHCN channel, the locked-open spHCN channel and the locked-closed spHCN channel show three different levels of binding to NBD-cAMP. (A) Current traces and original fluorescence images of the WT spHCN channel. Left top: Voltage step, laser pulse, and image collection protocol. The time of three images (a, b, and c) is indicated. Left bottom: Current traces before and after Cd2+. Right: Images collected before (top) and after (bottom) Cd2+ application. (B) Results of the locked-open spHCN channel. The same experimental protocol as used for the WT spHCN channel was applied. (C) Results of the locked-closed spHCN channel. The same experimental protocol as used for the WT spHCN channel was applied. (D) Normalized fluorescence intensity of the WT (black), locked-open (green), and locked-closed (magenta) spHCN channels before Cd2+ application. Results were normalized to the fluorescence intensity of image a. Before the application of Cd2+, all channels show a decrease in cAMP binding upon the hyperpolarization voltage step from 0 to −100 mV and a significant recovery after the depolarization voltage step from −100 to 0 mV (P < 0.001). (E) Normalized fluorescence intensity of the WT (black, n = 6), locked-open (green, n = 8), and locked-closed (magenta, n = 8) spHCN channels after Cd2+ application. After the application of Cd2+, the WT and the locked-closed channels still show similar trend of decrease in fluorescence intensity, whereas the locked-open channel shows a significant increase (P < 0.001) in fluorescence intensity.
|