Shi YP et al. (2019), Extracellular protons accelerate hERG channel d...

XB-ART-56951

J Gen Physiol 2019 Feb 04;1512:231-246. doi: 10.1085/jgp.201812137.

Show Gene links Show Anatomy links

Extracellular protons accelerate hERG channel deactivation by destabilizing voltage sensor relaxation.

Shi YP , Thouta S , Cheng YM , Claydon TW .

???displayArticle.abstract???
hERG channels underlie the delayed-rectifier K+ channel current (IKr), which is crucial for membrane repolarization and therefore termination of the cardiac action potential. hERG channels display unusually slow deactivation gating, which contributes to a resurgent current upon repolarization and may protect against post-depolarization-induced arrhythmias. hERG channels also exhibit robust mode shift behavior, which reflects the energetic separation of activation and deactivation pathways due to voltage sensor relaxation into a stable activated state. The mechanism of relaxation is unknown and likely contributes to slow hERG channel deactivation. Here, we use extracellular acidification to probe the structural determinants of voltage sensor relaxation and its influence on the deactivation gating pathway. Using gating current recordings and voltage clamp fluorimetry measurements of voltage sensor domain dynamics, we show that voltage sensor relaxation is destabilized at pH 6.5, causing an ∼20-mV shift in the voltage dependence of deactivation. We show that the pH dependence of the resultant loss of mode shift behavior is similar to that of the deactivation kinetics acceleration, suggesting that voltage sensor relaxation correlates with slower pore gate closure. Neutralization of D509 in S3 also destabilizes the relaxed state of the voltage sensor, mimicking the effect of protons, suggesting that acidic residues on S3, which act as countercharges to S4 basic residues, are involved in stabilizing the relaxed state and slowing deactivation kinetics. Our findings identify the mechanistic determinants of voltage sensor relaxation and define the long-sought mechanism by which protons accelerate hERG deactivation.

???displayArticle.pubmedLink??? 30530765
???displayArticle.pmcLink??? PMC6363419
???displayArticle.link??? J Gen Physiol

Species referenced: Xenopus
Genes referenced: kcnh2

???attribute.lit??? ???displayArticles.show???

	Figure 1. hERG mode shift is reduced by acidic pH. (A) Typical ionic recordings elicited with activating voltage-step durations of 250 ms and deactivating voltage-step durations of 750 ms (following depolarization to +60 mV for 250 ms), mimicking physiological durations of a cardiac action potential. Arrows mark where current measurements were made. (B) Plots of the normalized G-V for activation and deactivation, fitted to a Boltzmann function, yielded V1/2 values of −12.4 ± 1.1 mV and −66.0 ± 0.9 mV, respectively. The mean mode shift was −53.6 ± 1.8 (n = 6). (C) Typical ionic recordings from the activation and deactivation protocols in A at pH 6.5. (D) Boltzmann fits of the normalized G-V for activation and deactivation yielded V1/2 values of −5.9 ± 0.7 mV and −43.0 ± 0.7 mV, respectively, with a mode shift of −37.1 ± 1.1 mV (n = 6), which was reduced by 31% compared with that at pH 7.4 (P < 0.0001, Student’s t test). Dashed lines represent baseline.
	Figure 2. The pH sensitivity of mode shift behavior is similar to that of deactivation kinetics. (A and B) hERG channels were recruited into the relaxed state by applying a 15-s depolarizing step to +60 mV, following which deactivation kinetics were assessed during a repolarizing step to −110 mV. Current decay during repolarization was fitted to exponential function, which yielded values for τfast of 201 ± 11 ms (pH 8.5, n = 4), 192 ± 20 ms (pH 7.4, n = 5), 99 ± 16 ms (pH 6.5, n = 5), 79 ± 5 ms (pH 5.5, n = 5), and 73 ± 7 ms (pH 4.5, n = 5). (C) Plot of the pH dependence of the τfast of deactivation. Fitting the data with a Hill function yielded a pKa of pH 6.8 (Hill coefficient, n = 1.9). (D and E) Representative traces recorded during the voltage protocols (insets) designed to measure the steady state voltage dependence of activation and deactivation at pH 7.4 (D) and pH 6.5 (E). Arrows mark where current measurements were made. Plots of the normalized voltage dependence of activation and deactivation G-V at each pH are shown at right. Boltzmann fits of the data recorded at pH 7.4 yielded V1/2 values for activation and deactivation of −34.5 ± 1.3 mV and −56.4 ± 1.2 mV, respectively (n = 12). Corresponding values measured at pH 6.5 were −27.3 ± 0.8 mV and −38.8 ± 1.2 mV (n = 5). The mode shift calculated from these data was reduced by 74% from −21.9 ± 1.4 mV at pH 7.4 to −11.5 ± 0.5 mV at pH 6.5. (F) Plot of the pH dependence of the mode shift measured as in D and E. Fitting the data with a Hill function yielded a pKa of pH 7.0 (Hill coefficient, n = 1.1; also see Table 2). Dashed lines represent baseline.
	Figure 3. Voltage sensor mode shift. (A and B) Typical on-gating (Igon; A) and off-gating (Igoff; B) currents recorded at pH 7.4 in response to the voltage protocol shown (insets). (C) Plot of the voltage dependence of Qon and Qoff. Qon-V was calculated from the integral of the off-gating current measured during a 500-ms step to −100 mV (A, inset). Qoff-V was determined from integrals of the on-gating current measured during a 500-ms step to 0 mV (B, inset). Data were fitted with a Boltzmann function, which yielded V1/2 values for Qon and Qoff of −48.2 ± 0.5 mV (n = 3) and −75.5 ± 2.3 mV (n = 4), respectively. The mean gating mode shift was −27.3 ± 2.4 mV. (D and E) Typical on-fluorescence (Fon; D) and off-fluorescence (Foff; E) reports from TMRM attached at L520C at pH 7.4 in response to the voltage protocols shown (insets). (F) Plot of the voltage dependence of normalized Fon and Foff. Fon-V was constructed by measuring relative fluorescence change at the end of the 2-s depolarizing voltage step. Foff-V was constructed by measuring relative fluorescence change at the end of the 4-s repolarizing voltage step. Data were fitted with a Boltzmann function, which yielded V1/2 values for Fon and Foff of −44.7 ± 1.0 mV (n = 4) and −76.1 ± 3.2 mV (n = 5), respectively. The mean fluorescence mode shift was −31.4 ± 3.4 mV. The mean V1/2 values measured for Qon and Fon were not statistically different (P > 0.17), nor were the equivalent values for Qoff and Foff (P > 0.74). Dashed lines represent baseline.
	Figure 4. Voltage sensor mode shift is reduced by acidic pH. (A and B) Typical gating currents recorded at pH 6.5 in response to the voltage protocol shown (insets), as in Fig. 3. (C) Plot of the voltage dependence of Qon and Qoff. Data were fitted with a Boltzmann function, which yielded V1/2 values for Qon and Qoff of −44.8 ± 5.0 mV (n = 8; not significantly different from pH 7.4, P > 0.5) and −51.8 ± 3.3 mV (n = 5; P < 0.0007 compared with pH 7.4), respectively. The mean gating current mode shift was reduced to −7.0 ± 6.0 mV. Dashed lines show pH 7.4 fits from Fig. 3 C for comparison. (D and E) Typical on-fluorescence (Fon; D) and off-fluorescence (Foff; E) reports from TMRM attached at L520C at pH 6.5 in response to the voltage protocols shown (insets). Arrows mark where fluorescence measurements were made. Dashed lines represent baseline. (F) Plot of the voltage dependence of Fon and Foff. Fon-V was constructed by measuring relative fluorescence change at the end of the 2-s depolarizing voltage step. Foff-V was constructed by measuring relative fluorescence change at the end of the 4-s repolarizing voltage step. Data were fitted with a Boltzmann function, which yielded V1/2 values for Fon and Foff of −42.8 ± 3.7 mV (n = 5; not significantly different from pH 7.4, P > 0.5) and −57.2 ± 1.9 mV (n = 5; P < 0.001 compared with pH 7.4), respectively. The mean fluorescence mode shift was reduced to −14.4 ± 4.2 mV. Dashed lines show pH 7.4 fits from Fig. 3 F for comparison.
	Figure 5. Neutralization of D509 mimics the effect of protons on voltage sensor mode shift. (A, B, D, and E) Typical on-fluorescence (A and D) and off-fluorescence (B and E) changes from TMRM attached at L520C in D509A mutant channels in response to the voltage protocols shown (insets) at pH 7.4 (A and B) and pH 6.5 (D and E). Arrows mark where fluorescence measurements were made. (C and F) Plots of the voltage dependence of Fon and Foff at pH 7.4 (C) and pH 6.5 (F). Data were fitted with a Boltzmann function, which yielded V1/2 values for Fon and Foff of −52.2 ± 8.1 mV (n = 4) and −57.8 ± 3.7 (n = 3), respectively, at pH 7.4. The corresponding values at pH 6.5 were −42.7 mV ± 3.5 (n = 3) and −49.6 mV ± 4.4 (n = 4), respectively. The mean voltage sensor mode shift was −5.6 ± 8.9 mV at pH 7.4 and −6.9 mV ± 2.1 at pH 6.5. Dashed lines represent baseline.
	Figure 6. Mode shift is greatly diminished and is pH independent in D509A mutant channels. (A and D) Typical ionic current recordings from D509A mutant channels at pH 7.4 (A) or pH 6.5 (D) in response to the voltage protocols shown (insets) designed to measure the voltage dependence of activation. The membrane was held at −80 mV before 250-ms steps were applied up to +100 mV (in 10-mV increments), followed by a repolarizing step to −60 mV. Arrows mark where current measurements were made. (B and E) Typical ionic current recordings of D509A channel deactivation at pH 7.4 (B) or pH 6.5 (E). The membrane was held at −80 mV before a 250-ms depolarizing step to +100 mV, followed by 750-ms repolarizing steps down to −130 mV (in 10-mV increments). (C and F) Plots of the voltage dependence of activation and deactivation at pH 7.4 (C) and pH 6.5 (F). Data were fitted with a Boltzmann function, which yielded V1/2 values for activation and deactivation of +17.1 ± 0.8 mV (n = 5) and +1.6 ± 1.1 mV (n = 5), respectively, at pH 7.4. The corresponding values at pH 6.5 were +21.6 ± 1.5 mV (n = 5) and +4.8 ± 1.2 mV (n = 5), respectively. The mean mode shift was −15.5 ± 0.1 mV at pH 7.4 and −16.8 mV ± 0.6 at pH 6.5 (P > 0.3; not significantly different, Student’s t test; see Table 1). Dashed lines represent baseline.
	Figure 7. Model of hERG voltage sensor relaxation. (A) Markov gating scheme used to model hERG voltage sensor gating at pH 7.4 and pH 6.5 (top). Details are provided in the text, and transition rates are described in Table 4. The transition highlighted red represents that which was modified to simulate voltage sensor behavior at pH 6.5. Simulated on-gating currents using the model along with predicted on- and off-gating currents are shown for pH 7.4 (lower left) and pH 6.5 (lower right). Boltzmann fits yielded V1/2 values of −52.8 mV and −73.0 mV for Qon-V and Qoff-V, respectively, at pH 7.4 (mode shift = −20.2 mV). At pH 6.5, equivalent values were −47.9 mV and −54.7 mV (mode shift = −6.8 mV). The salient features of gating currents at pH 6.5 could be modeled by accelerating the rate of de-relaxation. (B) Cartoon representation of proposed reconfigurations of S2, S3, and S4 voltage sensor transmembrane domains in the resting, activated, and relaxed state. K1 and K6 refer to K525 and K538, respectively. R1–R5 describe R528–R537. Putative reconfigurations associated with entry into the relaxed state are shown, which may involve stabilizing interactions of D509 with outer S4 positive charges either directly or via a water molecule network. Neutralization of D509 or external protonation is predicted to destabilize these interactions and consequently the relaxed state.

References [+] :

Abbruzzese, Modification of hERG1 channel gating by Cd2+. 2010, Pubmed, Xenbase