Zhang M et al. (2004), Gating charges in the activation and inactivati...

XB-ART-2736

J Gen Physiol 2004 Dec 01;1246:703-18. doi: 10.1085/jgp.200409119.

Show Gene links Show Anatomy links

Gating charges in the activation and inactivation processes of the HERG channel.

Zhang M , Liu J , Tseng GN .

???displayArticle.abstract???
The hERG channel has a relatively slow activation process but an extremely fast and voltage-sensitive inactivation process. Direct measurement of hERG's gating current (Piper, D.R., A. Varghese, M.C. Sanguinetti, and M. Tristani-Firouzi. 2003. PNAS. 100:10534-10539) reveals two kinetic components of gating charge transfer that may originate from two channel domains. This study is designed to address three questions: (1) which of the six positive charges in hERG's major voltage sensor, S4, are responsible for gating charge transfer during activation, (2) whether a negative charge in the cytoplasmic half of S2 (D466) also contributes to gating charge transfer, and (3) whether S4 serves as the sole voltage sensor for hERG inactivation. We individually mutate S4's positive charges and D466 to cysteine, and examine (a) effects of mutations on the number of equivalent gating charges transferred during activation (z(a)) and inactivation (z(i)), and (b) sidedness and state dependence of accessibility of introduced cysteine side chains to a membrane-impermeable thiol-modifying reagent (MTSET). Neutralizing the outer three positive charges in S4 and D466 in S2 reduces z(a), and cysteine side chains introduced into these positions experience state-dependent changes in MTSET accessibility. On the other hand, neutralizing the inner three positive charges in S4 does not affect z(a). None of the charge mutations affect z(i). We propose that the scheme of gating charge transfer during hERG's activation process is similar to that described for the Shaker channel, although hERG has less gating charge in its S4 than in Shaker. Furthermore, channel domain other than S4 contributes to gating charge involved in hERG's inactivation process.

???displayArticle.pubmedLink??? 15545400
???displayArticle.pmcLink??? PMC2234031
???displayArticle.link??? J Gen Physiol
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: kcnh1 kcnh2 rps3a tbx2

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

	Figure 1. . Basic structure design of voltage-gated K (Kv) channels and unique features in the hERG channel. (A) Two-dimensional transmembrane topology of a Kv channel pore-forming subunit, highlighting negative charges in S1, S2, S3a, and S3b (−2, −4, and −5 are conserved in all Kv channels, while −1, −3, and −6 are unique to EAG family members including hERG), and positive charges in S4 (+1 to +6 for Shaker/0 or + for EAG and hERG). The gray cylinder in the extracellular S5-P linker denote a putative α-helix formed by residues 583–594 in the hERG channel. (B) Alignment of amino acid sequences of S1, S2, S3a, S3b, and S4 of KvAP, Shaker, hERG, and b(ovine)EAG. Channel domains are denoted on top. Charged residues are boxed, and their generic numbers are marked below (−1 to −6, +1 to +6). (C) Alignment of amino acid sequences from the end of S5 to the beginning of S6 of KvAP, Shaker, hERG, and bEAG. Channel domains are denoted below. Asterisks connected by horizontal lines on top denote putative H bond donor/acceptor pairs around the outer mouth in Shaker and KvAP (Doyle et al., 1998; Larsson and Elinder, 2000), which are missing in the hERG and bEAG. Gray shade highlights hERG residues 583–594 that form a putative α-helix. The hERG residues that are critical for the channel's fast inactivation process are highlighted in bold (Liu et al., 2002). Boxes indicate bEAG residues that are different from hERG at these critical positions. (D) hERG residues examined in this study and the corresponding residues in the Shaker channel.
	Figure 2. . Effects of substituting S4's lysine (K) or arginine (R) with cysteine (C) on the channel function. (A) Original current traces from WT and mutants (specified on the left). The currents were elicited using the voltage clamp protocol diagrammed on top. From a negative holding voltage (Vh, −80 to −120 mV) 1-s test pulses to different test voltages (Vt) in 10-mV increments were applied once every 15 s. These were followed by repolarization to Vr. For each channel, the range of Vt (selected to cover the voltage range of channel activation) and the Vr level are marked adjacent to the current traces. The test pulses were preceded by a brief (10-ms) hyperpolarizing prepulse (−20 mV from Vh) whose purpose was to estimate the background channel conductance at Vh. Open arrows point to significant prepulse currents in K525C and K538C, reflecting channel opening at Vh. (B) Isochronal (1-s) activation curves of WT and mutant channels. For each cell, peak amplitudes of tail currents during Vr (Itail, as shown in A) were normalized by the maximum tail current (Imax) obtained from the same cell after a strong Vt that maximally activated the channels. The relationship between normalized tail currents (Itail/Imax) and Vt was fit with a simple Boltzmann function to estimate the half-maximum activation voltage (V0.5) and slope factor (k): Itail/Imax = 1/[1 + exp((V0.5 − Vt)/k)]. To facilitate comparison, WT data are shown as gray symbols in all the panels for mutant channels (n = 6–14 each).
	Figure 3. . Cysteine substitutions of S4's positive charges had little effect on the inward rectification property and reversal potential of hERG. (A) Original current traces were elicited by the voltage clamp protocol diagrammed on top. From Vh −80 or −90 mV, a strong depolarization pulse to +60 mV for 0.2 s was used to activate and inactivate the channels. This was followed by repolarization pulses from +30 to −120 mV in 10-mV increments (current traces shown were from +20 to −120 mV in 20-mV steps) during which channels rapidly recovered from inactivation and then deactivated. Recordings were made in 2 mM [K]o (estimated EK −105 mV, assuming [K]i 125 mM). (B) Current–voltage relationships from experiments as shown in A. The peak or plateau amplitudes of tail currents were normalized by the maximum outward tail current in each cell (occurring at Vr −50 to −30 mV) and averaged. Gray open circles and black circles are for WT and mutant, respectively (n = 6–14 each).
	Figure 4. . Estimating the number of equivalent gating charges involved in channel activation (za) using the limiting slope method. (A) Original current traces of WT-hERG recorded in 20 mM [K]o. Tail currents (measured at −120 mV) were elicited by two protocols. (1) 1-s pulses were applied in 10-mV increments to Vt covering the whole voltage range of channel activation (based on Fig. 2 B). These are illustrated by the top family of current traces. (2) For a Vt range from below the activation threshold to ∼20 mV above the threshold, long (10-s) pulses were applied in 2-mV increments to gain a high resolution analysis of increase in channel open probability (Po) at Vt around the activation threshold. These are illustrated by the middle family of current traces, with the boxed area enlarged in the bottom family of current traces. The 10 μA calibration applies to the top and middle families of current traces, and the 1 μA calibration applies to the bottom family of current traces depicted in the box. The peak amplitudes of tail currents were normalized by the maximum tail current following the most depolarized Vt to estimate Po at the end of preceding depolarizing pulses. The same recording conditions and similar voltage clamp protocol were used for all the mutant channels, with Vt levels adjusted depending on the voltage range of activation. (B) Po values are plotted on a logarithmic scale against Vt for WT (gray open circles) and mutants (black open circles, mutant type specified in each panel). The relationships between Ln(Po) and Vt were subjected to linear regression analysis to estimate the limiting slope at low Po levels (Po ∼ 10−3). The value of za was calculated according to za = slope*(RT/F), where RT/F = 25 mV. The lines superimposed on the data points are calculations based on the mean limiting slopes. (C) Summary of za for WT and mutant channels (n = 3–6 each, mean + SEM).
	Figure 5. . Efficacy of intracellular application of hydrophilic reagents using the oocyte injection device during voltage clamp. (A) Original current traces recorded from an experiment on WT hERG at time points a, b, and c corresponding to those denoted along the time course in B. (B) Average time course of changes in WT-hERG peak tail current amplitude (n = 4). Current was elicited by 1-s depolarization pulses to +20 mV once every 30 s and tail current was recorded at −80 mV. After control data were obtained, the oocyte was impaled with a fine-tipped injecting pipette filled with MTSET (100 mM) solution (impale, gray symbols). After confirming current stability, MTSET was injected (10 nl, estimated cytoplasmic concentration = 2 mM, assuming oocyte volume = 0.5 μl, black symbols/light gray shade). 10 min after MTSET injection, the injecting pipette was pulled out, refilled with TEA (100 mM) solution, and reimpaled into the oocyte. TEA was injected as described above (10 nl, cytoplasmic concentration = 2 mM, open triangles/dark gray shade). The curve superimposed on the open triangle data points after the initial sigmoid phase represents single exponential fit to the mean time course of WT-hERG suppression by TEAi, with time constant (τ) marked.
	Figure 6. . Sidedness and voltage dependence of MTSET accessibility to cysteine side chain introduced into position 525. (A–C) Effects of MTSETo on K525C depended on the voltage clamp protocol during MTSETo application. Shown in each panel is a time course of changes in the half-time (T1/2) of K525C deactivation measured at −120 mV. In A and B, the membrane voltage was held at a constant Vh (−80 or 0 mV, denoted by the horizontal bars) for 10 min. MTSETo 1 mM was applied during the first 5 min (denoted by gray shades) and washed out for 5 min. Afterwards, the Vh was returned to −90 mV and the pulsing was resumed to monitor the deactivation kinetics. Insets, K525C tail currents recorded before (thin solid traces) and after (dashed trace or thick solid trace) MTSETo treatment. In C, the membrane was constantly pulsing from Vh −90 to +40 mV for 1 s once every 30 s before, during (gray shade), and after MTSETo exposure. The additional horizontal axis for data points during MTSETo exposure denotes the cumulative open-state exposure (1-s exposure to 1 mM MTSETo at +40 mV once every 30 s). The curve superimposed on the data points represents single exponential fit with τ = 4 mM·s. (D) Time course of changes in T1/2 of K525C deactivation when the oocyte was impaled and injected with MTSET (MTSETi 2 mM, dark gray shade). 18 min after MTSET injection, the membrane voltage was held at 0 mV for 10 min (denoted by the black horizontal bar), during which the oocyte was exposed to MTSETo 1 mM for the first 5 min (light gray shade) followed by wash out for 5 min. The tail current deactivation rate after MTSETo treatment was monitored. In all panels, the coordinates are T1/2 values normalized to the control values, and the abscissas are time after MTSET application to the bath (A–C) or into the oocyte (D).
	Figure 7. . Sidedness and state dependence of MTSET accessibility to cysteine side chains introduced into positions 528, 531, 534, 537, and 538 (mutant types marked on top). For each channel, the top graph (A) is an average time course of changes in current amplitude before, during (gray shades), and after exposure to extracellular MTSET (1 mM). For R528C, the membrane voltage was held at +20 mV for 10 min and MTSETo was applied during the first 5 min and then washed out for 5 min. For the other mutants, the membrane was pulsing from Vh (−80 mV for all, except K538C, Vh = −90 mV) to Vt (as specified in the insets of lower graphs) for 1 s every 30 s. The solid circles during MTSETo exposure denote closed state–preferred exposure because the channels spent 29 s of the 30-s cycle in the closed states. The bottom graphs (B) show the average time courses of current amplitude before (open triangles) and after intracellular injection of MTSET (estimated initial cytoplasmic concentration 2 mM, gray shades). The closed circles denote closed state–preferred exposure; channels stayed in the closed states at Vh −80 or −90 mV and only activated by 0.5-s test pulses to Vt (as denoted in the insets) once every 30 s. The open circles denote open/inactivated state –preferred exposure; channels stayed in open/inactivated states for 20 s at denoted Vt and only spend 10 s back to Vh once every 30 s. The curves superimposed on the data points are single exponential fits. Insets, original current traces before (thin traces) and 8–16 min after MTSETi injection using the closed state–preferred exposure protocol (thick traces). Arrows indicate the changes in peak tail current amplitudes.
	Figure 8. . Accessibility of cysteine side chain introduced into position 466 in S2 to MTSETi and MTSETo. (A and B) Time courses of changes in D466C peak tail current amplitude (elicited by 1-s pulse to +60 mV followed by repolarization to −60 mV) normalized to the control amplitude (open circles). Gray circles indicated data points after impalement with an injecting pipette. Black circles denote data points after MTSET injection (estimated initial cytoplasmic concentration 2 mM). Approximately 15 min after injection and in the continuous presence of intracellular MTSET (light gray shade), membrane voltage was held at −80 (A) or +20 (B) mV for 10 min (black horizontal boxes), during which the oocytes were exposed to 1 mM MTSETo for the first 5 min (dark gray shade) followed by washout. Open triangles depict data points after the washout of MTSETo. Insets, superimposed tail currents recorded from the same experiments as shown in the main graphs at time points a–c. (C) Time course of changes in D466C tail current amplitude in response to MTSETi (estimated initial concentration 2 mM). To test whether 466C modification by MTSETi was state dependent, two voltage clamp protocols were used: (1) closed state–preferred MTSETi exposure, short (0.5 s) depolarization pulses to +40 mV applied once every 60 s, and (2) open/inactivated state–preferred exposure, long (10 s) depolarization pulses to +40 mV applied once every 15 s. Data were averaged from five and six experiments, respectively.
	Figure 9. . Effects of cysteine substitution of S4's positive charges on the kinetics and voltage dependence of fast inactivation in the hERG channel. (A) Original current traces elicited by the voltage clamp protocol diagrammed on top; 0.2-s depolarizing pulse to +60 mV (except R531C for which depolarization was to +80 or +100 mV) was used to activate and then inactivate the channels. This was followed by a short repolarizing pulse (15 ms to −80 mV) to allow channels to recover from inactivation without significant deactivation. The membrane voltage was then depolarized to Vtest ranging from +40 to −20 mV in 10-mV increments, during which channels reinactivated. The time course of reinactivation could be well described by a single exponential function. Shown are current data (as gray circles), superimposed with the fitted single exponential function (black traces). (B) Summary of time constants (τ) of inactivation. Values of τinactivation are plotted on a logarithmic scale. The relationship between Ln(1/τinactivation) and Vtest was subject to linear regression analysis to estimate the slope. The slope is used to calculate zi, according to Eq. 3 in the text. *, P < 0.001 K525C vs. WT; #, P < 0.05 R537C vs. WT. Experiments were done in 2 mM [K]o.
	Figure 10. . (Top) Alignment of S4 amino acid sequences of Shaker and hERG. The positive charges are denoted by generic numbers below. Boxes highlight voltage-sensing (gating) charges in each channel. Gray shade highlights the Shaker arginine that is absent in hERG. (Bottom) Diagrams illustrating putative movements of Shaker's S4 and hERG's S2 and S4, relative to the membrane barrier, during channel activation. The Shaker diagram is based on (Starace and Bezanilla, 2004), with the first four positive charges switched from internally exposed positions in closed states to externally exposed in the open states. Although E293 in the Shaker S2 may contribute to gating charge transfer, it is not included in the diagram because there is no data on the accessibility of 293C to MTSETo or MTSETi. The hERG diagram is based on data from the present study, with one exception. Although we cannot detect any MTSETo accessibility of 528C (+3 in the diagram) in the open state, this side chain is accessible to external p-chloromercuibenzene sulfonate (pCMB, a smaller thiol-modifying reagent than MTSET) in the open state (Mitcheson, J.S., personal communication). The crevice around S2 is deep in the hERG channel, so that cysteine side chain at position −3 is readily accessible to extracellular MTSETo (Liu et al., 2003). Cysteine at position −4 (466C) is accessible to MTSETo in the closed state but not in the open state. MTSETi also modifies 466C, although the effect is much less than that of MTSETo in the closed state, and there is no clear state dependence in MTSETi modification of 466C. These observation are depicted by a narrow inwardly accessible crevice around S2 at the −4 level in both closed and open states.

References [+] :

Aggarwal, Contribution of the S4 segment to gating charge in the Shaker K+ channel. 1996, Pubmed, Xenbase