|
Figure 1. . (Top) Alignment of amino acid sequences from the end of S5 to the beginning of S6 of HERG, major classes of voltage-gated K (Kv) channels, and KcsA. Transmembrane and pore (P)-loop regions are underlined. S5 and S6 correspond to M1 and M2 of KcsA, respectively. S5-P linker (“turret” of KcsA) and P-S6 linker (P-M2 of KcsA) are marked. Gaps (…) are introduced to improve alignment. Amino acids in the Shaker channel implicated in hydrogen bonding and critical for maintaining the outer mouth in the open state are highlighted by gray shade: E418, W434, and Y445 (Shaker position numbers shown on top). Note that these amino acids are conserved in KcsA and all major classes of Kv channels, but not in HERG. The HERG positions examined here (571–613 of S5-P linker, and 631–638 of P-S6 linker) are marked above its sequence. Positions 583–597 are shown as an insert. H587 in HERG is circled (see text for discussion). (Bottom) Transmembrane topology of one Kv channel subunit. The structure of the unusually long extracellular S5-P linker of HERG (marked by a dotted line and “?”) is the major focus of this study.
|
|
Figure 2. . Effects of cysteine substitution at positions 583 and 585 on the I-V relationship and reversal potential (Erev) of HERG. These experiments are done in 2 mM [K]o and 96 mM [Na]o, using the voltage clamp protocol shown on top of B. Original current traces recorded from WT HERG (A), I583C (B) and W585C (C) before DTT treatment (a), and after DTT treatment (b; 5 mM DTT for 10–15 min, followed by ∼10-min wash) from the same oocytes are shown. Tail current amplitudes are measured, leak-subtracted, and normalized by the maximal outward tail currents measured from the same oocytes. Average I-V relationships are summarized in panels c. Dotted lines denote the Vr range (30 to −60 mV) where the slopes of the I-V relationships are used for the calculation of “rectification factors” (Fig. 6). Arrows in panel c of B point to Erev of I583C before and after DTT treatment.
|
|
Figure 3. . MTSET has no effects on WT HERG (A), and its effects on I583C (B) and W585C (C) are enhanced by DTT pretreatment. In each panel, the upper part illustrates superimposed current traces under the control conditions (IC, thin trace) and after MTSET treatment (IMTSET, thick trace). IC is measured right before MTSET exposure, and IMTSET is measured after MTSET washout. An example of the time points at which IC and IMTSET are sampled is denoted in the time course of A. WT HERG has no DTT-treatment. I583C and W585C are either not DTT-treated (−DTT), or DTT-pretreated (+DTT, 5 mM for up to 4 h). −DTT and +DTT data are from different cells. The lower parts show time courses of changes in the peak amplitude of tail current before, during and after MTSET exposure from the same experiments as shown in the upper parts. Time zero marks the beginning of MTSET (1 mM) superfusion. Open symbols are data points recorded before and after MTSET exposure, whereas closed symbols are those during MTSET exposure. Triangles denote data from cells without DTT treatment, and circles are data from cells pretreated with DTT. (A) WT HERG recorded in 2 mM [K]o, channel is activated by 1-s depolarization pulses to 20 mV, and tail currents are recorded at −80 mV. (B) I583C recorded in 2 mM [K]o, activated by 1-s depolarization to 60 mV, tail currents recorded at −80 mV. (C) W585C recorded in 98 mM [K]o, activated by 1-s depolarization to 20 mV, tail currents recorded at −120 mV.
|
|
Figure 4. . Effects of MTSES on WT HERG (A), I583C (B) and W585C (C). The format of this figure is the same as that in Fig. 3. (A) WT HERG recorded in 2 mM [K]o, activated by 1-s depolarization pulses to 20 mV, tail currents recorded at −80 mV. (B) I583C recorded in 98 mM [K]o, activated by 1-s depolarization to 20 mV, tail currents recorded at −80 mV. (C) W585C recorded in 98 mM [K]o, activated by 1-s depolarization to 20 mV, tail currents recorded at −120 mV.
|
|
Figure 5. . Effects of cysteine substitution at positions 583 and 585 on the voltage dependence of activation. The voltage clamp protocol is shown in the middle of column A. Superimposed original current traces recorded before DTT treatment (panels a), and after DTT treatment (panels b) from the same oocytes are depicted. Dotted lines denote the zero current level. Panels c illustrate 1-s isochronal activation curves averaged from 5–7 oocytes per group. For each oocyte, tail current amplitudes after depolarization to different levels of Vt are measured and normalized by the maximal tail current after Vt to 60 mV. This gives an estimate of the fraction of channels activated at the end of Vt. The relationship between Vt and “fraction activated” is fit with a single (WT, and I583C after DTT) or double (I583C before DTT, and W585C before/after DTT) Boltzmann function: Fraction activated = A1/{1 + exp[(V1 − Vt)k1]} + A2/{1 + exp[(V2 − Vt)/k2]}, where Ai, Vi, and ki are the fraction, half-maximum activation voltage and slope factor of ith Boltzmann component. For single Boltzmann function, i = 1 only. The superimposed activation curves are calculated from appropriate Boltzmann function with mean parameter values.
|
|
Figure 6. . Summary of effects of cysteine substitutions in the S5-P and P-S6 linkers of HERG on the degree of C-type inactivation (A), K+ ion selectivity (B), half-maximum activation voltage (C), response to MTSET (D), and response to MTSES (E). For all panels, white histogram bars denote data from WT HERG and WT-like mutants, gray histogram bars are data from mutants with altered behavior (MUT) after DTT treatment (5 mM, up to 4 h), and open circles are data from MUT without DTT treatment. Dotted lines are drawn across the WT data to facilitate comparison. Data are plotted against WT residues along the abscissa, with selected position numbers marked. The 583–597 segment is boxed. The channel domains (S5-P and P-S6 linkers, with P-loop in between) are also denoted. Asterisk indicates no expression (N573C, K595C, and K638C). (A) The degree of C-type inactivation is measured by a “rectification factor” calculated from the normalized tail I-V relationship as shown in Fig. 2: Rectification factor = [(I+30 − I−60)/90], where I+30 and I−60 are the normalized tail currents at +30 and −60 mV, respectively. (B) K+ ion selectivity is measured by the permeability ratio of K+ to Na+ (PK/PNa). This is calculated from the reversal potential (Erev) measured in 2 mM [K]o and 96 mM [Na]o based on the constant field equation: Erev = RTln[(α[K]o + [Na]o)/(α[K]i + [Na]i)], where α = PK/PNa, [K]i and [Na]i are assumed to be 125 and 10 mM, respectively, R is the gas constant, and T is the absolute temperature. PK/PNa is plotted on a logarithmic scale. Current amplitudes of P605C, Y611C, N633C, and T634C in 2 mM [K]o are too small for a reliable measurement of Erev. (C) Half-maximum activation voltage (V0.5) is obtained by fitting a Boltzmann function to 1-s isochronal activation curves as described for Fig. 5. When there are two Boltzmann components, the V0.5 values reported here correspond to those of the negative (major) component. (D) Effects of MTSET are evaluated by ratio of IMTSET to IC. These current amplitudes are measured as shown in Fig. 3. (E) Effects of MTSES are evaluated in the same manner as described for MTSET, with IC and IMTSES measured as shown in Fig. 4. The negative values of IMTS/IC for G584C, D609C (MTSET but not MTSES), T613C, and S631C indicate a change in tail current direction at −80 mV from outward (IC) to inward (IMTS) (see Fig. 8).
|
|
Figure 7. . Analysis of α-helical periodicity in the S5-P linker based on perturbations of activation gating induced by cysteine mutations (data from DTT-treated channels). The methods of calculating the Fourier transform power spectrum (P(ω) vs. ω) and α-periodicity index (α-PI) are described in materials and methods. (A) Fourier transform power spectrum of |ΔΔGo| values for the segment of 583–597 (value for the nonfunctional K595C is taken from <V>). (B) Windowed α-periodicity index analysis of S5-P linker of HERG. The α-PI values are calculated with a 15-residue sliding window and plotted against WT residues with selected position numbers along the abscissa (putative α-helix “583–597” is boxed).
|
|
Figure 8. . Effects of MTSET and MTSES on the I-V relationship and Erev of four cysteine mutants: G584C (A), T613C (B), S631C (C), and D609C (D). In all cases oocytes have been DTT treated. The recording conditions, voltage clamp protocol, and data analysis are the same as those described for Fig. 2. After control data (open circles) are obtained, the cells are exposed to MTSET (1 mM) or MTSES (1 mM) while the channels are activated by constant pulses from Vh −80 to 20 mV for 1 s once per 30 or 60 s till changes in currents reach a steady state. The MTS reagents are washed out for ≥10 min and then currents after MTS modification (closed circles) are recorded.
|
|
Figure 9. . (A) Summary table of effects of cysteine substitutions in the S5-P and P-S6 linkers on HERG channel function and response to MTSET. -S-S-, cysteine thiol side chains capable of forming disulfide bonds (+), or not (−). ?, data not conclusive. Phenotype, WT-like (W, strong C-type inactivation and high K+ selectivity) or mutant (M, disrupted C-type inactivation and low K+ selectivity), evaluated when thiol side chains are in reduced state (−SH) or in disulfide bonded state (-S-S-). M?, cysteine mutants have mutant behavior but the status of cysteine side chains unclear. /, not applicable because the introduced thiol side chains cannot form disulfide bonds. MTSET effect, decrease in channel conductance (downward arrow) or a switch of channel phenotype from WT-like to mutant (S). –, no effect. The 583–597 segment is boxed. G584C, W585C, G590C, I593C, T613C, and S631C are highlighted by different symbols below the table. (B) Plots of α-helical wheel (left) and α-helical net (right) of the 583–597 segment. Position numbers are labeled. Hydrophobic side chains are shaded gray. Hydrophilic side chains and glycine have no shade. Squares denote high-impact positions, while circles denote intermediate-impact positions. (C) Proposed structural model of HERG's pore domain. (Left) The helix formed by residues 583–597 is represented by a short cylinder. Symbols correspond to those in A and represent G584, W585, G590, I593, T613, and S631. For clarity, only one each of 583–597 helix, T613 and S631 are drawn for the channel. The HERG channel pore is drawn as a right side up “teepee” to reflect the following points: (1) The outer mouth of HERG may be narrower than that of Shaker or KcsA due to (a) a lack of important hydrogen bonds that can stabilize the outer mouth in the open state, and (b) the insertion of extra sequences from the S5-P linkers. (2) The inner mouth of HERG may be wider than that of Shaker or KcsA because it can accommodate drug molecules of large sizes (Mitcheson et al., 2000). (Right) An enlarged view of the boxed area. The model has the 583–597 helix orientated parallel to the plane of the cell membrane, with its NH2 end situated close to the pore entrance, so that G584 (star), like the two positions at the two ends of the P-loop, T613 (triangle) and S631 (circle), sits at the edge of extracellular pore entrance, with residues W585, G590, and I593 (diamonds) capable of interacting with a bound scorpion peptide toxin, ErgTx (Pardo-Lopez et al., 2002). The 574–582 segment forms an extracellular loop of low-impact positions. The 598–612 segment returns from the carboxyl end of the 583–597 helix to the pore, and along the way makes contacts with the outer surface of the channel at high- and intermediate-impact positions (604 and 609, respectively).
|
