Xu X et al. (2008), KCNQ1 and KCNE1 in the IKs channel complex make...

XB-ART-38160

J Gen Physiol 2008 Jun 01;1316:589-603. doi: 10.1085/jgp.200809976.

Show Gene links Show Anatomy links

KCNQ1 and KCNE1 in the IKs channel complex make state-dependent contacts in their extracellular domains.

Xu X , Jiang M , Hsu KL , Zhang M , Tseng GN .

Abstract
KCNQ1 and KCNE1 (Q1 and E1) associate to form the slow delayed rectifier I(Ks) channels in the heart. A short stretch of eight amino acids at the extracellular end of S1 in Q1 (positions 140-147) harbors six arrhythmia-associated mutations. Some of these mutations affect the Q1 channel function only when coexpressed with E1, suggesting that this Q1 region may engage in the interaction with E1 critical for the I(Ks) channel function. Identifying the Q1/E1 contact points here may provide new insights into how the I(Ks) channel operates. We focus on Q1 position 145 and E1 positions 40-43. Replacing all native cysteine (Cys) in Q1 and introducing Cys into the above Q1 and E1 positions do not significantly perturb the Q1 channel function or Q1/E1 interactions. Immunoblot experiments on COS-7 cells reveal that Q1 145C can form disulfide bonds with E1 40C and 41C, but not E1 42C or 43C. Correspondingly, voltage clamp experiments in oocytes reveal that Q1 145C coexpressed with E1 40C or E1 41C manifests unique gating behavior and DTT sensitivity. Our data suggest that E1 40C and 41C come close to Q1 145C in the activated and resting states, respectively, to allow disulfide bond formation. These data and those in the literature lead us to propose a structural model for the Q1/E1 channel complex, in which E1 is located between S1, S4, and S6 of three separate Q1 subunits. We propose that E1 is not a passive partner of the Q1 channel, but instead can engage in molecular motions during I(Ks) gating.

PubMed ID: 18504315
PMC ID: PMC2391252
Article link: J Gen Physiol
Grant support: [+]

Species referenced: Xenopus laevis
Genes referenced: kcna2 kcne1 kcnq1 nr4a1 s100a10 slc26a4.3 tbx2 ttn

Article Images: [+] show captions

	Figure 1. 2-D diagram of KCNQ1 and KCNE1 subunits. For Q1, transmembrane segments S1–S6, P-loop, voltage-sensing domain (VSD), and pore-domain (PD) are marked. Numbers in ovals denote positions of native cysteine (Cys) in Q1 that have been replaced by Ala in the Q1(-Cys)-WT construct. Ile at position 145 is replaced by Cys to create the Q1(-Cys)-I145C construct. Region of interactions identified in previous studies is outlined by a dotted rectangle, and specific interacting residues are linked by solid lines (Tapper and George, 2001; Melman et al., 2002; Panaghie et al., 2006). Potential interaction between the extracellular end of Q1 S1 and the extracellular end of E1 transmembrane domain (TMD) investigated in this study is denoted by a dotted line.
	Figure 2. Validating the oocyte expression system for study of Q1/E1 interaction. Each oocyte is injected with cRNA for E1-WT alone (3 ng/oocyte, to monitor currents through oocyte endogenous xQ1 associated with E1-WT) (Sanguinetti et al., 1996), Q1-WT alone (10 ng/oocyte), or Q1-WT + E1-WT (10 and 3 ng per oocyte, respectively, cRNA molar ratio 1:1). Membrane currents are recorded on days 3 and 5 after cRNA injection. (A) Representative current traces elicited by the diagrammed protocol from oocytes injected with marked cRNA(s). (B) Mean current amplitudes at +60 mV from the three groups of oocytes. Note that the E1-alone currents are almost invisible at the same gain suitable for the other two groups, and the SE bar is buried in the line of the histogram bars. In all the experiments reported in this study, we inject the same amount of cRNA for Q1 and/or E1 variant per oocyte as listed above, including E1-WT alone, and currents are recorded on days 4–5 (unless otherwise noted). To quantify and compare current amplitudes in oocytes expressing Q1 plus E1 vs. Q1 alone, data from the former group are corrected for mean current amplitude of oocytes injected with E1-WT alone (representing contribution from oocyte xQ1/E1-WT in the same batch of oocytes).
	Figure 3. Q1 and E1 mutations investigated in this study preserve Q1 channel function and Q1/E1 interactions. Current traces are recorded from three batches of oocytes, expressing Q1-WT (top), Q1(-Cys)-WT (middle), or Q1(-Cys)-I145C (bottom), alone or with E1 WT or mutants at 1:1 cRNA molar ratio. The type of Q1 variant is marked on the left of each row, while the type of E1 variant is marked on the top of each column. In the case of Q1(-Cys)-I145C coexpressed with Cys-substituted E1 mutants, oocytes are treated with DTT (10 mM for 10 min, or 25 mM for 5 min) and thoroughly rinsed before current recording. The voltage clamp protocol is diagrammed in the inset of top row. To illustrate the effects of E1 variants on the channel function of each of the Q1 variants, in each row the same representative current trace recorded from an oocyte expressing respective Q1 variant alone is shown as a gray trace, superimposed with current traces from oocytes coexpressing E1 variants. To illustrate the current enhancing effects of E1 variants on the Q1 channels, the current traces are scaled to reflect the average ratio of current amplitudes of Q1+E1 vs. Q1 alone in the same batch of oocytes (the exact numbers of fold-increase in current amplitudes are listed in Table II). Q1(-Cys)-I145C coexpressed with E1-G40C manifests a unique time-independent component (Istep) and a time-dependent component (It-d). To ascertain that this is not due to interference from oocyte leak conductance, current traces in the third row are HMR-sensitive current (HMR ΔI = control current – current recorded in the presence of 10 uM HMR; Wu et al., 2006a). Dotted lines denote the zero current level.
	Figure 4. Testing disulfide bond formation between Q1 145C and Cys introduced into E1 positions 40–43. (A) COS-7 cells are transfected with cDNA(s) listed above (leftmost lane: no cDNA negative control). 48 h after transfection, cells are treated with NEM (to protect free thiol groups) followed by whole cell lysis and solubilization. Whole cell lysates are run on 7.5% and 14% nonreducing SDS gels. The former is probed for Q1 (with a V5 mAb targeting V5 epitope attached to the carboxyl end of Q1) and the latter is probed for E1 (with E1 pAb). Size marker positions are listed on the left, and sizes of Q1- or E1-specific bands are noted on the right (in kD). (B) Whole cell lysates prepared from COS-7 cells transfected with cDNA(s) listed above are each divided into two aliquots. One aliquot is treated with DTT (10 mM, room temperature, 10 min) while the other is incubated with buffer under the same conditions without DTT. All samples are run on a nonreducing SDS gel and probed for Q1 with V5 mAb.
	Figure 5. (A) Confirming the presence of E1 G40C in the 80- and 75-kD Q1-positive bands detected in whole cell lysate of COS-7 cells coexpressing Q1(-Cys)-I145C and E1-G40C. COS-7 cells are transfected with the cDNA(s) listed on top. An aliquot of the whole cell lysates is loaded onto the “direct input” lanes. The other aliquot is subjected to immunoprecipitation with V5 mAb. The immunoprecipitates are DTT treated and loaded onto the IP lanes. The Q1 and E1 are probed using a goat pAb against Q1 (C-20) and a rabbit pAb against E1. (B) Testing N-glycosylation in free E1 proteins and disulfide-linked Q1/E1 complex. Whole cell lysates from COS-7 cells coexpressing Q1(-Cys)-I145C and E1-G40C are treated with PNGase F or with buffer under the same conditions (“+” and “−” PNGase F) for 48 h at 37°C. The samples are fractionated by nonreducing SDS-PAGE and probed for Q1 and E1.
	Figure 6. Effects of DTT treatment on the voltage dependence of activation of different combinations of Q1/E1 variants. The type of Q1 and E1 variants is marked on top of each panel: (A) E1-G40C coexpressed with Q1(-Cys)-I145C or Q1(-Cys)-WT, (B) E1-K41C coexpressed with Q1(-Cys)-I145C or Q1(-Cys)-WT, and (C) Q1(-Cys)-I145C coexpressed with E1-WT, E1-L42C, or E1-E43C. Voltage clamp protocol (diagrammed in the right bottom panel) and data analysis are the same as those described for Table I. The 2-s isochronal activation curves before and after DTT treatment are denoted by open and closed circles, respectively.
	Figure 7. (A and B) Testing the state dependence of appearance of a constitutive component in oocytes expressing Q1(-Cys)-I145C/E1-G40C after DTT washout. Data from two experiments are shown, with time courses of appearance of the constitutive component (Istep) shown on the left, and superimposed current traces shown on the right. In both experiments, before DTT treatment, the oocytes manifest slowly activating currents without any constitutive component (dotted traces in right panels). The oocytes are exposed to DTT 10 mM for 10 min (gray shade, time axis breaks between start of DTT application and last few data points in DTT). In the presence of DTT, the current amplitudes increase but the currents remain slowly activating without a significant constitutive component (gray traces in the right panels). In A, during DTT washout the oocyte is repetitively pulsed from Vh −100 to +60 mV for 2 s, followed by a step to −60 mV for 2 s, once every minute (voltage clamp diagram in inset). The appearance of Istep follows a single exponential time course, with a time constant of 3.1 min (black thick curve superimposed on the open circle data points). In B, during DTT washout the oocyte is held at Vh −100 mV without pulsing for 15 min. The current induced by the first pulse after resuming pulsing is highlighted by the thick black current trace (right) and the small Istep component during the first pulse is highlighted by a solid black circle (left). The dotted curve is a scaled single exponential time course similar to that shown in Fig. 7 A. (C) Testing whether membrane depolarization by elevating [K]o facilitates disulfide bond formation between Q1 145C and E1 40C. COS-7 cells are transfected with cDNA(s) listed on the top two rows. Cells are incubated in medium containing 5 mM [K]o for 48 h. Cells in lanes 4 and 5 are treated with 10 mM DTT for 10 min to break disulfide bonds, followed by thorough rinsing to remove DTT. Cells in lane 4 are incubated in 5 mM [K]o medium, while cells in lane 5 are incubated in 100 mM [K]o medium for 10 min. All cell preparations are treated with NEM to protect free thiol groups, followed by whole cell lysis. Whole cell lysates are run on nonreducing SDS gel and probed for Q1 with V5 mAb.
	Figure 8. Estimating the rate of disulfide bond formation between Q1 I145C and E1 40C after DTT treatment. The experiment is divided into four phases. During the control and DTT (10 mM) washin phases, the membrane voltage is stepped from Vh −100 to +60 mV for 2 s and then to −60 mV for 2 s once every 60 s. The current amplitude at the beginning and end of the step to +60 mV (Istep and Itotal, respectively) are monitored. After the DTT effects reach a steady state, the DTT washout phase begins during which the membrane is held at −100 mV for 15 min without depolarizing pulses. Then the ‘post-DTT’ phase begins that consists of two parts. During the first part, short (0.2 s) pulses to +60 mV applied once every 15 s are used to monitor the development of the constitutive component as indicated by the increase in Istep. Once the constitutive component reaches a steady state, the voltage clamp protocol is switched back to the 2-s depolarization pulses applied once every 60 s (as described above) to monitor the stability of Itotal and Istep. (A) Average time course of changes in Itotal and Istep. Data are pooled from four experiments. In each experiment, the amplitudes of Itotal and Istep are normalized by their final values at the end of the experiment to allow data averaging. The black curve superimposed on data points obtained during the post-DTT 0.2-s pulse period represents an exponential fit to the time course of development of the constitutive component. The time constants are converted to cumulative depolarization time at +60 mV (0.85 ± 0.14 s, n = 4). (B) Original current traces from one representative experiment. Note the change in time calibration for the 0.2-s pulse current traces. The horizontal line denotes zero current level.
	Figure 9. (A) Time course of changes in current amplitudes of Q1(-Cys)-I145C/E1-K41C before, during, and after DTT (10 mM) exposure. Data are averaged from six experiments. Oocyte membrane is pulsed from Vh −100 mV to +60 mV for 2 s, followed by repolarization to −60 mV for 2 s, once every 60 s. Current amplitudes at the end of the +60-mV step is measured and normalized by the control current before DTT exposure. The duration of DTT exposure is denoted by gray shade. The time course of current reduction after DTT washout is fit with a single exponential function (smooth curve superimposed on data points after DTT). The time constant (τ) is converted to cumulative time at Vh −100 mV (when channels are in the resting state). Mean and SEM values of τ are listed in inset. (B) Time course of changes in current amplitude of Q1(-Cys)-WT expressed alone before, during (gray shade), and after exposure to 10 mM DTT. Since there is no Cys in the Q1 channel, this experiment reveals a direct current-suppressing effect of DTT on the Q1 channel. The rapid onset of DTT effect and rapid/complete reversal reflect the time course of DTT equilibration in the bath solution during wash-in and rate of DTT clearance during washout.
	Figure 10. A structural model of E1 association with the Q1 channel, and dynamic interactions between the extracellular ends of their transmembrane domains. (A) Top view of the arrangement of transmembrane segments in a tetrameric Q1 channel based on the crystal structure of Kv1.2 (2A79.pdb). The four Q1 subunits are color coded light blue, pink, white, and light gray. S1–S4 of the voltage-sensing domains (VSDs), S5–S6 of the pore domains (PDs), and the S4–S5 linkers (rectangles connecting S4 and S5) are marked. The E1 transmembrane domain is in a putative “KCNE-binding pocket,” where E1 makes contacts with S4, S6, and S1 of three separate Q1 subunits. (B) Top view of a tetrameric Kv1.2 structure (using the 2A79.pdb coordinates), color coded in the same manner as in A. L290 at the extracellular end of S4 of the light blue subunit and UNK50 at the extracellular end of S1 of the pink subunit that serve as surrogates for A226 and I145 in Q1, respectively, are highlighted by the orange space-filled model. A291 at the extracellular end of S4 in the light blue Q1 subunit is also shown as yellow space-filled model. S6 of the light gray subunit is marked. The E1 transmembrane domain is signified by the open gray circle. The close-up view of the region demarcated by white dashed lines is shown in C and D. (C) In the activated state of the Q1/E1 channel complex, the E1 helix is oriented in such a manner that E1 position 40 faces toward Q1 position 145 (UNK50 in 2A79), and E1 position 43 faces toward Q1 position 226 of the adjacent subunit (L290 in 2A79). (D) Upon channel deactivation, the E1 helix is postulated to rotate ∼100° counterclockwise so that in the resting state E1 position 41 faces toward Q1 position 145. In C and D, the Cβ–Cβ distances are deduced by the rates of disulfide formation (Fig. 8 A and Fig. 9 A) (Nakajo and Kubo, 2007) and the experimentally determined relationship between the rate of disulfide formation and the distance between surface Cys pair (Careaga and Falke, 1992) (more in text).

References [+] :

Abbott, A superfamily of small potassium channel subunits: form and function of the MinK-related peptides (MiRPs). 1998, Pubmed