Clinically Relevant KCNQ1 Variants Causing KCNQ1-KCNE2 Gain-of-Function Affect the Ca2+ Sensitivity of the Channel

Figures

• 

  Figure A1. A Markov model considering the Ca2+ sensitivity of the KCNQ1 WT channel. (A) The five-state kinetic model in the lower part (white background) of the scheme applies to homomeric KCNQ1 channels with saturating concentrations of Ca2+ and CaM (high Ca-CaM). To account for the effects of reduced Ca2+, three closed states (*) were added without direct access to the open states IO and AO (low Ca-CaM; blue background). From IC*, channels have to pass through AC* to enable channel opening. (B) Model parameter used to describe WT KCNQ1 channel behavior with low and high intracellular Ca-CaM. k5 and k8 determine the access to *states; the ratio of k5/(k5+k6) mirrors the amount of current reduction induced by decreased levels of calcified CaM. (C) With low Ca-CaM, the potential- and time-dependent accumulation of channels in the AC state (black traces) is significantly reduced due to the high occupancy of the IC* state (red traces). “R”, “I” and “A” indicate the resting, intermediate or activated positions of the voltage sensor; “C” and “O” indicate the closed or open states of the channel. Simulated KCNQ1 current traces resulting from this model are shown in Figure S1 (deactivation…

• 

  Figure A2. Structure models of the KCNQ1-CaM complex in the PIP2-free and the PIP2-bound state and location of amino acid residues R116, V185 and P369. (A,B) Visualization of KCNQ1 (transmembrane helices S1 to S4 in turquoise, S4–S5 linker in light green, S5 in yellow, S6 and helix A in orange and helices B and C in red) and CaM (in grey) in the tetrameric KCNQ1-KCNE3-CaM complex in the PIP2-free state (A, based on pdb 6V00) and in the PIP2-bound state (B), based on pdb 6V01) and orientation in the membrane (gray lines). KCNE3 has been removed for clarity. Amino acid residues R116, V185 and P369 are highlighted in magenta and shown as sticks in a single KCNQ1 subunit. On the left, KCNQ1 and CaM are shown as a tetrameric channel complex. On the right, one KCNQ1 subunit with the highlighted amino acid residues R116, V185, and P369 and one CaM subunit are shown. The structural rearrangement of S6 and helix A, which form two helices connected by a loop in the PIP2-free state (A, right panel) to a continuous straight helix in the PIP2-bound state, is shown (B, right panel). The insets in the middle show a magnification to highlight the location of R116, V185 and P369, as well as the close…

• 

  Figure 1. KCNQ1 variant V185M leads to high current densities in cells expressing KCNQ1-KCNE2 (Q1E2) heteromers. Whole-cell membrane currents were recorded in CHO cells expressing wild-type (WT) or V185M KCNQ1 channels alone or together with KCNE2. (A) Typical current traces recorded with a 500 ms test pulse to 40 mV. (B) Distribution of current densities determined with test pulses shortly after break-through. Asterisks indicate significant differences between groups; *** p < 0.001. (C) Typical families of current traces obtained with the indicated standard double pulse protocol, consisting of a 2 s variable P1 pulse and a constant P2 pulse to 0 mV. (D) Current amplitudes (means ± SEM) at the end of the 2 s P1 pulse plotted against P1 potential. (E) Conductance–voltage (GV) relation: normalized instantaneous P2 current amplitudes (means ± SEM) plotted against P1 potential. Numbers of experiments in (B,D,E) are given in parentheses; scale bars in (A,C) denote 0.5 nA and 0.5 s; standard extracellular (ECS) and intracellular solution (ICS). WT: WT KCNQ1; V185M: V185M-KCNQ1.

• 

  Figure 2. All three disease-associated KCNQ1 variants lead to reduced early current run-down. (A) Scheme of the membrane topology of a KCNQ1 subunit with the location of the three studied amino acid substitutions indicated as colored dots. Transmembrane segments S1–S4 constitute the voltage-sensing domain, and S5 and S6 the pore domain. The cytoplasmic C-terminus harbors the proximal helices A and B and the distal helices C and D. (B) Whole-cell membrane currents were recorded in CHO cells expressing WT or mutant KCNQ1 channels using standard ECS and ICS. The time course of -current run-down was traced with a 500 ms test pulse to 40 mV every 5 s. Normalized current amplitudes (means ± SEM) for V185M, R116L and P369L are plotted together with the WT data of closely paralleled experiments. (C) Mean relative current amplitudes 3.5 to 4 min after the first measurement in the whole-cell configuration. Asterisks indicate significant differences to the combined WT data; *** p < 0.001. (D) Typical families of current traces, recorded after a 5 min tracing period with the indicated standard double pulse current–voltage (IV) protocol. Scale bars denote 250 pA and 0.5 s. (E) Mean (± SEM) normalized…

• 

  Figure 3. In cells co-expressing calmodulin (CaM), high intracellular Ca2+ abolishes early current run-down and characteristic WT KCNQ1 properties. Whole-cell recordings were performed with different Ca2+ concentrations of ICS in CHO cells expressing the KCNQ1 channels, together with CaM. (A) Time course of current amplitudes traced with a 500 ms test pulse to 40 mV. Mean normalized current amplitudes for WT and V185M-KCNQ1 channels from experiments with standard low-Ca2+ ICS and high-Ca2+ (5 µM) ICS. Numbers of experiments are given in the bar plot, which shows mean (+SEM) relative current amplitudes after 3.5 to 4 min. Asterisks indicate significant differences; *** p < 0.001. (B) Example of WT KCNQ1 current traces, recorded in a cell co-expressing CaM > 5 min in whole-cell configuration using 5 µM-Ca2+ ICS. Scale bars denote 1 nA and 500 ms. (C) Mean (± SEM; n = 7) normalized instantaneous P2 current amplitudes (GV relation, filled circles) and current amplitudes measured at the end of the 1 s P2 pulse to 0 mV (filled triangles) are plotted against the P1 potential. Filled symbols: data for WT KCNQ1 channels co-expressed with CaM and recorded with high-Ca2+ ICS, as shown in (B). Open…

• 

  Figure 4. Ca2+ sensitivity of WT KCNQ1 channels is blunted in the mutant channels. Whole-cell recordings were performed using high-Ca2+ ICS in CHO cells expressing KCNQ1 channels without overexpression of CaM. Cells were pre-incubated for at least 30 min in normal (i) or EGTA-Ringer solution (ii). (A) Time course of WT KCNQ1 current amplitudes traced with a 500 ms test pulse to 40 mV (left panel). Normalized current amplitudes (means ± SEM) for experiments in standard or EGTA-Ringer as ECS. Initial current densities were significantly lower in EGTA-Ringer compared to standard ECS (right panel; ** p < 0.01). (B) Time course of normalized current amplitudes (means ± SEM) for experiments on V185M-, R116L- and P369L-KCNQ1 channels in standard or EGTA-Ringer as ECS. (C) Comparison of the effects of pre-incubation in EGTA-Ringer on the changes in relative current amplitude for WT and mutant KCNQ1 channels. Asterisks directly above a box indicate significant differences in relative current amplitudes of mutant KCNQ1 channels compared to the WT data obtained under the same experimental conditions. Other asterisks indicate significant differences between conditions “i” and “ii” (see schematic…

• 

  Figure 5. BAPTA experiments confirm the different sensitivities of WT and mutant KCNQ1 channels to the inhibiting effect of low intracellular Ca2+. Two hours prior to the start of the recording, cells were incubated for 1 h in medium containing 10 µM BAPTA-AM followed by a 1-h wash period in normal CHO culture medium in the incubator at 37 °C. Whole-cell recordings were performed in EGTA-Ringer with high-Ca2+ ICS in CHO cells expressing WT or mutant KCNQ1 channels. (A) Overlays of the first and then every fourth current trace recorded with a 500 ms test pulse to 40 mV, followed by a 1 s pulse to −40 mV; test pulse interval 5 s. To illustrate the current run-up at 5 min after establishing the whole-cell configuration, the respective last trace is shown in black. Vertical bars denote 0.5 nA. (B) Boxplot of initially recorded current amplitudes, normalized to the respective maximal current amplitude during a 5 min recording period. Asterisks indicate significant differences to WT data; *** p < 0.001. Numbers of experiments are given in parentheses. (C) Time course of current run-up in BAPTA-containing cells during Ca2+ loading via the recording pipette. Data are shown as means ± SEM of…

• 

  Figure 6. High intracellular Ca2+-CaM levels counteract the differences in current densities between WT and mutant Q1E2 channels. Whole-cell recordings performed in standard Ringer on cells co-expressing KCNE2 with WT or mutant KCNQ1. In two sets of experiments, CaM was additionally co-expressed (condition i and iv, see schematic drawings). The ICS contained either a high (i and iii) or a low (ii and iv) Ca2+ concentration. (A) Overlays of the first and then every fourth current trace recorded with 500 ms test pulses to 40 mV from cells co-expressing CaM and using high-Ca2+ ICS (“i”). The respective last trace within 10 min after establishing the whole-cell configuration is shown in black. Vertical bars denote 2 nA. (B) Time course of current amplitudes for the experiments shown in (A) (open circles). In addition, a second WT Q1E2 experiment is given (black triangles) to illustrate the variability in the time courses of the current increase. (C) Time course of current amplitudes from exemplary experiments performed with standard condition “ii”. (D) Boxplot of initial Q1E2 current densities (thin boxes, open circles for all outliers) and maximal Q1E2 current densities (broad boxes, filled…

• 

  Figure 7. High intracellular Ca2+ activates WT Q1E2 channels expressed in GH3/B6 anterior pituitary cells. Whole-cell recordings were performed in native GH3/B6 somato-mammotroph cells (Control) or cells overexpressing KCNE2, together with either WT or P369L-KCNQ1. In one set of WT KCNQ1 experiments, CaM was additionally co-expressed. The ICS contained either a high (5 µM) or a low (20 nM) Ca2+ concentration. (A) Family of current traces recorded from a native GH3/B6 cell with the pulse protocol shown in (B) using high-Ca2+ ICS. (B) Membrane currents recorded with low-Ca2+ ICS in a cell overexpressing WT Q1E2. Peak Na+ inward currents are truncated. (C) Membrane currents recorded with high-Ca2+ ICS in a cell overexpressing WT Q1E2. Prior to the IV protocol (family of current traces, left panel), activation of WT Q1E2 channels was traced for 5 min by applying a voltage ramp ranging from −120 mV to −40 mV every 2 s. An overlay of the first and then every fifth current trace is shown (right panel). Scale bars in (A–C) denote 1 nA and 100 ms. (D) Boxplot of initial current densities at −50 mV (thin boxes) and maximal current densities after 4−5 min in the whole-cell configuration (broad…