Stoichiometry of the KCNQ1 - KCNE1 ion channel complex

Abstract

The KCNQ1 voltage-gated potassium channel and its auxiliary subunit KCNE1 play a crucial role in the regulation of the heartbeat. The stoichiometry of KCNQ1 and KCNE1 complex has been debated, with some results suggesting that the four KCNQ1 subunits that form the channel associate with two KCNE1 subunits (a 42 stoichiometry), while others have suggested that the stoichiometry may not be fixed. We applied a single molecule fluorescence bleaching method to count subunits in many individual complexes and found that the stoichiometry of the KCNQ1 - KCNE1 complex is flexible, with up to four KCNE1 subunits associating with the four KCNQ1 subunits of the channel (a 44 stoichiometry). The proportion of the various stoichiometries was found to depend on the relative expression densities of KCNQ1 and KCNE1. Strikingly, both the voltage-dependence and kinetics of gating were found to depend on the relative densities of KCNQ1 and KCNE1, suggesting the heart rhythm may be regulated by the relative expression of the auxiliary subunit and the resulting stoichiometry of the channel complex.

Fig. 1.

KCNQ1 forms a tetramer. (A) Schematic diagram of mEGFP - KCNQ1 and mEGFP_NotIKCNQ1 constructs. mEGFP was inserted in the middle of the cytoplasmic N-terminal domain in mEGFP_NotIKCNQ1. (B) A single frame from a TIRF movie of a Xenopus oocyte expressing mEGFP - KCNQ1. Green circles indicate immobile spots suitable for counting. (Scale bar: 2 μm). (C) Examples of four bleaching steps from two spots of mEGFP - KCNQ1. Dotted lines indicate the fluorescence intensity of single mEGFPs. (D) Distributions of observed bleaching step numbers (gray bars) from oocytes expressing mEGFP - KCNQ1. Distribution profiles are well fit by a binomial distribution with p (probability of GFP to be fluorescent) of 64% (white bars).

Fig. 2.

Up to four KCNE1 subunits within a KCNQ1 complex. (A) Images from Xenopus oocytes coexpressing mCherry - KCNQ1 and KCNE1 - mEGFP - Kv1.4C. mCherry image (left) and mEGFP image (center) are superimposed (right). Overlapping spots are marked by white arrowheads. (Scale bars: 2 μm). (B) Examples of multiple bleaching steps of mEGFP fluorescence from two representative spots with both red and green fluorescence. Illumination at 593 nm to excite mCherry for 5 s (red bars) was followed by illumination at 488 nm to excite mEGFP (green bars). Top example shows two bleaching steps while bottom example shows four bleaching steps. Arrows indicate the fluorescence levels. (C) (Left) Observed frequency distribution of the number of mEGFP bleaching steps from the mCherry(KCNQ1) + EGFP(KCNE1) overlapping spots (mean ± sem, n = 22 (optical patches from four oocytes of two different batches). (Right) Theoretical probabilities for monomers, dimers, trimers, and tetramers with p (probability of mEGFP being fluorescent) = 80%.

Fig. 3.

Stoichiometry of the KCNQ1∶KCNE1 complex depends on the relative density of expression of KCNQ1 and KCNE1. Average number of bleaching steps of mEGFP fluorescence for each optical patch is plotted against the relative density of mEGFP (KCNE1) to mCherry (KCNQ1) in the patch. X-axis is in a logarithmic scale. Data were taken from 38 optical patches from seven oocytes of four different batches. Same color indicates multiple patches from same oocyte. The ratio of two RNAs (mCherry - KCNQ1/KCNE1 - mEGFP - Kv1.4C) for the oocyte represented by the purple squares was 10, that of yellow-green triangles was 100, and the remainder (circles) were 1,000. If relative density is lower than 1, there are more red spots (KCNQ1) than green spots (KCNE1). If relative density is higher than 1, there are more KCNE1 subunits than KCNQ1. Dashed lines indicate expected values for 1–4 KCNE1 subunits if the stoichiometry is fixed.

Fig. 4.

Gating properties of KCNQ1 - KCNE1 complexes with different ratios of expression of KCNE1 to KCNQ1. (A) KCNQ1 currents with KCNE1 coexpressed at various RNA concentrations, 2 d after RNA injection. Voltages were stepped from -100 mV to various voltages (up to +80 mV) before being stepped to -30 mV for tail current measurement. Maximum tail current amplitude for each KCNE1 concentration is plotted below (n = 6 for each concentration). (B) Normalized G-V curves for different concentrations of KCNE1 RNA. G-V curves are fitted to the Boltzmann function. (C) Current traces and G-V curves of a linked construct of one KCNE1 subunit linked to two KCNQ1 subunits (the E1 - Q1 - Q1 tandem) alone (black) and coexpressed with free KCNE1 (red) or KCNE3 (blue). (D) Current traces and G-V curves of a linked construct of one KCNE1 subunit linked to one KCNQ1 subunit (the E1 - Q1 tandem) alone (black) and coexpressed with KCNE1 (red) or KCNE3 (blue).

Fig. 5.

Model of density-dependent stoichiometry of the KCNQ1 - KCNE1 complex. Stoichiometry of the KCNQ1 - KCNE1 complex depends on how many free KCNE1 subunits (green circles) are available for the exogenous KCNQ1 channels (red circles). Endogenous KCNQ1 subunits (xKCNQ1) in Xenopus oocytes are depicted as gray circles. Because we could not count xKCNQ1 and do not know if xKCNQ1 forms heteromultimers with human KCNQ1, the behavior of xKCNQ1 is unknown. However, the smaller value of p (probability of fluorescent GFP) shown in Fig. 1 suggests the involvement of xKCNQ1 subunit in human KCNQ1 channel.