The molecular nature of the 17β-Estradiol binding site in the voltage- and Ca2+-activated K+ (BK) channel β1 subunit

Figures

• 

  Figure 1. BK channel expression in HEK 293 cells. (a) Immunofluorescence of channels formed by the α subunit; green: anti-KCa1.1 and Alexa Fluor 488 antibodies. (b) Immunofluorescence of the β1 subunit; red: anti-MaxiK- β1 and Alexa Fluor 568 antibodies. (c) Immunofluorescence of channels formed by co-expression of α/β1 subunits; yellow: merged image of green and red antibodies in cells expressing α and β1 subunits. Blue: DAPI nuclear stain. (d) Quantification of membrane expression by flow cytometry (% cells). α: 49.2 ± 2.7; β1: 41.51 ± 0.8, α in α/β1 co-transfected cells: 41.4 ± 2.8; β1 in α/β1 co-transfected cells: 46.1 ± 2.6. Error bars: standard error of mean (SEM).

• 

  Figure 2. E2 binding to BK channel. (a) Representative images of HEK 293 cells without transfection, (b) transfected with α subunit, (c) cells transfected with α/β1 subunits, and (d) cells transfected with β1 subunit, showing bright field and FITC channels. All cells were treated with 10 μM E2-BSA-FITC. (e) Membrane binding quantification of E2-BSA-FITC by flow cytometry using a normalized Median Fluorescence Intensity (nMFI). α: 1.14 ± 0.07 (black bar), β1: 3.21 ± 0.43 (red bar), α/β1: 2.89 ± 0.29 (squared pattern red bar). Error bars: SEM, ***P < 0.001.

• 

  Figure 3. E2 modulates VSD equilibrium in BKα/β1 channels. (a) Representative gating current recordings from α channels (black traces), and α/β1 channels (red traces) in absence or presence of E2, as indicated. (b) Gating charge-voltage relationships for α (◆) and α/β1 channel complexes (). Q-V relationships for α exposed to E2 (♢) and α/β1 subjected to E2 (), are also shown. (c) Quantification of V0.5 obtained from the Q-V curves from α channels (black bars) and α/β1 channels (red bars), with or without E2 as indicated (mean ± SEM). Non-parametric t-tests were used to compare results between channels and experimental conditions. *P < 0.05.

• 

  Figure 4. Molecular modelling of E2 interaction with BK β1 subunit. (a) Docking simulation of E2 into the β1 subunit. The sampling conformations were performed using the whole β1 and the E2 conformations with the better 10 docking scores are shown. (b) The E2 molecule with the best docking score (amino acid residues at less than 3 Ǻ from E2) is shown close to hydrophobic residues W163, and F166. (c) An upper view of the whole BK α/β1/E2 channel complex. Note that the β1 subunit resides between two adjacent voltage sensor domains. (d) Lateral view of the BK α/β1/E2 channel complex. Color code: Blue: S0; Red:S1; Grey:S2. For the sake of clarity, the β1 transmembrane segments were colored differently. TM1: violet; TM2:green. The E2 molecule is shown in CPK. (e) Backbone RMSD profile of β1 TM segments of the β1/E2 complex in a membrane system (orange), and RMSD profile of β1 TMs in the complex α/β1/E2 (blue). Last 200 ns of two trajectories are shown.

• 

  Figure 5. Mutant BK β1 channel expression in HEK 293 cells. (a) Immunofluorescence of HEK cells showing plasma membrane expression of β1 mutant W163I expressed alone, (b) or in combination with α subunit. (c) Immunofluorescence of β1 mutant F166A expressed alone, (d) β1 mutant F159A (e) β1 mutant T169A and (f) of β1 mutant L172A. (g) Quantification of levels of membrane expression measured by flow cytometry (% cells), β1W163I: 43.21 ± 4.1 (blue bar); β1, α/β1W163I: 31.71 ± 2.4 (blue and black squared pattern) using an antibody against β1 when α/β1W163I were co-transfected; β1F166A: 21.74 ± 1.9% (green bar); β1F159A: 21.10 ± 0.3% (purple bar); β1T169A: 23.98 ± 1.1 (orange bar); β1L172A: 35.63 ± 1.1 (yellow bar); antibody against α when α/β1W163I were co-transfected (blue and white squared pattern bar): 32.59 ± 2.5; Error bars: SEM. Red: anti-MaxiK-β1and Alexa Fluor 568 antibodies; blue: DAPI nuclear stain.

• 

  Figure 6. E2 binding to different β1 subunit mutants. (a) Cells transfected with β1 mutant β1W163I. (b) Cells transfected with α/β1W163I. (c) Cells transfected with β1F166A. (d) Cells transfected with β1 mutant β1F159A. (e) Cells transfected with β1 mutant β1T169A. (f) Cells transfected with β1 mutant β1L172A. Upper: bright-field; bottom: fluorescence field. (g) Membrane binding quantification of E2-BSA-FITC by flow cytometry in nMFI of β1W163I: 0.97 ± 0.1 (blue bar); α/β1W163I: 1.05 ± 0.1 (blue and white squared pattern bar); β1F166A: 0.84 ± 0.1 (green bar); β1F159A: 3.33 ± 0.1 (purple bar); β1T169A: 3.30 ± 0.6 (orange bar); β1L172A: 3.37 ± 0.6 (yellow bar). Error bars: SEM, ***P < 0.001.

• 

  Figure 7. E2 effect on BK channel macroscopic currents. (a) Representative macroscopic ion current recordings from α channels (black traces), α/β1 channels (red traces), α/β1W163I (blue traces) and α/β1F166A (green traces) without 10 μM E2 at time 0 min or with E2 at 30 min. (b,c) Conductance-voltage relationships for α (black), α/β1 (red), α/β1W163I (blue) and α/β1F166A (green) channel complexes before (circles) and after exposition to 10 μM E2 (squares). Lines represent the best Boltzmann fit. (d) Quantification of V0.5 obtained from the G-V curves from BK α channels (black bars), α/β1 channels (red bars), α/β1W163I channels (blue bars) and α/β1F166A channels (green bars)with or without E2 as indicated (mean ± SEM). Differences between channels and experimental conditions were analysed using t-test. Half activation voltages (V0.5) were: α: 164 ± 6 mV (black filled bar); α/β1: 173 ± 3 mV (red filled bar); α/β1W163I: 164 ± 4 mV (blue filled bar), α/β1F166A: 160 ± 4 mV (green filled bar); α + E2: 163 ± 3 mV (black empty bar); α/β1 + E2: 119 ± 2 mV (red empty bar); α/β1W163I + E2: 172 ± 4 mV (blue empty bar), α/β1F166A + E2: 164 ± 5 mV (green empty bar). ****P < 0.0001. n = 5–8.

• 

  Figure 8. E2 is unable to modulate VSD equilibrium in α/β1W163I and α/β1F166A mutant channels. (a) Left: representative gating current recordings from α/β1W163I channels (blue traces), in the absence or presence of E2, as indicated. Right: Q/Qmax versus membrane potential relationships forα/β1W163I in control (filled blue circles) and after the addition of E2 to a final concentration of 10 μM (open blue circles). n = 3. (b) Left: representative gating current recordings from α/β1F166A channels (green traces), in the absence or presence of E2, as indicated. Right: Q/Qmax versus membrane potential curves for α/β1F166A in control (filled green circles) and after the addition of E2 to a final concentration of 10 μM (open green circles), n = 3. The red line corresponds to the Boltzmann function that best fit the α/β1 channel data.