6,7-Dichloro-1H-indole-2,3-dione-3-oxime functions as a superagonist for the intermediate-conductance Ca2+-activated K+ channel KCa3.1

Figures

• 

  Fig. 1. NS309 exhibits superagonistic effects on KCa channels. (A) Ca2+-concentration response curve for KCa3.1 activation measured from inside-out patches in the absence or presence of 1 μM NS309, displaying a calcium EC50 of 430 nM for the control curve (95% CI: 365–502 nM; nH 2.9; n = 3 for all data points) and 31 nM in the presence of NS309 (95% CI: 22–41 nM, P < .0001; nH 2.0; n = 2 for 20 and 40 nM data points, and n = 3 for remaining concentrations). (B) NS309 concentration response curves obtained via whole-cell patches in 250 nM free calcium (black) and 3 μM free calcium (red), with measured NS309 EC50 values of 74 nM (95% CI: 53–96 nM; nH 1.3; n = 3) and 8.6 nM (95% CI: 2.6–20 nM; nH 1.0; n = 3; P < .0001), respectively. P values for curve comparison were calculated using an extra sum-of-squares F test (GraphPad Prism 10; GraphPad Software, La Jolla, CA). (C) Whole-cell recordings of KCa2.1, KCa2.2, KCa2.3, and KCa3.1 in saturating calcium conditions (10 μM intracellular calcium), with the effect of a 10 μM NS309 perfusion shown in red. D. Scatter plot showing current increases for KCa2.1, KCa2.2, KCa2.3, and KCa3.1 induced by 10 μM of NS309 at 10 μM intracellular calcium (n =…

• 

  Fig. 2. Search for small-molecule binding sites in the KCa3.1 channel using Schrödinger’s SiteMap. (A) Seven distinct possible binding sites for KCa3.1 modulation were identified by the SiteMap function of Schrödinger suite. KCa3.1 is shown in light blue. Calmodulin is depicted in pale-yellow. (B) Standard precision GLIDE Docking of NS309 in sites A (green) and B (orange), with binding free energy scores reported at −6.9 and −7.05 kcal/mol respectively. Van der Waal’s interactions are shown as dashed pink lines and H-bonds are shown in solid purple.

• 

  Fig. 3. RosettaLigand Docking of NS309 in sites A and B. (A) Convergence of the top 10 poses within site A. (B) Top-ranked pose in site A, with VDW forces in pink and H-bonds in purple. (C) Convergence of the top 10 poses within site B. (D) Top-ranked pose in site B, with VDW interactions shown in pink and H-bonds in purple.

• 

  Fig. 4. Effect of NS309 on KCa3.1 mutations within sites A and B. (A) Concentration response curves of NS309 on wild-type KCa3.1 in black (EC50 = 74 nM, 95% CI: 55−95 nM, nH 1.3), L185A in yellow (EC50 = 1.27 μM, 95% CI: 945 nM to 1.83 μM, nH 1.4, P < .0001), S181A-L185A double mutant in purple (EC50 = 1.1 μM, 95% CI: 730 nM to 1.61 μM, nH 1.6, P < .0001), N201F in green (EC50 = 195 nM, 95% CI: 138−264 nM, nH 1.3, P < .0001), N201A in red (EC50 = 85 nM, 95% CI: 64−112 nM, nH 1.2, P = .4679), and R287A in blue (EC50 = 98 nM, 95% CI: 77−125 nM, nH 1.5, P = .1309). Each data point on curves represents an n of 3, displayed as mean ± SD. (B) Effect of NS309 at 25 and 250 nM concentrations on WT KCa3.1. (C) Effect of NS309 at 250 nM, 1 μM, and 10 μM concentrations on the S181A-L185A double mutant. For both recordings, a 200 ms voltage-ramp protocol from −120 to +40 mV was applied every 10 seconds.

• 

  Fig. 5. Effect of NS309 on the KCa2.2 site A double mutant S288A-L292A. (A) Concentration response curves of NS309 on wild-type KCa2.2 in black (EC50 = 1.7 μM, 95% CI: 912 nM to 4.2 μM) and S288A-292A double mutant in purple (EC50 >83 μM). Each data point on curve represents an n of 3, displayed as mean ± SD. (B) Effect of NS309 at 1 and 10 μM concentrations on WT KCa2.2. (C) Effect of NS309 at 10, 40, and 120 μM concentrations on the S288A-L292A double mutant. For both recordings, a voltage-ramp protocol was applied, going from −120 to +40 mV over 200 ms.

• 

  Fig. 6. KCa3.1 and NS309 molecular dynamics simulations setup. (A) Side view of KCa3.1 in the lipid bilayer. (B) Channel (purple) with 4 calmodulin (teal) and 4 bound NS309 molecules viewed from the intracellular side. View of KCa3.1 within the lipid bilayer. (C) Inset from panel A showing one of the NS309 binding sites viewed from the side and from below showing the proximity to S181 and L185 in the S4–S5 linker of the channel and E54 in calmodulin. (D) Representative frames of the course of the MD simulations showing hydrogen bonds (dotted yellow lines) between NS309 and E54 of calmodulin and S181 of KCa3.1, and hydrophobic contacts (dotted black lines) with L185. (E). Histograms of donor to acceptor distances between NS309 and residues S181 of KCa3.1 and E54 of calmodulin as well as carbon-carbon distances for hydrophobic contacts between KCa3.1 residue L185 and NS309.

• 

  Fig. 7. Water molecules interacting with NS309. (A) Number of water molecules per ligand over the time course of the MD simulation. Two to 6 water molecules typically make hydrogen bonds with each NS309 molecule. (B) A representative MD frame of NS309 making hydrogen bonds with S181, E54 in calmodulin, and 2 water molecules is shown next to the time course plot. (C) Evolution of hydrogen bonding with the indole group. Frame 15 of the simulation shows a water molecule mediating interactions with S181. Frame 306 shows a hydrogen bond with the hydroxyl group of S181, displacing the water molecule. Frame 432 shows a hydrogen between NS309 and the carbonyl oxygen of S181 which remains stable for the rest of the simulation.

• 

  Fig. 8. Diameter of the hydrophobic gate during the MD simulations. (A) Representative diameter of the hydrophobic gate following system equilibration. V282 is shown in a stick representation on all 4 chains. Average diameter of the hydrophobic gate over the course of the trajectory for each of the 3 simulations run for (B) zero NS309, (C) 2 NS309, and (D) 4 NS309 molecules bound to the KCa3.1 channel. A diameter of 7 Å is shown as a black dashed line indicating the minimum gate diameter likely to conduct a hydrated K+ ions.

• 

  Fig. 9. Characterization of 2 gain-of-function KCa3.1 mutations. (A) Whole-cell recordings of wild-type and constitutively open V282G and A279G mutant channels in response to various concentrations of intracellular calcium. Zero calcium was achieved by using a fluoride containing pipette solution. (B) Location of residues A279 (dark blue) and V282 (gold) in the lower pore of KCa3.1. (C) Whole-cell recordings of V282G and A279G mutants using 250 nM intracellular calcium. NS309 at 10 μM (green) does not increase currents. The KCa3.1 blocking peptide toxin charybdotoxin at 300 nM (red) inhibits currents in both mutants. All shown traces are 1 representative recording of at least 3 independent experiments.