Figure 1.
Schematic of the two-pronged screen designed to identify compounds that bind and regulate EAG1 channels.
(A) EAG1 channel subunits form tetramers around a central, K+ conducting pore. The ribbon representation is a homology model of mEAG1#505-702, obtained with SWISS-MODEL based on the crystal structure of the carboxy-terminal region of mHCN2 [42]. Enlarged view on the right shows a tryptophan residue at position 649 in the CNBD employed as a reporter of ligand binding. (B) For the electrophysiology arm of the screen, mEAG1 channels were expressed heterologously and the FOL pools were applied to inside-out patches containing these channels. The currents from EAG1 channels were recorded in the absence (black), presence (red) and after washout of FOL compounds. The conductance-voltage relationship was determined from the tail currents and normalized to the maximal conductance in the absence of drug. (C) For the fluorescence portion of the screen, mEAG1#505-702 was expressed and purified from bacteria. The FOL compounds were applied to mEAG1#505-702 in solution, the sample was placed in a quartz cuvette and excited at 280 nm wavelength. The mEAG1#505-702 emission spectra were recorded in the absence and presence of the FOL compounds (black and red traces, respectively).
Figure 2.
Summary of the screen results.
(A) The change in relative conductance of the electrophysiology portion of the screen is plotted in a black to green color gradient, with pools that did not change the relative conductance colored black, and those with a 100% reduction in conductance colored in bright green. (B) The shift of the Vhalf of the conductance versus voltage relationship from the electrophysiology portion of the screen is plotted in a black to green color gradient, for shifts to more positive potentials indicated by shades of green (with a shift of 16 mV indicated by the brightest green). Shifts of the conductance versus voltage relationship to more negative potentials are plotted in a black to red gradient, with the brightest red indicating a shift of −16 mV. The pools that repeatedly destroyed membrane seals and were removed from further analysis are indicated by grey boxes. (C) Results of the fluorescence based portion of the screen presented as RPC for each of the 96 FOL pools plotted in a black to green color gradient with pools that had no effect on the tryptophan colored in black, and largest effects colored in bright green. Grey boxes indicate the pools that had a larger effect on free tryptophan fluorescence than the fluorescence of the protein.
Figure 3.
Chemical structures of the compounds identified as hits.
(A) Structures of flavonoids luteolin and myricetin identified as EAG1 channel regulators by both electrophysiology and tryptophan fluorescence portions of the screen. (B) Structure of 5-chloroindole identified as EAG1 channel regulator by electrophysiology screen. (C) Structures of benzofurans 2-(6-methoxy-1-benzofuran-3-yl) acetic acid and benzofuran-2 carboxylic acid identified as binding partners of EAG1 channels by the tryptophan fluorescence portion of the screen.
Figure 4.
Luteolin modulated currents from EAG1 channels.
Current traces (A) and conductance-voltage relationship (B) for EAG1 channels recorded in the inside-out patch configuration in the absence (black), presence (red) and after washout (grey) of 30 µM luteolin. (C) The tail current recorded at −120 mV, following a voltage step to −50 mV, in the absence (black) and presence (red) of 30 µM luteolin, fit with single exponentials to give time constants of 3.5 ms before, and 6.2 ms after application of luteolin. (D) Plot of the change in the Vhalf versus luteolin concentration, fit with a Hill equation with a binding affinity of 9.1 µM, and a Hill coefficient of 4.
Figure 5.
Luteolin reduced mEAG1#505-702 fluorescence in a concentration dependent manner.
(A) The inner filter corrected, and background subtracted emission spectra of mEAG1#505-702 recorded without and with the indicated concentrations of luteolin. (B) The inner filter corrected, and background subtracted emission spectra of free tryptophan in solution recorded without and with the indicated concentrations of luteolin. (C) Plots of change in the peak emission fluorescence intensity versus total luteolin concentration for mEAG#505-702 (filled squares) and free tryptophan (open squares), fit with equation (5). The peak fluorescence intensity corresponded to fluorescence intensity at 338 nm for mEAG1#505-702 and at 357 nm for free tryptophan. The change in the peak fluorescence intensity was calculated by subtracting averaged peak emission intensity for low concentrations of luteolin (intensities at 0, 0.01 and 0.1 µM luteolin at 338 nm) from the peak emission intensities. The binding affinity of luteolin was ≥100 µM for mEAG1#505-702.
Figure 6.
5-Chloroindole modulated currents from EAG1 channels.
Current traces (A) and conductance-voltage relationship (B) for EAG1 channels recorded in the inside-out patch configuration in the absence (black), presence (red) and after washout (grey) of 100 µM 5-chloroindole. (C) The tail current at −120 mV, following a voltage step to −50 mV, in the absence (black) and presence (red) of 100 µM 5-chloroindole, and fit with single exponentials to give time constants of 3.0 ms before, and 4.1 ms after application of 5-chloroindole. (D) Plot of the change in the Vhalf versus 5-chloroindole concentration, fit with a Hill equation. The binding affinity and Hill coefficient of 5-chloroindole were 420 µM and 1.4 respectively.
Figure 7.
2-(6-methoxy-1-benzofuran-3-yl) acetic acid and benzofuran-2 carboxylic acid reduced fluorescence of mEAG1#505-702.
The inner filter corrected, and background subtracted emission spectra of mEAG1#505-702 recorded without (black) and with (red) 30 µM 2-(6-methoxy-1-benzofuran-3-yl) acetic acid (A), and 50 µM benzofuran-2 carboxylic acid (B). Also shown are the inner filter corrected, and background subtracted emission spectra of free tryptophan recorded without (grey) and with (green) 30 µM 2-(6-methoxy-1-benzofuran-3-yl) acetic acid (A) and 50 µM benzofuran-2 carboxylic acid (B).