Silverå Ejneby M , Wu X , Ottosson NE , Münger EP , Lundström I , Konradsson P , Elinder F .

	Figure 1. The lipoelectric mechanism. (A) A compound binds with its hydrophobic anchor in the lipid membrane. The charged effector, attached to the anchor, electrostatically affects the positively charged voltage sensor (S4). (B) Compounds in A affect the voltage dependence of the channel opening. The direction of the shift depends on the valence of the charge. (C) Structures of DHAA and AA. The nomenclature of the ring structure and location of carbon 7 (C7) is shown in DHAA. (D) pH dependence for the effect of 100 µM DHAA on the 3R Shaker KV channel (updated from Ottosson et al., 2015). The carboxyl group is shown in either its protonated or deprotonated form. Error bars represent mean ± SEM (n = 4–8). (E) A negatively charged compound, located close to the voltage sensor S4, either facilitates or hinders channel opening by affecting the rotation of the voltage sensor S4. The direction depends on which side of S4 an arginine (R) is located.
	Figure 2. Effect of DHAA derivatives with different stalk lengths on the 3R Shaker KV channel. (A) Top view of the Shaker KV channel VSD. Gating charges (arginines) in blue: the WT Shaker KV channel has R362 as top charge (=R1); the 3R Shaker KV channel has two additional charges (M356R and A359R). (B) Nomenclature for the position of the charged group (p1–11; top) and the specific carboxyl-acid stalks used (bottom). (C) K currents at 10 mV (top) and normalized G(V) curves (bottom), before (black) and after (red) application of 100 µM DHAA(p1) (G(V) shift = ‒13.0 mV) and Wu149(p6) (G(V) shift = ‒5.3 mV). pH = 7.4. (D) G(V)-shifting effects of DHAA (black) and AA (red) derivatives with different stalk lengths (100 µM, pH = 7.4). Mean ± SEM (n = 3–11). Data fitted with Eq. 4. Length constant λ = 5.1 ± 0.5 atoms (DHAA anchor) and λ = 6.1 ± 0.7 atoms (AA anchor), respectively.
	Figure 3. No correlations between G(V) shifts, pKa, and log p-values for DHAA derivatives. (A) Correlation between the G(V) shift of the 3R Shaker KV channel induced by the DHAA derivatives (100 µM, pH = 7.4) from Fig. 2 D and their calculated pKa values. Error bars represent mean ± SEM (n, see Table 1). (B) Correlation between the G(V) shift of the 3R Shaker KV channel induced by the DHAA derivatives (100 µM, pH = 7.4) from Fig. 2 D and their calculated solubility (log P) for the uncharged molecule. Error bars represent mean ± SEM (n, see Table 1). Correlation analyses with Pearson’s correlation test and linear regressions were not significant. P < 0.05 considered significant.
	Figure 4. Role of the stalk length with a fully charged effector. (A) Structures of permanently charged DHAA derivatives with a sulfonic acid group. (B) G(V)-shifting effects of DHAA derivatives (100 µM) on the 3R Shaker KV channel. White circles denote carboxyl groups (structures in Fig. 2 B) at pH 10.0. Black circles denote permanently charged groups (structures in A) at pH 7.4. Mean ± SEM (n = 3–7). Gray dashed line is for DHAA derivatives with carboxyl groups at pH 7.4 (Fig. 2 D). (C) pH dependence for G(V) shifts of indicated compounds (100 µM) on the 3R Shaker KV channel. Mean ± SEM (n = 3–11). Data fitted to Eq. 3. DHAA: pKa = 7.2, ΔVMax = −23.4 mV. Wu180: pKa = 7.3, ΔVMax = −19.3 mV. Wu179: pKa = 7.5, ΔVMax = −20.2. Wu176: pKa = 6.8, ΔVMax = −18.2 mV. (D) G(V)-shifting effects of DHAA derivatives (100 µM) on the WT Shaker KV channel. Symbols as in B. Mean ± SEM (n = 3–7).
	Figure 5. The cutoff model. (A) Electrostatic energy for a charge q1 = 1 e at (0,4) à and q2 = ‒1 e at (x,z) à . (B) Schematic illustration of the cutoff model. At one anchor position, the charge q2 is attracted to q1, independent of the stalk length l (i). At another anchor position, the charge q2 is attracted to the extracellular solution, independent of the stalk length (ii). At certain anchor positions, the charge q2 is attracted either to q1 for shorter stalk lengths (iii) or to the extracellular solution for longer stalk lengths (iv). (C) Critical stalk lengths (i.e., the length at which the charge q2 is attracted equally well to the charge q1 or to the extracellular solution) for a stalk fixed at (x,z) à . Charge q1 = 1 e at d = (0,2), (0,4), (0,6) à and q2 = ‒1 e at the end of the stalk with different lengths (according to the color coding). (D) Positions for the cutoff anchor point when the stalk length is 4 à (l = 4 à ). Charge q1 = 1 e at d = (0,2), (0,4), (0,6) à and q2 = ‒1 e at the end of the stalk.
	Figure 6. The valence of the charge is critical for the effect. (A) G(V) shifts for DHAA derivatives on the 3R Shaker KV channel (100 µM, pH = 7.4). Mean ± SEM (n = 3–4 for Wu110 and Wu111, n = 5–6 for Wu161 and Wu109). (B) Effect of the uncharged Wu110 on the normalized G(V) curve of the 3R Shaker KV channel. G(V) shift = 0.0 mV. (C) Effect of the permanently charged Wu161 on the normalized G(V) curve of the 3R Shaker KV channel. G(V) shift = ‒34.6 mV.
	Figure 7. Divalent DHAA derivatives are less potent than monovalent DHAA derivatives. (A) Cutoff area for charge-dependent effects for a stalk length of 4 à (l = 4 à ). Charge q1 = 1 e at d = (0,4). q2 = ‒1 e (black line) and q2 = ‒2 e (red dashed line). In the area between the two lines (at point X), q2 = ‒1 e will be attracted toward q1 = 1 e (representing the voltage sensor S4) in the membrane, and q2 = ‒2 e will be attracted toward the water (with a small effect on S4). (B) Structure of Wu162. (C) pH dependence of the G(V) shift induced by 100 µM Wu162(p3) on the 3R Shaker KV channel (top). Mean ± SEM (n = 3–5). Calculated pH dependence for Wu162(p3) and structures of the microspecies (bottom). (D) Molecular structure of Wu148. (E) G(V) shift induced by 100 µM Wu148(p4/p5) on the 3R Shaker KV channel compared with DHAA(p1), Wu117(p4), and Wu152(p4) (Fig. 2 D). pH = 7.4. Error bars represent mean ± SEM (n = 6). (F) pH dependence of the G(V) shifts induced by 100 µM Wu148(p4/p5) on the 3R Shaker KV channel. G(V) shifts are compared with Wu148(p4/p5) at pH 7.4, one-way ANOVA with Dunnett’s multiple comparison test. *, P < 0.05. Mean ± SEM (n = 4–6).
	Figure 8. Wu161(p3) is five times more potent than DHAA. (A) Normalized G(V) curves. G(V) shift = ‒12.4 mV. The 3R Shaker KV channel. pH = 7.4. (B) Concentration–response curves for Wu161(p3). pH = 7.4. Mean ± SEM (n = 3–6). c½ = 44 µM, ΔVMax = ‒13.9 mV (WT, Eq. 3). c½ = 36 µM, ΔVMax = ‒45.5 mV (3R, Eq. 3). (C) Wu161 and DHAA on the 3R (10 µM) and WT (100 µM) Shaker KV channels, respectively. pH = 7.4. Mean ± SEM (n = 3–10). Data for DHAA from Ottosson et al. (2015).
	Figure 9. The role of S4 charges for the effect of Wu161 and other DHAA derivatives. (A) Schematic picture of mutated S4 charges on the Shaker KV channel. The top gating charge (an arginine, R362 in WT) was moved step-by-step further out on S4 (left), or removed (R362Q; right). (B) Wu161-induced G(V) shifts on the Shaker KV channel S4 arginine mutants. 100 µM, pH = 7.4. Mean ± SEM (n = 3–5). Shifts are compared with R362Q (dashed line), one-way ANOVA with Dunnett’s multiple comparison test. Blue: nonsignificant, P > 0.05. Green: larger effect than for R362Q, P < 0.001. Orange: smaller effect than for R362Q, P < 0.001. Data for the mutations in control solution at pH 7.4 according to Eq. 2: R362Q (V½ = 3.4 ± 2.9 mV, s = 12.0 ± 0.8 mV, n = 4), R362 (WT) (V½ = −21.4 ± 1.8 mV, s = 10.8 ± 1.1 mV, n = 3), L361R/R362Q (V½ = −50.5 ± 1.4 mV, s = 17.9 ± 0.3 mV, n = 4), I360R/R362Q (V½ = 24.6 ± 2.3 mV, s = 12.2 ± 1.0 mV, n = 5), A359R/R362Q (V½ = 32.3 ± 3.8 mV, s = 13.0 ± 1.8 mV, n = 4), L358R/R362Q (V½ = −12.1 ± 0.9 mV, s = 11.6 ± 0.3 mV, n = 4), S357R/R362Q (V½ = −34.7 ± 2.1 mV, s = 8.1 ± 0.2 mV, n = 4), M356R/R362Q (V½ = 19.3 ± 2.5 mV, s = 9.0 ± 0.1 mV, n = 4), and A359R/M356R (3R; V½ = 25.4 ± 2.0 mV, s = 12.3 ± 0.9 mV, n = 6). (C) Correlation between G(V) shifts for the PUFA DHA (Ottosson et al., 2014) and for Wu161 on the Shaker KV channel S4 arginine mutants (100 µM, pH = 7.4). Error bars represent mean ± SEM (n = 4–14). Slope (2.1 ± 0.2) is significantly different from 0 (Pearson correlation test and linear regression are both significant). (D) Effects of DHAA(p1), Wu32(p1) (Ottosson et al., 2017), Wu161(p3), and PUFA DHA (Ottosson et al., 2014) on the R362Q, WT, and 3R Shaker KV channels (100 µM, pH = 7.4). Error bars represent mean ± SEM (n = 4–15).
	Figure 10. Combined stalk and anchor modifications of the DHAA molecule. (A) DHAA derivatives studied. (B) Concentration–response curves for the DHAA derivatives in A on the 3R Shaker KV channel, pH = 7.4. Mean ± SEM (n = 3–9). Eq. 3: c½ = 98 µM, ΔVMax = ‒26.5 mV (DHAA); c½ = 36 µM, ΔVMax = ‒45.5 mV (Wu161); c½ = 37 µM, ΔVMax = ‒46.0 mV (Wu50); and c½ = 6.1 µM, ΔVMax = ‒53.6 mV (Wu181). Wu50 and DHAA adapted from Ottosson et al. (2015). (C) Currents at 10 mV (top) and normalized G(V) curves (bottom) for the 3R Shaker KV channel, pH = 7.4. G(V) shift = ‒20.4 mV.
	Figure 11. DHAA derivatives open the human M channel. (A) Currents before and after application of 100 µM Wu161. Steps to voltages between ‒120 and 20 mV (in 5-mV increments) started at t = 0.7 s. At t = 2.7 s, the voltage was switched to ‒10 mV. Traces at voltage = ‒20 mV in red. (B) G(V) curves for the cell in A. G(V) shift = ‒23.8 mV. (C) G(V) shifts for compounds with different stalk lengths (100 µM). Mean ± SEM (n = 3–6). DHAA(p1) at pH = 10. Permanently charged Wu164(p2), Wu161(p3), and Wu154(p5) at pH = 7.4. (D) Compounds with different valences (100 µM). pH = 7.4. Mean ± SEM (n = 3–5). ***, P < 0.001, significantly different from 0. (E) Concentration–response curves. Mean ± SEM (n = 3–4). Eq. 3: c½ = 62 µM, ΔVMax = ‒35.0 mV (Wu161); c½ = 12 µM, ΔVMax = ‒42.1 mV (Wu181).
	Figure 12. Summary of suggested mechanisms. Proposed binding of four different compounds to the S3/S4 cleft. The positive charge in blue represents R362R (=R1) in the top of S4. The negative charge in red represents the charged group of the four different compounds. For P1 and P3 compounds and for PUFAs, the negative charge is suggested to be attracted to the S4 charge. For P4 compounds, the charge is suggested to be attracted to the extracellular water.
	Figure 13. Molecular structures of Wu109, Wu110, Wu111, Wu117, and Wu148, including some reactions. See Appendix for a detailed description. Full compound names: (((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)methyl)sulfamic acid (Wu109); N-(((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)methyl)methanesulfonamide (Wu110); N-(((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)methyl)acetamide (Wu111); ((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carbonyl)glycine (Wu117); and ((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carbonyl)-l-aspartic acid (Wu148). RCl, R = SO3H, -SO2Me, -COMe; TEA, triethylamine.
	Figure 14. Molecular structures of Wu149, Wu150, Wu151, Wu152, Wu153, Wu154, Wu157, and Wu158, including some reactions. See Appendix for a detailed description. Full compound names: 4-((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)butanoic acid (Wu149); 3-((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)propane-1-sulfonic acid (Wu150); 4-(4-((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)butanamido)butanoic acid (Wu151); 3-((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)propanoic acid (Wu152); 3-(3-((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)propanamido)propanoic acid (Wu153); 2-((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)ethane-1-sulfonic acid (Wu154); 4-((1R,4aR,4bR,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,4b,10,10a-decahydrophenanthrene-1-carboxamido)butanoic acid (Wu157); and (((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthrene-1-carboxamido)methyl)phosphonic acid (Wu158).
	Figure 15. Molecular structures of Wu50, Wu161, Wu162, Wu176, Wu179, Wu180, and Wu180, including some reactions. See Appendix for a detailed description. Full compound names: ((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)methyl hydrogen sulfate (Wu161); ((1R,4aS,10aR)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)methyl dihydrogen phosphate (Wu162); (E)-3-((1S,4aS)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)acrylic acid (Wu176); 3-((1S,4aS)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)propanoic acid (Wu179); 2-((1S,4aS)-7-isopropyl-1,4a-dimethyl-1,2,3,4,4a,9,10,10a-octahydrophenanthren-1-yl)acetic acid (Wu180); and ((1R,4aS,10aR)-6,7,8-trichloro-1,4a-dimethyl-2,3,4,9,10,10a-hexahydrophenanthrene-1-yl)methyl hydrogen sulfate (Wu181).

Börjesson, An electrostatic potassium channel opener targeting the final voltage sensor transition. 2011, Pubmed, Xenbase

Börjesson, An electrostatic potassium channel opener targeting the final voltage sensor transition. 2011, Pubmed , Xenbase
Börjesson, Structure, function, and modification of the voltage sensor in voltage-gated ion channels. 2008, Pubmed
Börjesson, Lipoelectric modification of ion channel voltage gating by polyunsaturated fatty acids. 2008, Pubmed , Xenbase
Börjesson, Electrostatic tuning of cellular excitability. 2010, Pubmed , Xenbase
Delemotte, Intermediate states of the Kv1.2 voltage sensor from atomistic molecular dynamics simulations. 2011, Pubmed
Elinder, Actions and Mechanisms of Polyunsaturated Fatty Acids on Voltage-Gated Ion Channels. 2017, Pubmed
Henrion, Tracking a complete voltage-sensor cycle with metal-ion bridges. 2012, Pubmed , Xenbase
Hessa, Membrane insertion of a potassium-channel voltage sensor. 2005, Pubmed
Hessa, Recognition of transmembrane helices by the endoplasmic reticulum translocon. 2005, Pubmed
Hoshi, Biophysical and molecular mechanisms of Shaker potassium channel inactivation. 1990, Pubmed , Xenbase
Humphries, Neuronal and Cardiovascular Potassium Channels as Therapeutic Drug Targets: Promise and Pitfalls. 2015, Pubmed
Imaizumi, Molecular basis of pimarane compounds as novel activators of large-conductance Ca(2+)-activated K(+) channel alpha-subunit. 2002, Pubmed
Jensen, Mechanism of voltage gating in potassium channels. 2012, Pubmed
Jentsch, Neuronal KCNQ potassium channels: physiology and role in disease. 2000, Pubmed
Kamb, Molecular characterization of Shaker, a Drosophila gene that encodes a potassium channel. 1987, Pubmed
Kobayashi, K(+)-channel openers suppress epileptiform activities induced by 4-aminopyridine in cultured rat hippocampal neurons. 2008, Pubmed
Liin, Polyunsaturated fatty acid analogs act antiarrhythmically on the cardiac IKs channel. 2015, Pubmed , Xenbase
Liin, Polyunsaturated fatty acids are potent openers of human M-channels expressed in Xenopus laevis oocytes. 2016, Pubmed , Xenbase
Long, Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment. 2007, Pubmed
Nerbonne, Molecular physiology of cardiac repolarization. 2005, Pubmed
Ottosson, Resin-acid derivatives as potent electrostatic openers of voltage-gated K channels and suppressors of neuronal excitability. 2015, Pubmed , Xenbase
Ottosson, A drug pocket at the lipid bilayer-potassium channel interface. 2017, Pubmed
Ottosson, Drug-induced ion channel opening tuned by the voltage sensor charge profile. 2014, Pubmed , Xenbase
Sakamoto, Molecular mechanisms underlying pimaric acid-induced modulation of voltage-gated K+ channels. 2017, Pubmed
Salari, Isopimaric acid - a multi-targeting ion channel modulator reducing excitability and arrhythmicity in a spontaneously beating mouse atrial cell line. 2018, Pubmed
Takeda, Potassium channels as a potential therapeutic target for trigeminal neuropathic and inflammatory pain. 2011, Pubmed
Tigerholm, Dampening of hyperexcitability in CA1 pyramidal neurons by polyunsaturated fatty acids acting on voltage-gated ion channels. 2012, Pubmed
Wu, Pharmacodynamics of potassium channel openers in cultured neuronal networks. 2014, Pubmed
Yazdi, The Molecular Basis of Polyunsaturated Fatty Acid Interactions with the Shaker Voltage-Gated Potassium Channel. 2016, Pubmed