Fig. 10: Carnosic acid is structurally optimal for KCNQ3 channel opening.

a Structure of carnosic acid highlighting the carboxyl group. b In silico docking of carnosic acid to wild-type and mutant KCNQ3 models. c Structures and electrostatic surface potentials (red, negative; blue, positive) of carnosic acid and derivatives as indicated. d In silico docking of carnosic acid and derivatives to wild-type KCNQ3 model. e Mean traces for KCNQ2, KCNQ3* or KCNQ2/3 expressed in oocytes in the absence (Control) or presence of carnosic acid derivatives (100 µM). Scale bars lower left; voltage protocol upper inset; n = 4–10 per group. f Mean normalized tail current (G/Gmax) for KCNQ2, KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of compounds as in (e); n = 4–10 per group. g Mean unclamped oocyte membrane potential for oocytes expressing KCNQ2, KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of compounds as in (e); n = 4–10 per group; DMC dimethylcarnosol. h Structures and electrostatic surface potentials (red, negative; blue, positive) of carnosic acid derivatives as indicated. i In silico docking of carnosic acid derivatives to wild-type KCNQ3 model. j Mean traces for KCNQ2, KCNQ3* or KCNQ2/3 expressed in oocytes in the absence (control) or presence of carnosic acid γ-lactone (10 µM). Scale bars lower left; voltage protocol as in (b); n = 5–10 per group. k Mean normalized tail current (G/Gmax) for KCNQ2, KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of carnosic acid γ-lactone (10 µM); n = 5–10 per group. l Mean unclamped oocyte membrane potential for oocytes expressing KCNQ2, KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of carnosic acid γ-lactone (CAγL) (10 µM); n = 5–10 per group. m Mean traces for KCNQ3* or KCNQ2/3 expressed in oocytes in the absence (Control) or presence of methyl carnosate (100 µM). Scale bars lower left; voltage protocol as in panel e; n = 5–10 per group. n Mean normalized tail current (G/Gmax) for KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of methyl carnosate (100 µM); n = 5–10 per group. o Mean unclamped oocyte membrane potential for oocytes expressing KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of methyl carnosate (100 µM); n = 5–10 per group. MC methyl carnosate. p Structures and electrostatic surface potentials (red, negative; blue, positive) of carnosic acid derivatives as indicated. q In silico docking of carnosic acid derivatives to wild-type KCNQ3 model. r Mean traces for KCNQ2, KCNQ3* or KCNQ2/3 expressed in oocytes in the absence (Control) or presence of carnosic acid derivatives (10–100 µM). Scale bars lower left; voltage protocol upper inset; n = 10 per group. s Mean normalized tail current (G/Gmax) for KCNQ2, KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of compounds as in (r); n = 10 per group. t Mean unclamped oocyte membrane potential for oocytes expressing KCNQ2, KCNQ3* or KCNQ2/3 in the absence (black) or presence (red) of compounds as in (r); n = 10 per group; CAD Carnosic acid diacetate; PA pisiferic acid. u Mean ΔV_0.5act for KCNQ3* in response to carnosic acid and derivatives; n = 5–10 per group. v Mean ΔV_0.5act for KCNQ2/3 in response to carnosic acid and derivatives; n = 5–10 per group. KCNQ3* response to carnosic acid (grey; from (u)) shown for comparison. Error bars indicate SEM. n indicates number of biologically independent oocytes. Statistical comparisons by one-way ANOVA.