Figures and data
Kv2.1/Kv8.1 heteromers are resistant to RY785 and sensitive to GxTX
A) Exemplar traces from a voltage-clamped Kv2.1-CHO cell transfected with Kv8.1. Black and red traces are currents before and after application of 1 µM RY785 respectively. Brown trace is current after subsequent application of 1 μM RY785 and 100 nM GxTX.
B) Exemplar traces from a Kv2.1-CHO cell transfected with Navβ2.
C) Exemplar traces from a Kv2.1-CHO cell transfected with AMIGO1.
D) Current remaining after application of 1 μM RY785 or 1 μM RY785 + 100 nM GxTX. Bars represent mean. Each point represents current from one cell at the end of a 200 ms voltage step to -9 mV. Dunnett tests with Kv8.1 + RY785 as control.
RY785 blocks Kv2.1/Kv8.1 heteromers in a concentration-dependent manner.
A) Current amplitudes during an RY785 concentration-effect experiment on a Kv2.1-CHO cell transfected with Kv8.1. Circles represent tail current amplitudes 2-4 ms into a step to –9 mV following a 200 ms activating step to 71 mV. Voltage protocol was repeated in 5 s intervals. Solution exchanges occurred during the gaps in the time axis. For exemplar current traces, see Figure 2 Supplement and Figure 3.
B) Mean normalized tail current amplitudes with increasing concentrations of RY785. Error bars represent SEMs. Black curve is a fitted Hill function with nH = 1 (IC50 = 5.1 ± 1.0 μM, base = 1.0 ± 0.1 %).
C) Vehicle control tail current with repeated solution exchanges (washes) into 0.35 μM RY785, mimicking solution exchanges in panel B. Vehicle control solution exchanges were followed by exchange into 35 μM RY785 (wash #5). Error bars represent SEMs from n = 4 cells.
D) Tail current recovery following solution exchange from 35 μM RY785 into 0.35 μM RY785 (washout). Bars represent mean current amplitudes from n = 5 cells.
RY785 can affect Kv2.1/Kv8.1 current kinetics. 0.1 1 10 100 [RY785] (μM)
A) Kinetics of currents from a Kv2.1-CHO cell transfected with Kv8.1 are altered by RY785. Traces normalized to max.
B) Latency to peak current during steps to +71 mV. The time axis of this plot is aligned with that of Panel A. Bars represent means. Unpaired Wilcoxon rank tests.
C) Time constant of deactivation at -9 mV is constant after washes with 0.35 μM RY785. Time constants are derived from fits of a monoexponential function (Equation 3 with A2 set equal to 0) to tail currents like those shown in Panel A. Fits were from the peak of each tail current to 200 ms after the voltage step. Brown and Forsythe test p = 0.98. ANOVA p = 0.98. Statistics were performed on natural logarithms of time constants. n = 4 cells.
D) RY785 can alter time constant of deactivation. Bars represent means. Brown and Forsythe test p = 3×10-12. Unpaired Welch test p = 1×10-7. Dunnett tests with 0.35 μM RY785 (initial) as control. Unpaired Wilcoxon rank test comparing 35 μM RY785 to 0.35 μM RY785 (washout) p = 0.001. Statistics were performed on natural logarithms of time constants.
RY785-resistant current is consistent with Kv2.1/Kv8.1 heteromers
Kv2.1/8.1 data (purple) are from Kv2.1-CHO cells transfected with Kv8.1, and are in 1 μM RY785. Kv2.1/control (black) were transfected with Navβ2, and are in vehicle control solution. Before measurements, repeated voltage steps to -9 mV were given until currents stabilized. p values are from two-tailed unpaired Wilcoxon rank test.
A) Exemplar currents during a step to -9 mV.
B) Time constants from exponential fit (Equation 1). Bars represent geometric mean.
C) Sigmoidicity from exponential fit (Equation 1). Bars represent geometric mean.
D) Conductance-voltage activation relation. Conductance was measured from initial tail currents at -9 mV. Mean ± SEM. Kv2.1/Kv8.1 n = 7 cells Kv2.1 n = 6 cells. Lines are Boltzmann fits (Equation 2) (Kv2.1/Kv8.1: V1/2 = 6 ± 1 mV, z = 1.6 ± 0.1 e0; Kv2.1/control: V1/2 = -6.3 ± 1 mV, z = 1.7 ± 0.1 e0).
E) Activation V1/2 values from individual cells. Bars represent means.
F) Activation z values. Bars represent means.
G) Exemplar currents during a 10 s step to -9 mV.
H) Percent of current inactivated after 10 seconds at -9 mV. Bars represent means.
I) Steady state currents at -9 mV after holding at indicated voltages for 10 seconds. Normalized to the max and min. Mean ± SEM. Kv2.1/Kv8.1 n = 5 cells Kv2.1 n = 5 cells. Lines are Boltzmann fits (Equation 2) (Kv2.1/Kv8.1: V1/2 = -66 ± 1 mV, z = 1.8 ± 0.1 e0; Kv2.1/control: V1/2 = -54.7 ± 0.8 mV, z = 3.1 ± 0.3 e0).
J) Inactivation V1/2 values from individual cells. Bars represent means.
K) Inactivation z values. Bars represent means.
A subunit from each KvS subtype is resistant to RY785
A) Exemplar traces from a voltage-clamped Kv2.1-CHO cell transfected with Kv5.1. Black and red traces are currents before and after application of 1 µM RY785 respectively. Brown trace is current after subsequent application of 1 μM RY785 + 100 nM GxTX
B) Exemplar traces from a Kv2.1-CHO cell transfected with Kv6.4.
C) Exemplar traces from a Kv2.1-CHO cell transfected with Kv9.3.
D) Current remaining after application of 1 μM RY785 or 1 μM RY785 + 100 nM GxTX. Bars represent mean. Each point represents current from one cell at the end of a 200 ms voltage step to -9 mV. Unpaired Wilcoxon rank tests.
The Kv2 conductances of mouse superior cervical ganglion neurons do not have KvS-like pharmacology.
A) Exemplar currents from a voltage-clamped SCG neuron. Black and red traces are currents before and after application of 1 µM RY785 respectively. Brown trace is current after subsequent application of 1 μM RY785 + 100 nM GxTX.
B) Tail current amplitude 10 ms after voltage was stepped from +5 mV to -45 mV normalized to current amplitude before RY785. Paired Wilcoxon rank tests, n = 7 neurons, N = 3 mice.
C) Subtracted currents from A. Kv2-like current is the RY785-sensitive current (black trace minus red in A). KvS-like current is the GxTX-sensitive current remaining in RY785 (red trace minus brown in A).
D) Conductance-voltage activation relation of Kv2-like current in SCG neurons. Conductance was measured from tail currents at -45 mV. V1/2 = -11 ± 1 mV, z = 2.1 ± 0.2 e0 Mean ± SEM. n = 7 neurons, N = 3 mice.
E) The faster time constant of a double exponential (Equation 3) fit to channel deactivation at -45 mV. Bar represents mean.
The Kv2 conductances of mouse dorsal root ganglion neurons have KvS-like pharmacology.
A) Exemplar currents from nonpeptidergic nociceptors, GFP+ neurons from MrgprdGFP mice.
B) Tail current amplitude 10 ms after voltage was stepped from +6 mV to -44 mV normalized to current amplitude before RY785 or vehicle treatment. Wilcoxon rank tests were paired, except the comparison of RY785 to vehicle which was unpaired. RY785 then GxTX: n = 7 neurons, N = 4 mice. Vehicle then GxTX: n = 6 neurons, N = 4 mice.
C) Exemplar subtracted currents from A. Kv2-like is the initial current minus RY785 (black trace minus red in A left panel). KvS-like is the current in RY785 minus GxTX (red trace minus brown in A left panel). Kv2-like+KvS-like is the current in vehicle minus RY785 + GxTX (blue trace minus brown in A right panel).
D) Voltage dependance of activation of subtraction currents in MrgprdGFP+ neurons. Pink points represent Kv2-like currents, brown points represent KvS-like currents, and light blue points represent Kv2+KvS-like currents after vehicle treatment. Conductance was measured from initial tail currents at -44 mV. Kv2-like: V1/2 = -18 ± 1 mV, z = 2.7 ± 0.3 e0, KvS-like: V1/2 = -18 ± 1 mV, z = 3 ± 0.2 e0, Kv2 + KvS-like: V1/2 = -19 ± 1 mV, z = 2.9 ± 0.1 e0. Mean ± SEM. KvS-like and Kv2-like n = 7 neurons N = 4 mice, Kv2+KvS-like n = 6 neurons N = 4 mice.
E) The faster time constant of a double exponential fit (Equation 3) to channel deactivation at -44 mV. p value represents paired Wilcoxon rank test.
F) Fractional KvS-like conductance relative to the total RY785 + GxTX-sensitive conductance. KvS-like is only sensitive to GxTX. Bar represents mean.
The Kv2 conductances of human dorsal root ganglion neurons have KvS-like pharmacology.
A) Exemplar currents from human dorsal root ganglion neurons.
B) Tail current amplitude 10 ms after voltage was stepped from +6 mV to -44 mV normalized to current amplitude before RY785 or vehicle treatment. Wilcoxon rank tests were paired. RY785 then GxTX: n = 3 neurons. Vehicle then RY785 and GxTX: n = 4 neurons. All neurons from same human.
C) Exemplar subtracted currents from A. Kv2-like is the initial current minus RY785 (black trace minus red in A left panel). KvS-like is the current in RY785 minus GxTX (red trace minus brown in A left panel). Kv2+KvS-like is the current in vehicle minus RY785 + GxTX (blue trace minus brown in A right panel).
D) Voltage dependence of activation of subtraction currents in human dorsal root ganglion neurons. Pink points represent Kv2-like currents, brown points represent KvS-like currents, and blue points represent Kv2 + KvS-like currents after vehicle treatment. Conductance was measured from initial tail currents at -44 mV. Mean ± SEM. Kv2 + KvS-like n = 3 neurons N = 1 human, KvS+Kv2-like n = 4 neurons N = 1 human.
E) Fractional KvS-like conductance relative to the total RY785 + GxTX-sensitive conductance. KvS-like is only sensitive to GxTX. Bar represents mean.
Structural models of RY785 docked to a Kv2.1 homomer and 3:1 Kv2.1:Kv8.1 heteromer.
A) Model of RY785 docked to a Kv2.1 homomer (PDB 8SD3). Left: transmembrane view. Right: Extracellular view. Individual subunits colored distinctly. RY785 represented by the Corey– Pauling–Koltun space-filling model and coloring scheme. Potassium (purple spheres) from PDB 8SD3 are shown in selectivity filter.
B) A zoom in of central cavity in transmembrane view. Colors indicate subunit of residues making Van der Waals contacts with RY785. Human Kv2.1 sequence numbering. Asterisks indicate residues that vary in Kv8.1.
C) Model of RY785 docked to a 3:1 Kv2.1:Kv8.1 heteromer. RY785 makes no contact with Kv8.1.
D) A replica of panel B with heatmap coloring to indicate sequence conservation of Kv2.1 to KvS subunits.
E) Sequence comparison of pore-lining S6 residues of Kv2.1 and all human KvS subunits. Zappo coloring indicates physiochemical identity. A dot indicates sequence conservation with Kv2.1.
KvS subunits colocalize with Kv2.1 on the surface of CHO cells.
Kv2.1-CHO cells transfected with the designated KvS subunits (c = no KvS transfection) were immunolabeled for Kv2.1 (green) and Kv5.1 or Kv9.3 (magenta). Immunolabeling for Kv5.1 and Kv9.3 were detected both intracellularly and on the apparent cell surface where they colocalized with Kv2.1 labeling. The anti-Kv5.1 mAb recognizes an extracellular epitope and was used on non-permeabilized cells, confirming surface expression of Kv5.1. Scale bars = 5 μm.
Rapid unblock of RY785 from Kv2.1/Kv8.1 heteromers
A) Exemplar traces showing current recovery following solution exchange from 35 μM RY785 into 0.35 μM RY785. Voltage protocol is the same as in Figure 2. Cells were held at -89 mV without pulsing during solution exchange.
B) Tail current amplitudes, as in Figure 2, normalized to pulse #5. n = 5 cells treated first with 35 μM RY785 (dark red) then 0.35 μM RY785 (orange). n = 4 control cells treated first with 0.35 μM RY785 and washed again with 0.35 μM RY785 (pink). The transient increase in current amplitudes upon solution exchange also occurs with Kv2.1 (Marquis and Sack, 2022).
Effect of vehicle control on Kv2.1.
Left: Exemplar traces from a Kv2.1-CHO cell transfected with Navβ2. Black and green traces are currents before and after application of vehicle control respectively. Right: Current remaining after application of vehicle control. Black bar represents mean. Each point represents current from one cell at the end of a 200 ms -9 mV voltage step.
Nonpeptidergic nociceptors express Kv5.1 and Kv9.1 mRNA transcripts.
Exemplar images of DRG sections from a MrgprdGFP mouse labeled with RNAscope in situ hybridization for KCNF1 (Kv5.1) (top magenta) or KCNS1 (Kv9.1) (bottom magenta). Scale bars are 50 μm.
RY785 resistant currents from Kv2.1-CHO cells transfected with Kv5.1 or Kv9.1 deactivate slower than currents from untransfected Kv2.1-CHO cell.
A) Exemplar traces of channel deactivation at -49 mV after a 50 ms step to +11 mV. Traces are normalized to max current during the -49 mV step.
B) The faster time constant of a double exponential fit (Equation 3) to channel deactivation. Dunnett tests versus control (ctl). Bars represent means. Ctl n = 7. Kv9.1 n = 4. Kv5.1 n = 6.
The Kv2 conductances of mouse dorsal root ganglion neurons have KvS-like pharmacology in the absence of the cocktail of inhibitors.
A) Exemplar currents from peptidergic nociceptor, GFP+ neurons from CGRPGFP mice. Inward sodium currents are cropped so that only outward currents are shown.
B) Tail current amplitude 10 ms after voltage was stepped from +6 mV to -44 mV normalized to current amplitude before RY785. Wilcoxon rank tests were paired. RY785 then GxTX: n = 9 neurons, N = 3 mice.
C) Time course of inhibition of tail currents in A in control and during drug application.
D) Fractional KvS-like conductance relative to the total RY785 + GxTX-sensitive conductance. KvS-like is only sensitive to GxTX. Bar represents mean.
