Kv3 Channels Contribute to the Excitability of Subpopulations of Spinal Cord Neurons in Lamina VII

E-type classification of lumbosacral spinal neurons in Lamina VII. A, Neurons were clustered according to firing phenotype at maximal firing; steady state firing frequency, adaptation index, interspike interval coefficient of variation (ISI CV), delays, bursts, pauses, and accommodation were used as clustering variables. Upper panel, Principle component scatter plot summarizing clusters. Lower panel, Parallel plot comparison across clustering variables, error bars represent SEM. B, Frequency-current plot summarizing the firing frequency elicited by 1-s current steps from −70 mV for different e-types, note the fast-firing e-type 3. C, Four different e-types: steady regular, continuous adapting, fast bursting, and steady delayed. Example traces for each e-type with 10, 50, 100, and 200 pA 1-s current injections. No firing phenotype was produced for e-type 4 at 10 pA. AP waveforms were also clustered to define AP-types and AP-types were correlated with e-types (Extended Data Fig. 2-1). PCA, principal component analysis; pA, picoAmps; mV, millivolts; Hz, Hertz; AP, Action potential.

Distinct morphologies were associated with each e-type. A, Comparison of the number of neurons with VRP neurites, typically associated with autonomic motoneurons. B, Sholl analysis of reconstructed morphologies; ramification index and critical value between e-types. C, Angles between the end of neurites and the soma were calculated to measure directional specificity (°) and spatial coverage (μm). Note most e-types had a large ventro-lateral projection except for e-type 3. Left panel, Polar plots indicating average dendritic length within 45° bins of dendritic end angle. Middle panel, Kernel density estimates (KDE) indicating the directional likelihood of dendrites for each e-type. Right panel, Representative examples of reconstructed Neurobiotin-filled neurons for each e-type.

E-type 3 accurately followed high frequencies of stimulation. A, 10-ms square current pulses were applied to neurons at different rates (20, 50, and 100 Hz) and at increasing amplitudes. Failure rate was quantified as a failure to fire an AP during a pulse. E-type 3 successfully followed faster stimulation frequencies significantly better than other e-types. B, Representative examples of APs evoked at different stimulation frequencies and at different current amplitudes for each e-type. pA, picoAmps; Hz, Hertz. *p < 0.05; **p < 0.01.

TEA impairs fast firing of e-type 3. A, left panel, Firing frequency-current plots for each e-type (1–4) in control aCSF and in presence of 0.5 mm TEA. Note significant reduction in firing rate at 140 pA in e-type 3. Right panel, Representative traces of neuronal firing in control aCSF and 0.5 mm TEA. B, Firing frequency of each e-type over a range of current amplitudes in control aCSF and in presence of 10 nm DTX.pA, picoAmps; Hz, Hertz; TEA, tetraethylammonium chloride; DTX, dendrotoxin. **p < 0.01.

TEA-sensitive Kv currents greatly contribute to outward currents in e-type 3. A, Plateau outward currents evoked during a voltage step to 14 mV for each e-type in the presence of 0.5 mm TEA and 10 nm DTX. E-type 3 had significantly more TEA-sensitive current than e-type 1. B, left panel, Voltage-current plots for each e-type (1–4). TEA-sensitive and DTX-sensitive (TEA_sens, DTX_sens) currents were obtained by subtracting currents in TEA and DTX from currents in control aCSF, the residual revealing the currents blocked by each compound. Middle panel, Activation plots indicating current activation over voltage steps. Current was measured during deactivating current tails and normalized to Imax. Right panel, Representative traces of outward currents evoked by a voltage step to 14 mV for each e-type. E-type 3 sustained a larger total current and TEA-sensitive current. pA, picoAmps; mV, millivolts; TEA, tetraethylammonium chloride; DTX, dendrotoxin; sens, sensitive; Itail, tail current; Imax, maximum tail current. *p < 0.05.

Kv3 blockade potentiates descending and local inhibitory synaptic responses. A bipolar electrode positioned in the lateral white matter was used to evoke descending synaptic currents. Neurons were held at 0 mV to isolate IPSCs. A, left panel, eIPSC amplitude between control and 0.5 mm TEA (each line represents an individual cell). Right panel, PPR between IPSCs 100 ms apart in 0.5 mm TEA. Bottom panel, Representative example of potentiated IPSC amplitude in TEA. B, Effect of bath application of 10 nm DTX on eIPSC amplitude. C, left panel, Super-resolution Airyscan images showing clear overlap of example synaptic boutons with Kv3.1b IF with GlyT2 IF. right panel, Kv3.1b co-localization with inhibitory markers GlyT2 and VGAT represented as the number of Kv3.1b positive boutons. D, Cumulative (line) and relative frequency (bar) plots of eIPSC (left panel) and sIPSC (right panel) amplitudes in the presence of 0.5 mm TEA and 10 nm DTX. E, Representative traces of eIPSCs (left panel) and sIPSCs (right panel) in control and 0.5 mm TEA. pA, picoAmps; TEA, tetraethylammonium chloride; DTX, dendrotoxin; eIPSC, evoked IPSC;sIPSC, spontaneous IPSC; GlyT2, glycine transporter 2; vGAT, vesicular GABA transporter; IF, immunofluorescence. *p < 0.05; ***p <0.005; ns- non-significant.