Mechanosensory Stimulation Evokes Acute Concussion-Like Behavior by Activating GIRKs Coupled to Muscarinic Receptors in a Simple Vertebrate

KO responses in free-swimming and restrained tadpoles. A, KO responses videoed in a dish (left) with swimming speeds plotted against post-clash swimming lengths (right, 18 responses in six tadpoles, p < 0.01, two-tailed Pearson correlation). Post-clash swimming lasts for 0.33 ± 0.02 s (0.21–0.63 s). Tracing shows the tadpole swimming track. B, Simulating head-on clashes in a physically restrained tadpole (left) and the distribution of post-tap swimming lengths. Gray represents single taps (half-cycle sinewave at 20 Hz) and black represents multiple taps (five sinewave cycles at 20 Hz).

Responses of dINs to head taps. A, A slow tap at 0.125 Hz ends swimming (motor nerve, m.n.) and evokes a mixture of EPSPs and IPSPs (arrow heads in inset) in a dIN on the right side of the hindbrain (r.dIN). B, A fast tap at 2.5 Hz briefly excites another dIN in the right side hindbrain (*) then stops swimming with delayed, prolonged inhibition in the RMP and decreased R_inp (test steps: −100 pA). C, D, RMPs and R_inp in control (unfilled columns) and at the trough period of inhibition after taps. While slow taps at 0.125–0.25 Hz decrease RMPs in 8 dINs (from −52.2 ± 1.9 to −54.4 ± 1.9 mV, gray) without changing the R_inp, fast taps at 2.5–10 Hz in 6 dINs decrease RMPs from −52.6 ± 0.7 to −59.1 ± 2 mV with a reduction in the R_inp by 29.6 ± 8.5% (black, *p < 0.05, paired t test or related sample Wilcoxon signed rank test). Dashed lines indicate RMP in A, B. Inset on the top shows experimental setup.

KO responses evoked by electrical skin stimulation. A, Increasing skin stimulation (arrow, five pulses at 30 Hz) intensity shortens swimming in one tadpole. B, Summary of poststimulation swimming lengths at different stimulation current intensities, capped below 320 µA, as shown in A. Swimming thresholds at rest (Thr.) are 2–14 µA. C, Three stimuli at 40 Hz (arrow, 55 µA) evokes a KO response in one tadpole without evoking any skin impulse, while a single stimulus (80 µA) evokes a skin impulse (arrowhead) without shortening swimming. Insets are stretched from time around stimulation. D, Paired comparison of poststimulation swimming lengths as in C showing their dependence on the number of stimuli (n = 29 pairs from four tadpoles, p < 0.001, related samples Wilcoxon signed rank test). A single stimulus always evokes a skin impulse. E, Experimental setup for the direct stimulation of the trigeminal ganglion (tg). A dorsal view of the CNS and some swimming myotomes are shown (fb, forebrain; mb, midbrain; hb, hindbrain; nll, lateral line nerve; stim., stimulation electrode; oc, otic capsule; sc, spinal cord; m, myotome). Dashed lines indicated severed nerves and oc. F, KO responses evoked by stimulating the tg directly after removing nlls (p < 0.01 in each of five tadpoles, two-tailed independent sample t test). G, Bath application of SR95531, a GABA_AR antagonist, does not affect the KO response in a tadpole. KO stimuli (arrow) consist of five pulses at 30 Hz. H, Summary of SR95531 experiments in six tadpoles (p > 0.05, related samples Friedman’s two-way ANOVA by ranks). I, KO sites mapped with electrical skin stimuli (10 at 30 Hz). Each site has been applied KO stimuli for more than five times intercalated with trials without KO stimuli. Filled circles indicate KO stimuli shortened swimming (p < 0.05, paired t test or related samples Wilcoxon signed rank test) and empty circles indicate KO stimuli have no effect.

Properties of KO inhibition and their relation with the longitudinal location of neurons. A, KO inhibition in a dIN 468 µm from mid/hindbrain border following electrical skin stimulation (five pulses at 30 Hz, arrow). B, KO inhibition after 10 electrical skin stimuli (25 Hz, arrow) in a dIN 670 µm from mid/hindbrain border. Dashed lines indicate RMP in A, B. C, A dIN filled with neurobiotin in the hindbrain (hb, lateral view). Fb, forebrain; mb, midbrain; p, pineal eye. Black dashed lines outline the central nervous system. White dashed line marks the border between mb and hb (0 in E). D, Photo of the cell in C (white rectangle). Arrow head marks a short ascending axon. E, The distribution of somata (circles) and axons (lines) of 13 neurons with >100% peak conductance increases during KO. Dashed vertical line marks the position of obex. F, Longitudinal location of neurons plotted against their peak conductance increases during KO (48 dINs, 25 non-dINs, p < 0.001, two-tailed Spearman rank correlation coefficient 0.58). G, KO inhibition amplitudes plotted against conductance increases (p < 0.0001, correlation coefficient 0.834). In E–G, red symbols represent dINs; black symbols represent non-dINs. Hollow symbols are from tap experiments and solid ones from electrical skin stimulation. H, Increasing the number of electrical skin stimuli (numerals above bars) while maintaining the stimulation strengths and frequencies induces larger rises in the membrane conductance in 10 dINs. The recordings from one dIN are shown in insets.

KO inhibition induced by electrical head skin stimulation (arrows) is mediated by GIRKs. A, A voltage-clamp recording shows the slow outward KO current and its onset (inset, *) after KO stimulation (five pulses at 30 Hz). A voltage ramp rising from −100 to 0 mV is applied at the peak of the current. Membrane potential is held at −60 mV at rest. B, I-V data points averaged from five recordings during the voltage ramp in A for estimating the reversal of the currents and their rectification. C, Effects of microperfusing 50 µM Ba²⁺ on KO inhibition in a dIN. D, Summary of blockade of peak conductance rises by 50–100 µM Ba²⁺ during KO in two dINs and six non-dINs (paired t tests). E, Microperfusing 3 µM Tertiapin-Q (T-Q) weakened the peak conductance increase during KO in one dIN and five non-dINs (paired t tests). F, One dIN recording with a pipette solution containing 10 µM GRK2i at the beginning (control) and after 28 min of equilibration. G, GRK2i weakened the peak conductance increase during KO inhibition (n = 8, paired t test). H, Summary of peak membrane conductance increase during KO in control and after BAPTA equilibration (n = 8, related samples Wilcoxon signed rank test, p = 0.33). *, significance at p < 0.05 and ** at p < 0.01 in D, E, G. Dashed lines indicate RMP or clamping current at −60 mV in A, C, F.

The pharmacological blockade of KO responses and KO inhibition. A, KO responses induced by KO stimulation (arrow, five pulses at 30 Hz) applied every 100 s. Swimming (m.n.) is evoked by dimming an LED. Gray bar indicates time of the application of 10 µM methoctramine (methoc., 100 s). B, Example KO responses before (1), shortly after methoc. application (2), and recovery (3). C, KO responses are blocked 200 s after methoc. application (p < 0.01, n = 9 tadpoles, related samples Wilcoxon signed rank test) but without recovery 25 min after wash. D, KO inhibition following skin stimulation (arrows, five at 30Hz) in a dIN is reduced by the microperfused 10 µM methoc. Swimming is initiated by dimming an LED light. E, Conductance increases during KO inhibition are weakened by methoc. (n = 6, p < 0.05, related samples Wilcoxon signed rank test). F, Summary of post-KO swimming lengths measured at 0, 200, and 2500 s from the application of different antagonists (numerals in brackets indicate number of tadpoles tested, related samples Wilcoxon signed rank tests). Atropine (10–50 µM) is a nonselective muscarinic receptor blocker. PD102807 (0.5 µM) is a selective M₄ muscarinic receptor blocker. CGP54626 (10 µM) is a GABA_B receptor blocker and LY341495 (10 nM) blocks mGluR II and III. *p < 0.05 and **p < 0.01 in C, E, F.

Swimming activity after KO stimulation. A, Distribution of swimming lengths after KO stimulation (range: 0–3.81 s, n = 197 trials). Trials are chosen from 20 tadpoles where membrane conductance during KO has increased by >100% (50/197 without swimming preceding KO). B, Normalized swimming frequency before and after KO stimuli. Gray traces are six trials with frequency increases (p < 0.05, t test, 10 post-KO swimming cycles compared with control). C, Motor responses following KO stimulation (arrow). Four single stimuli (arrow heads, 220 µA) are applied to evoke swimming during KO inhibition. Note failures in dIN spiking. D, Swimming thresholds (normalized to control, dashed line) tested with single skin stimulation after KO. E, Swimming thresholds are higher after KO stimulation and their recovery >5 min later (n = 10 sites in six tadpoles, p < 0.001, related samples Friedman’s two-way ANOVA by ranks). F, Lengths of swimming evoked by single suprathreshold skin stimulation after KO stimulation (filled circles). Empty circles show the same normalized data (secondary axis).