Two types of V2a neuron distinguished by intrinsic properties. a Schematics illustrating multi-layer (top) and single-layer (bottom) models of connectivity underlying timing and amplitude control. + excitatory interneurons, – inhibitory interneurons, MN motor neurons, L left side, R right side. Arrows are glutamatergic connections and the resistor symbol denotes electrical coupling. b Top, schematic illustrating a larval zebrafish viewed from the side. Region at midbody (segments 10–16) where recordings were performed is shaded. Rostral, left; dorsal, up. Bottom, schematic illustrating muscle segments and each of the four identified primary motor neurons per spinal hemisegment (dotted circles) that innervate either epaxial (pI-II) or hypaxial (pIII-IV) muscle (horizonal blue or brown lines). hm, horizontal myoseptum. c Single confocal optical section from one spinal segment illustrating type I (purple arrowheads) and type II (green arrowheads) V2a neurons targeted for recordings. Soma positions are normalized to the dorsal and ventral aspects of spinal cord (d-v). Rostral is to the left. d Top, contrast inverted epifluorescent images of fills after electrophysiological recordings. Bottom, reconstructions illustrating the trajectory of the main axon. The descending axon of the type I V2a neuron continues out of the field of view (arrowhead). Scale to the left indicates normalized dorso-ventral soma position. e As in panel d, but for a bifurcating type II V2a neuron. In top image, black arrowheads mark axon bifurcation point and the continuation of the ascending axon out of the field of view. f Quantification of input resistance (Rin) and rheobase according to V2a neuron type. Type I V2a (n = 105), type II V2a (n = 145). Source data are provided in a Source Data file. g Top, instantaneous firing frequencies are plotted above the respective recordings from a type I V2a neuron. Bottom, examples of voltage deflections in response to a hyperpolarizing current step (–25 pA) and firing responses to current steps around threshold (left trace) and just above threshold (right trace). Voltage; I, current. Source data are provided in a Source Data file. h, As in panel g, but for a type II V2a neuron

V0d inhibitory interneuron recruitment and targeting by V2a types. a Single confocal optical section from one spinal segment in compound transgenic larvae illustrating the location of V0ds targeted for recordings (yellow arrowheads). Note, only GFP fluorescence is illustrated. Scale to the left indicates normalized dorso-ventral soma position. b Contrast inverted epifluorescent images after recording illustrating the ipsilateral soma (ipsi) and contralateral axon (contra). The axon crosses at the white arrowheads. Soma position is normalized to dorso-ventral aspects of spinal cord, as in panel a. c Simultaneous current-clamp recordings from a V0d neuron (V) and extracellular recordings of peripheral motor nerves (pn) during fictive swimming following a brief electrical stimulus to the skin (open arrowhead). Filled circles highlight suprathreshold swimming frequencies, while open circles highlight examples of subthreshold frequencies. Spikes are truncated to better reveal the underlying synaptic drive. Dashed line indicates resting membrane potential, −54 mV. d Quantification of the reliability of firing as a function of swimming frequency for V0d neurons (n = 26). Values are superimposed on V2a data from Fig. 2c. Data are reported as means ± standard deviations. Source data are provided in a Source Data file. e Left, quantification of the mode of V2a synaptic connectivity to V0d neurons expressed as a percentage of total responses (I to V0d, n = 11; II to V0d, n = 17). e electric, m mixed, g glutamatergic, s slow. Right, quantification of the peak amplitudes of fast EPSPs (electric, mixed, or glutamatergic) measured from type I V2a to V0d neurons (purple, n = 11) and type II V2a to V0d neurons (green, n = 10). * significant difference following Mann–Whitney U-test, U(19) = 84, p < 0.05. Bars represent means. Source data are provided in a Source Data file