Coordination of fictive motor activity in the larval zebrafish is generated by non-segmental mechanisms.

Timothy D Wiggin,Jack H Peck,Mark A Masino,Filippo Del Bene

doi:10.1371/journal.pone.0109117

Abstract

The cellular and network basis for most vertebrate locomotor central pattern generators (CPGs) is incompletely characterized, but organizational models based on known CPG architectures have been proposed. Segmental models propose that each spinal segment contains a circuit that controls local coordination and sends longer projections to coordinate activity between segments. Unsegmented/continuous models propose that patterned motor output is driven by gradients of neurons and synapses that do not have segmental boundaries. We tested these ideas in the larval zebrafish, an animal that swims in discrete episodes, each of which is composed of coordinated motor bursts that progress rostrocaudally and alternate from side to side. We perturbed the spinal cord using spinal transections or strychnine application and measured the effect on fictive motor output. Spinal transections eliminated episode structure, and reduced both rostrocaudal and side-to-side coordination. Preparations with fewer intact segments were more severely affected, and preparations consisting of midbody and caudal segments were more severely affected than those consisting of rostral segments. In reduced preparations with the same number of intact spinal segments, side-to-side coordination was more severely disrupted than rostrocaudal coordination. Reducing glycine receptor signaling with strychnine reversibly disrupted both rostrocaudal and side-to-side coordination in spinalized larvae without disrupting episodic structure. Both spinal transection and strychnine decreased the stability of the motor rhythm, but this effect was not causal in reducing coordination. These results are inconsistent with a segmented model of the spinal cord and are better explained by a continuous model in which motor neuron coordination is controlled by segment-spanning microcircuits.

Highlights

Locomotion in vertebrates is organized by spinal neural circuits called central pattern generators (CPGs) that are capable of driving patterned motor neuron output even in the absence of patterned synaptic input [1]
There was no main effect of recording in the same-segment versus different segments on phase consistency (Two-way ANOVA; F(1, 27) = 1.3; p = 0.27; Fig. 5E), but a post-hoc test did show that Middle-5 same segment recordings are significantly more coordinated than Middle-5 different segment recordings (Corrected t-test; t = 2.2; p = 0.04). These results indicate that motor neuron coordination within a body segment is significantly reduced by spinal transection, which is inconsistent with the segmental CPG model
Our findings are inconsistent with larval zebrafish having a segmented locomotor CPG

Summary

Introduction

Locomotion in vertebrates is organized by spinal neural circuits called central pattern generators (CPGs) that are capable of driving patterned motor neuron output even in the absence of patterned synaptic input [1]. Lesion studies have demonstrated that rhythmic, coordinated motor output can be evoked using tonic excitatory drive from as few as 2 spinal intact segments in chicks [13], rats [14], lamprey [15], and salamanders [16]. Complementing this experimental evidence, computational models of the lamprey locomotor CPG that use spinal segments as the units of circuit reiteration accurately reproduce swimming output [5,17]

Methods

Results

Discussion

Conclusion