Mechanisms of Robo1/2-mediated motor neuron cell body and axon positioning relative to the spinal cord midline

点击翻译按钮获取中文摘要

PROJECT SUMMARY: Neurons are wired up during development through the action of attractive and repulsive molecular cues that guide migrating neuronal cell bodies and axons to their destinations by activating receptors on the leading process of the neuron, the growth cone. Growth cones are frequently exposed to both attractive and repulsive cues at the same time, and axonal responses to these signals are dictated by one of two main modes of cue integration: parallel or hierarchical signaling. In parallel signaling, there are few interactions between signaling pathways, and growth cone behavior is dictated by a sum of total attractive and repulsive signals from different directions. Hierarchical cross-talk, however, relies heavily on the intersection of signaling cascades, and the response to select cues can dominate growth cone behavior by silencing signaling from other cues. The receptors Robo1 and Robo2 mediate repulsion from secreted guidance cues of the Slit family, and DCC is a receptor for the attractive ligand Netrin-1. Previous research on the guidance of spinal commissural axons supports the idea that Slit-Robo1/2 signaling can silence DCC-mediated attraction to Netrin-1. Thus, Robo signaling might dominate over DCC signaling in commissural neurons, but whether this hierarchy of the Robo and DCC pathways is at work in other cell types has remained elusive. Recent studies have identified several defects in spinal motor neuron migration and motor axon guidance of Robo1/2 double mutant mice. In mice lacking Robo1 and Robo2, motor neuron cell bodies aberrantly migrate from the ventral horn into the commissure, and motor axons cross the spinal cord midline. The floor plate at the ventral midline expresses both Netrin-1 and Slits, and it has remained unclear whether motor neuron invasion of the midline in Robo1/2 mutant mice is caused by a loss of Slit repulsion or a gain of Netrin-1 attraction due to absence of Robo1/2-mediated DCC silencing. Here, we address this question by combining mouse genetics, in vivo phenotype analysis, and in vitro motor neuron migration and axon guidance assays. Our results demonstrate that at least some of the Robo1/2 mutant motor neuron phenotypes are caused by a loss of Slit-mediated midline repulsion, not a gain of Netrin-1 attraction. Hence, Slit-Robo1/2 and Netrin-1-DCC can signal in parallel in motor neurons.