Search

Background: Trigeminal sensory neurons detect thermal and mechanical stimuli in the skin through their elaborately arborized peripheral axons. We investigated the developmental mechanisms that determine the size and shape of individual trigeminal arbors in zebrafish and analyzed how these interactions affect the functional organization of the peripheral sensory system., Results: Time-lapse imaging indicated that direct repulsion between growing axons restricts arbor territories. Removal of one trigeminal ganglion allowed axons of the contralateral ganglion to cross the midline, and removal of both resulted in the expansion of spinal cord sensory neuron arbors. Generation of embryos with single, isolated sensory neurons resulted in axon arbors that possessed a vast capacity for growth and expanded to encompass the entire head. Embryos in which arbors were allowed to aberrantly cross the midline were unable to respond in a spatially appropriate way to mechanical stimuli., Conclusions: Direct repulsive interactions between developing trigeminal and spinal cord sensory axon arbors determine sensory neuron organization and control the shapes and sizes of individual arbors. This spatial organization is crucial for sensing the location of objects in the environment. Thus, a combination of undirected growth and mutual repulsion results in the formation of a functionally organized system of peripheral sensory arbors.

Trigeminal sensory axons project to several epithelial targets, including those of the maxillary and mandibular processes. Previous studies identified a chemoattractant activity, termed Maxillary Factor, secreted by these processes, which can attract developing trigeminal axons in vitro. We report that Maxillary Factor activity is composed of two neurotrophins, neurotrophin-3 (NT-3) and Brain-Derived Neurotrophic Factor (BDNF), which are produced by both target epithelium and pathway mesenchyme and which are therefore more likely to have a trophic effect on the neurons or their axons than to provide directional information, at least at initial stages of trigeminal axon growth. Consistent with this, the initial trajectories of trigeminal sensory axons are largely or completely normal in mice deficient in both BDNF and NT-3, indicating that other cues must be sufficient for the initial stages of trigeminal axon guidance.

The molecules of the collapsin/semaphorin gene family have been thought to play an essential role in axon guidance during development. Semaphorin III/D is a member of this family, has been shown to repel dorsal root ganglion (DRG) axons in vitro, and has been implicated in the patterning of sensory afferents in the spinal cord. Although semaphorin III/D mRNA is expressed in a wide variety of neural and nonneural tissues in vivo, the role played by semaphorin III/D in regions other than the spinal cord is not known. Here, we show that mice homozygous for a targeted mutation in semaphorin III/D show severe abnormality in peripheral nerve projection. This abnormality is seen in the trigeminal, facial, vagus, accessory, and glossopharyngeal nerves but not in the oculomotor nerve. These results suggest that semaphorin III/D functions as a selective repellent in vivo.

GAP-43 has been termed a "growth" or "plasticity" protein because it is expressed at high levels in neuronal growth cones during development and during axonal regeneration. By homologous recombination, we generated mice lacking GAP-43. The mice die in the early postnatal period. GAP-43-deficient retinal axons remain trapped in the chiasm for 6 days, unable to navigate past this midline decision point. Over the subsequent weeks of life, most GAP-43-deficient axons do enter the appropriate tracts, and the adult CNS is grossly normal. There is no evidence for interference with nerve growth rate, and cultured neurons extend neurites and growth cones in a fashion indistinguishable from controls. Thus, the GAP-43 protein is not essential for axonal outgrowth or growth cone formation per se, but is required at certain decision points, such as the optic chiasm. This is compatible with the hypothesis that GAP-43 serves to amplify pathfinding signals from the growth cone.