Here, we review our studies in rats of target recognition by developing cortical axons focusing on their innervation of the basilar pons, a major hindbrain target. The corticopontine projection develops by a ‘delayed interstitial budding’ of collaterals from layer 5 corticospinal axons, rather than by a direct ingrowth of primary axons or by bifurcation of the growth cone. Branches form de novo from the axon cylinder in the pathway overlying the basilar pons and extend directly into it. Cocultures of cortex and basilar pons in 3-dimensional collagen matrices show that a diffusible chemotropic signal released by the basilar pons directs the growth of collateral branches from layer 5 axons in a target and neuron specific manner. ‘Delayed’ co-cultures suggest that a diffusible, pontine-derived signal also initiates the selective branching of layer 5 axons. In vivo experiments support this chemotropic mechanism. First, corticospinal axons form collateral branches at novel locations directly over ectopic aggregations of basilar pontine neurons induced by X-irradiation; no branches form at positions that would normally overlie the appropriate region of basilar pons which is absent because of the X-irradiation. Thus, the basilar pons, rather than local cues in the axon pathway, appears to control the location of corticospinal axon branching. Second, in a series of experiments in which different subsets of corticospinal axons are prevented from innervating the basilar pons, remaining corticospinal axons extend collaterals in a directed manner to regions of the basilar pons deprived of cortical input, a behavior consistent with a response to a diffusible chemoattractant emanating from these regions. In conclusion, our findings suggest that a diffusible, target-derived chemotropic molecule(s) underlies target recognition in this developing system by initiating the formation and directing the growth of pontine collateral branches of primary layer 5 corticospinal axons.
GENETIC CONTROL OF IMMUNE RESPONSES IN VITRO