La voie olivocérébelleuse (OCP) comprend les fibres grimpantes (CFs), terminaisons axonales des neurones de l'olive inferieure (ION), et leurs cibles, les cellules de Purkinje (PCs). La OCP suit une topographie hautement organisée. A la suite d'une lesion unilatérale de la OCP mature, l'application locale du facteur trophique ‘BDNF’ dans le hemicervelet dénervé induit la reinnervation fonctionnelle des PCs par les CFs. L'objectif de ce travail a était de comprendre les mécanismes activés par le BDNF permettant la plasticité post-lésionnelle dans le OCP mature. Avec un modèle ex vivo chez la souris, nous avons montré que l’injection de BDNF dans le cervelet dénervé augmente la croissance des branchements transverses des CFs intactes. Cette réponse est médiée par l'augmentation de l’expression de Pax3 dans l'ION intact. La surexpression du Pax3 dans l’ION augmente le niveau de PSA-NCAM dans le hemicervelet dénervé, probablement sur les CFs. Cette expression de PSA-NCAM est nécessaire et suffisante pour la réinnervation CF-PC. Nous proposons que la plasticité activée par le BDNF dans l'OCP mature implique le Pax3 et le PSA-NCAM dans l’ION, qui sous-tendent la genèse des branchements des CFs et la reconnaissance correcte des PC dénervés. Pendant le développement de la OCP, la plasticité post-lésionnelle spontanée est plus importante, permettant la compensation anatomique et fonctionnelle. Dans notre modèle ex vivo au stade immature, nous avons montré que cette plasticité spontanée implique l'expression de Pax3 et de PSA-NCAM. Ces résultats suggèrent que la reinnervation post-lésionnelle dans la OCP mature active certains mécanismes de la plasticité développementale. / In the olivocerebellar pathway (OCP) the afferent climbing fibres (CFs), which are the terminal axon projections of the inferior olivary nucleus (ION), innervate cerebellar Purkinje cells (PCs). Following unilateral transection of mature OCP, the addition of the neurotrophic factor BDNF into the denervated cerebellum induces functional CF reinnervation of PCs. What mechanism underlies the BDNF-activated plastic window in the mature OCP and whether recapitulates developmental plasticity remains unknown. Using an optimized ex vivo model of the mouse OCP, we have found that the addition of BDNF into the de-afferented hemicerebellum induces both the outgrowth and elongation of transverse branches from intact CFs. This BDNF-induced plastic response is mediated by the up-regulation of the expression of transcription factor Pax3 in the intact ION. Increased pax3 gene in the ION up-regulates polysialic acid-neural cell adhesion molecule (PSA-NCAM), most likely in the olivocerebellar axons, which was found to be necessary and sufficient for CF reinnervation to PCs. We propose that the BDNF-activated plastic mechanism in the mature OCP involves the afferent Pax3 and PSA-NCAM, which underlies the sprouting of CFs and their appropriate recognition of denervated PCs. Early postnatal OCP shows a spontaneous plasticity following lesion that compensates anatomically and functionally for PC denervation. Using our ex vivo model of the OCP, we found that developmental post-lesion plasticity intrinsically activates and depends on the expression of Pax3 and PSA-NCAM. These results suggest that BDNF treatment in mature OCP reactivates some steps of developmental plasticity mechanisms.

Several seemingly conflicting sets of data have been reported on the regenerative capacity of central noradrenergic neurons, following transection of the ascending noradrenergic fiber tract in neonatal rats (Iacovitti et al., Dev Brain Res 1: 21-33, 1981; Jonsson and Sachs, Brain Res Bull 9: 641-650, 1982). In order to more fully investigate changes in noradrenergic neurons in the brain after such a transection, rats were lesioned at various times after birth, sometimes in conjunction with administration of the neurotoxin, 6-hydroxydopa (6-OHDOPA). Animals were sacrifced at 7, 10, 14, 28, 42 or 56 days after birth, in order to assess the pattern of noradrenergic neuronal damage, as well as the recovery rate. Dorsal bundle lesions were associated with neocortical and hippocampal hypoinnervation by noradrenergic fibers, and sprouting of a collateral fiber group, with production of noradrenergic hyperinnervation of the cerebellum and pons-medulla. Recovery of the norepinephrine (NE) content to control levels occurred in the neocortex at 8 weeks, when the dorsal bundle was lesioned at birth. When the lesion was produced at a later time (3 days or 5 days after birth), less recovery in the neocortex and hippocampus was found. Histofluorescent fiber number, as observed with a glyoxylic acid method, correlated with NE changes. It appears that 6-OHDOPA (20 μg/g IP) does not modify long-term recovery from a dorsal bundle lesion, when rats are co-treated at 3 days after birth. However, the length of the proximal noradrenergic fiber stump may be an important factor affecting the capacity for recovery from injury. These results suggest that a shorter fiber stump, as produced with a dorsal bundle lesion at the level of the pons, may be associated with a greater degree of recovery from injury. Also, the younger the rat at the time of injury, the greater appears to be the capacity for regeneration. These results demonstrate that regeneration can occur in one part of the brain without modification of a collateral hyperinnervation of a different part of the brain. Therefore, our findings discount a programming of central noradrenergic fibers to express a specific number of nerve terminal arborizations, (i.e., the "pruning hypothesis").