The stages of exuberant growth described above probably apply to other cortical axons (see Innocenti, 1991, 1995). Therefore, this mode of development raises several general questions about the mechanisms regulating the exuberant axonal growth as well as the selection for maintenance versus elimination of juvenile axonal branches and/or synapses.
The available information is mainly based on retrograde transport studies of the callosal axons and, to a lesser extent, of terminating callosal axons but using techniques that do not resolve the individual axons. Most of the relevant studies have been reviewed elsewhere (Innocenti, 1991). Only some essential findings will be summarized here, in particular some recent analysis of single callosal axons in visually deprived kittens (Zufferey et al., 1999).
Visual Activity Thus far, this is the most intensely explored factor regulating the development of callosal connections. In particular, deprivation of vision by bilateral suture of the eyelids appears to delete a large fraction of the callosal connections that are normally maintained, in addition to those that are normally eliminated (reviewed in Innocenti, 1991; see also Boire et al., 1995). Furthermore, the callosal axons that are not eliminated show severely stunted terminal arbors (Zufferey et al., 1999). Eye enucleation, thalamectomy, or lesion of thalamic radiation also plays a role (reviewed in In-nocenti, 1991; see also Fish et al., 1991; Miller, Wind-rem, and Finlay, 1991).
On the whole, the maintenance or elimination of the juvenile callosal connections seems to require a signal conveyed via thalamic afferents. This signal might involve concurrent activation of thalamic and callosal afferents, as appears to be the case for intra-areal connections (Lowel and Singer, 1992; Schmidt et al., 1997). The importance of information from the retina in the development of callosal connections was stressed in a recent study in which redirecting retinal afferents to the auditory thalamus caused reorganization of the callosal connections in the auditory areas (Pallas, Littman, and Moore, 1999).
The mode of action of visual activity remains to be understood. Manipulation of visual activity thus far has failed to maintain the majority of the transient juvenile axons. Perhaps visual activity cannot override cellular specificities based on biochemical markers. However, the role of spontaneous activity generated along the visual pathways or by other cortical afferents has never been explored.
Interestingly, visual activity seems to modify the fate of juvenile callosal axons very early, as soon as they are establishing the very first synaptic boutons. Indeed, normal vision until the age of two weeks—that is, extending into the beginning of synaptogenesis—is sufficient to prevent the loss of callosal connections that normally follows binocular deprivation by eyelid suture (Innocenti, Frost, and Illes, 1985). At the end of the first postnatal week the arbors of callosal axons have not fully differentiated. Therefore, normal vision might play a constructive role in the differentiation of the arbor by validating or eliminating synapses soon after they have formed (discussed in Aggoun-Zouaoui et al., 1996). Consistent with this possibility, it was found that in kittens that were binocularly deprived of vision by eyelid suture, the differentiation of callosal axons interconnecting visual areas 17 and 18 of the two hemispheres is arrested at the early stages of intracortical branching and synaptogenesis (Zufferey et al., 1999). Nevertheless, surprisingly, if vision is allowed until these early stages, the process of axonal differentiation, including an almost normal branching and synaptogenesis, can proceed in the absence of vision (Zufferey et al., 1999).
Asymmetric Development of the Hemispheres Interestingly, when the two hemispheres develop differently from each other, the callosal connections can develop asymmetrically. This was first demonstrated by Cynader, Lepore, and Guillemot (1981) in kittens that were raised with chiasmatic transection and monocular occlusion. This resulted in the development of cal-losal connections from the seeing hemisphere to the hemisphere that receives from the occluded eye and decreased connections in the opposite direction. The consequences of unilateral thalamectomy or of unilateral lesion of the optic tract were studied in different systems and species (Cusick and Lund, 1982; Melzer, Rothblatt and Innocenti, 1987; Rhoades et al., 1987; Koralek and Killackey, 1990; Miller et al., 1991). The results are compatible with the hypothesis that the development of callosal connections is modified by information coming from the periphery through the thalamus. Unfortunately, the unavoidable transection of corticothalamic fibers and of other corticofugal and corticopetal axons weakens the interpretation. Interestingly, in all the above-mentioned studies, callosal connections developed asymmetrically. Asymmetrical callosal connections were also obtained after unilateral early lesions of the cortical gray matter of areas 17 and 18 produced with perinatal injections of ibotenic acid in the cat (Inno centi and Berbel, 1991). The injections deleted the in-fragranular layers as well as, partially, layer III, resulting in a cytoarchitectonic abnormality similar to congenital microgyria in humans. Callosal connections from the injured cortex to the normal cortex were also deleted. Instead, connections in the opposite direction were maintained. Transient callosal connections from the auditory to the injured visual cortex were also maintained.
The possibility that competition between callosal and other corticocortical axons might regulate the fate of the former was tested with early lesions of the somatosensory areas (Caminiti and Innocenti, 1981). It was found that a small number of transient callosal projections from SI to contralateral SII could be maintained when the target SII was deprived of association afferents from ipsilateral SI. This was not obtained with the deprivation of callosal afferents from contralateral SII. These results brought evidence compatible with the view that competition with other corticocortical afferents might regulate the selection of callosal connections. Nevertheless, a reorganization of the thalamocortical connections as a cause could not be excluded.
The formation of callosal connections after experimental manipulations that affect hemispheric symmetry might provide tools for understanding the development of interhemispheric interactions in lateralized brains such as the human (Rosen, Sherman, and Gallaburda, 1989).
Was this article helpful?