Changing role of forebrain astrocytes during development, regenerative failure, and induced regeneration upon transplantation.

When the cerebral midline is lesioned in the embryo or neonate, the would-be callosal axons form neuromas. We have shown that an untreated Millipore implant inserted between the neuromas in young acallosal animals can support the migration of immature astrocytes that, in turn, support the de novo growth of commissural axons between the hemispheres. Since callosal neuromas persist into adulthood, we asked whether a critical period exists after which reactive glia no longer promote axon growth. We found that a critical period does exist and have documented a variety of changes in reactive gliosis that, in part, may lead to the axon growth-refractory state. In acallosal mouse postnates given untreated implants on or prior to day 8, glial fibrillary acidic protein (GFAP)+, stellate-shaped astrocytes migrated and attached to the implant by inserting foot processes into the pores of the filter. This form of gliotic response established an axon growth-promoting substratum within 24-48 hours after implantation. During this critical stage there was no evidence of scar formation or necrosis at or around the implant surface. However, when acallosal mice were implanted on or later than postnatal day 14, extensive tissue degeneration occurred, and a mixed population of astrocytes and fibroblasts invaded the surface of the filter, producing a dense scar. Reactive cells within the scar did not promote axonal outgrowth. To determine whether glia from neonates can influence the host environment and/or induce axonal regeneration in acallosal animals after the critical period, we harvested immature astrocytes on Millipore from critical-period mouse forebrains and transplanted the glia-coated prostheses into the brains of post-critical-period acallosal animals. Such transplants reduced glial scarring in the host, inhibited extensive bleeding and secondary necrosis, and promoted axonal regeneration. Our studies suggest that when controlled with a prosthesis, gliosis during the critical period is a beneficial process that can promote the reconstruction of malformed axon pathways; that in older animals a variety of changes in reactive glia and the extracellular matrix may work together to hinder axon regeneration after the critical period; and that axonal regeneration in the postcritical CNS may be stimulated by reintroducing an immature glial environment at the lesion site.