This study demonstrated the involvement of the tumor suppressor protein p53 in differentiation and programmed cell death of neurons and oligodendrocytes, two cell types that leave the mitotic cycle early in development and undergo massive-scale cell death as the nervous system matures. We found that primary cultures of rat oligodendrocytes and neurons, as well as of the neuronal PC12 pheochromocytoma cell line, constitutively express the p53 protein. At critical points in the maturation of these cells in vitro, the subcellular localization of p53 changes: during differentiation it appears mainly in the nucleus, whereas in mature differentiated cells it is present mainly in the cytoplasm. These subcellular changes were correlated with changes in levels of immunoprecipitated p53. Infection of cells with a recombinant retrovirus encoding a C-terminal p53 miniprotein (p53 DD), previously shown to act as a dominant negative inhibitor of endogenous wild-type p53 activity, inhibited the differentiation of oligodendrocytes and of PC12 cells and protected neurons from spontaneous apoptotic death. These findings suggest that p53, upon receiving appropriate signals, is recruited into the nucleus, where it plays a regulatory role in directing primary neurons', oligodendrocytes, and PC12 cells toward either differentiation or apoptosis in vitro.

The poor ability of mammalian central nervous system (CNS) axons to regenerate has been attributed, in part, to astrocyte behavior after axonal injury. This behavior is manifested by the limited ability of astrocytes to migrate and thus repopulate the injury site. Here, the migratory behavior of astrocytes in response to injury of CNS axons in vivo was simulated in vitro using a scratch-wounded astrocytic monolayer and soluble substances derived from injured rat optic nerves. The soluble substances, applied to the scratch-wounded astrocytes, blocked their migration whereas some known wound-associated factors such as transforming growth factor-␤ 1 (TGF- ␤

A covalent dimer of interleukin (IL)‐2, produced in vitro by the action of a nerve‐derived transglutaminase, has been shown previously to be cytotoxic to mature rat brain oligodendrocytes. Here we report that this cytotoxic effect operates via programmed cell death (apoptosis) and that the p53 tumor suppressor gene is involved directly in the process. The apoptotic death of mature rat brain oligodendrocytes in culture following treatment with dimeric IL‐2 was demonstrated by chromatin condensation and internucleosomal DNA fragmentation. The peak of apoptosis was observed 16‐24 h after treatment, while the commitment to death was already observed after 3‐4 h. An involvement of p53 in this process was indicated by the shift in location of constitutively expressed endogenous p53 from the cytoplasm to the nucleus, as early as 15 min after exposure to dimeric IL‐2. Moreover, infection with a recombinant retrovirus encoding a C‐terminal p53 miniprotein, shown previously to act as a dominant negative inhibitor of endogenous wild‐type p53 activity, protected these cells from apoptosis.

Morphogenesis and tissue repair require appropriate cross-talk between the cells and their surrounding milieu, which includes extracellular components and soluble factors, e.g., cytokines and growth factors. The present work deals with this communication needed for recovery after axotomy in the central nervous system (CNS). The failure of CNS axons to regenerate after axonal injury has been attributed, in part, to astrocyte failure to repopulate the injury site. The goal of this work was to provide an in vitro model to mimic the in vivo response of astrocytes to nerve injury and to find ways to modulate this response and create a milieu that favors astrocyte migration and repopulation of the injury site. In an astrocyte scratch wound model, we blocked astrocyte migration by tumor necrosis factor alpha (TNF-alpha). This effect could not be reversed by astrocyte migration-inducing factors such as transforming growth factor beta 1 (TGF-beta 1) or by any of the tested extracellular matrix (ECM) components (laminin and fibronectin) except for vitronectin (Vn). Vn, added together with TNF-alpha, counteracted the TNF-alpha blockage and allowed a massive migration of astrocytes (not due to cell proliferation) beyond that allowed by Vn only. Heparan sulfate proteoglycans (HSPG) were shown to be involved in the migration. The results may be relevant to regeneration of CNS axons, and may also provide an example that an extracellular component (Vn) can overcome and neutralize a negative effect of a growth factor/cytokine (TNF-alpha) and can act in synergy with other features of this cytokine to promote a necessary function (e.g., cell migration) that is otherwise inhibited.

It is now widely accepted that injured nerves, like any other injured tissue, need assistance from their extracellular milieu in order to heal. We compared the postinjury activities of thrombin and gelatinases, two types of proteolytic activities known to be critically involved in tissue healing, in nonregenerative (rat optic nerve) and regenerative (fish optic nerve and rat sciatic nerve) neural tissue. Unlike gelatinases, whose induction pattern was comparable in all three nerves, thrombin-like activity differed clearly between regenerating and nonregenerating nervous systems. Postinjury levels of this latter activity seem to dictate whether it will display beneficial or detrimental effects on the capacity of the tissue for repair. The results of this study further highlight the fact that tissue repair and nerve regeneration are closely linked and that substances that are not unique to the nervous system, but participate in wound healing in general, are also crucial for regeneration or its failure in the nervous system.