One means of treating the injured brain may be via regulation of expression of specific neurotrophic factors, in particular those synthesized by astrocytes. Our studies focus on (1) identification of novel neurotrophic factors; and (2) the analysis of the regulation of neurotrophic factor and neuropeptide gene expression during development and in response to injury. One novel neurotrophic factor under investigation is pigment epithelium-derived factor (PEDF). PEDF not only functions as a survival factor for cerebellar granule cell neurons but also can protect them against both glutamate toxicity and apoptotic cell death. A lentiviral construct carrying the PEDF cDNA provides sufficient PEDF to infected cerebellar granule cells for both survival and anti-glutamate protection. PEDF's ability to block glutamate toxicity involves modulation of the changes in intracellular calcium caused by glutamate. PEDF also activates microglia, which produce an as yet unidentified factor which inhibits astrocyte proliferation and may thus be useful when brain injury results in gliosis due to astrocyte division. The factor also inhibits proliferation of gliomas. DNA microarray analysis has shown that PEDF stimulates synthesis of several other neurotrophic factors by cerebellar granule cells, as well as stimulating microglia to produce a variety of cytokines. Since astrocytes can synthesize a number of neurotrophic factors, primary cultures of astrocytes are being used to determine factors which regulate production of trophic factors such as nerve growth factor (NGF) and glial-derived neurotrophic factor (GDNF) in response to 6-OHDA lesion of rat substantia nigra, a Parkinsonian-like model. Reactive astrocytes are prepared from 6-OHDA-lesioned brain: monoclonal antibodies raised against epitopes expressed only by reactive astrocytes in vivo distinguish between normal adult and reactive astrocytes in culture. 6-OHDA lesions of the substantia nigra induce reactive astrocytes at the site of lesion as well as in the terminal fields in striatum and cortex. Whereas all these reactive astrocytes express significantly more glial fibrillary acidic protein, S-100-beta, and vimentin, a major difference is found in terms of expression of the adhesion molecule PSA-NCAM. PSA-NCAM expression is turned on in astrocytes in the lesioned SN, but not the contralateral side, nor even in the terminal fields: thus, PSA-NCAM expression can distinguish subtypes of reactive astrocytes. Reactive astrocytes cultured from SN express PSA-NCAM whereas those from striatum do not. Thus, although comparable differences are seen between what occurs in vivo and in the cultures of reactive astrocytes, the changes do not all parallel those of the classical markers of reactive gliosis. Several co-culture models are being tested, using nigral dopamine neurons as well as others, to determine whether reactive astrocytes are differentially supportive of process outgrowth and whether the differences are due to expression of PSA-NCAM.
