Current studies of the central nervous system (CNS) are assigning an increasing number of activities to astrocytes, many of which are potentially relevant to stroke. However, nearly all of these suggested functions are based on observed correlations, and many of these were made on cultured cells, whose properties may differ from those in vivo. As an alternative approach to understanding astrocyte function, we are studying transcriptional regulation of the human gene encoding glial fibrillary acidic protein (GFAP), the major component of astrocyte intermediate filaments. By studying GFAP transcription, insights may be gained into mechanisms governing development, reaction to injury, and cell specificity. A second goal is to use identified astrocyte-specific transcription elements to direct expression of other genes in astrocytes. This enables testing of the roles of specific factors in CNS function, and may produce disease models. Transcriptional studies have focused primarily on identifying factors that act at a consensus AP-1 site that is essential for GFAP transcription. Since proteins encoded by the jun and fos proto-oncogene families are known to modulate transcription via AP-1 sites, their presence in a GFAP expressing astrocytic cell line was examined. Analyses included gel mobility shift assays, """"""""shift Westerns"""""""" and detection of the specific mRNAs by Northern analysis. Preliminary results show a correlation between GFAP transcription and the presence of c-Jun, JunD, and Fra-2, but not with JunB, c-Fos, FosB or Fra-1. Working with a human astrocytoma cell line, we have found a strong correlation with GFAP gene activity and the presence of JunD and Fra2. This result raises the possibility that the cell specificity of GFAP expression is due in part to a requirement for this particular combination of AP-1 components. However, cotransfection of GFAP reporter genes along with various Jun and Fos expression vectors has failed to demonstrate such a requirement. Applications of our GFAP gene analyses to studies of brain function have been carried out in collaborations with other laboratories. These projects include effects on stroke and development of overproduction of the inflammation-related cytokines TGF- beta 1 and TNF-alpha; investigation of the role of astrocytes in brain function by analyzing the effects of their timed ablation mediated by the herpes simplex virus thymidine kinase or E. coli cytosine deaminase transgenes; production of mouse glioma models through expression of oncogenes; gene therapy for Parkinson's disease via expression of tyrosine hydroxylase; a feasibility study for constructing a regulatable transgene; and studies of the functional role of GFAP itself through its overexpression and gene knock-out.
