The phenotypic effects of selectively decreasing the levels of S100 beta in cultured glial cells were analyzed. Two separate antisense approaches were utilized for inhibition of S100 beta production: analysis of clonal isolates of rat C6 glioma cells containing an S100 beta antisense gene under the control of a dexamethasone-inducible promoter, and analysis of C6 cells treated with S100 beta antisense oligodeoxynucleotides. Both antisense methods resulted in a decrease in S100 beta levels in the cell, as measured by RIA. The inhibition of S100 beta production correlated with three alterations in cellular phenotype: (a) a flattened cell morphology; (b) a more organized microfilament network; and (c) a decrease in cell growth rate. The studies describe here provide direct evidence for an involvement of S100 beta in glial cell structure and function, and suggest potential in vivo roles for S100 beta in regulation of glial cell morphology, cytoskeletal organization, and cell proliferation.

S100 beta produced in Escherichia coli from a synthetic gene (Van Eldik, L. J., J. L. Staecker, and F. Winningham-Major. 1988. J. Biol. Chem. 263:7830-7837) stimulates neurite outgrowth and enhances cell maintenance in cultures of embryonic chick cerebral cortex neurons. In control experiments, the neurite extension activity is reduced by preincubation with antibodies made against bovine brain S100 beta. When either of the two cysteines in S100 beta are altered by site-directed mutagenesis, the resultant proteins maintain the overall biochemical properties of S100 beta, but lose both the neurite extension and neuronal survival activities. However, another S100 beta mutant, in which the relative position of one of the two cysteines was changed, had neurotrophic activity similar to that of the unmodified protein. These and other results indicate that (a) specific neurite extension activity and neuronal survival activity are two related activities inherent to the S100 beta molecule; (b) a disulfide-linked form of S100 beta is required for full biological activity, and (c) the relative position of the cysteines can be modified. These data suggest potential in vivo roles for S100 beta in the development and maintenance of neuronal function in the central nervous system, and demonstrate the feasibility of the longer term development of selective pharmacological agents based on the S100 beta structure.