S100B (member of a family of proteins that are 100% soluble in ammonium sulfate at neutral pH) has been widely used as astrocyte marker in animal models and in human brain diseases. Recent studies revealed S100B-immunopositivity in oligodendrocytes and O2A oligodendroglial progenitor cells. It is unknown, however, if oligodendrocytes produce S100B themselves, or if the S100B-immunolabeling is caused by binding or absorption of the protein. To address this question, S100B expression and protein release were analyzed in a highly pure oligodendrocytic OLN-93 cell line (from rat), in the astrocytic C6 cell line (from rat) and primary astrocytes. S100B was gene expressed in all cultures, as revealed by reverse transcriptase polymerase chain reaction (RT-PCR) analysis. OLN-93 cells and glial fibrillary acidic protein (GFAP)-negative astrocytes expressed the multiligand receptor for advanced glycation end products (RAGE). S100B protein levels were determined in supernatants and cell homogenates by immunoluminometry under normal conditions and after serum and glucose deprivation (SGD). SGD led to a several-fold increased release of S100B (after 6 and 24 h), which was particularly pronounced in primary astrocytes. Increased S100B in cell homogenates was most notable in OLN-93 cells under SGD, indicating activated S100B synthesis. These cells also showed the highest percentage of dead cells, as determined by propidium iodide-positivity, after SGD. Incubation with 0.5, 2 and 5 microg/l exogenous S100B was not toxic to OLN-93 cells. In conclusion, OLN-93 cells produce more S100B under SGD than astrocytes and are more susceptible to cell death upon SGD, which provokes leakage of S100B. Our data indicate active S100B secretion from astrocytes under SGD since highly elevated levels of S100B were detected in the supernatant despite a low percentage of dead cells. The experimental results provide further evidence for a production/release of S100B in/from oligodendrocytes, e.g. in metabolic stress conditions like cerebral ischemia. Studies on S100B in bodily fluids should be carefully interpreted in order to avoid misleading hypotheses concerning the specific involvement of astrocytes, due to the various cellular sources of S100B.