To date, the electrophysiological properties of glial cells located in reactive scar tissue are unknown. To address this issue two subtypes of hippocampal glial cells, located in thin vital slices of normal or gliotic brain tissue, were analysed for their voltage controlled ion channels using the patch-clamp technique. Reactive gliosis was induced in adult rats by a single peritoneal injection of kainic acid. The intensity of the following seizures was rated ascending from 1 to 6. Rats which exhibited seizures of level 3 or higher showed, within three days, a marked loss of pyramidal cells (60% in CA1 and CA3) and an increase in the density of glial fibrillary acidic protein immunostaining, representing an apparent increase in the number and size of astrocytes in all layers of the hippocampal CA1 subfield. Reactive and normal astrocytes of one subtype, electrophysiologically characterized by time-independent potassium currents, did not significantly differ in membrane potential and potassium conductivity. Glutamine synthetase-positive, but mostly glial fibrillary acidic protein-negative, glial cells (presumably representing immature astrocytes) were also included in this study. This subtype of glial cells showed several voltage- and time-dependent potassium currents and, under control conditions, tetrodotoxin-sensitive voltage-gated Na+ channels, which were almost completely lost after reactive gliosis. Another part of this study focuses on the sensitivity of reactive and control glial cells for extracellular ATP. Several in vitro studies suggest that P2 purinergic receptors on glial cells could trigger the induction of reactive gliosis. In contrast to results described on cultured astrocytes, we found in situ that hippocampal glial cells were not sensitive to ATP or stable P2 receptor agonists in control or in gliotic brain slices. In summary, the presence of at least two different subtypes of hippocampal astrocytes was demonstrated for control as well as for gliotic brain tissue. A dramatic down-regulation of tetrodotoxin-sensitive sodium channels in one subpopulation of reactive astrocytes was shown. This result supports the hypothesis that the presence of active neurons could be required to maintain glial voltage-gated sodium channels. Furthermore, we conclude that there is no longtime expression of P2 purinoceptors on hippocampal astrocytes in situ, and therefore the involvement of astrocytic ATP receptors in the genesis of reactive gliosis is unlikely.