Both preclinical and clinical evidence suggested that antidepressant drugs upregulate hippocampal cell proliferation and neurogenesis. In addition, direct evidence was recently published that hippocampal de novo cell proliferation is necessary for antidepressant action. Within this frame, we used primary cultures of rat cerebellar granule cells (CGC) as an in vitro model of central nervous system (CNS) to investigate whether a neurogenic response could be elicited also in the cerebellum, upon chronic treatment with selective serotonin reuptake inhibitors (SSRIs). Furthermore, we assayed the presence of neural precursor cells in CGC, possibly responsive to proliferation and differentiation stimuli. We found that 1 microM fluoxetine increased cell proliferation, as assayed by [3H]-thymidine incorporation. CGC immunocytochemical analysis with neural cell-specific markers revealed the presence of granule neurons, glial cells, and a cell component that we named "round cells." Because only round cells displayed proliferation ability, as revealed by 5-bromo-2'-deoxyuridine (BrdU) labeling, they were further characterized. For this purpose, round cells were isolated and expanded by culturing in a serum-free medium, containing basic fibroblast growth factor (bFGF), before immunocytochemical analysis. We found that round cells were not immunoreactive for glial, neuronal, and oligodendrocyte markers, whereas they were immunoreactive for several immature neuronal markers. Accordingly, round cells could be induced to differentiate into astrocytes, neurons, and oligodendrocytes, either by withdrawing the mitogen bFGF or by exposing them to fluoxetine. These findings suggest that round cells in CGC possess the features and potentials of neural precursors, able to differentiate in mature neural cells upon a pharmacological simulum.