Myelosuppression is one the most frequent side effects of chemotherapy. New agents that more selectively target cancer cells have been developed in attempt to improve the effects and to decrease the side effects of cancer treatment. Roscovitine is a purine analogue and cyclin-dependent kinase inhibitor. Several studies have shown its cytotoxic effect in cancer cell lines in vitro and in xenograft models in vivo. In this study, we investigated the effect of roscovitine on hematopoietic progenitors in vitro and in vivo in mice. The clonogenic capacity of hematopoietic progenitors was studied using burst-forming unit-erythroid (BFU-E), colony-forming unit granulocyte, macrophage (CFU-GM) and colony-forming unit granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM). In vitro, bone marrow cells were exposed to roscovitine (25-250 microM) in Iscove's modified Dulbecco's media for 4 h or to roscovitine (1-100 microM) in MethoCult media for 12 days. No effect on colony formation was observed after exposure to roscovitine for 4 h; however, concentration- and cell type-dependent effects were observed after 12 days. Roscovitine in concentration of 100 microM inhibited the growth of all types of colonies, while lower concentrations have shown differential effect on hematopoietic progenitors. The most sensitive were CFU-GEMM, followed by BFU-E and then CFU-GM. In vivo, mice were treated with single dose of roscovitine (50, 100 or 250 mg/kg) and the effect on bone marrow was studied on day 1, 3, 6, 9 or 12 after the treatment. In the second part of experiment, the mice were treated with roscovitine 350 mg/kg/day divided into two daily doses for 4 days. The bone marrow was examined on day 1 and 5 after the last dose of roscovitine. On day 1, BFU-E decreased to less than 50% of the controls (P = 0.019). No decrease in BFU-E formation was observed on day 5. No significant effect was observed on CFU-GM and CFU-GEMM growth after the treatment with multiple doses of roscovitine. Single doses of roscovitine or dimethylsulfoxide did not affect the colony formation. We also studied the distribution of roscovitine to the bone marrow after a dose of 50 mg/kg was administered intraperitoneally. Only 1.5% of the drug was detected in the bone marrow. Thus, the roscovitine effect on hematopoietic progenitors in bone marrow in vivo is only transient. One reason may be that only a small fraction of roscovitine reaches the bone marrow. Another explanation may be the short half-life observed for roscovitine that might not allow enough cell exposure to the drug. However, the toxicity of roscovitine to hematopoietic progenitors in vitro is within the same exposure range as cytotoxicity to cancer cells. Thus, precaution should be taken in clinical trials, especially when combinations with myelosuppressive cytostatics are used.