Western blotting, indirect immunolocalization, flow cytometry, and a functional assay for drug-induced strand breakage were utilized to examine topoisomerase (topo) II levels during granulocytic maturation in HL-60 human progranulocytic leukemia cells and in samples of normal human marrow. Indirect immunofluorescence revealed that the intensity of the signal for topo II in unsynchronized log phase HL-60 cells varied widely. Indirect immunolabeling combined with propidium iodide staining and two-parameter flow cytometry revealed that topo II levels increased an average of 2-fold as cells progressed from G1 to G2/M. When HL-60 cells were induced to mature toward granulocytes, topo II levels progressively decreased and became undetectable by functional assays, by indirect immunoperoxidase staining, and by Western blotting with an antibody which identified Mr 170,000 and Mr 180,000 forms of topo II. Similar changes were detected during normal granulocytic maturation in human marrow in vivo. Western blotting revealed that levels of the Mr 170,000 (proliferation-associated) isoform of topo II were highest in marrow fractions enriched in progranulocytes and myelocytes, intermediate in unfractionated marrow from normal volunteers, and undetectable in mature granulocytes. The Mr 180,000 topo II polypeptide was also diminished or absent from mature granulocytes. In further experiments, marrow samples from normal volunteers were subjected to flow cytometry after labeling of topo II and various cell surface markers. Levels of the Mr 170,000 topo II polypeptide in CD34-positive cells (multipotent and committed progenitors from several hematopoietic lineages) were indistinguishable from levels observed in the HL-60 leukemia cell line. These results suggest that topo II levels in highly proliferative normal human myeloid cells in vivo approach levels found in corresponding neoplastic cell lines in vitro. Conversely, as the same cells mature into granulocytes in vivo or in vitro, levels of both molecular weight forms of topo II diminish. These results provide a framework for the further investigation of topo II levels and drug sensitivity in human leukemia.