Adhering platelets on the cell surface can give misleading results when doing flow cytometry analysis of platelet/megakaryocyte-specific glycoprotein (GP) antigens to enumerate megakaryocytes (MK) in mobilized peripheral blood (PB), apheresis products, or normal bone marrow (BM). For adequate quantification and characterization of human MK, we examined samples with parallel flow cytometry and immunocytochemistry. MK expression of GP IIb/IIIa (CD41a), GP Ib (CD42b), GP IIIa (CD61), CD45, CD33, and CD11b, and their light scatter properties were evaluated. Fresh samples of low density mononuclear cells (MNC) or purified CD34+ cells contained 10-45% of platelet-coated cells. Platelet-coated cells decreased dramatically after several days of incubation in a serum-free medium supplemented with stem cell factor, IL-3, IL-6, and/or GM-CSF. Between d 9-12, flow cytometry detected a distinct CD41a+ MK population, 8.3 +/- 1.3% in BM CD34 cell cultures (n = 7) and 13.1 +/- 2.1% in PB CD34 cell cultures (n = 14), comparable to immunocytochemistry data (7.8 +/- 1.9% and 16.4 +/- 2.6%, respectively). CD41a stained a higher proportion of MK than CD42b or CD61, while CD42b+ or CD61+ cells contained more morphologically mature MK than CD41a+ cells in cultures containing aplastic serum. When fluorescence emission of CD41a was plotted against forward-light scatter (FSC), subpopulations of small and large MK were observed. Such subpopulations overlapped in CD41a intensity and side-light scatter (SSC) property. Most MK co-expressed CD45 (98.8% positive) but not CD33 (80.7% negative) or CD11b (88.9% negative). Our data indicate that flow cytometry can be used effectively to identify MK. However, caution should be taken with samples containing adherent platelets.