A new method is described which combines the identification of DNA replicating and apoptotic cells in a single measurement by flow cytometry. The detection of DNA replicating cells is based on incorporation of 5-bromo-2'-deoxyuridine or 5-iodo-2'-deoxyuridine, followed by selective photolysis at the site of incorporation of the halogenated DNA precursors. Single-strand breaks in DNA resulting from the photolysis are subsequently labeled with digoxygenin or biotin-conjugated dUTP in a reaction catalyzed by exogenous terminal deoxynucleotidyl transferase. The double-stranded DNA breaks in apoptotic cells resulting from activation of the endonuclease can be labeled in this reaction as well. However, in contrast to the photolysed DNA, the low molecular weight fraction of DNA of apoptotic cells is extractable from the cells, and the degree of DNA elution can be modulated by cross-linking with formaldehyde. Thus, apoptotic cells can be distinguished and quantified by virtue of their fractional DNA content. Replication of less than 1% of a genome of a cell in the presence of 5-bromo-2'-deoxyuridine (equivalent of a 5-min 10 microM 5-bromo-2'-deoxyuridine pulse) can be detected by the selective photolysis method. The method was applied to study apoptosis and proliferation of human leukemic HL-60 cells and normal, mitogen-stimulated lymphocytes. Whereas apoptosis of HL-60 cells induced by the DNA topoisomerase I inhibitor camptothecin was selective to DNA replicating cells, apoptosis induced by hyperthermia showed no such selectivity. Lymphocytes that preferentially underwent apoptosis in cultures stimulated by phytohemagglutinin did not initiate DNA replication. By offering the possibility for identification of both DNA replicating and apoptotic cells in a single measurement, the method may find an application in studies of the prognostic value of both cell proliferation and death in human tumors and the apoptotic response of DNA replicating vs. nonreplicating cells to different treatments.