Identification of a population of bipotent stem cells in the HL60 human promyelocytic leukemia cell line.

HL60, the human promyelocytic leukemia cell line, consists of cells that resemble promyelocytes but are transformed into more mature myeloid forms (polymorphonuclear leukocytes) that express most of the functional characteristics of terminally differentiated myeloid forms when incubated in the presence of dimethyl formamide (HCONMe(2)). When HL60 cells are exposed to the phorbol diester 12-O-tetradecanoylphorbol 13-acetate [phorbol 12-myristate 13-acetate (PMA)], they shift from suspension to adherent on the surface of the tissue culture vessel, assume the appearance of macrophages, and acquire macrophage-associated surface markers, the enzyme-nonspecific esterase, the isozyme acid phosphatase, and the myeloid-specific esterase. Although data from other laboratories suggest that HL60 consists of stem cells that are bipotent with respect to myeloid or macrophage differentiation, unequivocal proof of bipotency or evidence that ruled out the presence of two different types of stem cells, each committed to different lines of hematopoietic differentiation, has been lacking. To resolve these two alternatives, we have developed a cloned population of HL60 cells and studied the properties of these cells when exposed to inducers of myeloid (HCONMe(2)) or macrophage (PMA) differentiation. After 120 hr of incubation with HCONMe(2), 95% of the cells acquire myeloid markers and lack specific macrophage markers, whereas the reverse is true in the presence of PMA, clearly establishing that HL60 is able to commit itself to the development of two different programs of hematopoietic differentiation. Moreover, the commitment to macrophage differentiation is irreversible after 6 hr of incubation in PMA whereas development of the myeloid program requires continuous exposure of the cells to HCONMe(2). Commitment of HL60 to macrophage differentiation by PMA is not affected by subsequent addition of HCONMe(2), whereas HCONMe(2)-induced myeloid differentiation is overridden by exposure of the cells to PMA. These data suggest that the HL60 cell may provide a model system for studying the process that generates several classes of precursor cells, each committed to different lines of hematopoietic differentiation, as well as the process that mediates the maturation of these committed precursor cells to their terminally differentiated state.