Differential manipulation of normal and chronic myeloid leukemia stem cell proliferation in vitro.

The long-term marrow culture (LTC) system allows the sustained production of primitive normal and neoplastic (chronic myeloid leukemia [CML]) hematopoietic cells in vitro for many weeks. This is achieved in the absence of exogenously added hematopoietic growth factors because of the presence in the cultures of supportive "stromal" cells of the fibroblast-endothelial-adipocyte lineages. These latter cells form a confluent adherent layer with which the most primitive hematopoietic cells become associated and which locally regulates their behavior. The LTC system has thus been considered as a model of the microenvironment of the bone marrow and used to delineate potentially physiologically relevant mechanisms that regulate the proliferation, self-renewal and differentiation of primitive normal hematopoietic cells. It has also been used to analyze the molecular basis of the altered proliferative behavior that characterizes primitive neoplastic cells from patients with CML. Most of the information obtained to date has emerged from experiments designed to shift the balance of stimulatory and inhibitory factors present in order to favor either the cycling or quiescence of primitive normal or CML cells in LTC. This has been achieved either by addition of soluble factors (or antagonists) to the LTC medium or by the use of genetically engineered factor-producing stromal cells. Such experiments have allowed the identification of a number of cytokines that promote one or the other of these responses (i.e., primitive progenitor cycling or quiescence), including some that are involved in control mechanisms endogenous to the LTC system. Recent studies suggest that the retention of primitive normal cells in a reversible G(o) state in this system is mediated by the cooperating action of limiting concentrations of at least two endogenously produced inhibitory factors (transforming growth factor-beta (TGF-beta) and macrophage inflammatory protein-1 alpha (MIP-1 alpha)), either of which, however, if added exogenously at a sufficient concentration, can exert this action on its own. Interestingly, the heightened turnover characteristic of primitive CML cells appears to be due to a selective unresponsiveness to only one of these two inhibitors (MIP-1 alpha). These findings are consistent with a complex model of the extrinsic regulation of primitive hematopoietic cells in which a multiplicity of intracellular signaling intermediates within the target cells converge at different points ultimately to control their entry into S phase. Our findings further suggest that only some of these pathways may be affected by intracellular expression of the BCR-ABL fusion gene.