Transforming growth factor beta3 inhibits chronic myelogenous leukemia hematopoiesis by inducing Fas-independent apoptosis.

OBJECTIVE: Transforming growth factor beta3 (TGF-beta3) is a potent suppressor of human hematopoietic progenitor cells. In this article, we compare the activity of TGF-beta3 on highly purified CD34+ cells and more immature CD34-DR(-) cells from chronic myelogenous leukemia (CML) patients in chronic phase and normal donors. MATERIALS AND METHODS: Primitive hematopoietic progenitors were stimulated in liquid cultures and clonogenic assays by early-acting growth factors such as stem cell factor (SCF) and interleukin 11 (IL-11) and the intermediate-late-acting stimulating factors IL-3, granulocyte-macrophage colony-stimulating factor, and erythropoietin. Molecular analysis of bcr/abl mRNA was performed on single CML colonies by nested reverse transcriptase polymerase chain reaction. Moreover, cell cycle analysis and assessment of apoptosis of normal and leukemic CD34+ cells were performed by propidium iodide (PI) alone and simultaneous staining with annexin V and PI, respectively. RESULTS: The colony-forming efficiency of CML CD34+ cells was generally inhibited by more than 90% regardless of whether the colony-stimulating factors were used alone or combined. When compared to normal CD34+ cells, leukemic cells were significantly more suppressed in 6 of 8 culture conditions. The inhibitory effect of TGF-beta3 on CD34+ cells was exerted within the first 24 hours of incubation as demonstrated by short-term preincubation followed by IL-3-and SCF-stimulated colony assays. Evaluation of bcr/abl transcript on residual CML colonies incubated with TGF-beta3 demonstrated a small subset of neoplastic CD34+ cells unresponsive to the inhibitory effect of the study cytokine. TGF-beta3 demonstrated a greater inhibitory activity on primitive CD34+DR cells than on more mature CD34+ cells. Again, CML CD34+DR(-) cells were significantly more inhibited by TGF-beta3 than their normal counterparts in 3 of 8 culture conditions. Kinetic analysis performed on CD34+ cells showed that TGF-beta induces cell cycle arrest in G(1) phase. However, this mechanism of action is shared by normal and leukemic cells. Conversely, TGF-beta3 preferentially triggered the programmed cell death of CML CD34-cells without increasing the proportion of leukemic cells coexpressing CD95 (Fas receptor), and this effect was not reversed by functional blockade of Fas receptor. Conclusion. We demonstrate that TGF-beta3 exerts a potent suppressive effect on CML cells that is partly mediated by Fas-independent apoptosis.

Arabinosyl cytosine in chronic myeloid leukaemia: evidence for high cytokinetic sensitivity of myeloblasts.

The effect of a single and of repeated i.v. push dose(s) of Arabinosyl Cytosine (ARA-C) has been investigated in 9 chronic myeloid leukaemia (CML) patients in non-blastic phase. This was done by determining separately the relative compartment size, the mitotic index (IM), and the in vitro 3H-TdR labelling index (IL) of marrow and blood myeloblasts (MB) and promyelocytes plus myelocytes (PMC + MC), before and at intervals after the drug. After a single dose of ARA-C, the IL of marrow MB declines rapidly, and recovers thereafter, often with an overshoot at 15 h. After 2 to 4 doses of ARA-C, the IL of marrow and blood MB rises by a factor of 2 to 3, and is maintained at a plateau during further treatment. The behaviour of the IL of blood MB is not always the same as that of marrow MB. The IM of marrow MB does not rise proportionally to the IL, and sometimes is even found to be decreased. It is suggested that these kinetic perturbations reflect an accumulation of MB in S-phase where many but not all of them are trapped and sooner or later die off. With a few exceptions, ARA-C induces only milder kinetic perturbations in marrow and blood PMC+MC. The overall results of this study are in agreement with the generally accepted mechanism of action of ARA-C (S-phase specific effector agent), and with studies that indicate that the effect of ARA-C depends on the growth pattern and on the degree of maturation of the target cells. It is suggested that a proper evaluation of ARA-C on a cell population should take into account the existence of different cell pools, provided with different proliferative activity and potential, and with variable degrees of maturation.