Foxo Transcription Factor Activity Is Retained in Quiescent Chronic Myeloid Leukaemia Stem Cells and Activated by Tyrosine Kinase Inhibitors to Mediate “induced-quiescence” in More Mature progenitors.

Quiescent CML stem cells were isolated phenotypically (Lin⁻CD34⁺38⁻) by FACS from the total CD34⁺ stem/progenitor cell population and D-FISH analysis performed to determine the percentage of Ph⁺ cells. The expression levels of Spred1, FOXO1, FOXO3a, FOXO4 and Cyclin D1 were evaluated by Real-Time-PCR. Protein levels and localization were studied by Western blot, immunofluorescence and FACS. EMSA assay was used to evaluate FOXO3a DNA binding activity. K562 cells were transfected with wt FOXO3a and the constitutively active triple mutant (TM) form of FOXO. After transfection, proliferation and apoptosis were tested using incorporation of H³ thymidine and annexin V detection by FACS, respectively.

We found that BCR-ABL, through activation of the PI3K/Akt pathway, induces phosphorylation and cytoplasmic localization of FOXO in CD34⁺ CML cells, thereby blocking its transcriptional activity as demonstrated by EMSA and by the inhibition of FOXO target genes, including Spred1. Incubation with TKIs decreased phosphorylation and induced re-localization of FOXO to the nucleus in CD34⁺ CML cells, thus restoring FOXO transcriptional activity. This resulted in decreased levels of Cyclin D1 and reduced ROS. Similar effects and a dramatic reduction of cell proliferation, accompanied by significant apoptosis, were observed by forcing the expression of FOXO3a in K562 through transfection of the TM form, confirming the role of FOXO in inducing cell cycle arrest. Incubation experiments using LY294002, rapamycin and TKIs established that the reactivation of FOXO by TKI is mediated by the PI3K/Akt pathway. Interestingly, we found that phosphorylation of FOXO1, 3a and 4 was higher (i.e. cytoplasmic and inactive) in proliferating CD34⁺38⁺ CML cells, as compared to more primitive and quiescent CD34⁺38⁻90⁺ CML cells, although the activity of BCR-ABL, measured by analysis of p-CrKL was found to be similar in both populations, indicating that although present, BCR-ABL was not inducing phosphorylation of FOXO in the quiescent population. In this sense the quiescent CML stem cells resembled normal HSC in terms of FOXO regulation. In conclusion our data indicate that TKIs initiate a process in CML stem and progenitor cells that maintains their quiescence and therefore potential resistance to TKIs themselves. The anti-proliferative activity of TKIs against primary CML CD34⁺ cells is mediated, at least in part, by the re-activation of FOXO1, 3a and 4. BCR-ABL appears to play a different role in more mature progenitor cells compared to primitive quiescent stem cells, suggesting the possibility of an incomplete activity of BCR-ABL at the stem cell level or alternatively, the possibility that FOXO activity at this level provides the dominant signal responsible for intrinsic quiescence.

No relevant conflicts of interest to declare.