The Tyrosine Kinase Inhibitor TK258 Differentially Targets Proliferating Human Hematopoietic Progenitor Cells but Does Not Eliminate the Quiescent Fraction.

NOD-SCID mouse models have shown that, as with normal hematopoietic stem cells, leukemic stem cells (LSC) exist in a quiescent state and are thus capable of sustaining disease in vivo. Targeted therapy of LSC may therefore have a significant impact on disease eradication, relapse rates and toxicity reduction. Inhibition of FLT3 and c-kit is a current therapeutic strategy for AML but the efficacy of this approach on quiescent cells is unknown. Furthermore, although both FLT3 and c-kit play crucial roles in the development of normal human hematopoietic progenitor (HPC) and stem cells, the hematopoietic consequences of FLT3 and c-kit inhibition remain undefined. Here we show that the FLT3, c-kit and PDGF directed tyrosine kinase inhibitor TK258 specifically targets slowly dividing human HPCs, but not rapidly proliferating HPCs or quiescent cells displaying an immature phenotype. For normal CD34+ cells, treatment of short-term cultures with TK258 resulted in a dose dependent inhibition of cell expansion. Moreover, treatment of normal CD34+ cells with TK258 within the clinical dose range [0.2μM to 1.5μM] inhibited myeloid colony growth but had no effect on erythroid colony output, suggesting a differential effect on lineage-committed progenitors. To further elucidate the effect of TK258 on dividing and also quiescent CD34+ cells, normal CD34+ cells were labelled with a non-toxic fluorescent cell membrane dye, PKH26, prior to TK258 treatment in vitro. Coupled with flow cytometry, this dye enabled the discrimination of progenitor cells with different proliferative capacities. Thus, using this technique, we defined a CD34+/CD133+ quiescent population of cells that retained the same PKH fluorescence intensity as from day 0 of culture. TK258 treatment in culture generated a significant increase in the CD34+/CD133+ quiescent fraction (15.8% treated versus 5.9% untreated of total viable cells). In contrast with untreated cells this fraction maintained c-kit expression, suggesting maintenance of an immature population by TK258. Total viable cell recoveries were higher in the TK258 treated quiescent fraction (24% treated versus 13.5% untreated, p=0.037). Rapidly dividing progenitors were unaffected by TK258, while progenitors with a slower division rate showed significant sensitivity to the compound. The anti-proliferative effect of TK258 was mediated via G0/G1 cell cycle arrest, but not by apoptosis of CD34+ cells. We assessed the functional read-out of TK258 pre-treated progenitors in CFU assays, by sorting cells on the basis of their PKH26 fluorescence. These assays demonstrated a higher total colony output and diverse colony type for TK258 pre-treated ‘quiescent’ cells compared to untreated controls, in keeping with the preservation of a more immature population by TK258. Rapidly proliferating populations displayed a lower CFU potential, attributable to a reduction in myeloid output, which was not significantly altered by TK258 pre-treatment. CAFC assays and gene expression profiling of PKH26-sorted subsets are in progress. Taken together, these data demonstrate that TK258 has significant inhibitory effects on normal human hematopoietic progenitor cells in vitro and that the sensitivity of progenitor cells to the compound can be predicted from their proliferative index. Importantly, immature hematopoietic quiescent cells are preserved, which may have implications for the activity of tyrosine kinase inhibitors on non-cycling human leukemia cells.