GATA1s Exerts Fetal Stage-Specific Oncogenic Effects in Human Hematopoietic Stem and Progenitor Cells

Abstract 2355

Approximately 10% of all neonates with Down syndrome (DS) have a transient leukemia (TL), which is characterized by the accumulation of immature megakaryoblasts in the peripheral blood, liver, and bone marrow. In the majority of cases, TL is asymptomatic and resolves spontaneously. However in 20 and 30% of DS children, TL can progress to acute megakaryoblastic leukemia (AMKL) within the first 4 years of life. Acquired mutations in the hematopoietic transcription factor GATA1 are consistently present in both entities. In mice Gata1s overexpression leads to hyperproliferation of a transient fetal population of megakaryocytic progenitors (MPs), while ostensibly similar adult MPs are largely unaffected. Thus, growing evidence supports that TL and DS-AMKL originate in utero from fetal hematopoietic cells, suggesting that intrinsic genetic programs in fetal hematopoietic stem and progenitor cells (HSPCs), which may differ from those in adult HSPCs, are essential for leukemogenesis. However, if human fetal HSPCs or MPs are equally sensitive to the effects of GATA1s as their murine counterparts, remained elusive.

To address this issue, we first explored the intrinsic potential of adult (from mobilized peripheral blood; PB), neonatal (from cord blood; CB) and fetal (from livers of aborted fetuses, gestation weeks 13 to 18; FL) HSPCs to differentiate along the megakaryocytic lineage. In erythropoietin (EPO) and thrombopoietin (TPO) containing medium allowing erythroid and megakaryocytic differentiation, FL-HSPCs generated a higher percentage of CD41+ cells (29%) than PB- (16%), CB-HSPCs (24%). FL cells showed an 8-fold higher proliferation rate than PB cells and 1.4 higher proliferation rate than CB cells in stem cell expansion medium. The in vitro derived megakaryocytes from FL- and CB-HSPCs were much smaller in size than adult megakaryocytes from PB, consistent with previous reports.

Next we explored the effect of GATA1s overexpression on human PB-, CB- and FL-HSPCs using a lentiviral system. Cells were sorted and differentiated along megakaryocytic lineage in liquid and semi-solid media. In liquid culture, GATA1s transduced FL-MPs expanded 10-fold by day 10 (p=0.016), whereas MPs from CB and PB expanded 8.4-fold (p=0.013) and 3.1-fold (p=0.035), respectively. GATA1 overexpression caused only a moderate proliferation of FL-MPs (6.3-fold, p=0.041), CB-MPs (6.8-fold, p=0.049) and PB-MPs (2.2-fold, p=0.11). In semi-solid media (MegacultC), GATA1s induced a remarkable increase of megakaryocytic colonies (CFU-MKs) in FL (383 ± 20) compared to CB (116 ± 7) and PB (6.5 ± 3.5). GATA1s-transduced FL-MPs gave rise to 4.9 times more CFU-MKs than empty vector (p=0.002) and 1.77 times more than GATA1-transduced cells (p=0.018). In CB-MPs, the increase was 4.37 times (p=0.003) and 1.69 times (p=0.04), respectively. In PB the difference between GATA1s, GATA1 and empty vector control was insignificant (p=0.18, p=0.51). The CFU-MKs of FL- and CB-MPs were not only more abundant but also larger in size. We obtained 90% CD41+CD42b+ cells in FL GATA1s megakaryocytic differentiation culture by day 11, 88% in FL GATA1 and only 19% in FL empty vector control culture.

Thus, we demonstrated that FL-HSPCs have a higher intrinsic proliferative potential and are prone to differentiate along the megakaryocytic lineage. Furthermore, FL-MPs are more sensitive to GATA1s overexpression, leading to their hyperproliferation in liquid media and abundant formation of large CFU-MKs in semi-solid medium. However, without trisomy 21 in human cells we do not see oncogenic transformation and arisal of TL-like blasts, underscoring the requirement of the triad of factors (GATA1s mutation, fetal origin of the cells and trisomy 21) for DS-AMKL and DS-TL.