Hematopoietic stem cells (HSCs) in adults are largely quiescent, periodically entering and exiting cell cycle to replenish the progenitor pool or to self-renew, without exhausting their number. Expression profiling of quiescent HSCs in our and other laboratories suggests that high expression of the zinc finger transcription factor GATA-2 correlates with quiescence. We show here that TGFβ1-induced quiescence of wild-type human cord blood CD34+ cells in vitro correlated with induction of endogenous GATA-2 expression. To directly test if GATA-2 has a causative role in HSC quiescence we constitutively expressed GATA-2 in human cord blood stem and progenitor cells using lentiviral vectors, and assessed the functional output from these cells. In both CD34+ and CD34+ CD38− populations, enforced GATA-2 expression conferred increased quiescence as assessed by Hoechst/Pyronin Y staining. CD34+ cells with enforced GATA-2 expression showed reductions in both colony number and size when assessed in multipotential CFC assays. In CFC assays conducted with more primitive CD34+ CD38− cells, colony number and size were also reduced, with myeloid and mixed colony number more reduced than erythroid colonies. Reduced CFC activity was not due to increased apoptosis, as judged by Annexin V staining of GATA-2-transduced CD34+ or CD34+ CD38− cells. To the contrary, in vitro cultures from GATA-2-transduced CD34+ CD38− cells showed increased protection from apoptosis. In vitro, proliferation of CD34+ CD38− cells was severely impaired by constitutive expression of GATA-2. Real-time PCR analysis showed no upregulation of classic cell cycle inhibitors such as p21, p57 or p16INK4A. However GATA-2 expression did cause repression of cyclin D3, EGR2, E2F4, ANGPT1 and C/EBPα. In stem cell assays, CD34+ CD38− cells constitutively expressing GATA-2 showed little or no LTC-IC activity. In xenografted NOD/SCID mice, transduced CD34+ CD38−cells expressing high levels of GATA-2 did not contribute to hematopoiesis, although cells expressing lower levels of GATA-2 did. This threshold effect is presumably due to DNA binding by GATA-2, as a zinc-finger deletion variant of GATA-2 shows contribution to hematopoiesis from cells irrespective of expression level. These NOD/SCID data suggest that levels of GATA-2 may play a part in the in vivo control of stem and progenitor cell proliferation. Taken together, our data demonstrate that GATA-2 enforces a transcriptional program on stem and progenitor cells which suppresses their responses to proliferative stimuli with the result that they remain quiescent in vitro and in vivo.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author