Identification of Human Hematopoietic Stem Cell-Specific STAT5 Target Genes Involved in Self-Renewal and Transformation.

Abstract 568

The transcription factor STAT5 fulfils an essential role in self-renewal of both mouse and human HSCs and persistent activation of STAT5 contributes to leukemic transformation. In patients with acute myeloid leukemia, increased STAT5 activity has been observed in over 60% of the cases. Yet, little is known about mechanisms that are involved. To gain further insight into these processes we studied whether STAT5-imposed long-term self-renewal is exclusively restricted to HSCs, or whether long-term self-renewal can also be imposed on progenitor cells. Human cord blood (CB) cells were transduced with control and STAT5-ER retroviral vectors allowing the induction of STAT5 activity by treatment of cells with 4-hydroxytamoxifen (4-OHT). Four populations were isolated: hematopoietic stem cells (HSC, CD34⁺CD38^low), common myeloid progenitors (CMP, CD34⁺CD38⁺CD123⁺CD45RA^-), granulocyte-macrophage progenitors (GMP, CD34⁺CD38⁺CD123⁺CD45RA⁺) and megakaryocyte-erythroid progenitors (MEP, CD34⁺CD38⁺CD123^-CD45RA^-). MS5 bone marrow stromal cocultures were initiated and STAT5 activity was induced by 4-OHT. In HSCs, STAT5 overexpression induced a long-term proliferative advantage as well as a significant increase in cobblestone formation. This coincided with elevated levels of Colony Forming Cells (CFCs) that were maintained over 5 weeks. In contrast, STAT5 was unable to induce cobblestone formation in progenitor cocultures and only a transient STAT5-induced increase in cell numbers was observed. CFC numbers dropped significantly after 2 weeks and progenitor initiated cultures could not be maintained longer than 3 weeks regardless of STAT5 activity. Myelopoiesis was blocked and an increase in erythroid differentiation in STAT5-ER-transduced HSC, CMP, and MEP populations was observed, while the differentiation potential of the GMP remained unaffected.

Next, we aimed to identify HSC-specific STAT5 target genes by performing microarray analysis on HSC, CMP, GMP and MEP populations transduced with our STAT5-ER vectors. To limit STAT5 mediated effects on erythropoiesis GATA1 was downmodulated in STAT5-transduced CB cells by a lentiviral RNAi approach, which completely abrogated erythropoiesis but maintained enhanced HSC self-renewal. Microarrays were performed on GATA1 downmodulated STAT5-transduced CB cells and controls, and these data sets were compared to the HSC-specific STAT5 target gene lists. This combined approach resulted in the identification of 36 GATA1-independent STAT5 target genes in the HSC population. One of the identified genes was HIF2a. The involvement of HIF2a in STAT5 phenotypes was studied functionally by using a lentiviral HIF2a RNAi approach in STAT5 transduced CB cells. These studies revealed that expansion of STAT5/HIF2a RNAi-transduced cells on MS5 bone marrow stromal cocultures was reduced, coinciding with reduced CFC and LTC-IC frequencies, while differentiation was not affected.

In summary, our data show that hematopoietic stem cells, but not progenitors are the exclusive target for STAT5-induced long-term self-renewal. Furthermore, we show that HIF2a is a novel STAT5 target gene which plays an important role in STAT5-induced stem cell phenotypes.