Towards a B-Lymphoid Model of E2A-PBX1-Mediated Leukemogenesis: Evaluating the Impact of Hematopoietic Cell of Origin on the Transformation Properties of a Leukemogenic Transcription Factor

Abstract

The t(1;19) chromosomal translocation is present in 5% of acute lymphoblastic leukemia (ALL) cases and leads to expression of the oncogenic transcription factor, E2A-PBX1. Although t(1;19) is exclusively associated with pre-B ALL in clinical cases, murine models produce myeloid or T-lymphoid leukemias, which are not representative of the clinical disease. In this work, we have advanced progress towards the development an E2A-PBX1-driven experimental leukemia model. We initially determined that lineage-negative (lin-) hematopoietic progenitors expressing E2A-PBX1 expression fail to repopulate the B-lymphoid lineage when transplanted into irradiated recipient mice. Furthermore, E2A-PBX1 expressing, lin- fetal liver progenitors (FLPs) fail to differentiate into B-lymphocytes ex vivo. The majority of E2A-PBX1-expressing FLPs manifested an immature phenotype and displayed stem cell factor (SCF)-dependency and enhanced self-renewal. Additionally, these cells retained myeloid potential upon transplantation or stimulation with granulocyte macrophage colony-stimulating factor (GM-CSF). DNA binding was required for the differentiation block, suggesting that E2A-PBX1 target genes are incompatible with B-lineage specification. E2A-PBX1 FLPs had a stem cell like gene expression profile, including up-regulation of the leukemic transcription factors, Hoxa9 and Meis1. These findings explain why E2A-PBX1-driven bone marrow transplant models fail to generate B-lymphoid disease and suggest that future efforts in developing a model of E2A-PBX1-driven pre-B ALL leukemia should focus on expressing E2A-PBX1 subsequent to B-lymphoid commitment. In an attempt to override the B-lymphoid differentiation block, we next expressed E2A-PBX1 in primary pre-B cells. E2A-PBX1 induced an apoptotic response in pre-B cells, which was consistent with previous observations. Since pre-B ALL induction requires secondary genetic events, we attempted to abrogate these E2A-PBX1-mediated effects by modulating expression of the Cdkn2a locus. Loss of Cdkn2a through deletion or Bmi1 overexpression failed to ameliorate the apoptotic response, suggesting that E2A-PBX1 mediated apoptosis occurs independently of Cdkn2a in murine pre-B cells. However, in the absence of Cdkn2a, co-expression of constitutively active MerTK or Ras attenuated the E2A-PBX1 mediated apoptosis. Cumulatively, these results support the notion that t(1;19) occurs subsequent to B-lymphoid commitment and requires multiple secondary genetic lesions. Data presented in this thesis represents crucial initiating steps towards the development of a pre-B ALL model mediated by E2A-PBX1.