Identification and Modeling of TEL-AML1 (ETV6-RUNX1) Molecular Signature In Acute Lymphoblastic Leukemia

Abstract 2500

Despite high cure rates for childhood ALL, treatment failure remains a formidable problem. TEL-AML1 (ETV6-RUNX1), the product of t (12;21) translocation, is the most common cytogenetic abnormality in childhood cancer, and is detected in 25–30% of precursor B-cell (pre-B) acute lymphoblastic leukemia (ALL) in children, and in small percentage of adult ALL. Considerable experimental and clinical evidence indicate that the TEL-AML1 fusion is insufficient by itself for leukemic transformation. First, monozygotic twins had the same prenatal TEL-AML1 sequence, but different latency and/or no leukemia development. Second, we have previously established transgenic zebrafish expressing TEL-AML1. These TEL-AML1 transgenic zebrafish have hematopoietic stem cell (HSC) expansion, and develop pre-B ALL at low penetrance, and after prolonged latency. Similarly, transgenic mice expressing TEL-AML1 did not develop pre-B ALL, but showed HSC expansion. Third, retroviral-mediated TEL-AML1 gene transfer into murine HSCs resulted in a preleukemic state, and the incidence of leukemia increased with additional mutation. Here, we report on the identification of a TEL-AML1-specific leukemic signature in ALL cases, and modeling of these mutations in dual and compound transgenic zebrafish. First, we performed meta-analyses of ÓALL compendiumÕ of molecular signatures and expression profiles of 990 ALL cases from six studies, measuring the expression of 14,145 genes in 475 arrays. The normalized data were imported into the Ingenuity Pathway Analysis (IPA) software to identify TEL-AML1-related pathways. We determined a TEL-AML1-specific signature that was organized into modules that are induced or suppressed, based on involvement in several biological pathways comprising the hallmarks of cancer. These analyses identified modules of cell differentiation, cell proliferation, apoptosis, autophagy, cell cycle regulation, and lymphocyte development as the most common modules associated with TEL-AML1. The signature was confirmed using additional microarray analyses, in combination with Q-PCR, and western blotting from the same cases. Similar analyses of marrow cells from transgenic zebrafish with ALL identified 1,128 upregulated and 936 down-regulated genes with 27 genes common with the human TEL-AML1 signature. We next isolated the zebrafish homologues of several induced signature genes to generate dual and compound transgenic zebrafish, and to investigate the effects of their overexpression on TEL-AML1 leukemia development. Analyses of established transgenic fish demonstrate leukemia-enhancing effects of signals within the proliferation and cell cycle modules. These studies provide a model to understand the role of the TEL-AML1 fusion and the secondary mutations required for leukemia development, and might present a rational for leukemia combination therapy.