Bioimaging of Vascular Microenvironment Reveals Presence of Fli1a, Not VEGFR2, in Acute Lymphoblastic Leukemia Model in Syngeneic Zebrafish

Abstract 2551

Despite great advances in the curability of children with acute lymphoblastic leukemia (ALL), outcomes for pediatric and adult patients with relapsed ALL remain poor. This highlights the need for new approaches to ALL treatment and novel tools for screening of potential therapeutics. Dysregulated angiogenesis has been implicated both in the pathogenesis of leukemia and development of chemoresistance. However, the complexity of the bone marrow microenvironment and the precise contributions by the various components cannot be easily dissected in cell lines, tissue blocks, or mouse models. For instance, leukemia-associated angiogenesis was mainly characterized by immunostaining of bone marrow biopsies. High resolution in vivo bioimaging of fluorescence-tagged tumors and their microenvironment has recently become feasible due to establishment of transplantable tumor models in clonal syngeneic zebrafish. We have applied this model to dissection of cellular mechanisms of angiogenesis during stages of progression of T-ALL, This novel animal model enables us to 1) track leukemic cell proliferation and dissemination in vivo in different temperio-spatial and vascular contexts, and 2) follow up on cellular angiogenic events in response to leukemia progression including those occurring in response to chemotherapy for leukemia. In zebrafish, angiogenesis is similar to that of mammals and has been well characterized by using VEGFR2 and Fli1a transgenic reporter systems. In embryos, the fli1a:EGFP expression pattern mirrors that of VEGFR2 fluorescence in vascular endothelial cells. However, later in the development, in larvae and adult fish, the pattern of Fli1a or VEGFR2 expression diverges. This divergence further extends to angiogenesis in areas adjacent to T-ALL with more prominent development of Fli1a vasculature. Leukemic patches are characterized by a microenvironment where Fli1a is predominant and VEGFR2 is absent. We are now using cyclophosphamide treatment of leukemic fish to dissect the role of microenvironment and whether angiogenic factors are modified. These data yield new insights into molecular mechanisms of leukemogenesis in conjunction with angiogenesis. Furthermore, our findings would have predicted the lack of efficacy of VEGFR inhibitors in leukemia therapy. Our model offers an advantage for cost-efficient in vivo large scale screening system for antiangiogenic drugs for acute leukemia.