Fetal Vs. Adult Hematopoietic Progenitors Respond Differently to NOTCH1: Implications for Pediatric Vs. Adult T-ALL.

Abstract 2381

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy of T-cell progenitors which affects both children and adults. Whereas pediatric T-ALL is curable in 80–90% of cases, only 40% of adults with T-ALL survive beyond 5 years. Fetal liver and adult bone marrow-derived hematopoietic stem/progenitor cells (HSPCs) are known to differ in terms of their gene expression programs and functional properties. Despite this work, the extent to which differential programming of fetal and adult HSPCs may impact the biology of their respective leukemias in children and adults remains unexplored.

NOTCH1 is a prominent oncogene in T-ALL and activated by mutation in over 50% of cases. The retroviral NOTCH1 mouse bone marrow transplant model of T-ALL is well established; however, most investigators use 8–12 week-old adult mice as bone marrow donors for these experiments and thus these studies could be interpreted as having modeled adult disease. In order to explore the possibility that fetal program HSPCs would more accurately model pediatric disease, we transduced lineage-negative fetal liver HSPCs with activated NOTCH1 (ΔE) retrovirus, transplanted them into syngeneic (C57BL/6) recipients, and compared the behavior of the resulting leukemias to those generated from lineage-negative 8-week-old adult bone marrow HSPCs.

Primary transplant recipient mice developed nearly identical T-ALL disease in terms of penetrance, latency, disease distribution/burden, and immunophenotype. Serial transplantation of these leukemias into secondary recipients, however, revealed stark differences in that whereas “adult” leukemias were readily transplantable, “fetal” leukemias were largely non-transplantable. In order to quantitate leukemia-initiating cell (LIC) frequencies in these two situations, we performed secondary transplants into highly permissive, immunodeficient (NOD/Scid/Il2rg^−/−) recipients at limiting dilution and observed fetal leukemias to exhibit 500-fold lower LIC activity than adult leukemias (1 in ∼4500 cells vs. 1 in ∼9 cells, respectively).

To identify potential mechanisms that might underlie this difference in LIC activity, we compared the behaviors of fetal liver vs. adult bone marrow-derived HSPCs shortly after transduction with NOTCH1(ΔE) virus. Interestingly, NOTCH1 induced fetal HSPCs to cycle rapidly whereas adult HSPCs were largely quiescent. We also noted that non-transduced cells in fetal HSPC cultures were also cycling rapidly, and through a series of fetal/adult mixing and conditioned media experiments, we determined that NOTCH1 induces an autocrine IGF1 signaling circuit in fetal, but not adult HSPCs. This observation was also confirmed to hold true for CD34+ human cord blood vs. adult bone marrow HSPCs. Expression profiling/qRT-PCR and chromatin immunoprecipitation (ChIP) studies further revealed NOTCH1 to induce IGF1 transcription and altered chromatin structure (increased H3K4me3 and decreased H3K27me3 marks) specifically in fetal, but not adult HSPCs.

These findings suggest that developmental stage-specific programming in fetal vs. adult progenitors underlies their differential responses to oncogenic NOTCH1 signaling, and also the biological aggressiveness of resulting leukemias. Therapeutic targeting of adult-specific pathways may thus achieve improved clinical responses in adults with T-ALL and perhaps also the minority of pediatric patients with more aggressive, possibly “adult-like” disease.