Generation of Synthetic T-ALL By De Novo Transformation of Human Cord Blood Progenitors with a 4-Oncogene Cocktail

Abstract Background: T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive form of blood cancer that can arise in both children and adults. Numerous studies have explored the effects of putative T-ALL oncogenes in mouse models and have contributed significantly to our understanding of disease pathogenesis. Nonetheless, it is clear there are important differences between mouse and human cells, particularly with respect to cellular transformation, and additional work is therefore needed to generate more accurate models of human disease. We sought here to create human T-ALL in the lab from normal CB progenitors by lentiviral transduction with a combination of known T-ALL oncogenes. Methods: Human CD34+ hematopoietic progenitor cells were isolated from pooled cord blood by magnetic bead/flow cytometric sorting (MACS/FACS). Sorted cells were then transduced by lentiviral vectors encoding a combination of four known T-ALL oncogenes including activated NOTCH1. NOTCH1 virus was marked with a GFP reporter (N1/GFP) while the other three accessory oncogenes were marked with a Cherry reporter (3xOnc/Cherry). Transduced cells were cultured on OP9-DL1 stromal feeders briefly prior to transplantation into NOD/SCID-IL2Rg-null (NSG) mice to assess leukemogenesis, or for longer periods to study their behavior in vitro. Results: Initial transduction efficiencies were typically 3-5% for each virus with 1-2% doubly-transduced N1/GFP+, 3xOnc/Cherry+ cells (hereafter referred as 4xOnc cells). After 28 days culture in vitro, the 4xOnc population reproducibly expanded and outcompeted singly- and non-transduced populations, accounting for more than 70% of cells in mixed cultures. By absolute cell counts, non-transduced cells stopped expanding within the first few weeks; however, 4xOnc cells kept expanding even after 6 weeks of culture. To test leukemogenesis in vivo, CD45+ cells were FACS sorted after 10 days of culture on OP9-DL1 feeders (including doubly-, singly-, and non-transduced populations) and injected intrahepatically into NSG neonates. Engraftment of human cells was followed monthly by flow cytometry of peripheral blood. Engraftment of GFP+ Cherry+ 4xOnc cells was first detected 2 months after transplantation whereas no engraftment of singly- or non-transduced cells was detected. The level of engraftment was below 5% and did not increase substantially even after 6 months following transplantation. At day 203 post-transplant, the primary recipient was sacrificed and 4xOnc cells were recovered from bone marrow, spleen and thymus where the levels of engraftment were approximately 10%. 4xOnc cells from the primary recipient were then serially transplanted into secondary recipients. Engraftment of 4xOnc cells in secondary recipients was observed 5 weeks after transplant. Unlike the primary recipient, however, the percentage of 4xOnc cells in the peripheral blood of secondary recipients gradually increased and these animals developed clinically morbid disease by 20 weeks post-transplant. At the time of necropsy, splenomegaly, lymphadenopathy, and enlarged thymus were observed and the bone marrow contained 80-90% 4xOnc cells. By flow cytometric analyses, 4xOnc cells expressed CD2, CD3, CD7, CD38, and TdT supporting acute T-cell leukemia. Also, TCR gamma clonality assay was performed with genomic DNA from 4xOnc cells from secondary recipients and revealed of 5-7 distinct clonal populations. These in vitro and in vivo findings were observed with multiple experimental replicates and with different pools of cord blood. Conclusion: Our in vitro and in vivo results suggest that NOTCH1, in combination with 3 accessory oncogenes are sufficient to transform normal human blood cells into clonal T-ALL-like malignant cells. Although we cannot exclude the possibility of the spontaneous acquisition of additional co-operating genetic or epigenetic abnormalities, this model provides a significant step forward to reveal the mechanisms involved in human T-ALL pathogenesis. Disclosures No relevant conflicts of interest to declare.