Laboratory for Human Disease Models

Since we started up our lab in 2006, we aimed to achieve the followings: 1. Through understanding how human HSCs generate diverse blood and immune cells and which genes define robustness of hematopoietic stemness, we want to make allogeneic stem cell transplantation safer in the future. 2. Unlike genetically-engineered mouse models, patient leukemias are so heterogenous in genetics, treatment response and clinical outcome. We therefore sought to recapitulate patient disease using patient cells, and then connect genetic and transcriptomic heterogeneity with cellular dynamics and function. 3. Key to suppress tumor cell activity should be immunity. We want to understand “memory” or “longevity” of human immunity and engineer immune cells as a new and curative treatment for eliminating leukemia with genetic complexity and clinical aggressiveness.

This year, our effort finding CXCR4 as a critical molecule structuring immune cell memory in T cells has been accepted (Nature Communications, in press). In the work, by analyzing T cell proteomic profiles of patients with long-term remission after stem cell transplantation, we found correlation between immune cell memory signature and high expression of CXCR4. At the same time, we focused on two cell surface antigens, CD25 and CD96, as therapeutic targets on surface of leukemic cells. By making antibodies for the two antigens, we established CXCR4-expressing CAR-T cells targeting those antigens. With engineering CXCR4 expression, CAR-T cells could better eradicate leukemic cells in peripheral blood, bone marrow and spleen of patient-derived xenograft models compared with conventional/CXCR4(-) CAR-T cells. By scRNAseq and proteome/CyTOF analyses, we found that CXCR4-expressing CAR-T cells could achieve memory status. We are now examining the engineering of CXCR4 expression in CAR-T cells can also be applied for treatment against acute lymphoblastic leukemia of T and B cell lineages.