Modeling normal and malignant hematopoiesis in vitro. To screen for extrinsic regulators and differentiation therapy

Abstract: The incredible thing with blood stem cells, also known as hematopoietic stem cells (HSC), is that they can restore normal hematopoiesis in patients that need a new blood system. Since a prerequisite for successful transplantation is immune compatibility, it requires large donor registries to find a suitable match for a recipient. Unfortunately, there is still a shortage of immune-compatible donors in these registries. Thus one potential approach to improve the quantity and quality of these registries is to expand HSC in umbilical cord blood units. However, robust in vitro expansion of human HSC is not possible yet. Thus, in vitro expansion of HSC is a high-value objective in hematological research.A common reason patients need a new blood system is blood cancer. An added benefit of transferring someone’s else blood system to a cancer patient is that the donor’s immune cells can help eradicate the cancer cells. Because the transfer of a new blood system is not without risks, as the donor’s immune cells also target normal tissues, physicians will only transplant when the cancer treatment is not potent enough to eradicate the cancer cells. Even though the cancer therapies of some AML subtypes are effective today, the prognosis of most cancer types would improve with new therapies. Thus, developing new therapies is another high-value objective in hematological research.Here we addressed both of these objectives by using a mesenchymal stroma-based co-culture model for culturing primary acute myeloid leukemia (AML) cells and HSC to identify differentiation therapy of AML (papers I and II), improve culture conditions of HSC (paper III), and for investigating synthetic lethality in AML (paper IV). Thus, the common thread of the four papers included in this thesis is the use of OP9M2 cells to model normal and malignant hematopoiesis.In paper I, we identified a natural product that induces differentiation in AML through activation of the PKC pathway. Moreover, we show that AML with FLT3-ITD or FLT3 mutations are resistant to differentiation, highlighting the importance of neutralizing the effect of mutated FLT3 in differentiation therapy. This study illustrates how small molecule screening and genetic profiling are powerful tools for developing personalized treatments.Paper II is a small molecule screening protocol based on the OP9M2 co-culture model of primary AML cells. With a flow-cytometry readout, the protocol is highly adjustable to different study objectives, including screening for novel therapeutic agents, drug repurposing, drug synergism, patient selection, mechanism of action analysis, and drug resistance. Methods such as these will continue to be crucial for developing new therapies to improve outcomes for many patient groups.In Paper III, we identified potential regulators of HSC, which we screened by shRNA knockdown in the OP9M2 model. However, it did not identify any candidates, likely due to a sub-optimal screening methodology. Still, the list of potential regulators could be helpful for similar studies. Improving in vitro culture conditions remains a high-value objective as cellular therapies will continue to be essential for treating hematological diseases.Paper IV shows that STAG1 and STAG2 have a synthetic lethal interaction in primary AML cells. Thus, targeting STAG1 or STAG2 in STAG1- or STAG2-null AML is potentially a new precision medicine for molecular targeted therapy. This study shows how an in-depth understanding of disease heterogeneity and subtype-specific weaknesses is critical for developing precision medicine.