Visualizing the Effects of Chemicals On Hematopoietic Stem Cell Engraftment in an Endogenous Niche

Abstract 506

Hematopoietic stem and progenitor cells (HSPC) self-renew and give rise to all blood cell types throughout adulthood. Definitive HSPC arise from the hemogenic endothelium of the dorsal aorta, are released into circulation, and then seed an intermediate hematopoietic tissue before colonizing the adult marrow. In mammals this intermediate tissue is the fetal liver, and in the zebrafish it is the caudal hematopoietic tissue (CHT), a vascular plexus in the ventral tail of the embryo. We have generated the first highly specific zebrafish transgenic reporter of HSPC, using the previously described mouse Runx1 +23 kb intronic enhancer driving GFP (or mCherry) fluorescent protein. We have demonstrated that these Runx1+23 positive cells are capable of long-term engraftment and multi-lineage contribution. Using time-lapse live imaging in the embryo, we followed HSPC as they migrate to the CHT niche. Together with endothelial (kdrl(flk1):DsRed2) or stromal (cxcl12a(sdf-1a):DsRed2) reporter lines, we could visualize stem cell behavior directly in the endogenous niche. Upon arrival, HSPC underwent a number of distinct steps to engraftment, including: 1) adherence to the vessel wall; 2) extravasation; 3) migration to the abluminal space; 4) triggering of niche formation—endothelial cells actually remodel around a single HSPC to create a niche; 5) cell division decisions. To determine if endothelial niche formation is conserved in mammals during ontogeny, we performed live imaging of mouse fetal liver explants at embryonic day 11.5, the earliest stage of seeding by HSPC. We observed rare c-kit+/Ly6a(Sca1):GFP+ HSPC become centered in a rosette of CD31+/Lyve1+ sinusoidal endothelial cells. This dynamic remodeling of endothelial cells around an HSPC in the niche was strikingly similar to the cellular behaviors we observed in zebrafish. We hypothesized that chemical genetics could reveal the molecular mechanisms and signaling pathways that are associated with the distinct steps of HSPC engraftment. As proof-of-concept, we tested the CXCR4 antagonist AMD3100 because the CXCR4-CXCL12 receptors and ligands are expressed in the CHT, and found that it suppressed CHT hematopoiesis. Next, we performed a chemical genetic screen by applying ∼2400 individual compounds of known action to zebrafish embryos during colonization of the CHT. We found 40 compounds that increased and 107 compounds that decreased CHT hematopoiesis. Applying selected chemical hits in our live imaging assay we found that certain compounds actually modulated distinct steps during engraftment. We identified a role for sphingosine-1-phosphate signaling during extravasation. We observed that regulators of the transcription factor hypoxia inducible factor (HIF)-1α modulated migration into abluminal spaces. The HIF-1α stabilizer dimethyloxalylglycine (DMOG) promoted migration into hypoxic abluminal spaces, while conversely the HIF-1α inhibitor YC-1 promoted migration into normoxic luminal spaces. We found the plant alkaloid Lycorine promoted endothelial niche formation, creating more locations for HSPC and allowing longer residence times in the CHT. The transforming growth factor (TGF)-β receptor inhibitor SB-431542 increased the rate of HSPC division after they had arrived in the niche. Our studies provide the first genetic approach to understanding engraftment, and the chemicals found could be used therapeutically for patients receiving marrow transplantation.