Tracking Reconstitution Dynamics in Mice Co-Transplanted with Hematopoietic Stem Cells From Nine Distinguishable Donor Types

Abstract 1890

For decades, the functional properties of mouse hematopoietic stem cells (HSCs) have been studied using transplantation-based experiments. These assays require the transplantation of donor cells that are distinguishable from those of the recipient. Most current studies utilize mouse strains congenic for the leukocyte common antigen CD45 and specific monoclonal antibodies to distinguish donor and recipient derived blood cells by flow cytometry. In recent years, there has been an increased interest in clonal tracking of HSC output, in particular to study HSC heterogeneity, post-transplant regeneration, activation versus dormancy of HSCs in steady-state hematopoiesis, and for gene therapy purposes. Recent progress has been made in cellular barcoding, which has the potential to simultaneously track multiple uniquely marked clones. While powerful, this technique is limited in that it requires an in vitro transduction step, which dramatically decreases the quantity and quality of HSCs in the test cells. Furthermore, cellular barcoding is not currently amenable to flow cytometry based analysis and is therefore of much lower resolution. Conversely, transplantation of single HSCs is a powerful and direct approach to study the output of individual HSCs, however, these experiments are costly and inefficient. In addition, these experiments by design exclude effects due to competition between different HSCs. It would be particularly informative if multiple distinguishable HSCs could be co-transplanted and their individual contribution to hematopoiesis tracked over-time using flow cytometry, however, to date no such system exists. To address this need, we developed and implemented a simple strategy to generate multiple distinguishable HSCs sources, by crossing mice with different CD45 alleles (CD45.1, CD45.2 or CD45.1/5.2) with mice transgenic for the fluorescent proteins dsRed, GFP, or CFP. In this manner, HSCs could be isolated from up to nine unique unmanipulated donors. As a proof of principle, we co-injected varying numbers of purified HSCs from all nine donor types and tracked the contribution of each donor to long-term hematopoiesis. We observed that initially, multiple HSC types contributed to blood cell output in a dose-dependent manner but that ultimately one or two tended to dominate hematopoiesis. We also tested the regenerated HSC pool by transplanting regenerated bone marrow into groups of secondary recipients. Within each group, the hematopoietic output from each donor type was similar in almost all cases, confirming that the regenerated HSC pool is stable in function. Next, we determined the composition of donor types within the primitive Lin-Sca+cKit+CD48- (LSKCD48-) subset of bone marrow and compared this with the output of mature blood cells in the same recipients. This analysis revealed that the donor composition of the LSKCD48- pool correlated best with myeloid blood cell output. Although several instances were observed where a particular donor type was observed in the blood but not the LSK48- cells, these were limited to those with exclusively lymphoid cells, suggesting that the HSCs of that donor type were exhausted. Of particular interest, we also observed rare examples where a particular donor type was clearly detectable in the LSK48- cells but not in the blood, suggesting the presence of a dormant and/or defective HSC clone. In summary, we demonstrate a simple and powerful method to track the hematopoietic output of multiple unique HSCs simultaneosly. This approach has the potential to improve our understanding of the competitive dynamics of post transplant regeneration between different HSC clones, and may shed light on how different HSCs clones contribute to life-long hematopoiesis.