Transcriptomic identification of functionally potent human hematopoietic stem and progenitor cells Free

Allogeneic hematopoietic cell transplantation is a potentially curative treatment for malignant and non-malignant hematologic disorders, wherein healthy hematopoietic stem (HSC) and progenitor cells (HSCs) reconstitute a functional blood and immune system following ablation of diseased cells. Functionally potent HSCs/HPCs from bone marrow, mobilized peripheral blood, or umbilical cord blood drive recovery of the hematopoietic system following transplantation. Improving initial characterization of the engraftment potential of donor grafts and recovery time will lead to improved patient outcomes, particularly in the case of umbilical cord blood transplantation, which is limited by low total numbers of cells. However, we still lack a complete understanding of the molecular programs that regulate potency, which is the ability of cells to home to and engraft in a hematopoietic niche, then self-renew and differentiate driving multi-lineage repopulation. Here, we use transcriptomic approaches to elucidate gene programs associated with hematopoietic cell potency in CD34+ cells, which are enriched for HSCs/HPCs.

1) We compared transcriptomic differences between immunophenotypically defined HSCs/HPCs, which have different potentials. 2) We examined the transcriptomic profile of “homing capable” human CD34+ cells from transplantation recipient mouse bone marrow compared to all CD34+ cells used for transplantation. 3) We elucidated the transcriptome of Aldehyde dehydrogenase Bright (AldhBr) CD34+ cells, which are clinical indicators of potent cord blood units, compared to more functionally deficient populations. 4) We used single cell RNA-seq to compare the transcriptome of fresh cord blood HSCs/HPCs with those that are predicted to have lost potency due to cell culture. 5) We analyzed the gene expression profile of potent HSCs from cord blood based on their high long-term engraftment SCID repopulating cell frequencies in an in vivo mouse model compared to those with low engraftment capacity. We validated a novel target using recombinant protein or siRNA mediated knock down in CD34+ expansion assays.

Results: Integration of all five data sets identified genes and molecular networks that are highly represented in potent HSCs/HPCs defined by different aspects of potency (homing capacity, cell state, and long-term engraftment capacity). Unsurprisingly, we found that pathways associated with stress, interferon responses, and proliferation were associated with altered potency. We also found a consensus across several or all screens that genes associated with mitochondrial regulation, aldehyde dehydrogenase activity, and hypoxia regulated programs were enriched in potent HSCs/HPCs. By computational integration and manual filtering, we identified a list of genes highly associated with potency that includes PRSS2, AVP, SPP1, MDK, SPINK2, CXCL11, ENO1, BEX2 and PIM1, as well as several mitochondrial genes with expression that negatively correlates with potency such as MT-CO2. Interestingly, many of the potency associated genes code for secreted factors, which may indicate a level of HSC/HPC auto-regulation with regards to potency. We found that treatment with a subset of the identified secreted factors or siRNA-mediated knockdown of genes encoding those proteins potency lead to increased or reduced HPC CFU capacity, respectively, confirming their importance in hematopoietic function ex vivo.

We found gene programs associated with hematopoietic cell potency, and validated a subset as important for hematopoietic function ex vivo. The identified gene networks could be used to develop a new potency assay based on gene expression that would simplify donor unit selection for cell therapies. For example, following potency assay validation, cord blood units could be selected for transplantation based on expression levels of PRSS2, BEX2, ENO1, and AVP. Next, we identified molecular programs that can be stabilized to improve cell function for use in therapies, or that could be targetable vulnerabilities in diseased hematopoiesis. These findings could also be important for other cell therapies such as gene editing or CAR-T or CAR-NK therapies derived from HSCs/HPCs, which also require the creation or maintenance of potent cells. In all, these transcriptomic screens are the first step toward elucidating the complex network of molecular programs that contribute to hematopoietic cell function and potency.