Simple combinations of markers that reliably demarcate human long-term repopulating hematopoietic stem cells (LT-HSCs) would facilitate our understanding of their unique molecular state and the development of stem cell-based therapeutics. The complex CD34+CD38-CD45RA-CD90+CD49f+ phenotype is now routinely accepted to define a subset of cord blood CD34+ cells enriched in LT-HSCs (Notta, Science 2011). The same phenotype with the addition of rhodamine can enrich human LT-HSCs within mobilized peripheral blood (MPB) CD34+ cells, but lower stem cell enrichment is observed (Huntsman, Blood 2015). The hematopoietic cytokine thrombopoietin (TPO) and its receptor cMPL act as primary regulators of HSC function. In recent clinical studies, the synthetic TPO agonist eltrombopag was shown to improve long-term trilineage hematopoiesis in patients with immune aplastic anemia, providing evidence that activation of cMPL receptor stimulates residual multipotent LT-HSCs within the bone marrow (BM). Building on these observations, we hypothesized that cMPL might also be a relevant marker to delineate LT-HSC activity within human adult CD34+ cell populations and provide the basis for a novel LT-HSC purification scheme.
To assess the utility of surface cMPL expression for purification of human adult LT-HSCs, we first performed cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) of human MPB CD34+ cells from a healthy volunteer. CITE-seq was conducted with antibodies designed to react with human cMPL receptor and previously described HSC epitopes (CD34, CD38, CD90, CD45RA, and CD49f). We identified 7 distinct clusters in dimension-reduction (UMAP) analysis. By confirming expression levels of HSPC gene signatures and HSC surface markers, such as CD38, we assigned each CD34+ cell cluster to a distinct HSPC subpopulation, including a single LT-HSC population (cluster 2) (Fig A). Notably, CD34+ cell subsets expressing higher levels of surface cMPL expression were highly enriched in transcriptionally defined HSC population (cluster 2) and consistently co-expressing the known human LT-HSC phenotypic markers, including lower levels of surface CD38 and CD45RA expression and higher levels of surface CD90 and CD49f expression (Fig B).
To determine whether surface cMPL expression can enrich functional LT-HSCs, we next performed serial xenotransplantation assays. We used fluorescence-activated cell sorting to partition human MPB CD34+ cells into cMPL high (top 10%) and cMPL low (bottom 10%) populations, and transplanted these cells into immunodeficient NBSGW mice. Serial peripheral blood (PB) sampling of primary xenografted mice revealed a striking difference in patterns of hematopoietic reconstitution over time between the two groups. In mice transplanted with cMPL high cells, human cell chimerism increased gradually throughout the 16-week engraftment period, while a progressive decline in engraftment was observed in the cMPL low group. Notably, mean human cell engraftment within the PB, BM and spleen of mice transplanted with cMPL high cells was 209-fold (p < 0.01), 37-fold (p < 0.0001) and 283-fold (p < 0.001) higher than mean human cell chimerism in the cMPL low group at 16 weeks post-transplantation, respectively. These data suggest that cMPL high and cMPL low populations are distinctly enriched in LT-HSCs and hematopoietic progenitors, respectively. Next, to quantitatively compare the frequency of self-renewing LT-HSCs within the cMPL high and cMPL low HSPC subsets, human CD45+ cells obtained from the BM of primary mice were injected into secondary NBSGW mice at limiting dilution and BM engraftment was measured at 16 weeks post-transplantation (total period of engraftment: 32 weeks). After accounting for engraftment parameters (i.e., cell dose and engraftment levels) in both primary and secondary mice, extreme limiting dilution analysis computed self-renewing LT-HSC frequencies of 1 in 1,267 within the cMPL high CD34+ cell population, and 1 in 149,010 in the cMPL low CD34+ cell fraction, representing a 116-fold enrichment of LT-HSCs using cMPL as a single marker purification strategy.
Collectively, these data suggest that cMPL expression can effectively delineate LT-HSCs in human adult CD34+ HSPCs and thus serve as an additional relevant marker for the development of human adult HSC targeting therapeutics.
Disclosures
No relevant conflicts of interest to declare.
