Hematopoietic stem cells (HSCs) have the capacity to differentiate into all hematopoietic lineages and at the same time self-renew to maintain the HSC pool. HSCs have been thoroughly investigated using immunophenotypic-, molecular- and functional-analysis resulting in the development of protocols for high-purity prospective isolation of human HSCs. However, within the current state-of-the-art HSC populations, 90% of the cells lack stem cell activity, confounding molecular analysis of HSC function. Thus, identification of novel immunophenotypic markers to delineate the HSC population would improve our understanding of HSC biology.

To identify cell-surface markers with the potential to discriminate between functionally different cells within the HSC population, we performed antibody screens measuring the expression of 340 markers on human cord blood (CB) and bone marrow (BM). Candidate markers that divide the HSC population were included in single-cell CITE-seq experiments together with conventional HSC and progenitor markers for combined analysis of immunophenotype and RNA sequencing. This allowed us to correlate the molecular signature of each single-cell with the expression of 40 cell-surface proteins in CD34+ and CD34+CD38- populations of fetal liver (FL), CB, young- and old BM.

Following sequencing, the cells were clustered based on molecular signature. Fourteen distinct groups with HSC-, multipotent progenitor-, and early committed progenitor profiles were identified. To investigate how the molecularly defined groups correlate to established populations within CD34+ HSPCs, the surface marker expression from the CITE-seq experiment was included in the analysis. The immunophenotypically defined GMP, MEP and CMP populations showed high molecular heterogeneity with cells at different stages of differentiation. The immunophenotypic HSCs (CD38-CD90+CD45RA-) correlated with the molecularly defined HSC population with a 75.6% overlap. To find novel surface markers for prospective isolation of HSCs pseudo-time analysis was used, allowing for correlation of surface marker expression with differentiation status. Interestingly, both CD35 and CD11a correlated with differentiation, with CD35 expression decreasing and CD11a expression increasing with pseudo-time. These two novel HSC marker-candidates are currently being functionally validated by transplantation analysis.

To compare the progenitor composition of CD34+ HSPCs at different stages of life, young BM was used as a baseline control. Interestingly, compared to young BM CB CD34+ cells contained a higher frequency of multipotent progenitor cells and a decreased proportion of committed progenitors. In contrast, old CD34+ BM was reduced in multipotent progenitor frequencies with a corresponding relative increase of committed progenitors. However, both CB and old BM showed similar proportions of molecularly defined HSCs as compared to young BM. These results indicate that ageing causes a depletion of the earliest hematopoietic progenitor populations while the HSC pool remains intact.

Together, using single cell CITE-seq we can describe the immunophenotypic- and molecular-heterogeneity of the HSC and progenitor populations and identify two novel cell-surface marker candidates for prospective isolation of HSCs.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution