Session: 501. Hematopoietic Stem and Progenitor Cells and Hematopoiesis: Basic and Translational: Poster II
Hematology Disease Topics & Pathways:
Hematopoiesis, Biological Processes
To comprehensively characterize the development and aging of HSPCs, we collected samples from YS (6-7 post-conception weeks (PCW)), FL (6-17 PCW), fetal BM (14-17 PCW), cord blood, pediatric and adult BM (1-91 years old). CD34+ HSPCs were isolated and subjected to CITE-seq, which allows simultaneous scRNA-seq and quantification of 198 cell surface proteins. A total of 98,266 single HSPCs passed transcriptomic and proteomic quality controls, and the two modalities were then integrated using the weighted nearest neighbor algorithm to generate a single multimodal landscape.
The integrated landscape was clustered and manually annotated using three orthogonal methods: (i) cluster marker gene identification, (ii) label transfer analysis from published atlas datasets and (iii) in-silico flow cytometry using the cell surface proteome data. Overall, 23 different HSPC states were identified across eight hematopoietic lineages (megakaryocyte, erythroid, eosinophil/basophil/mast cell, B-lymphoid, NK/T-lymphoid, monocyte/macrophage, dendritic cell and plasmacytoid dendritic cell). Importantly, the proteome data particularly contributed to a greater resolution of the most immature HSC compartments by successfully identifying phenotypic long-term HSCs (pLT-HSCs; CD34+ CD38− CD45RA− CD90+ CD49f+) and short-term HSCs (CD34+ CD38− CD45RA− CD90+ CD49f−) through our in-silico gating approach.
A comparison of cell type proportions showed that erythroid progenitors were significantly abundant in FL, whereas fetal and pediatric BM samples were significantly enriched in B-cell progenitors, which then progressively decreased with age in adulthood. Robust detection of pLT-HSCs (>10 cells) was first observed in FL at 7 PCW. Thereafter, pLT-HSCs accounted for an average of 1.8% and 1.1% of CD34+ HSPCs in FL and fetal BM, respectively. Intriguingly, pLT-HSCs significantly increased in pediatric BM (average of 6.8%), then diminished (average of 4.1%) in adults under 60 years of age and ultimately increased significantly (average of 14.3%) in adults aged 60 years or older, indicating that the age-specific hematopoietic changes involve quantitative shifts not only in lineage-committed progenitors but also in the most immature pLT-HSCs.
To elucidate the molecular programs associated with the increase in pLT-HSCs in pediatric BM and the further increase in aged BM, we performed differential expression analysis. Interestingly, during the transition from fetal to pediatric BM, interferon-α (IFNα) signaling was most significantly upregulated, which is known to promote perinatal expansion and maturation of HSPCs in mice. Since IFN cytokines are key mediators of pathogen defense, exposure of children to the non-sterile postnatal environment may play a role in this transcriptional change. In aged pLT-HSCs, the NF-κB signaling was most significantly upregulated, indicating that two different inflammatory pathways (i.e., IFNα and NF-κB) are involved at different stages of HSC development and aging.
In summary, our results provide insights into the quantitative and qualitative changes in immature HSPCs across the human lifespan, which ultimately govern the age-specific hematopoietic landscape. Our comprehensive HSPC atlas will be a valuable resource for future studies on HSPC development and aging, as well as on HSPC-derived diseases such as leukemia.
Disclosures: Laurenti: CSL Behring: Research Funding.
See more of: Oral and Poster Abstracts