Background: Aging is characterized by accumulation of senescent cells in the hematopoietic stem and progenitor cell (HSPC) compartment. Our previous studies have demonstrated that these senescence HSPCs are associated with elevated central carbon metabolism. In this study, we have defined the functional properties of these senescent HSPC.

Methods: HSPCs (CD34+ cells) were isolated from the bone marrow of healthy human subjects (range: 20 - 74 years; n = 18; old defined as >60 years and young defined as between 19 to 35 years) and separated by FACS in three distinct subpopulations with high, medium, and low glucose uptake (GU) capacity (GUhigh, GUmed, GUlow). Single-cell RNA-sequencing (scRNAseq) studies, normalization and clustering of the sequencing data, cell cycle regression (to excludes changes due to cell cycle), gene ontology (GO) of biological processes, gene set enrichment analysis (GSEA), pseudotime ordering using Monocle 3, and developmental trajectory analyses were performed.

Results: Our previous studies revealed that the GUhigh subset is coupled with significantly higher expression of genes involved in myeloid development, cell cycle arrest, senescence associated secretory phenotype (SASP), anti-apoptosis and survival pathways, and significantly lower expressions of genes involved in lymphoid development. Single cell transcriptomic studies showed that the GUhigh and GUmed subsets consist mainly of HSPCs involved in myeloid development and GUlow subset in lymphoid development. In this study, we have compared the differential gene expressions between old versus young subjects in each of the three subsets with distinct glucose metabolic levels: GUhigh, GUmed and GUlow. The major findings from GSEA (using the gene sets: Hallmarks, Wikipathways, Reactomes, and KEGG) that are relevant to cellular senescence are summarized in Figure 1. These results confirm that the deregulations inherent in senescent cells are found mainly in the GUhigh subset of elderly human subjects. Comparing the frequencies of old versus young GUhigh cells along the cell cycle progression ranks showed a significant reduction of cells in the late G1 phase among GUhigh cells. The differentiation trajectories of HSPCs between old and young human subjects, especially of the GUhigh subset, develop differently (Fig. 1B). When comparing the trajectories of the GUhigh subsets (i.e. senescent HSPCs) between old versus young subjects, we found a significant increase in upregulation of genes inherent in non-primed HSPCs (ADGRG6, HLF, PTPRC) and significant down regulation of genes characteristic for primed HSPCs (HBB, GYPC, CSF2RB, CD74, CD47, LAIR1) in the elderly.

Conclusions: Our series of proteome and single cell transcriptome analyses have demonstrated that the GUhigh subpopulation is highly enriched in senescent HSPCs. Cell cycle analysis has revealed a significantly shortened G1 phase in these senescent cells. Developmental trajectories have demonstrated that these cells are in dedifferentiated state. Above all, we have provided evidence for a dependency of senescent HSPCs on elevated glucose metabolism and on anti-apoptotic factors for survival. These mechanisms may represent targets for elimination of senescent HSPCs.

Legend for Figure 1:

A. Results of Gene Set enrichment analysis (GSEA) comparing the GUhigh subsets of CD34+ cells derived from old versus young human subjects (in first column).The enrichment scores (x102) (ES) are reflected in the size of the circles and the statistical significance (p-values) in the color scale of the circles. Negative ES, i.e. enriched in the GUhigh of young adults, are found in "Mitotic spindle", "DNA repair", "B cell receptor signaling pathway", and in "Apoptosis". The corresponding comparisons between old versus young subjects for GUmed and GUlow subsets are depicted in the second and third columns.

B. Developmental trajectories of the HSPCs derived from old versus young human subjects. 1 = early progenitor cells, E = erythroid cells, Meg = megakaryocytes, G = granulocytes, Ly = lymphoid cells. HSPCs, especially the GUhigh subset, from the elderly have a distinctly different developmental trajectory than the corresponding subset from the young.

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution