Fig 1.
Frequencies of human and murine hematopoietic stem and progenitor cells at different ontogenic stages.
(A-B) Gating strategy and HSPC frequencies for human donors. (A) Flow cytometric profile and gating strategy for h-CLPs (CD10+CD45RA+), h-HSCs (CD123low/-CD45RA-CD90+), h-MEPs (CD123low/-CD45RA-CD110+) and h-GMPs (CD123low/-CD45RA+) in one aged bone marrow sample. Pre-gated on viable CD34+Lin- singlets. (B) Frequencies of depicted early hematopoietic cell types within the primitive CD34+Lin- fraction. Each point represents one donor. H-HSCs: aged n = 44, young n = 52, CB n = 58. H-GMPs: aged n = 42, young n = 52, CB n = 45. H-MEPs: aged n = 29, young n = 46, CB n = 13. H-CLPs: aged n = 42, young n = 52, CB n = 49. (C-D) Gating strategy and HSPC frequencies for mouse donors. (C) Flow cytometric profile and gating strategy for m-HSCs (Sca1+KithighCD150+CD48-), m-CLPs (Sca1+/-cKitlowFlt3+IL7Ra+), m-MkPs (Sca1-cKithighCD150+CD41+), m-pGM/GMPs (Sca1-cKithighCD41-CD150-CD105-), m-pMegEs (Sca1-cKithighCD41-CD150+CD105-), m-pCFU-Es (Sca1-cKithighCD41-CD150+CD105+) and CFU-Es (Sca1-cKithighCD41-CD150-CD105+) in one old murine bone marrow sample. Pre-gated on viable cKit+Lin- singlets. (D) Frequencies of depicted early hematopoietic cell types within the primitive Lin-cKit+ fraction. Each point represents one donor. Young n = 31, aged n = 9. Analyses were performed with unpaired t-tests. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001. Reference lines depict means ± SEM.
Fig 2.
(A-B) Proliferation of human HSCs at different ontogenic stages. (A) Fractions of candidate h-HSCs (cloning frequency) that proliferate in vitro. Cloning frequency was determined as the frequency of wells in which single cells had been sorted and two or more cells was observed after eleven days in culture. (B) Clonal size of proliferating h-HSCs, scored as wells containing either 2–50 cells or >50 cells among wells that showed proliferation. Error bars represent +SEM. CB n = 6, young n = 4, aged n = 4. (C) Lymphoid potential of candidate HSCs. 10 h-HSCs each were sorted into multiple wells (20–60/donor) and scored after 5 weeks in culture. Graph depicts lineage distribution from wells with proliferation. Error bars represent -SEM. CB n = 6, young n = 10, aged n = 6. Analyses were done with unpaired t-tests. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. (D-E) Up- and downregulated GO terms in aged human BM HSCs. Differentially expressed genes between young and aged human HSCs with adjusted p-values < 0.1 were subjected to analysis of enriched gene ontology (GO) terms using DAVID. (D) Top 20 GO terms of upregulated genes in aged HSCs. (E) GO terms of conserved downregulated genes in aged HSCs.
Fig 3.
Enrichment of age- and lineage-specific gene expression signatures for human HSPCs.
(A-D): Age-associated enrichment of lineage-affiliated signatures. Conventional GSEA against lineage-associated gene sets of h-CLPs (A), h-MEPs (B) and h-GMPs (C) for differential enrichment between young (left) and aged (right) BM h-HSCs (permutation type: phenotypes, FDR < 0.25). Significantly enriched signatures are marked with blue borders if enriched to the left, and with red borders if enriched to the right. (D) High throughput GSEA of progenitor-associated signatures to young and aged h-HSCs using BubbleMap. Saturation of bubbles reflects significance (permutation type: phenotypes, B-Y FDR < 0.25). Color represents enrichment to the left side (blue, young) or to the right side (red, aged). The size of the bubbles reflects quantification of the nominal enrichment score (NES). The number of probes in respective gene sets are indicated within brackets. (E-K): Lineage-associated enrichment of age-affiliated signatures. (E-J) Conventional GSEA against age-associated h-HSC gene sets for differential enrichment to the depicted lineages (permutation type: gene sets, FDR < 0.05). Significantly enriched signatures are marked with blue borders if enriched to the left, and with red borders if enriched to the right. (K) High throughput GSEA of age-associated h-HSC signatures between h-CLPs, h-MEPs, and h-GMPs using BubbleMap. Saturation of bubbles reflects significance (permutation type: gene sets, B-Y FDR < 0.05). Color represents enrichment to the left (blue) or right (red) side.
Fig 4.
Enrichment of age- and lineage-specific gene expression signatures for mouse HSPCs.
(A-I): Age-associated enrichment of lineage-affiliated signatures. Conventional GSEA against lineage-associated gene sets generated by comparison of the three progenitor populations m-CLPs (A), m-preMegEs (B) and m-pGM/GMPs (C) for differential enrichment between young (left) and aged (right) BM m-HSCs. Significantly enriched signatures are marked with blue borders if enriched to the left, and with red borders if enriched to the right (permutation type: gene sets, B-Y FDR < 0.05). (D) High throughput GSEA of the progenitor-associated signatures used in A-C to young and aged m-HSCs using BubbleMap (permutation type: gene sets, B-Y FDR < 0.05). (E-H) Conventional GSEA against lineage-associated gene sets generated by comparison of the four progenitor populations m-CLPs (E), m-MkPs (F), preCFU-Es (G) and m-pGM/GMPs (H) for differential enrichment between young (left) and aged (right) BM m-HSCs (permutation type: gene sets, FDR < 0.05). (I) High throughput GSEA of the progenitor-associated signatures used in E-H to young and aged m-HSCs using BubbleMap (permutation type: gene sets, B-Y FDR < 0.05). (J) Lineage-associated enrichment of age-affiliated signatures. High throughput GSEA of age-associated m-HSC signatures between m-CLPs, m-preMegEs, MkPs, preMegEs, preCFU-Es, and m-GM/GMPs using BubbleMap (permutation type: gene sets, B-Y FDR < 0.05).
Fig 5.
Conserved up- and downregulated GO terms in aging BM HSCs between humans and mice.
Conserved genes among the the top 500 most significantly up- and downregulated genes, respectively, in human and mice aged HSCs were subjected to analysis of enriched GO terms using DAVID. (A) Top 20 GO terms of conserved upregulated genes in aged HSCs. (B) GO terms of conserved downregulated genes in aged HSCs. X-axis indicates p-values of the depicted GO terms.
Fig 6.
Summary of conserved concepts in aged BM HSCs between humans and mice.
Aging of the HSC-enriched compartment of the bone marrow is associated with a decreased lymphoid specification, whereas megakaryocytic/erythroid specification is increased.