Quiescent hematopoietic stem cells accumulate DNA damage during aging that is repaired upon entry into cell cycle

Abstract

Hematopoietic stem cells (HSCs) maintain homeostasis and regenerate the blood system throughout life. It has been postulated that HSCs may be uniquely capable of preserving their genomic integrity in order to ensure lifelong function. To directly test this, we quantified DNA damage in HSCs and downstream progenitors from young and old mice, revealing that strand breaks significantly accrue in HSCs during aging. DNA damage accumulation in HSCs was associated with broad attenuation of DNA repair and response pathways that was dependent upon HSC quiescence. Accordingly, cycling fetal HSCs and adult HSCs driven into cycle upregulated these pathways leading to repair of strand breaks. Our results demonstrate that HSCs are not comprehensively geno-protected during aging. Rather, HSC quiescence and concomitant attenuation of DNA repair and response pathways underlies DNA damage accumulation in HSCs during aging. These results provide a potential mechanism through which premalignant mutations accrue in HSCs.

Figure 1

DNA damage accumulates in HSCs during aging. (A) Representative alkaline comets of young and old HSCs. (B-C) Olive Tail Moment (B) and percent of DNA in tail (C) of 710 HSCs from young mice, 447 HSCs from old mice, and 77 HSCs dosed with 2Gy IR. *** p<0.001 (see also Figure S1 and Table S1)

Figure 2

Age-associated DNA damage accrual is greatest in the HSC compartment. (A) Representative alkaline comets of HSCs, multipotent progenitors (MPP^Flk2- and MPP^Flk2+) and oligopotent progenitors (GMP and CLP) isolated from young mice. (B-C) Olive Tail Moment (B) and percent of DNA in tail (C) of HSC (n=1620), MPP^Flk2- (n=714) MPP^Flk2+(n=324) GMP (n=333) and CLP (n=713) from young mice. HSCs (n=292) that received 2Gy of irradiation were also scored. (D) Representative alkaline comets of HSCs, multipotent progenitors (MPP^Flk2- and MPP^Flk2+) and oligopotent progenitors (GMP and CLP) isolated from old mice. (E-F) Olive Tail Moment (E) and percent of DNA in tail (F) of HSC (n=424), MPP^Flk2- (n=578) MPP^Flk2+(n=479) GMP (n=309) and CLP (n=503) from old mice. The same irradiated controls (292 HSCs with 2Gy) are shown, as all samples were arrayed on one slide. *** p<0.001 (see also Figure S1, S2)

Figure 3

HSCs recognize and repair DNA damage upon stimulation into cell cycle regardless of age. (A) Olive Tail Moment of HSCs isolated from young and old mice at steady state (n=749 and 694 respectively) or after 24 hours in culture (Young 24 hour (n=385) Old 24 hour (n=649)). 294 irradiated HSCs were used as a positive control. (B) Olive Tail Moment of aged HSCs after receiving two doses of PBS (n= 1107) or 5-FU (n=1195) and irradiated control cells (n=176) (C) Olive Tail Moment of donor derived HSCs from either young donor HSC (n=2310) or aged HSC (n=1746) 12 months post competitive transplant and irradiated control cells (n=379) *** p<0.001 (see also Figure S3)

Figure 4

Clonal analysis of single HSCs from young and old mice (A) Individual HSC clones scored daily for six days from three young mice and three old mice. Numbers of cells scored daily are presented in a color scale from white=1 cell to red>32 cells. Each clone was then cultured an additional 6 days and scored for types of cells generated from each clone. (B) Summary of clones derived from 318 young and 337 old single HSCs after six days in culture. Each clone was assayed at time points 0.5, 1, 2, 3, 4, 5, and 6 days and the composite data is presented. (C) Cell division kinetics of young and old HSCs. (D) Overall colony size at day 12 of clones derived from single young or old HSCs. (E) Colony composition of colonies generated from single HSCs isolated from young and old mice. *** p<0.001

Figure 5

Attenuation of DNA damage response and repair pathways in HSCs compared to downstream progenitor populations. (A) Fold change comparisons of genes involved in DNA damage response and checkpoints (DDRC), nucleotide excision repair (NER), base excision repair (BER), mismatch repair (MMR), non-homologous end joining (NHEJ), and homologous recombination (HR) in progenitor populations compared to their age-matched HSCs. Each column represents an individual replicate, and the log2 fold change compared to the average expression of the HSCs is shown. Significant expression changes defined as >1.5 fold and p < 0.05, are designated with a bold black boarder. (B) Frequency of genes that show significant up-regulation (red), significant down-regulation (blue) or no significant change (grey) in each pathway for comparisons between HSCs and the indicated progenitors. The global frequencies of the total number of genes showing significant differential regulation out of the total 17,872 genes examined on the arrays is also shown (Global). (C) Statistical analysis of the significance of the changes in each of the indicated DNA damage response and repair pathways in progenitor cells compared to the HSCs from either young or old mice. p-values are presented by a color scale and odds ratios of less than one are indicated with a hash through the box. (see also Figure S4)

Figure 6

Attenuation of DNA damage response and repair genes in quiescent HSCs (A) Fold change comparisons between genes involved in DDR and repair in fetal liver HSCs compared to adult HSCs young or old. Each column represents an individual replicate and the log2 fold change compared to the average expression of the fetal liver HSCs is shown. Significant expression changes, defined as >1.5 fold and p < 0.05, are designated with a bold black boarder. (B) Frequency of genes that show significant up-regulation (red), significant down-regulation (blue) or no significant change (grey) in each pathway for comparisons between fetal liver HSCs and either young or old HSCs. The global frequencies of the total number of genes and those with significant differential regulation (up or down-regulated) out of the total 17,872 genes examined on the arrays are also included. (C) Analysis of the changes of the overall pathways involved with DNA damage response and repair in young and old HSCs compared to cycling fetal liver HSCs. p-values are presented by a color scale and odds ratios of less than one are indicated with a hash through the box. (see also Figure S5)

Figure 7

Dynamic expression profiles of DNA damage response and repair genes in HSCs after stimulation (A) Fold change comparisons of genes involved in DDR and repair in freshly purified HSCs in comparison to HSCs stimulated into cycle at different time points (3, 6, 12 and 24 hours). Each column represents an individual replicate and the log2 fold change compared to the average expression of the age-corresponding HSCs is shown. Genes with average expression fold changes of >1.5 and p<0.05 are signified by a bold black boarder. (B) Frequency of genes that show significant up-regulation (red), significant down-regulation (blue) or no significant change (grey) in each pathway in comparisons between HSCs at steady state or 3, 6, 12, or 24 hours post-stimulation for both young and old. (C) Analysis of significance of the changes of the overall pathways involved with DNA damage response and repair in stimulated HSCs compared to steady state HSCs from respective young or old mice. p-values are presented by a color scale and odds ratios of less than one are indicated with a hash through the box. (D) Fold-change profiles of several DNA damage response genes.