Hematopoietic Stem Cell Aging Promotes Expansion of Tet2 Mutant Clones By Cell Intrinsic Mechanisms

The incidence of clonal hematopoiesis of indeterminate potential (CHIP) increases exponentially with aging. Still, rare clones harboring the prototypical clonal hematopoiesis (CH) mutations are nearly ubiquitous in middle aged individuals, suggesting that cell intrinsic or micro-environmental changes observed with aging can promote the expansion of malignant clones to pathologically relevant levels, increasing the risk of hematological malignancies and a multitude of sequelae associated with CHIP.

To address if aging can alter the fitness dynamics regulating the rate of expansion of malignant clones, we utilized a CRISPR based approach in hematopoietic stem and progenitor cells (HSPC) from mice of different ages to introduce mutations in the Tet2 and Dnmt3a genes or in the Rosa26 safe harbor locus (as competitors), and analyzed their rate of expansion after transplantation in aged matched mice.

Our data show that Dnmt3a KO cells expand with similar rates in young and old recipient mice after primary and secondary transplant. In contrast, Tet2 KO cells expand at a significantly faster rate in old HSPC in old mice (20 months old) in comparison to young HSPC in young mice (2 months old) and in all hematopoietic populations in the peripheral blood and the bone marrow. Importantly, by heterochronic transplantation of young or old donor HSPC in old or young recipient mice respectively, we observe that the rate of expansion of the Tet2 mutant cells is solely determined by the age of the donor, suggesting that cell intrinsic age-associated changes in the HSPC population facilitate the expansion of the Tet2 KO clones.

By using donor mice of different ages, we observed that the rate of expansion of the Tet2 KO cells is significantly accelerated by 7 months of age in both male and female mice. Aging is associated with shifts in the relative frequencies of HSPC sub-populations - therefore we sorted myeloid biased HSC, lymphoid biased HSC, short-term (ST) HSC and multipotent progenitor cells (MPP) from old mice, and after Tet2 CRISPR, transplanted the cells into young recipient mice. Both myeloid and lymphoid biased HSC show a similar accelerated rate of Tet2 KO cell expansion, suggesting that the increased myeloid bias with age is not correlated with the Tet2 KO adaptation. In ST-HSC and MPP, we did not observe a selection for the Tet2 clones in the short term. As expected, we noticed a dwindling engraftment in mice transplanted with control ST-HSC and MPP with time; however, Tet2 KO cells exhibited much more robust and prolonged maintenance in these multi-potent progenitor populations, suggesting that the loss of Tet2 can extend the limited self-renewal of progenitor cells.

Next, we performed competitive transplantations with donor cells with differing genotype (WT or Tet2 KO) or age (2 months or 20 months) to directly compare their relative fitness. Our results show that young and old Tet2 KO HSPC have similar fitness levels, and young Tet2 KO HSPC having a small advantage over young control cells. In contrast, young WT HSPC have a much larger fitness advantage than old WT HSPC. Thus, the enhanced Tet2 KO expansion with age is due to an intrinsic defect of the WT population that cannot effectively compete with Tet2 mutant clones.

To elucidate the molecular mechanisms for the fitness advantage of Tet2 KO cells with aging, we performed scRNAseq on WT or Tet2 KO HSPC from 2, 12 and 20 month old mice isolated 12 weeks after transplantation into age matched recipient mice. By performing GSEA analyses, we found that aging increased signatures related to mTORC1 activity, ribosomal biogenesis and mitochondrial oxidative phosphorylation, as well as pathways related to RUNX1 and p53 activity, all of which were reverted in Tet2 KO cells to young-like levels.

In conclusion, our study supports the hypothesis that TET2 mutations observed in CH are selectively advantageous in aging due to cell intrinsic changes occurring in HSC. While the specific age-associated alteration(s) in HSC leading to the adaptation of TET2 mutant clones remain to be defined, we find that several transcriptional changes observable with aging are reverted by the loss of Tet2, likely endowing mutant cells with a larger fitness differential than that observable in a young, highly fit HSC population. Interventions aimed at maintaining HSC clones with higher fitness in aging can be a powerful preventative factor against the emergence of CHIP and its negative consequences.