Mitochondrial Membrane Potential Defines the HSC Population with Serial Repopulating Potential in Young but Not Aged Animals

The process of haematopoiesis is tightly regulated by the different cellular components in the bone marrow (BM) microenvironment. During aging there is a phenotypical expansion of haematopoietic stem cells (HSC) in the BM niche and a shift towards a myeloid bias in differentiation. We have recently shown that the metabolic response of haematopoietic progenitors to infection is impaired in the aged BM microenvironment and this is driven by senescent changes in the BM stromal cells rather than just intrinsic changes of the progenitor cells (1). Here we dissect the intrinsic changes of HSCs in aged BM to define the true regenerative HSC population using a metabolic signature. BM was isolated from young (8-12 weeks) female C57Bl/6 mice prior to comparing mitochondrial membrane potential (ΔΨm) in HSCs, defined as Lin-, Sca1-, c-Kit+, CD150+, CD48- cells. Within the HSC gated population cells were separated into two groups based on TMRM levels: TMRM high (TMRMhi) and TMRM low (TMRMlo) cells. To determine if the observed TMRM signature affects HSC function, a competitive transplant model was used. TMRMhi and TMRMlo HSCs were FACS purified from 8-12 week old (young) PepCboy (CD45.1+) mice, and adoptively transferred into young C57Bl/6 (CD45.2) mice together with competitive LSKs from C57Bl/6 (CD45.2) mice. 12 weeks post engraftment BM was analysed. TMRMhiPepCboy CD45.1 HSCs did not show evidence of engraftment. However, TMRMlo cells did engraft and analysis showed HSC populations gave rise to CD45.1 cells with both TMRMhi and TMRMlo signatures. A secondary transplant was performed with the CD45.1 HSCs from the C56Bl/6 mice previously engrafted with TMRMlo cells. This showed subsequent repopulation potential of these TMRMlo HSCs. In aged mice (18-24 months) we found that the frequency of HSCs with the TMRMlo signature is significantly increased. We therefore wanted to determine if all these cells maintain the function and particularly their repopulation ability. Whilst we found the TMRMlo HSCs from aged PepCboy mice can repopulate the BM of young C57Bl/6 mice in a competitive transplant, this ability was lost in the secondary transplant performed as described for the young mice. This suggests a decline in HSC function with age despite the expansion of the TMRMlo cell population. Next, we therefore wanted to further define the TMRMlo HSC populations and to identify the changes that occur with age and in particular to determine if there are distinct cell populations amongst the TMRMlo HSCs in aged mice. To do this single TMRMlo and TMRMhi HSCs were FACS purified and the single cell sequencing SmartSeq2 protocol was used to further define these cell populations in young and aged mice. In conclusion, here we describe that in young animals low TMRM level of mononuclear cells in the FACS sorting HSC gate contain the cell population with serial transplantation repopulating potential. However, we report that this TMRMlo HSC population is expanded in aged mice but these cells have reduced stem cell function. Finally, we use single cell RNA sequencing to further define the sub-populations of TMRMlo HSCs from aged mice compared to those from young mice to explore the differences we observe. (1) Hellmich C, Wojtowicz E, Moore JA, Mistry JJ, Jibril A, Johnson BB, et al. p16INK4A dependent senescence in the bone marrow niche drives age-related metabolic changes of hematopoietic progenitors. Blood Adv. 2022.