Session: 506. Bone Marrow Microenvironment: Poster I
Hematology Disease Topics & Pathways:
Research, Translational Research, Hematopoiesis, Biological Processes
To investigate the cellular origins of retrotransposon-mediated inflammation within the bone marrow, we performed single-cell RNA sequencing on a cohort of five young (2 months) and five aged (>24 months) C57BL/6 mice. We identified six major BM environment cell types (chondrocytes, endothelial cells, fibroblasts, mesenchymal stromal cells, osteolineage cells, and pericytes) and amongst these cell types, we found that mesenchymal stromal cells (MSCs) robustly reactivated retrotransposon expression during aging. We further validated these findings by purifying MSCs from young and aged mice for bulk RNA-sequencing, additionally finding that type I interferon genes were the most significantly enriched genes in aged MSCs by Gene Ontology enrichment analysis. Our data thus suggested a link between aging-associated reactivation of retrotransposons in the BM environment and innate immune activation.
We next sought to understand whether retrotransposon reactivation can non-cell-autonomously promote the expansion of HSCs harboring mutations in Dnmt3a, the most commonly mutated gene in CH. To do so we utilized mice deficient in Mpp8, which is an epigenetic regulator that normally acts to maintain retrotransposon silencing. We transplanted Vav1-cre;Dnmt3afl-R878H/+ (Dnmt3a-mutated) HSCs and WT HSCs at a 1:5 ratio into busulfan-conditioned recipient Mpp8-/- mice, which recapitulate conditions of retrotransposon activation in the BM environment, as well as WT recipient mice. No difference was observed in peripheral blood donor chimerism over the course of five months, likely due to the well-described differentiation defects that characterize Dnmt3a-deficient HSCs. However, in a terminal BM analysis at 5 months, Dnmt3a-mutated cells demonstrated more robust expansion in Mpp8-/- recipients as compared with WT recipients across hematopoietic stem and progenitor cell compartments (HSC p=0.06, LSK p=0.06, CMP p=0.03, GMP p=0.49, MEP p=0.34) and the absolute frequency of Dnmt3a-mutated HSCs was higher in Mpp8-/- recipients as compared to WT recipients (p=0.057). Together, these results provide the first evidence of a cell-extrinsic role for retrotransposon reactivation in driving CH.
To further investigate whether retrotransposons can directly induce a type I interferon response in MSCs, we ectopically overexpressed two retrotransposons that we found to be upregulated in MSCs during aging, Lx8 and B1, using a murine BM stromal cell line, MS5. We found that both overexpressed retrotransposons robustly induced the expression of IFN-beta as well as downstream type I interferon stimulated genes (ISGs). Additionally, when we co-cultured MS5 cells in the presence of interferon-alpha with a competing mixture of WT and Dnmt3a-mutated HSCs for 7 days ex vivo, we found that the Dnmt3a-mutated cells outcompeted WT cells (p=0.07).
Together, our data support a role for a retrotransposon-mediated type I interferon response in cell-extrinsically promoting the preferential outgrowth of DNMT3A-mutated HSCs during aging. Given the high prevalence of CH amidst an expanding aging population, a better understanding of the mediators of HSC expansion in CH is key to mitigating its associated diseases.
Disclosures: No relevant conflicts of interest to declare.