Intrinsic molecular programs and extrinsic factors including proinflammatory molecules are understood to regulate hematopoietic aging. This is based on foundational studies using genetic perturbation to evaluate causality. However, individual organisms exhibit natural variation in the hematopoietic aging phenotypes and the molecular basis of this heterogeneity is poorly understood. Here, we generated individual single-cell transcriptomic profiles of hematopoietic and nonhematopoietic cell types in 5 young adult and 9 middle-aged C57BL/6J female mice, providing a web-accessible transcriptomic resource for the field. Among all assessed cell types, hematopoietic stem cells (HSCs) exhibited the greatest phenotypic variation in expansion among individual middle-aged mice. We computationally pooled samples to define modules representing the molecular signatures of middle-aged HSCs and interrogated, which extrinsic regulatory cell types and factors would predict the variance in these signatures between individual middle-aged mice. Decline in signaling mediated by adiponectin, kit ligand (KITL) and insulin-like growth factor 1 (IGF1) from mesenchymal stromal cells (MSCs) was predicted to have the greatest transcriptional impact on middle-aged HSCs, as opposed to signaling mediated by endothelial cells or mature hematopoietic cell types. In individual middle-aged mice, lower expression of Kitl and Igf1 in MSCs was highly correlated with reduced lymphoid lineage commitment of HSCs and increased signatures of differentiation-inactive HSCs. These signatures were independent of expression of aging-associated proinflammatory cytokines including interleukin-1β (IL-1β), IL-6, tumor necrosis factor α and RANTES. In sum, we find that Kitl and Igf1 expression are coregulated and variable between individual mice at the middle age and expression of these factors is predictive of HSC activation and lymphoid commitment independently of inflammation.
© 2024 American Society of Hematology. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.