As people and animals age, hematopoietic stem cells (HSCs) gradually lose their ability to maintain a balanced blood system. Although their numbers may increase, their regenerative capacity declines, potentially contributing to anemia, weakened immunity, blood clots, and other age-related blood disorders. Researchers at Tohoku University used artificial intelligence to identify genes that may drive these age-related changes and then validated their findings through animal experiments.

The team analyzed individual blood stem cells from mice at different stages of life and discovered that aging activates two distinct genetic programs. One maintains the cells in an immature stem-cell state, while the other favors platelet production. These changes developed progressively over time, suggesting that stem-cell aging is a continuous process rather than something that suddenly occurs in old age.

Using Geneformer, an AI model trained on gene-expression data from around 30 million cells, researchers analyzed approximately 160,000 young and aged blood stem and progenitor cells. The model identified 143 potential genes involved in shifting young stem cells toward an aged state.

Laboratory screening ultimately highlighted Pbx1, a gene-regulating factor, as a key driver. Increasing Pbx1 in young stem cells reproduced several characteristics of aged cells, including reduced red blood cell production and increased platelet formation. Researchers suggest that Pbx1 may contribute to this imbalance by suppressing Gata1, a gene important for red blood cell development.

Rather than simply becoming weaker, aged blood stem cells appear to enter a distinct and stable cellular state with altered patterns of activity. By combining AI prediction, large-scale screening, multi-omics analysis, and transplantation experiments, the study provides new insight into the molecular mechanisms underlying blood stem-cell aging.

Future research will determine whether this pathway also occurs in humans and whether it contributes to anemia, thrombosis, clonal hematopoiesis, and blood cancers.

Source: https://www.tohoku.ac.jp/en/press/ai_helps_identify_regulatory_hub_for_aging_in_hematopoietic_stem_cells.html