Spatial Transcriptomic Clocks Reveal Cell Proximity Effects in Brain Aging

Academic Background

With aging, cognitive function declines, and the risk of neurodegenerative diseases significantly increases. Brain aging is a complex process accompanied by numerous cellular changes. However, how aging cells affect neighboring cells and how this contributes to tissue decline remains unclear. Additionally, existing tools have yet to systematically address this issue in aging tissues. To this end, researchers have developed a spatially resolved single-cell transcriptomic atlas, combined with machine learning models, to reveal spatial and cell-type-specific transcriptional signatures of aging, rejuvenation, and disease.

Source of the Paper

This paper was co-authored by Eric D. Sun, Olivia Y. Zhou, Max Hauptschein, Nimrod Rappoport, Lucy Xu, Paloma Navarro Negredo, Ling Liu, Thomas A. Rando, James Zou, and Anne Brunet, affiliated with institutions such as Stanford University and the University of California, Los Angeles. The paper was published in Nature in 2024.

Research Process

1. Construction of the Spatial Transcriptomic Atlas

Using multiplexed error-robust fluorescence in situ hybridization (MERFISH) technology, researchers performed single-cell resolution transcriptomic analysis on coronal and sagittal sections of the mouse brain. They selected 300 genes, including cell-type markers, aging-related pathway genes, and other functional genes. The study spanned 20 different age groups of mice, generating spatial transcriptomic data for 4.2 million cells.

2. Development of Spatial Aging Clocks

To quantify the biological age of individual cells, researchers developed a machine learning-based spatial aging clock model. The model, using a spatial smoothing method (SpatialSmooth), preserved spatial information and maximized the performance of single-cell aging clocks. The model performed excellently across 14 cell types, accurately predicting cell age.

3. Analysis of Cell Proximity Effects

Using spatial aging clocks, researchers quantified the influence of specific cell types on the age of neighboring cells. They found that T cells have a significant pro-aging proximity effect, while neural stem cells (NSCs) have a significant pro-rejuvenating proximity effect. Through a deep graph neural network (GNN) model, researchers further validated the existence of these cell proximity effects.

4. Investigation of Potential Mediating Mechanisms

Through gene expression analysis, researchers discovered that T cells mediate their pro-aging effect via the interferon signaling pathway, while NSCs mediate their pro-rejuvenating effect through the secretion of exosomes and growth factors. These findings provide new insights into the molecular mechanisms underlying cell proximity effects.

Key Results

1. Spatial Transcriptomic Atlas: Researchers constructed a spatial transcriptomic atlas of the mouse brain across the entire lifespan, revealing significant changes in different cell types and brain regions during aging.
2. Spatial Aging Clocks: The developed spatial aging clock model accurately predicted cell age and demonstrated good generalization across multiple external datasets.
3. Cell Proximity Effects: T cells have a significant pro-aging effect on neighboring cells, while NSCs have a significant pro-rejuvenating effect. These effects were validated across different brain regions and cell types.
4. Potential Mediating Mechanisms: T cells mediate their pro-aging effect through the interferon signaling pathway, while NSCs mediate their pro-rejuvenating effect through the secretion of exosomes and growth factors.

Conclusion

This study, by constructing a high-resolution spatial transcriptomic atlas and spatial aging clocks, revealed cell-type and brain-region-specific changes during brain aging. The study also identified significant effects of T cells and NSCs on neighboring cells and explored the potential molecular mechanisms underlying these effects. These findings provide new directions for developing interventions targeting aging and disease.

Research Highlights

1. High-Resolution Spatial Transcriptomic Atlas: The study constructed the first spatial transcriptomic atlas of the mouse brain across the entire lifespan, providing unprecedented spatial and single-cell resolution.
2. Spatial Aging Clocks: The developed spatial aging clock model accurately predicted cell age and demonstrated good generalization across multiple external datasets.
3. Cell Proximity Effects: The study found that T cells and NSCs have significant pro-aging and pro-rejuvenating effects on neighboring cells, offering new perspectives on cell-cell interactions.
4. Potential Mediating Mechanisms: The study revealed that T cells and NSCs mediate their proximity effects through different molecular pathways, providing new targets for interventions against aging and disease.

Research Significance

This study not only provides a high-resolution spatial transcriptomic atlas of brain aging but also develops spatial aging clock models, offering powerful tools for studying cell-cell interactions and aging mechanisms. The identified proximity effects of T cells and NSCs and their potential mediating mechanisms provide new directions for developing interventions against aging and disease. Additionally, the machine learning framework developed in this study can be applied to other tissues and species, further advancing research on aging and disease.