The vascular network was long viewed as a passive delivery system for oxygen and nutrients. Over the past few years it has become clear that endothelial cells actively maintain homeostasis and regeneration in every organ, and that their transcriptomic programs differ sharply from organ to organ. Studies of aging in individual organs have revealed common themes: stronger inflammatory and interferon responses, weaker angiogenic programs. But the question remained open: how exactly does aging affect the organ-specific functions of the endothelium, and is this tied to a breakdown in tissue resilience? This work sits at the intersection of two fields: research on endothelial heterogeneity and molecular aging. Unlike neighboring studies that focused on a single organ or a single cell type, this one is a systematic multi-organ analysis at the level of individual cells.
The researchers isolated endothelial cells (CD31⁺) from the lungs, kidneys, liver, heart and brain of mice aged 3 to 26 months and performed single-cell RNA sequencing (scRNA-seq) on the 10x Genomics platform. After quality control and data integration with Harmony, 258,574 cells entered the analysis. For the lungs and brain, intermediate age points (6, 12, 15, 17 months) were included to track the dynamics of change. Clustering revealed canonical vascular subtypes (arterial, capillary, venous) and organ-specific subpopulations. To assess age-related changes, the team used differential gene expression, functional category analysis (Gene Ontology), and pseudotime reconstruction with tradeSeq. They also ran functional tests: measuring lung vascular permeability after lipopolysaccharide (LPS) administration, assessing the uptake activity of liver endothelium, immunohistochemistry, and RNA in situ hybridization.
Aging progressively reduced the expression of organ-specific genes in every organ except the brain, while vascular subtypes largely retained their identity. In the lungs, alveolar capillary endothelial cells lost genes responsible for barrier function (Rap1 signaling, cell-cell junctions) and at the same time ramped up antigen-presentation genes (MHC-I/II) and interferon response genes. With age, Evans blue dye leakage increased after LPS-induced lung injury, pointing to weakened barrier resilience. In the liver, sinusoidal endothelial cells lost genes tied to uptake and clearance and acquired markers of capillarization (CD34) and fibrosis. In the heart, angiogenic and proliferative programs declined, correlating with a drop in the number of Ki67⁺ endothelial cells. In the brain, the picture was complex: alongside degradation of barrier genes, compensatory mechanisms were activated. Based on the expression of interferon-induced genes, the team built a cellular aging index that revealed heterogeneity even among cells of the same age. The functional decline of the endothelium began before a fully senescent state appeared: in the lungs, only about 3% of cells reached a senescence-like state even in old mice.
The work shows that endothelial aging is not just an accumulation of damage but a systemic loss of organ specialization, and that it shows up earlier than the classic markers of cellular senescence. Unlike previous studies that focused on individual organs or shared inflammatory signatures, this is the first multi-organ analysis at the single-cell level with functional validation. That makes it possible to separate universal mechanisms of aging from organ-specific consequences. What is more, the study found heterogeneity in aging among cells of the same chronological age, which calls into question population-level approaches to estimating biological age.
Future studies will likely aim to pin down the causal link between interferon signaling and the loss of organ programs. It is already clear that neutralizing IFNγ in old mice reduces MHC-II expression on the endothelium but does not affect MHC-I. Whether the interferon response is a driver or a consequence of endothelial dysfunction remains an open question. A promising direction could be the search for therapeutic windows to target pre-senescent endothelial states, which, judging by the data, are more reversible than fully senescent cells.
The results bear directly on age-related diseases of the vessels, lungs, liver, heart and brain. Understanding how the endothelium loses its specialization could help in developing strategies to maintain tissue resilience and prevent age-related organ dysfunction.
The next step could be to test the causal role of interferon signaling in the loss of organ programs using genetic models and pharmacological inhibitors. It will also be important to study how reversible pre-senescent endothelial states are and whether organ specialization can be restored by targeting the microenvironment.
In the biology of aging, the question of which molecular events are primary and which are compensatory remains unresolved. This work links the loss of endothelial organ specialization to a breakdown in tissue resilience, but it does not establish causation. This echoes the broader problem of aging heterogeneity: cells of the same age can be at different stages of functional decline, which demands a rethink of the very concept of biological age.
In old mice, after interferon-gamma was neutralized, MHC-II levels on the endothelium of the lungs and kidneys dropped, but MHC-I stayed unchanged. This suggests that different components of antigen presentation are regulated differently. What is more, in germ-free old mice the age-related induction of MHC genes persisted, meaning the microbiota is not an obligatory trigger of this process.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.