https://www.cell.com/cell/fulltext/S0092-8674(26)00455-1
chatGPT:
Summary
This is a Cell review article, not a primary experimental paper. Its central proposal is that “mesenchymal drift” (MD) can act as a unifying framework for understanding aging.
The authors define mesenchymal drift as a gradual process in which cells lose their original lineage identity and acquire or intensify mesenchymal-like features. At tissue level, this means loss of functional parenchymal cells, expansion or activation of mesenchymal/stromal programs, extracellular matrix remodeling, fibrosis, inflammation, and organ dysfunction. They present this not as a binary switch, but as a spectrum of hybrid states, similar to partial EMT rather than full conversion.
The article argues that MD may connect the traditional hallmarks of aging. Instead of viewing genomic instability, epigenetic alterations, mitochondrial dysfunction, senescence, inflammation, stem-cell exhaustion, dysbiosis, and altered communication as separate processes, the authors suggest that many of them converge on — and are reinforced by — cellular identity erosion toward mesenchymal/fibrotic states.
The paper’s core model is a feedback-loop model:
aging damage and stress induce mesenchymal drift; mesenchymal drift then worsens genomic instability, inflammation, senescence, mitochondrial dysfunction, ECM stiffening, stem-cell exhaustion, and barrier breakdown.
For example, the authors describe genomic instability as both a driver and consequence of MD: DNA damage can promote EMT/EndoMT/FMT, while mesenchymal programs can increase replication stress, chromosomal instability, and defective DNA repair.
They also emphasize that MD spans multiple cell types and transitions: EMT, endothelial-to-mesenchymal transition, fibroblast-to-myofibroblast transition, pericyte-to-myofibroblast transition, macrophage-to-myofibroblast transition, hepatic stellate cell activation, vascular smooth muscle cell plasticity, and related stromal or inflammatory conversions. The figure on page 3 illustrates this as diverse cell origins converging on a mesenchymal spectrum under cues such as TGF-β, WNT, NOTCH, hypoxia, and inflammatory cytokines.
A major therapeutic implication is that partial reprogramming might counteract MD. The authors argue that OSKM/Yamanaka-factor or chemical reprogramming can initiate mesenchymal-to-epithelial transition-like reversal, repress mesenchymal transcription factors such as SNAIL, TWIST, and ZEB, restore youthful epigenetic patterns, and potentially reset multiple aging hallmarks at once.
They cite evidence that partial reprogramming can improve DNA repair, restore H3K9me3, reduce DNA methylation age, improve mitochondrial function, activate autophagy, reduce senescence/SASP markers, remodel ECM, and improve regeneration in several tissues.
What is novel?
The novelty is conceptual rather than experimental.
The main new idea is to elevate mesenchymal drift from a disease-associated phenomenon — familiar from EMT, fibrosis, cancer invasion, endothelial dysfunction, and wound healing — into a proposed central integrator of aging biology.
The paper’s novelty can be broken down into four points:
-
A unifying framework: It reframes aging as not just accumulation of molecular damage, but as progressive destabilization of cell identity toward mesenchymal/fibrotic states.
-
A bridge between hallmarks: It tries to explain how many hallmarks interact rather than simply listing them. MD becomes a “hub” linking genomic instability, epigenetic drift, mitochondrial dysfunction, senescence, inflammation, ECM remodeling, and stem-cell exhaustion.
-
A directional model of tissue aging: The authors suggest aging has a biased trajectory: tissues drift toward fibrosis, ECM stiffening, inflammatory signaling, and loss of differentiated function.
-
A therapeutic axis: The paper positions partial reprogramming as a possible inverse process to MD — not merely “making cells younger,” but specifically restoring lineage identity and reversing mesenchymal-state erosion.
Critique
The paper is strong as a synthesis, but weaker as a proof of causality.
Its biggest strength is that it brings together a large body of evidence from EMT, fibrosis, inflammation, senescence, epigenetics, mitochondrial dysfunction, and partial reprogramming into a coherent biological story. This is useful because aging tissues often do show fibrosis, stromal activation, ECM stiffening, inflammatory signaling, and loss of specialised cell function.
However, the central claim risks becoming too broad. If mesenchymal drift includes EMT, EndoMT, FMT, PMT, macrophage transitions, stellate-cell activation, fibroblast activation, vascular smooth-muscle plasticity, and general ECM remodeling, then MD may become a label for many different processes rather than a sharply defined mechanism. The paper acknowledges MD as a spectrum, but that makes it harder to know exactly what counts as MD and what does not.
A second weakness is causal hierarchy. The authors argue that MD can both arise from and reinforce hallmarks, but this does not prove that MD is upstream of aging rather than downstream of damage. In many cases, inflammation, DNA damage, mitochondrial dysfunction, senescence, and tissue injury could be the primary drivers, with MD being a common repair/fibrosis response. The authors themselves state that more longitudinal and experimental work is needed to decide whether MD is a primary driver or an amplifying node.
A third issue is measurement. For MD to become a true aging hallmark or therapeutic target, it needs robust biomarkers: single-cell signatures, spatial markers, lineage-tracing evidence, ECM-state measures, or blood-based correlates. The review discusses this need, but does not yet provide a standardized MD score that can be used across tissues and species.
A fourth concern is therapeutic ambiguity. Mesenchymal plasticity is not always bad. EMT-like and fibroblast activation programs are needed for wound healing, development, repair, and regeneration. Blocking MD too broadly could impair repair or immune containment. The authors recognize this problem, noting that therapies must distinguish adaptive transient plasticity from chronic pathological drift.
Finally, the partial reprogramming section is plausible but somewhat optimistic. Partial reprogramming may indeed reverse some aging-associated features, but it also raises unresolved issues: tumor risk, loss of cell identity if overdosed, delivery, tissue specificity, dosing cycles, and whether observed rejuvenation reflects true aging reversal rather than stress adaptation or selective survival of healthier cells.
Bottom line
This is a useful and ambitious review. Its main contribution is to propose mesenchymal drift as a systems-level aging mechanism linking cell identity loss, fibrosis, inflammation, ECM remodeling, and multiple hallmarks of aging.
The idea is compelling because fibrosis and stromal activation are widespread in aging. But the framework still needs sharper definitions, better biomarkers, longitudinal validation, and experiments showing that reducing MD extends healthspan independently of merely suppressing injury, inflammation, or fibrosis.