https://www.cell.com/cell/fulltext/S0092-8674(26)00581-7
Obviously this fits with my hypothesis that some senescent cells are cells in the process of differentiation and getting established as somatic cells
chatGPT:
Summary
This paper argues that cellular senescence is not merely a damage or aging programme, but can be a normal developmental programme required for forming the brain’s barrier systems. The authors study mouse brain development and identify p21-positive cells in three major brain-barrier-associated lineages: choroid plexus epithelial cells, vascular endothelial cells, and brain-resident macrophages. They show that these cells display multiple senescence-associated features, including cell-cycle exit, senescence-like transcriptional signatures, lysosomal dysfunction markers, and secretory programmes.
The central finding is that these senescent states are cell-type specific and temporally different. Endothelial cells and macrophages enter a transient, inflammatory, pro-remodelling senescent-like state during vascularisation and BBB formation. By contrast, choroid plexus epithelial cells enter a persistent, non-inflammatory senescent state that appears to support CSF production, epithelial polarity, barrier integrity, and long-term choroid plexus function.
Mechanistically, the authors suggest a division of labour. p21-positive endothelial cells are linked to angiogenic tip-cell behaviour and vessel maturation, while p21-positive macrophages provide local patterning and matrix-remodelling cues. Selective ablation of p21-positive macrophages reduced vessel density and branching, whereas broader ablation of all p21-positive cells caused more severe defects, including abnormal vascular patterning, haemorrhage, ventricular collapse, and impaired CSF regulation.
For the choroid plexus, ablation of p21-positive cells reduced choroid plexus size, ventricular fluid, Aqp1 expression, endothelial supply, collagen IV, plexin B2, integrin β1, and altered ZO-1/F-actin organisation. This supports the claim that these cells contribute to blood-CSF barrier formation and CSF homeostasis.
Novelty
The main novelty is the claim that persistent senescence can be physiological rather than pathological. The paper challenges the usual framing in which developmental senescence is transient and beneficial, while persistent senescence is harmful. Here, the choroid plexus epithelial cells appear to maintain a long-lived, non-inflammatory senescent state that supports a stable epithelial barrier.
A second novel point is the distinction between different flavours of senescence within the same developing organ. The authors do not treat senescence as a single binary state. Instead, they propose a modular programme: inflammatory, transient senescence in vascular/macrophage populations, and non-inflammatory, persistent senescence in choroid plexus epithelium.
A third important contribution is the connection between senescence and brain-barrier development, especially the BBB and blood-CSF barrier. Previous developmental senescence work has often focused on tissue remodelling in structures such as limb, kidney, placenta, or endolymphatic sac. This paper extends that idea to neurodevelopmental barrier formation.
Critique
The paper is strong in combining single-cell transcriptomics, histology, lineage tracing, and ablation experiments. It also appropriately recognises that p21 alone is not a definitive senescence marker, so it uses a multidimensional definition including transcriptional signatures, Ki67 loss, SA-β-gal, lipofuscin, and SASP-like gene expression.
However, there are several limitations.
First, the causal interpretation is partly complicated by the use of p21-positive cell ablation. p21 marks senescent cells, but also marks other cell states, including differentiation, stress response, and cell-cycle regulation. The paper itself notes that p21-positive Cajal-Retzius cells had some senescence-like transcriptional features but lacked key senescence hallmarks, suggesting p21 can identify non-senescent developmental states too. This makes pan-p21 ablation potentially broad and difficult to interpret cleanly.
Second, the ablation phenotype is severe and systemic. Widespread p21-positive cell ablation caused embryonic lethality, growth defects, craniofacial abnormalities, swelling, haemorrhage, and fluid accumulation. That makes it hard to separate direct brain-barrier effects from general developmental disruption, vascular compromise, or systemic oedema. The authors acknowledge that extracranial fluid accumulation likely reflects both impaired CSF regulation and systemic effects of widespread p21-positive cell ablation.
Third, “senescence” may be somewhat stretched conceptually. The endothelial tip-cell and macrophage states may be better described as senescence-like developmental programmes rather than fully senescent terminal states, especially because they are transient and integrated into angiogenesis. This is not necessarily a flaw, but it means the terminology needs care: these are not equivalent to aged, damaged, chronically pro-inflammatory senescent cells.
Fourth, the persistent choroid plexus senescence claim is intriguing but needs more functional ageing data. Showing SA-β-gal and p21 in 12-month-old mouse choroid plexus supports persistence, but does not fully prove that the state remains beneficial across ageing rather than eventually becoming maladaptive. A key follow-up would be to test whether selectively disrupting this programme in adult choroid plexus impairs CSF production, barrier integrity, waste clearance, or brain ageing phenotypes.
Fifth, the study is in mice. The choroid plexus and BBB are conserved, but the relevance to human brain development, neurodevelopmental disorders, ageing, and human choroid plexus pathology remains to be shown.
Bottom line
This is a conceptually important paper. Its strongest contribution is to reframe senescence as a context-dependent developmental tool, not merely a damage programme. The most persuasive part is the distinction between transient inflammatory senescence in vascular/macrophage lineages and persistent non-inflammatory senescence in choroid plexus epithelium. The main caution is that p21-based ablation and p21-based lineage tracing do not perfectly isolate senescent cells, so some phenotypes may reflect loss of broader p21-positive developmental populations rather than senescence per se.