With Age, Macrophages Chew up the Blood-CSF Barrier

Protective borders between the brain and the blood weaken with age. This includes the barrier formed by tightly linked epithelial cells in the choroid plexus, which generates cerebrospinal fluid. In the February 26 Neuron online, scientists led by Yidong Shen at the Chinese Academy of Sciences in Shanghai put forward an explanation for the decay of this blood-CSF barrier. In aging mice, senescent cells in the choroid plexus recruited circulating macrophages. These immune cells released a protease, cathepsin S, that chewed up claudin 1, part of the glue that seals adjoining epithelial cells, Shen and colleagues report. As these tight junctions deteriorated, more blood proteins and T cells infiltrated the CSF.

  • As mice aged, macrophages in the choroid plexus secreted more cathepsin S, breaking down the blood-CSF barrier.
  • Inhibiting cathepsin S protected the barrier and preserved memory.
  • Senescent cells triggered the age-related rise in cathepsin S.

By inhibiting cathepsin S, the authors claimed they could prevent breakdown of the barrier. This intervention lowered neuroinflammation, promoted neurogenesis, and maintained memory in aging animals, suggesting broad benefits to the brain. “Our findings … identify cathepsin S as a potential target to improve brain homeostasis in aged animals,” the authors noted.

The findings come hot on the heels of another paper that reported the blood-brain barrier breaks down with age as the endothelial cells lining small blood vessels in the brain lose a protective coat of glycoproteins (Feb 2025 news). 

Loose Junctions? In 2-month-old mice (left), claudin 1 (CLDN1) is abundant and epithelial cell tight junctions are sealed. In 18-month-old mice (right), senescent cells in the choroid plexus trigger macrophages to release cathepsin S (CTSS), which chews up claudin 1 and creates leaks. [Courtesy of Chen et al., Neuron, 2025.]

The blood-CSF barrier has been far less studied than the blood-brain barrier. Scientists have detected age-related changes in gene expression in the choroid plexus, and have linked these to increased inflammation (Aug 2014 news; Dani et al., 2021). Nonetheless, it remains unclear how age affects this barrier’s function.

To study this, joint first authors Yifan Chen, Yifei Zhou, and Yaqing Bai compared the abundance of tight junction proteins in the choroid plexus of wild-type mice at 2, 12, 18, and 24 months old. Claudin 1 was the only protein that changed, steadily decreasing with age (image below). This transmembrane protein protrudes from epithelial cells, zipping up with claudin 1 on neighboring cells to form a tight seal.

Because expression of claudin 1 did not change with age, the authors explored whether proteases might be responsible for its loss. They found that macrophages in the choroid plexus secreted cathepsin S, which rose 10-fold from 2 to 24 months of age. This rise correlated with the fall in claudin 1. The authors confirmed that the protease chopped up the endothelial membrane protein both in vitro and in ex vivo choroid plexus tissue.

Culprit: Cathepsin. In 2-month-old mice (left), claudin 1 (green) abounds at tight junctions (arrowheads) in the choroid plexus, but by 18 months (middle), it is scarce. Inhibiting cathepsin S for three months preserved claudin 1 in 18-month-old mice (right). Nuclei are blue. [Courtesy of Chen et al., Neuron, 2025.]

Could stopping cathepsin S protect tight junctions? Indeed, when the authors fed mice the cathepsin S inhibitor aloxistatin from 15 to 18 months of age, claudin 1 protein was as abundant as in 2-month-old mice (image above). The blood-CSF barrier stayed intact, keeping out blood proteins and T cells. Neuroinflammation in neighboring cortical regions, as judged by TNF-α staining, was half that of age-matched controls.

Previous work had shown that the choroid plexus regulates adult neurogenesis in the nearby subventricular zone (Silva-Vargas et al., 2016). Inhibiting cathepsin S with aloxistatin also boosted neurogenesis in 18-month-old mice to twice the level of that in controls. Treated mice solved the Barnes maze twice as fast as did age-matched controls, suggesting their memory was sharper.

The authors found the same benefits when they infused a different cathepsin S inhibitor, LY3000328, into the aging mouse brain, and when they expressed a third inhibitor, cystatin C, in the choroid plexus via adenovirus. Conversely, overexpressing cathepsin S via adenovirus in 2-month-old mice caused the blood-CSF barrier to break down and memory to falter by 5 months.

What revs up cathepsin S in older mice? The authors suspected senescent cells, because these increase with age and are known to exacerbate inflammation (Freund et al., 2010). Indeed, when Chen and colleagues analyzed cells in the choroid plexus using single-cell RNA-Seq, they found greater senescence in numerous cell types, including epithelial, endothelial, and smooth muscle cells, as well as fibroblasts and pericytes. In cell culture, senescent cells recruited macrophages and triggered them to secrete cathepsin S, they report. In line with this, inducing senescence in the choroid plexus of young mice weakened the blood-CSF barrier.

Would eliminating senescent cells help? Treating year-old mice with senolytics for three months kept the blood-CSF barrier strong, and boosted neurogenesis and memory. However, the effects were not as robust as when the scientists directly inhibited cathepsin S. “Targeting key downstream factors of senescent cells could provide a more precise and potent strategy to counteract aging,” the authors suggested.

The findings could apply to neurodegenerative diseases. Previous work has reported numerous structural and gene expression changes in the choroid plexus in Alzheimer’s disease, in part due to the effects of Aβ (Sep 2015 news; for review see Solár et al., 2020). A trial of senolytics is ongoing for AD.—Madolyn Bowman Rogers

Comments

No Available Comments

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Do Specialized Glycoproteins Prop Up Blood-Brain Barrier?
  2. Choroid Plexus May Hold a Key To Aging Brain
  3. Barriers Between Blood and CSF, Brain Yield to Aβ—Not a Bad Thing?

Therapeutics Citations

  1. Dasatinib + Quercetin

Paper Citations

  1. Dani N, Herbst RH, McCabe C, Green GS, Kaiser K, Head JP, Cui J, Shipley FB, Jang A, Dionne D, Nguyen L, Rodman C, Riesenfeld SJ, Prochazka J, Prochazkova M, Sedlacek R, Zhang F, Bryja V, Rozenblatt-Rosen O, Habib N, Regev A, Lehtinen MK. A cellular and spatial map of the choroid plexus across brain ventricles and ages. Cell. 2021 Apr 27; PubMed.
  2. Silva-Vargas V, Maldonado-Soto AR, Mizrak D, Codega P, Doetsch F. Age-Dependent Niche Signals from the Choroid Plexus Regulate Adult Neural Stem Cells. Cell Stem Cell. 2016 Nov 3;19(5):643-652. Epub 2016 Jul 21 PubMed.
  3. Freund A, Orjalo AV, Desprez PY, Campisi J. Inflammatory networks during cellular senescence: causes and consequences. Trends Mol Med. 2010 May;16(5):238-46. Epub 2010 May 3 PubMed.
  4. Solár P, Zamani A, Kubíčková L, Dubový P, Joukal M. Choroid plexus and the blood-cerebrospinal fluid barrier in disease. Fluids Barriers CNS. 2020 May 6;17(1):35. PubMed.

External Citations

  1. aloxistatin 

Further Reading

Primary Papers

  1. Chen Y, Zhou Y, Bai Y, Jia K, Zhang H, Chen Q, Song M, Dai Y, Shi J, Chen Z, Yan X, Shen Y. Macrophage-derived CTSS drives the age-dependent disruption of the blood-CSF barrier. Neuron. 2025 Feb 19; Epub 2025 Feb 19 PubMed.

Annotate

To make an annotation you must Login or Register.