In the adult brain, multiple cell types are known to produce factors that regulate blood-brain barrier (BBB) properties, including astrocytes. Yet several recent studies disputed a role for mature astrocytes at the BBB. To determine if astrocytes contribute a nonredundant and necessary function in maintaining the adult BBB, we used a mouse model of tamoxifen-inducible astrocyte ablation. In adult mice, tamoxifen induction caused sparse apoptotic astrocyte cell death within 2 hr. Indicative of BBB damage, leakage of the small molecule Cadaverine, and the large plasma protein fibrinogen into the brain parenchyma indicative of BBB damage was detected as early as astrocyte ablation was present. Vessels within and close to regions of astrocyte loss had lower expression of the tight junction protein zonula occludens-1 while endothelial glucose transporter 1 expression was undisturbed. Cadaverine leakage persisted for several weeks suggesting a lack of barrier repair. This is consistent with the finding that ablated astrocytes were not replaced. Adjacent astrocytes responded with partial nonproliferative astrogliosis, characterized by morphological changes and delayed phosphorylation of STAT3, which restricted dye leakage to the brain and vessel surface areas lacking coverage by astrocytes 1 month after ablation. In conclusion, astrocytes are necessary to maintain BBB integrity in the adult brain. BBB-regulating factors secreted by other cell types, such as pericytes, are not sufficient to compensate for astrocyte loss.

224/250 words)In the adult brain, multiple cell types are known to produce factors that regulate blood-brain barrier properties, including astrocytes. Yet several recent studies disputed a role for mature astrocytes at the blood-brain barrier. To determine if astrocytes contribute a non-redundant and necessary function in maintaining the adult blood-brain barrier, we used a mouse model of tamoxifen-inducible astrocyte ablation. In adult mice, tamoxifen induction caused sparse apoptotic astrocyte cell death within 2 hours. Indicative of BBB damage, leakage of the small molecule Cadaverine and the large plasma protein fibrinogen into the brain parenchyma indicative of BBB damage was detected as early as astrocyte ablation was present. Vessels within and close to regions of astrocyte loss had lower expression of the tight junction protein zonula occludens-1 while endothelial glucose transporter 1 expression was undisturbed.Cadaverine leakage persisted for several weeks suggesting a lack of barrier repair. This is consistent with the finding that ablated astrocytes were not replaced. Adjacent astrocytes responded with partial non-proliferative astrogliosis, characterized by morphological changes and delayed phosphorylation of STAT3, which restricted dye leakage to the brain and vessel surface areas lacking coverage by astrocytes one month after ablation. In conclusion, astrocytes are necessary to maintain blood-brain barrier integrity in the adult brain. Blood-brain barrier-regulating factors secreted by other cell types, such as pericytes, are not sufficient to compensate for astrocyte loss. Main Points (250 characters)Mature astrocytes are necessary for maintenance of endothelial tight junctions in the adult brain.Ablated astrocytes are not replaced by proliferation or process extension of neighboring astrocytes resulting in long-term blood-brain barrier damage.

Cover Illustration: A select tracing of GFAP‐positive processes (yellow) of an astrocyte that is adjacent to ablated astrocytes indicated by lack of Glt1 (magenta) in mouse cortex. After conditional ablation of a small subset of astrocytes, neighboring astrocytes increased the length, volume, and complexity of their GFAP+ processes. (See Heithoff, B., et al, https://doi.org/10.1002/glia.23908)