Vascular cognitive impairment and dementia (VCID) is the second leading cause of dementia behind Alzheimer’s disease (AD) and is a frequent co-morbidity with AD. Despite its prevalence, little is known about the molecular mechanisms underlying the cognitive dysfunction resulting from cerebrovascular disease. Astrocytic end-feet almost completely surround intraparenchymal blood vessels in the brain and express a variety of channels and markers indicative of their specialized functions in the maintenance of ionic and osmotic homeostasis and gliovascular signaling. These functions are mediated by end-foot enrichment of the aquaporin 4 water channel (AQP4), the inward rectifying potassium channel Kir4.1 and the calcium-dependent potassium channel MaxiK.

Using our HHcy model of VCID we examined the time-course of astrocytic end-foot changes along with cognitive and neuroinflammatory outcomes. We found that there were significant astrocytic end-foot disruptions in the HHcy model. AQP4 becomes dislocalized from the end-feet, there is a loss of Kir4.1 and MaxiK protein expression, as well as a loss of the Dp71 protein known to anchor the Kir4.1, MaxiK and AQP4 channels to the end-foot membrane. Neuroinflammation occurs prior to the astrocytic changes, while cognitive impairment continues to decline with the exacerbation of the astrocytic changes.

We have previously reported similar astrocytic changes in models of cerebral amyloid angiopathy (CAA) and therefore, we believe astrocytic end-foot disruption could represent a common cellular mechanism of VCID and may be a target for therapeutic development.

Document Type


Publication Date


Notes/Citation Information

Published in Neuroscience, v. 341, p. 42-51.

© 2016 Published by Elsevier Ltd on behalf of IBRO.

This manuscript version is made available under the CC‐BY‐NC‐ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/.

The document available for download is the author's post-peer-review final draft of the article.

Digital Object Identifier (DOI)


Funding Information

Research reported in this manuscript was funded by fellowship F31NS092202 (EMW) and grant 1RO1NS079637 (DMW) from the National Institutes of Health.