ABSTRACT
Vascular dementia (VaD) is the second leading cause of dementia behind Alzheimer’s disease, but its
pathogenesis is poorly understood. Cerebral blood vessels undergo age-related degenerative changes resulting
in blood-brain barrier (BBB) disruption and hypoperfusion, leading to neurodegeneration and cognitive decline.
Pericytes (PC) are key components of the BBB whose behavior is regulated by extracellular matrix components
such as the heparan sulfate proteoglycan, perlecan. We hypothesize that age-related overexpression of
cerebrovascular perlecan, coupled with reduced proteolytic activity of the cysteine protease cathepsin
B, which processes perlecan into a neuroprotective laminin globular domain 3 (LG3) peptide, plays a
central pathological role in VaD, by reducing PC coverage of blood vessels, leading to diminished BBB
integrity. First, cerebral blood vessels in aged mice show increased perlecan expression that correlates with
BBB disruption and reduced PC coverage. Second, young mice exposed to chronic mild hypoxia (CMH; 8% O2
for 4 days) display similar increased perlecan expression that correlates closely with reduced PC coverage and
BBB disruption. Third, perlecan inhibits PC proliferation and migration in vitro, in keeping with similar effects
described for smooth muscle cells, in which inhibitory effects are lost with removal of heparan sulfate (HS) side
chains. Fourth, cerebrovascular perlecan expression chronically increased in the brain after ischemic stroke.
Fifth, perlecan is increased in conditions with abnormal brain vasculature including arteriovenous malformation.
Sixth, the endothelial nitric oxide synthase (eNOS) deficient mouse model of VaD display reduced PC coverage
and BBB damage; our preliminary data show these mice have increased brain perlecan and reduced cathepsin
B, with progression to VaD. Seventh, our preliminary data shows cathepsin B KO mice express significantly more
brain perlecan and less PC coverage, supporting the role of cathepsin B in perlecan processing and perlecan’s
role in PC coverage. Therefore, we hypothesize that the fully intact perlecan molecule and its LG3 fragment, as
regulated by cathepsin B proteolysis, have opposing deleterious and beneficial effects, respectively, on vascular
health and integrity in cerebrovascular diseases such as VaD. Specifically, we hypothesize that (i) in response
to aging +/- hypoxia or diminished eNOS, upregulation of cerebrovascular perlecan in association with
decreased cathepsin B activity, and thereby less LG3, leads to reduced PC coverage and BBB disruption,
culminating in neurodegeneration and cognitive decline, and (ii) perlecan LG3 ‘replacement therapy’
may prevent or mitigate this process. To test this hypothesis we propose to: 1, Determine how age and sex
influence levels of full-length perlecan (and its HS modification), the LG3 fragment, and cathepsin B in the eNOS-
deficient and CMH mouse models of spontaneous VaD, 2, Define the molecular basis of the positive (integrin
binding sites) and negative (HS side chains) effects of perlecan and LG3 on PC behavior, and 3, Determine the
therapeutic potential of LG3 in the eNOS-deficient and CMH mouse models of spontaneous VaD.
