Premium
Single nucleus and bulk homogenate RNA‐sequencing comparison of vascular endothelial growth factor family associations with Alzheimer's disease
Author(s) -
Moore Annah M.,
Mahoney Emily R.,
Dumitrescu Logan,
De Jager Philip L.,
Koran Mary Ellen I.,
Petyuk Vladislav A.,
Robinson Renã A.S.,
Ruderfer Douglas M.,
Cox Nancy J.,
Schneider Julie A.,
Bennett David A.,
Jefferson Angela L.,
Hohman Timothy J.
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.046170
Subject(s) - biology , cognitive decline , neuroscience , gene expression , gene , disease , medicine , genetics , dementia
Background Previous work from our group and others has suggested that the VEGF family of genes and proteins may be involved in the pathophysiology of Alzheimer’s disease (AD) and associated cognitive decline. Using dorsolateral prefrontal cortex (DLPFC) RNA sequencing (RNA‐seq) data (N = 531), we have characterized robust associations between gene expression of four family members (VEGFB, PGF, FLT1 and FLT4) and AD diagnosis, neuropathology, and associated cognitive decline. More specifically, increased brain expression of these genes was associated with worse outcomes. Single nucleus RNA sequencing (snRNA‐seq) of DLPFC became available for a small subset of this cohort (N = 48). We set out to assess associations between gene expression of these family members and cognitive decline using snRNA‐seq and compare to homogenate RNA‐seq findings to better understand the cell specificity of observed effects. Method Longitudinal cognitive and snRNA‐seq data from astrocytes, endothelial cells, microglia, oligodendrocytes, oligodendrocyte progenitor cells, pericytes, excitatory and inhibitory neurons were obtained from the Religious Orders Study and Rush Memory and Aging Project cohort studies. Linear regression and mixed‐effects regression were used to assess VEGF gene expression associations with global cognition and cognitive change, respectively. Models covaried for age at death, sex and postmortem interval and were subjected to the Benjamini‐Hochberg procedure to correct for multiple comparisons within cell types. Result PGF and FLT4 expression across cell types was too low to be considered in statistical models. FLT1 expression was largely restricted to endothelial cells. while VEGFB expression was highest in neurons and astrocytes. Interestingly, inhibitory neuron VEGFB expression approached a positive association with cross‐sectional cognition (p.fdr = 0.055), in opposition to the direction of our bulk homogenate finding. In contrast, oligodendrocyte VEGFB was negatively associated with cross‐sectional (p.fdr = 0.039) and longitudinal (p.fdr = 0.053) cognition, consistent with our previous findings in bulk. Conclusion Single nucleus VEGF expression analyses suggest that homogenate VEGFB associations with AD outcomes are driven by oligodendrocytes. VEGFB is a neuroprotective factor, and the observed upregulation in VEGFB could represent a stress response to AD pathology. The highly restricted expression of FLT1 to endothelial cells suggests that the relationship between FLT1 and cognitive trajectory reflects alterations in endothelial signaling.