Related Experiment Video
Updated: Aug 10, 2026

Immunostaining for DNA Modifications: Computational Analysis of Confocal Images
Published on: September 7, 2017
Peripheral blood DNA methylation partially reports on immune cell type composition in the frontal brain
Mandy Meijer1,2,3,4, Maggie Po Yuan Fu1,2,3,4, Erick Isaac Navarro-Delgado1,2,3,4
1Department of Medical Genetics, Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada.
None:
Most existing DNA methylation (DNAm) studies have used peripheral surrogate tissues to research molecular mechanisms underlying brain disorders and diseases. Initial studies comparing brain to blood primarily at the individual CpG level analysis have generally pointed to a limited overlap of epigenetic patterns, consistent with DNAm being largely tissue- and even cell type-specific. Expanding on these studies, we employed a more complex analysis strategy aimed to 1) identify single DNAm sites associated with deconvolution-estimated brain cell type proportions, the principal measure in this study, in both the frontal brain and peripheral blood, 2) combine blood DNAm sites to predict brain cell type proportions through multivariate models, and 3) examine the association of blood DNAm, age, and epigenetic age acceleration (EAA) on brain cell type proportions. Epigenome-wide association studies for seven brain cell type proportions in matched frontal brain and peripheral blood samples (n = 104) revealed that ∼10% of brain cell type-associated DNAm sites had correlating DNAm levels in peripheral blood (p < 0.05). However, only three peripheral blood DNAm sites were significantly associated with endothelial and stromal brain cell type proportions (adjusted p < 0.05). Brain cell type proportion predictions trained with machine learning approaches using peripheral blood DNAm showed the strongest, although still modest correlations with microglia proportions estimated through cell deconvolution using brain DNAm. Further, deconvolution-estimated blood immune cell type proportions showed a nominally significant association with estimated brain stromal cell proportions, driven primarily by NK cells; however, this association was dependent on chronological age and did not survive age-residualization. Lastly, brain EAA was not associated with brain and blood cell type proportions (adjusted p < 0.05). Collectively, these results suggested that in the context of broad tissue-specificity of DNAm patterns, DNAm levels in peripheral blood might actually inform on some immune brain cell type proportions. The correlations between DNAm profiles specific to immune cell types in blood and brain were consistent with a potential link between peripheral immune and central nervous system immune functions.

