Post-mortem brain studies provide important molecular information for understanding genetic- epigenetic- and expression-based determinants of human brain development and dysregulation in neuropsychiatric illness and other brain disorders. While the sample sizes of postmortem studies have increased dramatically over the past decade, the majority of large-scale human studies use homogenate brain tissue, highlighting the difficulty in obtaining specific cell types of interest in a large nmber of samples. However, we have observed large variability in the proportion of neuronal cells in over 1000 post-mortem human brain tissue homogenates, potentially due to dissection issues, effects of disease, subject age and/or individual/random variation. Failure to incorporate cellular composition, e.g. the relative proportions of functional cell types, into both epigenetic and gene expression studies in homogenate brain tissue can result in both widespread false positives and negatives. This project involves isolating four pure cellular populations (inhibitory and excitatoy neurons, oligodendrocytes, and astrocytes) across four brain regions (cerebellum, DLPFC, hippocampus and caudate nucleus) in five individuals. We will generate genome-scale DNA methylation maps in each cell type, region, and individual using whole genome bisulfite sequencing (WGBS). These data will better characterize epigenetic differences between related cell types like GABA and glutamatergic interneurons to identify what regions are differentially regulated between related classes of cells with similar functions. Perhaps more importantly, we will use these pure DNA methylation profiles to create open-source statistical software that performs in silico estimation of the relative proportion of each cell type in brain tissue homogenate provided by other researchers. This approach has been successfully applied to statistically deconvoluting whole blood into the relative proportion of blood cell types. We propose to greatly expand previous work in the brain by quantifying more functionally enriched cell types across more brain regions using fewer discriminatory markers. We hope to create an accurate assay for inexpensively estimating cellular composition of brain tissue homogenate using a series of bisulfite sequencing experiments. The easy estimation of cellular composition would have widespread use in postmortem brain research, both among epigenetic and gene expression studies. For example, during RNA extraction steps, a small amount of DNA can be concurrently extracted and used to estimate cellular composition proportions. The sample selection for the studies can utilize this composition information by excluding samples with outlying composition estimates, and sample selection for biological groups can be matched on cell composition proportions, such as ensuring a similar distribution of composition across outcome groups. This is especially important because differences in cell composition between outcome groups result in widespread false positives. We hope to introduce a powerful tool to complement important factors like tissue quality and sample characterization in the analysis of postmortem human brain data.

Public Health Relevance

Post-mortem human brain tissue provides an important tool for the study of mental illness and brain disorders. However, many studies utilize homogenate brain tissue, composed of varying proportions of cell types like neurons and glia - failure to account for this cellular composition may result in widespread false positives and negatives. This project provides resources for the generation of genome-scale epigenetic maps in individual brain cell populations, estimating the relative proportion of these cell types in brain tissue homogenate to maximize its utility and, exploring the epigenetic differences between related cell types like excitatory and inhibitory neurons.

Agency
National Institute of Health (NIH)
Institute
National Institute of Mental Health (NIMH)
Type
Exploratory/Developmental Grants (R21)
Project #
1R21MH105853-01A1
Application #
8970099
Study Section
Special Emphasis Panel (ZRG1-MDCN-P (57))
Program Officer
Senthil, Geetha
Project Start
2015-09-03
Project End
2017-08-31
Budget Start
2015-09-03
Budget End
2016-08-31
Support Year
1
Fiscal Year
2015
Total Cost
$370,000
Indirect Cost
$170,000
Name
Lieber Institute, Inc.
Department
Type
DUNS #
963044529
City
Baltimore
State
MD
Country
United States
Zip Code
21205
Zhu, Ying; Sousa, André M M; Gao, Tianliuyun et al. (2018) Spatiotemporal transcriptomic divergence across human and macaque brain development. Science 362:
Amiri, Anahita; Coppola, Gianfilippo; Scuderi, Soraya et al. (2018) Transcriptome and epigenome landscape of human cortical development modeled in organoids. Science 362:
Rhie, Suhn K; Schreiner, Shannon; Witt, Heather et al. (2018) Using 3D epigenomic maps of primary olfactory neuronal cells from living individuals to understand gene regulation. Sci Adv 4:eaav8550
Wang, Daifeng; Liu, Shuang; Warrell, Jonathan et al. (2018) Comprehensive functional genomic resource and integrative model for the human brain. Science 362:
Li, Mingfeng; Santpere, Gabriel; Imamura Kawasawa, Yuka et al. (2018) Integrative functional genomic analysis of human brain development and neuropsychiatric risks. Science 362:
An, Joon-Yong; Lin, Kevin; Zhu, Lingxue et al. (2018) Genome-wide de novo risk score implicates promoter variation in autism spectrum disorder. Science 362:
Kozlenkov, Alexey; Li, Junhao; Apontes, Pasha et al. (2018) A unique role for DNA (hydroxy)methylation in epigenetic regulation of human inhibitory neurons. Sci Adv 4:eaau6190
Jaffe, Andrew E (2016) Postmortem human brain genomics in neuropsychiatric disorders--how far can we go? Curr Opin Neurobiol 36:107-11
PsychENCODE Consortium; Akbarian, Schahram; Liu, Chunyu et al. (2015) The PsychENCODE project. Nat Neurosci 18:1707-12