Epigenetic modifications confer stable transcriptional patterns in the brain, and both normal and abnormal brain function involve specialized brain regions. We examined DNA methylation by whole-genome bisulfite sequencing in neuronal and non-neuronal populations from four brain regions (anterior cingulate gyrus, hippocampus, prefrontal cortex, and nucleus accumbens) as well as chromatin accessibility in the latter two. We find pronounced differences in both CpG and non-CpG methylation (CG-DMRs and CH-DMRs) only in neuronal cells across brain regions. Neuronal CH-DMRs were highly associated with differential gene expression, whereas CG-DMRs were consistent with chromatin accessibility and enriched for regulatory regions. These CG-DMRs comprise ~12 Mb of the genome that is highly enriched for genomic regions associated with heritability of neuropsychiatric traits including addictive behavior, schizophrenia, and neuroticism, thus suggesting a mechanistic link between pathology and differential neuron-specific epigenetic regulation in distinct brain regions.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Raw and processed data generated are available through NCBI GEO under accession number GSE96615. Processed data is available through a UCSC hub, at http://genome.ucsc.edu/cgi-bin/hgTracks?db = hg19&hubUrl = https://s3.us-east-2.amazonaws.com/brainepigenome/hub.txt.
Hoffmann, A., Sportelli, V., Ziller, M. & Spengler, D. Epigenomics of major depressive disorders and schizophrenia: Early life decides. Int. J. Mol. Sci. 18, E1711 (2017).
Negi, S. K. & Guda, C. Global gene expression profiling of healthy human brain and its application in studying neurological disorders. Sci. Rep. 7, 897 (2017).
Mo, A. et al. Epigenomic signatures of neuronal diversity in themammalian brain. Neuron 86, 1369–1384 (2015).
Marques, S. et al. Oligodendrocyte heterogeneity in the mouse juvenile and adult central nervous system. Science 352, 1326–1329 (2016).
Lake, B. B. et al. Neuronal subtypes and diversity revealed by single-nucleus RNA sequencing of the human brain. Science 352, 1586–1590 (2016).
Tasic, B. et al. Adult mouse cortical cell taxonomy revealed by single cell transcriptomics. Nat. Neurosci. 19, 335–346 (2016).
Zeisel, A. et al. Brain structure. Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq. Science 347, 1138–1142 (2015).
Luo, C. et al. Single-cell methylomes identify neuronal subtypes and regulatory elements in mammalian cortex. Science 357, 600–604 (2017).
Davies, M. N. et al. Functional annotation of the human brain methylome identifies tissue-specific epigenetic variation across brain and blood. Genome Biol. 13, R43 (2012).
Illingworth, R. S. et al. Inter-individual variability contrasts with regional homogeneity in the human brain DNA methylome. Nucleic Acids Res. 43, 732–744 (2015).
Ladd-Acosta, C. et al. DNA methylation signatures within the human brain. Am. J. Hum. Genet. 81, 1304–1315 (2007).
Viana, J. et al. Schizophrenia-associated methylomic variation: molecular signatures of disease and polygenic risk burden across multiple brain regions. Hum. Mol. Genet. 26, 210–225 (2017).
Watson, C. T. et al. Genome-wide DNA methylation profiling in the superior temporal gyrus reveals epigenetic signatures associated with Alzheimer’s disease. Genome Med. 8, 5 (2016).
Kaut, O. et al. Aberrant NMDA receptor DNA methylation detected by epigenome-wide analysis of hippocampus and prefrontal cortex in major depression. Eur. Arch. Psychiatry Clin. Neurosci. 265, 331–341 (2015).
Kundaje, A. et al. Integrative analysis of 111 reference human epigenomes. Nature 518, 317–330 (2015).
Jaffe, A. E. et al. Mapping DNA methylation across development, genotype and schizophrenia in the human frontal cortex. Nat. Neurosci. 19, 40–47 (2016).
Ellis, S. E., Gupta, S., Moes, A., West, A. B. & Arking, D. E. Exaggerated CpH methylation in the autism-affected brain. Mol. Autism 8, 6 (2017).
Mo, A. et al. Epigenomic landscapes of retinal rods and cones. eLife 5, e11613 (2016).
Stroud, H. et al. Early-life gene expression in neurons modulates lasting epigenetic states. Cell 171, 1151–1164.e16 (2017).
Kozlenkov, A. et al. Substantial DNA methylation differences between two major neuronal subtypes in human brain. Nucleic Acids Res. 44, 2593–2612 (2016).
Kozlenkov, A. et al. Differences in DNA methylation between human neuronal and glial cells are concentrated in enhancers and non-CpG sites. Nucleic Acids Res. 42, 109–127 (2014).
Lister, R. et al. Global epigenomic reconfiguration during mammalian brain development. Science 341, 1237905 (2013).
Sanchez-Mut, J. V. et al. Human DNA methylomes of neurodegenerative diseases show common epigenomic patterns. Transl. Psychiatry 6, e718 (2016).
Morel, L. et al. Molecular and functional properties of regional astrocytes in the adult brain. J. Neurosci. 37, 8706–8717 (2017).
von Bartheld, C. S., Bahney, J. & Herculano-Houzel, S. The search for true numbers of neurons and glial cells in the human brain: A review of 150 years of cell counting. J. Comp. Neurol. 524, 3865–3895 (2016).
Hansen, K. D., Langmead, B. & Irizarry, R. A. BSmooth: from whole genome bisulfite sequencing reads to differentially methylated regions. Genome. Biol. 13, R83 (2012).
Montaño, C. M. et al. Measuring cell-type specific differential methylation in human brain tissue. Genome. Biol. 14, R94 (2013).
Vermunt, M. W. et al. Large-scale identification of coregulated enhancer networks in the adult human brain. Cell Rep. 9, 767–779 (2014).
Andersson, R. et al. An atlas of active enhancers across human cell types and tissues. Nature 507, 455–461 (2014).
Hansen, K. D. et al. Increased methylation variation in epigenetic domains across cancer types. Nat. Genet. 43, 768–775 (2011).
Hansen, K. D. et al. Large-scale hypomethylated blocks associated with Epstein-Barr virus-induced B-cell immortalization. Genome Res. 24, 177–184 (2014).
McLean, C. Y. et al. GREAT improves functional interpretation of cis-regulatory regions. Nat. Biotechnol. 28, 495–501 (2010).
Guo, J. U. et al. Distribution, recognition and regulation of non-CpG methylation in the adult mammalian brain. Nat. Neurosci. 17, 215–222 (2014).
Ziller, M. J. et al. Genomic distribution and inter-sample variation of non-CpG methylation across human cell types. PLoS Genet. 7, e1002389 (2011).
Buenrostro, J. D., Wu, B., Chang, H. Y. & Greenleaf, W. J. ATAC-seq: A method for assaying chromatin accessibility genome-wide. Curr. Protoc. Mol. Biol. 109, 1–9 (2015).
Fullard, J. F. et al. Open chromatin profiling of human postmortem brain infers functional roles for non-coding schizophrenia loci. Hum. Mol. Genet. 26, 1942–1951 (2017).
Pinello, L., Farouni, R. & Yuan, G. C. Haystack: systematic analysis of the variation of epigenetic states and cell-type specific regulatory elements. Bioinformatics 34, 1930–1933 (2018).
Fukuchi, M. & Tsuda, M. Convergence of neurotransmissions at synapse on IEG regulation in nucleus. Front. Biosci. (LandmarkEd.) 1052–1072 (2017).
Hu, T. M., Chen, C. H., Chuang, Y. A., Hsu, S. H. & Cheng, M. C. Resequencing of early growth response 2 (EGR2) gene revealed a recurrent patient-specific mutation in schizophrenia. Psychiatry Res. 228, 958–960 (2015).
Pfaffenseller, B. et al. Differential expression of transcriptional regulatory units in the prefrontal cortex of patients with bipolar disorder: potential role of early growth response gene 3. Transl. Psychiatry 6, e805 (2016).
Larson, E. B. et al. Striatal regulation of ΔFosB, FosB, and cFos during cocaine self-administration and withdrawal. J. Neurochem. 115, 112–122 (2010).
Yin, Y. et al. Impact of cytosine methylation on DNA binding specificities of human transcription factors. Science 356, eaaj2239 (2017).
Durek, P. et al. Epigenomic profiling of human CD4+ T cells supports a linear differentiation model and highlights molecular regulators of memory development. Immunity 45, 1148–1161 (2016).
Schmidt, F. et al. Combining transcription factor binding affinities with open-chromatin data for accurate gene expression prediction. Nucleic Acids Res. 45, 54–66 (2017).
Finucane, H. K. et al. Partitioning heritability by functional annotation using genome-wide association summary statistics. Nat. Genet. 47, 1228–1235 (2015).
Schizophrenia Working Group of the Psychiatric Genomics Consortium. Biological insights from 108 schizophrenia-associated genetic loci. Nature 511, 421–427 (2014).
Okuno, H. Regulation and function of immediate-early genes in the brain: beyond neuronal activity markers. Neurosci. Res. 69, 175–186 (2011).
Hertzberg, L., Katsel, P., Roussos, P., Haroutunian, V. & Domany, E. Integration of gene expression and GWAS results supports involvement of calcium signaling in schizophrenia. Schizophr. Res. 164, 92–99 (2015).
Jaffe, A. E. et al. Developmental and genetic regulation of the human cortex transcriptome illuminate schizophrenia pathogenesis. Nat. Neurosci. 21, 1117–1125 (2018).
Feinberg, A. P. Phenotypic plasticity and the epigenetics of human disease. Nature 447, 433–440 (2007).
Harrow, J. et al. GENCODE: the reference human genome annotation for The ENCODE Project. Genome Res. 22, 1760–1774 (2012).
Gardiner-Garden, M. & Frommer, M. CpG islands in vertebrate genomes. J. Mol. Biol. 196, 261–282 (1987).
Rosenbloom, K. R. et al. The UCSC Genome Browser database: 2015 update. Nucleic Acids Res. 43, D670–D681 (2015).
Krueger, F. & Andrews, S. R. Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics 1571–1572 (2011).
Kuhn, R. M., Haussler, D. & Kent, W. J. The UCSC genome browser and associated tools. Brief. Bioinform. 14, 144–161 (2013).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Zhang, Y. et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008).
Lawrence, M. et al. Software for computing and annotating genomic ranges. PLoS Comput. Biol. 9, e1003118 (2013).
Law, C. W., Chen, Y., Shi, W. & Smyth, G. K. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 15, R29 (2014).
Robinson, M. D. & Oshlack, A. A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol. 11, R25 (2010).
Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140 (2010).
Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015).
Leek, J. T. & Storey, J. D. Capturing heterogeneity in gene expression studies by surrogate variable analysis. PLoS Genet. 3, 1724–1735 (2007).
Patro, R., Duggal, G., Love, M. I., Irizarry, R. A. & Kingsford, C. Salmon provides accurate, fast, and bias-aware transcript expression estimates using dual-phase inference. Nat. Methods 14, 417–419 (2017).
Soneson, C., Love, M. I. & Robinson, M. D. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Res. 4, 1521 (2015).
R. Core Team. A Language and Environment for Statistical Computing. (R Foundation for Statistical Computing, Vienna, Austria, 2016).
Wood, S. N. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J. R. Stat. Soc. B 73, 3–36 (2011).
Bulik-Sullivan, B. K. et al. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat. Genet. 47, 291–295 (2015).
Holm, S. A simple sequentially rejective multiple test procedure. Scand. J. Stat. 6, 65–70 (1979).
Mathelier, A. et al. JASPAR 2016: a major expansion and update of the open-access database of transcription factor binding profiles. Nucleic Acids Res. 44(D1), D110–D115 (2016).
Tripathi, S. et al. Meta- and orthogonal integration of influenza “OMICs” data defines a role for UBR4 in virusbudding. Cell Host Microbe 18, 723–735 (2015).
Durinck, S. et al. BioMart and Bioconductor: a powerful link between biological databases and microarray data analysis. Bioinformatics 21, 3439–3440 (2005).
Durinck, S., Spellman, P. T., Birney, E. & Huber, W. Mapping identifiers for the integration of genomic datasets with the R/Bioconductor package biomaRt. Nat. Protoc. 4, 1184–1191 (2009).
Gentleman, R. C. et al. Bioconductor: open software development for computational biology and bioinformatics. Genome. Biol. 5, R80 (2004).
Huber, W. et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat. Methods 12, 115–121 (2015).
Wickham, H. ggplot2: Elegant Graphics for Data Analysis, (Springer, New York, NY, USA, 2016).
Lawrence, M., Gentleman, R. & Carey, V. rtracklayer: an R package for interfacing with genome browsers. Bioinformatics 25, 1841–1842 (2009).
This work was supported by funding awarded to A.P.F. (U01MH104393) through the enhanced Genotype-Tissue Expression (eGTEx) project supported by the Common Fund of the Office of the Director of NIH. This work was supported by National Cancer Institute under U24CA180996. We would like to thank H. Zhang from the Flow Cytometry Cell Sorting Core Facility at Johns Hopkins School of Public Health for flow sorting. The core facility is supported by CFAR: 5P30AI094189–04, 1S10OD016315–01, and 1S10RR13777001. Brain tissues were received from the NIH NeuroBioBank at the University of Maryland and University of Pittsburgh.
Integrated supplementary information
Supplementary Figure 1 Isolation and global methylation analysis of neuronal (NeuN+) and non-neuronal (NeuN−) nuclei from frozen brain tissues.
(a) Nuclei in this representative example were isolated by fluorescence activated nuclei sorting (FANS) from prefrontal cortex (BA9) and debris, doublets, and auto-fluorescent nuclei were gated out. The remaining nuclei were separated based on detection of AlexaFluor 488-conjugated anti-NeuN antibody. FANS was repeated a minimum of 8 times per brain region. (b) Example of nuclei stained with anti-NeuN and DAPI both pre- and post-sort; staining repeated twice with similar results. (c) Proportion of NeuN+ nuclei isolated via FANS from tissue samples from indicated brain regions; arrows indicate two punches from the same tissue sample. Tissues are from BA9, anterior cingulate cortex (BA24), hippocampus (HC) and nucleus accumbens (NAcc) from six individuals as indicated. (d) Proportion of NeuN+ nuclei isolated via FANS plotted against post-mortem interval from tissue samples from indicated brain regions (BA9, n = 11; BA24, n = 13; HC, n = 13; NAcc, n = 8 individuals). Overlaid is a line showing the conditional mean, and shaded 95% confidence interval, for this relationship as estimated using a linear model within each tissue. (e) Average autosomal CpG (mCG) and (f) non-CpG (mCH) methylation of NeuN+ (n = 6 individuals for BA9, HC, NAcc and n = 5 for BA24) and NeuN− nuclei (n = 6 individuals for BA9, HC, NAcc and n = 4 for BA24); significant differences in global mC between tissues within dinucleotide contexts and NeuN status are indicated. P-values in (e,f) are adjusted for multiple comparisons using Tukey’s honest significant difference test.
Supplementary Figure 2 Differences in DNA methylation between brain regions are restricted to neuronal nuclei.
Principal component analyses (PCA) of distances derived from average autosomal CpG methylation levels in 1 kb intervals in (a) bulk tissues, (b) NeuN+ nuclei, (c) bulk tissues and sorted nuclei combined, and (d) NeuN− nuclei. Tissues, sample type, and individuals are indicated as shown. (e) Correlation matrix between samples based on average methylation of CpGs shared across all samples. (a-e) Tissues are from anterior cingulate cortex (BA24; NeuN+ n = 5 individuals; NeuN− n = 4, bulk n = 6), dorsolateral prefrontal cortex (BA9; NeuN+ n = 6, NeuN− n = 6, bulk n = 6), hippocampus (HC; NeuN+ n = 6, NeuN− n = 6, bulk n = 6), and nucleus accumbens (NAcc; NeuN+ n = 6, NeuN− n = 6, bulk n = 6).
(a) Pearson correlations for DNA methylation between samples from the same tissue within either a donor (bulk samples composed of both NeuN+ and NeuN−; n = 27) or a cell type (NeuN+, n = 23; or NeuN−, n = 22 samples). (b) Between sample variation within a brain region and sample type, as a function of methylation level. (c) Proportion of CpGs with an estimated absolute methylation difference greater than 10% (chosen because CG-DMRs all have an absolute methylation difference > 10%) between the indicated brain regions, stratified by sample type (NeuN+, NeuN−, bulk).
Supplementary Figure 4 Methylation profiles are distinct between neuronal and non-neuronal nuclei and among neurons from distinct brain regions.
(a) Hierarchical clustering of samples based on the average per sample methylation of the top 20,000 cell type-specific CG-DMRs. Examples of (b) cell type-specific CG-DMRs and a cell type-specific block of differential methylation (NeuN+ is green; NeuN− is purple) and (c) neuronal CG-DMRs. Methylation values are shown with CG-DMRs shaded pink and overlap with protein-coding genes depicted below each graph. (d) Average CpG methylation values for NeuN+ nuclei within each brain region over a 3 kb window centered on the CG-DMRs (n = 208) identified among BA9, BA24, and HC. For BA9 and HC, n = 6 individuals and n = 5 individuals for BA24. CG-DMRs were grouped by k-means clustering based on their methylation patterns. Metagene plots for each group are shown to the right with the number of CG-DMRs indicated. (e) As in (c) with normalized ATAC sequencing coverage for NeuN+ nuclei from NAcc and BA9 (individual means are transparent lines; tissue means are opaque lines). Regions of differential accessibility (DARs), CG-DMRs and CG-blocks are shaded pink.
Neuronal (NeuN+) non-CpG methylation was measured in 1 kb bins (n = 2,881,044) across the autosomes. Bins with little to no coverage of cytosines were removed. (a) Non-CpG methylation in the CA context on the positive strand compared across two biological replicates. (b) Non-CpG methylation in the CA context compared between the positive and negative strand within an individual sample. (c) Non-CpG methylation on the positive strand compared between the CA context and the CT context within an individual sample. Solid red line is y = x, dashed red line is the best regression line through the origin (correlation and equation shown on graph).
Supplementary Figure 6 Non-CpG methylation is consistent across strand and context and with CpG methylation.
Hierarchical clustering of samples based on the z-score of methylation over (a) CA(-) DMRs, (b) CT( + ) DMRs, and (c) CT(-) DMRs. (d) Hierarchical clustering of samples based on z-scores of methylation for CA( + ) DMRs that overlap CG-DMRs.
(a) Sample correlation matrix of ATAC-seq read counts [log2(counts per million)] in open chromatin regions (green = NeuN+ nuclei, purple = NeuN− nuclei, orange = nucleus accumbens (NAcc), blue = prefrontal cortex (BA9)). For NeuN+ and NeuN− samples from BA9 and NAcc, n = 6 individuals each. (b) Mean-difference plot of peak accessibility data comparing: NeuN+ to NeuN− nuclei. Differentially accessible regions (DARs) are shown in orange. Data were randomly sampled (100% of DARs, 10% of non-DARs).
Supplementary Figure 8 DNA methylation and chromatin accessibility are correlated with gene expression.
(a) Average methylation levels around protein coding genes (n = 19,823) with different expression levels in neuronal nuclei isolated from nucleus accumbens (NAcc) and prefrontal cortex (BA9). Each gene length is split into 100 bins and data extend two lengths up and downstream. Gene expression is split into quartiles based on RPKM. (b,c) Scatterplots showing Pearson correlation of differential gene expression [log2(fold change)] (n = 16,538) with: (b) differential mCA and mCG, and (c) differential accessibility [log2(fold change)] from NAcc vs BA9 neuronal nuclei. Differentially expressed protein coding genes (DEGs) are shown in orange. Correlation values (with 95% CI) are given below each graph. (d,e) Scatterplots showing Pearson correlations of differential gene expression [log2(fold change)] with differential epigenetic features from NAcc vs BA9 neuronal nuclei that overlap (d) promoters and (e) gene bodies. DEGs are shown in orange and the number of genes (n) with a feature(s) is indicated. (f) Relationships between expression of protein coding genes and chromatin accessibility, mCA, and mCG within the gene body in NAcc and BA9 neuronal nuclei. Contours show point densities; red line shows smoothed trend. (a-f) n = 6 individuals for both NAcc and BA9 neuronal samples.
Each feature depicted in the figure was included separately in an SLDSR analysis including 53 baseline features. Results from our differential and the 3 non-differential epigenomic features (Brain H3K27ac30, CNS51, chromHMM15) are shown by trait; the same results are displayed in Fig. 5 without the ability to identify individual traits. (a) Coefficient z-scores. Enrichment is only reported for traits where at least one feature has a z-score significantly larger than 0 (one-sided z-test with alpha = 0.05, P-values corrected within each trait using Holm’s method). (b) Enrichment + /− 2 standard errors. GWAS sizes for each trait are reported in Supplementary Table 24.
Supplementary Figure 10 SLDSR analysis of differential epigenomic features, controlling for nondifferential epigenomic features.
Each of the differential epigenomic features (neuronal CG-DMRs, DARs and cell-type CG-DMRs and DARs) was included separately in a SLDSR analysis with the 3 non-differential epigenomic features (Brain H3K27ac30, CNS51, chromHMM15) included in the baseline (53 features). Results are shown by trait; the same results are displayed in Fig. 5 without the ability to identify individual traits. (a) Coefficient z-scores. Enrichment is only reported for traits where at least one feature has a z-score significantly larger than 0 (one-sided z-test with alpha = 0.05, P-values corrected within each trait using Holm’s method). (b) Enrichment + /− 2 standard errors. GWAS sizes for each trait are reported in Supplementary Table 24.
Supplementary Figure 11 SLDSR analysis of differential epigenomic features, controlling for nondifferential epigenomic features by excluding their overlap from baseline.
Each of the differential epigenomic features (neuronal CG-DMRs, DARs and cell-type CG-DMRs and DARs) was including separately in a SLDSR analysis including the 3 non-differential epigenomic features (Brain H3K27ac30, CNS51, chromHMM15) in the baseline. For the 3 non-differential epigenomic features included in the baseline, we removed any overlap with the differential epigenomic features. Results are shown by trait; the same results are displayed in Fig. 5 without the ability to identify individual traits. (a) Coefficient z-scores. Enrichment is only reported for traits where at least one feature has a z-score significantly larger than 0 (one-sided z-test with alpha = 0.05, P-values corrected within each trait using Holm’s method). (b) Enrichment + /− 2 standard errors. GWAS sizes for each trait are reported in Supplementary Table 24.
GREAT Analysis of CG-DMRs where NAcc is hypomethylated compared to other three brain regions.
List of non-CpG DMRs (CA- and CT-DMRs).
GO analysis of differentially expressed genes in NAcc vs. BA9 in neurons.
Differentially accessible regions between NeuN-positive and NeuN-negative samples.
Differentially accessible regions between NAcc and BA9 neurons.
Differentially accessible regions between NAcc and BA9 in NeuN-negative samples (non-neurons).
Transcription factor motif enrichments with neuronal DARs overlapping hyper- or hypomethylated neuronal NAcc CG-DMRs.
SLDSR analysis results.
About this article
Current Nutrition Reports (2019)
Integrated DNA methylation and gene expression profiling across multiple brain regions implicate novel genes in Alzheimer’s disease
Acta Neuropathologica (2019)