Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jul 9;20(1):135.
doi: 10.1186/s13059-019-1747-7.

Cell type-specific epigenetic links to schizophrenia risk in the brain

Isabel Mendizabal  1 , Stefano Berto  2 , Noriyoshi Usui  2   3   4 , Kazuya Toriumi  2   5 , Paramita Chatterjee  1 , Connor Douglas  2 , Iksoo Huh  1   6 , Hyeonsoo Jeong  1 , Thomas Layman  1 , Carol A Tamminga  7 , Todd M Preuss  8 , Genevieve Konopka  9 , Soojin V Yi  10
Affiliations

Cell type-specific epigenetic links to schizophrenia risk in the brain

Isabel Mendizabal et al. Genome Biol. .

Abstract

Background: The importance of cell type-specific epigenetic variation of non-coding regions in neuropsychiatric disorders is increasingly appreciated, yet data from disease brains are conspicuously lacking. We generate cell type-specific whole-genome methylomes (N = 95) and transcriptomes (N = 89) from neurons and oligodendrocytes obtained from brain tissue of patients with schizophrenia and matched controls.

Results: The methylomes of the two cell types are highly distinct, with the majority of differential DNA methylation occurring in non-coding regions. DNA methylation differences between cases and controls are subtle compared to cell type differences, yet robust against permuted data and validated in targeted deep-sequencing analyses. Differential DNA methylation between control and schizophrenia tends to occur in cell type differentially methylated sites, highlighting the significance of cell type-specific epigenetic dysregulation in a complex neuropsychiatric disorder.

Conclusions: Our results provide novel and comprehensive methylome and transcriptome data from distinct cell populations within patient-derived brain tissues. This data clearly demonstrate that cell type epigenetic-differentiated sites are preferentially targeted by disease-associated epigenetic dysregulation. We further show reduced cell type epigenetic distinction in schizophrenia.

Keywords: Brain cell type; DNA methylation; Epigenetics; Neurogenomics; Neuron; Oligodendrocyte; Schizophrenia; Transcriptome.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1
Fig. 1
Experimental design and FANS workflow example. a Postmortem brain tissue from BA46 was matched between cases with schizophrenia and unaffected individuals. Tissue pieces were processed to isolate nuclei and incubated with antibodies directed toward NeuN or OLIG2. The nuclei were sorted using fluorescence-activated nuclei sorting (FANS) to obtain purified populations of cell types. The nuclei were processed to obtain genomic DNA (gDNA) and nuclear RNA from the same pools. Nucleic acids then underwent whole-genome sequencing (WGS), whole-genome bisulfite sequencing (WGBS), or RNA sequencing (RNA-seq). b NeuN-positive (NeuN+) nuclei represent neurons within the cerebral cortex as few human NeuN-negative (NeuN−) cells in the cortex are neurons [23, 24] (e.g., Cajal-Retzius neurons). OLIG2-positive (OLIG2+) nuclei represent oligodendrocytes and their precursors [25, 26]. Isolation of nuclei expressing either NeuN conjugated to Alexa 488 or OLIG2 conjugated to Alexa 555. The nuclei were first sorted for size and complexity, followed by gating to exclude doublets that indicate aggregates of nuclei and then further sorted to isolate nuclei based on fluorescence. “Neg” nuclei are those that are neither NeuN+ nor OLIG2+. c Example percentage nuclei at each selection step during FANS. Note that while in this example more nuclei were OLIG2+, in other samples, the proportions might be reversed. d Immunocytochemistry of nuclei post-sorting. The nuclei express either NeuN or OLIG2 or are negative for both after FANS. DAPI labels all nuclei
Fig. 2
Fig. 2
Divergent patterns of DNA methylation in NeuN+ and OLIG2+ cell types in the human brain. a Clustering analysis based on whole-genome CpG methylation values completely discriminated between NeuN+ (N = 25) and OLIG2+ (N = 20) methylomes. Additional NeuN+ (colored in turquoise) and those labeled as NeuN− (pink) are from [27]. b Density plots showing the distribution of fractional methylation differences between OLIG2+ and NeuN+ at differentially methylated positions (DMPs) and other CpGs (non-DMPs). We observed a significant excess of NeuN+-hypermethylated DMPs compared to OLIG2+ (binomial test with expected probability = 0.5, P < 10−15). c Heatmap of the top 1000 most significant DMRs between OLIG2+ and NeuN+. Fractional methylation values per individual (column) and DMR (row) show substantial differences in DNA methylation and clear cell type clustering. d Genic annotation of DMRs and coverage with Illumina 450K Methylation Arrays. Counts of different genic positions of DMRs are shown. DMRs containing at least one CpG covered by a probe in the array are indicated. Only 20.8% of the DMRs contain one or more CpG targeted by Illumina 450K Array probes. e DMRs are enriched for brain enhancers. Enrichment of enhancer states at DMRs compared to the 100 matched control DMR sets from 127 tissues [28]. Random sets are regions with similar characteristics as, including the total number of regions, length, chromosome, and CG content. f Correspondence between cell type-specific methylation sites in NeuN+ and OLIG2+ with NeuN+ and NeuN− ChIP-seq datasets [9]. Neuron-specific ChIP-seq peaks show an excess of sites with NeuN+-specific hypomethylated sites (positive DSS statistic, see the “Methods” section) whereas non-neuron peaks showed significant enrichment for sites specifically hypomethylated in OLIG2+ (negative DSS statistic). g Distribution of cell type-specific methylation differences at CpGs within H3K27ac ChIP-seq peaks in NeuN+ and NeuN− nuclei. Positive values of DSS statistic indicate hypomethylation in NeuN+ compared to OLIG2+, whereas negative values indicate hypermethylation (see the “Methods” section). Dashed lines indicate the significance level for DSS analyses
Fig. 3
Fig. 3
Gene expression signatures in NeuN+ and OLIG2+ nuclei. a Heatmap of cell type DEGs with covariates indicated. b Cell deconvolution of bulk RNA-seq data from the CommonMind Consortium and BrainSeq compared with NeuN+ and OLIG2+ (control samples). Y-axes show the weighed proportion of cells that explain the expression of bulk RNA-seq. c Gene set enrichment for cell type markers from single-nuclei RNA-seq. Enrichment analyses were performed using Fisher’s exact test. Odds ratios and FDRs (within parentheses) are shown. d Correspondence between the expression change and methylation change in cell types. The X-axis represents differential DNA methylation statistics for genes harboring DMRs in promoters. The Y-axis indicates the log2(fold change) of expression between the two cell types. The negative correlation supports the well-established impact of promoter hypomethylation on the upregulation of gene expression
Fig. 4
Fig. 4
Cell type DNA methylation patterns associated with schizophrenia. a DMPs associated with schizophrenia. Quantile-quantile plots of genome-wide P values for differential methylation between schizophrenia and control based on NeuN+ (left) and OLIG2+ (right) WGBS data. b Concordance between WGBS data and microarray-based data. Y-axis shows the ratio of sites showing the concordant direction in schizophrenia vs. control in our study at each P value bin compared with the Jaffe et al. study [7] (X-axis). Concordance was tested using a binomial test (stars indicate P < 0.05). Boxplots correspond to the directional concordance in 100 sets of association results after case-control label permutations. NeuN+ (left) and OLIG2+ (right). c szDMPs show altered cell type differences. Barplot shows the percentage of sites with larger cell type differences in controls than in schizophrenia and vice versa at different CpG classes. Absolute OLIG2+ vs. NeuN+ methylation differences are larger in controls than cases in szDMPs compared to cell type DMPs and non-DMP or background CpGs. szDMPs were detected as differentially methylated between cases and controls at FDR < 0.2 in NeuN+ (14 sites) and OLIG2+ samples (83 sites). Top 1000 szDMPs were selected as the top 1000 loci according to best P values in each cell type (N = 2000). Cell type DMPs were detected by comparing OLIG2+ vs. NeuN+ methylomes at Bonferroni P < 0.05. Background CpGs were sampled from CpGs showing non-significant P values for both case-control and OLIG2+ vs. NeuN+ comparisons. Stars represent P values for binomial tests with all comparisons showing P < 10−7. d Top 1000 szDMPs are enriched for SZ GWAS signals. szDMPs identified in our methylation study in both cell types consistently co-localize with genetic variants with moderate to large effect sizes for schizophrenia risk than expected. The table shows the empirical P values of szDMPs at each odds ratio (OR) percentile of different traits from genome-wide association studies (GWAS). The actual ORs corresponding to the schizophrenia percentiles are indicated at the top. Specifically, for each szDMP, we identified all SNPs reported by the GWAS study within a 1-kb window and counted the number of SNPs at different quantiles of odds ratio (OR). We used quantiles of OR so that we can compare the different diseases and traits among them. We repeated this step using the same number of random non-szDMPs 100 times. To obtain empirical P values, we calculated the number of times non-szDMP sets showed more SNPs in each OR quantile than szDMPs. SNPs with moderate-to-high OR in schizophrenia GWAS consistently showed low empirical P values for both cell type DMPs, implying that SNPs with large effect sizes in GWAS studies are closer to szDMPs than expected. Interestingly, this pattern was not observed for other traits, implying the co-localization is exclusive to the disease
Fig. 5
Fig. 5
Gene expression associated with schizophrenia in NeuN+ and OLIG2+. a Heatmap of szDEGs for each cell type with covariates indicated. b The first principal component of the DEGs was associated with diagnosis. Red dotted line corresponds to P = 0.05. c Volcano plot showing szDEGs. X-axis indicates the log2(fold change), and Y-axis indicates log10(FDR). szDEGs (FDR < 0.01) are colored. d Comparisons of differentially expressed genes in schizophrenia from the current study with the BrainSeq and CMC data. We used genes that are classified as differentially expressed in all three datasets (each dot represents a gene, 63 genes are common to NeuN+, CMC, and BrainSeq, and 49 to OLIG2+, CMC, and BrainSeq). The X-axes represent the fold change between control and schizophrenia in CMC or BrainSeq datasets, and the Y-axes represent the log2(fold change) between control and schizophrenia in the current datasets, for either NeuN+-specific or OLIG2+-specific genes. Regression line and confidence interval are shown for each comparison. e Barplot highlighting the enrichment for trait-associated genetic variants. Bars correspond to NeuN+ (cyan) and OLIG2+ (magenta) szDEGs. Red dashed line corresponds to the FDR threshold of 0.05. X-axis shows the acronyms for the GWAS data utilized for this analysis (ADHD, attention deficit hyperactivity disorder; ASD, autism spectrum disorders; BIP, bipolar disorder; ALZ, Alzheimer’s disease; MDD, major depressive disorder; SZ, schizophrenia; CognFun, cognitive function; EduAtt, educational attainment; Intelligence, intelligence; BMI, body mass index; CAD, coronary artery disease; DIAB, diabetes; HGT, height; OSTEO, osteoporosis). f Enrichment map for szDEGs (up-/downregulated) and the top 1000 szDMPs (X-axis shows genic annotation). Enrichment analyses were performed using Fisher’s exact test. Reported odds ratios and FDRs within parentheses for NeuN+ (top) and OLIG2+ (bottom)
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Birnbaum R, Weinberger DR. Genetic insights into the neurodevelopmental origins of schizophrenia. Nat Rev Neurosci. 2017;18:727. doi: 10.1038/nrn.2017.125. - DOI - PubMed
    1. Gusev A, Mancuso N, Won H, Kousi M, Finucane HK, Reshef Y, Song L, Safi A, Schizophrenia Working Group of the Psychiatric Genomics C. McCarroll S, et al. Transcriptome-wide association study of schizophrenia and chromatin activity yields mechanistic disease insights. Nat Genet. 2018;50:538–548. doi: 10.1038/s41588-018-0092-1. - DOI - PMC - PubMed
    1. Loh PR, Bhatia G, Gusev A, Finucane HK, Bulik-Sullivan BK, Pollack SJ, Schizophrenia Working Group of Psychiatric Genomics C. de Candia TR, Lee SH, Wray NR, et al. Contrasting genetic architectures of schizophrenia and other complex diseases using fast variance-components analysis. Nat Genet. 2015;47:1385–1392. doi: 10.1038/ng.3431. - DOI - PMC - PubMed
    1. Schizophrenia Working Group of the Psychiatric Genomics C Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511:421–427. doi: 10.1038/nature13595. - DOI - PMC - PubMed
    1. Fromer M, Roussos P, Sieberts SK, Johnson JS, Kavanagh DH, Perumal TM, Ruderfer DM, Oh EC, Topol A, Shah HR, et al. Gene expression elucidates functional impact of polygenic risk for schizophrenia. Nat Neurosci. 2016;19:1442–1453. doi: 10.1038/nn.4399. - DOI - PMC - PubMed

Publication types