If you have epigenetics on the brain like us, concentrate your mammalian mainframe on the recently published marvels on the brain’s epigenome from the NIH’s Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative – Cell Census Network (BICCN). Recent work brought together under the guise of the Brain Cell Census now reports on the examination of the cellular composition of the developing and adult human brain at various levels to allow a deeper appreciation of the comprising wires, connections, and supports!

We now bring you all the thought-provoking details from the most (epigenetically) exciting articles from this vast haul of research papers. Our first article employs epigenomic analysis to study DNA methylation and chromatin conformation in single cells across adult human brain regions to gain an appreciation of the gene-regulatory programs controlling cell function. Meanwhile, a second article describes a single-cell chromatin conformation “atlas” of the human brain that offers insight into the gene-regulatory programs that control the cellular composition of the brain and the relevance of disease risk variants that lie outside of protein-coding regions.

Exploring Epigenetic Gene Regulatory Mechanisms in Single Brain Cells to Understand Human Brain Function

Gaining an appreciation of how the healthy brain functions requires an awareness of the gene-regulatory programs controlling the constituting cells; therefore, brainy researchers led by Joseph Ecker (Salk Institute for Biological Studies) examined human brain cell epigenomes by evaluating single-cell DNA methylation and chromatin conformation profiles in neurons and non-neurons from regions spanning the adult brain. Their brain-busting Brain Cell Census study employs snmC-seq3 (DNA methylation) and snm3C-seq (DNA methylation and chromatin conformation) to identify epigenetically distinct cell types and reveal changes in various epigenetic parameters across cells and areas of the mammalian mainframe.

Let’s hear the details of this epigenetic teardown of the mammalian mainframe by Tian, Zhou, and Colleagues:

• Clustering based on DNA methylation divides cells into telencephalic excitatory neurons, inhibitory and/or non-telencephalic neurons, and non-neuronal cells, which are further divided into 40 major types and 188 subtypes
  • Major non-neuronal cells distribute evenly across the brain, while neuronal cells exhibit spatial specificity
• Clustering based on chromatin conformation data separates all major cell types through chromatin contacts only and highlights the diversity of chromatin conformation across brain regions
• Investigating cell-type-specific genome folding by evaluating contacts per cell at genome distances reveals more short-distance interactions in neurons but longer-range contacts in non-neural cells
  • Excitatory neurons display more shorter-range interactions than inhibitory neurons
  • Neurons display an enrichment for short-range intradomain/interdomain interactions, while non-neural cells display enriched intracompartment interactions and depleted intercompartment interactions
• Compartment scores, domain boundary probabilities, and loop strength distinguish between cell types and determine the hierarchy of similarities, indicating cell-type specificities of 3D structures
  • However, the differential ability of 3D features to distinguish cells underscores their varying roles in gene regulation across cell types (e.g., with loops more specific than domains)
• Cell-type-specific DNA methylation profiles depict distinct epigenetic signatures for brain cell identities and provide critical insight into understanding gene-regulatory programs in brain cells
  • Integrating DNA methylation with chromatin conformation reveals distinct cell-type regulatory dynamics
  • These data help to link brain diseases/traits and differentially-methylated regions
• Integrating single-cell DNA methylation profiles with brain region data maps the cells onto a “regional methylation space” where cells closer together possess DNA methylation “neighbors” from similar brain regions
  • These findings suggest region-specific regulatory mechanisms that underlie functional diversity
• Distinct DNA methylation patterns at CpG sites in specific brain cells support the creation of single-cell methylation barcodes (scMCodes) that determine brain cell types at the single-cell level via DNA methylation

Overall, this multimodal map of DNA methylation and chromatin conformation in single cells in the human adult brain provides insight into the complexity of cell-type-specific gene regulation and may fuel additional studies into brain cell diversity and gene regulation mechanisms. Furthermore, scMCodes could support precision medicine by developing non-invasive diagnostics for brain disorders.

Chromatin Accessibility Atlas Links Brain-cell Specific Regulatory Regions to Neuropsychiatric Disorders

Understanding neuropsychiatric disorders remains a challenging task as disease-associated sequence variants in the human genome often lie outside protein-coding regions and lack known function. These disease risk variants may perturb transcriptional regulatory elements, prompting altered chromatin accessibility in relevant brain cell types. These facts prompted a cerebral circle of researchers led by Bing Ren (University of California, San Diego) to employ single-nucleus assay for transposase-accessible chromatin using sequencing (snATAC-seq) to generate a comprehensive single-cell chromatin accessibility atlas of 42 distinct human brain regions. Their Brain Cell Census study defines ~100 brain cell types and uncovers the epigenetic state of over 500,000 candidate cis-regulatory DNA elements (cCREs).

Let’s hear more from Li and Colleagues on the potential of a chromatin accessibility atlas of the mammalian mainframe:

• Clustering of snATAC-seq profiles classifies cells into glutamatergic and GABAergic neurons and non-neuronal cells
  • Additional iterative clustering classified these classes into 42 subclasses, which gives 107 distinct cell types
• Most neuronal cell types, but only some glial cell types (e.g., Bergmann glia), display strong regional specificity throughout the adult human brain
  • Overall, distinct neuronal subtypes located in differing brain subregions display a stark separation
• The search for cCREs identified 62,045 open chromatin regions per cell type and 544,735 open chromatin regions in total across all 107 cell types
  • cCREs constitute 8.8% of the human genome, with 95.3% located at least 2 kbp away from annotated gene promoter regions and 22% overlapping endogenous retrotransposable elements
  • Most cCREs display highly variable chromatin accessibility across the brain cell types
• Integrating chromatin accessibility with single-cell transcriptome/DNA methylome data links cCREs to target genes
  • Leveraging the chromatin accessibility atlas predicts disease-relevant brain cell types for 19 neuropsychiatric traits and disorders, including schizophrenia, bipolar disorder, Alzheimer’s disease, and major depression
• The development of machine learning models employing this data supports the prediction of the regulatory roles of non-coding risk variants in these neuropsychiatric disorders
  • Models are freely available through an interactive web portal named the “cis-element atlas” or “CATlas”

This single-cell chromatin accessibility atlas provides further insight into the gene-regulatory programs that shape the diversity of cells – neurons and non-neurons alike – throughout the human adult brain. However, even more interestingly, the atlas supports the interpretation of the functional roles of disease risk variants located outside of protein-coding regions, which may offer novel means to address neuropsychiatric disorders.

The Brain Cell Census: More Data Downloads for your Mammalian Mainframe

For the vast impact and overall relevance of all the studies that constitute the Brain Cell Census, head over to Science (“A quest into the human brain“) and Nature (“This is the largest map of the human brain ever made“) and download all the facts into your very-own mammalian mainframe!

For more on how deciphering epigenetic gene regulatory mechanisms in single cells can aid our appreciation of human brain function, see Tian, Zhou, and Colleagues in Science, October 2023; and for more on how a chromatin accessibility atlas can link brain-cell specific regulatory regions to neuropsychiatric disorders, see Li and Colleagues in Science, October 2023.