//tags: Single Cell, Atlas, Chromatin, Epigenetics, DNA methylation, chromatin conformation
Mapping DNA methylation and 3D chromatin conformation at single-cell resolution across the human body has now created a comprehensive epigenetic atlas.
DNA methylation and 3D chromatin conformation represent two of the dynamic epigenetic mechanisms that cooperate to control the lineage-specific gene expression patterns that characterize the distinct cell types that make up the human body (Allis and Jenuwein). Early "bulk" efforts to map these distinct epigenetic regulatory layers generally failed to capture cell-type-specific patterns due to a lack of resolution; however, the advancing field of single-cell epigenetics and transcriptomics has enabled epigenetic mapping in complex tissues, though these tissues have been limited in scope or studied in isolation (Tian et al., Luo et al., Nagano et al., and Tan et al.).
In a significant advance towards understanding the relationships among distinct chromatin regulatory mechanisms in single cells, researchers from the lab of Jesse R. Dixon and Joseph R. Ecker recently applied single-nucleus methyl-3C sequencing (snm3C-seq) - which simultaneously assays DNA methylation and 3D chromatin conformation in individual nuclei - to a panel of adult human tissues. Their recent Science study now reports on a comprehensive epigenetic atlas of DNA methylation and 3D chromatin conformation from nearly 87,000 single nuclei across 16 tissues, comprising 35 major cell types and 206 cell subtypes (Zhou and Wu et al.). The team behind this fascinating research hopes that characterizing these epigenetic features at this unprecedented scale and cell-type resolution will provide a useful resource that will significantly accelerate related research aims.
Paired-Tag technology from Epigenome Technologies generates joint epigenetic and transcriptomic profiles at single-cell resolution and detects histone modifications and RNA transcripts in nuclei with efficiency comparable to single-nucleus RNA-seq/ChIP-seq assays. Could an in-depth analysis of the histone modification and transcriptomic profiles of the same single cells afforded by the integration of Paired-Tag have provided another level to this comprehensive epigenetic atlas covering an enormous range of human cells?
As a summary of the findings arising from the construction of this comprehensive epigenetic atlas, the authors first found that canonical DNA methylation patterns (i.e., those at CpG sites) varied widely across the cell types under study; however, certain cell lineages possessed domains characterized by partial DNA methylation patterns that corresponded to generally repressive regions of the genome. More interestingly, the study found that a low-level non-CG methylation pattern previously associated only with neurons actually occurred across a broader range of cell types; additionally, the ability of non-CG methylation patterns to distinguish major cell types identified in this study suggests they reflected cell identity. The study highlighted the need for future studies regarding how these patterns become installed and how to interpret them in non-brain tissues.
[! Multi-panel image showing CH methylation cell embeddings and quantifications](img/image0.png Panels G through I highlight the broader biological occurrence and regulatory relevance of non-CG DNA methylation (mCH) beyond the brain. Summarizing the context, while mCH was previously thought to be restricted to neurons, this atlas reveals that low-level mCH patterns occur across a wide range of cell types and reflect distinct cell identities. Panel G features t-SNE plots of all 86,689 single cells using mCH profiles, colored by major cell types, global mCH levels, or tissue sources, while Panel H demonstrates that mCH can also resolve cellular subtypes within endocrine pancreas epithelial cells and skeletal muscle cells. Finally, Panel I illustrates the functional relevance of these patterns by showing correlations between mCH levels across transcriptional start sites and gene bodies versus the expression of differentially expressed genes across pancreatic endocrine subtypes. From (Zhou and Wu et al.)
Moving to the characterization of 3D chromatin conformation (carried out across an unprecedented number of cell types), this study next revealed a vast number of chromatin loops that varied between cell types and correlated with cell-type-specific gene expression levels. Interestingly, they also revealed that distinct 3D chromatin conformation elements characterized distinct cell lineages, with strong local chromatin domains in neurons and long-range compartment interactions in hematopoietic cells provided as examples. While the basis for these differences remains unclear, the authors of this study speculated that compartment- or domain-dominant phenotypes may result from differences in global rates of chromatin loop extrusion.
Importantly, the real insight came when finally combining DNA methylation and 3D chromatin conformation, which provided proof for a greater diversity of cells/cell subtypes; for example, the combination of these epigenetic layers revealed the existence of overlapping but distinct skeletal muscle fiber and Schwann cell states, which the authors believe may reflect cells transitioning between differentiation states. They note that reducing data sparsity in future single-cell multiomic technologies could clarify the nature of these cell types, and that understanding the mechanisms that give rise to these divergent epigenomic patterns may be a fruitful area for future expansion.
[! Multi-panel image showing cell embeddings and relationships](img/image1.png Combining single-nucleus DNA methylation and 3D chromatin conformation mapping reveals unexpected cellular diversity and confluence, such as overlapping yet distinct skeletal muscle fiber and Schwann cell states that may reflect ongoing cell transitions. Panels A through C highlight the degree of concordance between these two epigenetic modalities in defining cell subtypes. Specifically, Panel A uses the Adjusted Rand index (ARI) to quantify clustering consistency between mCG and chromatin contact data across major cell types. Panels B and C then use t-SNE projections of skeletal muscle cells to compare mCG-based versus chromatin contact-based groupings, highlighting cell groups and global mCG levels to demonstrate how integrating these layers refines cellular classification. From (Zhou and Wu et al.)
The authors hope that this deeper understanding of the diversity of DNA methylation and 3D chromatin conformation patterns across multiple cell types from tissues throughout the human body, and the creation of a comprehensive epigenetic atlas, will serve as a platform for future research on the epigenomic basis of gene regulation, cell differentiation, and human disease.
Implementing Paired-Tag technology from Epigenome Technologies, which generates joint epigenetic and transcriptomic profiles at single-cell resolution and detects histone modifications and RNA transcripts in individual nuclei with efficiency comparable to single-nucleus RNA-seq/ChIP-seq assays, could provide deeper insight into these research aims. Could the simultaneous single-cell analysis of histone modifications and transcriptomic profiles have defined an even broader range of cells/cell subtypes in the cells of the human body and contributed towards the construction of an improved epigenetic atlas?