In a groundbreaking study recently published in Nature, researchers have unveiled an innovative approach to simultaneously map the DNA methylome and transcriptome within the spatial context of complex tissues. This novel methodology, termed spatial-DMT (Spatial DNA Methylome and Transcriptome profiling), was applied to the mouse brain, revealing unprecedented insights into the spatial heterogeneity and gene regulatory mechanisms governed by DNA methylation patterns. Such dual-omic spatial analyses mark a significant leap forward in understanding how epigenetic regulation integrates with gene expression to define cellular identities and regional brain functions.
The study focused on postnatal day 21 (P21) mouse brain tissue sections spanning cortical and hippocampal regions. DNA methylation of cytosines not only in the canonical CpG context but also in non-CpG contexts—particularly mCA (methylated cytosine followed by adenine)—was scrutinized. It is well-established that mCH (where H = A, C, or T) methylation, especially mCA, is abundantly enriched in neuronal genomes, contributing to brain-specific regulatory phenomena. By leveraging spatial-DMT, the authors quantified and spatially mapped methylation landscapes alongside transcriptomic profiles, unveiling distinct epigenetic and transcriptomic patterns aligned with the brain’s diverse anatomical domains.
Initial global methylation analyses revealed that mCA and mCG (methylated CpG) levels are not uniform across brain regions. Notably, regions such as the dentate gyrus (DG), and hippocampal sectors CA1/2 and CA3, exhibit relatively lower methylation levels compared to the cerebral cortex. These contrasts correlated spatially with distinct methylome and transcriptome clusters that convincingly recapitulated anatomical structures observable in histological images. This direct spatial correspondence underscores the biological relevance and precision of the spatial-DMT technique in capturing epigenetic landscapes in situ.
By clustering DNA methylation and RNA transcription data independently and also integrating both data modalities through a weighted nearest neighbor (WNN) approach, the study achieved refined spatial delineation of brain subregions. These clusters aligned with known cytoarchitectural features, reinforcing how epigenetic modifications and gene expression programs translate into functional territories within the brain. Moreover, the spatially resolved expression of canonical marker genes—such as Prox1, Satb1, and Bcl11b—emerged with distinct regional specificity, linking transcription factors to their respective neuroanatomical contexts.
Prox1, a transcription factor pivotal for neurogenesis and granule cell maintenance, was predominantly expressed in the DG. Satb1, implicated in cortical neuron differentiation and laminar organization, localized strongly within the cerebral cortex. Meanwhile, Bcl11b, essential for neuronal progenitor differentiation, was enriched in the hippocampus, further validating the spatial-DMT findings with established biological knowledge. These expression patterns parallel methylation modifications at their gene loci, suggesting a tight interplay between methylation states and transcriptional regulation.
Importantly, the regulatory roles of mCG and mCA methylation were dissected by correlating their spatial levels with gene expression changes for signature genes in distinct brain clusters. This analysis uncovered gene-specific epigenetic regulation modalities. For instance, Prox1 and Bcl11b expression levels showed significant associations with both mCG and mCA methylation, indicating a dual-modification regulatory mechanism that may fine-tune gene activity in DG and CA1/2 regions. Contrastingly, Ntrk3, a gene encoding a receptor tyrosine kinase with critical roles in neuronal signaling, exhibited expression patterns strongly correlated with mCG methylation but comparatively independent of mCA status.
Satb1 expression in the cortex displayed a similar pattern, predominantly linked to mCG methylation without a clear association with mCA. These findings highlight the versatility and gene-specific nature of methylation-mediated gene regulation. Intriguingly, the silencing of Cux1, a transcription factor involved in neuronal development, showed complex relationships: in the CA3 region, negative correlations with both mCG and mCA hypermethylation were observed, while in CA1/2, its expression correlated negatively only with mCA hypermethylation. This underscores the nuanced epigenetic control that can vary dramatically within neighboring brain regions.
Across the spectrum of genes analyzed, negative correlations between DNA methylation and gene expression were more frequent than positive correlations, reflecting the generally repressive influence of DNA methylation on transcription. Yet, the presence of positive correlations in certain contexts highlights the multifaceted roles methylation may play, including potentially activating or facilitating transcription under specific genomic or cell-type landscapes. The spatial-DMT data thus unravel a complex regulatory milieu where epigenetic modifications contribute heterogeneously to gene expression dynamics.
Delving deeper into cellular heterogeneity, the researchers characterized methylation and transcription patterns across neuronal and glial populations. Genes such as Syt1 and Rbfox3 were found expressed broadly across neurons in various cortical layers, consistent with their fundamental neuronal roles. Conversely, the expression of Cux1, Cux2, and Satb2 was enriched in upper cortical layers, while Bcl11b was predominant in deeper layers, illustrating layer-specific regulatory programs at both the epigenetic and transcriptomic levels.
Cell-type specific epigenetic markers also emerged, evidenced by oligodendrocyte genes Mbp and Plp1 and fibrous astrocyte marker Gfap showing selective enrichment in corpus callosum and hippocampal regions. This finding elucidates how spatial-DMT can capture the spatial distribution of glial populations and their distinct epigenetic landscapes, an area traditionally challenging to resolve given the complex cellular milieu of the brain.
To enhance cellular resolution further, spatial-DMT datasets were integrated with reference single-cell RNA sequencing (scRNA-seq) data. This facilitated the mapping of spatial clusters to predefined cell types, confirming the identities of oligodendrocyte-enriched clusters (W0, W3) and telencephalon-projecting excitatory neurons (W5). The integration demonstrated high correspondence between aggregated gene expression and DNA methylation profiles across pixels and independent single-cell datasets. Thus, spatial-DMT effectively bridges histological structure with molecular identity at near single-cell resolution.
This integrative approach also revealed classical cortical laminar neuron subtypes, such as TEGLU7 to TEGLU3, with layer-specific spatial distributions in agreement with known cortical layers 2/3, 4, 5, and 6. In the hippocampus, TEGLU24 and TEGLU23 mapped neatly to CA1/2 and CA3 regions, respectively, while the granule neuron marker DGGRC2 localized to the DG. Moreover, the thalamic region was enriched with DEGLU1 neurons, further illustrating the fine spatial granularity achievable.
Collectively, these findings highlight spatial-DMT as a pioneering platform that overcomes limitations of conventional single-cell and anatomical methods by providing simultaneous, spatially resolved epigenomic and transcriptomic profiles. Such dual-omic spatial resolution offers a powerful window into the molecular architecture of complex tissues, illuminating how epigenetic mechanisms sculpt gene expression landscapes that define cell identities and functional regionalization. As brain-wide epigenomic atlases emerge, approaches like spatial-DMT will undoubtedly accelerate discovery in neurobiology, developmental biology, and disease pathogenesis.
Future applications of this technology promise to elucidate disease-associated epigenetic dysregulation with spatial precision, shedding light on how localized epigenetic anomalies may drive neurological disorders. Moreover, the capacity to distinguish regulatory roles of different DNA methylation contexts (mCG vs. mCA) offers avenues to decode the epigenetic grammar governing neural gene expression. This foundational work sets the stage for extended studies involving other tissues and complex organs, facilitating comprehensive spatial multi-omic profiling that integrates epigenetics, transcriptomics, proteomics, and beyond.
In conclusion, the advent of spatial-DMT represents a paradigm shift in spatial biology, providing an unprecedented integrated view of epigenetic and transcriptional regulation in situ. By revealing intricate, gene- and region-specific methylation-expression relationships, this approach paves the way for deciphering the molecular codes underlying cellular diversity and tissue architecture. With its demonstrated robustness and precision, spatial-DMT is poised to become an indispensable tool for biomedical research, offering transformative insights into biology’s spatial dimension.
Subject of Research: Spatial joint profiling of DNA methylome and transcriptome in brain tissues.
Article Title: Spatial joint profiling of DNA methylome and transcriptome in tissues.
Article References:
Lee, C.N., Fu, H., Cardilla, A. et al. Spatial joint profiling of DNA methylome and transcriptome in tissues. Nature (2025). https://doi.org/10.1038/s41586-025-09478-x
Tags: Cortical and Hippocampal Gene RegulationDNA Methylation in Complex TissuesDual-Omic Spatial Analysesepigenetic regulation and gene expressionMethylation Patterns in NeuronsMouse Brain Development StudiesNon-CpG Methylation in Brain FunctionPostnatal Mouse Brain ResearchSpatial DNA Methylome ProfilingSpatial Heterogeneity in GeneticsSpatial-DMT MethodologyTranscriptome Analysis in Brain Tissue
