Three-dimensional genome organization is crucial for gene regulation and cell identity. The architectural proteins CTCF and the cohesin complex mediate chromatin looping, shaping cell-type-specific transcription. While their functions have been extensively characterized in embryonic stem cells and other proliferative systems, the regulation of transcription by these two ubiquitously expressed factors in terminally differentiated cells, such as endothelial cells (ECs), remains poorly understood. Moreover, interpretation of prior studies is further complicated by the essential requirement of CTCF and cohesin for cell viability, as most rely on near-complete depletion. Consequently, the extent to which CTCF and cohesin function redundantly, cooperatively, or independently to regulate cell-type-specific programs in differentiated cells remains a major knowledge gap. To address this, we reduced CTCF or the core cohesin subunit RAD21 to approximately 50% protein levels in an isogenic system of human induced pluripotent stem cells (iPSCs) and iPSC-derived endothelial cells (iEcs), enabling direct comparison between undifferentiated and lineage-committed cell states. CUT&Tag and RNA sequencing revealed a striking, EC-specific loss of CTCF and RAD21 occupancy at gene promoters following partial depletion; RAD21 reduction had a more pronounced impact on endothelial gene expression than CTCF depletion. RAD21 loss also led to a marked redistribution of FLI1 binding, a transcription factor essential for endothelial identity and function. Collectively, our findings highlight critical contributions of RAD21 in EC-specific gene expression through coordinated control of chromatin architecture and modulating transcription factor occupancy, providing mechanistic insight into cohesin-dependent regulation of vascular transcriptional programs.
Partially methylated domains (PMDs) represent extensive genome-wide hypomethylated regions that drive genomic instability and proto-oncogene activation. However, their complex regulatory influence on tumor-specific transcriptional networks remains largely elusive. In this study, we developed a robust computational pipeline to characterize aberrant PMD landscapes across pan-cancer. We specifically focused on the epigenetic features associated with transcriptional dysregulation in esophageal squamous cell carcinoma (ESCC). By analyzing gene expression patterns, we classified PMDs into four distinct classes and revealed that several immunosuppressive genes were recurrently activated within Tumor-PMDs across various cancers. While these PMDs were generally marked by elevated repressive H3K27me3 and diminished active histone modifications, the activated genes within them paradoxically exhibited a precise, localized enrichment of H3K9ac and H3K27ac near their transcription start sites (TSSs). This localized epigenetic rewiring suggests a potential correlation where collaborative histone modification landscapes may shape oncogenic chromatin remodeling to license transcriptional dysregulation. Furthermore, we identified a robust co-expression axis between the prognostic gene ATP6V0D2 and established immune checkpoints (HAVCR2, TDO2). Elevated ATP6V0D2 expression was found to potentially contribute to ESCC progression and poor prognosis by activating the complement system and epithelial–mesenchymal transition (EMT) pathways. Capitalizing on these findings, virtual screening and in vitro validation suggested that caffeic acid could act as a potential ATP6V0D2-binding compound, effectively suppressing the migration and viability of ESCC cells. This study further nominates caffeic acid as a candidate compound for further mechanistic and therapeutic investigation in ESCC.
Ketamine produces rapid antidepressant effects, but the molecular processes associated with their persistence remain poorly defined in humans. We examined longitudinal DNA methylation changes in 10 adults with major depressive disorder undergoing ketamine-assisted psychotherapy using two pre-treatment assessments (T1, T2) and one post-treatment assessment (T3). Participants received three sessions of intranasal racemic ketamine within a structured psychotherapy protocol. Buccal DNA methylation was measured using the Illumina MethylationEPIC v2.0 BeadChip. Depressive symptoms improved markedly following treatment, with mean Beck Depression Inventory–II scores declining from the severe to the mild range (p < .001). Of 802,595 CpG loci evaluated, 20,770 met the nominal T3-versus-T2 screening criterion. Sequential filtering for methylation-change magnitude and movement relative to pre-treatment fluctuation identified 962 retained CpG loci, including 517 retained gene-annotated loci. A supplementary linear mixed-effects analysis modeled repeated measurements within participants across the 603 preselected gene-annotated magnitude-filtered loci; because these loci were selected using the same contrast and cohort, this analysis was not treated as independent confirmation. Exploratory enrichment analysis of this 603-locus set identified NOTCH1-related Reactome terms driven by NOTCH1, DLL1, and ADAM17. Within the retained signature, multiple core and adjacent Notch-related components formed a coherent signaling module spanning ligand, receptor processing, transcriptional regulation, and pathway cross-talk. These findings identify a structured peripheral methylation signature associated with ketamine-assisted psychotherapy. They do not establish brain-specific pathway activation or causality, but provide a longitudinal basis for investigating regulatory adaptations accompanying clinical improvement.
Chromatin-associated RNAs (caRNAs) serve as central hubs for regulation within the cell nucleus and are involved in key biological processes such as transcriptional homeostasis, chromatin remodeling, and transgenerational inheritance. In recent years, the development of various genome-wide RNA-chromatin interaction capture technologies has made it possible to resolve the spatial interaction networks between caRNAs and chromatin with high precision. This paper systematically reviews the concepts, functions, and interaction patterns of caRNA, and provides a comprehensive overview of the principles, characteristics, technical biases, and applicable scenarios in the fields of targeted and non-targeted identification technologies. Building on this foundation, this paper conducts a comparative analysis of caRNA regulatory mechanisms in animals and plants across four functional dimensions: the organization of three-dimensional chromatin topology, RNA modification-mediated epigenetic regulatory switches, the transmission of environmental and transgenerational signals, and the pathological mechanisms associated with caRNA network dysregulation. Through this comparative analysis, this paper reveals the common characteristics of caRNA-mediated functional regulation in animals and plants, as well as lineage-specific adaptive strategies, and discusses the challenges and future directions in this field regarding single-cell resolution, AI-assisted data analysis, and translational applications.
Epigenetic clocks have been widely used to estimate biological age across various tissues, but their accuracy in breast tissue remains suboptimal. Classical models such as Horvath's and Hannum's clocks perform poorly in predicting chronological age in breast tissue, underscoring the need for a tissue-specific approach. In this study, we introduce a Breast Tissue-specific Epigenetic Clock (BTEC), developed using DNA methylation data from 553 healthy breast tissue samples across seven different studies. BTEC significantly outperformed existing clocks, demonstrating superior correlation with chronological age (r = 0.91) and lower prediction errors (MAE = 3.0 y) without requiring dataset-specific regressions adjustments. Notably, BTEC showed consistent performance across ancestry groups, in contrast to existing clocks which exhibited variable ancestry-related biases. We also found that residual-based epigenetic age acceleration, commonly used to correct for prediction bias, is highly dependent on the choice of reference dataset and does not provide a generalizable correction. BTEC's chronological age predictions for tumor-adjacent samples showed distortions, with an average deviation of -2.3 y, which was even more pronounced in tumor samples, where the average difference between predicted and chronological age was -14 to -15 y. When analyzed by molecular subtype, the distortion was greater in the more aggressive HER2+ and TNBC tumors compared to HR+ tumors. Overall, our findings indicate that breast tumors do not generally exhibit accelerated epigenetic aging. The probes used by BTEC were associated with known oncogenes such as JAK1, TP73, PDGFRA, CCND1 and STAT5B, as well as genes involved in cancer-related pathways, including Wnt, Hedgehog and ErbB signaling. Importantly, extreme deviations in epigenetic age, as measured by BTEC-derived epigenetic age acceleration (EAA), were associated with patient survival in HR+ breast cancer patients in an exploratory analysis, highlighting the prognostic value of tissue-specific epigenetic aging measures. These findings demonstrate that BTECs not only improve age prediction in breast tissue but also capture biologically meaningful alterations in tumor epigenetic aging with potential clinical implications.
