Tumor recurrence and metastasis remain the principal causes of mortality in patients with head and neck squamous cell carcinoma (HNSCC). However, the key molecular drivers underlying HNSCC progression and their regulatory mechanisms have not yet been fully elucidated. In this study, we investigated the biological role of the long noncoding RNA MIR4435-2HG in HNSCC. Integrated analyses across multiple public databases demonstrated that MIR4435-2HG is significantly upregulated in HNSCC and is closely associated with pathological T stage, clinical stage, and histological grade. Through further bioinformatic screening and functional validation, upstream transcription factor 1 (USF1) was identified as a critical downstream effector of MIR4435-2HG. Mechanistically, MIR4435-2HG directly interacts with USF1 and promotes its protein stability by facilitating USF1 succinylation, thereby enhancing tumor cell migration and invasion through the induction of epithelial - mesenchymal transition. Moreover, dual-luciferase reporter assays and RT-qPCR analyses revealed that CREB1 transcriptionally upregulates MIR4435-2HG expression, establishing a positive feedback regulatory circuit. Collectively, our findings demonstrate that MIR4435-2HG promotes USF1 succinylation and stabilization, leading to the formation of a MIR4435-2HG - USF1-AKT - CREB1 signalling loop that drives epithelial - mesenchymal transition and facilitates the invasive and metastatic potential of HNSCC cells.
Cutaneous melanoma (CM) is highly invasive with limited prognostic tools, and the prognostic relevance of lactylation, a novel post-translational modification, remains unexplored. We developed a lactylation-related prognostic model using relative methylation orderings from The Cancer Genome Atlas (TCGA) melanoma cohort (n = 458), which was validated in two dependent cohorts (n = 241). Comprehensive analyses of immune infiltration, therapy efficacy, drug sensitivity, and tumour mutation burden were conducted between high- and low-risk groups. Results showed that 284 lactylation-related genes collected classified TCGA melanoma patients into two clusters with significant survival differences (Kaplan-Meier log-rank test, p = 0.046). Between the two clusters, 279 differentially methylated loci were identified (false discovery rate < 0.05 and |Δβ| >0.1). Based on these loci, 1,743 locus pairs were significantly associated with overall survival (Univariate Cox regression, false discovery rate < 0.05). From these significant pairs, the CM 11 locus-pair prognostic model (CM-LP11) consisting of 11 pairs was constructed, which could effectively stratify patients into high- and low-risk groups with significantly different overall survival across all datasets (all p < 0.05, log-rank test). The high-risk group exhibited an immunosuppressive microenvironment with reduced CD8+ T cells and B cells and increased monocytes (p < 0.05, Wilcoxon test). Functional analysis of methylation differences between risk groups showed significant immune response-related pathways such as antigen processing and presentation. The high- and low-risk patients also exhibited distinct sensitivity to cytotoxic chemotherapy, targeted therapy, and immunotherapy, with the low-risk group showing higher immunophenoscore, cytolytic activity, and tumor mutational burden. CM-LP11 represents a robust lactylation-related prognostic biomarker that predictive survival outcomes and therapeutic responses in CM patients, proving insights for personalized treatment strategies.
Colorectal cancer (CRC) is a leading cause of cancer-related deaths worldwide. Although epigenetic alterations are common in CRC, the epigenetic changes that occur during colorectal carcinogenesis remain unclear. Thus, we sought to elucidate the role of NOS1 methylation in colorectal carcinogenesis, which is essential for understanding disease pathology. We used the UCSC Xena database to comparatively analyze NOS1 expression and methylation status between tumour and adjacent normal tissues across 33 cancer types. Low expression and hypermethylation of NOS1 have been identified in 12 cancer types, with CRC demonstrating this characteristic epigenetic regulation. NOS1 promoter methylation status was examined using genomic DNA extraction and MassARRAY EpiTYPER methylation analysis. NOS1 hypermethylation was confirmed among CRCs in in-house dataset 1 (p < 0.001), and among CRCs and advanced adenomas in in-house dataset 2 (p < 0.001). An upward trend in methylation changes was identified from non-advanced adenoma to advanced adenoma to CRC (p for trend < 0.001). Quantitative real-time PCR was used to analyze NOS1 expression in in-house Dataset 3, and significantly low NOS1 expression was also identified in CRCs (p < 0.001). Kaplan-Meier estimator and Cox proportional hazard models were used to assess the prognostic and predictive roles of NOS1. NOS1 hypermethylation was significantly associated with better disease-specific survival (DSS) in CRC patients. NOS1 hypermethylation is an epigenetic driver of colorectal tumorigenesis and is associated with better survival.
Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide, yet current surveillance with alpha-fetoprotein (AFP) and ultrasound (US) lacks sufficient accuracy, particularly for early-stage disease. This study aimed to develop and clinically validate a blood-based digital PCR (dPCR) assay targeting novel methylation biomarkers for HCC detection. Genome-wide methylation profiles from HCC, normal, and other cancer types were analysed to identify candidates, which were then screened and verified in cancer cell lines, primary tumour tissues, and plasma specimens. A total of 186 plasma samples were evaluated, including 66 healthy controls, 60 patients with chronic liver disease (CLD) without HCC, and 60 patients with HCC. FAR1 methylation emerged as an HCC-specific biomarker, showing significantly higher levels in liver cancer cell lines and HCC tumour tissues compared with controls. A dPCR assay targeting FAR1 achieved 70.0% sensitivity (42/60; 95% CI, 56.8%-81.2%) and 96.8% specificity (122/126; 95% CI, 92.1-99.1%), with an area under the curve (AUC) of 0.90 (95% CI, 0.85-0.96). Combining FAR1 with AFP improved sensitivity to 88.3% (53/60; 95% CI, 77.4%-95.2%) while maintaining 96.8% specificity. Notably, early-stage HCC (Barcelona Clinic Liver Cancer [BCLC] stage 0-A) was detected with 85.4% sensitivity (35/41; 95% CI, 77.4%-94.4%). This blood-based dPCR assay demonstrated high diagnostic performance, underscoring its potential as a noninvasive tool to enhance early detection and clinical management of HCC in high-risk populations.
Epigenetic aging prior to midlife is gaining interest as an intervenable period to address health and cognitive aging. Epigenetic changes may index DNA methylation aging rates, but methylation profiles may not be substitutable across tissues. We compared DNA methylation clocks and age acceleration in saliva, buffy coat (BC), and peripheral blood mononuclear cells (PBMC) collected in 91 individuals (7 unpaired, 20 siblings, 64 twins; 18 monozygotic (MZ), 14 dizygotic (DZ) pairs) from the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging (CATSLife1; Mean age = 30.90 years [range = 28.07-41.13]; 50.5% female). Across 15 DNA methylation clocks, chronological age and DNA methylation ages were moderately associated (mean Spearman correlations: r = 0.37, Saliva; r = 0.40, BC; r = 0.34, PBMC). In mixed-effects models, saliva showed higher DNA methylation ages (B = 3.77-19.72 years vs BC, p < 0.001), whereas PBMC and BC were comparable (B = -0.06-0.39 years vs BC, p ≥ 0.436). The exception was the next-generation clock DunedinPace showing comparability (p = 0.486). Similar patterns were observed for age acceleration estimates. Altogether, MZ pairs (meta-analytic r = 0.49, 95%CI = 0.32,0.66) and DZ pairs (meta-analytic r = 0.38, 95%CI = 0.25,0.52) were moderately correlated (Spearman), but MZ pairs showed heterogeneity across tissues (p < 0.034): saliva was lower (mean r = 0.33, SD = 0.25) than BC (mean r = 0.64, SD = 0.10) and PBMC (mean r = 0.49, SD = 0.18). Next-generation PCGrimAge and DunedinPace clocks showed consistent zygosity correlations across tissues, while multi-tissue clocks (e.g. ZhangQ) showed comparable MZ-DZ correlations. While saliva-based DNA methylation is not a direct substitute for blood-based DNA methylation, BC and PBMC show comparability; nevertheless, all tissue types may be appropriate for DNA methylation aging studies when compared within tissues.
Birthweight is heritable and strongly associated with epigenetic differences at birth. It is unclear whether a genetic birthweight score is associated with DNA methylation (DNAm) and, if so, whether through direct genetic effects (i.e. child genotype) or indirect genetic effects (i.e. parental genotype, independent of child genotype), which are suspected to be mediated by the prenatal environment (e.g. metabolic factors). We constructed polygenic scores (PGS) for birthweight predisposition in mothers and children using an existing ‘pre-adjusted’ GWAS assessing maternal indirect effects (maternal genotype on offspring birthweight, correcting for offspring genotype) or fetal effects (offspring genotype on offspring birthweight, correcting for maternal genotype) in 2116 families from the Generation R Study. Offspring DNAm levels at 829 birthweight-related sites were then regressed on both maternal and fetal effect PGS. Additionally, we aggregated 829 DNAm sites into a methylation profile score for birthweight (MPS-BW) and tested which pregnancy health factors (n = 13) might mediate genetic effects. We identified six DNAm sites associated with maternal indirect effects and the birthweight MPS revealed indirect genetic associations, as well. Paternal genotype was not associated with offspring DNAm in this study but was limited by lower sample size and lack of GWAS on paternal indirect effects, thus requiring a posteriori disentanglement of direct and indirect effects. Gestational age partly mediated maternal indirect effects on DNAm, but no other mediators were identified. Results did not depend on offspring sex. This study presents first evidence of maternal genetic effects on offspring birthweight being associated with offspring epigenetic patterns at birth.
Chronic kidney disease (CKD) stage 5 is frequently accompanied by systemic inflammation, and peripheral blood mononuclear cells (PBMCs) play an important role. To define the epitranscriptomic features of PBMC small RNAs in CKD stage 5, we profiled N6-methyladenosine (m6A) using small RNA modification microarrays. A total of 158 miRNAs, 149 pre-miRNAs, and 197 tsRNAs showed differential m6A modification. Enrichment analysis implicated PI3K-Akt, p53 signalling, and leukocyte transendothelial migration. Integrating target prediction with GEO transcriptomic datasets identified IRF1 and RUNX2 as key targets. MeRIP confirmed reduced m6A in miR-205-3p and miR-93-5p, accompanied by upregulation of RUNX2 and downregulation of IRF1 by qPCR. These results define an altered m6A-modification profile of PBMC small RNAs in CKD stage 5 and highlight miRNA-target gene axes with potential biomarker utility.
The complex pathophysiology of essential hypertension (EH) continues to pose challenges to blood control. This study investigated the association of blood DNA methylation and gene expression with EH in blood and arterial tissues to identify potential pathogenic methylation sites, genes, and mechanisms. We identified 767 methylation sites, 159 blood genes, and 111 arterial genes associated with EH. Among these, 41 genes were consistently associated with EH in the same direction in both blood and arterial tissues. Of these, eight genes (FDFT1, FES, GNL3, NME6, SLC22A5, UBA7, ULK3, and ZNF589) were annotated from the EH-relevant DNA methylation sites. Drug target analysis identified 64 compounds targeting FDFT1, FES, ULK3, and SLC22A5. Integration of summary data-based mendelian randomization (SMR) results revealed 72 and 79 significant DNA methylation-related, gene-proximal regulatory pathways associated with EH in single blood tissue and cross-tissue analyses, respectively. Notably, methylation sites of the FES significantly influenced changes in FES expression in both blood (cg03209642: b_meSMR = -2.199; cg05211768: b_meSMR = -1.613) and arterial tissue (cg06330618: b_meSMR = -0.572), influencing the risk of EH (FES in blood: b_eSMR = -0.013; arterial tissues: b_eSMR = -0.029). The eight methylated genes are associated with the risk of EH in both blood and arterial tissue. NME6, SLC22A5, GNL3 and UBA7 have been found to be associated with EH. The methylation-mediated regulation of FES may represent a potential new pathway in the blood and arterial tissues. These findings offer valuable insights for the development of EH biomarkers and therapeutic strategies.
According to the omnigenic model for the inheritance of common traits proposed by Jonathan Pritchard’s group in 2017, there are a very large number of genetic variants underlying common traits spread across the genome. Causal variants are of two types: core genes that have direct roles in disease pathogenesis that contribute only a small amount to total heritability and peripheral genes that are much more in number than core genes and that cumulatively contribute maximally to trait heritability. The same group proposed in 2019 that peripheral genes have indirect effects by acting in trans. The authors propose that most heritability is due to weak trans-expression quantitative trait loci (trans-eQTL) single nucleotide polymorphisms (SNPs) whose effects are mediated by peripheral genes that influence the expression of core genes. In 2018, we provided evidence for an epigenetic-based omnigenic model for the inheritance of common psychiatric disorders. In the current article, we discuss the possible role of genes underlying trans-acting factors, namely, non-coding RNAs and transcription factors, in the pathogenesis of common psychiatric disorders.
Proper adjustment for population substructure is essential in epigenome-wide association studies (EWAS), particularly in cohorts with diverse ancestries. EPISTRUCTURE offers a genotype-free approach to ancestry inference, originally developed using a European reference population from the Cooperative Health Research in the Region of Augsburg (KORA) study. However, its effectiveness in genetically diverse, multi-ancestry cohorts remains insufficiently evaluated. For EWAS using cord-blood samples from the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study, we systematically assessed the ancestry adjustment performance of EPISTRUCTURE principal components (PCs) derived from the widely used KORA-based reference set versus new reference sets generated from genotyping data of the multi-ancestry HAPO cohort. HAPO-based reference sets were defined by varying SNP - CpG R2 thresholds (e.g. RS30: R2>0.3) to identify ancestry-informative CpGs. We applied these reference sets for population substructure adjustment in EWAS of three newborn adiposity traits, birthweight, cord C-peptide, and sum of skinfolds, to evaluate their impact on association detection and biological interpretation. Compared to the KORA reference, the HAPO RS30 reference consistently produced lower genomic inflation and identified more biologically relevant associations for birthweight and cord blood C-peptide in EWAS of HAPO cord blood samples (n = 3,116). Pathway enrichment analyses revealed strong immune and metabolic signals, including pathways uniquely captured by EWAS when using the HAPO-derived reference for ancestry adjustment. Trait enrichment using the EWAS Catalog further confirmed associations with fetal growth, maternal metabolic traits, and glucose regulation. Our findings demonstrate that reference sets derived from multi-ancestry cohorts like HAPO better capture underlying population substructure and improve ancestry adjustment in diverse EWAS settings.
Cutaneous melanoma, a highly aggressive and therapy-resistant skin cancer, is characterized by its remarkable cellular plasticity, enabling tumour cells to switch between different phenotypic states. This plasticity contributes to tumour heterogeneity and is regulated by key transcription factors. Long non-coding RNAs (lncRNAs) are emerging as crucial regulators in melanoma progression, yet much remains to be explored regarding their role in phenotype switching. In this study, we analysed long non-coding RNAs (lncRNAs) across different murine melanoma cell lines, identifying a set of lncRNAs potentially involved in regulating melanoma phenotypic state through cis-regulation of neighbouring protein-coding genes. We demonstrated that the lncRNA Dlx4os regulates genes associated with melanoma plasticity, favouring a mesenchymal-like, undifferentiated state. Dlx4os knockdown redirected melanoma cells to a more differentiated and less malignant phenotype, confirmed by differential expression of phenotypic state markers (Sox10, Mitf, Tgfβ, Sox6, Mlana), reduced their invasive and migratory potential, and delayed tumour progression in vivo. Furthermore, we identified a human orthologue of Dlx4os. Our findings highlight the potential of Dlx4os as both a biomarker and therapeutic target, capable of modulating melanoma’s phenotypic plasticity to influence treatment response and metastasis.
Fetal repair of spina bifida aperta (fSBA) is an established intervention that improves neurological and neurodevelopmental outcomes. The present exploratory study examines whether molecular signatures related to stress regulation are detectable in newborns following this procedure. Specifically, we investigated DNA methylation and gene expression of two stress-regulatory genes, NR3C1 and FKBP5. Within a clinical trial (ID: NCT04027374), we analyzed postpartum saliva samples from newborns who had undergone fSBA repair (fSBA group; n = 30) and compared them with two control groups: newborns exposed to antenatal glucocorticoids for lung maturation (LMI group; n = 12) and healthy controls (HC group; n = 27). Pyrosequencing and qRT-PCR were used for epigenetic and transcriptional analyses. Significant group differences were observed in FKBP5 methylation, particularly at intron 7 CpG sites 5-7. The fSBA group showed lower methylation at site 5 but higher methylation at sites 6-7 compared to controls. No significant methylation differences were detected for NR3C1. Conversely, NR3C1 gene expression was elevated in the fSBA group, whereas FKBP5 expression did not differ between groups. These findings suggest gene- and site-specific molecular variation in newborns following fetal surgery. Given the exploratory nature of the study, the results are not suited to draw specific clinical implications but may inform future work aimed at understanding stress-related molecular alterations surrounding fetal interventions. Larger and longitudinal studies are warranted to clarify the robustness, developmental course, and potential clinical relevance of these molecular patterns.
Clinically prior hyperglycemia may lead to long-lasting adverse cardiovascular effects, a process referred to as ‘glycemic memory.’ Epigenetic modifications, specifically DNA methylation changes, may play a key role in this phenomenon. This study investigated if prior high glucose delivery to cardiomyocytes, led to worsened cardiovascular effects upon pressure overload and to ascertain the gene expression and corresponding DNA methylation signatures linked to glycemic memory. Using inducible and cardiomyocyte-specific glucose transporter 4 (GLUT4) overexpressing mice. We induced glucose delivery for 2 weeks, then returned to basal uptake for 2 weeks, followed by sham or transverse aortic constriction (TAC) surgery as a secondary stress. Mice were followed for an additional 8 weeks and assessed for contractile function, cellular remodeling, and molecular changes. TAC led to an exacerbated hypertrophic response and cardiac dysfunction in the transgenic mice. Subsequent analysis identified molecular changes akin to heart failure, worsened cardiac fibrosis, and oxidative stress. Using bulk RNA-sequencing and reduced representation bisulfite sequencing, we discovered differential gene expression and DNA methylation signatures that persisted even after cellular glucose levels reverted to normal. Significant changes across both expression and methylation-identified enriched pathways related to adverse cardiac events, supporting a glycemic memory response. Glycemic memory led to cardiac structural and functional exacerbation, mimicking heart failure, when subjected to a secondary stress. Our data identified transcriptome, and preliminary DNA methylome changes which may potentially be molecular signatures of future therapeutic targets associated with this heart failure susceptibility resulting from enhanced glucose delivery.
Advancing age and deregulated obesity are well-established risk factors for breast cancer risk. While epigenetic ageing, a measure of biological ageing, has been linked to breast cancer risk, prior studies have mainly focused on blood-based measures. We aimed to study well-established epigenetic age estimates in normal breast tissue in relation to obesity-related metabolic markers. We analysed breast tissue from 91 cancer-free women, using DNA methylation age (mAge) clocks, including first-generation clocks (Hannum, Horvath), mortality-related clocks (Grim, Pheno), and methylation-based telomere length (DNAmTL). Pearson correlations assessed relationships among the clocks. Linear regression was used to associate mAge clocks with obesity-related blood markers (leptin, adiponectin, IGF-1, IGFBP-3), adjusting for chronological age, race, body mass index (BMI), and smoking, applying a false discovery rate threshold of 0.1. We explored effect modification by menopausal status using interaction terms. In the normal breast tissue, the strongest correlation of mAge estimates was observed between Horvath-mAge and Grim-mAge (r = 0.85). BMI was positively associated with all mAge estimates except Pheno-mAge. Older Grim-mAge and longer DNAmTL were associated with higher levels of adiponectin, IGF-1, and IGFBP-3, while older Horvath-mAge was associated only with IGF-1. Associations of DNAmTL with IGF-1 and adiponectin differed statistically by menopausal status. Epigenetic age in breast tissue, particularly Grim and DNAmTL, is associated with obesity-related metabolic markers, independent of chronological age, BMI, and smoking. Although limited by a moderate sample size, our findings indicate a potential biological pathway linking mAge-related metabolic dysregulation that may provide insight into breast cancer risk.
Ribosomal DNA copy number (rDNAcn) and DNA methylation are important modulators of the human genome, both studied in relation to overall cellular function, biological ageing, and disease development. Despite the overlapping roles, their relationship remains poorly understood, especially in the early stages of life, characterized by rapid growth and high cellular demands. Even though previous studies have associated rDNA methylation with cancer and ageing, no study to date has examined the interplay between rDNAcn and whole-genome DNA methylation. In an epigenome-wide association study of 45S rDNAcn variation in 194 newborns, we show strong positive associations between rDNAcn and single DNA methylated CpGs, measured with the Illumina EPIC array. Out of the 122 Bonferroni-significant CpGs, 63.5% were also Bonferroni-significant in a replication cohort of 167 newborns, in which a second EWAS was conducted using DNA methylation data from the Illumina 450K array. The identified CpGs were dispersed over the autosomes and were not functionally related to the rDNA-forming nucleolar-associated domains. The top CpGs were annotated to genes (GFI1, USP46, ABHD14B, CHL1, CGREF1) that are functionally linked to cancer and cellular proliferation. In downstream analyses, the 122 rDNAcn-related CpGs revealed 31 differentially methylated regions and 253 nominally significant correlations with cord blood gene transcripts in an eQTM analysis. Pathway enrichment analyses showed an overrepresentation of the following pathways: 'RNA Polymerase III transcription' (R-HSA-76071, R-HSA-76046, R-HSA-74158, R-HSA-749476, R-HSA-73780, R-HSA-73980, R-HSA-76066, R-HSA-76061, hsa03020), ‘cytosolic sensors of pathogen-associated DNA’ (R-HSA-1834949), ‘RNA polymerase II transcribes snRNA genes’ (R-HSA-6807505), and 'translation initiation' (R-HSA-72613, R-HSA-72737). Our findings reveal a close link between rDNAcn variation and DNA methylation in early life. Disruptions in this interplay may influence cellular functions critical for early development, potentially shaping health and disease trajectories later in life.
In recent years, the incidence of gestational diabetes mellitus (GDM) has been steadily increasing, posing risks to the long-term health of both mother and child. We aim to characterize blood glucose levels of pregnant women and predict the risk of GDM during early pregnancy through plasma cell-free mRNA and non-coding RNA (cfRNA). Here, we collected plasma samples from 108 pregnant women (54 with GDM and 54 controls) at around 16 weeks of gestation. Following high-throughput sequencing, we performed differentially abundant genes analysis and evaluated correlations between cfRNA profiles and blood glucose levels. Based on these findings, we developed a predictive model utilizing cf-mRNA and cf-lncRNA signatures. We found that ribosomal genes (RPL/RPS) are decreased in GDM, negatively correlated with 1hGlu and 2hGlu, and enriched in protein synthesis metabolic pathways. Additionally, placental-derived cfRNA contributed less to plasma in GDM, with placental-specific gene IGF2 significantly negatively correlated with blood glucose. Furthermore, 35 blood glucose correlated-cfRNA genes accurately predict GDM, with area under the curve of 0.84 in internal testing and 0.73 in external validation cohort. Our study reveals significant alterations in protein metabolic pathways and placenta-derived RNAs in plasma cfRNA prior to GDM diagnosis.
The transplantation of neural progenitor cells derived from induced pluripotent stem cells (iPSCs) has therapeutic potential for the treatment of neurological diseases. However, the functional integration of transplanted iPSC-derived neurons into host neural networks remains controversial. Optogenetic and chemogenetic tools offer the means to assess such integration. However, constructing modifiable iPSC-derived neurons requires efficient gene editing. Here, we used CRISPR/Cas9 (targeting the AAVS1 safe harbor) and PiggyBac transposon systems to insert optogenetic and chemogenetic receptors (ChR2/hM4Di) into human iPSCs. While both systems successfully integrated genes into the genomes of HEK293T cells and iPSCs, receptor expression was detected only in HEK293T cells. Bisulfite sequencing revealed extensive methylation of the TRE3G BI promoter (95.3-98.2%) in iPSCs, in contrast to low methylation (5.9%) in HEK293T cells. For PiggyBac, the methylation of CMV/EF1α promoters in iPSCs exhibited integration site-dependent variability (0-95.2%). Notably, even hypomethylated clones failed to show gene expression, suggesting that additional regulatory mechanisms, such as histone modifications or chromatin remodeling, may contribute to transcriptional silencing. Differentiation into neural stem cells does not reverse methylation nor restore protein expression. Our findings demonstrate that the CRISPR/Cas9 and PiggyBac systems enable the integration of optochemical receptor genes into iPSCs. However, promoter methylation or other epigenetic and non-epigenetic gene-silencing mechanisms could pose barriers to efficient protein expression from the integrated transgene in iPSCs.
Heart failure arises from maladaptive remodelling driven by genetic and epigenetic networks. Using a systems genetics framework, we mapped how DNA variants and CpG methylation shape cardiac transcriptomes during beta adrenergic stress in the Hybrid Mouse Diversity Panel, a cohort of over 100 fully inbred mouse strains. Expression QTLs (eQTLs), methylation QTLs (mQTLs) and methylation-driven eQTLs (emQTLs) were generated from over 13k expressed genes and 200k hypervariable CpGs in left ventricles. We discovered hundreds of regulatory ‘hotspots’ that control large portions of the genome, including several that regulate over 10% of the transcriptome and/or methylome. Approximately 16% of these hotspots overlapped with prior GWAS or EWAS signals. We focus on a hotspot on chromosome 12 and identify the serpine peptidase inhibitor Serpina3n, as the most likely driver gene in this hotspot. Experimental knockdown of Serpina3n in neonatal rat ventricular cardiomyocytes blunted hypertrophy induced by a variety of hypertrophic signals, while altering predicted target expression and modulating the activity of Nppa and Nppb. Together, these findings position Serpina3n as a major regulator of stress-responsive cardiac gene programs, highlighting how integration of genetic and epigenetic signals can pinpoint key drivers of heart failure.
DNA methylation (DNAm) signatures are highly cell type-specific, yet most epigenome-wide association studies (EWAS) are performed on bulk tissue, potentially obscuring critical cell type-specific patterns. Existing computational tools for detecting cell type-specific DNAm changes are often limited by the accuracy of cell type deconvolution algorithms. Here, we introduce CEAM (Cell-type Enrichment Analysis for Methylation), a robust and interpretable framework for cell type enrichment analysis in DNA methylation data. CEAM applies over-representation analysis with cell type-specific CpG panels from Illumina EPIC arrays derived from nuclei-sorted cortical post-mortem brains from neurologically healthy aged individuals. The constructed CpG panels were systematically evaluated using both simulated datasets and published EWAS results from Alzheimer’s disease, Lewy body disease, and multiple sclerosis. CEAM demonstrated resilience to shifts in cell type composition, a common confounder in EWAS, and remained robust across a wide range of differentially methylated positions, when upstream modeling of cell type composition was modeled with sufficient accuracy. Application to existing EWAS findings generated in neurodegenerative diseases revealed enrichment patterns concordant with established disease biology, confirming CEAM’s biological relevance. The workflow is publicly available as an interactive Shiny app (https://um-dementia-systems-biology.shinyapps.io/CEAM/) enabling rapid, interpretable analysis of cell type-specific DNAm changes from bulk EWAS.
DNA methylation (DNAm), capturing biological gestational age (GA) and epigenetic gestational age acceleration (EGAA), can be modified by environmental exposures. The Asthma&Allergy array is a new DNAm array developed with content focused on asthma and allergy loci. The association between content on the Asthma&Allergy array and chronological GA and EGAA has not been evaluated alone or in the context of perinatal exposures. We performed an epigenome wide association study(EWAS) based on chronological GA at single CpG sites and regions. We further constructed a multi-CpG site methylation model to predict chronological GA in cord blood from 391 newborn children from a Detroit-based birth cohort. Associations between perinatal environmental factors with GA, epigenetic gestational age (EGA), and EGAA were assessed. We identified 2,435 CpG sites associated with chronological GA. HLA class II (HLA-DRB1,HLA-DQB1,HLA-DRB6) were the most significantly associated with chronological GA. Our multi-CpG site model attained predictive accuracy (cross-validated Pearson's correlation=0.75) comparable to other EGA methods. Using genes implicated in region-based analyses (n=395 regions), the pathways most significantly enriched with chronological GA-associated CpGs included T helper 1(Th1) and 2(Th2) activation, macrophage classical activation, and IL10 signaling, which were also enriched in at least one of the other published epigenetic clocks. In multi-exposure models, prenatal indoor pet exposure and unplanned C-section were associated with EGA deceleration, while infant's first-born status was associated with EGAA. Our findings highlight enrichment for T cell modulated pathways and antigen presentation as biological processes enriched in chronological GA, as well as novel perinatal factors that may impact EGAA.
