Epigenomic profiling of isolated blood cell types reveals highly specific B cell smoking signatures and links to disease risk

Abstract

Background Tobacco smoking alters the DNA methylation profiles of immune cells which may underpin some of the pathogenesis of smoking-associated diseases. To link smoking-driven epigenetic effects in specific immune cell types with disease risk, we isolated six leukocyte subtypes, CD14+ monocytes, CD15+ granulocytes, CD19+ B cells, CD4+ T cells, CD8+ T cells, and CD56+ natural killer cells, from whole blood of 67 healthy adult smokers and 74 nonsmokers for epigenome-wide association study (EWAS) using Illumina 450k and EPIC methylation arrays. Results Numbers of smoking-associated differentially methylated sites (smCpGs) at genome-wide significance (p < 1.2 × 10−7) varied widely across cell types, from 5 smCpGs in CD8+ T cells to 111 smCpGs in CD19+ B cells. We found unique smoking effects in each cell type, some of which were not apparent in whole blood. Methylation-based deconvolution to estimate B cell subtypes revealed that smokers had 7.2% (p = 0.033) less naïve B cells. Adjusting for naïve and memory B cell proportions in EWAS and RNA-seq allowed the identification of genes enriched for B cell activation-related cytokine signaling pathways, Th1/Th2 responses, and hematopoietic cancers. Integrating with large-scale public datasets, 62 smCpGs were among CpGs associated with health-relevant EWASs. Furthermore, 74 smCpGs had reproducible methylation quantitative trait loci single nucleotide polymorphisms (SNPs) that were in complete linkage disequilibrium with genome-wide association study SNPs, associating with lung function, disease risks, and other traits. Conclusions We observed blood cell-type-specific smCpGs, a naïve-to-memory shift among B cells, and by integrating genome-wide datasets, we identified their potential links to disease risks and health traits.

Epigenetic age oscillates during the day

“…Next, we asked if these oscillating WBC subtypes could have contributed to the observed periodicity of epigenetic age. To this end, we reanalyzed two datasets (Reinius et al, 2012;Wang et al, 2023) containing samples of whole blood as well as six sorted WBC subtypes and estimated their epigenetic ages using Horvath pan-…”

Waterpipe and cigarette epigenome analysis reveals markers implicated in addiction and smoking type inference

“…Eleven studies were retrieved: four on cord blood ( Howe et al, 2019 , Bergens et al, 2019 , De Queiroz Andrade et al, 2021 , Danielewicz et al, 2021 ), one on saliva and peripheral blood leukocytes ( Barcelona et al, 2019 ), one on leukocyte subtypes ( Wang et al, 2023 ), one on bronchoalveolar cells ( Ringh et al, 2019 ), one on nucleus accumbens brain cells ( Markunas et al, 2021 ), two on peripheral blood ( Christiansen et al, 2021 , Domingo-Relloso et al, 2020 ), and one study was irrelevant ( Osborne et al, 2020 ). The two studies involving 850 K profiling on peripheral blood ( Christiansen et al, 2021 , Domingo-Relloso et al, 2020 ) are not yet reported in the EWAS data hub, and, among them, the study by Christiansen et al ( Christiansen et al, 2021 ) was used as our 850 K reference because it conducted a meta -analysis of multiple cohorts, and already covered the second study by Domingo-Relloso et al .…”

Epigenomic insights into common human disease pathology

“…By contrast, the tobacco-related hypomethylation of GPR15 cg19859270 is lymphocyte specific [ 277 ]. Furthermore, regarding cellular proportions, smokers are observed to have a 7.2% reduction in naïve B cells (CD19 + ) [ 278 ].…”