Impact of tobacco smoking on estimated telomere shortening and epigenetic age acceleration among human blood immune cell types

BACKGROUND: Tobacco smoke exposure has been shown to dramatically alter the methylome of whole blood DNA, and we have previously observed that effects are very different among blood cell types. In the present work, we compare smoking effects on DNA methylation-based telomere length (DNAmTL) shortening, age acceleration (DNAmAA) of 6 DNAmAge clocks (Horvath, Hannum, Skin_Blood, PhenoAge, FitAge, GrimAge2), aging pace (DunedinPACE), and Stochastic Epigenetic Mitotic Timer of Cancer (stemTOC) in six major immune cell types isolated from the whole blood samples of the same individuals from 74 nonsmokers and 69 smokers. RESULTS: Telomere shortening in all cell types was significantly associated with smoking status, with smokers displaying greater telomere shortening. Examining the difference in DNAmTL shortening between smokers and nonsmokers across cell types, CD8+ T cells had the greatest DNAmTL shortening, while CD15+ granulocytes had the least shortening. Among six DNAmAge models tested, the strongest association between DNAmAge Acceleration (DNAmAA) and smoking status was observed for the GrimAge2 model, followed by the FitAge model and then the PhenoAge model, with smokers displaying greater age acceleration. For the GrimAge2 and FitAge models, as expected, smokers revealed greater age acceleration in whole blood samples. Across all cell types, the effect size in epigenetic age acceleration associated with smoking was significantly correlated with smoking effect as measured by mean AHRR cg05575921 demethylation values, with myeloid cell types showing the greatest effect and T cells least. However, the Hannum, Horvath, and Skin_Blood models (which were trained only by chronological age) were generally not significantly associated with smoking status in isolated cell types. We observed the DunedinPACE model was significantly associated with smoking status in all cell types and whole blood samples. Smokers had significantly higher aging pace in all cell types relative to nonsmokers. The stemTOC was not significantly associated with smoking status in isolated cell types. CONCLUSIONS: The present study demonstrates that tobacco smoking is significantly associated with methylation-based measures of telomere length shortening, biological age acceleration, and aging pace in six major immune cell types and whole blood. The effect of smoking on these outcomes differs among cell types, suggesting shifts in cell composition may play an important role in human aging as measured by DNA methylation models and epigenetic aging may play a role in smoking-related diseases.