Longitudinal analysis of epigenome-wide DNA methylation reveals novel smoking-related loci in African Americans

Jiaxuan Liu, Wei Zhao, Farah Ammous, Stephen T Turner, Thomas H. Mosley, Xiang Zhou, Jennifer A. Smith

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Changes in DNA methylation may be a potential mechanism that mediates the effects of smoking on physiological function and subsequent disease risk. Given the dynamic nature of the epigenome, longitudinal studies are indispensable for investigating smoking-induced methylation changes over time. Using blood samples collected approximately five years apart in 380 African Americans (mean age 60.7 years) from the Genetic Epidemiology Network of Arteriopathy (GENOA) study, we measured DNA methylation levels using Illumina HumanMethylation BeadChips. We evaluated the association between Phase 1 smoking status and rate of methylation change, using generalized estimating equation models. Among the 6958 CpG sites examined, smoking status was associated with methylation change for 22 CpGs (false discovery rate q < 0.1), with the majority (91%) becoming less methylated over time. Methylation change was greater in ever smokers than never smokers, and the absolute differences in rates of change ranged from 0.18 to 0.77 per decade in M value, equivalent to a β value change of 0.013 to 0.047 per decade. Significant enrichment was observed for CpG islands, enhancers, and DNAse hypersensitivity sites (p < 0.05). Although biological pathway analyses were not significant, most of the 22 CpGs were within genes known to be associated with cardiovascular disease, cancers, and aging. In conclusion, we identified epigenetic signatures for cigarette smoking that may have been missed in cross-sectional analyses, providing insight into the epigenetic effect of smoking and highlighting the importance of longitudinal analysis in understanding the dynamic human epigenome.

Original languageEnglish (US)
JournalEpigenetics
DOIs
StatePublished - Jan 1 2019

Fingerprint

DNA Methylation
African Americans
Smoking
Methylation
Epigenomics
CpG Islands
Molecular Epidemiology
Longitudinal Studies
Hypersensitivity
Cardiovascular Diseases
Cross-Sectional Studies
Genes
Neoplasms

Keywords

  • DNA methylation
  • epigenome-wide association study
  • longitudinal analysis
  • Smoking

ASJC Scopus subject areas

  • Molecular Biology
  • Cancer Research

Cite this

Longitudinal analysis of epigenome-wide DNA methylation reveals novel smoking-related loci in African Americans. / Liu, Jiaxuan; Zhao, Wei; Ammous, Farah; Turner, Stephen T; Mosley, Thomas H.; Zhou, Xiang; Smith, Jennifer A.

In: Epigenetics, 01.01.2019.

Research output: Contribution to journalArticle

Liu, Jiaxuan ; Zhao, Wei ; Ammous, Farah ; Turner, Stephen T ; Mosley, Thomas H. ; Zhou, Xiang ; Smith, Jennifer A. / Longitudinal analysis of epigenome-wide DNA methylation reveals novel smoking-related loci in African Americans. In: Epigenetics. 2019.
@article{4bc952705c994ecd98376dfeaaf53639,
title = "Longitudinal analysis of epigenome-wide DNA methylation reveals novel smoking-related loci in African Americans",
abstract = "Changes in DNA methylation may be a potential mechanism that mediates the effects of smoking on physiological function and subsequent disease risk. Given the dynamic nature of the epigenome, longitudinal studies are indispensable for investigating smoking-induced methylation changes over time. Using blood samples collected approximately five years apart in 380 African Americans (mean age 60.7 years) from the Genetic Epidemiology Network of Arteriopathy (GENOA) study, we measured DNA methylation levels using Illumina HumanMethylation BeadChips. We evaluated the association between Phase 1 smoking status and rate of methylation change, using generalized estimating equation models. Among the 6958 CpG sites examined, smoking status was associated with methylation change for 22 CpGs (false discovery rate q < 0.1), with the majority (91{\%}) becoming less methylated over time. Methylation change was greater in ever smokers than never smokers, and the absolute differences in rates of change ranged from 0.18 to 0.77 per decade in M value, equivalent to a β value change of 0.013 to 0.047 per decade. Significant enrichment was observed for CpG islands, enhancers, and DNAse hypersensitivity sites (p < 0.05). Although biological pathway analyses were not significant, most of the 22 CpGs were within genes known to be associated with cardiovascular disease, cancers, and aging. In conclusion, we identified epigenetic signatures for cigarette smoking that may have been missed in cross-sectional analyses, providing insight into the epigenetic effect of smoking and highlighting the importance of longitudinal analysis in understanding the dynamic human epigenome.",
keywords = "DNA methylation, epigenome-wide association study, longitudinal analysis, Smoking",
author = "Jiaxuan Liu and Wei Zhao and Farah Ammous and Turner, {Stephen T} and Mosley, {Thomas H.} and Xiang Zhou and Smith, {Jennifer A.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/15592294.2019.1581589",
language = "English (US)",
journal = "Epigenetics",
issn = "1559-2294",
publisher = "Landes Bioscience",

}

TY - JOUR

T1 - Longitudinal analysis of epigenome-wide DNA methylation reveals novel smoking-related loci in African Americans

AU - Liu, Jiaxuan

AU - Zhao, Wei

AU - Ammous, Farah

AU - Turner, Stephen T

AU - Mosley, Thomas H.

AU - Zhou, Xiang

AU - Smith, Jennifer A.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Changes in DNA methylation may be a potential mechanism that mediates the effects of smoking on physiological function and subsequent disease risk. Given the dynamic nature of the epigenome, longitudinal studies are indispensable for investigating smoking-induced methylation changes over time. Using blood samples collected approximately five years apart in 380 African Americans (mean age 60.7 years) from the Genetic Epidemiology Network of Arteriopathy (GENOA) study, we measured DNA methylation levels using Illumina HumanMethylation BeadChips. We evaluated the association between Phase 1 smoking status and rate of methylation change, using generalized estimating equation models. Among the 6958 CpG sites examined, smoking status was associated with methylation change for 22 CpGs (false discovery rate q < 0.1), with the majority (91%) becoming less methylated over time. Methylation change was greater in ever smokers than never smokers, and the absolute differences in rates of change ranged from 0.18 to 0.77 per decade in M value, equivalent to a β value change of 0.013 to 0.047 per decade. Significant enrichment was observed for CpG islands, enhancers, and DNAse hypersensitivity sites (p < 0.05). Although biological pathway analyses were not significant, most of the 22 CpGs were within genes known to be associated with cardiovascular disease, cancers, and aging. In conclusion, we identified epigenetic signatures for cigarette smoking that may have been missed in cross-sectional analyses, providing insight into the epigenetic effect of smoking and highlighting the importance of longitudinal analysis in understanding the dynamic human epigenome.

AB - Changes in DNA methylation may be a potential mechanism that mediates the effects of smoking on physiological function and subsequent disease risk. Given the dynamic nature of the epigenome, longitudinal studies are indispensable for investigating smoking-induced methylation changes over time. Using blood samples collected approximately five years apart in 380 African Americans (mean age 60.7 years) from the Genetic Epidemiology Network of Arteriopathy (GENOA) study, we measured DNA methylation levels using Illumina HumanMethylation BeadChips. We evaluated the association between Phase 1 smoking status and rate of methylation change, using generalized estimating equation models. Among the 6958 CpG sites examined, smoking status was associated with methylation change for 22 CpGs (false discovery rate q < 0.1), with the majority (91%) becoming less methylated over time. Methylation change was greater in ever smokers than never smokers, and the absolute differences in rates of change ranged from 0.18 to 0.77 per decade in M value, equivalent to a β value change of 0.013 to 0.047 per decade. Significant enrichment was observed for CpG islands, enhancers, and DNAse hypersensitivity sites (p < 0.05). Although biological pathway analyses were not significant, most of the 22 CpGs were within genes known to be associated with cardiovascular disease, cancers, and aging. In conclusion, we identified epigenetic signatures for cigarette smoking that may have been missed in cross-sectional analyses, providing insight into the epigenetic effect of smoking and highlighting the importance of longitudinal analysis in understanding the dynamic human epigenome.

KW - DNA methylation

KW - epigenome-wide association study

KW - longitudinal analysis

KW - Smoking

UR - http://www.scopus.com/inward/record.url?scp=85062958241&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062958241&partnerID=8YFLogxK

U2 - 10.1080/15592294.2019.1581589

DO - 10.1080/15592294.2019.1581589

M3 - Article

C2 - 30764717

AN - SCOPUS:85062958241

JO - Epigenetics

JF - Epigenetics

SN - 1559-2294

ER -