[This site contains pages, such as this, to describe your study.  In addition, pre-created "Document Libraries" (much like file folders on a PC) are set up for the storage of raw and analytical datasets, data dictionaries, data preparation documents, and other relevant study documentation.  For help in using the features of this site, along with guidelines for data submission, please consult the Data Submission Guidelines and Help Topics]

AS315 - Epigenetic Mechanisms of PM-mediated CVD Risk


Investigator Names and Contact Information

Eric Whitsel, MD, MPH  [eric_whitsel@med.unc.edu]


Epidemiologic studies, including several within the Women’s Health Initiative (WHI), have linked exposure to ambient particulate matter (PM) air pollution with sub-clinical and clinical cardiovascular disease (CVD).  Although PM inhalation also has been linked with increases in inflammatory, oxidative, endothelial, metabolic, and coagulation biomarkers, the molecular mechanisms producing such effects remain unclear.  We and others have consistently found that PM exposure is associated with altered global and gene-promoter DNA methylation measured in peripheral leukocyte DNA, an environmentally inducible and dynamic epigenetic mechanism that controls gene expression.  DNA methylation has also been associated with CVD, implicating it as a primary molecular mechanism capable of mediating the cardiovascular health effects of environmental exposures.  However, genome-wide DNA methylation has been examined at only one time point relative to a given exposure or outcome and conclusive evidence of temporality is lacking.  We therefore propose a two-stage study of associations between PM air pollution, DNA methylation, and cardiovascular measures among independent subsets of the exam site- and race-stratified, randomly selected 6% minority oversample of approximately 4,300 WHI CT participants who had fasting blood draws and resting, standard, twelve-lead electrocardiograms (ECGs) repeated at three-year intervals from 1993 to 2004.  Stage 1 will focus on the interrogation, discovery and ranking of 450,000 DNA methylation sites potentially sensitive to PM in year 2000 blood samples from 800 of the participants (400 with low and 400 with high PM exposures).  In up to three blood samples collected serially from the remaining 3,500 participants, Stage 2 will focus on the longitudinal replication of the ten most PM-sensitive DNA methylation sites identified by Stage 1, the temporal relationship between PM and DNA methylation at those sites, and that between site-specific DNA methylation and several types of cardiovascular measures.  The proposed epigenetic data analyses will be well-powered and appropriately adjusted for both ancestral admixture and multiple comparisons.  Findings based on them will improve understanding of pathophysiological mechanisms underlying PM-related CVD in post-menopausal women and support inference to the larger, dynamic population of WHI CT participants from which the minority oversample was drawn.  From this perspective, the DNA methylation data generated by the proposed ancillary study may be of substantial value in addressing a variety of hypotheses posed by other investigators.


Specific Aims
I.    Characterize associations between PM exposure and DNA methylation in a genome-wide analysis
      a)               quantify the extent of DNA methylation at approximately 450,000 DNA methylation sites
      b)   rank the PM sensitivity of the DNA methylation sites
II.   Characterize temporal patterns and modification of associations between PM exposure and DNA methylation in a gene-specific analysis
a)      perform replication analyses of the ten most PM-sensitive DNA methylation sites identified by Aim I in an independent subset of WHI participants by measuring the extent of DNA methylation changes at these sites using a highly quantitative, bisulfite-pyrosequencing method
b)      evaluate modification of longitudinal PM effects on site-specific DNA methylation by a diet rich in anti-oxidant micronutrients, as reflected by high and low plasma concentrations of
i.        α- and β-carotene
ii.   α- and γ-tocopherol
iii.  β-cryptoxanthine
iv.  lycopene
v.   lutein                                                                                                       
vi.  zeaxanthin
vii. retinol
III. Characterize temporal patterns of associations between DNA methylation at sites identified by Aims I-II and measures of
a)      lipid metabolism:  TC, HDL-C, TG, LDL-C, HDL-2, HDL-3 and apolipoprotein (a)
b)      glucose homeostasis:  glucose, insulin and HOMA-IR
c)      coagulation:  fibrinogen, factor VII antigen, and factor VII clotting activity
d)      ECG:  heart rate, heart rate variability, conduction, repolarization, ischemia, ectopy and arrhythmia