Abstract P260: Obesity is Associated With DNA Methylation in Population-Based Adolescents
Introduction: The association between obesity and DNA methylation in adolescents are extremely limited.
Hypothesis: We hypothesize that gene specific methylations are associated with obesity in population-based adolescents.
Methods: From the population-based Penn State Child Cohort follow-up exam (N=421), we randomly selected 20 obese cases from those with BMI percentile > 95, and 20 nonobese controls from those with BMI percentile < the median (61). White Blood Cell DNA was subjected to enhanced reduced representation bisulfite sequencing providing single nucleotide resolution of DNA methylation in CpG sites and surrounding regions. Briefly, 20 ng of genomic DNA was digested by MspI, which recognizes and cleaves CpG sites, followed by end repair, adenylation and adapter ligation using NEXTflex Bisulfite-Seq Library Prep Kit and NEXTflex Bisulfite-Seq Barcodes with a modification of bead size selection to capture MspI fragments of 70-320 bp size. The resulting libraries were bisulfite-converted by using EZ DNA Methylation Kit, followed by 18 to 20 cycles of PCR amplification using the NEXTflex Bisulfite-Seq U+PCR Master Mix and NEXTflex Primer Mix. Purified and quantified libraries were pooled at 6 samples per sequencing lane and read by 1X50 bp on HiSeq 2500. CASAVA-demultiplexed .fastq files were subjected to downstream analyses. Base calls of bisulfite treated sequencing reads with phred quality scores < 20 were trimmed and the adaptor was cut using trim_galore v0.3.3. Resulting reads were mapped to the hg19 assembly and methylation calls were performed using Bismark v0.10.1. Analysis was performed using custom R scripts and methylKit v0.9.2. Bases with less than 10x coverage were excluded. Bases with fewer than 5 subjects in either the obese or nonobese groups were also excluded. Significance at each base was assessed with t-test. A gene-level analysis was conducted by selecting only bases with a p-value < .05 and identifying genes with the largest number of differentially methylated bases after normalizing by gene length.
Results: We identified 4136 bases with statistically significant differential methylation and an average difference of 25% or more. The genes with the most differentially methylated bases after normalizing to the gene length are GADD45B (8 bases), NKX6-2 (5 bases), UNCX (7 bases), KCNG2 (5 bases). Other significant genes include GNAS (8 bases), KIF26A (8 bases), TBX3 (8 bases), PDGFB (7 bases), B3GALT6 (6 bases), DLG3 (6 bases), GRIN3B (6 bases). Methylkit also identified the following significantly differentially methylated genes:
LINC01444, GPR123, PTP4A3, SH3PXD2A-AS1, ZNF835. One base on the PHYKPL gene maintained significance (p=.01) after multiple correction at the chromosome level.
Conclusion: Obesity is associated DNA methylation in otherwise healthy adolescents. Future analysis will focus on validating significant findings in a larger cohort.
Author Disclosures: A. Berg: None. Y.I. Kawasawa: None. A. Salzberg: None. E.O. Bixler: None. F. He: None. D. Liao: None.
- © 2015 by American Heart Association, Inc.