Poster Presentation Australian Epigenetics Alliance Conference 2022

Sequence-dependent allele-specific methylation is causally associated with body fat traits in Norfolk Island genetic isolate (#104)

Luu Phuc-Loi 1 , Rodney Lea 2 , Larisa Haupt 2 , Cao May Van 2 , Heidi Sutherland 2 , Susan Clark 1 , Lyn Griffiths 2
  1. Epigenetics Research Laboratory, Garvan Institute of Medical Research, Sydney, NSW, Australia
  2. Centre for Genomics and Personalised Health, School of Biomedical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia

Background: Sequence-dependent allele-specific methylation (SD-ASM) occurs at heterozygous genetic variants and can have specific effects on gene expression and disease risk. Isolated populations can offer an ideal natural experimental setting for studying SD-ASM in humans due to reduced genetic and environmental diversity and the existence of multi-generational pedigrees. In this study, we characterize whole methylome SD-ASM in the Norfolk Island genetic isolate and explore causal links to body fat traits.

Methods: This study focused on blood DNA of 67 subjects comprising a 3-generation pedigree from the Norfolk Island genetic isolate. The full spectrum of SD-ASM in the pedigree was determined using whole genome sequencing (WGS) and whole genome bisulphite sequencing (WGBS) data obtained from the Illumina XTen platform. SD-ASM is called using 200 bp windows surrounding a heterozygous SNP across WGBS with at least 30% difference between the methylation and unmethylation state. Gene expression and disease phenotype data were also available for a subset of the subjects and were used to uncover causal links to different phenotypes.

Results: We identified 59,245 SD-ASM loci (FDR<0.05). Enrichment analysis showed that SD-ASM occurs mostly in gene regulatory regions (p<0.05). Furthermore, SD-ASM was enriched at GWAS loci for multiple disease traits including BMI and waist-hip index. To explore this further, we tested the causal relationship of genotype, expression, and methylation to BMI, and weight phenotypes in the pedigree. Seven SD-ASM loci exhibited significant mQTL and eQTL effects that were correlated with BMI and weight. One locus in METRNL was causally linked to both phenotypes. This finding is interesting in relation to body fat because METRNL is involved in the hormonal regulation of brown fat cell differentiation and energy homeostasis via glucose tolerance.

Conclusion: This study characterizes methylome-wide SD-ASM in Norfolk Island isolate and indicates that some are causally associated with body fat phenotypes.