DNA Methylome Profiling In Monozygotic Twin Pairs Discordant For Waist-to-hip Ratio

Objective:Obesity is prevalent worldwide and highly associated with a variety of adverse medical conditions,making it one of the major contributing factors to the disease burden.Obesity has been shown to associate with widespread changes in the methylome through large-scale epigenome-wide association studies（EWAS）,suggesting that DNA methylation may play an important regulatory role in the occurrence and development of obesity.However,most of the previous studies were conducted in populations of predominantly European ancestry using body mass index（BMI）as the phenotype.Epidemiological studies indicate that abdominal obesity,as assessed by waist-to-hip ratio（WHR）,is more informative for predicting the risk of obesity sequelae than overall adiposity（measured by BMI）.Furthermore,since DNA methylation and obesity phenotypes are both influenced by genes,age,and early environment,the confounding from these factors cannot be ignored when studying the relationship between them.Monozygotic（MZ）twins share almost all of their genetic components and early environmental factors,the use of MZ twin pairs discordant for obesity phenotypes in EWAS can yield more reliable results by perfectly controlling for the effects from different genetic backgrounds.Accordingly,we performed an EWAS based on WHR-discordant MZ twin pairs among Shandong Han nationality,aiming to identify DNA methylation alterations associated with abdominal obesity and to determine the causation underlying.The results are expected to provide new insight into the study of the early diagnosis and precise prevention and control of obesity and obesity-related diseases.Methods:The subjects were taken from the most recent wave（2012-2013）of twin sampling in the Qingdao Twin Registry following specific inclusion and exclusion criteria.The basic information and clinical data of the twins were collected through face-to-face questionnaires,physical examination,and blood sampling.Zygosity was determined by sex,ABO blood type,and 16 multiple short tandem sequence repeat DNA markers.We used reduced representation bisulfite sequencing technique to detect peripheral blood leukocyte genome-wide DNA methylation and Re FACTor was introduced to correct for cell type heterogeneity.Next-generation sequencing technology was used to comprehensively quantify gene expression levels.The generalized estimated equation model was fitted to estimate the relationship between the methylation level of single Cp G site and WHR with adjustment for age,sex,BMI,diastolic blood pressure,cell-type composition,and twin pairing using R software and packages.The Inference about Causation through Examination of FAmilia L CONfounding（ICE FALCON）was used to explore the temporal relationship between DNA methylation sites and WHR.Differentially methylated regions（DMRs）were detected using the comb-p python library.Genomic Regions Enrichment of Annotations Tool（GREAT）was conducted to identify biological pathways associated with obesity.Finally,the gene expression profiling data was further used to obtain differentially methylated genes that are regulated at the RNA level,thus validating the DNA methylation results.Results:A total of 60 WHR discordant MZ twin pairs were included in the current sample,including 32 male and 28 female pairs.The mean age of twins was 52.60±7.24 years,and the mean WHR was around 0.90±0.07.Single-site-based EWAS identified 92 Cp G sites with a level of P<10^-4,among which 8 Cp G sites with P<1×10^-5,and 1 Cp G site achieved genome-wide significance(P<5×10^-8).These 92 Cp Gs were annotated to 32 genes and intergenic regions,involving TUSC5,PCDHB6,ZCCHC14,AMPD3,CACN1C et al.ICE FALCON showed positive causality between 3 Cp Gs and WHR,negative causality between 2 Cp Gs and WHR,as well as bi-directional causality between 4 Cp Gs and WHR（P<0.05）.Region-based analysis identified 14 DMRs significantly associated with WHR（slk-corrected P<0.05）,located in/near 15 genes,among which 4 DMRs located in PCDHB6,LOC105373021,AMPD3,HOXD3 genes covered the top Cp G sites with P<10^-4 in EWAS.GREAT analysis revealed a large number of significantly enriched biological pathways related to obesity,including the Notch signaling pathway,Transcription regulation by b ZIP transcription factor,Hedgehog signaling pathway,Nicotinic acetylcholine receptor signaling pathway,mitochondrial L-carnitine shuttle pathway,etc（FDR<0.05）.Gene expression analysis identified significant correlations between gene expression levels of 5 differentially methylated genes and WHR,including CORO7（r=0.781,P=0.022）,ZCCHC14（r=-0.666,P=0.036）,KCNMB4（r=0.568,P=0.043）,C19orf139（r=-0.667,P=0.035）,and FUT7（r=-0.724,P=0.028）,validating the results of DNA methylation from a transcriptomic perspective.Conclusions:This study explored the association between DNA methylation and WHR based on discordant MZ twin design and identified multiple Cp G sites,DMRs,genes and pathways potentially related to WHR in Shandong Han nationality.The results are supposed to provide the scientific basis for early diagnosis and precise prevention and control of obesity and obesity-related diseases.