Font Size: a A A

Genome-Wide DNA Methylation Of Fasting Blood Glucose In Middle-Aged And Elderly Monozygotic Twins

Posted on:2023-12-04Degree:MasterType:Thesis
Country:ChinaCandidate:W Q YaoFull Text:PDF
GTID:2544306833952029Subject:Epidemiology and Health Statistics
Abstract/Summary:PDF Full Text Request
Purpose:Type 2 diabetes mellitus(T2DM)is widespread in the world and has become one of the serious public health burdens,which needs to be solved urgently.As an important indicator for the T2DM diagnosis,high fasting blood glucose(FBG)levels can increase morbidity and mortality even when it is below the diagnostic thresholds.The occurrence and development of T2DM were associated with deoxyribonucleic acid methylation(DNAm),and both T2DM and DNAm were influenced by genetic factors and environmental exposure.Therefore,the findings of studies on association of DNAm with T2DM or glycemic metabolism indicators were limited.Monozygotic(MZ)twins with discrepant disease/traits share the same genome sequence and early life environment,which can eliminate the confounding effects caused by these factors in the DNAm study of diseases with complex mechanisms,thus improving the accuracy of statistical analysis.Therefore,we conducted an epigenome-wide association study(EWAS)based on MZ twins with discrepant FBG levels to systematically explore the DNAm sites,to determine differentially methylated regions(DMRs),to identify genes,and to screen biological pathways related to the occurrence and development of T2DM.Moreover,the causal relationship between single Cp G and FBG was further explored.Additionally,the differentially methylated results were further validated based on gene expression profile data of MZ twins with inconsistent FBG levels.Methods:The subjects came from Qingdao twin registration system,and informed consent were signed.The glucose oxidase method and semi-automatic analyzer(Hitachi 7600,Japan)were utilized to detect FBG level.After fasting for 10-12 hours,venous blood was extracted and zygosity was identified step by step through the differences of the gender,ABO blood type and short tandem repeat DNA marker technique.DNAm spectrum was sequenced by the reduced representation bisulfite sequencing technique.Re FACTor software was used to correct the influence of cell type composition on DNAm.Gene expression profile was sequenced by performing the next-generation sequencing.The relationship between FBG and DNAm at cytidine phosphate guanosines(Cp Gs)was analyzed by generalized estimation equation(GEE)through R software and some covariables were adjusted such as age,sex,diastolic blood pressure,cell type composition and twin pair number.Genomic Cp Gs were annotated to the nearest gene by using R/Bioconductor-Package-bioma Rt.Genomic regions enrichment associated with FBG was analyzed online by genomic regions enrichment of annotations tool(GREAT).FBG-related DMRs were identified by comb-P software.The Inference about causation through examination of familial confounding factors(ICE FALCON)was used to infer the causal relationship between DNAm site and FBG.The weighted gene correlation network analysis(WGCNA)was used to further explore differentially expressed gene associated with FBG based on gene expression profile.The pathway enrichment associated with FBG was performed by DAVID tool.Finally,the results were compared and integrated with DNAm analysis to verify the differential methylation genes as well as biological functions and pathways.Results:A total of 52 MZ twins with inconsistent FBG levels were included in DNAm analysis,including 27 male pairs and 25 female pairs.The mean age of twins was 52.12±7.44years,and the median and 95%range of FBG levels were 5.44(3.76,7.40)mmol/L.(1)The relationship between DNAm of 30 top Cp Gs and FBG levels reached the P-value<1×10-6,and among which 12 Cp Gs reached P-value<1×10-7.The top Cp Gs with P-value<1×10-6were located at 17 gene,and there were 4,2,3,2,5,2,and 2 Cp Gs that located at/near TLCD1,SYNPO,MZF1,PTPRN2,SLC6A18,ASTN2,and IQCA1,respectively.The remaining 10 Cp Gs were located at/adjacent to 10 genes,including ALDH1L2,RNF126,PPP2R2A,LAMC3,ENSG00000270492(The corresponding gene name was not found),LINC02064,FAM175B,GRIN1,LONRF2P1,and PDE2A.GREAT analysis showed that FBG-related pathways such as sequence-specific DNA binding,mitogen-activated protein kinase p38 binding,metalion binding,insulin receptor signaling pathway,regulation of biosynthetic process,cell fate commitment and so on.DMRs analysis identified 32 DMRs associated with FBG level.Among which,the DNAm of 18 DMRs and 12 DMRs were positively and inversely associated with FBG levels,separately.But the correlation between DNAm of 2DMRs and FBG level was difficult to determine.These DMRs located at/near 24 genes,attentively,4 DMRs covered the top Cp G sites with P-valve<1×10-6.(2)The causal inference analysis based on Cp Gs with P<1×10-6 revealed that the DNAm of the 7 Cp Gs(Cp G13-16,Cp G22-23,and Cp G30)located at/near 3 genes(SLC6A18,IQCA1,and PDE2A)were mutually causal with FBG,FBG changes followed by DNAm changes,and vice versa.Whereas,DNAm changes at 8 Cp Gs(Cp G5-6,Cp G9,Cp G,17 Cp G20,Cp G24,Cp G27,and Cp G29)located at/near 5 genes(MZF1,PTPRN2,ENSG00000270492,GRIN1,and SLC6A18)caused FBG changes.And FBG changes caused DNAm changes at 5 Cp Gs(Cp G2,Cp G7,Cp G11,and Cp G18-19)located at/near 2 genes(TLCD1 and ASTN2).The causal association between the DNAm of residual Cp Gs and FBG was not statistically significant.(3)Based on 12 MZ twins with inconsistent FBG levels,nineteen DNAm modules were identified by WGCNA.Darkolivegreen module had the most significant positive correlation with FBG levels(r=0.61,P=0.001)and this module was further analyzed.As shown in the hierarchical clustering tree,the positions of Darkolivergreen module and FBG level were in the same branch.Meanwhile,the correlation heat map and correlation scatter plot also showed that Darkolivergreen module was related to FBG.Additionally,85 hub genes in Darkolivegreen module were determined.DAVID analysis showed that the enrichment pathways of Darkolivegreen module genes including G-protein coupled receptor signaling pathway,olfactory transduction,neuropeptide signaling pathway,dopamine binding,retinol metabolism,regulation of cell fate commitment,calcium ion binding,etc.(4)In EWAS,the Cp Gs with P-value<1×10-4 were located at/near 123 genes,78 of which were verified by gene expression profiling.And 18 genes including ALDH1L2,TLCD1,RNF126,SYNPO,MZF1,PTPRN2,PPP2R2A,SLC6A18,ASTN2,LAMC3,IQCA1,FAM175B,GRIN1,PDE2A,MRPL23,CASZ1,GON4L,and CSNK1E were also annotated at the top Cp G sites(P-value<1×10-6)or DMRs.Among which,causal relationships between DNAm of 8validated gene including 20 top Cp Gs and FBG were identified by ICE FALCON method including TLCD1,MZF1,PTPRN2,SLC6A18,ASTN2,IQCA1,GRIN1,and PDE2A.Additionally,11 common biological pathways,such as dopamine binding,regulation of biosynthetic process,neuron fate specification,and cell fate commitment,were also found between DNAm analysis and WGCNA.Conclusions:We identified multiple DNAm sites,DMRs,genes,DNAm modules,and biological pathways that may be associated with FBG.Some important clues were provided by these findings for further elucidating the pathogenesis of T2DM in the population.
Keywords/Search Tags:Causation, DNA methylation, Gene expression profile analysis, Fasting glucose, Monozygotic twins
PDF Full Text Request
Related items