Computational Analyses And Functional Research Of Type 2 Diabetes And Obesity Associated SNPs And Genes Identified By GWAS

Type 2 diabetes(T2D)is often associated with obesity,and obesity is an important risk factor for T2D.T2D is a common complex disease,characterized by chronic hyperglycemia due to insulin resistance of peripheral tissues(skeletal muscle,liver,adipose tissue).and insufficient compensatory insulin secretion by pancreatic beta cells.The heritability of T2D is about 80%.Obesity results from a prolonged energy imbalance owing to intake exceeds expenditure,which may be defined as a condition in which there is an excessive accumulation of body fat and abnormally high body fat percentage caused by an increase in the number and volume of adipocytes at the cellular level.Usually,obesity is defined by body mass index(BMI).In the past decades,genome-wide association study(GWAS)is a major method to identify susceptibility loci for complex diseases,and has increased the number of associated loci for T2D and obesity from a few to hundreds,but the exact pathogenesis of these loci is not fully understood.The future challenge of T2D and obesity genetic studies are to elucidate functional mechanisms through which these GWAS associated loci modulate T2D and obesity risk.Aims of our study are to identify potentially functional SNPs in order to provide targets and clues for subsequent experimental research via computational analyses and functional annotation,and to offer the promising guidance for future research with respect to the etiology and pathogenesis of T2D and obesity.1.Computational analyses of type 2 diabetes associated SNPs and genesTotal of sixty-one GWASs and meta-analyses reported 192 SNPs and 161 genes that were associated with T2D at the stringent threshold of P<5×10-8 for genome-wide significance.1314 SNPs were in strong LD with T2D GWAS lead SNPs by SNAP(r2≥0.80).RegulomeDB identified a total of 222 SNPs with score less than 3 may have a relatively high degree of evidence for potential regulatory function.169 SNPs were identified as eQTLs,and 133 SNPs were located at protein binding sites,including CTCF,EP300,HNrF4A,TCF7L2,FOXA1 and FOXA2,which are required for normal pancreatic development and maintaining β-cell function.For 3DSNP,41 SNPs had score>100.Three GWAS lead SNPs and 26 proxy SNPs may affect miRNA binding by miRNASNP(v2.0).LncRNAs at T2D GWAS loci were identified,14 SNPs located in TFBSs of IncRNAs,and 11 lead SNPs may affect miRNA-lncRNA interaction.M6A-HPCS,RAMPred and iRNAm5C-PseDNC uncovered that some lead SNPs sites as well as the surrounding regions of them may habour multiple m6A(19),m1A(81)and m5C(24)sites.RBPmap analyses showed that some lead SNPs might influence the differential binding of diabetes related RNA binding proteins YBX2,TRA2B,RBMS1 and IGF2BP2.71(-37%)lead SNPs were identified as CpG-SNPs,and 17 lead SNPs were identified as cis-mQTLs.Using PhosSNP 1.0 database,four T2D-associated lead SNPs and nine proxy SNPs were identified as phosSNPs that might influence protein phosphorylation.GO and KEGG signaling pathway enrichment analyses via WebGestalt showed that T2D GWAS genes were mainly enriched in biological regulation process,binding function and several KEGG signaling pathway including well-established maturity onset diabetes of the young and type II diabetes mellitus pathways.PPI network was constructed by STRING,and identified highly interconnected "hub" genes and bottleneck genes TCF7L2,MTNR1B,SLC30A8.Clustering analysis of the PPI network was performed using MCODE in Cytoscape to select functional modules,and two modules were identified.CircRNAs that related to T2D GWAS loci were identified by CircInteractome.2.Computational analyses of obesity associated SNPs and genesA total of fifty-nine GWASs and meta-analyses reported 445 SNPs involved in 389 genes that were associated with obesity.And 5,614 proxy SNPs were considered to be in strong LD with obesity GWAS lead SNPs by SNAP.RegulomeDB identified a total of 680 SNPs with score less than 3 may have a relatively high degree of evidence for potential regulatory function.Some obesity associated SNPs were identified as eQTLs,and some were located at the binding sites of obesity related transcription factors.For 3DSNP,26 SNPs had score>100.Nine GWAS lead SNPs and 35 proxy SNPs may affect miRNA binding.LncRNAs that related to obesity GWAS loci were identified,12 lead SNPs located in TFBSs of related lncRNAs,20 lead SNPs may affect miRNA-lncRNA interaction.M6A-HPCS,RAMPred and iRNAm5C-PseDNC uncovered that some lead SNPs sites as well as the surrounding regions of them may exist multiple m6A(31,one of them located in rs17150703),m1A(126)and m5C(35,one of them located in rs62105306)sites.RBPmap analyses showed that some lead SNPs might influence the differential binding of obesity related RNA binding proteins YBX2,TRA2B,RBMS1,RBFOX1,RBM6 and IGF2BP2.283(-63%)lead SNPs were identified as CpG-SNPs,and 29 lead SNPs were identified as cis-mQTLs.Using PhosSNP 1.0 database,ten obesity-associated lead SNPs and thirteen proxy SNPs were identified as PhosSNPs.GO enrichment analyses via PANTHER showed that obesity GWAS genes mainly involved in biological processes categories including cellular process,metabolic process,biological regulation and developmental process,and molecular function categories including binding,catalytic activity and receptor activity,and cellular component categories including cell part,organelle,macromolecular complex and membrane.Further analyses found that obesity GWAS genes significantly enrnched in 104 GO functional categories and 15 KEGG signaling pathways.PPI network was constructed by STRING,and identified highly interconnected"hub"genes and bottleneck genes,such as MC4R,FTO and TMEM18.These hub genes were mainly involved in AMPK signaling pathway,Neurotrophin signaling pathway,PI3K-Akt signaling pathway,Circadian rhythm,T2D,and several cancer associated pathways.Clustering analysis of the PPI network was.performed using MCODE in Cytoscape to select functional modules,and three modules were identified.CircRNAs that related to obesity GWAS loci were identified by CircInteractome3.Experimental validation of influence of four T2D GWAS SNPs on the binding of transcription factorsOne of the molecular mechanisms by which SNPs increase disease susceptibility is to modulate gene expression via affecting the binding of transcription factors.In the present study,we firstly analyzed effects of T2D GWAS SNPs on transcriptional factors binding motifs using JASPAR,and experimentally demonstrated that GWAS SNP rs4607103 acts as an allele-specific enhancer to bidirectionally regulate ADAMTS9/ADAMTS9-AS2 expression via differential binding of TCF7L2.GWAS SNP rs4430796 acts as an allele-specific enhancer to modulate HNF1B expression via differential binding of PAX6.GWAS SNP rs849135 acts as an allele-specific enhancer to bidirectionally regulate expression of JAZF1 and JAZF1-AS1 via differentially bind to CREB1.4.Experimental validation of influence of three T2D GWAS lead SNPs and live proxy SNPs on the binding of miRNAsEffects of 3 T2D GWAS SNPs and 5 proxy SNPs on miRNA binding were verified through experimental methods,with negative results,suggesting the limitation of computational analyses and the necessity and importance of combining computational analyses with experimental verification.