| Research background and objectiveType 2 diabetes mellitus is becoming a global pandemic.Diabetes has become a major public health problem in China,a country now has the largest number of diabetes patients around the world.The etiology and pathogenesis of type 2 diabetes are still not fully understood,which makes it difficult for people to prevent and treat it.It has been increasingly accepted that type 2 diabetes arise from numerous environmental and genetic factors working together.Each of these risk factors can,via largely undefined mechanisms,lead to skeletal muscle,adipose and hepatic insulin resistance,and/orβ-cell dysfunction.Ultimately,insulin resistance accompanied by inadequate insulin secretory responses results in postprandial and fasting hyperglycaemia.In turn,diabetes-related hyperglycaemia and associated metabolic abnormalities can further alter signal transduction and gene expression（glucolipotoxicity）,thus contributing to a vicious cycle.The rapid increase in the prevalence of type 2 diabetes may partially attribute to environmental and lifestyle factors,including overnutrition,obesity,physical inactivity and so on,furthermore,overwhelming data support that type 2 diabetes has an evident genetic component and genetic factors influence the disease susceptibility.First,twin studies demonstrated a markedly higher concordance for type 2 diabetes in monozygotic compared with dizygotic twins.Second,type 2 diabetes clusters within families and first-degree relatives have,compared with the general population,higher risk to develop the disease.Finally,certain ethnic minorities and indigenous groups with low population admixture show exceptionally high type 2 diabetes prevalence.In recent years,the domestic and foreign scholars have conducted a large number of genetic epidemiological studies,applying a variety of strategies and approaches,in order to locate and identify susceptible genes of type 2 diabetes and better understand its genetics architecture and pathogenesis.We review some commonly used methods in the following sections.（1）Linkage analysis:Linkage analysis was the primary methods to link genotype and phenotype in early time.Genome-wide scans have been performed to search for microsatellite markers associated with type 2 diabetes in more than 20ethnic groups（such as the Finns,Mexico Americans,Caucasian American,British,Japanese,French,Chinese northern and Southern Han people and so on）The results show that multiple susceptibility loci of type 2 diabetes positioning on different chromosomes and some of them have been duplicated in different populations.（2）Candidate-gene association study:Candidate gene approach directly tests the effects of genetic variants of a potentially contributing gene in an association study.These studies,which may include members of an affected family or unrelated cases and controls,can be performed relatively quickly and inexpensively and may allow identification of genes with small effects.However,the candidate gene approach is limited by how much is known of the biology of the disease being investigated.As researchers select potential candidate genes based on the results of linkage analysis,chromosomal location or functional information of gene products.Since insulin gene was reported twenty years ago,more and more candidate genes were studied in different populations.Candidate-gene association studies showed that the genes for the peroxisome proliferator activated receptor gamma（PPARG）and the potassium inwardly rectifying channel subfamily J member 11（KCNJ11）were 2 candidate susceptibility genes.Both genes encode targets of anti-diabetes medications and harbor missense variants associated with T2D.To date,only these two loci（PPARG,KCNJ11）were robustly implicated in T2D susceptibility.（3）Genome-wide association study,GWAS:Limited success of candidate gene approach and linkage analysis in identifying the genetic background of type 2diabetes has caused many research groups to apply the genome-wide association studies approach in large case-control cohorts.In a typical GWAS,hundreds of thousands of SNPs are genotyped for thousands of individuals.By comparisons of differences in the DNA variations between the normal and affected individuals,the SNPs can be ordered according to their degrees of association.The common approach is to select dozens of the most significant SNPs in the list for further investigations.Nearly 60 susceptibility loci associated with type 2 diabetes have been successfully identified and replicated since 2007.Substantial progress in our knowledge of the genetic basis of T2D has been elucidated by T2D GWAS,but there remains a large portion of unexplained genetic heritability.This may attribute to some limitations of GWAS.Recently,scientists have focused on performing further analysis by utilizing the genome-wide genotyping data to identify more susceptibility genes of complex diseases.Many strategies and methods have been applied in the following GWAS,such as gene-gene and gene-environment interaction,pathway analysis,and epistasis study and so on.The application of these strategies and methods compensates the limitation of the traditional GWAS and provides new insights into genetics basis of complex diseases.（4）SNP-SNP interaction:GWAS have been successful in identifying individual variants in a variety of genes that may play a role in the etiology of T2D.However,because of practical and statistical challenges,none of the GWAS have considered interactions among the thousands of variants.Gene-gene interaction mainly refers to the nonallelic gene interactions,also known as epistatic effect,which may be one of the most important factors influencing susceptibility of complex diseases.Complex diseases are affected by minor genes,these effects are more likely to be detected by the interaction analysis.In addition to increasing the power to detect associations,it is hoped that detecting interactions between loci will allow us to elucidate the biological and biochemical pathways that underpin disease.One study showed that there was a two-locus interaction between the UCP2 and PPARγgenes among 23 loci in the candidate genes of Type 2 diabetes In another research the results suggest that the single nucleotide polymorphisms from the obesity candidate genes may contribute to the risk of T2D in an interactive manner.（5）Pathway-based analysis:The pathway-based association analysis is to take a pathway as a basic unit of analysis.The pathway may come from KEGG（Kyoto Encyclopedia of Genes and Genomes）or Gene Ontology database,which was defined based on existing knowledge in biological processes.The pathway-based approach aims to simultaneously study association of a group of genetic variants in the same biological pathway,which help us to holistically unravel complex genetic structure of common disease to gain insight into the biological processes and disease mechanism.It is well known that genes do not work in isolation;instead,complex molecular networks and cellular pathways are often involved in disease susceptibility and disease progression.Therefore,by taking into account prior biological knowledge about genes and pathways,we may have a better chance to identify the genes and mechanisms that are involved in disease pathogenesis.In previous study we found that adipocytokine genes and related signaling pathways have close relationship with insulin signaling transduction,and variants in these genes or pathways may lead to insulin resistance,which plays an important role in the pathophysiology of type 2 diabetes.Therefore,we select TNFαpathway and 6adipocytokine genes and propose（1）to investigate associations between single SNPs of these genes and type 2 diabetes;（2）to identify SNP-SNP interactions;（3）to assess the association of the TNFαpathway with type 2 diabetes mellitus by using a pathway-based approach;（4）to further estimate the combined effects of these SNPs and predictive power for type 2 diabetes.We hope to find novel variants associated to the disease.The results of this study will provided the theoretical basis for further revealing the pathogenesis of type 2 diabetes mellitus,identifying potential drug targets,predicting higher risk individuals and preventing type 2 dianbetes mellitus.Materials and MethodsThis study employed a case-control study design.The cases were patients with type 2 diabetes which were recruited from endocrinology departments of 8 hospitals in Guangdong province（affiliated hospital of Guangdong medical college,the people’s hospital of Maoming,the people’s hospital of Gaozhou and etc.）.The non-diabetic controls were recruited from people who came for general health exams.The cases and controls were matched according to the region and age.Questionnaire surveys were conducted on subjects who met the inclusion criteria by trained investigators.Age,gender,native place,occupation,history of disease,course of disease,smoking history,family history,complications,diet,exercise and etc.were collected.ACD（anticoagulant citrate dextrose）peripheral blood was collected in the morning,part of blood sample was employed for biochemical detection and another part was used to extract DNA.DNA was extracted from 4 ml ACD（anticoagulant citrate dextrose）peripheral blood.We selected 23 tagging single nucleotide polymorphisms（rs1799964,rs1800629,rs1860545,rs4149621,rs653667,rs10781522,rs3750512,rs2261434,rs10864490,rs7075976,rs9888128,rs12269373,rs11986055,rs3810901,rs3211908,rs584507,rs3802682,rs6584467,rs10205923,rs16822633,rs9521509,rs10136000,rs2494752）across the region of 14 genes（TNF,TNFR1,TNFR2,TRAF2,MTOR,JNK1,CHUK,IKBKB,CD36,PRKCQ,NF-κB,IRS1,IRS2,AKT1）of TNFαpathway and 12 SNP（rs3806318,rs1805096,rs2275735,rs2232851,rs822396,rs3774261,rs4730153,rs2058540,rs10244329,rs2060715,rs7408174,rs3219175）across the region of 6 adipocytokines and receptor genes（ADIPOQ,ADIPOR1,LEP,LEPR,RETN,NAMPT）.We genotyped the SNPs by SNPscanTM multiple SNP genotyping assays.We checked the demographic,biochemical characteristics and SNP data and integrated them for statistical analysis.Comparisons of all variables between T2DM and control subjects were carried out by chi-square test for nominal variables or t-test for continuous variables.We tested the association of the single SNPs with type 2 diabetes using Pearson chi-square test,Cochran-Armitage trend test,MAX3,MERT（the maximin efficiency robust test）,GMS（genetic model selection）,logistic regression,etc.Logistic regression was also used for SNP-SNP interactions analysis.We tested the pathway analysis using SNP set analysis based on logistic kernel machine regression.The joint effect of multiple variants gene locus was estimated by genetic risk score.We used the ROC curve to evaluate the predictive ability of GRS.Searching risk SNP combination of type 2 diabetes was performed by genetic algorithm.All statistical analyses were performed by statistical packages including SPSS15.0,PLINK 1.07,R2.14.2,Med Calc 11.5 and MATLAB 7.9.0.Results（1）Association analysis of TNFαpathway and type 2 diabetes:In association analysis based on single SNPs,rs2494752 had statistical significance in allele association analysis,genotype association analysis,robust test and after adjusting covariates.Its minor allele in Caucasian is A,while in Chinese Han is G in our research,which is the same as Japanese;rs6584467 showed statistical significance in robust test（MERT method）,rs9521509 showed statistical significance in dominant genetic model without adjusting the covariates,but both of them showed no statistical significance after adjusting the covariates;On the other side,rs3211908showed no statistical significance without adjusting the covariates,while showed statistic significance in additive model after adjusting the covariates.We got six statistically significant gene pairs from SNP-SNP interaction analysis,the corresponding gene pairs were:TNF and AKT1,IKBKB and IKBKB,IKBKB and TNFR1,TRAF2 and NF-Κb,TRAF2 and PRKCQ,JNK1 and TNFR1。There was no statistical significance in pathway analysis no matter the covariates were added or not,the empirical P values from bootstrap method was close to the P values from hypothesis testing.（2）The association analysis of polymorphisms of adipocytokine and corresponding receptor genes with type 2 diabetes:Rs2060715 showed statistical significance in the comparison of genotype frequency（after adjusting covariates）,genotype association analysis under dominant genetic model with or without adjusting the covariates,the P value of MAX3 robust testing was close to significant level,that’s to say,the results for rs2060715 had a good consistence.But for rs3774261 and rs2232851,statistic significance were only found in genotype association analysis with the covariates were adjusted.The interaction analysis of 12 SNPs from six adipocytokines genes and their corresponding receptor genes displayed three pairs of genes with statistic significance,they were LEPR and ADIPOR1,ADIPOR1 and RETN,LEP and LEP,the rs2060715and rs2232851 were both included.SNP-SNP interaction analysis between adipocytokines genes,receptor genes and genes in TNFαpathway get 19 pairs with statistical significance,among them,TRAF2,LEPR,ADIPOR1,ADIPOQ and TNF had more interaction with other genes,indicating that these genes may play important roles in glycolipid metabolism.Moreover,three interactions with statistical significance existed between TNF and LEP 4 SNPs,indicating the closer relationship between these two genes.（3）Joint effect of multiple locus variants to the risk of type 2 diabetesFive SNPs associated with type 2 diabetes（rs2494752,rs3211908,rs3774261,rs2060715 and rs2232851）were used to calculated the GRS（genetic risk score）and w GRS（weighted genetic risk score）.Both of the results had statistical significance,and the scores of GRS and w GRS were all higher than the controls.After adjusting gender,BMI and triglyceride,the every unit increase of GRS will result in 45.8%increase in the risk of type 2 diabetes,the OR=1.46,95%CI is 1.25-1.70,P<0.001.Three logistic regression models were constructed based on the covariates,GRS and w GRS:Model 1（only the covariates）,Model 2（the covariates and the GRS）and Model 3（the covariates and the w GRS）,the AUC（area under curve）was calculated to measure the predictive abilities of these three models.The AUC of Model 1 is0.691（95%CI 0.659-0.723）;The AUC of Model 2 is 0.706（95%CI 0.675-0.737）;The AUC of Model 3 is 0.710（95%CI 0.679-0.741）;When comparing Model 2 to Model 1,we found that the values of AUC possess no statistic significance（P=0.08）,while comparing Model 3 to Model 1,the values of AUC have statistic significance（P=0.03）.It means that the w GRS can approve the predictive ability for T2DM.In addition,when compared Model 3 to Model 2,the values of AUC also possessed no statistic significance（P=0.31）.（4）Searching for risk SNP combinations of T2DM using genetic algorithm:An optimal gene barcode was found and thirteen SNPs were included:rs4149621,rs3750512,rs2261434,rs3211908,rs2494752,rs3806318,rs1805096,rs2275735,rs822396,rs3774261,rs10244329,rs7408174 and rs3219175.LEPR、ADIPOQ and RETN all had two SNPs be included by the optimal gene barcode.Besides,three SNPs out of five SNPs associated with type 2 diabetes were also included by the optimal gene barcode.The sensitivity of optimal gene barcode was65.17%,the specifity was 80.00%,the youden index was 45.17%,and the predicting accuracy was 72.83%.The ROC curve was drawn according to predicted values,the AUC was 0.565,95%C was 0.538-0.592,the results of hypothesis testing showed P<0.001（comparing to AUC=0.5）,which means that the gene barcode can be used to prediction.Comparing to GRS,the value of AUC increased a litter,but showed no statistical significance.Conclusions（1）Rs2494752 and rs3211908 of TNFαpathway are associated with type 2diabetes.In adipocytokine and receptor genes,rs37774261 is associated with type 2diabete,while the evidence for this association for rs2060715 and rs2232851 are not sufficient,especially for rs2232851,more verification of this relationship from other independent samples are needed.（2）The SNPs from adipocytokine and receptor genes have pairs of interactions with SNPs from TNFαpathway,these interactions are statistical significant,these SNP-SNP interactions will increase the risk of type 2 diabetes.It means that the SNP-SNP interaction is the constituent part of genetic structure of type 2 diabetes.（3）The association between SNP set of TNFαpathway and type 2 diabetes are not confirmed.（4）SNP combinations can both increase the predictive ability of T2DM,that’s to say,the combined effects of SNPs can increase the risk of T2DM. |
PDF Full Text Request