Bivariate Genome-wide Association Study Of Cognitive Function And Fasting Plasma Glucose

Background:Cognitive dysfunction and diabetes mellitus are common comorbidities.Genetic factors play an important role in the occurrence of both,and they have common pathophysiological changes and genetic mechanisms.At present,many domestic and foreign studies have investigated the heritability and genetic susceptibility sites of cognitive function and fasting glucose,but the common genetic effects of the two have not been fully studied.Therefore,the present study aims to explore the genetic correlation between cognitive function and fasting plasma glucose using bivariate genetic analysis,and to use the bivariate genome-wide association study（GWAS）to investigate the genetic association between cognitive function and fasting plasma glucose to explore the common Single Nucleotide Polymorphisms（SNPs）,genes and pathways.Methods:The subjects were selected from the twin registration system of Qingdao,and all subjects gave informed consent.Cognitive function was assessed using the Montreal Cognitive Assessment Scale（Mo CA）,and fasting plasma glucose levels were measured using an automated biochemical analyzer according to standard procedures.The zygosity was identified by gender,ABO blood type,microsatellite DNA scanning and genotyping techniques.Illumina Infinium Omni2.5 Exome-8V1.2 chip was used for genotyping.Based on the linkage disequilibrium information of the 1000 Genomes Project Phase 3 Reference Panel,IMPUTE2 was used for genotype imputation.Cholesky decomposition model were constructed using MX software to evaluate the influence of genetic and environmental factors on phenotypic associations between cognitive function and fasting plasma glucose.Genome-wide Efficient Mixed Model Analysis（GEMMA）was used for SNP-based analysis.The regional association plots were drawn using Locus Zoom.Vegas2（Versatile Gene-Based Association Study-2）software was used to carry out the research on the basis of gene.Pascal was used to calculate the path score and the pathway enrichment analysis was performed.Results:A total of 381 pairs of twins were included in this study,of which 139 pairs of dizygotic were analyzed for GWAS.Subjects had a median age（interquartile）of 50（12）years,a median cognitive function score（interquartile）of 22（6）points,and a median fasting glucose score（interquartile）of 5.10（1.21）mmol/L.After adjusting for age and sex,the genetic correlation between cognitive function and fasting glucose was-0.25（-0.37,-0.13）.139 pairs of dizygotic twins were included in the GWAS study,and the results were as follows:（1）In the SNP-based study,a total of 1,339,805 SNPs were included in this study,among which 6 SNPs reached the significance level(P<5×10^-8),46 SNPs reached the suggestive significance level(P<1×10^-5).The lowest P value was rs4516415 adjacent to Spermatogenesis associated serine rich 2 like（SPATS2L）gene(P=5.49×10^-9).After imputation,47 SNPs exceeded the significant level(P<5×10^-8),323 SNPs exceeded the recommended association level(P<1×10^-5).The SNPSs reached the significance level were all located in or near SPATS2L gene,which were rs4516415,rs295134,rs10931893,rs159320,rs295114 and rs1900706.Regional association analysis showed that chromosome 2q33.1 was significantly associated with cognitive function and blood glucose.After imputation,47 SNPs reached the significance level,323 SNPs reached the suggestive significance level,and the lowest P value was rs60106404 near SPATS2L gene（P=4.91×10^-10）.（2）In gene-based analysis,no genes above significant level were found before and after imputation,but there were 287 and 276 genes whose P values were lower than the suggestive significance level（P<0.05）,respectively.186 genes reached the recommended association level before and after interpolation The top 5 genes with the lowest P value before and after imputation were SPATS2L,WDPCP,C2orf69,TLR6,SNRNP27,and SPATS2L,TYW5,C2orf69,C2orf47,LANCL1.（3）In the pathway analysis,there were 706 and 726 pathways associated with cognitive function and fasting glucose before and after imputation（P<0.05）,including 603 pathways before and after imputation.The pathway with the lowest P value were G1 PATHWAY before and after imputation.Other pathway associated with cognitive function and fasting glucose were FGFR1 LIGAND BINDING AND ACTIVATION,ACTIVATED POINT MUTANTS OF FGFR2,SNARE INTERACTIONS IN VESICULAR TRANSPORT,SIGNALING BY NOTCH and others.Conclusions:There is a low degree of genetic correlation between cognitive function and blood glucose in twin population in Qingdao.We identified multiple SNPs,genes,and pathways associated with cognitive function and fasting plasma glucose.Our findings enrich the related results of cognitive function and fasting glucose in the field of genetic research and provide a basis for discovering more molecular biological mechanisms related to cognitive dysfunction and diabetes in northern Chinese population.