Correlation Between The Risk Of Eating Disorders And Single Nucleotide Polymorphism

Objective Eating disorders usually started in adolescence and early adulthood and is common in woman.The EDs patients’ eating behaviors were seriously unhealthy,often characterized by focusing too much on weight or shape,which causes severe restrictive eating or overeating,suffered from the eating behaviors and the consequent social dysfunction.According to the latest diagnostic criteria of the American psychiatric association,eating disorders are classified into four major subtypes: anorexia nervosa,bulimia nervosa,binge eating disorder,and other undifferentiated eating disorders.Patients with eating disorders often suffer from other mental illnesses such as obsessive-compulsive disorder,depression and anxiety disorders.Patients with eating disorders were in poor prognosis,complications such as cardiac arrhythmia caused by chronic malnutrition,or suicide due to the comorbidity of depression or other adverse consequences may be the cause of eating disorders.Etiology researches had been one of the hot areas in eating disorders.According to the current research results,the etiology of eating disorders was not only related to social,family environment and personality traits,but the genetic factors also had a great influence on the pathogenesis.According to the views of genetics,eating disorders were caused by multiple neural pathological pathways,including serotonin(5-HT)signaling pathway,dopamine(DA)signaling pathway,brain derived neurotrophic factor(BDNF),ghrelin and estrogen.The purpose of this study was to use meta-analysis to evaluate the correlation between Single Nucleotide polymorphisms and the onset of eating disorders.Included Single Nucleotide polymorphisms were from serotonin,dopamine,brain derived neurotrophic factor and ghrelin pathway.It is hoped to provide an effective reference for the etiology,diagnosis,treatment and prognosis of eating disorders.Materials and methods Search the published papers about the correlation between SNP and eating disorder until December 31,2017.Database included Web of Science,Pub Med and China National Knowledge Internet(CNKI),language was limited in Chinese and English.The included literatures were all designed with case-control studies,and the control groups were all from healthy population.Literature qualities were assessed using the Newcastle-Ottawa scale,and the literatures which scores above 6 stars were included in the data extractions and analysis.The articles included were summarized according to SNP.Using Rev Man 5.3 software,the combination of OR value and the estimation of 95% confidence interval were respectively carried out.For the SNP with undefined genetic model,the allele model,dominant model,recessive model,co-dominant model and super-dominant model were used for meta-analysis.The corresponding model is used to analyze the genetic model with the genetic model of the SNP already known.Subgroup analysis was carried out in accordance with the subjects and subtypes of the disease.Sensitivity analysis was carried out by the method of elimination of the included literatures one by one,and the funnel plot was used to estimate the publication bias of the included literatures.Results At last,there were 141 studies on the correlation between different gene sites and eating disorders included.Five single nucleotide polymorphisms were selected in this meta-analysis: the serotonin transporter gene linkage polymorphism region(5-HTTLPR),serotonin receptors 2A(5-HT2ARs)promoter gene-1438(A/G),catechol oxygen methyltransferase(COMT)gene(rs4680),BDNF gene(rs6265)and ghrelin(408 C/A).There were 13 related literatures in the 5-HTTLPR gene,including 2515 patients with eating disorders and 2847 healthy controls.According to allele model,5-HTTLPR had no correlation with Caucasian population in AN(OR=1.09,P =0.23)OR BN(OR=1.12,P =0.57),but increased the risk of BN in Mongolians(OR=1.55,P <0.001).According to dominant model,5-HTTLPR was neither related with AN in Caucasians(OR=1.06,P =0.55)nor in Mongolians(OR=1.54,P =0.05),and was not related to the BN in Caucasians(OR=1.23,P =0.46).A total of 18 articles were retrieved from the 5-HT2A(-1438A/G)gene related literature,and 2,463 patients with eating disorders and 2713 healthy controls were enrolled.In Caucasian race-1438 loci alleles of A gene mutation is AN risk factor [allele model(OR = 1.36,P = 1.36),dominant model(OR = 1.39,P = 1.39),recessive model(OR = 1.76,P = 1.76),co-dominant model(OR = 2.02,P = 2.02)].But has nothing to do with the AN in Mongolians [allele model(OR = 0.91,P = 0.91),dominant model(OR = 0.87,P = 0.87),recessive model(OR = 0.88,P = 0.88),codominant model(OR = 0.81,P = 0.81),super-dominant model(OR = 0.99,P = 0.99)].5-HT2A(-1438A/G)is a risk factor of BN in Caucasians [allele model(OR = 1.42,P = 1.42),dominant model(OR = 1.81,P < 0.000),recessive model(OR = 1.66,P < 0.000)],but the researches on Mongolian people were so little that cannot be analyzed.A total of 10 related literatures on the DA pathway COMT gene rs4680(Val158Met),and included 2,637 patients with eating disorders and 3776 healthy controls.Subgroup analysis showed that the onset of AN in Caucasians has nothing to do with rs4680 COMT gene A gene mutations [allele model(OR = 0.84,P = 0.84),dominant model(OR = 0.90,P = 0.90),recessive model(OR = 0.99,P = 0.99),and co-dominant model(OR = 1.04,P = 1.04),super-dominant model(OR = 0.96,P = 0.96)].The mutation of this gene will reduce the risk of the onset of BN in Caucasians [allele model(OR=0.76,P =0.02),dominant model(OR=0.65,P =0.03),and co-dominant model (OR=0.58,P =0.03)].The literatures on the Mongolians were not enough to carry out meta-analysis.The meta-analysis on BDNF gene rs6265(Val66Met)included 14 articles,with 4,754 patients with eating disorders and 5215 healthy controls.According to the results of the subgroup analysis,rs6265 carried A(Met)alleles would increase the risk of AN [allele model(OR = 1.13,P = 1.13),dominant model(OR = 1.17,P = 1.17)] and BN [allele model(OR = 1.36,P < 0.000),recessive model(OR = 1.74,P = 1.74),and co-dominant model(OR = 1.91,P = 1.91)] in Caucasians.But hasn’t increased the AN risk of the Mongolians.There were too few related studies on BN in Mongolian ethnic group,and no meta-analysis could be carried out.The meta-analysis on ghrelin gene 408C/A(Leu72Met)included 4 articles,with 713 patients with eating disorders and 702 healthy controls.Subgroup analysis was carried according to disease subtype,ghrelin gene 408 C/A(Leu72Met)loci A gene mutation does not increase the risk of AN [allele model(OR = 1.44,P = 1.44),dominant model(OR = 1.30,P = 1.30),recessive model(OR = 1.09,P = 1.09),and co-dominant model(OR = 1.18,P = 1.18),super-dominant model(OR = 0.76,P = 0.76)].However,the SNP increased the risk of BN(OR=1.46,P =0.02),dominant model(OR=1.78,P =0.003)and BED[alleles model(OR=2.31,P =0.005),and dominant model(OR=2.55,P =0.003)].ConclusionIn the 5-HT pathway,5-HTTLPR was not associated with the incidence of eating disorders.But 5-HT2A(-1438A/G)gene polymorphism is associated with increased risk of eating disorders,especially in the Caucasians.The single nucleotide polymorphism in the DA pathway rs4680 was not associated with the incidence of eating disorders.The genetic polymorphism of BDNF gene rs6265 is related to the incidence of eating disorders,especially in Caucasians.Ghrelin 408C/A is not related to AN,but may increase the risk of BN and BED.More research on Asian populations is needed in the future to provide more evidence for the genetic causes of eating disorders.