Association Between RELN Polymorphisms And Schizophrenia

ObjectivesTo investigate the association between SNPs of RELN gene and schizophrenia,to study the relationship between the level of serum reelin and schizophrenia,and to analyze whether there is gender-specific in the above associations.MethodsThis study applied the case-control design to investigate the association between RELN SNPs and schizophrenia among 761 cases and 775 controls.The genotype of four SNPs(rs1062831,rs736707,rs3808039 and rs362746)in RELN gene was determined by MALDI-TOF-MS,the best genetic model was determined by the SNPStats website,the association between each SNP of RELN gene and schizophrenia in all subjects,males and females was analyzed by SPSS 24.0 software,and the genetic power was calculated by Quanto software.Thirty-one cases and thirty-one controls were included to study the association between the level of serum reelin and schizophrenia.The concentration of serum reelin was determined by enzyme linked immunosorbent assay(ELISA).Serum protein levels between case group and control group were compared by SPSS 24.0 software,and scatter plot was drawn by Graph Pad Prism 8.Results1.A total of 1536 subjects were included in this study,including 443 males(58.2%)and 318 females(41.8%)in the case group,435 males(56.1%)and 340 females(43.9%)in the control group.There was no significant difference in the gender structure between the two groups(χ2=0.681,P=0.409).The mean age of the case group and control group was 34.61±12.02 and 34.74±11.41 years,respectively,and there was no significant difference in the level of average age between the two groups(t=0.221,P=0.825).The reelin level was measured in 62 subjects(31 cases and 31 controls),including 16 males(51.6%)in the case group and 17 males(54.84%)in the control group.There was no significant difference in gender structrue between the two groups(χ2=0.065,P=0.799).The mean age of the case group and control group was 24.35 ± 3.85 and 24.48 ± 3.98 years,respectively,and there was no statistical difference(t=0.130,P=0.897).The two groups were matched by age and sex.2.Hardy-Weinberg equilibrium test showed that the 4 loci(rs1062831,rs736707,rs3808039 and rs362746)in the control group conformed to the Hardy-Weinberg equilibrium(P>0.05).3.Among all subjects,the genotype distribution of rs362746 was statistically different between the case group and control group(χ2=11.470,P=0.003),while the genotype and allele of other loci(rs1062831,rs736707 and rs3808039)were not significantly different between the two groups.In both men and women subjects,genotypes and allele distributions of the four loci were not found to be statistically significant between the case group and the control group after Bonferroni correction(adjusted by age).4.Genetic model analysis showed that the best genetic model of rs362746 was the recessive model.Under the recessive genetic model,there was a statistical association between rs362746 and schizophrenia(OR=1.958,P=0.001).Participants carrying the GG genotype had a 1.958 times of developing schizophrenia risk than participants carrying AG and AA genotype.The best genetic models of rs1062831,rs736707,and rs3808039 were overdominant,recessive,and overdominant,respectively.Under each model,no association between the three loci and schizophrenia was found among all subjects,males and females.5.There were no statistically significant differences in the distribution of haplotypes consisting of 2,3,and 4 adjacent SNPs(rs1062831,rs736707,rs3808039 and rs362746)in the RELN gene(P>0.05).6.There was significant difference in serum reelin level between the case group and control group(P<0.01),and the serum reelin level was significantly lower in the case group than in the control group.Conclusion1.The rs362746 polymorphism of RELN gene was associated with schizophrenia.Under the best genetic model(recessive model),participants carrying the GG genotype had a higher risk of developing schizophrenia risk than participants carrying AG and AA genotype,and there was no association between other loci(rs1062831,rs736707,and rs3808039)and schizophrenia.2.After the gender stratification,the four loci(rs1062831,rs736707,rs3808039 and rs362746)of RELN gene were not found to be associated with schizophrenia after Bonferroni correction.3.Different haplotypes of four SNPs(rs1062831,rs736707,rs3808039 and rs362746)in RELN gene were not associated with schizophrenia.4.The serum level of reelin was lower among patients with schizophrenia than that in the control group.