Association Study Of SNPs With Schizophrenia In Combined Family And Case-control Samples

Schizophrenia is one of the most common and devastating mental disorders in humans. It is not clear what causes the disease through the pathophysiological mechnism by which the disease is triggered and developed. Twin sutdy shows that the concodance rate of schizophrenia is 4-6 fold higher in monozygotic than in dizygotic twins; adoptive study also supports the involvement of a genetic conponant in the disease. Genomie-wise association (GWA) and candidate gene studies have identified a number of genes contributing to the etiology of schizophrenia but which gene exactly causes the disease has not been clarified due to poor replication of an initial finding. Therefore, how to confirm a susceptibility gene remains an important topic of schizophrenia research.ObjectivesThe present study aimed to investigate 4 known genes associated with schizophrenia, including PLA2G4A,TNXB,KPNB3and PEMT genes, using bioinformatics tools and in the combined family and case-control samples in order to reveal their genetic association with schizophrenia and their genetic machnism of gene-gene interaction for the idsease..MethodsThe study samples were composed of 371 parental-offspring family trios, 950 unrelated patients with schizophrenia and 1095 unrelated control subjects. All the subjects are of Han Chinese origin. These patients were recruited through 3 psychiatric hospitals based in Jilin Provence, including Siping Psychiatric Hospital, Changchun Kaixuan Hospital and Jilin Psychiatric Hospital during the period between 2000 and 2009. Diagnosis of these patients as having schizophrenia was made by at least 2 consultant psychiatrists according to the ICD-10 or CCMD-II-R criteria. The control subjects were randomly recruited from the individuals who attended health suiveilence at the Centre for Disease Control and Prevention of Jilin Province during the same period.All the subjects gave informed consent to donating a 5-ml blood sample for genetic analysis. Genomic DNA was extracted by the phenol-chloroform method to establish a DNA sample biobank. Four SNPs, which are present in the PLA2G4A(rs10798059),TNXB (rs204887),KPNB3(rs626716) and PEMT(rs4646396) genes, were selcted to genotype using PCR-based restriction fragment length polymorphism (RFLP) analysis to detect a disease association signal. The goodness-of-fit chi-square (χ2) was applied to test the genotypic distributions of each SNP for Hardy-Weinberg equilibrium (HWE) in both unrelated patients with schizophrenia and the unrelated control subjects. The UNPHASED (version 3.0.12) program was used to test for allelic and genotypic assocaition, the gene-gene interaction and the quantitative trait association (with negative syndrome only). SPSS for Windows 13.0 was applied to test for allelic and genotypic association with positive symptoms, schizophrenia subtypes and premorbid personality.Results⑴Hardy-Weinberg equilibriumSNP rs10798059 in the PLA2G4A gene was not in HWE due to a high frequecy of the AA homozygous genotype in the patient group, suggesting the possibilify of an involvement of the PLA2G4A gene in schizophrenia. A disease association could be observed if the samples size were enlarged. The genotypic distributions of rs10798059 did not deviate from HWE in the control group(p>0.05). The genotypic distributions of the other 3 SNPs were in HWE in both the patient group (p>0.05)and the control group(p>0.05).⑵Allelic and genotypic associations of SNPs with schizophreniaIn the case-control sample, the 4 SNPs studied did not show allelic and genotypic associations with schizophrenia. In the famiily trio sample, rs10798059 and rs204887 did not show allelic and genotypic association with the disease, but the disease association was shown for rs626716 (p=0.006 for allelic and p=0.008 for genotypic) for rs4646396 (p=0.005 for allelic and p=0.0019 for genotypic).⑶Allelic and genotypic association with clinical phenotypesNone of these 4 SNPs studied showed assocaition with clinical subtypes and premorbid personality. In analysis of allelic association, however, rs10798059 was associated with delusion of observation (p=0.003) and delution of negation (p=0.011); rs4646396 was associated with delusion of observation (p=0.001), experience of being revealed(p=0.004), delusion of persecution (p=0.047) and delution of jealousy(p=0.039); rs204887 was associated with delusion of negation (p=0.011) and violent disorder (p= 0.013); and rs626716 was associated only with delusion of observation (p=0.021).In analysis of genotypic association, rs10798059 was associated with delusion of observation (p=0.014)and delusion of negation (p= 0.013); rs4646396 was associated with experience of being revealed (p=0.040) and delusion of jealousy(p=0.004); rs204887 was associated with delusion of negation (p=0.03)and violent disorder (p=0.042); and rs626716 was associated only with delusion of observation(p=0.035).Quantitative analysis showed that only rs204887 was associated with abulia in the male sample(p=0.035) but not either in the female sample or in the combined family and case-control samples.⑷The roles of the gene-gene interaction in developing schizophreniaWe applied the KPNB3 and PEMT genes, which have been reported to be associated with schizophrenia in previous studies, as a conditional locus to analyse the gene-gene interaction using the trans-phase interaction model of UNPHASED.When the KPNB3 gene was used as a conditional locus, allelic interaction was shown for the rs626716(C)-rs4646396(C) combination (p=0.0091) and the rs626716(C)-rs4646396(T) combination (p=0.025) in the case-control sample;allelic interaction was also shown for the rs626716(C)–rs10798059(G) combination (p=0.010) , the rs626716(T)–rs10798059(G) combination (p=0.0041) and the rs626716(C)– rs4646396(C) combination (p=0.002)in the family trio sample. In the combined case-control and family samples, allelic interaction was shown for the rs626716(C)–rs10798059(G) combination (p=0.015) , the rs626716(C)–rs204887(C) combination (p=0.030) and the rs626716(T)– rs4646396(T) combination(p=0.0068).When the KPNB3 gene was used as a conditional locus, genotypic interaction was shown for the rs626716(C/T)-rs4646396(C/C) combination (p=0.033) and the rs626716(C/T)- rs4646396(T/T) combination (p=0.028) in the case-control sample; genotypic interaction was also shown for the rs626716(T/T)– rs10798059(G/A) combination (p=0.028) , the rs626716(C/T)– rs204887(T/T) combination (p=0.0032), rs626716(C/T)– rs4646396(C/C) combination (p=0.0018) and the rs626716(T/T)– rs4646396(T/T) combination (p=0.029). In the combined case-control and family samples, genotypic interaction was shown for the rs626716(T/T)–rs10798059(G/A) combination (p=0.046),the rs626716(C/T)–rs204887(T/T) combination (p=0.0027), the rs626716(T/T)–rs204887(T/T) combination (p=0.0033), the rs626716(C/T)– rs4646396(C/C) combination (p=0.031) and the rs626716(T/T)– rs4646396(T/T) combination (p=0.0002), respectively.When the PEMT gene was used as a conditional locus, allelic interaction was not shown either for the rs4646396-rs10798059 combination (p>0.05)or the rs4646396- rs204887 combination (p>0.05) in the case-control samples, but the interaction was shown for the rs4646396(C)– rs10798059(G) combination in the family trio sample(p=0.030);In the combined case-control and family samples, allelic association was shown for the rs4646396(C)– rs10798059(G) combination (p=0.031),the rs4646396(T)– rs10798059(G) combination (p=0.0145),the rs4646396(C)– rs204887(C) combination (p=0.017) and the rs4646396(T)– rs204887(C) combination (p=0.0071). Similarly, genotypic interaction was not shown either for the rs4646396-rs10798059 combination (p> 0.05)or the rs4646396- rs204887 combination (p>0.05) in the case-control samples, but the interaction was shown for the rs4646396(C/C)–rs204887(C/C) combinationin (p=0.019) in the family trio sample. In the combined case-control and family samples, genotypic association was shown for the rs4646396(C/C)– rs204887(C/C) combination (p=0.025) and the rs4646396(T/T– rs204887 (C/C) combination (p=0.036).ConclusionBased on the above results, it can be concluded:①that the KPNB3(rs626716)and PEMT(rs4646396) loci are associated with schizophrenia;②that the PLA2G4A(rs10798059)locus was not associated with schizophrenia but the higher frequency of the AA genotype in patients with schizophrenia suggests that the PLA2G4A association with schizophrenia cannot be ruled out in the Chinese population;③that the TNXB(rs204887)locus was associated only with abulia but not with the disease itself;④that the PLA2G4A (rs10798059) locus was associated with delusion of observation and delusion of negation; the TNXB (rs204887) locus was associated with delusion of observation and violent disorder; the KPNB3 (rs626716) locus was associated with delusion of observation; and the PEMT (rs4646396) locus was associated with both experience of being revealed and delusion of jealousy;⑤that none of the 4 genes studied was associated with premorbid personality and clinical subtypes of schizophrenia;⑥that schizophrenia is a complex disease, and polygenic and heterogeneous characteristcs appear to be closely linked;⑦that when the KPNB3 gene was used as a conditional locus, the interaction between KPNB3 rs626716 C/T or T/T genotype and the TNXB rs204887 T/T genotype was associated with increased risk of schizophrenia;⑧that when the KPNB3 gene was also used as a conditional locus, the interaction between KPNB3 rs626716 T/T genotype and the PEMT rs4646396 T/T genotype was associated with increased risk of schizophrenia;⑨that when the PEMT gene was used as a conditional locus, the interaction between the PEMT gene and the other 3 genes was associated with increased risk of schizophrenia.Taken together, both the KPNB3 gene and the PEMTgene are very likely to play a role in the pathogenesis of schizophrenia and the interaction between these 2 genes may be involved in increased risk of schizophrenia. Both polygenic transmission and genetic heterogeneity may co-exist and make it more complicated than expected to reveal the exact mechanism of triggering the onset of the disease. On the basis of recent progress of schizophrenia research, the present study analysed the combined case-control and family samples for genetic association with schizophrenia. Such a novel strategy is very useful to rule out false positive results as reported in previous studies and to illustrate the precise mechanism of the pathogenesis of schizophrenia.