The Correlation Of DISC1 Pathway Gene NDEL1, PCM1 Polymorphisms With Schizophrenia In Chinese Southern Han Population

BackgroundAt present, many large twin studies and adoptee studies confirmed that the schizophrenia is a typical polygenic disease. Research in Finland found that the heritability of schizophrenia was up to 83%。Wang Weimin have found that the heritability of schizophrenia in the Han population in China was 70.85%. This indicated that the genetic factor play a dominant role (more than 70%) in schizophrenia susceptibility, while the environmental factor play a secondary role (less than 30%). By complex segregation analysis, linkage analysis, affected sib pair analysis, and genome-wide association study(GWAS) method, a large number of schizophrenia susceptibility loci (1q21-22,6p22-24,8p21-22,13q14-32 etc.) and susceptible genes (DISCI, COMT, BDNF, PLXNA2, CSF2R, PCM1, ZNF804A, RELN) were found. Disrupted in schizophrenia 1 (DISCI) is a currently definite schizophrenia susceptibility gene, it was found in a high developed mental disease Scotland pedigree which have a balanced chromosomal translocation. The research on the correlation between DISCI and schizophrenia showed DISCI and schizophrenia has significant correlation. DISCI protein is a scaffold protein,the protein interaction group and a large number of interacting proteins called the DISC1 pathway. Bradshaw’s study displayed that DISC1 and DISC binding proteins were likely one of the molecular mechanism of schizophrenia and many other mental illness. As two important DISC binding proteins, the association between NDEL1, PCM1 and schizophrenia were confirmed by a number of studies. Results Lipska’s study showed that compared with control, expression of NDEL1 in schizophrenia patients in the hippocampus and dorsolateral prefrontal cortex significantly reduced. Tomppo’s study in the Finland population showed significant correlation between rs 17806986 of NDEL1 gene and schizophrenia. In the same year, Ikeda found that SNP:rs3744652, rs8064655 of NDEL1 gene were not related to schizophrenia in a Japanese population. The latest study found that the NDEL1 oligopeptidase activity in plasma of schizophrenia patients was significantly lower than that of healthy people, suggesting that the peptidase activity of plasma NDEL1 may become biomarker for the diagnosis of schizophrenia. Datta’ study found significantly correlation between rs208747, rs445422, rsl3276297, rs370429 of PCM1 gene and schizophrenia in England population. While Hashimoto’s study showed that rs445422, rs3780103, rs3214087, rs370429 genotype frequency and allele frequency had no significant difference between schizophrenia and normal control in the study of Japanese population. The above literature shows that there are significant association between NDEL1 gene and PCM1 gene and schizophrenia exists in the European population, while the association between the two genes and schizophrenia was not confirmed by the repeated study in the Japanese population. The reasons are the following:first, the false positive association really right not be repeated; secondly, true associations were not confirmed in repeated research with low statistical performance; finally, a true association was not confirmded in the the repeated study in different populations due to the genetic heterogeneity of complex diseases. There are not repeated studies of these associations in Chinese Han population. In view of repetitive study is necessary to ensure the association between gene and disease, and, wse design this case-control study with the aim of clarifying the relationship between NDEL1, PCM1 and schizophrenia in Chinese southern Han population.Objective1 To repeat and comfirm the correlation between rs17806986 of NDEL1 and rsl3276297 and rs208747 of PCM1 and schizophrenia in Chinese southern Han people;2 To investigate the correlation between some tagSNPs of NDEL1, PCM1 and schizophrenia in Chinese southern Han people;3 To investigate the correlation between schizophrenia related SNP polymorphisms and schizophrenia patient’s age of first onset, frequency of hospitalization, family history and clinical typing.Objects and methods1 Subjects A total of 150 subjects,100 schizophrenia and 50 matched healthy control, were randomly enrolled in this case-control study. All case participants were randomly selected from the department of psychiatry, Shenzhen Kangning Hospital from 2013 June to 2014 June. The case group included 53 male and 47 female, mean age was 32.72±10.14 years. All healthy control participants were randomly selected from the medical examination department of Shenzhen Longgang Central Hospital from 2013 June to 2014 June. The control group included 28 males,22 females, mean age was 32.24±8.73 years. Requirements of the case group are as follows:The diagnosis of schizophrenia is made by two psychiatric physician according to DSM-IV; The biological parents of participants are Han people; The age of patients are 18-60 years old, male or female is not limited; The patients, their family members or other legal guardians understand the full content and nature of the research and signed the informed consent. The patients who have mood disorders, mental retardation, schizoaffective disorder or other cognitive impairment and serious physical diseases are excluded. Requirements of healthy control group are as follows: Their biological father, mother were Han people; The age of subjects are 18-60 years old, male or female is not limited; Gender, age, and other general demographic data are matched with case group; No relation to the case group; The subjects understand the full content and nature of the research and signed the informed consent. The subject who have a history of neuropsychiatric disease, with schizophrenia family history or suffering from severe physical disease or mental disorders are excluded.2 Gene polymorphism detection(1) The tagSNP selection:The within and 10KB upstream and downstream Chinese in Beijing Han population (CHB) genotype data packets (Release28, Phase Ⅰ+Ⅱ+Ⅲ) of NDEL1 gene and PCM1 gene are retrieved and downloaded from the international human genome haplotype map project website (http://hapmap.ncbi.nlm. nih. gov/). The downloaded genotype data packets are analyzed through Haploview 4.2 in order to screen the haplotype tagging SNP (TagSNP) of NDEL1 gene and PCM1 gene.(2) Genomic DNA preparation:The venous blood of every object was collected by using the EDTA-2K vacuum vessel through elbow vein. The genomic DNA was extracted from peripheral whole blood cell by using of whole blood genomic DNA rapid extraction kit (solution) which was made by BioTeke Corporation. The purity of DNA was identificated by using 1% agarose gel electrophoresis. The concentration and purity of DNA samples was measured by using of nucleic acid and protein analyzer.(3) Genotype detection:The genotyping of selected SNPs were carried out using MassARRAY(?) molecular weight array technology from Sequenom Company. The technique integrate primer extension, cutting reaction and sensitive, reliable matrix assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS) technology to realize the genotype detection. First of all, amplifying a DNA fragment containing the SNP locus through the PCR, and then use the shrimp alkaline phosphatase (SAP) to remove the residual deoxyribonucleoside three phosphoric acid (dNTP) and primer in the system, then add a single base extension primer, the 3’end base of primer is close to the SNP site, with four 2’,3’-dideoxy nucleotide (ddNTP) instead of dNTP. In this way, the probe extends only a base at SNP loci, alleles corresponding to ddNTP and SNP sites on the connection site. Finally, the molecular weight differences between the primers and extended product was detected through MALDI-TOF MS to determine the base type at the point, and then determine the genotype of the SNP site.(4) Statistical analysis:Differences in age between the two groups were analyzed by independent samples T test through SPSS 17.0 software. Differences in gender distribution between the two groups were analyzed by x2 test through SPSS17.0 software. Hardy Weinberg equilibrium test of control group were analyzed with goodness of fit χ2 test on SPSS 17.0 software. Using the χ2 test to compare the SNP allele frequency differences between case and control on SPSS 17.0 software. Non conditional Logistic regression analysis was used to analyze the association between each SNP locus and schizophrenia in five different genetic mode using (codominant inheritance, autosomal dominant, recessive, overdominance, additive genetic correlation) on the SNPstats software (http://bioinfo.iconcologia.net/SNPstats). For multiple testing problems of the research, we used the method of BH under False Discovery Rate (FDR) standard to complete the multiple correction. The linkage disequilibrium between different SNPs was analyzed by Haploview4.2 software. Haploview 4.2 software was used to set haplotype block, and then, compared different haplotypes between the two groups. Differences in schizophrenic patients’ clinical features between different genotype were analyzed by SPSS 17.0 software,χ2 test was used to analyze classification variables; if the continuous variable data distribution accords with the normal distribution, choice one-way analysis, if the data distribution does not accord with the normal distribution, selected Kruskal-Wallis H multiple groups of rank sum test.Result1 Association between NDEL1 and schizophrenia:(1) Allele frequency:The results showed that there are not significant association between the frequency of SNP allele and schizophrenia. After stratification of gender, there are still not significant association between the frequency of SNP allele and schizophrenia in both male and female.(2) Genotype distribution:The genetic model with lowest AIC and BIC value is the most suitable genetic model of this SNP. Then correct the P value of each SNP under it’s most suitable genetic model by BH method based on FDR, the corrected P value represent as PFDR.Results showed that there are significant correlation between rs17806986 and schizophrenia in the ultra dominant inheritance mode (PFDR=0.027, OR=2.96,95%CI:1.42-6.17), shows that G/C genotype significantly increased the risk of schizophrenia. The risk of G/C genotype carriers with schizophrenia is 2.96 times as the carriers of G/G or C/C. genotype carriers. The remaining SNPs showed no significant association with schizophrenia after FDR correction. Correlation genotype distribution of rs 17806986 polymorphism with schizophrenia was analyzed between case group and control group of different gender. The results showed that the rs17806986 genotype distribution at the ultra dominant genetic model was significantly correlated with schizophrenia in male population (P=0.015, genotype G/C compared to G/G and C/C, OR=3.26,95%CI:1.22-8.72). The risk of G/C carriers with man schizophrenia was 3.26 times as the carriers of G/G genotype or C/C genotype. There was not significant association between rs17806986 genotype distribution and women schizophrenia.(3) Linkage disequilibrium test and haplotype association analysis:The r2 between rs17806986 and the other four SNPs were less than 0.33, indicating that there was not linkage disequilibrium between rs17806986 and the rest four SNPs. The r2 between rs8065812, rsl391766, rs8064655, rs931672 were close to 1, suggesting that the four SNP were closely linked. Rs8065812, rsl391766, rs8064655 and rs931672 4 constituted a haplotype block, which was mainly composed of two haplotypes:CGCC and TCTT There was not significant correlation in the general population and male or the female population between the two haplotypes and schizophrenia.(4) Clinical characteristics in schizophrenia patients with rs17806986:There was not significant correlation between rs17806986 genotype distribution and schizophrenia patients’ age of first onset, frequency of hospitalization, family history and clinical typing.2 Association between PCM1 and schizophrenia:(1) Association between allele frequency and schizophrenia:The result showed that there was not significant association between allele frequencies of 5 selected SNPs in PCM1 and schizophrenia. After gender stratification, there were not significant associations between 5 SNPs allele frequencies in both male and female people.(2) Association between genotype distribution and schizophrenia:There are not significant association between rs2285310, rs208062, rs208022, rs208756 genotype distribution and schizophrenia under 5 different genetic models. There are only C/C, T/C genotype found in rsl 3276297, which lead to the SNP can’t be analyzed under 5 kinds of genetic models. So the non conditional Logistic regression method (SNPstats software) was used to analyze the association between C/C, T/C genotype and schizophrenia. Results show that there are not significant association in both male and female between rs13276297 genotype distribution and schizophrenia. After gender stratification, there are not significant association between 5 SNPs genotype distribution and schizophrenia under 5 different genetic models.(3) Linkage disequilibrium test and haplotype based association analysis:There are strong linkage disequilibrium between rs208756-rs2285310 (r2=0.71), rs208756-rs208022 (r2=0.56) and rs208022-rs2285310 (r2=0.37), while SNPs does not exist or exist only weak linkage disequilibrium. There are not any haplotype block between this 5 SNPs.Conclusion(1) NDEL1 gene may be a susceptibility gene of schizophrenia in Chinese southern Han population, the genotype distribution of rs17806986 site in intron 1 showed significant association with schizophrenia and the the G/C genotype can increase the schizophrenia risk by 2.96 times. This correlation is only found in male, no correlation was found between the female. There was not significant correlation between schizophrenia related SNP polymorphisms and schizophrenia patient’s age of first onset, frequency of hospitalization, family history and clinical typing.(2) The significant association between rs13276297 gene polymorphism of PCM1 with schizophrenia which was found in European populations was not confirmed in Chinese southern Han people. There was not significant difference between allele frequency, genotype distribution of other 4 SNPs and schizophrenia. The PCM1 gene may not be schizophrenia susceptibility gene in Chinese southern Han.