Genetic Research Into A Schizophrenia-associated Gene On Chromosome 22q11-12

Schizophrenia is a serious mental disorder with a lifetime prevalence of 1% in thegeneral population worldwide. It was characterized by the abnormal mental functions anddisturbed behaviors, which characteristically appear as a series of clinical features, suchas positive and negative symptoms, and disturbances in basic cognitive functions. Boththe incidence and the prevalence of psychiatric diseases are gradually increasing year byyear with the development of social economy and the change of a life style and theprevalence of schizophrenia is the highest one among them. The investigation of the causefor schizophrenia has always drawn great attention from the public.In the past decades, people have studied all the risk factors related to schizophrenia,especially the genetic factor to which a high attention was paid. Work on geneticepidemiology shows that schizophrenia is not a simple Mendelian disease but it is verylikely to be a complex disease with a polygenic mechanism.The research into schizophrenia susceptibility genes has been a hot spot of mentaldisease. The completion of the Human Genome Project (HGP) has provided a goodopportunity for mapping all the genes involved in human diseases, includingschizophrenia. Basically, there are two steps for mapping a disease-related gene in thehuman genome, the linkage-based genome-wide scan and the regional mapping withlinkage disequilibrium (LD) analysis. Genome-wide scanning shows the consistentlinkage findings on some chromosome regions, such as 1q21-22, 5q22-23, 6p24-21,8p22-21, 13q14-33 and 22q11-12.At present, great attention had been paid to the small interstitial deletions ofchromosome 22q11(del22q11), which clinically causes the velo-cardio-facial syndrome(VCFS) or the del22q11 syndrome. The relationship between the del22q11 syndrome andschizophrenia had been noted by many psychiatrists and researchers. Several studiesshowed that over 20-30%of patients with VCFS had expressed psychotic symptoms.Moreover, linkage studies have also suggested a schizophrenia susceptibility genepossibly located within the del22q11 region.The present study had focused on identifying the candidate susceptibility genes onthe 22q11-12 region using a family-based LD analysis and case-control analysis, and wasdesigned to construct a SNP-based LD map. Technique route was followed in such a waythat from positive region determined by genome scanning to gene map, and SNP-basedLD map.A total of 296 family trios of Chinese Han descent, consisting of healthy fathers,healthy mothers and affected offspring with schizophrenia, and 400 unrelated patientswith schizophrenia and 419 unrelated healthy individuals were recruited. The SNP-basedLD map was constructed between the PRODH gene and the APOL1 gene based on thegene map on the 22q11-12 region and the SNPs on each known gene using bioinformaticsmethods. Six SNPs were chosen in this region. SNPs were genotyped using PCR-basedRFLP analysis. Genotyping data were put into the SPSS database and pedigree database.The Hardy-Weinberg (H-W) equilibrium was tested for genotypic distributions of SNPsusing the chi-square (X2) goodness-of-fit test. The LD between paired SNPs wasestimated with UNPHASED programs (Version 2.404). The haplotype-based haplotyperelative risk (HHRR) test and the transmission/disequilibrium test (TDT) were applied todetect allelic association between an SNP and schizophrenia. Haplotypes consisting oftwo or more SNPs were tested by the UNPHASED program. To elucidate geneticheterogeneity and reduce the false negative results, in addition, schizophrenic patientswere sub-grouped based on their clinical symptoms, and the genetic association betweenSNPs and clinical subgroups were then analyzed. Positive loci were replicated withdifferent family trios to rule out the false positive results.The patients were divided into two groups according to the clinical psychoticsymptoms. We analyzed psychotic symptoms versus allelic and genotypic frequency ofeach SNP.To investigate the association between the UFD1L and CLDN5 genes andschizophrenia, the case-control study was designed. The following are the details ofmethodology and major results obtained in this study.1. DNA markersSix SNPs present in five genes on the 22q11-12 region were chosen by accessing thedatabases at http://www.ncbi.nlm.nih.gov/, http://www.ncbi.nlm.nih.gov/SNP andhttp://snp.cshl.org/ web sites. The candidate SNPs included SNP1 present in the PRODHlocus, SNP2 and SNP3 in the UFD1L locus, SNP4 in the CLDN5 locus, SNP5 and SNP6in the APOL3 and APOL1 locus respectively.2. The H-W equilibriumThe X2 goodness-of-fit test showed that the genotypic distributions of all 6 SNPswere not deviated from the H-W equilibrium, and thus this sample pool was suitable forthe genetic analysis.3. LD between paired SNPsThe estimated LD showed that SNP2 and SNP3 were in the same LD block.4. Association between SNPs and schizophrenia4.1 The TDT analysisThe TDT analysis showed that both SNP2 and SNP4 were associated withschizophrenia (P=0.006 and P=0.007, respectively). The SNP2 is a C to G base changeand heterozygous parents have excessively transmitted allele G to their affected offspring,suggesting that the haplotype containing SNP2 (C) may carry a disease-resistant variantfor schizophrenia. The SNP4 is a C to G base change and heterozygous parents haveexcessively transmitted allele G to their affected offspring, suggesting that the haplotypecontaining SNP4 (C) may carry a disease-resistant variant for schizophrenia.4.2 The HHRR analysisThe HHRR analysis revealed allelic association between two SNPs andschizophrenia, one the SNP2 and the other one the SNP3. These findings were consistentwith TDT results.4.3 Analysis for clinical subgroupsSince the time at which SNPs occurred differs, each SNP may have its own geneticheritage with a different haplotype and non-random association. They may affect eachother in allele frequency distribution, so as to reduce the power to detect their associationwith the illness. The advantage of clinical subgroup analysis was to limit the interferencefrom different LD signals to reduce the false negative results and validate the hypothesisof genetic heterogeneity. The results showed that SNP2 and SNP4 were associated withschizophrenia.4.4 Analysis for haplotype transmission.The analysis for multi-SNP hapoltype transmission is essential to look for a specifichaplotype or chromosome possibly carrying a gene for schizophrenia.4.4.1 A number of haplotype systems were constructed with these 6 SNPs studied,including 5 two-SNP haplotype systems and 4 three-SNP haplotype systems. Thetwo–SNP haplotype systems included SNP1-SNP2, SNP2-SNP3, SNP3-SNP4,SNP4-SNP5 and SNP5-SNP6. The global χ 2 test showed that the SNP1-SNP2haplotype system, the SNP2-SNP3 haplotype system and the SNP4-SNP5 haplotypesystem were associated with schizophrenia.4.4.2 The three-SNP haplotype systems included SNP1-SNP2-SNP3, SNP2-SNP3-SNP4,SNP3-SNP4-SNP5 and SNP4-SNP5-SNP6. The global χ 2 test showed that theSNP1-SNP2-SNP3 haplotype system and the SNP2-SNP3-SNP4 haplotype system wereassociated with schizophrenia.4.4.3 The 1-df test for individual haplotypes showed that the SNP1(G)-SNP2(C)haplotype and the SNP1(G)-SNP2(C)-SNP3(T) haplotype were excessivelynon-transmitted (P<0.05). They may be derived originally from the same haplotypesystem and carry the same disease-resistant allele providing a protective mechanism fromthe attack of schizophrenia. The SNP1(G)-SNP2(G) haplotypes was excessivelytransmitted (P<0.05), this haplotype may contain a susceptibility allele for schizophrenia.4.4.4 The 1-df test for individual haplotypes also showed that the SNP3(T)-SNP4(C)haplotype and the SNP2(C)-SNP3(T)-SNP4(C) haplotype were excessivelynon-transmitted (P<0.05). These two haplotypes may be derived from the same haplotypesystem and carry the same disease resistant allele for schizophrenia. These 2 individualhaplotypes all contain the SNP4(C) allele that shows a negative association withschizophrenia. They may contain the same disease-resistant allele. TheSNP2(G)-SNP3(T)-SNP4(C) haplotype was excessively transmitted and this haplotypemay contain a susceptibility allele for schizophrenia.4.4.5 Similar findings were shown with the SNP4(C)-SNP5(A) haplotype and theSNP3(T)-SNP4(C)-SNP5(A) haplotype. The major segment of two haplotypes wasSNP4(C)-SNP5(A), suggesting that they may contain the same disease resistant allele.4.4.6 Similar findings were also shown with the SNP4(G)-SNP5(A) haplotype and theSNP3(T)-SNP4(G)-SNP5(A) haplotype. The major segment of two haplotypes wasSNP4(G)-SNP5(A), suggesting that they may contain the same disease susceptibilityallele for schizophrenia.5. The association between SNPs and psychotic symptoms of schizophreniaBecause the genotypic distributions of all 6 SNPs were in the H-W equilibrium, wewere able to analyze psychotic symptoms versus allelic and genotypic frequency of eachSNP using a case-control design. To investigate the association between SNPs andpositive and negative symptoms of schizophrenia, the SPSS program and theUNPHASED program were applied for the statistical analysis. The results were shown asfollows:5.1 The SNP3 was associated with genuine auditory hallucination and incoherence ofthinking (P=0.045 and P=0.009). The frequency of allele T was greater in the patientswith genuine auditory hallucination than those without the symptom, lower in the patientswith incoherence of thinking than those without the symptom.5.2 The SNP4 was associated with incoherence of thinking and aggressive behavior(P=0.016 and P=0.013). The frequency of allele G was greater in the patients withincoherence of thinking than those without the symptom. The frequency of allele C waslower in the patients with aggressive behavior than those without the symptom.5.3 The SNP6 was associated with delusion of being revealed (P=0.016). The frequencyof allele A was lower in the patients with delusion of being revealed than those withoutthe symptom.5.4 The quantitative trait analysis showed that the SNP4 was associated with poverty ofthought (P=0.038).6. The association between SNPs and sex of schizophreniaThe HHRR analysis showed that the parents have excessively transmitted SNP2 allele Gto their female affected offspring (P=0.022), and SNP4 allele G to their male affectedoffspring (P=0.002). These results suggested that SNP2 was associated with femaleschizophrenia and SNP4 was associated with male schizophrenia.7. Analysis for case-controlThe SNP2 was associated with schizophrenia. There was a significant difference infrequency of allele and genotype between the patient group and the control group. Thefrequency of allele G was greater in the patient group than that in the control group.The association could be also showed between SNP4 and schizophrenia. There was asignificant difference in frequency of allele and genotype between the patient group andthe control group. The frequency of allele G was greater in the patient group than that inthe control group.Taken together, the present study demonstrates that there may be one or more diseasesusceptibility genes located on the 22q11-12 region. The UFD1L gene and the CLDN5gene may be associated with schizophrenia. SNP2 may be associated with schizophreniain female and SNP4 may be associated with schizophrenia in male. The 3 SNPs, includingSNP3, SNP4 and SNP6, were associated with some positive symptoms of schizophrenia.The SNP4 was also found to be associated with poverty of thought that was classified as anegative symptom of schizophrenia. These findings are very important for elucidating thegenetic mechanisms of schizophrenia at a molecular level, and also for the developmentof genetic diagnosis, new drugs for the treatment of the illness and prediction ofschizophrenia risk.