Shih-Hsin Hsu

Role of the DLGAP2 Gene Encoding the SAP90/PSD-95-Associated Protein 2 in Schizophrenia

Aberrant synaptic dysfunction is implicated in the pathogenesis of schizophrenia. The DLGAP2 gene encoding the SAP90/PSD-95-associated protein 2 (SAPAP2) located at the post-synaptic density of neuronal cells is involved in the neuronal synaptic function. This study aimed to investigate whether the DLGAP2 gene is associated with schizophrenia. We resequenced the putative promoter region and all the exons of the DLGAP2 gene in 523 patients with schizophrenia and 596 non-psychotic controls from Taiwan and conducted a case-control association analysis. We identified 19 known SNPs in this sample. Association analysis of 9 SNPs with minor allele frequency greater than 5% showed no association with schizophrenia. However, we found a haplotype (CCACCAACT) significantly associated with schizophrenia (odds ratio:2.5, p<0.001). We also detected 16 missense mutations and 1 amino acid-insertion mutation in this sample. Bioinformatic analysis showed some of these mutations were damaging or pathological to the protein function, but we did not find increased burden of these mutations in the patient group. Notably, we identified 5 private rare variants in 5 unrelated patients, respectively, including c.−69+9C>T, c.−69+13C>T, c.−69+47C>T, c.−69+55C>T at intron 1 and c.−32A>G at untranslated exon 2 of the DLGAP2 gene. These rare variants were not detected in 559 control subjects. Further reporter gene assay of these rare variants except c.−69+13C>T showed significantly elevated promoter activity than the wild type, suggesting increased DLGAP2 gene expression may contribute to the pathogenesis of schizophrenia. Our results indicate that DLGAP2 is a susceptible gene of schizophrenia.

Genetic analysis of the DLGAP1 gene as a candidate gene for schizophrenia.

Schizophrenia is a severe chronic mental disorder with high genetic components in its etiology. Several studies indicated that synaptic dysfunction is involved in the pathophysiology of schizophrenia. Postsynaptic synapse-associated protein 90/postsynaptic density 95-associated proteins (SAPAPs) constitute a part of the N-methyl-d-aspartate receptor-associated postsynaptic density proteins, and are involved in synapse formation. We hypothesized that genetic variants of the SAPAPs might be associated with schizophrenia. Thus, we systemically sequenced all the exons of the discs, large (Drosophila) homolog-associated protein 1 (DLGAP1) gene that encodes SAPAP1 in a sample of 121 schizophrenic patients and 120 controls from Taiwan. We totally identified six genetic variants, including five known SNPs (rs145691437, rs3786431, rs201567254, rs3745051 and rs11662259) and one rare missense mutation (c.1922A>G) in this sample. SNP- and haplotype-based analyses showed no association of these SNPs with schizophrenia. The c.1922A>G mutation that changes the amino acid lysine to arginine at codon 641 was found in one out of 121 patients, but not in 275 control subjects, suggesting it might be a patient-specific mutation. Nevertheless, bioinformatic analysis showed this mutation does not affect the function of the DLGAP1 gene and appears to be a benign variant. Hence, its relationship with the pathogenesis remains to be investigated.

Genetic and functional analyses of early growth response (EGR) family genes in schizophrenia.

OBJECTIVE Early growth response genes (EGR1, 2, 3, and 4) encode a family of nuclear proteins that function as transcriptional regulators. They are involved in the regulation of synaptic plasticity, learning, and memory, and are implicated in the pathogenesis of schizophrenia. METHODS We conducted a genetic association analysis of 14 SNPs selected from the EGR1, 2, 3, and 4 genes of 564 patients with schizophrenia and 564 control subjects. We also conducted Western blot analysis and promoter activity assay to characterize the EGR genes associated with schizophrenia RESULTS We did not detect a true genetic association of these 14 SNPs with schizophrenia in this sample. However, we observed a nominal over-representation of C/C genotype of rs9990 of EGR2 in female schizophrenia as compared to female control subjects (p=0.012, uncorrected for multiple testing). Further study showed that the average mRNA level of the EGR2 gene in the lymphoblastoid cell lines of female schizophrenia patients was significantly higher than that in female control subjects (p=0.002). We also detected a nominal association of 4 SNPs (rs6747506, rs6718289, rs2229294, and rs3813226) of the EGR4 gene that form strong linkage disequilibrium with schizophrenia in males. Reporter gene assay showed that the haplotype T-A derived from rs6747506 and rs6718289 at the promoter region had significantly reduced promoter activity compared with the haplotype A-G. CONCLUSION Our data suggest a tendency of gender-specific association of EGR2 and EGR4 in schizophrenia, with an elevated expression of EGR2 in lympoblastoid cell lines of female schizophrenia patients and a reduced EGR4 gene expression in male schizophrenia patients.

Chronic methamphetamine treatment reduces the expression of synaptic plasticity genes and changes their DNA methylation status in the mouse brain

Methamphetamine (METH) is a highly addictive psychostimulant that may cause long-lasting synaptic dysfunction and abnormal gene expression. We aimed to explore the differential expression of synaptic plasticity genes in chronic METH-treated mouse brain. We used the RT(2) Profiler PCR Array and the real-time quantitative PCR to characterize differentially expressed synaptic plasticity genes in the frontal cortex and the hippocampus of chronic METH-treated mice compared with normal saline-treated mice. We further used pyrosequencing to assess DNA methylation changes in the CpG region of the five immediate early genes (IEGs) in chronic METH-treated mouse brain. We detected six downregulated genes in the frontal cortex and the hippocampus of chronic METH-treated mice, including five IEGs (Arc, Egr2, Fos, Klf10, and Nr4a1) and one neuronal receptor gene (Grm1), compared with normal saline-treated group, but only four genes (Arc, Egr2, Fos, and Nr4a1) were confirmed to be different. Furthermore, we found several CpG sites of the Arc and the Fos that had significant changes in DNA methylation status in the frontal cortex of chronic METH-treated mice, while the klf10 and the Nr4a1 that had significant changes in the hippocampus. Our results show that chronic administration of METH may lead to significant downregulation of the IEGs expression in both the frontal cortex and the hippocampus, which may partly account for the molecular mechanism of the action of METH. Furthermore, the changes in DNA methylation status of the IEGs in the brain indicate that an epigenetic mechanism-dependent transcriptional regulation may contribute to METH addiction, which warrants additional study.

Exonic resequencing of the DLGAP3 gene as a candidate gene for schizophrenia

We resequenced the exonic regions of the DLGAP3 gene, which encodes SAP90/PSD95-associated protein 3, in 215 schizophrenic patients and 215 non-psychotic controls. Seven known single-nucleotide polymorphisms (SNPs) were identified, but not associated with schizophrenia. Nevertheless, we identified several rare missense mutations and some of them might be associated with the pathogenesis of schizophrenia.

Rare mutations and hypermethylation of the ARC gene associated with schizophrenia.

Activity-regulated cytoskeleton-associated protein (ARC), which interacts with the N-methyl-d-aspartate receptor (NMDAR) complex, is a critical effector molecule downstream of multiple neuronal signaling pathways. Dysregulation of the ARC/NMDAR complex can disrupt learning, memory, and normal brain functions. This study examined the role of ARC in susceptibility to schizophrenia. We used a resequencing strategy to identify the variants of ARC in 1078 subjects, including patients with schizophrenia and normal controls. We identified 16 known SNPs and 27 rare mutations. SNP-based analysis showed no association of ARC with schizophrenia. In addition, the rare mutations did not increase the burden in patients compared with controls. However, one patient-specific allele in the putative ARC promoter region and seven patient-specific mutants in ARC exon regions significantly reduced the reporter gene activity compared with ARC wild-type. Methylation of a putative ARC promoter attenuated reporter activity in vitro, suggesting that ARC expression is regulated by DNA methylation. Pyrosequencing revealed eight hypermethylated CpG sites in the putative ARC promoter region in 64 schizophrenic patients compared with 63 controls. Taken together, our results suggest that both rare variants and epigenetic regulation of ARC contribute to the pathogenesis of schizophrenia in some patients.

Genetic and functional analysis of the gene encoding GAP-43 in schizophrenia

OBJECTIVES In earlier reports, growth-associated protein 43 (GAP-43) has been shown to be critical for initial establishment or reorganization of synaptic connections, a process thought to be disrupted in schizophrenia. Additionally, abnormal GAP-43 expression in different brain regions has been linked to this disorder in postmortem brain studies. In this study, we investigated the involvement of the gene encoding GAP-43 in the susceptibility to schizophrenia. METHODS We searched for genetic variants in the promoter region and 3 exons (including both UTR ends) of the GAP-43 gene using direct sequencing in a sample of patients with schizophrenia (n=586) and non-psychotic controls (n=576), both being Han Chinese from Taiwan, and conducted an association and functional study. RESULTS We identified 11 common polymorphisms in the GAP-43 gene. SNP and haplotype-based analyses displayed no associations with schizophrenia. Additionally, we identified 4 rare variants in 5 out of 586 patients, including 1 variant located at the promoter region (c.-258-4722G>T) and 1 synonymous (V110V) and 2 missense (G150R and P188L) variants located at exon 2. No rare variants were found in the control subjects. The results of the reporter gene assay demonstrated that the regulatory activity of construct containing c.-258-4722T was significantly lower as compared to the wild type construct (c.-258-4722G; p<0.001). In silico analysis also demonstrated the functional relevance of other rare variants. CONCLUSIONS Our study lends support to the hypothesis of multiple rare mutations in schizophrenia, and it provides genetic clues that indicate the involvement of GAP-43 in this disorder.

Resequencing of early growth response 2 (EGR2) gene revealed a recurrent patient-specific mutation in schizophrenia

Abnormal myelination is considered as part of the pathophysiology of schizophrenia. We resequenced the genomic DNA of the EGR2, which has a specific function in the myelination of peripheral nervous system, in 543 schizophrenic patients and 554 non-psychotic controls. We identified six known SNPs, which were not associated with schizophrenia. Nevertheless, we discovered 24 rare mutations, some of them were patient-specific, including a recurrent mutation (p.P173_Y174insP), which might be associated with the pathogenesis of schizophrenia.

Mutation analysis of the WNT7A gene in patients with schizophrenia

Aberrant WNT signaling has been implicated in the pathophysiology of schizophrenia. WNT7A, a member of the WNT gene family, is considered a potential candidate of schizophrenia. All exons of WNT7A in 570 schizophrenic patients and 563 controls were sequenced, and protein functional analysis was conducted. Five common variants were identified, but none were noted to be associated with schizophrenia. Nevertheless, nine rare mutations, including one schizophrenia-specific missense mutation (c.305G > A), were discovered. However, immunoblot analysis findings revealed that the c.305G > A mutation did not affect protein expression. These results suggest that WNT7A is unlikely to be associated with susceptibility to schizophrenia.

DNA methylation analysis of the EGR3 gene in patients of schizophrenia

DNA methylation has been implicated in the pathogenesis of schizophrenia. EGR3 is considered as a potential candidate gene for schizophrenia. We conducted in vitro DNA methylation reaction, Lucia luciferase activity assay, and pyrosequencing assay to assess the DNA methylation of the EGR3 expression underlying the pathophysiology of schizophrenia. We found that DNA methylation of the putative EGR3 regulatory regions attenuated Lucia luciferase activity. There was no difference in the DNA methylation pattern of EGR3 between in 50 schizophrenic patients and 47 controls. Our data suggest that DNA methylation regulated the expression of EGR3 might not be associated with schizophrenia.