The interaction of RELN-DNMT genes involving in neurotrophin signaling pathway contributes to schizophrenia susceptibility

Objective: Schizophrenia belongs to a severe mental illness with complicated clinical presentations, an ill-defined pathogenesis, and no known cause. Many genetic studies imply that polygenic interaction is important in the development of schizophrenia. The main mechanism of the RELN-BDNF-CREB-DNMT signaling pathway in neurodevelopment involves RELN, brain-derived neurotrophic factor (BDNF), transcription factor cyclic adenosine monophosphate response element binding protein (CREB), DNA methyltransferase 1 (DNMT1), as well as DNA methyltransferase 3B (DNMT3B). An early case-control research on 15 polymorphisms in the RELN, CREB, BDNF, DNMT1, and DNMT3B genes was done. A single gene variation has little effect on the pathogenesis of schizophrenia, but the combination of intergenic variation loci has a bigger impact because schizophrenia is a complex polygenic disorder. The objective of the current study sought to explore the impact of genetic interactions between RELN, BDNF, CREB, DNMT1, and DNMT3B on schizophrenia in order to further highlight the genetic factors influencing the risk of schizophrenia. Methods: Taking the case-control study design, with the Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (DSM-5) to be the evaluation norm, 134 individuals suffering from schizophrenia hospitalized in the Third People's Hospital of Zhongshan City within January 2018 to April 2020 (case group) were selected, and 64 healthy individuals (control group) from the same geographical area had been chosen as well. MassArray identified DNMT1 gene single nucleotide polymorphisms (rs2114724 and rs2228611) and DNMT3B gene SNPs (rs2424932, rs1569686, rs6119954, and rs2424908). Using the generalized multifactor dimensionality reduction (GMDR), the RELN-BDNF-CREB-DNMT pathway's gene interactions were examined for their impact on schizophrenia. Results: GMDR analysis showed that the three-order interaction model RELN (rs2073559, rs2229864)-DNMT3B (rs2424908) was the optimal model (p = 0.001), with the consistency of cross-validation of 10/10 and the test accuracy of 0.8711. Conclusion: The interaction between the RELN (rs2073559, rs2229864)-DNMT3B (rs2424908) may be related to schizophrenia, and large sample sizes should be verified in different population.