Introduction

Schizophrenia is a chronic and debilitating mental disorder affecting more than 20 million individuals worldwide, with significant socioeconomic consequences [1, 2]. One of the main features of schizophrenia is the dissociation of thoughts, concepts, identities, and emotions [3,4,5,6], associated with dysregulated neural pathways in the brain that lead to positive and negative symptoms [4, 6, 7]. The etiology of schizophrenia is multifactorial, involving genetic predisposition, environmental factors, and epigenetic modifications [6]. It is estimated that schizophrenia has a heritability range of 79–81% [8]. Genetic risk loci have been identified in non-coding areas such as promoters and introns, indicating that gene regulation is crucial for the onset of psychotic disorders [9]. In the case of schizophrenia, research has identified a number of genetic risk factors that are associated with altered gene expression patterns [10,11,12], which may be influenced by epigenetic modifications [9]. Twin studies have also provided evidence that epigenetic factors may be involved in the development of schizophrenia [13].

Epigenetic modifications, mainly DNA methylation or histone modifications, are involved in gene regulation. Among these, DNA methylation is the most extensively investigated [14]. DNA methylation occurs through the covalent binding of methyl groups to cytosine-guanine dinucleotides (CpG) [15]. Several genes, including GAD1 [16], COMT [17], RELN [18], BDNF [19], CHRNA7 [20], SOX10 [21], EGR1 [22], have been identified as potential candidates for DNA methylation in schizophrenic patients. DNA methylation is established and maintained by methyltransferases, such as DNMT1, DNMT3A, and DNMT3B [23]. DNMTs were first recognized as stable repressive marks, which are important mechanisms for gene silencing during progression and the cycle of life [24]. It has been suggested that patients with schizophrenia may have DNMT gene dysfunction.

Studies have shown that DNMTs are essential for the generation of mature, functioning germ cells, and appropriate embryonic development [25, 26]. Abnormal DNA methylation patterns have been linked to an increased risk of certain human diseases in individuals with a hereditary predisposition [27]. Genetic variations, such as single nucleotide polymorphisms (SNPs), have been strongly associated with schizophrenia in several family studies [28, 29]. Recent case–control studies indicated a significant association between minor alleles of DNMT1 (rs2114724 and rs2228611), DNMT3B (rs2424932 and rs1569686) and susceptibility of schizophrenia, while it is rare [30].

Previous studies have focused on the three active DNMT enzymes: DNMT1, DNMT3A, and DNMT3B. Some investigations have linked schizophrenia to polymorphisms within these enzymes, but there is still considerable ambiguity regarding the relationship between these polymorphisms and the clinical features of schizophrenia. DNMT SNPs prevalence in southern Chinese populations has not been reported yet. Therefore, the aim of the present study was to explore the effects of DNMT gene polymorphisms on schizophrenia susceptibility and symptoms in this population.

Materials and methods

Patient and control samples

Patients or their legal guardians provided informed consent after receiving authorization from the ethics committee of Zhongshan Third People’s Hospital for this study. The investigation focused on patients with schizophrenia who were treated at the hospital between 2018 and 2020. Participants were included if they met the following criteria: a diagnosis of schizophrenia as per the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, made by two attending physicians with consistent training, age between 16 and 60 years, and Han ethnicity. The study excluded patients with diabetes, hypertension, cancer, and other nervous system or mental diseases. However, other mental disorders were not excluded.

The control group at Zhongshan City’s Third People’s Hospital consisted of physical exam mentally healthy patients, employees, nurses, and volunteers. To be included in the control group, subjects had to meet the following criteria: no history of mental illness, no family history of mental illness, pass a physical exam, Han ethnicity, and between 18 and 60 years of age after psychological screening. The control group excluded adopted or single-parent children with unknown family histories, as well as individuals with serious somatic disorders such as diabetes, hypertension, or cancer.

Methods

The experimental methods used in this research included the collection of general data and clinical information, evaluation of psychotic symptoms, genomic DNA extraction, primer creation, PCR amplification, alkaline phosphatase treatment, single-base extension reaction, resin purification, chip sampling, mass spectrometric detection, and data analysis, which were based on past studies [31, 32].

The collected data were sorted and evaluated using the statistical program SPSS 20.0 (IBM, Armonk, NY, USA). The analysis of variance was performed, and results were presented as mean and standard deviation since the data exhibited normal distribution. Nonparametric statistical approaches were applied to non-normal data, and the results were expressed numerically as the median value (upper quartile, lower quartile). The genotype frequency was analyzed using the Chi-square test. regarding as categorical data. Hardy–Weinberg equilibrium was examined using SPSS 20.0 Chi-square test, and genotype and allele frequencies were compared between the schizophrenia and control groups. The linkage disequilibrium (LD) and pairwise LD coefficients in DNMT1 and DNMT3B were implemented with Haploview 4.2. ORs and confidence intervals (CIs) were computed for 95% of the data, and G*Power 3.1 was used for power analysis.

Results

Test of Hardy–Weinberg equilibrium for DNMT1 and DNMT3B

We performed a Hardy–Weinberg equilibrium analysis on the genotype data of 134 individuals with schizophrenia, focusing on six DNMT gene polymorphisms (rs2114724, rs2228611, rs2424932, rs1569686, rs6119954, and rs2424908). The results, presented in Tables 1 and 2, indicate that the genotype frequencies at these loci did not deviate significantly from the predicted population frequencies (p > 0.05). No outliers were identified as the cause of these results.

Table 1 Hardy–Weinberg equilibrium test for DNMT1 genotypes
Full size table
Table 2 Hardy–Weinberg equilibrium test for DNMT3B genotypes
Full size table

Results of genotype and allelic analysis for DNMT1 and DNMT3B

Tables 3 and 4 demonstrate that there were no significant differences in genotype frequencies between the two groups at the DNMT1 gene loci rs2114724 and rs2228611 (p > 0.05). Likewise, no significant differences were observed in the genotype frequencies between the two groups at the DNMT3B gene loci rs2424932, rs1569686, rs6119954, and rs2424908 (p > 0.05).

Table 3 Comparison of DNMT1 rs2114724 and rs2228611 genotype frequencies between the two groups
Full size table
Table 4 Comparison of DNMT3B rs1569686, rs2424908, rs2424932 and rs6119954 genotype frequencies between the two groups
Full size table

Table 5 displays no significant differences in the frequency of the rs2114724, rs2228611, rs1569686, rs2424908, rs2424932, and rs6119954 alleles between the two groups (p > 0.05).

Table 5 Comparison of rs2114724, rs2228611, rs1569686, rs2424908, rs2424932 and rs6119954 allele frequencies between the two groups
Full size table

Linkage disequilibrium analysis for DNMT1 and DNMT3B

Genotyping for two SNPs in DNMT1 and four SNPs in DNMT3B was successful, and Fig. 1 illustrates their linkage disequilibrium (LD). The two SNPs in DNMT1 formed a block within 25 kb (chr19: 10154572-10156401), contrasting the four SNPs in DNMT3B. Due to DNMT1's strong linkage, tagged SNPs were run in Haploview.

Fig. 1
figure 1

Linkage disequilibrium (LD) pattern of DNMT1 (a) and DNMT3B (b) SNPs from Haploview analysis. The D′ values for LD are represented by the numbers inside the boxes; the miss numbers were 100. The strength of the LD between pairwise pairings of SNPs is shown by the areas that are shaded in the LD map (white represent low LD; red represent high LD)

Full size image

Haplotypes of the SNPs in DNMT1 and DNMT3B

We examined the haplotype distribution of DNMT1 (rs2114724 and rs2228611) and DNMT3B (rs1569686, rs2424908, rs2424932, and rs6119954) in both the schizophrenia and control groups. Haplotypes with a frequency of at least 3% were selected for analysis to investigate their association with schizophrenia. Two distinct DNMT1 haplotypes were identified, with the CT haplotype being the most common in both controls (68.0%) and patients (63.8%). The most common DNMT3B haplotype in both groups was GTGT (59.3% and 60.9%, respectively). The haplotype frequencies at gene loci rs2114724–rs2228611, rs2424932–rs1569686–rs6119954–rs2424908, and overall did not differ significantly between patients and controls (p > 0.05, Table 6).

Table 6 Analysis of haplotype distribution between the schizophrenia and control groups
Full size table

Analysis of the association between genotypes and clinical symptoms of schizophrenia

Table 7 displays the associations between DNMT1 (rs2114724 and rs2228611) and DNMT3B gene (rs1569686, rs2424908, rs2424932, and rs6119954) and clinical mental symptoms of schizophrenia. Statistically significant differences in PANSS overall positive symptom scores, P3 (hallucinatory behavior), P6 (suspicious/persecution), G7 (motor retardation), and G15 (preoccupation) were observed when comparing patients with different DNMT1 gene rs2114724 and rs2228611 genotypes. Moreover, significant differences were found in PANSS general symptom scores G1 (somatic concern), G5 (mannerisms and posturing), and G8 (uncooperativeness) among patients with different DNMT3B gene rs6119954 genotypes (p < 0.05). Additionally, significant differences were noted in the PANSS general symptom score G10 (disorientation) among individuals with different DNMT3B gene rs2424908 genotypes (p < 0.05).

Table 7 Relationships between different DNMT1 and DNMT3B genotypes and PANSS scores in schizophrenia group
Full size table

Evaluation of statistical power

G*Power calculated power. This sample had 92.372% power to identify a meaningful relationship (α < 0.05) with an effect size index of 0.5.

Discussion

Accumulating evidence suggests that the development of schizophrenia is influenced by the interaction of heredity and environmental factors [33]. Epigenetic mechanisms, such as DNA methylation with DNA methyltransferases (DNMTs), are crucial for the effect of environmental factors on genetic susceptible. In the dynamic procedure methylation, the DNMT1 is responsible majorly for maintaining methylation during cell division [34, 35].

This study investigated the genotype and allele distribution of DNMT1 (rs2114724 and rs2228611) and DNMT3B (rs1569686, rs2424908, rs2424932, and rs6119954) in 134 patients with schizophrenia and 64 healthy controls. The results showed no statistical difference between the case and control groups, which contradicted to the few previous study in India population [36]. The study found that the TT genotype and T allele in DNMT1 rs2114724 and the AA genotype and A allele in DNMT1 rs2228611 were associated with schizophrenia. However, there was no significant difference in genotype and allele distribution in DNMT3B gene (rs2424932, rs1569686, rs6119954, and rs2424908) between the case group and the control group. In line with study from population with shared ethical background [37], indicated that the genotype frequency and allelic frequency of DNMT1 rs2114724 and rs2228611 were also correlated with schizophrenia. Few studies employing Chinese Han population reported similar findings with the current study. For instance, Zhai et al. [38] investigated the genotype and allele distribution of DNMT1 rs2114724 and rs2228611 in a Chinese population and found no significant difference between the schizophrenic patients and healthy controls. Furthermore, Zhang Chen et al. [39] revealed that there was no statistical difference in genotype and allele distribution of DNMT3B rs2424908 between the schizophrenic case and control groups. Although the significant difference in genotype and allele distribution of DNMT3B rs6119954 was identified in the schizophrenic patients, which indicated dysfunction of DNMR3B. There are couple of reasons contribute to the discrepancies. Firstly, racial genetic background will affect on the difference in allelic frequency, according to the previous findings [38, 39]. Secondly, technology employed for identifying the genetic variation attribute partly to difference of observational results. Previous study indicated that low effective values for identifying genetic variation in promoter upstream regions with Methylated DNA Immunoprecipitation (MeDIP) comparing to microarray [40]. Finally, sample size in the literatures may contribute to variation in the findings [41].

This study used linkage disequilibrium analysis to identify an absolute linkage block between rs2114724 and rs2228611 in DNMT1 and the four loci in DNMT3B. These loci were significantly correlated with positive symptoms of schizophrenia, including hallucinatory behavior, suspicion/persecution, motor retardation, and preoccupation. This is the first report linking rs2114724 and rs2228611 in DNMT1 with positive symptoms of schizophrenia. DNMT1 gene may affect the clinical symptoms of schizophrenia by regulating the expression of genes involved in the dopaminergic and GABAergic systems [41]. Cumulating evidence has suggested that DNMT1 as well as DNMT3 established and maintained dynamically the DNA methylation in dopaminergic, GABAergic, glutamatergic, serotonergic pathways of neurotransmission, which response for regulating the clinical phenotypes [42,43,44,45,46].

Although this study provides new insights into the role of DNMTs in schizophrenia, limitations exist, and further research is necessary to validate the findings in larger samples and different populations. Future studies should also explore the specific regulatory mechanisms of DNA methylation in the development of schizophrenia.