Male-Specific Association between Dopamine Receptor D4 Gene Methylation and Schizophrenia

Objective The goal of our study was to investigate whether DRD4 gene DNA methylation played an important role in the susceptibility of Han Chinese SCZ. Methods Using the bisulphite pyrosequencing technology, DNA methylation levels of 6 CpG dinucleotides in DRD4 CpG island were measured among 30 paranoid SCZ patients, 30 undifferentiated SCZ patients, and 30 age- and gender-matched healthy controls. Results Strong correlation was observed among the six CpG sites (r>0.5, P<0.01), thus average methylation levels were applied thereafter. Our results indicated that there was a significant association between DRD4 methylation and the risk of SCZ (P = 0.003), although there was no significant difference in DRD4 methylation between the two SCZ subtypes (P = 0.670). A breakdown analysis by gender showed that the significant association of DRD4 methylation and SCZ was driven by males (P<0.001) but not by females (P = 0.835). DRD4 methylation was significantly associated with p300 in male SCZ patients (r = −0.543, P = 0.005) but not in female SCZ patients (r = 0.110, P = 0.599). Moreover, receiver operating characteristic (ROC) curves showed DRD4 methylation was able to predict the status of SCZ in males [area under curve (AUC) = 0.832, P = 0.002] but not in females (AUC = 0.483, P = 0.876). Finally, a further expression experiment showed that DRD4 methylation in the gene body was positively associated with gene expression, although the exact mechanism of gene regulation remained unknown for this interesting DRD4 methylation. Conclusion The gender disparity in the DRD4 DNA methylation provides novel insights into the pathogenesis of SCZ.


Introduction
Schizophrenia (SCZ) is a complex mental disorder with a global lifetime prevalence of about 1% [1]. Paranoid SCZ and undifferentiated SCZ are two most common subtypes of SCZ according to DSM-IV criteria [2]. The hallmark symptoms of these subtypes comprise delusions, hallucinations, extremely disorganized behavior and negative symptoms [3]. SCZ is a complex disorder resulting from both genetic and environmental factors, including genetic vulnerability, neurotoxicity, unbalanced neurotransmitter, living environment, drug abuse and prenatal stressors [4,5,6]. Twin and family studies have revealed that SCZ is a heritable disorder [7,8], although heritability estimation varies due to the difficulty in separating the effects of genetics and environmental factors [9]. Dopamine (DA) is one of the most important neurotransmitters in human brain, and dysfunction of DA system is a fundamental event in SCZ development [10,11]. An inverted-U curve can describe the relationship between DA activity and cortical function, either up or down stimulation of DA can result in poor cerebral performance [12,10]. The dopamine receptor D4 (DRD4) is a subtype of dopamine receptor family that is activated by the DA [13]. The neuregulin and DA modulation of hippocampal function was dependent on DRD4 activation and genetically associated with SCZ [14]. It can regulate many neurological processes connecting with psychiatric disorders [15].
The variations of DRD4 were widely reported to be related to diverse human behavior phenotypes [16,17].
Epigenetic modification is one of the mechanisms underlying the interaction between environmental exposure and individual genetic background in the development of psychiatric disorders [18,19]. DNA methylation is a crucial way of epigenetic mechanisms that regulate expression of numerous functional genes in human nervous system [4,18,20]. However, there is a lack of epigenetic evidence for the involvement of DRD4 in SCZ pathogenesis.
The p300 wave, a component of an event-related potential, is often used as metrics of cognitive function and has been proved relative to SCZ cognitive impairment [10]. The p300 waveforms of monozygotic twins are almost the same and more unanimous than that in dizygotic twins [10,21]. This implied that p300 amplitude might be under genetic control and may serve as an endophenotype for SCZ [10]. In the present study, we explore the association between DRD4 methylation and SCZ clinical features such as cognitive symptomatology and medication. The goal of our study is to assess the association between DRD4 methylation and SCZ.

Samples and clinical data
The samples in the present study comprise 30 paranoid SCZ patients ( Patients with other psychiatric co-morbidities and drug abuse were also excluded from current study. All SCZ subjects were previously pharmacologically treated by one type of antipsychotic drug. SCZ subjects were prescribed antipsychotic drugs for at least 8 weeks. Healthy controls were age-and gender-matched healthy persons without any history of psychiatric diseases or disorders, and serious diseases such as cancers and cardiac diseases. In addition, we performed a longitudinal research to study the effect of risperidone therapy on the DRD4 DNA methylation level in peripheral blood. Blood of five additional newly onset male patients were drawn before and after risperidone (3-6 mg/d) treatment as the main antipsychotic medication for 8 weeks. Chemistry Analyzer (AU2700, Olympus, Japan) was used for the measurement of cysteine at the Clinical Chemistry Laboratory of Ningbo Kangning Hospital. The study protocol was approved by the ethical committee of Ningbo University. The informed written consent was obtained from all subjects. The informed written consent of the participants, whose capacity to consent was compromised, were obtained from their guardians and the scientific work complies with the current laws of China.

Symptomatology assessments
A series of tests were used for symptomatology assessments. These tests comprised the Positive and Negative Syndrome Scale (PANSS), Clinical Global Impressions (CGI), Wisconsin Card Sorting Test, and Chinese version of Wechsler Memory Scale (WMS) test. As an endophenotype for evaluating the cognitive function, p300 wave was also recorded in this study and analyzed by two experienced psychiatrists (JC and SG) by using the neural electricity device (NDI-200P+, Shanghai, China).

DNA methylation assay
Human genomic DNA was prepared from peripheral blood samples using the nucleic acid extraction automatic analyzer (Lab-Aid 820, Xiamen City, China). DNA was quantified using the PicoGreenH double strand (dsDNA) DNA Quantification Kit (Molecular Probes, Inc. Eugene, USA) and extracted from the fresh blood after being drawn from the involved individuals into a vacuum tube with 3.8% Sodium Citrate agent. Then the left blood and extracted DNA were stored at 280uC. As described previously [22,23], bisulphite pyrosequencing technology was used to determine methylation levels of 6 CpGs on the fragment within   DRD4 CpG island (Figure 1). PCR and pyrosequencing primers for DRD4 gene amplification were described in Table S1. More details on the pyrosequencing assay are available in the study by Guay et al [24].

Gene expression assay
Total RNA was extracted from peripheral blood. RNA was extracted with TRIZOL reagent (Invitrogen Life Technologies Co, USA) according to the manufacturer's protocol. RNA isolation and reverse transcription were described in our previous study [25]. Primers for qRT-PCR were designed according to the previous study [26]. The qRT-PCR reactions were conducted according to the previous workflow [25] under the thermal conditions reported before [26]. GAPDH was used as an endogenous control.

Statistical analysis
Categorical variables were compared using a Pearson chisquare test, whereas Student t-test was used to compare the mean group differences for continuous variables. A two-sided P-value ,0.05 was considered statistically significant. All statistical analyses were performed with PASW Statistics 18.0 software (SPSS, Inc., Somers, NY, USA).

Results
A total of 60 SCZ patients and 30 age-and gender-matched healthy controls are recruited in the current association study. The biological information of all SCZ subjects was shown in Table 1.
In this study, we have selected a locus containing 6 CpG dinucleotides to explore the association of DRD4 methylation with SCZ ( Figure 1). We found DNA methylation levels were well correlated among all the CpGs (Figure 1, r.0.5, P,0.01), therefore average methylation was used to represent DRD4 methylation in the following analyses. In addition, we didn't observe any non-CpG methylation in the sequenced fragment. No significant difference of methylation level was found between male (71.87613.33%) and female (67.34615.82%) SCZ patients (P = 0.210). Symptomatology assessments of 60 SCZ subjects were shown in Table S2. As shown in Figure S1, no significant difference of DRD4 methylation was found between paranoid SCZ and undifferentiated SCZ in males (P = 0.986), females (P = 0.591) and combined groups (P = 0.670). Thus, we combined the SCZ subtypes in the following analyses. There was significant difference between SCZ patients (69.82616.76%) and healthy controls (62.92619.96%) in DRD4 methylation (P,0.001). As shown in Figure 2, a breakdown analysis by gender indicated there was significantly higher DRD4 methylation in male SCZ patients comparing with healthy male controls (male cases versus male controls: 71.87613.33% versus 48.83616.79%, P,0.001), however, no difference was observed between female SCZ patients (67.34615.82%) and female healthy controls (66.31612.30%, P = 0.835, Figure 2). Hypomethylation of DRD4 was observed in the male controls (48.83616.79%) compared with the female controls (66.31612.30%, P,0.05). There is no difference in DRD4 methylation levels between smokers and non-smokers in male SCZ patients (P = 0.529, Figure 3). DRD4 methylation was significantly inversely correlated with p300 in the male SCZ patients (Figure 4, r = 20.543, P = 0.005), but not in the female SCZ patients (Figure 4, r = 0.110, P = 0.599). Meanwhile, receiver operating characteristic (ROC) curve showed that DRD4 methylation was able to predict the status of SCZ in males [ Figure 5    An analysis by therapeutic antipsychotics showed that there were no significant differences in DRD4 methylation among SCZ patients with different drug treatment ( Figure S2, P = 0.080). DRD4 methylation of subgrouped SCZ patients by therapeutic antipsychotics was significantly higher than that of healthy male controls ( Figure S3, P,0.05), in contrast that no difference of DRD4 methylation was observed between female SCZ patients treated by clozapine and healthy female controls ( Figure S3, P.0.05). To note, there were significant differences between medicated SCZ males on quetiapine and control males ( Figure S3, P,0.001) and significant differences between medicated SCZ females on quetiapine and control females ( Figure S3, P = 0.003), although both have equally small numbers of SCZ subjects (4 males and 5 females). Future work is needed to investigate the effect of quetiapine on DRD4 methylation in SCZ patients. A longitudinal research demonstrated there was no significant different DRD4 methylation between pre-therapy and post-therapy of risperidone in five additional male SCZ patients ( Figure 6, P = 0.352).
In addition, we explored the result of combining the data of this five additional schizophrenic samples with other male controls, the result was accorded with Figure 2 ( Figure S4, P,0.001). Furthermore, we have done an association analysis between cysteine levels and DRD4 methylation in both male and female SCZ subjects. The result showed no correlation between cysteine levels and DRD4 methylation in both genders of SCZ subjects ( Figure S5, Female: r = 20.125, p = 0.509; Male: r = 0.293, p = 0.116). The p300 remained significant after excluding one male sample that was very different from others ( Figure S6, Male: r = 20.427, p = 0.037).
In order to investigate the role of DRD4 methylation in the gene regulation, we further performed expression experiment among 10 male controls. Our results showed a significant positive correlation between DRD4 methylation and DRD4 gene expression in the healthy males (r = 0.713, P = 0.021, Figure 7). Our expression data suggests that DRD4 methylation is functionally relevant to expression level.

Discussion
SCZ is a type of complex disorder that is contributed by both genetic and environmental factors [27,28]. As reversible and heritable modifications, epigenetic marks such as DNA methylation are important to reflect the interactions between genetic factors and environmental exposures, and thus may provide us novel understandings for the development of complex diseases or disorders [29,30].
DA transmitter plays an important role in the psychological function, and the DA receptors are involved in the neural signal pathway modulating human behavior [15,16]. The dopaminergic system is involved in many kinds of human behavioral performances, such as aggression, delinquency and other behavioural problems [16]. The dysregulation of DA system pathway was hypothesized to be one of major causes of SCZ [31]. Dopaminergic system was the biological targets of psycholytic drugs in SCZ treatment. Genetic variations of genes in this system can influence the response to antipsychotic treatment [10,17].
DRD4 gene encodes a subtype of DA receptor that regulates many neurological processes connecting with psychiatric disorders [15]. Genetic association of DRD4 gene has been reported in SCZ research [29,32,33]. However, the role of DRD4 methylation in SCZ development is not known yet. In the present study, we found that DRD4 was likely to exert its role in the susceptibility of male SCZ through its aberrant methylation. This finding may provide potential implication for future SCZ diagnosis and the development of personalized antipsychotic therapy.
Gender differences were observed in the onset and prevalence of SCZ [7]. Male patients have a 5-year younger age of onset and have 40% greater risk than the females [7]. Estrogenic protection may contribute to the prevention of SCZ onset in females [34], but the exact reason that causes this gender difference always puzzles SCZ researchers. The gender difference in the SCZ development suggested estrogen had a protective effect against SCZ [35]. Several dopaminergic system genes have been proved to contribute to the etiology of SCZ [36]. DRD4 may be a member of these susceptibility genes. A DRD4 variant was shown to be associated with gender-specific abnormal behaviors [16]. Higher exposure to psychosocial risk factors of male individuals may help explain this phenomenon [16]. In addition, we observed that DRD4 methylation was associated with p300 in males. However, no association was shown between DRD4 methylation and cytosine level. Our results revealed a male-specific DNA methylation of DRD4 gene and a positive correlation between DRD4 methylation and DRD4 expression in the healthy males. These findings may help to understand the gender disparity in SCZ development. It also indicates that the therapeutic schedule of SCZ needs to be prescribed in a gender-specific way.
There are several limitations in our study that need to be taken with caution. Firstly, DNA methylation levels change across different tissues [37]. Here we use the DRD4 methylation in peripheral blood as a surrogate of brain tissue. The aberrant blood DRD4 methylation in male SCZ patients may not imply an aberrant DRD4 methylation in their brain tissues. Secondly, due to the strict matching criteria for the methylation study, we only recruited 30 paranoid SCZ patients, 30 undifferentiated SCZ patients, and 30 age-and gender-matched healthy controls. Future validation with larger sample size is warranted in other ethnic populations. Thirdly, we didn't investigate how a hypermethylation of DRD4 gene body activated gene expression. Future work will be needed to take this into account.
In summary, we found a gender dimorphism of DRD4 methylation in the susceptibility of SCZ. This epigenetic modification may give hints to elaborate the pathological mechanisms of SCZ, and provide new biological aspects of DRD4 gene. The aberrant methylation of DRD4 gene may be a valuable gender-specific biomarker to monitor the risk and development of SCZ.