DNA hypermethylation of serotonin transporter gene promoter in drug naïve patients with schizophrenia

Abstract

Introduction: Dysfunctional serotonin signaling has been linked to the pathogenesis of autism, obsessive compulsive disorder, mood disorders and schizophrenia. While the hypo-activity of serotonin signaling is involved in the pathogenesis of depression, anxiety and obsessive compulsive disorder; LSD, an agonist of serotonin type 2 receptor (5-HTR2A) induces psychosis. Therefore, anxiety and depressive disorders are treated by SSRIs which inhibit serotonin transporter (5-HTT) while psychotic disorders are controlled by drugs that block serotonin and/or dopamine receptors. Since genetic polymorphisms and epigenetic dysregulation of 5-HTT are involved in the pathogenesis of mental diseases, we analyzed DNA methylation of 5-HTT promoter in post-mortem brains and saliva samples of patients with schizophrenia (SCZ) and bipolar disorder (BD) to evaluate its potential application as a diagnostic and/or therapeutic biomarker in SCZ and BD.

Methods: Whole genome DNA methylation profiling was performed for a total of 24 samples (including two saliva samples) using the Illumina 27K (for 12 samples) and 450K DNA methylation array platform (for another 12 samples), followed by bisulfite sequencing to identify candidate CpGs for further analysis. Quantitative methylation specific PCR (qMSP) was used to assess the degree of CpG methylation of 5-HTT promoter in 105 post-mortem brains (35 controls, 35 SCZ and 35 BD) and 100 saliva samples (30 controls, 30 SCZ, 20 BD and 20 first degree relatives of SCZ or BD). The U133 2.0 Plus Human Transcriptome array for a total of 30 post-mortem brain samples (each group 10) followed by quantitative real-time PCR was used to study 5-HTT expression in 105 post-mortem brain samples.

Results: The qMSP analysis for 5-HTT promoter region showed DNA hypermethylation in post-mortem brain samples of SCZ patients (~30%), particularly in drug free patients (~60%, p=0.04). Similarly, there was a trend for DNA hypermethylation in antipsychotic free BD patients (~50%, p=0.066). qMSP analysis of DNA extracted from the saliva samples also exhibited hypermethylation of 5-HTT promoter in patients with SCZ (~30%, p=0.039), which was more significant in drug naïve SCZ patients (>50%, p=0.0025). However, the difference was not significant between the controls and unaffected first degree relatives of patients with SCZ (p=0.37) and versus patients using antipsychotic drugs (p=0.2). The whole genome transcriptome analysis of post-mortem brain samples showed reduced expression of 5-HTT in SCZ compared to the control subjects (~50%, p=0.008), confirmed by quantitative real-time PCR analysis (~40%, p=0.035) which was more significant in drug free SCZ patients (~70%, p=0.022).

Conclusion: A correlation between reduction in 5-HTT expression and DNA hypermethylation of the 5-HTT promoter in drug naïve SCZ patients suggests that an epigenetically defined hypo-activity of 5-HTT may be linked to SCZ pathogenesis. Furthermore, this epigenetic mark in DNA extracted from saliva can be considered as one of the key determinants in a panel of diagnostic and/or therapeutic biomarkers for SCZ.

Keywords: Brain; DNA methylation; Saliva; Schizophrenia; Serotonin transporter.

Figure 1.. The 5-HTT promoter sequence and the location of primers.

The upper lines show original DNA sequence and the lower lines bisulfite modified DNA sequence assuming that all of cytosines (C) followed by guanine (G) are methylated. The candidate CpGs for methylation are in bold and gray. DNA sequences flanking the MSP and qMSP primer sites amplifying methylated (M) and unmethylated (U) products are underlined. The site corresponding to the Probe # CG26741280 of the Illumina DNA methylation array is underlined and bolded in original DNA sequence. There are two SP1 binding sites in 5-HTT promoter region which are marked with yellow color (GGGCGG and CCGCCC).

Figure 2.. Bisulfite sequencing of the 5-HTT promoter region and MSP analysis.

A. Bisulfite sequencing of 5-HTT promoter region in representative samples for identification overall methylation pattern and partially methylated sites (indicated by arrows) for subsequent MSP and qMSP analysis. The original DNA sequence (top line) and bisulfite modified DNA sequence (bottom line) are placed at the top of sequence traces. The red segment of the bottom line shows a representative differentially methylated site. B. MSP analysis of a representative control (1), SCZ (2) and BD (3) using post-mortem brain samples. Placental DNA (P) was used as a negative control for methylation. C. MSP analysis of saliva DNA of a representative control subject (1) and a SCZ patient (2) using F1R1 qMSP primers. The β-Actin promoter amplified with primers designed from a CpG free region was used for the normalization of unmethylated (U) and methylated (M) products during qMSP analysis. L indicates a 100bp DNA ladder. D. Serial dilution of standard samples to optimize conditions for qMSP analysis in the test trials.

Figure 3.. Promoter DNA methylation status of 5-HTT in SCZ versus control subjects.

A. There was higher promoter DNA methylation in SCZ patients compared to the controls and first degree relatives of SCZ patients (SCZ FM) at site A. Notably, SCZ patients and drug naïve SCZ patients (DN SCZ) exhibit higher degree of DNA methylation. Drug use in SCZ patients (DU SCZ) reduced 5-HTT promoter DNA hypermethylation. B. Similar to site A, the 5-HTT promoter DNA methylation in SCZ was higher than controls at site B, particularly in drug naïve patients.

Figure 4.. Promoter DNA methylation and expression and analysis for 5-HTT gene in post-mortem brains of patients with SCZ and BD versus the control subjects.

A: The Y axis indicates promoter DNA methylation of 5-HTT (site B) normalized with the PCR product of β-Actin promoter in SCZ and BD versus controls. B: The Y axis indicates relative expression of 5-HTT normalized with β-Actin expression. As shown, in SCZ the expression of 5-HTT is significantly less than controls as well as BD patients.