Schizophrenia (SCZ) is one of the most common and severe mental disorders. Modified electroconvulsive therapy (MECT) is the most effective therapy for all kinds of SCZ, and the underlying molecular mechanism remains unclear. This study is aim to detect the molecule mechanism by constructing the transcriptome dataset from SCZ patients treated with MECT and health controls (HCs).
Transcriptome sequencing was performed on blood samples of 8 SCZ (BECT: before MECT; AECT: after MECT) and 8 HCs, weighted gene co-expression network analysis (WGCNA) was used to cluster the different expression genes, enrichment and protein-protein interaction (PPI) enrichment analysis were used to detect the related pathways.
Three gene modules (black, blue and turquoise) were significantly associated with MECT, enrichment analysis found that the long-term potentiation pathway was associated with MECT. PPI enrichment p-value of black, blue, turquoise module are 0.00127, <1×10-16 and 1.09×10-13, respectively. At the same time, EP300 is a key node in the PPI for genes in black module, which got from the transcriptome sequencing data.
It is suggested that the long-term potentiation pathways were associated with biological mechanism of MECT.
Schizophrenia (SCZ) is a stressful, chronic, incorrigible psychological disorder [
The effectiveness of electroconvulsive therapy (ECT) in SCZ and mood disorders have been verified [
To uncover the mechanisms of the action of ECT, several studies focus on the change of gene expression. Kaneko et al. [
All the 8 SCZ were recruited in the Fourth People’s Hospital of Yibin, PR China. The Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition (DSM-IV), Patient Edition (SCID-P) [
MECT was applied to all the patients by using the Thymatron IV instrument (Somatics, Lake Bluff, IL, USA). The treatment course of MECT was 6 times, 3 times a week. The patients were evoked using bilateral electrical stimulation with an initial electrical dose that based on 2/3 of their age, and subsequent dosing was performed according to seizure morphology adequacy. EEG can be used to monitor and assess the patient’s seizures. The indicators for reference include the peak heart rate, EEG endpoint, average seizure energy index and etc., all the parameters are recommended by instruction of Thymatron. Before treatment, patients received etomidate (0.16–0.2 mg/kg) to reach anesthesia status, succinylcholine (1.0 mg/kg) for muscle-relaxing and atropine sulfate (0.01 mL/kg) to reduce airway secretion by intravenous injection. The entire treatment process is monitored by professional anesthesiologists to prevent serious side effects such as asphyxia and arrhythmia.
All the patients received two times PANSS (baseline and end of the MECT treatment) to envaulted the severity of symptoms and the efficiency of the MECT, the PANSS reduction rate was defined as (PANSS score at baseline—PANSS score at the end of ECT treatment)/(PANSS score at baseline—30) ×100%, the reduction rate was indicated as: >75% indicated complete remission, 50–75% significant improvement [
The peripheral blood of the SCZ patients were collected before and after the MECT in anticoagulation tubes and stored in -80°C immediately. The whole blood of the HCs were collected in the same way. All the whole blood samples were used to extract the RNA using the protocol for TRIzol Reagent. The NanoDrop 2000 spectrophotometer (NanoDrop Technologies, Wilmington, Delaware) and Agilent 2100 (Agilent Technologies, Santa Clara, CA, USA) were used respectively to determine the RNA concentration and integrity number (IN), besides, agarose gel electrophoresis proved the integrity of all the RNA samples, spectrophotometer shows the OD260/280 of all the samples is between 1.8 to 2.0. To construct the RNA-transcriptome library, 5 ug of each high-quality RNA sample was used. The libraries were used for further transcriptome sequencing. In brief, mRNA was isolated according to the poly(A)-oligo(dT) and fragmented by fragmentation buffer, secondly, cDNA was synthesized via random hexamers and Illumina’s library construction protocol was used to handle matched cDNA. Libraries were size-selected for cDNA target fragments of 200–300 bp on 2% Low Range Ultra Agarose followed by PCR amplified using Phusion DNA polymerase for 15 PCR cycles. After quantification by TBS-380, a pairedend RNA-seq sequencing library was sequenced with the Illumina HiSeq 4000 system (2×150 bp read length).
SeqPrep (
The R package software edgeR [
To determine whether the genes in each module were associated with pathophysiological mechanism of SCZ, Kyoto Encyclopedia of Genes and Genomes (KEGG) and Disease Ontology (DO) enrichment analysis of those genes were applied. A hypergeometric test implemented in WebGestalt [
We carried out the permutation test using STRING [
In order to know whether the expression level of EP300 in different human brain areas have discrepancy, by examining 5 brain regions (cerebellum, parietal cortex brain, hippocampus, prefrontal cortex and striatum) gene expression data of SCZ patients and HCs (all data from
Multiple analysis was used to test those modules in different group and each module needs to meet the following points: 1) statistical differences for MEs between BECT and HC; 2) statistical differences for MEs between BECT and AECT group; 3) no statistical difference for MEs between AECT and HC group. Chi-square test were used to examine the differences of age, gender, marital status resident areas; t-test was used to examine the differences of educated years of patients and HCs, and PANSS scores (BECT and AECT) also use t-test to check the statistic difference. SPSS version 23 (IBM Corp., Armonk, NY, USA) was used to all the data analysis.
The demographic differences and statistic results are all shown in the
To investigate the expression of genes affected by MECT in SCZ. Firstly, 3000 DEGs (p<0.05) were obtained after analyzing the transcriptome sequence of 8 SCZ patients, then those DEGs of the 8 SCZ patients and 8 HCs were used to construct the gene co-expression network through WGCNA and 13 modules were obtained.
According to the above-mentioned WGCNA analysis, multiple comparisons were used to compare MEs in each module in SCZ patients (BECT vs. AECT) and HCs, it found that three modules (Black, Blue, and Turquoise) whose MEs were significantly associated with disease and treatment status (
Results of KEGG pathway are following points: One important pathway in black module was found: long-term potentiation (
The findings indicated that the richness in protein connections in each module and the resulting network was significantly different from any random networks; there were 51 direct edges in the black module gene network compared with only 32 edges expected (p=0.00127) (
By examining the gene expression level of EP300 in the brain regions mentioned above, we identified three areas including hippocampus (p=0.024), prefrontal cortex (p=0.012), cerebellum (p=1.35×10-4). After the false discovery rate (FDR), it found that gene expression level of EP300 in cerebellum was statistic different (FDR=0.026) (
It was found that gene expression changes in patients treated with MECT. Nishiguchi et al. [
Long-term potentiation (LTP) is also an important pathway in black module, it participated in variety of SCZ-related biological pathways. Like long-term plasticity of synaptic transmission, such as in LTP and long-term depression (LTD), provides a cellular correlate of experience-driven learning [
DO enrichment analysis showed that genes in the black module were related to valproic acid (VPA). VPA was found to induce broader epigenetic changes through different mechanisms, like DNA demethylation and histones acetylation [
In summaries, VPA plays an important role in SCZ and symptoms relief in SCZ may partly rely on MECT mediated VPA pathway alteration.
It has known to all that MECT also could improve the cognitive function. it was indicated that eight genes of some loci were associated with cognitive deficits of SCZ, EP300 is one of it [
In conclusion, our work still has some shortcomings and limitations. The insufficient amount of the samples is a great shortage; the influence of all the antipsychotics should be considered into the assessment of the severity of the symptoms, the impact of mono-antipsychotic, MECT and antipsychotics combined MECT treatment to gene expressions levels should be considered in the future researches, the gene expression changes in different time points in MECT therapy, the gene expression profile of MECT-resistance patients and etc. What is more, the transcriptome data of other SCZ related brain areas should also be detected to compare the DEGs to find more possible pathways. Besides, the pathways that revealed by our transcriptome data play a vital role in multiple neural functions, include cognitive function, neurodevelopment, synaptic plasticity and etc.
From all the above results, it is informed that many previous studies and findings were based on the change of single gene and/or protein of SCZ. Our study identified the long-term potentiation pathway and gene EP300 may play an important role in MECT and provides a new and macro perspectives for the future studies which devoted to understand the possible mechanism of MECT in SCZ treatment.
The online-only Data Supplement is available with this article at
This study was supported in part by a grant from National Nature Science Foundation of China (82001414, BX), Ministry of education Chunhui plan [2020 (703)], Sichuan provincial health and Family Planning Commission (16PJ562), Key projects of the Sichuan Provincial Education Department (18ZA0534), Luzhou Science and Technology Bureau [2017-S-40(4/18), 2016-S-67(7/23)], Youth Project of Affiliated Hospital of Southwest Medical University [2017-PT-9, 2011(37)], Southwest Medical University-Luzhou Government [2015LZCYD-S06(5/11), 2017LZXNYD-Z02] and Open Program of Nuclear Medicine and Molecular Imaging Key Laboratory of Sichuan Province.
The authors have no potential conflicts of interest to disclose.
Conceptualization: Wanhong Peng, Qingyu Tan, Ping Wang, Bo Xiang, Kezhi Liu, Xuemei Liang. Data curation: Wanhong Peng, Qingyu Tan, Minglan Yu, Ping Wang, Tingting Wang, Jixiang Yuan. Formal analysis: Wanhong Peng, Qingyu Tan, Ping Wang, Bo Xiang. Funding acquisition: Wanhong Peng, Dongmei Liu, Dechao Chen, Chaohua Huang, Youguo Tan, Kezhi Liu, Bo Xiang, Xuemei Liang. Investigation: Wanhong Peng, Qingyu Tan, Minglan Yu, Ping Wang, Tingting Wang, Jixiang Yuan. Methodology: Wanhong Peng, Qingyu Tan, Ping Wang, Bo Xiang. Project administration: Wanhong Peng, Dongmei Liu, Dechao Chen, Chaohua Huang, Youguo Tan, Kezhi Liu, Bo Xiang, Xuemei Liang. Resources: Wanhong Peng, Dongmei Liu, Dechao Chen, Kezhi Liu, Bo Xiang, Xuemei Liang. Validation Software: Wanhong Peng, Ping Wang, Bo Xiang. Supervision: Dongmei Liu, Dechao Chen, Chaohua Huang, Youguo Tan, Kezhi Liu, Bo Xiang, Xuemei Liang. Validation: all authors. Visualization: all authors. Writing—original draft: Wanhong Peng. Writing—review & editing: all authors.
Gene expression level of EP300 in different brain areas. A: The gene expression levels of EP300 in cerebellum areas between patients with SCZ and controls. B: The gene expression levels of EP300 were compared in hippocampus and prefrontal cortex between patients with SCZ and controls. The X-axis indicates the gene expression level of EP300 in different brain regions, and the Y-axis indicates the amount of gene expression.
Demographic and clinical data
Variables | SCZ (N=8) | HC (N=8) | χ2/t | p |
---|---|---|---|---|
Age, mean | 31.63±6.28 | 31.25±11.22 | t=0.30 | 0.75 |
Gender, women (%) | 4 (50) | 4 (50) | χ2=0 | 1 |
Educated years, N (%) | ||||
Primary | 0 (0) | 2 (25) | χ2=3.54 | 0.17 |
Middle | 5 (62.5) | 3 (37.5) | χ2=2.4 | 0.30 |
High & above | 3 (37.5) | 3 (37.5) | χ2=0 | 1.00 |
Baseline PANSS score | 83.63±7.11 | t=1.36 | 0.19 | |
AECT PANSS score | 39.88±6.92 | t=6.07 | <0.01 | |
Medicine, N (%) | ||||
Risperidone (3–6 mg/d) | 4 (50) | - | ||
Sulpiride (0.3–0.6 g/d) | 2 (25) | - | ||
Aripiprazole (10–15 mg/d) | 2 (25) | - | ||
Married (%) | 4 (50) | 4 (50) | ||
Urban & rural (%) | 6 (75) | 6 (75) |
Statistic demographic data results, there is three levels in educated years, primary refer to the 1–6 educated years, middle refer to 7–12 years and high & above refer to 13 or longer educated years. SCZ: schizophrenia, HCs: health controls, BECT: before MECT, AECT: after MECT, PANSS: Positive and Negative Syndrome Scale
Multiple comparisons of 13 module eigengenes
Module | BECT & AECT | BECT & HCs | AECT & HCs |
---|---|---|---|
Tan | 0.001 | 0.277 | 0.010 |
Black | 0.003 | <0.001 | 0.352 |
Pink | 0.014 | <0.001 | 0.045 |
Blue | 0.006 | 0.009 | 0.827 |
Turquoise | 0.026 | 0.004 | 0.405 |
Magenta | 0.087 | 0.159 | 0.742 |
Salmon | 0.317 | 0.311 | 0.991 |
Green | 0.539 | 0.547 | 0.991 |
Yellow | 0.537 | 0.220 | 0.532 |
Greenyellow | 0.758 | 0.917 | 0.680 |
Purple | 0.778 | 0.432 | 0.611 |
Brown | 0.754 | 0.714 | 0.498 |
Red | 0.959 | 0.867 | 0.908 |
HCs: health controls, AECT: patients before modified electroconvulsive therapy, BECT: patients after modified electroconvulsive therapy
Results of enrichment analysis for genes in the black module
Category | Pathway | ID | P | Padjust | Gene module |
---|---|---|---|---|---|
KEGG | Long-term potentiation | 4720 | 5.49e-05 | 0.0016 | Black |
Phosphatidylinositol signaling system | 4070 | 0.0004 | 0.0058 | Black | |
Regulation of actin cytoskeleton | 4810 | 0.0004 | 0.0058 | Black | |
Drug | Valproic acid | PA451846 | 0.0074 | 0.0222 | Black |