Evidence for Altered Metabolism of Sphingosine-1-Phosphate in the Corpus Callosum of Patients With Schizophrenia

Kayoko Esaki; Shabeesh Balan; Yoshimi Iwayama; Chie Shimamoto-Mitsuyama; Yoshio Hirabayashi; Brian Dean; Takeo Yoshikawa


Schizophr Bull. 2020;46(5):1172-1181. 

In This Article

Materials and Methods

Postmortem Brain Samples

Samples of the corpus callosum and BA8 were obtained from the Victorian Brain Bank Network at the Florey Institute for Neuroscience and Mental Health.[28–30] All procedures were carried out with written informed consent from the next of kin, and the study was approved by the Ethics Committee of RIKEN. After reviewing the clinical records, psychiatric diagnoses were confirmed based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria and the Diagnostic Instrument for Brain Studies, which allow for a consensus psychiatric diagnosis to be made after death.[31,32] The case histories for all nonpsychiatric subjects (controls) were extensively reviewed, and neuropsychopharmacological profiles were obtained. Treating clinicians and family members were also interviewed to exclude any history of psychiatric illness in the controls.

We first quantified sphingolipids in the postmortem brain samples from patients with schizophrenia and the sphingolipids with altered levels were subsequently evaluated in the patients with major depression and bipolar disorder, in comparison to the controls (Supplementary Figure S1). For sphingolipid analysis, we used frozen tissue samples from patients with (1) schizophrenia (n = 15, Supplementary Table S1), (2) major depressive disorder (n = 15, Supplementary Table S2), (3) bipolar disorder (n = 15, Supplementary Table S3), and (4) non-affected controls (n = 15, Supplementary Table S4) (sample set 1, Table 1). Within each control-schizophrenia-major depressive disorder-bipolar disorder tetrad (total of 15 tetrads), the age at death, sex, and postmortem interval were strictly matched; we also ensured there were no significant differences in those measures between the test groups and the controls (Table 1).[33]

For gene expression analysis, we used an expanded sample set from the corpus callosum and BA8 of patients with schizophrenia (n = 91) and controls (n = 90). The expanded sample set also included sample set 1 (sample set 2, Supplementary Table S5).[33] The overall flow of the experiment is shown in Supplementary Figure S1.


All animal experiments were performed in compliance with relevant laws and the guidelines approved by the Animal Ethics Committee at RIKEN. Male C57BL/6J mice (6 weeks old) were intraperitoneally administered phosphate-buffered saline (vehicle), haloperidol (0.1 mg/kg body weight), or risperidone (0.2 mg/kg body weight) for 4 weeks. After the mice were euthanized by cervical dislocation, the brains (vehicle, n = 10; haloperidol, n = 10; risperidone, n = 10) were excised as quickly as possible and stored at −80°C until the frontal cortex and corpus callosum were dissected for the downstream analyses. Methodology details are provided as Supplementary Information.

Measurement of Sphingolipids

Sphingolipids were extracted from the brain tissues of both humans and mice (Bligh and Dyer method)[34] and quantified using liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS).[18] Methodology details are provided as Supplementary Information.

Gene Expression Analysis

Expression of the target genes was measured by quantitative real-time reverse-transcription polymerase chain reaction (RT-PCR) or digital PCR using TaqMan Gene Expression Assays as described elsewhere.[35,36] Expression of the selected target genes in the corpus callosum were quantified by normalizing with the geometric mean of Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH) and Beta-2-Microglobulin (B2M) expression in the final evaluation. The overall flow of gene expression analysis is shown in Supplementary Figure S2.

Absolute quantification of mRNAs was performed in the corpus callosum and BA8/frontal cortex of samples from human controls (n = 6) and C57BL/6J mice (n = 3), using digital PCR. Methodology details are provided as Supplementary Information.


GraphPad Prism version 7 (GraphPad Software) was used for data analysis. Data are presented as the mean ± standard error of the mean (SEM) or as the mean ± standard deviation (SD). Outliers (more or less than the mean ± 2SD) were excluded. Significant changes between 2 groups were tested using the Mann-Whitney U test (2-tailed). Differences among more than 2 groups were analyzed using the nonparametric Kruskal–Wallis H test, followed by Dunnett's test for all data, including mouse lipid data, which did not show normality with the D'Agostino-Pearson test. Correlation was examined using Spearman's rank correlation coefficient. P < .05 was considered statistically significant and .05 ≤ P < .1 was considered a trend toward significance.