Gene expression profiling in peripheral blood mononuclear cells of early-onset schizophrenia

Abstract

Schizophrenia (SZ) is a severe chronic psychiatric disorder with wide prevalence and high morbidity. We know little about SZ's etiology and pathophysiology at present. The study of gene expression profile is useful for us to identify potential biomarkers at molecular level and explain possible pathogenesis of SZ. Therefore we recently compared gene expression profiles in PMBCs from EOS cases and healthy controls using microarrays. Here we will describe in detail the contents and quality control of the microarray experiment. The raw microarray data are accessible through GEO series accession number GSE54913.

Specifications
Organism/cell line/tissue	Homo sapiens/peripheral blood mononuclear cell
Sex	Male and female
Sequencer or array type	Arraystar LncRNA Array v2.0
Data format	Raw and processed
Experimental factors	Early-onset SZ cases vs. healthy controls(< 18 years)
Experimental features	Microarray gene expression profiling to identify differential expressed genes in SZ cases compared with controls
Consent	All the participants and their parents signed the informed consent
Sample source location	China

1. Direct link to deposited data

http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE54913.

The disease onset before age 18 is generally regarded as early-onset (EOS) when the patients confer more familial vulnerability and poor outcomes [1]. The neurodevelopmental hypothesis posits that the onset of SZ is associated with early development of the nervous system [2]. We paid attention to this period and speculate that the altered gene expression in these patients may be associated with the disease process.

Peripheral blood mononuclear cells (PBMCs) have represented an accessible tissue source for gene expression, as it is easily collected from patients. There already have many gene expression profiling studies using PBMCs, a consistent conclusion about the expression alteration of schizophrenia is lacked [3], [4].

2. Experimental design, materials and methods

We recently collected blood samples from 18 EOS cases and 12 controls. Then we generated whole-genome gene expression profiles on PBMCs from these samples by using microarray. 17,200 valid probes detected in our experiment were used to identify altered gene expressions.

2.1. Study population

A total of 18 first-onset SZ patients (8 males and 10 females, aged 14.78 ± 1.70 years) were included in our study. They were untreated and drug-naïve patients diagnosed by at least two experienced psychiatrists independently according to the Diagnosis and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) criteria for SZ. 12 healthy controls (6 males and 6 females, aged 14.75 ± 2.14 years) were recruited into the study. Teenagers with a history of other mental health or neurological diseases were enrolled into our study. All participants were unrelated Han Chinese recruited from the north of China. And both the participants and their parents signed the informed consent before participation. The study was approved by Medical Research Ethics Committee of Shanxi Medical University.

2.2. Microarray and quality control

Peripheral blood was collected. NanoDrop ND-1000 was used to quantify total RNA after RNA extraction, and RNA integrity was assessed by standard denaturing agarose gel electrophoresis. Agilent Array platform was employed to perform the microarray analysis. Following RNA amplification, hybridization and image scanning, signal intensities were normalized in the quantile method using GeneSpring GX v11.5.1 (Agilent Technologies), and low intensity mRNAs were filtered (mRNAs that at least 20 out of 30 samples have flags in Present or Marginal were chosen for further analysis). R [5] was used to perform the data processing and analyses of mRNA data. The sample preparation and microarray hybridization were performed based upon the manufacturer's standard protocols with minor modifications.

Log2-ration was used by quantile normalization. The distributions of the intensities after normalized among all samples were shown in Fig. 1, Identification of differentially expressed genes between SZ cases and controls was made using R package genefilter [6].We identified 84 differentially expressed genes through fold change and P value filtering (FC ≥ 2 and P_adjusted < 0.05) listed in Table 1.

Fig. 1.

Quality assessment of mRNA data after filtering. The box-plot shows the distribution of normalized signal intensity by array; the distributions of log2-ratios among all samples are nearly the same after normalization.

Table 1.

List of differentially expressed genes.

Up-regulated genes (82)	C11orf49 SLC18A1 NAT1 MYBPC1 KIF23 GTF2H1 ALDH4A1 IL28RA ERVFRDE1 ALDH3A1 ODF4 TSPAN16 CCNE1 TGFA SATB2 SLC45A1 IL1RL2 BBS5 CTLA4 EPPIN OSGIN1	NKAIN4 EYA2 OPRL1 C21orf56 SLC5A4 GJB5 CCDC134 MYL3 SCAP PRICKLE2 ENTPD3 RNF186 EIF4G1 UGT2B4 RASSF6 FGA PAICS SH3RF2 UBD ECM1 HOXD11	LCE2D UBAP2L RFPL4B CCL26 DARC POU6F2 PNMA2 CNGB3 DEFB135 FAM110B MAL2 SARDH NUP188 C9orf171 TMEM27 XAGE3 CUL4B PNCK SMIM9 ERAS	GAGE10 ATP2B3 TKTL1 USP9Y LDB1 ACP2 P4HA3 C11orf1 FAM19A2 C12orf68 HCFC2 RPL10L PRKAB2 CA12 C15orf2 ZP2 SALL1 C16orf46 SLC5A2 GPT2
Down-regulated genes (2)	IQCF6	POM121L12

(p < 0.05 with a fold change > 2).

3. Discussion

All the participants in our study were teenagers with similar age (< 18 years), and their brains were still developing. The SZ cases were neither under medication nor had a history of pharmacotherapy. We mainly described a dataset about gene expression profiles of the 30 samples measured by Arraystar.

Among the 84 DE genes listed above, SLC18A1 has been reported to be associated with SZ [7], [8]. In addition, CTLA4 was also identified showing a high expression level in SZ [9] which is consistent with the results from our study. Through our description above, we believe that this dataset will be useful for the exploration of SZ's pathogenesis in the future.

Conflict of interest

The authors have no conflicts of interest.