An eQTL study in the Japanese population [genotype]

A genome-wide eQTL analysis was performed in whole blood samples collected from 76 Japanese subjects. RNA microarray analysis was performed for 3 independent samples that were genotyped in a genome-wide scan. The correlations between the genotypes of 534,404 autosomal single nucleotide polymorphisms (SNPs) and the expression levels of 30,465 probes were examined for each sample. The SNP-probe pairs with combined correlation coefficients of all 3 samples corresponding to P < 3.10 × 10-12 (i.e., Bonferroni-corrected P < 0.05) were considered significant. SNP-probe pairs with a high likelihood of cross-hybridization and SNP-in-probe effects were excluded to exclude false positive results. We identified 102 cis-acting and 5 trans-acting eQTL regions. The cis-eQTL regions were widely distributed both upstream and downstream of the gene, as well as within the gene. RNA microarray data obtained from 3 independent samples originally recruited for other studies investigating the gene expression levels in psychiatric disorders were used in the present study. For the purpose of the present analyses, genomic DNA was collected from 24 subjects (13 men and 11 women, mean age [SD] = 39.9 [7.6] years) in sample 1, 24 subjects in sample 2 (12 men and 12 women, 34.1 [11.5] years), and 28 subjects (14 men and 14 women, 41.4 [11.8] years) in sample 3. Some of the subjects had depressive symptoms, but all were physically healthy and without clinically significant systemic disease (e.g., malignant disease, diabetes mellitus, hypertension, renal failure, or endocrine disorders). Subjects were recruited from the outpatient clinic of the National Center of Neurology and Psychiatry Hospital, Tokyo, Japan, through advertisements in free local information magazines or through our website announcement. All the subjects were biologically unrelated Japanese individuals who resided in the same geographical area (western Tokyo). The study protocol was approved by the ethics committee at the National Center of Neurology and Psychiatry, Japan. Written informed consent was obtained from every subject after the study was explained to them. Venous blood was collected between 1100 and 1200 h in PAXgene tubes (Qiagen, Valencia) from each subject and was incubated at room temperature for 24 h for RNA stabilization. RNA was extracted from whole blood according to the manufacturer’s guidelines by using the PAXgene Blood RNA System Kit (PreAnalytix GmbH, Hombrechtikon, Switzerland). The RNA was quantified by optical density readings at A260nm by using the NanoDrop ND-1000 (Thermo Scientific, Rockford). Gene expression analysis was performed using Agilent Human Genome 4 × 44 K arrays (Agilent Technologies, Santa Clara). Raw signal data for each of the 3 independent samples were analyzed separately by the GeneSpring GX software (Agilent Technologies). Data were filtered according to the expression level for quality control to eliminate genes that were below the 20th percentile threshold. The expression value of each gene was normalized to the median expression value of all genes in each chip. A total of 30,465 probes were included in the analysis. Genomic DNA was obtained from venous blood samples. Genotyping was performed by Riken Genesis (Yokohama, Japan) using the Illumina HumanOmni1-Quad BeadChip (Illumina, Inc., San Diego). A total of 713,495 autosomal SNPs were assessed for quality using the PLINK v1.07 software. All SNPs with a call rate below 95%, a deviation from Hardy-Weinberg equilibrium at an error level of P < 0.001, or a minor allele frequency of less than 10% were excluded. The remaining 534,404 SNPs were used for further analysis.

本研究针对76名日本受试者的全血样本开展全基因组表达数量性状基因座（expression quantitative trait locus, eQTL）分析。针对3组经全基因组扫描分型的独立样本，开展RNA微阵列分析。针对每一组样本，分别检测534,404个常染色体单核苷酸多态性（single nucleotide polymorphism, SNP）的基因型与30,465个探针的表达水平之间的相关性。将3组样本合并相关系数对应的P值小于3.10×10^-12（即邦费罗尼校正后P<0.05）的SNP-探针对视为显著关联。排除存在交叉杂交高风险及探针内SNP效应的SNP-探针对，以剔除假阳性结果。最终共鉴定得到102个顺式作用eQTL区域与5个反式作用eQTL区域。顺式eQTL区域广泛分布于基因上下游区域及基因内部。
本研究使用的3组独立样本的RNA微阵列数据，最初取自为探究精神疾病基因表达水平开展的其他研究。本次分析的受试者信息如下：样本1共24名受试者（男13名、女11名，平均年龄[标准差] = 39.9 [7.6]岁），样本2共24名受试者（男12名、女12名，平均年龄[标准差] = 34.1 [11.5]岁），样本3共28名受试者（男14名、女14名，平均年龄[标准差] = 41.4 [11.8]岁）。部分受试者存在抑郁症状，但所有受试者均身体健康，无临床意义明确的系统性疾病（如恶性肿瘤、糖尿病、高血压、肾衰竭或内分泌疾病）。受试者通过免费本地信息杂志广告或本团队官网公告招募自日本东京国立神经内科与精神病医院门诊。所有受试者均为无生物学亲缘关系的日本个体，且居住于同一地理区域（东京西部）。
本研究方案经日本国立神经内科与精神病医院伦理委员会批准。所有受试者在充分了解研究内容后签署书面知情同意书。
每位受试者于上午11:00至12:00采集静脉血，置于PAXgene采血管（Qiagen, 巴伦西亚）中，室温孵育24小时以稳定RNA。采用PAXgene全血RNA系统试剂盒（PreAnalytix GmbH, 宏雷克顿, 瑞士），按照制造商说明书从全血中提取RNA。使用NanoDrop ND-1000分光光度计（Thermo Scientific, 罗克福德）通过A260nm光密度读数对RNA进行定量。基因表达分析采用安捷伦人类基因组4×44K芯片（Agilent Technologies, 圣克拉拉）。使用GeneSpring GX软件（Agilent Technologies）分别对3组独立样本的原始信号数据进行分析。为进行质量控制，根据表达水平进行过滤，剔除表达量处于第20百分位数以下的基因。将每个基因的表达值标准化至每个芯片中所有基因的表达中位数。本次分析共纳入30,465个探针。
基因组DNA取自静脉血样本。基因分型由日本横滨的Riken Genesis公司采用Illumina HumanOmni1-Quad芯片（Illumina, Inc., 圣地亚哥）完成。使用PLINK v1.07软件对共计713,495个常染色体SNP进行质量评估。剔除分型成功率低于95%、哈迪-温伯格平衡偏差P<0.001或次要等位基因频率低于10%的SNP。最终剩余的534,404个SNP用于后续分析。