基于GEO数据库筛选代谢综合征潜在生物标志物

doi:10.3969/j.issn.1673-8640.2023.01.007

摘要/Abstract

摘要：

目的通过分析基因表达综合数据库（GEO）中代谢综合征患者外周血单个核细胞（PBMC）的基因数据集（GSE98895和GSE23561），寻找代谢综合征的关键基因。方法分析GSE98895数据集中代谢综合征患者与健康志愿者PBMC的差异表达基因。对差异表达基因进行基因本体论（GO）富集分析和京都基因与基因组数据库（KEGG）富集分析。构建差异表达基因的蛋白质互作（PPI）网络，筛选出度值最高的10个基因作为核心基因。通过GSE98895数据集、GSE23561数据集和人类蛋白图谱（HPA）数据库验证核心基因在代谢综合征患者PBMC中的表达，构建基于核心基因的环状RNA（circRNA）-微小RNA（miRNA）-mRNA互作网络。结果从GSE98895数据集中共获得155个代谢综合征患者与健康志愿者的差异表达基因，其中表达上调62个、表达下调93个。GO和KEGG富集分析结果显示，差异表达基因主要涉及转录调节、免疫应答、cAMP信号通路、T细胞受体信号通路等。PPI网络中度值最高的10个基因分别为：JUN、IFNG、MMP9、PIK3R1、LCK、NGF、PDGFRB、CD79B、CX3CR1和ABCA1。验证结果显示，代谢综合征患者PBMC中IFNG和LCK的表达均显著高于健康志愿者（P<0.05）。HPA数据库分析结果显示，IFNG和LCK主要表达于T细胞和自然杀伤（NK）细胞。构建circRNA-miRNA-mRNA互作网络，与IFNG相关的miRNA有7个、circRNA有43个；与LCK相关的miRNA有16个、circRNA有62个。结论代谢综合征患者PBMC中IFNG和LCK表达显著增加，对基于IFNG和LCK的circRNA-miRNA-mRNA互作网络进行深入研究，可能有助于发现代谢综合征发生、发展的分子机制和生物标志物。

关键词: 代谢综合征, 基因表达综合数据库, 差异表达基因, 生物标志物

Abstract:

Objective By analyzing the Gene Expression Omnibus （GEO） database （GSE98895 and GSE23561） of peripheral blood mononuclear cells （PBMC） from patients with metabolic syndrome，and to identify the key genes for metabolic syndrome. Methods The differentially expressed genes of PBMC in metabolic syndrome patients and healthy subjects in the GSE98895 dataset were analyzed. Gene Ontology （GO） and Kyoto Encyclopedia of Genes and Genomes （KEGG） pathway analysis of differentially expressed genes were carried out. Protein-protein interaction （PPI） network analysis was also performed. The top 10 hub genes with a high degree of connectivity were selected in the PPI network. The hub gene expression in PBMC of metabolic syndrome patients had been illustrated in the GSE98895 and GSE23561 datasets and human protein atlas （HPA） dataset. A circular RNA（circRNA）-microRNA（miRNA）-mRNA network for the pathogenesis of metabolic syndrome was constructed. Results A total of 155 differentially expressed genes between metabolic syndrome patients and healthy subjects were identified，among which 62 genes were up-regulated，and 93 genes were down-regulated. The results of the GO and the KEGG enrichment pathway analysis indicated that the differentially expressed genes were mainly involved in the regulation of transcription，immune response，cAMP signaling pathway and T cell receptor signaling pathway. Totally，10 genes （JUN，IFNG，MMP9，PIK3R1，LCK，NGF，PDGFRB，CD79B，CX3CR1 and ABCA1） with the highest degree scores were identified. The expressions of IFNG and LCK in PBMC of patients with metabolic syndrome were higher than those of healthy subjects （P<0.05）. HPA dataset analysis showed that IFNG and LCK were mainly expressed in T cells and natural killer（NK） cells. A circRNA-miRNA-mRNA network had been constructed for metabolic syndrome. Totally，7 miRNA and 43 circRNA associated with IFNG were identified，and 16 miRNA and 62 circRNA associated with LCK were identified. Conclusions IFNG and LCK have up-regulated expressions in PBMC of metabolic syndrome patients. An in-depth study of the circRNA-miRNA-mRNA network for IFNG and LCK may help to discover the molecular mechanism of occurrence and development and the biomarkers of metabolic syndrome.

Key words: Metabolic syndrome, Gene Expression Omnibus database, Differentially expressed gene, Biomarker

中图分类号:

R589

陈雪薇, 李一荣. 基于GEO数据库筛选代谢综合征潜在生物标志物[J]. 检验医学, 2023, 38(1): 32-38.

CHEN Xuewei, LI Yirong. Candidate biomarkers in metabolic syndrome based on GEO database[J]. Laboratory Medicine, 2023, 38(1): 32-38.

图/表 7

图1 代谢综合征患者与健康志愿者差异表达基因火山图注：表达上调（62个）；表达下调（93个）；表达无差异；图中竖向虚线为1倍差异阈值，横向虚线为错误发现率=0.05（阈值）。

图2 差异表达基因的GO和KEGG富集分析注：（a）GO富集分析；（b）KEGG富集分析

图3 Reactome数据库对差异表达基因的富集分析注：①NOTCH2为神经源性基因座缺口同源蛋白2（neurogenic locus notch homolog protein 2）；②RUNX1为Runt相关转录因子1（Runt-related transcription factor 1）；③FOXP3为叉头框蛋白P3（forkhead box P3）；④RUNX3为Runt相关转录因子3（Runt-related transcription factor 3）；⑤PI3K为磷脂酰肌醇3-激酶（phosphoinositide 3-kinase）；⑥Akt为蛋白激酶B（protein kinase B，PKB）。

图4 PPI网络和核心基因分析注：（a）PPI网络；（b）核心基因的互作子网。

图5 GSE98895和GSE23561数据集中代谢综合征患者与健康志愿者核心基因表达的比较注：（a）GSE98895数据集；（b）GSE23561数据集；与健康志愿者比较，*P<0.05，**P<0.01；健康志愿者；代谢综合征患者。

图6 HPA数据库中核心基因IFNG和LCK表达的分析注：（a）IFNG在PBMC中的表达；（b）LCK在PBMC中的表达；粒细胞；单核细胞； T细胞； B细胞；树突状细胞； NK细胞；总PBMC。

图7 circRNA-miRNA-mRNA互作网络注：（a）与IFNG结合miRNA的Venn图；（b）与LCK结合miRNA的Venn图；（c）由circRNA-miRNA-IFNG组成的互作网络；（d）由circRNA-miRNA-LCK组成的互作网络。

参考文献 19

[1]	BORGHI C, AGABITI-ROSEI E, JOHNSON R J, et al. Hyperuricaemia and gout in cardiovascular,metabolic and kidney disease[J]. Eur J Intern Med, 2020, 80:1-11. DOI URL
[2]	WANG C H, LUNDH M, FU A, et al. CRISPR-engineered human brown-like adipocytes prevent diet-induced obesity and ameliorate metabolic syndrome in mice[J]. Sci Transl Med, 2020, 12(558):eaaz8664. DOI URL
[3]	MASSILLO C, DUCA R B, LACUNZA E, et al. Adipose tissue from metabolic syndrome mice induces an aberrant miRNA signature highly relevant in prostate cancer development[J]. Mol Oncol, 2020, 14(11):2868-2883. DOI URL
[4]	HU C, JIA W. Multi-omics profiling:the way towards precision medicine in metabolic diseases[J]. J Mol Cell Biol, 2021. [Epub ahead of print]
[5]	刘晓妮, 王颖, 李桃桃, 等. 2型糖尿病合并代谢综合征患者胰岛素抵抗指数与慢性炎症指标相关性研究[J]. 检验医学, 2019, 34(9):826-830.
[6]	ULVEN S M, HOLVEN K B, RUNDBLAD A, et al. An isocaloric nordic diet modulates RELA and TNFRSF1A gene expression in peripheral blood mononuclear cells in individuals with metabolic syndrome-a SYSDIET sub-study[J]. Nutrients, 2019, 11(12):2932. DOI URL
[7]	FRANCESCHI C, GARAGNANI P, PARINI P, et al. Inflammaging:a new immune-metabolic viewpoint for age-related diseases[J]. Nat Rev Endocrinol, 2018, 14(10):576-590. DOI
[8]	CAI J, XU M, ZHANG X, et al. Innate immune signaling in nonalcoholic fatty liver disease and cardiovascular diseases[J]. Annu Rev Pathol, 2019, 14:153-184. DOI PMID
[9]	DARYABOR G, ATASHZAR M R, KABELITZ D, et al. The effects of type 2 diabetes mellitus on organ metabolism and the immune system[J]. Front Immunol, 2020, 11:1582. DOI PMID
[10]	KEWCHAROENWONG C, SAENWONGSA W, WILLCOCKS S J, et al. Glibenclamide alters interleukin-8 and interleukin-1β of primary human monocytes from diabetes patients against Mycobacterium tuberculosis infection[J]. Tuberculosis (Edinb), 2020, 123:101939. DOI URL
[11]	SCHNEIDER-MATYKA D, SZKUP M, OWCZAREK A J, et al. The relationship between the IFNG (rs2430561) polymorphism and metabolic syndrome in perimenopausal women[J]. Medicina (Kaunas), 2020, 56(8):384.
[12]	JAISWAL N, AGRAWAL S, AGRAWAL A. High fructose-induced metabolic changes enhance inflammation in human dendritic cells[J]. Clin Exp Immunol, 2019, 197(2):237-249. DOI PMID
[13]	LIU J, GUO Z, ZHANG Y, et al. LCK inhibitor attenuates atherosclerosis in ApoE^-/- mice via regulating T cell differentiation and reverse cholesterol transport[J]. J Mol Cell Cardiol, 2020, 139:87-97. DOI URL
[14]	LIN W, LIU H, TANG Y, et al. The development and controversy of competitive endogenous RNA hypothesis in non-coding genes[J]. Mol Cell Biochem, 2021, 476(1):109-123. DOI
[15]	DOLATI S, SHAKOURI S K, DOLATKHAH N, et al. The role of exosomal non-coding RNAs in aging-related diseases[J]. Biofactors, 2021, 47(3):292-310. DOI PMID
[16]	ORTIZ-DOSAL A, RODIL-GARCÍA P, SALAZAR-OLIVO L A. Circulating microRNAs in human obesity:a systematic review[J]. Biomarkers, 2019, 24(6):499-509. DOI URL
[17]	HUANG Y, ZHAO L, YAN Y, et al. PBMCs to stress-associated miR-18a-5p and miR-22-3p ratios as new indicators of metabolic syndrome[J]. Biomed Res Int, 2020, 2020:8159342.
[18]	BROZZI F, REGAZZI R. Circular RNAs as novel regulators of β-cell functions under physiological and pathological conditions[J]. Int J Mol Sci, 2021, 22(4):1503. DOI URL
[19]	ZAIOU M. Circular RNAs as potential biomarkers and therapeutic targets for metabolic diseases[J]. Adv Exp Med Biol, 2019, 1134:177-191. DOI PMID