Zhai Ming, Luan Peipei, Shi Yefei, Li Bo, Kang Jianhua, Hu Fan, Li Mingjie, Du Lei, Zhou Donglei, Jian Weixia, Peng Wenhui
Department of Cardiology, Shanghai Tenth People's Hospital, School of Medicine, Tongji University, 301 Middle Yanchang Road, Shanghai 200072, China.
Department of Endocrinology, Xinhua Hospital, Shanghai Jiaotong University, School of Medicine, 1665 Kongjiang Road, Shanghai 200092, China.
Int J Endocrinol. 2021 Jan 22;2021:8820089. doi: 10.1155/2021/8820089. eCollection 2021.
Low-grade chronic inflammation in dysfunctional adipose tissue links obesity with insulin resistance through the activation of tissue-infiltrating immune cells. Numerous studies have reported on the pathogenesis of insulin-resistance. However, few studies focused on genes from genomic database. In this study, we would like to explore the correlation of genes and immune cells infiltration in adipose tissue via comprehensive bioinformatics analyses and experimental validation in mice and human adipose tissue.
Gene Expression Omnibus (GEO) datasets (GSE27951, GSE55200, and GSE26637) of insulin-resistant individuals or type 2 diabetes patients and normal controls were downloaded to get differently expressed genes (DEGs), and GO and KEGG pathway analyses were performed. Subsequently, we integrated DEGs from three datasets and constructed commonly expressed DEGs' PPI net-works across datasets. Center regulating module of DEGs and hub genes were screened through MCODE and cytoHubba in Cytoscape. Three most significant hub genes were further analyzed by GSEA analysis. Moreover, we verified the predicted hub genes by performing RT qPCR analysis in animals and human samples. Besides, the relative fraction of 22 immune cell types in adipose tissue was detected by using the deconvolution algorithm of CIBERSORT (Cell Type Identification by Estimating Relative Subsets of RNA Transcripts). Furthermore, based on the significantly changed types of immune cells, we performed correlation analysis between hub genes and immune cells. And, we performed immunohistochemistry and immunofluorescence analysis to verify that the hub genes were associated with adipose tissue macrophages (ATM).
Thirty DEGs were commonly expressed across three datasets, most of which were upregulated. DEGs mainly participated in the process of multiple immune cells' infiltration. In protein-protein interaction network, we identified , , and as hub genes. GSEA analysis suggested high expression of the three hub genes was correlated with immune cells functional pathway's activation. Immune cell infiltration and correlation analysis revealed that there were significant positive correlations between and M0 macrophages, and M0 macrophages, Plasma cells, and CD8 T cells. Finally, hub genes were associated with ATMs infiltration by experimental verification.
This article revealed that , , and were potential hub genes associated with immune cells' infiltration and the function of proinflammation, especially adipose tissue macrophages, in the progression of obesity-induced diabetes or insulin-resistance.
功能失调的脂肪组织中的低度慢性炎症通过激活组织浸润免疫细胞，将肥胖与胰岛素抵抗联系起来。许多研究报道了胰岛素抵抗的发病机制。然而，很少有研究关注基因组数据库中的基因。在本研究中，我们希望通过全面的生物信息学分析以及在小鼠和人类脂肪组织中的实验验证，探索脂肪组织中基因与免疫细胞浸润的相关性。
下载胰岛素抵抗个体或2型糖尿病患者以及正常对照的基因表达综合数据库（GEO）数据集（GSE27951、GSE55200和GSE26637）以获取差异表达基因（DEGs），并进行基因本体（GO）和京都基因与基因组百科全书（KEGG）通路分析。随后，我们整合来自三个数据集的DEGs，并构建跨数据集的共同表达DEGs的蛋白质-蛋白质相互作用（PPI）网络。通过Cytoscape中的MCODE和cytoHubba筛选DEGs的中心调节模块和枢纽基因。通过基因集富集分析（GSEA）进一步分析三个最显著的枢纽基因。此外，我们通过在动物和人类样本中进行逆转录定量聚合酶链反应（RT qPCR）分析来验证预测的枢纽基因。此外，使用CIBERSORT（通过估计RNA转录本的相对子集进行细胞类型鉴定）的反卷积算法检测脂肪组织中22种免疫细胞类型的相对比例。此外，基于显著变化的免疫细胞类型，我们进行了枢纽基因与免疫细胞之间的相关性分析。并且，我们进行了免疫组织化学和免疫荧光分析，以验证枢纽基因与脂肪组织巨噬细胞（ATM）相关。
在三个数据集中共同表达了30个DEGs，其中大多数上调。DEGs主要参与多种免疫细胞浸润过程。在蛋白质-蛋白质相互作用网络中，我们鉴定出[具体基因1]、[具体基因2]和[具体基因3]为枢纽基因。GSEA分析表明，这三个枢纽基因的高表达与免疫细胞功能通路的激活相关。免疫细胞浸润和相关性分析显示，[具体基因1]与M0巨噬细胞、[具体基因2]与M0巨噬细胞、浆细胞和CD8 T细胞之间存在显著正相关。最后，通过实验验证枢纽基因与ATM浸润相关。
本文揭示了[具体基因1]、[具体基因2]和[具体基因3]是与免疫细胞浸润以及肥胖诱导的糖尿病或胰岛素抵抗进展中的促炎功能相关的潜在枢纽基因，尤其是与脂肪组织巨噬细胞相关。
