High expression of TREM2 promotes EMT via the PI3K/AKT pathway in gastric cancer: bioinformatics analysis and experimental verification

Abstract

Background: To date, the pathogenesis of gastric cancer (GC) remains unclear. We combined public database resources and bioinformatics analysis methods, explored some novel genes and verified the experiments to further understand the pathogenesis of GC and to provide a promising target for anti-tumor therapy. Methods: We downloaded the chip data related to GC from the Gene Expression Omnibus (GEO) database, extracted differentially expressed genes (DEGs), and then determined the key genes in the development of GC via PPI networks and model analysis. Functional annotation via GO and KEGG enrichment of DEGs was used to understand the latent roles of DEGs. The expression of the triggering receptor expressed on myeloid cells 2 (TREM2) gene in GC cell lines was verified via RT-PCR and western blotting. Moreover, the CCK-8, wound healing assay, and transwell migration and invasion assays were used to understand the changes in the proliferation, migration, and invasion abilities of GC cells after silencing TREM2. Western blotting verified the interaction between TREM2 and PI3K predict of the string website, as well as the effect of TREM2 on EMT. Finally, a lung metastasis model was used to explore the relationship between TREM2 and metastasis. Results: Our study identified 16 key genes, namely BGN, COL1A1, COL4A1, COL5A2, NOX4, SPARC, HEYL, SPP1, TIMP1, CTHRC1, TREM2, SFRP4, FBXO32, GPX3, KIF4A, and MMP9 genes associated with GC. The EMT-related pathway was the most significantly altered pathway. TREM2 expression was higher in GC cell lines and was remarkably associated with tumor invasion depth, TNM stage, histological grade, histological type, anatomic subdivision, and Helicobacter pylori state. Knockdown of TREM2 expression inhibited the proliferation, migration, and invasion of GC cells as well as the progression of EMT by PI3K/AKT signaling in vitro. In addition, lung metastasis were decreased in vivo. Conclusions: We identified some important genes associated with the progression of GC via public database analysis, explored and verified the effects of proto-oncogene TREM2 on EMT via the PI3K/AKT pathway. TREM2 may be a novel target in the GC therapy.

Keywords: EMT; PI3K/AKT pathway; TREM2; bioinformatics; gastric cancer (GC).

Figure 1

Identification of key genes and function and pathway analysis about GSE72305 and GSE103236. (A) The expression profile of gastric cancer related expressed genes. (B) Volcanic plot of gastric cancer-related differentially expressed genes (DEGs) respectively. (C, D) The biological process of EDGs by Gene Ontology (GO) and the biological pathways analysis of DEGs by Kyto Encyclopedia of Gene and Genomes (KEGG) enrichment about GSE72305 and GSE103236.

Figure 2

Screening of hub genes by PPI network and model analysis. (A) Protein-protein interaction network of differentially expressed genes in GSE72305 and GSE103236. The different colored lines indicate the different evidences demonstrating the interaction. (B, C) The heatmap about expression profile of nine common genes in GSE72305 and GSE103236. (D) Seven common DEGs in GSE72305 and GSE103236 by Venn diagrm software.

Figure 3

The survival analysis of the key genes in GC patients. The online Kaplan-Meier plotter tool was used to analysis the prognostic information of the 16 key genes. p<0.05 was considered statistically significant.

Figure 4

The TREM2 expression level between GC sample and normal tissue with immunohistochemistry analysis from the human protein ATLAS.

Figure 5

The expression of TREM2 in GC and affected cell proliferation. (A, B) TREM2 is high expression in GC cell lines at mRNA and protein levels. (C, D) Verified the ability of silence, shTREM2# 2 is more effectivity than shTREM2# 1 both in BGC-823 and SGC-7901 by RT-PCR and Western Blot. (E) The proliferation of BGC-823 and SGC-7901 were inhibited after knockdown of TREM2 measured by CCK8.

Figure 6

TREM2 affected the GC cell migration and invasion. Knockdown of TREM2, (A, B, C) the ability of migration assessed with wound healing assay and transwell migration assays (magnification, ×100), the ability of invasion measured by transwell invasion assays (magnification, X100). The statistical analysis was shown in the bar graphs (*, P<0.05, vs. control group; mean ± SD, n=3).

Figure 7

TREM2 promote EMT by PI3K/AKT pathway. (A) In GC cell line SGC-7901, the expression of E-cadherin was increased after knockdown of TREM2, measured by immunofluorescence staining (magnification, X100). (B) After knockdown of TREM2, the expression of E-cadherin was increased, the expression of N-cadherin, Vimentin, p-PI3K, p-AKT were decreased, the expression of PI3K and AKT not changed significantly. (C) The statistical analysis of proteins expression was shown in the bar graphs (*, P<0.05, vs. control group; mean ± SD, n=3). (D) Protein interaction network of TREM2 interacts with PIK3R2, PIK3CA, and PIK3CB proteins predicted by string website.

Figure 8

Knockdown of TREM2 inhibits GC cells metastasis in vivo. (A) The lung metastasis tumors about BGC-823 shCtrl group were increased than BGC-823 shTREM2# 2 group (n = 5/group). A representative photographs of gross lungs from indicated groups are shown (n = 5/group). (B) Representative H&E images of lung metastatic tumors.