Expression level of GAB2 in pan-cancer
In order to explore the expression level of GAB2 in pan-cancer, we detected the mRNA expression level of GAB2 using Oncomine database. As shown in Figure 1A, the expression level of GAB2 in gastric cancer, leukemia and lymphoma was significantly increased, and the expression level also was higher in breast cancer and head and neck cancer. Meanwhile, the expression of GAB2 was lower in cancers, including esophageal cancer, head and neck cancer, leukemia and lung cancer.
In addition, we further analyzed the expression of GAB2 in TCGA by using Timer software, and the results were shown in Figure 1B. GAB2 expression was significantly lower in BLCA, BRCA, CESC(Cervical and endocervical cancer), COAD, HNSC(Head and Neck squamous cell carcinoma), LUAD, LUSC, PRAD(Prostate adenocarcinoma), READ(Rectum adenocarcinoma), UCEC than that in normal tissue. However, we also found that the expression level of GAB2 was higher in CHOL(Cholangiocarcinoma), KICH(Kidney Chromopphobe), KIRC, LIHC(Liver hepatocellular carcinoma), PCPG(Pheochromocytoma and Paraganglioma), THCA than that in normal tissue.
Because some tumors in the TCGA data lacked normal tissue, we combined the TCGA and GTEx databases in GEPIA to analyze the difference in the expression of GAB2 in normal tissue and tumor tissue. The results were shown in Figure 1C. The expression levels of GAB2 were different between normal tissues and tumor tissues in DLBC(Lymphoid Neoplasm Diffuse Large B-cell Lymphoma), LGG(Brain Lower Grade Glioma) and UCS(Uterine Carcinosarcoma).
The relationship between the expression level of GAB2 and prognosis in cancer
We further explored the relationship between GAB2 expression levels and survival in pan-cancer patients. Kaplan-Meier Plotter tools were used to analyze the relationship between GAB2 mRNA levels and survival of patients with hematoma. We found that the Overall Survival(OS) of patients with high level of GAB2 mRNA was shorter than that with low level of GAB2 mRNA in BLUC, LUSC and UCEC. At the same time, we still found the opposite phenomenon, tumor patients with high GAB2 mRNA level had significantly better OS than tumor patients with low levels in BRCA, KIRC, LUAD, READ, SARC(Sarcoma), THCA and THYM. (Figure 2).
Phosphorylation analysis of GAB2 in pan-cancer
We analyzed the differences in phosphorylation levels at different sites of GAB2 in normal and tumor tissues from different tumors. As shown in Figure 3, Ualcan tool was used to analyze 6 types of tumor(breast cancer, LUAD, ovarian cancer, UCEC, COAD and KIRC). In breast cancer, the phosphorylation levels of S185, T508 and Y373 sites were different between normal and tumor tissues, and phosphorylation levels in normal tissues were higher than those in tumor tissues. A similar phenomenon was observed, phosphorylation levels is significantly higher in normal tissues at S330, S505, and T353 rather than S185, T508, and Y373 in Lung adenocarcinoma. However, in ovarian cancer, phosphorylation levels at S172 and T353 sites in tumor tissues were significantly higher than those in normal tissues. Meanwhile, the phosphorylation level of T353 was also different in the tumor and normal tissues of colon cancer. However, in renal clear cell carcinoma, the phosphorylation level of GAB2 was not statistically significant between tumor and normal tissues. We found that the phosphorylation level at T353 differed in lung adenocarcinoma, ovarian cancer, and colon cancer. Therefore, phosphorylated T353 may be a potential target for tumor therapy.
Correlation and enrichment analyses
In order to explore the function of GAB2, we chose to use the TCGA database to analyze the correlation between GAB2 and other genes in ovarian cancer. Finally, we selected the first 300 genes related to GAB2 for enrichment analysis. As shown in Figure 4A, the top 50 genes positively and negatively correlated with GAB2 were respectively displayed. Afterwards, Functional enrichment and Gene Ontology (GO) analysis revealed that GAB2 is involved in cell movement, including integrator complex; myofibril; contractile fiber part; region of cytosol; apical junction complex; clathrin-coated pit; postsynaptic cytosol; cortical actin cytoskeleton; lamellipodium;cell-substrate adherens junction and so on(Figure 4B). The main molecular functions, biological processes and cell composition involved in GAB2 are shown in Table 1.
Correlation analysis between GAB2 expression and infiltrating immune cells
Tumor infiltrating immune cells, which participate in the composition of tumor microenvironment, affect the survival of patients with various tumors. Therefore, taking CD8+T as an example, TIMER tool was applied to explore the correlation between the expression level of GAB2 and infiltrating immune cell in pan-cancer. The result is shown in Figure 5. The expression level of GAB2 was positively correlated with tumor cell purity in PCPG and SARC, but negatively correlated with tumor cell purity in PRAD. In addition, we found that GAB2 expression was significantly correlated with CD8+T cell infiltration levels in PCPG, PRAD and SARC.