mRNA and protein expression of NCOA4
We assessed NCOA4 expression levels using the Oncomine database. The NCOA4 expression levels of Brain and CNS cancer, Leukemia, Other cancer, Pancreatic cancer, Sarcoma were significant increased than those normal tissues. However, The NCOA4 expression levels of Breast cancer, Leukemia, Lung cancer, Lymphoma and Other cancer were obviously decreased compared with those normal tissues. (Figure1, A). And then we used the TIMER database to analyze the expression of NCOA4 in more human cancers. The result showed that NCOA4 expression levels differences in tumor tissue and adjacent normal tissue (Figure1, B): NCOA4 expression in BLCA (Bladder Urothelial Carcinoma), BRCA (Breast invasive carcinoma), COAD (Colon adenocarcinoma), KICH (Kidney Chromophobe), KIRP (Kidney renal papillary cell carcinoma), LUAD (Lung adenocarcinoma), LUSC (Lung squamous cell carcinoma), READ (Rectum adenocarcinoma), SKCM (Skin Cutaneous Melanoma), THCA (Thyroid carcinoma), UCEC (Uterine Corpus Endometrial Carcinoma) were lower than those of neighboring tissues. In addition, Data mining using the GEPIA and UALCAN, TNM plot databases was further confirmed that NCOA4 expression is decreased in LUAD (Figure1, C-F). We also used the HPA database for immunohistochemical analysis (Figure1, G). The protein expression level of NCOA4 was distinctly decreased in lung adenocarcinoma.
Relationship between expression of NCOA4 and clinicopathological characteristics of LUAD
In order to explore the reasons for the low expression of NCOA4 in lung adenocarcinoma, the UALCAN database was used to further analyze the correlation between NCOA4 expression and clinical information. In terms of patient’s age, significant low expression of NCOA4 was observed in 41-100 years (Figure2, A). As shown in Figure 2B, mining of the UALCAN database results suggested that NCOA4 expression was decreased in males and females. For the cancer stage, NCOA4 expression level was significantly decreased in stage 1, 2, 3,4 (Figure2, C). At the same time, analysis of patient’s race, TP53 muation status, patient’s smoking habits showed that the expression level of NCOA4 in patients with lung adenocarcinoma decreased relative to normal samples (Figure2, D-F).
Prognostic analysis of NCOA4 in LUAD
Next, we applied to Kaplan-Meier plotter to explore the correlation between NCOA4 expression and the prognosis of LUAD patient. As a result, patients of LUAD with high NCOA4 level had better overall survival rates than patients with low NCOA4 level (Figure3, A). In the meantime, low NCOA4 expression was correlated with poor PPS and poor FP as shown in Figure3B, C. In addition, the prognostic potential of NCOA4 was further analyzed using OncoLnc database (Figure3).
Correlation between expression of NCOA4 and infiltrating immune cells
Then, we explored the correlation between NCOA4 expression and six major tumor-infiltrating immune cells in the TIMER database. The results showed that NCOA4 expression levels were significantly associated with CD8+T cells (r=0.365, p=8.41e-17), macrophages cells (r=0.386, p=1.19e-18), neutrophil cells (r=0.329, p=1.09e-13), dendritic cells (r=0.336, p=2.30e-14) in LUAD (Figure4, A). The expression level of NCOA4 was remarkably positively correlated with CD8+T cells, macrophages cells, neutrophil cells and dendritic cells, indicating that the expression level of NCOA4 was positively related to lung adenocarcinoma immune infiltration. We also explored the difference of immune cells in different groups of samples through CIBERSORT. As shown in Figure4B, NCOA4 levels were found to be correlated with infiltration by sixteen types of tumor-infiltrating cells, including memory B cell, plasma B cell, resting memory CD4+T cell, activated memory CD4+T cell, follicular helper T cell, Tregs, gamma delta T cell, resting NK cell, Monocyte, Macrophage M0, Macrophage M1, resting Myeloid dendritic cell, activated Myeloid dendritic cell, activated Mast cell, Eosinophil, Neutrophil.
Relationship between NCOA4 expression level and different immune marker sets
Based on a set of immune markers in LUAD, we analyzed the link between NCOA4 expression and immune markers through the TIMER database (Table1). NCOA4 expression was significantly correlated with the levels of most markers in different types of immune cells in lung adenocarcinoma. It is interesting to note that NCOA4 expression showed a strong positive association with M2 macrophages and TAMs immune marker genes. After adjustment for tumor purity, we found a significant correlation between NCOA4 expression and markers of M2 macrophages and TAMs. Expression of CCL18, CD163, MS4A4A, and MRC1 of M2 macrophages and CCL2, CD86, CCR5and CD80 of TAMs showed a significantly positive correlation with NCOA4 expression (P<0.001). However, M1 macrophage markers, such as IRF5, ARG2 and PTGS2 showed a weak or no correlation with NCOA4 expression. M1 macrophage markers, such as CXCL10 and NOS2 show a correlation with NCOA4 expression. These findings elucidated that NCOA4 expression was correlated with immune infiltration, and it might serve as a potential immunotherapeutic target for LUAD treatment.
Prognostic value of NCOA4 from immune cells in LUAD
To investigate whether the expression of NCOA4 affects the prognosis of LUAD patients by directly influencing the infiltration of immune cells, we used the KM plotter database to analyze the prognosis of NCOA4 expression in different subgroups of immune cells in LUAD. As shown in Figure5, low expression of NCOA4 in enriched B cells, enrich CD8+T cells, enriched macrophages, decreased natural killer T cells, decreased type 1 T helper cells, enriched type 2 T helper cells cohorts in LUAD was associated with poor OS. Furthermore, there was no significant correlation between NCOA4 expression and decreased CD8+T cells, decreased macrophages, enriched natural killer T cells, enriched type 1 T helper cells, decreased type 2 T helper cells cohorts in LUAD.
Network construction of NCOA4 interacting genes and proteins.
We used the GeneMANIA database to constructe the gene-gene interaction network of NCOA4. There are 20 surrounding nodes represent genes associated with NCOA4. The middle node represents NCOA4(Figure6, A). The three genes most significantly associated with NCOA4 are PPARA, AR, RXRA. Then, in order to further investigate the biological role of NCOA4, we used the STRING database to constructe a PPI network.
Construction of NCoA4 co-expression network and GO, KEGG enrichment analysis.
In order to understand the biological function of NCOA4 in lung adenocarcinoma, we used the LinkedOmics database to build a co-expression network of NCOA4. As shown in Figure7A, 7576genes (red dots) are positively related to NCOA4 and 10520 genes (green dots) are negatively related to NCOA4(P<0.05). As shown in Figure 7B-C, the first 50 positive and negative genes associated with NCOA4 are shown as heat maps. NCOA4 expression is presented with genes strong positive correlation such as SPRR3(positive correlation, r=0.895, p=0.0348), MFSD1(r= 0.848, p=4.16e-29), CTSS (r=0.812, p=7.71e-25), but negative correlation with genes such as PHC1(negative correlation, r=-0.714, p=5.25e-17), DLG5(r=-0.703, p=2.48e-16), CEP164(r=-0.702, p=2.76e-16). In addition, we used the LinkedOmics to perform a rich analysis based on the NCOA4 co-expression gene network. As shown Figure7D-G, at the level of biological process, NCOA4 co-expression is rich in the positive regulation of neutrophil mediated immunity, translation initiation, antigen processing and presentation and so on. At the cell component level, NCOA4 co-expression is mainly rich in ribosome, primary lysosome, vesicle lumen, and at the molecular function level, NCOA4 co-expression is mainly rich in structural constituent of ribosome, electron transfer activity, oxidoreductase activity, acting on NAD(P)H. At the KEGG pathway, NCOA4 co-expression is mainly rich in Ribosome, Lysosome, Ferroptosis.
Correlation between NCAO4 and surface markers of lung cancer stem cells.
Tumor stem cells play a key role in tumorigenesis, development and metastasis. We first used machine learning to evaluate the stemness index of lung adenocarcinoma samples. We found that the low expression group of NCOA4 had a higher mRNAsi(Figure8, D). In addition, we used GEPIA to predict the correlation between NCOA4 and lung cancer stem cell surface markers. We found that NCOA4 was significantly positively correlated with CD44, CD166(ALCAM), and CD90(THY1)(Figure8, A-C). These results suggest that NCOA4 may interact with lung cancer stem cell surface markers, play an important role in targeted treatment of lung adenocarcinoma, and can be used as a reliable biomarker of lung adenocarcinoma.