Integrating proteomics and transcriptomics for the identification of potential targets in early colorectal cancer

Abstract

Colorectal cancer (CRC) is one of the most common malignancies worldwide. At present, CRC can often be treated upon diagnosis at stage I or II, or when dysplasia is detected; however, 60‑70% of cases are not diagnosed until they have developed into late stages of the disease or until the malignancy is identified. Diagnosis of CRC at an early stage remains a challenge due to the absence of early‑stage‑specific biomarkers. To identify potential targets of early stage CRC, label‑free proteomics analysis was applied to paired tumor‑benign tissue samples from patients with stage II CRC (n=21). A total of 2,968 proteins were identified; corresponding RNA‑Sequencing data were retrieved from The Cancer Genome Atlas‑colon adenocarcinoma. Numerous bioinformatics methods, including differential expression analysis, weighted correlation network analysis, Gene Ontology and protein‑protein interaction analyses, were applied to the proteomics and transcriptomics data. A total of 111 key proteins, which appeared as both differentially expressed proteins and mRNAs in the hub module, were identified as key candidates. Among these, three potential targets [protein‑arginine deiminase type‑2 (PADI2), Fc fragment of IgG binding protein (FCGBP) and phosphoserine aminotransferase 1] were identified from the pathological data. Furthermore, the survival analysis indicated that PADI2 and FCGBP were associated with the prognosis of CRC. The findings of the present study suggested potential targets for the identification of early stage CRC, and may improve understanding of the mechanism underlying the occurrence of CRC.

Figure 1

Flow chart for identifying potential targets of stage II CRC. CRC, colorectal cancer; TCGA, The Cancer Genome Atlas; COAD, colon adenocarcinoma; RNA-Seq, RNA-Sequencing; FDR, false discovery rate; WGCNA, weighted correlation network analysis; DEP, differentially expressed protein; PPI, protein-protein interaction; GO, Gene Oncology; GSEA, Gene Set Enrichment Analysis.

Figure 2

Distribution of DEPs. (A) A heatmap of DEPs of stage II CRC. The horizontal axis represents the name of the sample, while the vertical axis indicates the genes. (B) A volcano plot of DEPs of stage II CRC. The horizontal axis represents the 'log2Fold Change', while the vertical axis indicates the '-log10FDR'. P<0.05; Fold Change >4. DEPs, differentially expressed proteins; CRC, colorectal cancer; FDR, false discovery rate.

Figure 3

Module-trait association and members of a module versus gene significance of the hub module. (A) Module-trait association of data. The first row of each cube represents the correlation coefficient between the module and the trait, and the second row indicates the significance of the correlation coefficient. The correlation coefficient is presented by the color of the cube; red indicates positive correlation, while green represents negative correlation. (B) Members of the blue modules vs. gene significance for normal. (C) Members of the blue module vs. gene significance for stage II COAD. COAD, colon adenocarcinoma; ME, module eigengene.

Figure 4

GO enrichment analysis of key genes. (A) Top 15 GO terms listed by the count of genes associated with the GO term. The color of the dot indicated the adjusted P-value; red indicated higher significance in association with the GO term. (B) Top 15 GO terms. The horizontal axis presents the genes associated with the GO term; genes of the same GO term were clustered. GO, Gene Ontology.

Figure 5

Protein-protein interaction network of stage II CRC. The six key clusters are presented as different colors; the seed node of the cluster is presented as a diamond. The color of the inner ring of the node indicates the fold change in proteomics data. The intensity of the outer circle indicates the fold change in the mRNA. CRC, colorectal cancer; COAD, colon adenocarcinoma; PADI2, protein-arginine deaminase type-2; FCGBP, Fc fragment of IgG binding protein; GAR1, GAR1 ribonucleoprotein; PSAT1, phosphoserine aminotransferase; ETFDH, electron-transfer flavoprotein-ubiquinone oxidoreductase; METTL7A, methyltransferase-like 7A.

Figure 6

Pathological images of critical targets. The pathological and normal images of (A) PADI2, (B) FCGBP, (C) GAR1, (D) PSAT1 and (E) ETFDH. The information was obtained from the Human Protein Atlas. Magnification, x20. PADI2, protein-arginine deaminase type-2; FCGBP, Fc fragment of IgG binding protein; GAR1, GAR1 ribonucleoprotein; PSAT1, phosphoserine aminotransferase; ETFDH, electron-transfer flavoprotein-ubiquinone oxidoreductase; CRC, colorectal cancer.

Figure 7

Statistical analysis of critical targets. Statistical analysis of (A) PADI2 (P<0.001 in RNA-Seq and P<0.001 in proteomics), (B) FCGBP (P<0.001 in RNA-Seq and P<0.001 in proteomics), (C) GAR1 (P<0.001 in RNA-Seq and P=0.020 in proteomics), (D) PSAT1 (P<0.001 in RNA-Seq and P=0.003 in proteomics) and (E) ETFDH (P<0.001 in RNA-Seq and P=0.108 in proteomics). PADI2, protein-arginine deaminase type-2; FCGBP, Fc fragment of IgG binding protein; GAR1, GAR1 ribonucleoprotein; PSAT1, phosphoserine aminotransferase; ETFDH, electron-transfer flavoprotein-ubiquinone oxidoreduc-tase; RNA-Seq, RNA-Sequencing; COAD, colon adenocarcinoma.

Figure 8

Survival analysis. (A) PADI2 (P=0.0084) was associated with prognosis in stage II CRC. (B) FCGBP (P=0.0031) was associated with prognosis in stage II CRC. The information was obtained from the Human Protein Atlas. CRC, colorectal cancer; PADI2, protein-arginine deaminase type-2; FCGBP, Fc fragment of IgG binding protein.

Figure 9

Gene Set Enrichment Analysis of potential targets. (A) Biological function of PADI2. (B) Biological function of FCGBP. (C) Biological function of PSAT1. The size of the dot indicates the number of enriched genes. The red dots indicate that the function was facilitated by the potential targets, while dark green dots indicate that the function was inhibited by potential targets. PADI2, protein-arginine deaminase type-2; FCGBP, Fc fragment of IgG binding protein; PSAT1, phosphoserine aminotransferase; KEGG, Kyoto Encyclopedia of Genes and Genomes; NES, normalized enrichment score.