Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2019 Oct 14;235(3):2037–2048. doi: 10.1002/jcp.29067

Identification of hub genes and construction of transcriptional regulatory network for the progression of colon adenocarcinoma hub genes and TF regulatory network of colon adenocarcinoma

Shuxun Wei 1,†, Jinshui Chen 1,†, Yu Huang 1, Qiang Sun 1, Haolu Wang 2, Xiaowen Liang 2, Zhiqian Hu 1,✉, Xinxing Li 1,✉
PMCID: PMC6916361  PMID: 31612481

Abstract

The aim of this study was to identify key genes related to the progression of colon adenocarcinoma (COAD), and to investigate the regulatory network of hub genes and transcription factors (TFs). Dataset GSE20916 including 44 normal colon, 55 adenoma, and 36 adenocarcinoma tissue samples was used to construct co‐expression networks via weighted gene co‐expression network. Gene Ontology annotation and the Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis for the objective module were performed using the online Database for Annotation, Visualization and Integrated Discovery. Hub genes were identified by taking the intersection of differentially expressed genes between dataset GSE20916 and GSE39582 and validated using The Cancer Genome Atlas (TCGA) database. The correlations between microRNA (miRNA) and hub genes were analyzed using the online website StarBase. Cytoscape was used to establish a regulatory network of TF‐miRNA‐target gene. We found that the orange module was a key module related to the tumor progression in COAD. In datasets GSE20916 and GSE39582, a total of eight genes (BGN, SULF1, COL1A1, FAP, THBS2, CTHRC1, COL5A2, and COL1A2) were selected, which were closely related with patients’ survivals in TCGA database and dataset GSE20916. COAD patients with higher expressions of each hub gene had a worse prognosis than those with lower expressions. A regulatory network of TF‐miRNA‐target gene with 144 TFs, 26 miRNAs, and 7 hub genes was established, including model KLF11‐miR149‐BGN, TCEAL6‐miR29B2‐COL1A1, and TCEAL6‐miR29B2‐COL1A2. In conclusion, during the progression of COAD, eight core genes (BGN, SULF1, COL1A1, FAP, THBS2, CTHRC1, COL5A2, and COL1A2) play vital roles. Regulatory networks of TF‐miRNA‐target gene can help to understand the disease progression and optimize treatment strategy.

Keywords: colon adenocarcinoma, hub genes, Gene Ontology and Kyoto Encyclopedia of Genes and Genomes, transcription factors regulatory network, weighted gene co‐expression network

Hub genes and transcription factor regulatory network of colon adenocarcinoma

graphic file with name JCP-235-2037-g009.jpg

Abbreviations

DEGs

the differentially expressed genes

GO

Gene Ontology

KEGG

Kyoto Encyclopedia of Genes and Genomes

TCGA

The Cancer Genome Atlas

WGCNA

weighted gene co‐expression network analysis

1. INTRODUCTION

Colon cancer is one of the most common gastrointestinal malignancies worldwide (Arnold et al., 2017). Every year, the incidence of colorectal cancer increases by 4.2% globally (Chen et al., 2016; Maley et al., 2017). Currently, most patients with colorectal cancer cannot be diagnosed at early stages. Thus, early diagnosis and intervention are important to improve the prognosis of colorectal cancer. Colon adenocarcinoma (COAD) is the most common type of colorectal cancer (Mutch, 2007). The progression of COAD is through a multi‐stage process, from normal mucosa to adenoma, and finally to carcinoma (Arnold et al., 2017; Mutch, 2007). With the progression of COAD, gene mutations such as TP53, Ras, and Braf, and abnormal gene expressions including TROP2, ABCG2, and HIF‐1, have been reported to play crucial roles (Hall, Morris, & Sun, 2018; Recio‐Boiles & Cagir, 2019; Shawki, Ashburn, Signs, & Huang, 2018). However, the progression of COAD involving complex gene‐mechanisms has not yet been clearly explored.

Now many bioinformatics methods have been developed to identify potential prognostic biomarkers and therapeutic targets (Wan, Tang, Han, & Wang, 2018). For example, weighted gene co‐expression network analysis (WGCNA) can be used for finding clusters (or modules) of highly correlated genes (Langfelder & Horvath, 2008), for summarizing such clusters using the module eigengene or an intramodular hub gene, for relating modules to one another and to external sample traits, and for calculating module membership (MM) measures (Wan et al., 2018). WGCNA has been used to identify gene modules related to clinical outcomes in many cancers including colon cancer for evaluating recurrence (Ding, Li, Wang, Chi, & Liu, 2019). However, there is no study to systematically identify the regulatory network of transcription factor (TF)‐microRNA (miRNA)‐target gene related to the progression of COAD using WGCNA. miRNAs are small noncoding regulatory RNAs, which can bind to the target messenger RNAs (mRNAs) and repress the target mRNA expression. After regulated by TFs, miRNAs greatly affect the progression of COAD.

In this study, a co‐expression network of differentially expressed genes (DEGs) in COAD was constructed and the most significant modules in the network were used to reveal the tumorigenic genes. Then hub genes were identified by taking the intersection of DEGs between datasets GSE20916 and GSE39582 and validated using The Cancer Genome Atlas (TCGA) database. Finally, a regulatory network of TF‐miRNA‐target gene was established for the progression of COAD.

2. MATERIALS AND METHODS

2.1. Data preprocessing

The gene expression profiling datasets GSE20916 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE20916) and GSE39582 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi) were downloaded from the Gene Expression Omnibus (GEO) database (Barrett et al., 2012). The GSE20916 dataset contains 44 normal, 55 adenoma, 36 adenocarcinoma, and 10 carcinoma tissue samples. The GSE39582 dataset contains 19 normal and 566 adenocarcinoma tissue samples. All samples are hybridized by Affymetrix Human Genome U133 Plus 2.0 Array. The matrix of the dataset was normalized by the limma package. The inclusion criteria of the GSE20916 included samples with normal, adenoma, and adenocarcinoma tissue types. The exclusion criteria of GSE39582 included the samples without disease‐free survival or the disease‐free survival time is 0.

2.2. Construction of WGCNA and identifying preserved modules

The GSE20916 matrix that contained gene expression (including normal, adenoma, and COAD samples) was ranked by the standard deviation (SD), and top 5,000 genes (Table S1) were selected for the next WGCNA analysis. Then the ideal soft‐thresholding power was chosen and genes cluster into modules was based on the Topological Overlap Matrix. The correlation between module eigengene and clinical characteristics was investigated by Pearson's correlation tests. p < .05 and |r| > .7 were considered as the selection criteria of candidate modules. Module preservation was calculated by the module preservation function to detect the stability and repeatability of modules in GSE39582 (Table S2). The module with maximum preservation statistics Zsummary and minimum preservation statistics medianRank was selected as the key module.

2.3. Function enrichment analysis

To investigate the potential biological function and pathways in the key module, the Online Database for Annotation, Visualization and Integrated Discovery (DAVID; https://david-d.ncifcrf.gov/) was used to analyze and visualize Gene Ontology (GO) terms and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (Table S3).

2.4. Screening hub genes

MM shows the correlation between genes and modules, and the top 15 genes with the highest correlation with the key module were selected as Group 1 of candidate genes. DEGs between normal colon tissue and adenocarcinoma in the GSE39582 dataset were determined by limma package based on the standard of adjusted p < .01 and | log FC | > 2. Then the disease free survival (DFS) of DEGs was calculated by survival package and genes with p < .05 were defined as Group 2 of candidate genes (Table S4). Hub genes were identified by the intersection of the two candidate gene groups.

2.5. Validating for hub genes

Gene Expression Profiling Interactive Analysis (GEPIA; http://gepia.cancer-pku.cn/) was used to analyze the RNA sequencing data from TCGA. TCGA data of COAD were used to validate the expression of identified hub genes. Human Protein Atlas (http://www.proteinatlas.org) was used to validate the candidate hub genes by immunohistochemistry.

2.6. Construction of TF regulatory network

StarBase (http://starbase.sysu.edu.cn/index.php) was used to predict miRNAs that bind to hub genes based on the standard that CLIP Data > = 3 and expression was present in at least one tumor sample. Then miRNAs with most intersections in 7 databases were selected (Fig. S3A‐G). The plugin iRegulon of Cytoscape is used to predict TF regulation networks.

3. RESULTS

3.1. Data preprocessing

A total of 145 samples in GSE20916 and 585 samples in GSE39582 were downloaded from GEO dataset and were normalized by the limma package. Then 135 samples from GSE20916 and 536 samples from GSE39582 were selected for the next step. Fig S1 shows the flow chart of the study design. Our selected clinical characteristics included histological results (normal colon epithelial tissue, adenoma, and adenocarcinoma), gender (male and female), and age.

3.2. Construction of WGCNA and identifying preserved modules

The sample dendrogram was plotted using the WGCNA package (Figure 1a). The appropriate soft‐thresholding power was chosen with the ideal soft‐thresholding power of 10 (Figure 1b). Twenty‐six modules were identified by the Hierarchical Clustering and the Dynamic branch Cutting based on the minimum module size 30. Then 26 modules merge in 13 modules based on the similarity between module eigengenes above 0.8 (Figure 2). Further, the correlation between modules eigengenes and histological results (normal colon epithelial tissue, adenoma, and adenocarcinoma) was analyzed. There were four modules positively associated with histological results, and six modules negatively associated (Figure 3a; p < .01). Among these modules, brown, green, and orange modules had the highest correlations with histological results (Figure 3a; all |r| > .7), indicating that they play important roles in the tumorigenesis of COAD. Expression of 5,000 genes with the largest difference of SD in dataset GSE39582 was used to calculate the preservation of modules. Figure 4a,b shows the medianRank statistics and Zsummary statistics of each module. Among the brown, green, and orange modules with preservation, the orange module with the maximum statistics Zsummary and minimum statistics medianRank was defined as the key module.

Figure 1.

Figure 1

Open in a new tab

(a) The cluster was based on the expression data of GSE20916, which contained 44 normal tissue samples, 55 adenoma tissue samples, and 36 adenocarcinoma tissue samples. The top 5,000 genes with the highest SD values were used for the analysis by WGCNA. The color intensity was proportional to disease status (normal, adenoma, and adenocarcinoma), sex (male and female) and age (years old). (b) Analysis of the scale‐free fit index for various soft‐thresholding power (β). Analysis of the mean connectivity for various soft‐thresholding power. In all, 10 was the most fit power value. SD, standard deviation

Figure 2.

Figure 2

Open in a new tab

(a) The cluster dendrogram of module eigengenes. (b) The cluster dendrogram of genes in GSE20916. Each branch in the figure represented one gene, and every color below represented one co‐expression module

Figure 3.

Figure 3

Open in a new tab

(a) Heat‐map of the correlation between module eigengenes and the disease status of colon adenocarcinoma. The orange module was most positively correlated with status. (b) Bubble chart of GO and KEGG results of orange module. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes

Figure 4.

Figure 4

Open in a new tab

Module preservation was evaluated by medianRank and Zsummary statistics which correlated to connectivity and density of networks. The module with lower medianRank tends to exhibit stronger observed preservation than the module with a higher medianRank if both of them are preserved. Compared with the GSE20916 dataset, orange module with highest Zsummary (a)and lowest medianRank statistics (b) than brown and green modules

3.3. Enrichment analysis of the orange module

The online database of DAVID was used to explore the enrichment analysis of GO and KEGG in key orange module. Figure 3b shows the enrichment results of GO and KEGG in the orange module. The KEGG of orange module was mainly in the extracellular matrix receptor (ECM)–receptor interaction, focal adhesion, cytokine–receptor interaction, hematopoietic cell lineage, and transforming growth factor (TGF)‐β signaling pathway, while the biological process based on GO of orange module mainly included response to wounding, inflammatory response, cell adhesion, biological adhesion, and vasculature development. Figure 3b show the contents of cellular component and molecular function based on the GO of orange module.

3.4. Identification and validation of hub genes

A total of eight genes (BGN, SULF1, COL1A1, FAP, THBS2, CTHRC1, COL5A2, and COL1A2) shared by both Groups 1 and 2 of candidate genes were identified as hub genes (Figure 5a). We used TCGA data of COAD to validate the hub gene expression with the online tool of GEPIA. All of the hub genes are expressed differently in normal and cancer tissues of COAD by the criterion of |logFC| > 1 and p < .01 (Figure 5b). All of the hub genes significantly related to disease‐free survival of COAD (Figure 6) and GSE39582 (Figure S2). Compared to normal tissues, the protein expressions of these hub genes in tumor tissues were significantly higher based on the Human Protein Atlas database (Figure 7).

Figure 5.

Figure 5

Open in a new tab

(a) Venn plot of candidate hub genes commonly owned in GS20916 and GSE39582. (b) The transcriptional differences of hub gene levels between colon carcinoma tissues and the para‐cancer tissues in TCGA. TCGA, The Cancer Genome Atlas (*p < .001)

Figure 6.

Figure 6

Open in a new tab

The disease‐free survival analysis of hub gene levels in colon carcinoma patients in TCGA. TCGA, The Cancer Genome Atlas

Figure 7.

Figure 7

Open in a new tab

The translational differences of hub gene levels between colon carcinoma tissues and the paracancer tissues in the Human Protein Atlas database

3.5. Construction of TF regulatory network

A total of 42 miRNAs could bind to hub genes predicted by starBase. One hub gene FAP does not have any binding miRNA. The related TF and the TFs regulatory network are shown in Figure 8. The regulatory network of TF‐miRNA‐target gene was established, involving 144 TFs, 26 miRNAs, and 7 hub genes, such as model KLF11‐miR149‐BGN, TCEAL6‐miR29B2‐COL1A1, TCEAL6‐miR29B2‐COL1A2, TCEAL6‐miR29B2‐COL5A2, TEAD1‐miR30A‐CTHRC1, CREB3‐miR199A1‐SULF1, and GRHPR‐miR182‐THBS2.

Figure 8.

Figure 8

Open in a new tab

The transcriptional regulatory network of hub genes, miRNAs, and TFs. miRNAs, microRNAs; TFs, transcription factors

4. DISCUSSION

High‐throughput biological technologies, such as the systems biology algorithm of WGCNA, allow researchers to evaluate the relationships between genes and clinical features through a scale‐free network (Langfelder & Horvath, 2008; Wan et al., 2018). WGCNA has been widely used in disease, physiology, drug, genome annotation, and cancer progression (Chen et al., 2018). WGCNA has been used to identify gene modules and hub genes associated with TNM stages in gastric cancer, hepatocellular carcinoma (Zhu, Sun, Zhou, He, & Qian, 2018), and metastatic colorectal cancer (Wang et al., 2014). In this study, datasets GSE20916 and GSE39582 were utilized to construct co‐expression networks by WGCNA to identify the key orange module related to the progression of COAD.

Herein, we found eight hub genes for the progression of COAD, including BGN, SULF1, COL1A1, FAP, THBS2, CTHRC1, COL5A2, and COL1A2. These genes may contribute to the exploration of the progression of COAD and could be regarded as biomarkers and therapeutic targets of COAD. Most of these genes have been proved to relate to the tumorigenesis and metastasis of cancer. BGN is an extracellular matrix protein which plays vital roles in the regulation of cell morphology, growth, migration, and differentiation. The abnormal expressions of BGN in ovarian cancer and osteosarcoma were closely associated with chemoresistance. In HCT116 colon cancer cell line, BGN knockdown inhibited the proliferation and invasion (Xing, Gu, & Ma, 2015). Furthermore, BGN has been reported to promote the chemotherapy resistance via activating nuclear factor kappa‐light‐chain‐enhancer of activated B cells signal transduction in colon cancer (Liu, Xu, Xu, Cui, & Xing, 2018). The SULF1 gene encoded the heparin‐degrading endo‐sulfatase to inhibit the activation of cell growth factor. The expression of SULF1 is stable in normal tissues but downregulated in cancer cells. Moreover, cancer cell proliferation and migration could be inhibited by re‐expression of SULF1 (Lai, Sandhu, Shire, & Roberts, 2008). The activation of the TGFβ/SMAD transcriptional pathway underlays a novel tumor‐promoting role of SULF1 in hepatocellular carcinoma (Dhanasekaran et al., 2015). COL1A1 encodes the pro‐α 1 chains of type I collagen. COL1A1 was reported to be overexpressed in many cancers (Shi et al., 2019). COL1A1 could promote colorectal cancer metastasis by regulating the WNT/PCP pathway (Zhang, Wang, Zhang, Zhong, & Yang, 2018). THBS2 is a thrombospondin family protein and negatively regulates angiogenesis in cancer. Increasing the THBS2 expression in cancer inhibited tumor growth (Sun et al., 2014). THBS2 expression in colorectal cancer was associated with reduced angiogenesis and distant metastasis (Wang et al., 2016). Most recently, miR‐135b was reported to promote invasion and metastasis through THBS2 suppression (Nezu et al., 2016). Higher THBS2 expression in colorectal cancer tissue was found compared with normal tissue (Wang et al., 2016). Furthermore, RBP4 and THBS2 have been reported to serve as biomarkers for the diagnosis of colorectal cancer (Fei et al., 2017).

In addition, CTHRC1 is an extracellular matrix glycoprotein and regulates the Wnt‐PCP pathway in developmental morphogenesis (Ma et al., 2014). Recently, CTHRC1 was found to be highly expressed in many human cancers, promoting invasion and metastasis (Tan et al., 2013). CTHRC1 has been reported to promote proliferation and invasiveness of human colorectal cancer cells by activating Wnt/PCP signaling (Yang et al., 2015). COL5A2, also known as collagen type V alpha 2 chain, involves in the progression of colorectal cancer, breast cancer, and osteosarcoma. COL5A2 was reported to be correlated with poor clinical outcomes of bladder cancer patients, suggesting that it could serve as a cancer biomarker (Zeng, Liu, Liu, & Wang, 2018). COL5A2 expression was also found in colorectal cancer but not in normal colon, showing that COL5A2 was implicated in the carcinogenesis of colorectal cancer (Fischer, Stenling, Rubio, & Lindblom, 2001). Type I collagen consists of one α2 chain (COL1A2) and two α1 chains (COL1A1). COL1A2 was downregulated in many cancers, including melanoma, bladder cancer, and head and neck cancer. However, in pancreatic and ovarian cancer, COL1A2 was upregulated. The different expressions of COL1A2 cause distinct collagen‐mediated effects in different human malignancies (Ao, Guan, Wang, & Wang, 2018; Wu et al., 2019). Upregulation of COL1A2 was reported in blood and tumor tissues of patients with colorectal cancer (Rodia et al., 2018). Inhibition of COL1A2 expression in colorectal cancer cell could suppress cell proliferation and invasion (Yu et al., 2018).

In our study, Cytoscape and StarBase were used to construct a regulation network of TF‐miRNA‐mRNA. Although miRNAs like miR‐155‐5p, miR‐199a‐5p and miR‐206 were clearly associated with the development and progression of colon cancer, miRNAs associated with colon cancer are still not thoroughly explored (Parasramka et al., 2012; Qu et al., 2015; Sakurai et al., 2011). We established the first regulatory network of TF‐miRNA‐target gene for the progression of COAD, involving 144 TFs, 26 miRNAs, and 7 hub genes, such as model KLF11‐miR149‐BGN, TCEAL6‐miR29B2‐COL1A1, and TCEAL6‐miR29B2‐COL1A2.

In summary, we found a total of eight hub genes (BGN, SULF1, COL1A1, FAP, THBS2, CTHRC1, COL5A2, and COL1A2) closely related to the survival of patients with COAD. These genes and factors may contribute to the exploration of the progression of COAD and could be regarded as biomarkers and therapeutic targets of COAD. A regulatory network of TF‐miRNA‐target gene was established, involving 144 TFs, 26 miRNAs, and 7 hub genes, which could help to understand disease progression and optimize treatment strategy.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS

Z. H. and X. L. contributed to the conception or design of the work. S. W., Y. H., and J. C. contributed to acquisition, analysis, and interpretation of data for the work. Q. S., H. W., and X. L. contributed to manuscript preparation, revision, and review.

Supporting information

Supporting information

Click here for additional data file. (398.1KB, pptx)

Supporting information

Click here for additional data file. (8.6MB, xlsx)

Supporting information

Click here for additional data file. (22.4MB, xls)

Supporting information

Click here for additional data file. (71.1KB, xlsx)

Supporting information

Click here for additional data file. (47KB, xls)

ACKNOWLEDGMENTS

We would like to thank Dr. Chris Lou (Blogger, WeChat Official Accounts) for the bioinformatics technology support. This study was funded by the National Youth Science Foundation (8180032537).

Wei S, Chen J, Huang Y, et al. Identification of hub genes and construction of transcriptional regulatory network for the progression of colon adenocarcinoma hub genes and TF regulatory network of colon adenocarcinoma. J Cell Physiol. 2019;235:2037–2048. 10.1002/jcp.29067

Contributor Information

Zhiqian Hu, Email: huzhiq163@163.com.

Xinxing Li, Email: xingxin123456@sina.com.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available in Gene Expression Omnibus (GEO) datasets at https://www.ncbi.nlm.nih.gov/gds/.

REFERENCES

  1. Ao, R. , Guan, L. , Wang, Y. , & Wang, J. N. (2018). Silencing of COL1A2, COL6A3, and THBS2 inhibits gastric cancer cell proliferation, migration, and invasion while promoting apoptosis through the PI3k‐Akt signaling pathway. Journal of Cellular Biochemistry, 119(6), 4420–4434. [DOI] [PubMed] [Google Scholar]
  2. Arnold, M. , Sierra, M. S. , Laversanne, M. , Soerjomataram, I. , Jemal, A. , & Bray, F. (2017). Global patterns and trends in colorectal cancer incidence and mortality. Gut, 66(4), 683–691. [DOI] [PubMed] [Google Scholar]
  3. Barrett, T. , Wilhite, S. E. , Ledoux, P. , Evangelista, C. , Kim, I. F. , Tomashevsky, M. , … Soboleva, A. (2012). NCBI GEO: Archive for functional genomics data sets—update. Nucleic Acids Research, 41(D1), D991–D995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chen, P. F. , Wang, F. , Nie, J. Y. , Feng, J. R. , Liu, J. , Zhou, R. , … Zhao, Q. (2018). Co‐expression network analysis identified CDH11 in association with progression and prognosis in gastric cancer. OncoTargets and Therapy, 11, 6425–6436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chen, W. , Zheng, R. , Baade, P. D. , Zhang, S. , Zeng, H. , Bray, F. , … He, J. (2016). Cancer statistics in China, 2015. CA: A Cancer Journal for Clinicians, 66(2), 115–132. [DOI] [PubMed] [Google Scholar]
  6. Dhanasekaran, R. , Nakamura, I. , Hu, C. , Chen, G. , Oseini, A. M. , Seven, E. S. , … Roberts, L. R. (2015). Activation of the transforming growth factor‐beta/SMAD transcriptional pathway underlies a novel tumor‐promoting role of sulfatase 1 in hepatocellular carcinoma. Hepatology, 61(4), 1269–1283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Ding, M. , Li, F. , Wang, B. , Chi, G. , & Liu, H. (2019). A comprehensive analysis of WGCNA and serum metabolomics manifests the lung cancer‐associated disordered glucose metabolism. Journal of Cellular Biochemistry, 120, 10855–10863. [DOI] [PubMed] [Google Scholar]
  8. Fei, W. , Chen, L. , Chen, J. , Shi, Q. , Zhang, L. , Liu, S. , … Hu, X. (2017). RBP4 and THBS2 are serum biomarkers for diagnosis of colorectal cancer. Oncotarget, 8(54), 92254–92264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fischer, H. , Stenling, R. , Rubio, C. , & Lindblom, A. (2001). Colorectal carcinogenesis is associated with stromal expression of COL11A1 and COL5A2. Carcinogenesis, 22(6), 875–878. [DOI] [PubMed] [Google Scholar]
  10. Hall, M. J. , Morris, A. M. , & Sun, W. (2018). Precision medicine versus population medicine in colon cancer: From prospects of prevention, adjuvant chemotherapy, and surveillance. American Society of Clinical Oncology Educational Book, 38, 220–230. [DOI] [PubMed] [Google Scholar]
  11. Lai, J. P. , Sandhu, D. S. , Shire, A. M. , & Roberts, L. R. (2008). The tumor suppressor function of human sulfatase 1 (SULF1) in carcinogenesis. Journal of Gastrointestinal Cancer, 39(1‐4), 149–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Langfelder, P. , & Horvath, S. (2008). WGCNA: An R package for weighted correlation network analysis. BMC Bioinformatics, 9, 559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Liu, B. , Xu, T. , Xu, X. , Cui, Y. , & Xing, X. (2018). Biglycan promotes the chemotherapy resistance of colon cancer by activating NF‐kappaB signal transduction. Molecular and Cellular Biochemistry, 449(1‐2), 285–294. [DOI] [PubMed] [Google Scholar]
  14. Ma, M. Z. , Zhuang, C. , Yang, X. M. , Zhang, Z. Z. , Ma, H. , Zhang, W. M. , … Zhang, Z. G. (2014). CTHRC1 acts as a prognostic factor and promotes invasiveness of gastrointestinal stromal tumors by activating Wnt/PCP‐Rho signaling. Neoplasia, 16(3), 265–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Maley, C. C. , Aktipis, A. , Graham, T. A. , Sottoriva, A. , Boddy, A. M. , Janiszewska, M. , … Shibata, D. (2017). Classifying the evolutionary and ecological features of neoplasms. Nature Reviews Cancer, 17(10), 605–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Mutch, M. G. (2007). Molecular profiling and risk stratification of adenocarcinoma of the colon. Journal of Surgical Oncology, 96(8), 693–703. [DOI] [PubMed] [Google Scholar]
  17. Nezu, Y. , Hagiwara, K. , Yamamoto, Y. , Fujiwara, T. , Matsuo, K. , Yoshida, A. , … Ochiya, T. (2016). miR‐135b, a key regulator of malignancy, is linked to poor prognosis in human myxoid liposarcoma. Oncogene, 35(48), 6177–6188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Parasramka, M. A. , Dashwood, W. M. , Wang, R. , Saeed, H. H. , Williams, D. E. , Ho, E. , & Dashwood, R. H. (2012). A role for low‐abundance miRNAs in colon cancer: The miR‐206/Kruppel‐like factor 4 (KLF4) axis. Clinical Epigenetics, 4(1), 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Qu, Y. , Wang, H. , Sun, Z. , Tang, Y. , Han, X. , Yu, X. , & Liu, K. (2015). Up‐regulated miR‐155‐5p promotes cell proliferation, invasion and metastasis in colorectal carcinoma. International Journal of Clinical and Experimental Pathology, 8(6), 6988. [PMC free article] [PubMed] [Google Scholar]
  20. Recio‐Boiles, A. , & Cagir, B. (2019). Cancer, Colon. Treasure Island (FL). [Google Scholar]
  21. Rodia, M. T. , Solmi, R. , Pasini, F. , Nardi, E. , Mattei, G. , Ugolini, G. , … Lauriola, M. (2018). LGALS4, CEACAM6, TSPAN8, and COL1A2: Blood markers for colorectal cancer‐validation in a cohort of subjects with positive fecal immunochemical test result. Clinical Colorectal Cancer, 17(2), e217–e228. [DOI] [PubMed] [Google Scholar]
  22. Sakurai, K. , Furukawa, C. , Haraguchi, T. , Inada, K. , Shiogama, K. , Tagawa, T. , … Iba, H. (2011). MicroRNAs miR‐199a‐5p and ‐3p target the Brm subunit of SWI/SNF to generate a double‐negative feedback loop in a variety of human cancers. Cancer Research, 71(5), 1680–1689. [DOI] [PubMed] [Google Scholar]
  23. Shawki, S. , Ashburn, J. , Signs, S. A. , & Huang, E. (2018). Colon Cancer. Surgical Oncology Clinics of North America, 27(2), 269–287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Shi, Y. , Duan, Z. , Zhang, X. , Zhang, X. , Wang, G. , & Li, F. (2019). Down‐regulation of the let‐7i facilitates gastric cancer invasion and metastasis by targeting COL1A1. Protein & Cell, 10(2), 143–148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Sun, R. , Wu, J. , Chen, Y. , Lu, M. , Zhang, S. , Lu, D. , & Li, Y. (2014). Down regulation of Thrombospondin2 predicts poor prognosis in patients with gastric cancer. Molecular Cancer, 13, 225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Tan, F. , Liu, F. , Liu, H. , Hu, Y. , Liu, D. , & Li, G. (2013). CTHRC1 is associated with peritoneal carcinomatosis in colorectal cancer: A new predictor for prognosis. Medical Oncology, 30(1), 473. [DOI] [PubMed] [Google Scholar]
  27. Wan, Q. , Tang, J. , Han, Y. , & Wang, D. (2018). Co‐expression modules construction by WGCNA and identify potential prognostic markers of uveal melanoma. Experimental Eye Research, 166, 13–20. [DOI] [PubMed] [Google Scholar]
  28. Wang, D. D. , Xu, Y. , Tu, Y. L. , Tan, X. L. , Zhu, Z. M. , Han, M. M. , … Cai, S. W. (2014). Comparison analysis in synchronous and metachronous metastatic colorectal cancer based on microarray expression profile. Hepato‐gastroenterology, 61(136), 2215–2218. [PubMed] [Google Scholar]
  29. Wang, X. , Zhang, L. , Li, H. , Sun, W. , Zhang, H. , & Lai, M. (2016). THBS2 is a potential prognostic biomarker in colorectal cancer. Scientific Reports, 6, 33366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wu, J. , Liu, J. , Wei, X. , Yu, Q. , Niu, X. , Tang, S. , & Song, L. (2019). A feature‐based analysis identifies COL1A2 as a regulator in pancreatic cancer. Journal of Enzyme Inhibition and Medicinal Chemistry, 34(1), 420–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Xing, X. , Gu, X. , & Ma, T. (2015). Knockdown of biglycan expression by RNA interference inhibits the proliferation and invasion of, and induces apoptosis in, the HCT116 colon cancer cell line. Molecular Medicine Reports, 12(5), 7538–7544. [DOI] [PubMed] [Google Scholar]
  32. Yang, X. M. , You, H. Y. , Li, Q. , Ma, H. , Wang, Y. H. , Zhang, Y. L. , … Li, J. (2015). CTHRC1 promotes human colorectal cancer cell proliferation and invasiveness by activating Wnt/PCP signaling. International Journal of Clinical and Experimental Pathology, 8(10), 12793–12801. [PMC free article] [PubMed] [Google Scholar]
  33. Yu, Y. , Liu, D. , Liu, Z. , Li, S. , Ge, Y. , Sun, W. , & Liu, B. (2018). The inhibitory effects of COL1A2 on colorectal cancer cell proliferation, migration, and invasion. Journal of Cancer, 9(16), 2953–2962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Zeng, X. T. , Liu, X. P. , Liu, T. Z. , & Wang, X. H. (2018). The clinical significance of COL5A2 in patients with bladder cancer: A retrospective analysis of bladder cancer gene expression data. Medicine, 97(10), e91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Zhang, Z. , Wang, Y. , Zhang, J. , Zhong, J. , & Yang, R. (2018). COL1A1 promotes metastasis in colorectal cancer by regulating the WNT/PCP pathway. Molecular Medicine Reports, 17(4), 5037–5042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Zhu, Q. , Sun, Y. , Zhou, Q. , He, Q. , & Qian, H. (2018). Identification of key genes and pathways by bioinformatics analysis with TCGA RNA sequencing data in hepatocellular carcinoma. Molecular and Clinical Oncology, 9(6), 597–606. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting information

Click here for additional data file. (398.1KB, pptx)

Supporting information

Click here for additional data file. (8.6MB, xlsx)

Supporting information

Click here for additional data file. (22.4MB, xls)

Supporting information

Click here for additional data file. (71.1KB, xlsx)

Supporting information

Click here for additional data file. (47KB, xls)

Data Availability Statement

The data that support the findings of this study are available in Gene Expression Omnibus (GEO) datasets at https://www.ncbi.nlm.nih.gov/gds/.

Articles from Journal of Cellular Physiology are provided here courtesy of Wiley

RESOURCES