Introduction
Colon cancer is a common malignancy that occurs in the gastrointestinal tract. According to statistics from the National Cancer Institute, as of the first three quarters of 2022, there were about 150,000 cases in the United States, and it caused about 50,000 deaths. Many risk factors lead to the onset of colon cancer: family history, gene mutations, disease history, excessive drinking, smoking, and obesity [
1]. Existing studies found that gene mutation is a major risk factor for colon cancer. For instance, mutations in
K-RAS can stimulate colon cancer, and mutations in
TP53 can promote early colon cancer deterioration and transformation into metastatic colon cancer [
2]. Change in gene expression levels is also a key factor affecting the onset and prognosis of colon cancer. Numerous studies analyzed the gene transcriptional levels of colon cancer patients and found key genes such as
GLUT1,
MUC2,
Cyclooxygenase-2, etc. that are related to the prognosis of colon cancer [
3,
4]. Recent studies on colon cancer-related gene regulatory pathways found that MAPK signaling pathway, JAK/STAT, AKT/NF-κB and other signaling pathways can modulate the occurrence of colon cancer [
5‐
7]. Therefore, we believed that the analysis and mining of colon cancer-related genes would deepen our understanding of colon cancer and improve colon cancer treatment.
Serpin Peptidase Inhibitor 1 (
SERPINE1), also known as plasminogen activator inhibitor-1 (
PAI-1), is an encoding gene of serine protease inhibitors that can suppress fibrinolysis
in vivo [
8].
SERPINE1 is closely correlated with the morbidity of multiple cancers. Studies illustrated that
SERPINE1 can stimulate the progression of triple-negative breast cancer by promoting cytoskeletal reorganization and glycometabolism and can also accelerate the progression of pancreatic cancer by up-regulating the secretion and expression of
IL-8 [
9,
10]. The role of
SERPINE1 in the pathogenesis of colon cancer was studied previously, and a current study found that
SERPINE1 can promote the progression of colon cancer through the P38-MAPK pathway [
11]. Since the expression of
SERPINE1 is closely related to cancer, it is vital to understand the regulatory mechanism of
SERPINE1 expression in vivo. Current research believed that
miR-143 can reduce the expression of
SERPINE1, thereby preventing the progression of bladder cancer [
12]. It was also found that
SERPINE1 interacts with the membrane receptor LDL receptor-related protein 1 to enhance proliferation and migration of esophageal squamous carcinoma cells, and activate AKT and ERK signaling pathways [
13]. Due to its complex regulatory effect in vivo, the study of
SERPINE1 and its regulatory mechanism will shed light on further understanding of pathogenesis and the early screening of colon cancer.
In this study, multiple databases (including Tumor Immune Estimation Resource (TIMER), starBase, UALCAN, and The Human Protein Atlas) were used to find that SERPINE1 was markedly overexpressed in colon cancer, and SERPINE1 expression was correlated with clinical features of colon cancer patients. Survival curve was used to analyze patient survival rate. In this study, survival curve was utilized to analyze the prognostic effect of SERPINE1 on colon cancer, and it was found that patients with high SERPINE1 expression had poor prognoses. LinkedOmics is a multi-omics analysis tool based on The Cancer Genome Atlas (TCGA) database. The highly correlated genes (Integrin Alpha 5 (ITGA5), Matrix Metallopeptidase 19 (MMP19), and ADAM Metallopeptidase with Thrombospondin Type 1 Motif, 4 (ADAMTS4)) of SERPINE1 were obtained by Pearson correlation analysis. Gene set enrichment analysis (GSEA) was used to identify SERPINE1-associated kinases, miRNAs and transcription factors. Gene Multiple Association Network Integration Algorithm (GeneMANIA) is a simple co-expression gene analysis software that can be used to establish gene regulatory networks and analyze the biological functions of genes. In this study, GeneMANIA was utilized to obtain network based on MAPK1, miR-18a and SRF-Q6 and the biological functions of these 3 genes. The results of our research might reveal novel targets and strategies for the diagnosis and treatment of colon cancer, as well as provide a deeper insight into the pathogenesis of colon cancer, which would inform further research.
Discussion
SERPINE1 (
PAI-1) is proven by numerous studies to be a pivotal gene influencing the occurrence and progression of cancer. Current research believes that
SERPINE1 can catalyze the degradation of basement membrane and ECM, making it easier for cancer cells to invade surrounding normal tissue and promote the progression of cancer [
15]. In this study, through the analysis of TIMER database, it was found that
SERPINE1 had obvious differential expression in various cancers, and the most significant differential expression was present in colon cancer. Colon cancer as a common type of cancer attracted a large number of studies on changes in its gene expression. For example, changes in
FGFRL1 gene can promote the occurrence and progression of colon cancer [
16].
WISP can extend the survival time of patients through the WISP2/β-catenin pathway [
17]. Based on the analysis of the relevant data from The Human Protein Atlas and the UALCAN database, this study found that the expression of
SERPINE1 in colon cancer tissue was significantly up-regulated, demonstrating that
SERPINE1 may be a key gene involved in development and progression of colon cancer. This study also explored the expression of
SERPINE1 in patients with various clinical characteristics in the UALCAN database. The findings indicated that the expression of
SERPINE1 was generally higher in patients with various clinical characteristics than that in healthy individuals. By comparing
SERPINE1 with tumor-associated factors, it was revealed that
SERPINE1 level in stage1 of colorectal cancer substantially varied from other 3 stages, and
SERPINE1 was related to TMN stages of CRC [
18]. Furthermore, the differences in survival time of patients with high/low
SERPINE1 expression were also analyzed in combination with clinical characteristics, and the results showed that the survival time of patients with high
SERPINE1 expression was dramatically shorter. Li et al
. [
19] reported that
SERPINE1 is highly expressed in gastric adenocarcinoma and is dramatically implicated in patients’ prognoses. Forced expression of
SERPINE1 (PAI-1) is noticeably related to shorter survival of glioblastoma patients [
20]. Another study indicated that gastric cancer patients with high
SERPINE1 expression have shorter OS than those with low
SERPINE1 expression [
21].
The top 10 genes significantly associated with
SERPINE1 were assessed by GO and KEGG to predict the potential biological functions and signaling pathways involved. GO enrichment analysis reported that changes in
SERPINE1 expression mainly affected cellular components related to ribosomes, cell structure and cell adhesion, receptor ligand activity, and protein tyrosine kinase activity. KEGG enrichment analysis presented that alternations in
SERPINE1 expression mainly affected Rap1\PI3K-Akt and MAPK signaling pathways. Zhang et al
. [
22] reported that protein tyrosine kinase is poorly expressed in breast cancer cells and represses invasion and metastasis of breast cancer cells. Xiong et al
. [
23] found that 6 differentially expressed miRNA target genes in osteosarcoma involved in the PI3K-Akt signaling pathway. Forced expression of
ART3 stimulates proliferation of triple-negative breast cancer cells and modulates triple-negative breast carcinogenesis via activation of Akt and ERK pathways [
24].
Subsequently, it was found through the LinkedOmics database that there were a great number of genes associated with the expression of
SERPINE1, wherein the expression of
ITGA5,
MMP19 and
ADAMTS4 was the most pronouncedly related to the expression of
SERPINE1.
ITGA5 is integrin subunit α5, and there are little published data on proving that relationship between
SERPINE1 and the expression of
ITGA5, while it is reported that overexpression of
ITGA5 can promote the metastasis and spread of colon cancer [
25].
MMP19 (matrix metallopeptidase 19) is related to epithelial cell proliferation, enterocyte migration and poor prognosis [
26,
27].
ADAMTS4 is highly expressed in colon cancer cells, and research suggested that
ADAMTS4 can promote cancer progression through macrophages [
28,
29].
SERPINE1 is notably positively correlated with these three genes, that is,
SERPINE1 and these three genes may jointly played important roles in the progression of colon cancer.
Exploring biological pathways is an important method for cancer research. Studies found that biological pathways such as
MAPK pathway and
NF-κB pathway can affect the pathogenesis of colon cancer [
30,
31]. Our research found that
SERPINE1 might affect the pathogenesis of colon cancer by influencing cell adhesion, kinase activation and other pathways, and thus we further studied the kinases, miRNAs and transcription factors related to
SERPINE1. Kinases play a pivotal role in the pathogenesis of cancer and research suggested that the activation of the tyrosine kinase pathway can stimulate the pathogenesis of colon cancer [
32]. This study found that the
MAPK1 kinase pathway and
SRC kinase were important causes of colon cancer, while
MAPK1 is a common cancer-promoting pathway that is believed to promote the progression of prostate cancer and breast cancer [
33,
34]. In addition,
SRC is also a common cancer-promoting pathway, and it can promote the occurrence of breast cancer by inducing mitochondrial dysfunction [
35]. Besides, miRNAs are also common cancer-related regulatory factors. MiRNAs such as
miR-770‑5p and
miR-200 family are found to be related to the risk of colon cancer [
36,
37]. Here,
miR-18a and
miR-29a were found in the
SERPINE1-miRNA regulatory network.
MiR-18a can affect the onset of colon cancer through the Cdc42/filopodia pathway, and
miR-29a can affect the onset of colon cancer by regulating
b-3p-COL5A1 [
38,
39]. In addition to miRNA, transcription factors are also important regulators that affect pathogenesis. Research suggested that the aberrant expression of transcription factors such as
KLF14 and
SOX2 in colon cancer can affect the occurrence and progression of colon cancer by affecting the expression of downstream genes [
40,
41]. This study found that
SERPINE1 was associated with
SRF-Q6 and
ZIC, which are serum response factor and zinc finger protein, respectively. Research suggested that
SRF-Q6 can affect the occurrence of cancer through the SRF/MRTF pathway while
ZIC can induce colon cancer by influencing the expression of
GLUT142,
43. After conducting the above research, GeneMINIA was used to establish regulatory networks. The results demonstrated that
MAPK1 was mainly involved in immune-related pathways,
miR-18a was mainly involved in serotonin methylation-related pathways, and
SRF-Q6 was mainly involved in actin-related pathways.
In summary, this study dived into SERPINE1 (PAI-1) through bioinformatics methods and explored the genes and pathways related to SERPINE1 expression. SERPINE1 expression was found dramatically related to the expression of ITGA5, MMP19, and ADAMTS4, and affected biological pathways such as RNA transcription, cell adhesion, and kinase activation. This study also identified regulatory networks of SERPINE1 with kinases, miRNAs, and transcription factors through GSEA, such as MAPK1 kinase, miR-18a, and transcription factor SRF-Q6, etc. The contribution of this study is to fully explore the genes and pathways that might interact with SERPINE1. However, this study has the following two limitations. First, this paper is mainly based on bioinformatics methods for analysis, lacking of cell or clinical experiments for verification. Additionally, we conducted bioinformatics analysis directly based on the databases without special treatment, and there may be differences in sample quality, leading to some deviations in the results. We will design relevant experiments in the follow-up study to verify the clinical significance of SERPINE1, and will look for more sample data for optimization to reduce the deviation of results.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.