Identification of WWTR1 as an immune infiltration-correlated prognostic biomarker in colon cancer

Jian-Shun Ge ,Jie Zhang ,Rui Wang ,Hai-Ning Li ,Xin-Nong Liu* ,Xu-Dong Yin ,Yan-Bing Ding,Guo-Tao Lu,Lin-Lin Tian

1Institute of Digestive Diseases,The Affiliated Hospital of Yangzhou University,Yangzhou 225000,China.2Department of General Surgery,The Affiliated Hospital of Yangzhou University,Yangzhou 225000,China.3Department of Oncology,The Affiliated Hospital of Yangzhou University,Yangzhou 225000,China.4Department of Gastroenterology,The Affiliated Hospital of Yangzhou University,Yangzhou 225000,China.5Department of Oncology,Dalian Medical University,Dalian 116044,China.

Abstract WWTR1,a gene related to the TGF-β signaling pathway,has been elucidated to be involved in oncogenesis in multiple studies.There is,however,no research on its link to immune infiltration in colon cancer.The TCGA database has identified WWTR1,a gene related to the TGF-β signaling pathway,which is lowly expressed in colon cancer patients compared to normal subjects.Meanwhile,we produced the Kapan-Meier curve with GEO and the TCGA database,which revealed that colon cancer patients with high WWTR1 expression had a poor prognosis.We discovered that high expression of WWTR1 in colon cancer was associated with clinical stage,pathological T-stage,and lymphatic metastasis after examining the clinical characteristics of colon cancer patients.WWTR1 was found to be an independent predictive factor for colon cancer in a multivariate Cox regression study.Infiltration of immunological cells (B cells,CD8+T cells,CD4+T cells,Macrophage,Neutrophil,Dendritic cells) was linked to WWTR1 expression.In colon cancer,WWTR1 expression was also found to be favorably linked with major immune cell markers.According to an analysis of WWTR1 DCGs,GO,and KEGG enrichment analysis,WWTR1 expression levels were associated with ameboidal-type cell migration,focal adhesion,actin binding,Chemical carcinogenesis-reactive oxygen species,Non-alcoholic fatty liver disease,and Alzheimer disease.These findings imply that WWTR1 is a prognostically valuable and important biomarker for colon cancer,and imply that its expression is strongly linked to colon cancer immune infiltration,making it a potential new target for colon cancer biotherapy.

Keywords:WWTR1;colon cancer;The Cancer Genome Atlas;Transforming growth factor-β signaling;immune infiltration;prognosis

Introduction

One of the most prevalent malignant tumors in the gastrointestinal tract is colon cancer.Colon cancer has surpassed gastric cancer to become the first malignant tumor in the gastrointestinal tract,according to the most recent research report [1].Although there has been a recent increase in the overall incidence of early-onset colon cancer,the trends in the incidence of early-onset colon cancer vary widely among various populations [2].Surgical resection,chemotherapy,and targeted medication therapy are currently the most common treatments for colon cancer.All major treatment modalities have limitations,and surgical resection is usually the main treatment with adjuvant chemotherapy or targeted therapy [3].Immunotherapy,an emerging treatment strategy for colon cancer,can be used with other treatment modalities,according to reports,to improve colon cancer patients’ outcomes [4,5].As a result,the most critical difficulty we have today is finding useful indicators and possible immunotherapeutic targets.Many factors,such as metastasis,T stage,and Transforming growth factor (TGF)-signaling pathways,have been crucial in determining cancer prognosis in research [6–8].The TGF-β signaling pathway is an important signaling pathway for cancer cell growth and differentiation.Inhibiting tumor function in the early stages,such as cell cycle arrest and apoptosis,and promoting carcinogenesis in the late stages,such as metastasis and chemoresistance,are possible outcomes of this pathway[9–12].

We examined the expression of WWTR1 in colon cancer and its connection with clinical characteristics using the The Cancer Genome Atlas (TCGA) database in this work.Following that,we verified WWTR1’s predictive relevance in colon cancer and identified it as a gene linked to immune infiltration in colon cancer via TGF signaling.

Methods

Data and information

RNA-seq expression data and clinical information of 473 colon cancer tissues and 41 normal colon tissues were acquired from TCGA(https://portal.gdc.cancer.gov/) [13].The identification of differentially expressed genes with done with |logFC| >0.9.Then a total of 54 genes involved in TGF-β signaling were retrieved by the“HALLMARK_TGF_BETA_SIGNALING”gene set in the Gene set enrichment analysis (GSEA)(http://www.gsea-msigdb.org/gsea/msigdb/cards/HALLMARK_TGF_BETA_SIGNALING.html) [14].After survival analysis of the screened TGF-β differentially expressed signaling genes,we obtained 8 genes,including WWTR1,through the“VennDiagram”R software for further study.

Expression levels and prognostic value of WWTR1 in public datasets

The expression profiles and clinicopathological information,involving 473 colon cancer tissues and 41 normal tissues,were obtained from TCGA to validate the expression of WWTR1 in colon cancer.We examined the expression of WWTR1 in the“Expression DIY”module of the Gene Expression Profiling Interactive Analysis(GEPIA)database(http://gepia.cancer-pku.cn/index.html) [15].Additionally,other expression data,involved the“GSE12945”,“GSE40967”,“GSE17537”databases,on WWTR1 and clinicopathologic features of colon cancer patients,were screened from the Gene Expression Omnibus (GEO)database (https://www.ncbi.nlm.nih.gov/geo/) [16].Moreover,the protein levels of WWTR1 in colon cancer were assessed by The Human Protein Atlas (HPA) (https://www.proteinatlas.org/) [17].Then the association between WWTR1 expression and clinical features was assessed using the“ggpubr”package and Perl language.Next,based on the information on WWTR1 expression and overall survival data of colon cancer patients obtained from the TCGA,GSE12945,GSE40967,GSE17537,Kaplan-Meier (K-M) curves were plotted through Long-term Outcome and Gene Expression Profiling Database of pan-cancers (LOGpc) (http://bioinfo.henu.edu.cn/ DatabaseList.jsp).Immunohistochemical figures were obtained from the HPA database.Finally,we performed the univariate and multivariate Cox regression analyses to further investigate the prognostic value of WWTR1 in colon cancer.

Immune infiltration analysis of WWTR1

The Tumor Immune Estimation Resource (TIMER)(https://cistrome.shinyapps.io/timer/),a comprehensive website,was used to analyze the correlation between gene expression and the degree of immune cell infiltration with a wide range of cancers in TCGA [18].We evaluated WWTR1 expression in colon cancer and its correlation with various conventional immune cells.Then,we carried out a series of correlation analyses to uncover the effect of WWTR1 on markers of immune cells.GEPIA is an online database containing RNA-sequencing data on 275 colon cancer samples and 349 normal colon samples from the TCGA and Genotype-Tissue Expression(GTEx)projects.Immunocytic differences were further investigated using the CIBERSORT database for infiltration of colon cancer patients,and the results were plotted using the“ggplot2”package in R software for barplot corHeatmap.Based on the expression level of WWTR1,the 473 samples obtained from the TCGA-COAD cohort were categorized into two groups:high expression and low expression.The CIBERSORT database was used to investigate the infiltration between the two groups,and the results were plotted using the“ggpubr”package in R software.

Identification of differentially co-expressed genes (DCGs) of WWTR1 in colon cancer

19,828 genes are acquired from LinkedOmics database(http://www.Linkedomics.org/),and 12,598 genes were screened based on the criteria of correlation coefficient>0,FDR<0.05.Then,we used the R language“limma”program to filter the DCGS with|logFC| >2,and the first 50 genes,positively and negatively associated with WWTR1,were plotted respectively by the“heatmap”R package.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) enrichment analyses

We used DCGS to perform GO and KEGG by“clusterprofiler”R software and correctedP-value <0.05 were found to be statistically relevant.

Statistical analysis

R software and the Perl language (https://www.perl.org/) were used to perform all statistical analyses.Expression data were normalized by log2 transformation.The survival analysis was completed by log-rank test.The correlation of WWTR1 with several biomarkers of immune infiltration was calculated using Spearman’s correlation.The relevance of infiltrating immune cells was determined by the following guide for the value of partial cor:0.00–0.19:“very weak”,0.20–0.39:“weak”,0.40–0.59:“moderate”,0.60–0.79:“strong”,and 0.80–1.0:“strong”,0.40–0.59:“moderate”,0.60–0.79:“strong”,and 0.80–1.0:“very strong”.

Results

WWTR1 is an oncogene that has been discovered in many malignancies.Following a comprehensive examination of numerous databases,it was discovered that the expression of WWTR1 is closely linked to the expression of immune infiltration and immune cell markers in colon cancer.As a result,it could be a potential promising biomarker for colon cancer immunotherapy.

Filtration of differentially expressed genes in colon cancer

We obtained differentially expressed genes |LOG >0.9| from the TCGA database (Figure 1a).After identification transformation,the list of 14,198 colon cancer genes was decreased to 3,359 genes with official gene symbols.8 intersecting genes were identified(intersecting genes between DEGs in the TCGA and TGF-β signaling).(Figure 1b).Then we used heatmaps to show these 8 differentially expressed genes(Figure 1c).We discovered that WWTR1 had not been published in bioinformatics in colon cancer,so we decided to investigate it more.

Expression levels of WWTR1 in colon cancer

To verify whether WWTR1 expression affects patients with colon cancer,we analyzed the expression data of TCGA.The boxplot(Figure 2a) reveals that the level of WWTR1 expression was lower in colon cancer tissues compared to normal tissues (P<0.05).Moreover,GEPIA is commonly used to examine the expression of WWTR1 in colon cancer versus healthy individuals (Figure 2b).WWTR1 expression in tumor tissues is lower than in normal individuals,according to the findings.The HPA database was used to assess the expression of WWTR1 protein in colon cancer tissues to better understand how it is expressed.The results showed that,when compared to normal tissues with antibody CAB068248,colon cancer tissues displayed lower amounts of WWTR1 protein (Figure 2c).In addition,LOGpc was used to examine survival data for WWTR1 expression levels in colon cancer (Figure 2d–h).As shown in the figure,(TCGA,GSE12945,GSE40967,GSE17537),the expression level of WWTR1 was negatively correlated with the overall survival(OS) of colon cancer patients.(TCGA:HR=1.55,95%CI=1.0222 -2.3503,P=0.0391;GSE12945:HR=5.8066,95% CI=1.055 -31.959,P=0.0432;GSE40967:HR=1.422,95% CI=1.0604 -1.9068,P=0.0187;GSE17537:HR=2.471,95% CI=1.0223 -5.9725,P=0.0445).

Relationship between WWTR1 and prognosis of colon cancer

We evaluated the association between WWTR1 expression levels and different clinicopathological features of colon cancer to better understand the relationship between WWTR1 and colon cancer prognosis.WWTR1 expression was higher in relatively advanced tumor tissues in the clinical stage (Figure 3a).Similarly,WWTR1 expression was positively correlated with pathological T-stage (Figure 3b) and lymphatic metastasis (Figure 3c).Then,on diverse clinical parameters of colon cancer,we ran univariate Cox regression analysis and multifactor Cox regression analysis to confirm the link between WWTR1 and prognosis in colon cancer.Age(P=0.012,HR=3.613,95% CI=1.319 -9.897),clinical analysis (P<0.001,HR=7.767,95%CI=3.809-15.837),T stage(P<0.001,HR=16.540,95%CI=4.116 -66.466),lymphatic metastasis,(P<0.001,HR=2.325,95%CI=1.644-3.288),distant metastasis(P<0.001,HR=4.461,95% CI=2.738 -7.270),and WWTR1 expression (P=0.035,HR=1.319,95% CI=1.020 -1.705) were all found to be effective predictors for patients with colon cancer in a univariate Cox regression analysis (Figure 3d).WWTR1 expression level was also found to be an independent predictive factor in patients with colon cancer in a multifactorial Cox regression analysis (Figure 3e).In conclusion,WWTR1 expression was highly correlated with the prognosis of colon cancer patients with the above clinicopathological features.

Relationship between WWTR1 expression and immune infiltration in colon cancer patients

Immune infiltration has been linked to the development and progression of most tumors,according to literature.In addition,a relationship has been discovered in the case of colon cancer.The association between six immune infiltrates and WWTR1 expression levels was explored to identify a link between immune infiltration and colon cancer.From the images we know that (image):WWTR1 with B-cell infiltration (partial.cor=0.145,P=3.47e -03),Crohn’s disease (CD) 8+T cells (partial.cor=0.301,P=5.71e -10),CD4+T cells (partial.cor=0.531,P=1.10e -30),Macrophage(partial.cor=0.617,P=1.13e-43),Neutrophil(partial.cor=0.571,P=4.65e-36),and Dendritic cell (partial.cor=0.572,P=2.75e -36) were correlated.Also,it is noteworthy that WWTR1 expression was negatively correlated with tumor purity (partial.cor=-0.348,P=5.39e -13) (Figure 4a).We revealed that WWTR1 expression was closely associated with immune infiltration in colon cancer based on these findings.

In order to obtain a better understanding of the interaction between WWTR1 and immune infiltration,the TIMER database was used to validate the association between WWTR1 and immune marker.We identified a positive correlation between WWTR1 and the majority of immune cell markers after adjusting for purity (Table 1).CD19 and CD79A on B cells,HLA-DRA,ITGAX,NRP1,CD1C,HLA-DPB,and HLA-DQB1 on Dendritic cells,NOS2,IRF5 and PTGS2 on M1,CD163,MS4A4 and VSIG4 on M2,Monocyte CFS1R and CD86,CCR7,CEACAM8 and ITGAM on Neutrophils,CTLA4,GZMB,HAVCR2,LAG3 and PDCD1 on T cells (exhaustion),CD2,CD3D and CD3E on T cells(general),CD68 on tumor-associated macrophages (TAM),CCL2 and IL10 expression were observed to be tightly linked to WWTR1 expression in colon cancer (Figure 4b–j).In addition,the GEPIA database was used to confirm the associations between immune cell markers and WWTR1 expression levels in colon cancer.The results of GEPIA confirmed the accuracy of the previous data (Table 2).

Table 1 Correlation analysis between WWTR1 and relative gene biomarkers of immune cells in TIMER

Table 1 Correlation analysis between WWTR1 and relative gene biomarkers of immune cells in TIMER (continued)

Table 2 Correlation analysis between CDC20 and relative gene biomarkers of immune cells in GEPIA

Table 2 Correlation analysis between CDC20 and relative gene biomarkers of immune cells in GEPIA (continued)

The CIBERSORT database was then used to investigate the immune infiltration of patients in the TCGA-COAD cohort.Our findings revealed that the proportion of tumor-infiltrating immune cells (TICs)in patients varies significantly,with some having a higher proportion of Macrophages and others having a higher proportion of Neutrophils than others(Figure 5a).We analyzed the relationship between TICs in this dataset and noticed that many immune cells have significant positive correlations,such as Neutrophils and Mast cells activated,T cells CD4 memory resting and Plasma cells,T cells CD4 memory activated and T cells CD8,and Mast cells resting and Dendritic cells resting.Macrophages M0 and Plasma cells,T cells CD4 memory resting,T cells CD8,Dendritic cells resting,and Mast cells resting all had a strong negative connection (Figure 5b).The specific linkage needs to be investigated further.Finally,we used the CIBERSORT database to investigate the differences between the WWTR1 high and low expression groups in the tumor microenvironment.There were substantial differences between the two groups in B cells naive,B cells memory,T cells CD4 memory resting,T cells CD4 memory activated,NK cells activated,Monocytes,Macrophages M0,Macrophages M1,Macrophages M2,Dendritic cells activated,and Neutrophils (Figure 5c).In conclusion,the expression of WWTR1 showed a significant correlation with immune infiltration.The intrinsic mechanism needs further study and discussion.

Figure 1 Procedures to identify WWTR1.a,A volcano map of differentially DGCs in TCGA(logFC>0.9);b,Venn diagram of the intersection of 3,551 DEGs in TCGA with 46 official gene symbols in the TGF-β signaling pathway;c,Heatmap of the screened genes;DGCs,differentially co-expressed genes;TCGA,The Cancer Genome Atlas;TGF-β,Transforming growth factor-β.

Figure 2 The expression level and survival analysis of WWTR1 in colon cancer.a,The expression level of WWTR1 in colon cancer from TCGA cohort;b,The expression level of WWTR1 in colon cancer from GEPIA database;c,The protein expression of WWTR1 in colon cancer from HPA database(1,normal.2,tumor);d–g,Survival analysis of OS(n=438,n=29,n=579,n=55)in colon cancer;TCGA,The Cancer Genome Atlas;OS,overall survival.

Figure 3 The prognostic value of WWTR1 in colon cancer. Expression of WWTR1 correlated significantly with clinical stage (a),pathologic T stage (b),and lymphatic metastasis (c).Univariate COX analysis (d) and multivariate COX analysis (e) revealed the correlation between WWTR1 and clinicopathological features in colon cancer patients.

Figure 4 The relationship between WWTR1 expression levels and immune filtrates in colon cancer. a,WWTR1 expression level is significantly positive correlated with immune cells.Correlation analysis between WWTR1 expression and immunological markers in colon cancer:b,B cell;c,Monocyte;d,Dendritic cells;e,T cell(exhaustion);f,M1;g,M2;h,Neutrophils;I,T cell(general);j,TAM.

Figure 5 The proportion and correlation analysis of TICs in colon cancer and the correlation between TICs and WWTR1 expression. a,Percentage of 22 types of immune cells in each colon cancer sample;b,Correlation analysis between immune cells;c,High and low expression of WWTR1 and differences of immune cells;TIC,tumor-infiltrating immune cells.

Functional enrichment analysis of WWTR1 DCGs in colon cancerTo explore the mechanisms underlying the role of WWTR1 in colon cancer,19,828 DCGs were identified through the LinkedOmics database,which contains 11,579 positively associated genes and 8,249 negatively associated genes(Figure 6a).The heat map shows 50 DCGs positively and negatively correlated with WWTR1(Figure 6b–c).These associated genes were selected for GO and KEGG enrichment analysis and used to explore the important biological pathways and functions between them.Ameboidal-type cell migration,camera-type eye development,eye development,cell-substrate adhesion,visual system development,and sensory system development are among the pathways,in which,GO enrichment analysis identifies the main involvement of DCGs in biological processes,as shown in the following images.The most enriched categories in the cellular component enrichment analysis were focal adhesion and cell-substrate junction.At the molecular functional level,acting binding was thought to be the most enriched category (Figure 6d).DCGs regulates chemical carcinogenesis,reactive oxygen species,nonalcoholic fatty liver disease,diabetic cardiomyopathy,Alzheimer disease,oxidative phosphorylation,thermogenesis,Parkinson disease,Prion disease,Huntington disease,Amyotrophic lateral sclerosis,cardiac muscle contraction,and neurodegeneration pathways,according to KEGG enrichment analysis (Figure 6e).The strong association of WWTR1 with non-alcoholic fatty liver disease has been demonstrated in the present study by KEEG enrichment analysis [19–21].

Figure 6 Functional enrichment analysis of differentially co-expressed genes of WWTR1 in colon cancer. a,Volcano plots of differentially coexpressed genes of WWTR1.The heatmap visualized the top 50 positively (b) and negatively (c) differentially co-expressed genes.GO terms of biological process,cell component,and molecular function;d,KEGG terms associate with differentially co-expressed genes of WWTR1;GO,Gene Ontology;KEGG,Kyoto Encyclopedia of Genes and Genomes.

Discussion

Colorectal cancer has overtaken stomach cancer as the most prevalent gastrointestinal tumor.According to the World Health Organization’s International Agency for Research on Cancer (IARC) global statistics for 2018,1.85 million cases of colorectal cancer are diagnosed worldwide,ranking third in the spectrum of malignant tumor incidence,and 880,000 people die from colorectal cancer,ranking second in the spectrum of death [22,23].The most common treatment for colon cancer is surgery.However,surgical complications are a significant determinant in the patient’s prognosis [24].The surgical risk factors include,but are not limited to,preoperative albumin levels,surgical anastomosis technique,and suturing technique[25].Furthermore,adjuvant therapy plays a vital role in the treatment of colon cancer.Such as adjuvant therapies,chemotherapy,targeted therapy,and immunotherapy together with surgery form a combined treatment for colon cancer [26].Targeted therapy and immunotherapy are now emerging as novel treatment modalities for colon cancer.TGF-β signaling pathway is a class of cytokines with multiple biological activities,which regulate cell proliferation,differentiation,growth,and other life activities [27].On one hand,in pre-malignant cells,TGF-β signaling mainly acts as a tumor suppressor,and on the other hand,in advanced malignant cells,TGF-β signaling promotes tumor invasion and metastasis[28,29].It has been recently documented that TGF-β signaling has a significant role in regulating cancer proliferation,migration,and differentiation in a range of cancers,particularly breast,liver,gastric,and bladder cancers[30–33].TGF-beta signaling is strongly associated with cancer immunotherapy,and relevant studies have reported that TGF-β could act as a tumor promoter through several mechanisms,including immunosuppression,and it also modifies its function via regulation by TICs [34,35].Therefore,defining the connection between TGF-β signaling and immune infiltration has significant and original implications for the immunotherapy of various cancers.WWTR1 is a Hippo pathway transcriptional cofactor that regulates the TGF-β signaling pathway through a mechanism that is independent of Smad3,p38,and MRTF-mediated,but not MRTF-translocated [36].Hence,it is essential to investigate the prognostic value of WWTR1 and assess immunotherapy’s efficacy in colon cancer.

In this study,we demonstrated the prognostic value of WWTR1 in colon cancer through data from TCGA,GEO,and TIMER databases.The expression of WWTR1 was much lower in colon cancer tissues compared to normal tissues.In addition,patients with colon cancer with higher WWTR1 expression had lower OS.We found that age,clinical stage,T stage,N stage,M stage,and WWTR1 were of considerable value in predicting the prognosis of colon cancer in univariate Cox regression analysis.WWTR1 was shown to be an independent prognostic factor in the colon cancer cohort based on multivariate Cox regression analysis.

The colon is an organ containing a wide range of immune cells,anda number of factors regulate its immune microenvironment.CD,Ulcerative colitis,and various other chronic inflammatory diseases of the colon are considered to be major drivers of colon cancer development [37].It has been reported that infiltration of relevant immune cells such as CD8+T is tightly associated with cancer development [38].We found an intimate association of WWTR1 with immune cell infiltration,especially in Dendritic cells,M2,Monocyte,Neutrophils,T cell exhaustion,and TAM cells.Most of the immune cells in the immune microenvironment are used to be activated to fight cancer cells,but not in all of the immune cells [39].Immune evasion of tumors is an essential feature of cancer evolution [40–42].By inhibiting immunological checkpoints and promoting immune evasion,tumor cells could exploit associated immune cells to boost cancer cell growth and destruction [43–45].Consequently,immunotherapy is an essential and well-tolerated treatment option for patients with advanced colon cancer.In our research,we observed that the expression level of WWTR1 in colon cancer was positively correlated with markers of T cell exhaustion,such as PD-1,CTLA4,LAG3,TIM-3.These markers proteins are essential components of suppressive immunological checkpoints,which help tumor cells evade immune monitoring.This suggests that WWTR1 is critical in the T cell exhaustion process.Furthermore,most immune cell markers were shown to be strongly linked to WWTR1,showing that WWTR1 is involved in the regulation and recruitment of immune infiltrating cells in colon cancer.We used GO and KEGG analysis to learn more about WWTR1’s role in colon cancer.Meanwhile,Chemical carcinogenesis reactive oxygen species is identified as a critical route by KEGG analysis.This is the first study to use bioinformatics to examine the role of WWTR1 in the immunological microenvironment of colon cancer in relation to immune infiltration function.Despite this,our work has limitations.We evaluated the function of WWTR1 and the predictive level with related colon cancer patients using various databases,but more clinical trials are needed to confirm our findings.

Conclusion

In conclusion,high levels of WWTR1 expression are associated to a poor prognosis in colon cancer.Further research revealed a relationship between WWTR1 expression and immune infiltration in colon cancer,indicating WWTR1 could be a potential target for colon cancer biotherapy.