Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Sep 30;13(9):4574-4592.
doi: 10.21037/tcr-24-191. Epub 2024 Sep 18.

A novel gene signature based on endoplasmic reticulum stress for predicting prognosis in hepatocellular carcinoma

Xuezhi Du #  1 , Yingjie He #  1 , Penggang Dong #  1 , Caigu Yan  1 , Yaqing Wei  1 , Hao Yao  1 , Jinjin Sun  1
Affiliations

A novel gene signature based on endoplasmic reticulum stress for predicting prognosis in hepatocellular carcinoma

Xuezhi Du et al. Transl Cancer Res. .

Abstract

Background: Hepatocellular carcinoma (HCC) remains one of the most common human cancers, the death cases induced by HCC are increasing these years. Endoplasmic reticulum stress (ERS) occurs when misfolded proteins cannot be disposed of properly. It is reported that ERS plays a crucial role in the pathogenesis of human malignant tumors. The aim of this study is to construct a novel gene signature based on ERS for predicting prognosis in HCC.

Methods: The data of HCC patients were downloaded from public databases. The Cox regression analysis and least absolute shrinkage and selection operator (LASSO) regression analysis were performed to construct ERS-related gene signature. The cases were divided into high- and low-risk groups based on the ERS-related gene signature in The Cancer Genome Atlas (TCGA) cohort. Subsequently, the differences in messenger ribonucleic acid (mRNA) expression patterns, immune status, tumor mutation burden (TMB) and copy number variants (CNV) were investigated between high- and low-risk groups. Then, a predictive nomogram according to the ERS-related gene signature and clinicopathological variables was established. At last, we explored the biological functions of TMX1 which had the biggest coefficient and we investigated the effect of BRSK2 on apoptosis in HCC.

Results: In our study, a 9-gene ERS-related gene signature was constructed. The results showed that patients in the low-risk group had a better prognosis than the high-risk group patients. The results of receiver operating characteristic (ROC) curves revealed that the area under the curve (AUC) was 0.784 at 1 year, 0.780 at 2 years, 0.793 at 3 years in the training set. While in validation cohort, this index was 0.694 at 1 year, 0.622 at 2 years, 0.613 at 3 years respectively. The analysis of immune status revealed an immunosuppressive microenvironment in the high-risk group. The analysis of TMB and CNV revealed that the high-risk group patients had a higher genomic mutation frequency. In Univariate Cox regression analysis, the hazard ratio of RiskScore was 2.718 [95% confidence interval (CI): 2.173-3.399]. In Multivariate Cox regression analysis, the hazard ratio of RiskScore was 2.422 (95% CI: 1.805-3.25). Then, we established a nomogram according to the RiskScore and Eastern Cooperative Oncology Group performance status. The AUCs of the nomogram were 0.851 at 1 year, 0.860 at 2 years, and 0.866 at 3 years. At last, we found that TMX1 knockdown can inhibit the proliferation and migration of Huh7 and HepG2 cells. In addition, BRSK2 knockdown could promote the apoptosis induced by ERS.

Conclusions: In our study, a novel ERS-related gene signature was constructed to predict the prognosis of HCC patients. In addition, TMX1 and BRSK2 could promote the progression of HCC. This study may provide a new understanding for HCC.

Keywords: Prognostic gene signature; endoplasmic reticulum stress (ERS); hepatocellular carcinoma (HCC); immune status; nomogram.

PubMed Disclaimer

Conflict of interest statement

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-191/coif). The authors have no conflicts of interest to declare.

Figures

Figure 1
Figure 1
Confirmation of ERS-related genes. The ERS-related genes were identified by Molecular Signature Database. LASSO and multivariate Cox regression were performed for further screening. (A) Cross-validation to select the most suitable parameter in the LASSO regression model in the TCGA cohort. (B) The map of coefficient distribution of the LASSO regression model in the TCGA cohort. (C) The relationship between 9 genes related to ERS and OS was analysed by multivariate Cox regression in the TCGA cohort. ERS, endoplasmic reticulum stress; LASSO, least absolute shrinkage and selection operator; TCGA, The Cancer Genome Atlas; OS, overall survival; HR, hazard ratio; AIC, Akaike information criterion; CI, confidence interval.
Figure 2
Figure 2
Establishment and assessment of the ERS-related gene signature. The Kaplan-Meier survival curves were performed to compare OS between high- and low-risk groups using the “survival” R package in the TCGA and ICGC cohort. The “timeROC” R package was applied to plot ROC curves to assess the accuracy of the gene signature. (A) The distribution and median value of ERS-related RiskScore in the TCGA cohort. The distribution of survival status (B) and the expression status of ERS-related genes (C) in the TCGA cohort. The survival curve (D) and ROC curve (E) of high- and low-risk groups in the TCGA cohort. (F) The distribution and median value of ERS-related RiskScores in the ICGC cohort. The distribution of survival status (G) and the expression status of ERS-related genes (H) in the ICGC cohort. The survival curve (I) and ROC curve (J) of high- and low-risk groups in the ICGC cohort. ERS, endoplasmic reticulum stress; TCGA, The Cancer Genome Atlas; ICGC, International Cancer Genome Consortium; ROC, receiver operating characteristic; HR, hazard ratio; AUC, area under the curve; TPR, true positive rate; FPR, false positive rate; CI, confidence interval; OS, overall survival.
Figure 3
Figure 3
The relationship between RiskScore and clinicopathological parameters. The Wilcox test was performed to explore the relationship between the RiskScore and the clinicopathologic factor. The high RiskScore was significantly related to higher T stage (A), higher clinical stage (B), higher degree of vascular invasion (C) and higher neoplasm histologic grade (D) in the TCGA cohort. TCGA, The Cancer Genome Atlas.
Figure 4
Figure 4
DEGs between high- and low-risk group patients in TCGA cohort and their functional enrichment analysis. The “limma” R package was applied to screen DEGs with the criteria Padj<0.05 and |log2FC| >1. The “clusterProfiler” R package was used to perform GO and KEGG pathway analysis with DEGs. (A) The volcano plot displayed the significantly DEGs with Padj<0.05 and |log2FC| >1 between the high- and low-risk group in the TCGA cohort. The biological process (B), cellular component (C) and molecular function (D) in GO enrichment analysis of DEGs between high- and low-risk group patients in the TCGA cohort. (E) The results of KEGG enrichment analysis of DEGs between different groups in the TCGA cohort. FC, fold change; DEG, differentially expressed gene; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; TCGA, The Cancer Genome Atlas; BP, biological process; CC, cellular component; MF, molecular function.
Figure 5
Figure 5
The difference between immune landscape of two different groups in the TCGA cohort. The difference of ESTIMATE score (A), proportional differences of immune cells assessed by CIBERSORT algorithm (B) and MCP-counter algorithm (D) between high- and low-risk groups. The difference of 13 immune-related functions (C) and the relationship between RiskScore and part of immune cells estimated by ssGSEA (E-L) in high- and low-risk groups. *, P<0.05; **, P<0.01; ***, P<0.001; ns, no significance. ssGSEA, single sample gene set enrichment analysis; TCGA, The Cancer Genome Atlas; MCP, microenvironment cell population; APC, antigen-presenting cell; CCR, CC chemokine receptor; HLA, human leukocyte antigen; IFN, interferon.
Figure 6
Figure 6
Analysis of mutations and CNV between high- and low-risk groups in the TCGA cohort. The maf format file of TMB was downloaded by the “maftools” R package. Masked copy number segment of CNV was obtained from TCGA GDC portal (https://portal.gdc.cancer.gov/), and was analyzed by GISTIC2 with default parameters for arm-level and focal CNVs. (A,B) The distinct somatic mutations in high- and low-risk groups. (C,D) A distinct CNV spectrum identified by RiskScore related to ERS in high- and low-risk groups. ERS, endoplasmic reticulum stress; TMB, tumor mutation burden; TCGA, The Cancer Genome Atlas; CNV, copy number variation; GDC, Genomic Data Comments.
Figure 7
Figure 7
Construction of a predictive nomogram based on RiskScore. The R package “survival” was applied to perform Multivariate Cox regression of parameters with P<0.05. Time-dependent ROC curves and calibration plots were applied to evaluate the precision of the nomogram with the “survivalROC” and “rms” R packages. Construction of a predictive nomogram based on RiskScore in the TCGA cohort and (A) univariate and (B) multivariate Cox regression analyses between OS, RiskScore and clinicopathological parameters in the TCGA derivation cohort. (C) Nomogram based on RiskScore and ECOG performance status. (D-F) Calibration curves for 1-, 2- and 3 years for the nomogram. (G) The ROC curve to predict 1-, 2- and 3 years OS of the nomogram in TCGA cohort. ROC, receiver operating characteristic; TCGA, The Cancer Genome Atlas; OS, overall survival; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; CI, confidence interval; TPR, true positive rate; FPR, false positive rate; AUC, area under the curve; AJCC, American Joint Committee on Cancer.
Figure 8
Figure 8
The validation of biological functions of TMX1 in HCC. (A) The validation of knockdown efficiency of TMX1 in Huh7 and HepG2 cell lines. (B) The influence of TMX1 knockdown on proliferation of Huh7 and HepG2 cell lines detected by CCK8. (C,D) The influence of TMX1 knockdown on proliferation of Huh7 and HepG2 cell lines detected by EDU assay. The proliferating nucleus were stained by EDU solution and all nucleus were stained by DAPI solution. Finally, the data was collected by fluorescence microscopy. (E,F) The results of wound healing assay of Huh7 and HepG2 cell lines after TMX1 knockdown. For wound healing assay, 3×105 cells were put into per well of 6-well plates, then the siRNA of TMX1 were transfected into cells. After 24 h, the cells were scratched, and the images of 0, 12, 24 h were collected by microscope. (G) The results of transwell assay of Huh7 and HepG2 cell lines after TMX1 knockdown. The cells were stained by crystal violet and the data were cedullected by microscope. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. CCK8, cell counting kit 8; EDU, 5-ethynyl-2'-deoxyuridine; DAPI, 4',6-diamidino-2-phenylindole; NC, normal control; siRNA, small interfering RNA; HCC, hepatocellular carcinoma.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. 10.3322/caac.21660 - DOI - PubMed
    1. Craig AJ, von Felden J, Garcia-Lezana T, et al. Tumour evolution in hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 2020;17:139-52. 10.1038/s41575-019-0229-4 - DOI - PubMed
    1. Yang JD, Hainaut P, Gores GJ, et al. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol 2019;16:589-604. 10.1038/s41575-019-0186-y - DOI - PMC - PubMed
    1. Schwarz DS, Blower MD. The endoplasmic reticulum: structure, function and response to cellular signaling. Cell Mol Life Sci 2016;73:79-94. 10.1007/s00018-015-2052-6 - DOI - PMC - PubMed
    1. Schubert U, Antón LC, Gibbs J, et al. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature 2000;404:770-4. 10.1038/35008096 - DOI - PubMed

LinkOut - more resources