Glycosyltransferase-related long non-coding RNA signature predicts the prognosis of colon adenocarcinoma

Abstract

Purpose: Colon adenocarcinoma (COAD) is the most common type of colorectal cancer (CRC) and is associated with poor prognosis. Emerging evidence has demonstrated that glycosylation by long noncoding RNAs (lncRNAs) was associated with COAD progression. To date, however, the prognostic values of glycosyltransferase (GT)-related lncRNAs in COAD are still largely unknown.

Methods: We obtained the expression matrix of mRNAs and lncRNAs in COAD from The Cancer Genome Atlas (TCGA) database. Then, the univariate Cox regression analysis was conducted to identify 33 prognostic GT-related lncRNAs. Subsequently, LASSO and multivariate Cox regression analysis were performed, and 7 of 33 GT-related lncRNAs were selected to conduct a risk model. Gene set enrichment analysis (GSEA) was used to analyze gene signaling pathway enrichment of the risk model. ImmuCellAI, an online tool for estimating the abundance of immune cells, and correlation analysis were used to explore the tumor-infiltrating immune cells in COAD. Finally, the expression levels of seven lncRNAs were detected in colorectal cancer cell lines by reverse transcription-quantitative polymerase chain reaction (RT-qPCR).

Results: A total of 1,140 GT-related lncRNAs were identified, and 7 COAD-specific GT-related lncRNAs (LINC02381, MIR210HG, AC009237.14, AC105219.1, ZEB1-AS1, AC002310.1, and AC020558.2) were selected to conduct a risk model. Patients were divided into high- and low-risk groups based on the median of risk score. The prognosis of the high-risk group was worse than that of the low-risk group, indicating the good reliability and specificity of our risk model. Additionally, a nomogram based on the risk score and clinical traits was built to help clinical decisions. GSEA showed that the risk model was significantly enriched in metabolism-related pathways. Immune infiltration analysis revealed that five types of immune cells were significantly different between groups, and two types of immune cells were negatively correlated with the risk score. Besides, we found that the expression levels of these seven lncRNAs in tumor cells were significantly higher than those in normal cells, which verified the feasibility of the risk model.

Conclusion: The efficient risk model based on seven GT-related lncRNAs has prognostic potential for COAD, which may be novel biomarkers and therapeutic targets for COAD patients.

Keywords: colorectal cancer; glycosyltransferase; lncRNAs; overall survival; risk model.

Figure 1

Landscape of genetic and expression variation of glycosyltransferase genes in COAD. (A, B) Volcano and heatmap visually showed differentially expressed genes between normal and COAD. (C, D) GO and KEGG analysis of DEGs. (E) Protein–protein interaction (PPI) network showing the interaction between DEGs among glycosyltransferases. (F) The SNV mutation frequency of DEGs in the COAD cohort. (G) The CNV variation frequency of DEGs in COAD. The height of the column represented the alteration frequency. N, normal tissues; T, tumor; BP, biological process; CC, cellular component; MF, molecular function; FC, fold change.

Figure 2

Construction of a prognostic GT-related lncRNA risk model. (A) Screening of optimal parameters (lambda) in the LASSO regression model. (B) LASSO coefficient profiles of 11 candidate GT-related lncRNAs. (C) Forest map of seven GT-related lncRNAs significantly correlated with outcome and identified by multivariate cox regression. (D) Kaplan–Meier curve for the OS of COAD patients in the high- and low-risk group. (E) Distribution of survival status and risk score in the patient cohort. (F) ROC curves at 1, 3, and 5 years in COAD patients.

Figure 3

Kaplan–Meier survival curve of the selected GT-related lncRNAs and co-expression network between lncRNA and mRNA. (A–G) The overall survival curve of LINC02381, AC002310.1, ZEB1AS1, AC020558.2, AC105219.1, MIR210HG, and AC009237.14 of COAD patients in the high- and low-risk groups. (H) Co-expression network of GT genes and prognostic lncRNAs. Red nodes represent GT-related lncRNAs, while blue nodes represent GT genes. (I) The relationships among GT genes, GT-related lncRNAs, and risk type in the Sankey diagram.

Figure 4

Correlation between risk model and clinicopathological factors. (A–D) Distribution of stage I/II and III/IV, T1/2 and T3/4, M0 and M1, and N0 and N1/2 tumors between high- and low-risk group. (E–H) Expression levels of seven prognostic GT-related lncRNAs in T, M, N, and S stage groups. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 5

Independent prognostic role of risk model signature. (A, B) Univariate and multivariate Cox regression analyses of risk scores and clinical characteristics. (C–E) Determination of the area under the ROC curve (AUC) of the risk score and clinical features based on the ROC curve. (F–H) The Kaplan–Meier curve shows the prognostic value of the risk model for COAD patients categorized by age, stage, and T stage.

Figure 6

Construction and calibration of the nomogram. (A) Construction of a nomogram based on age, stage, and risk score as independent prognostic factors. (B) Time-dependent ROC analysis based on the nomogram and clinical characteristics. (C–E) The calibration plot for internal validation of the nomogram within 1, 3, and 5 years, respectively. AUC, the area under the ROC curve.

Figure 7

Gene set enrichment analysis (GSEA) and the correlation between risk model and tumor-infiltrating immune cells. (A) Enrichment analysis of signaling pathways in the risk model. (B) The expression levels of immune-related genes between the high- and low-risk groups. (C) The response of immune checkpoint blockade (ICB) therapy between the high- and low-risk group. R, response; NR, no response. (D) The fraction of tumor immune infiltrating cells in the high- and low-risk group. (E) Correlation of risk score with five tumor-infiltrating immune cell subtypes. *p < 0.05, **p < 0.01, ***p < 0.001. Th1, T helper 1; NKT, natural killer T.

Figure 8

The expression level of LINC02381 (A), MIR210HG (B), AC009237.14 (C), AC105219.1 (D), ZEB1-AS1 (E), AC002310.1 (F), AC020558.2 (G) between normal cell line NCM460 and HCT116, DLD1, and HT-29 CRC cell lines.