A novel fatty acid metabolism-related signature identifies MUC4 as a novel therapy target for esophageal squamous cell carcinoma

Abstract

Fatty acid metabolism has been identified as an emerging hallmark of cancer, which was closely associated with cancer prognosis. Whether fatty acid metabolism-related genes (FMGs) signature play a more crucial role in biological behavior of esophageal squamous cell carcinoma (ESCC) prognosis remains unknown. Thus, we aimed to identify a reliable FMGs signature for assisting treatment decisions and prognosis evaluation of ESCC. In the present study, we conducted consensus clustering analysis on 259 publicly available ESCC samples. The clinical information was downloaded from The Cancer Genome Atlas (TCGA, 80 ESCC samples) and Gene Expression Omnibus (GEO) database (GSE53625, 179 ESCC samples). A consensus clustering arithmetic was used to determine the FMGs molecular subtypes, and survival outcomes and immune features were evaluated among the different subtypes. Kaplan-Meier analysis and the receiver operating characteristic (ROC) was applied to evaluate the reliability of the risk model in training cohort, validation cohort and all cohorts. A nomogram to predict patients' 1-year, 3-year and 5-year survival rate was also studied. Finally, CCK-8 assay, wound healing assay, and transwell assay were implemented to evaluate the inherent mechanisms of FMGs for tumorigenesis in ESCC. Two subtypes were identified by consensus clustering, of which cluster 2 is preferentially associated with poor prognosis, lower immune cell infiltration. A fatty acid (FA) metabolism-related risk model containing eight genes (FZD10, TACSTD2, MUC4, PDLIM1, PRSS12, BAALC, DNAJA2 and ALOX12B) was established. High-risk group patients displayed worse survival, higher stromal, immune and ESTIMATE scores than in the low-risk group. Moreover, a nomogram revealed good predictive ability of clinical outcomes in ESCC patients. The results of qRT-PCR analysis revealed that the MUC4 and BAALC had high expression level, and FZD10, PDLIM1, TACSTD2, ALOX12B had low expression level in ESCC cells. In vitro, silencing MUC4 remarkably inhibited ESCC cell proliferation, invasion and migration. Our study fills the gap of FMGs signature in predicting the prognosis of ESCC patients. These findings revealed that cluster subtypes and risk model of FMGs had effects on survival prediction, and were expected to be the potential promising targets for ESCC.

Keywords: Esophageal squamous cell carcinoma; Fatty acid metabolism; Immune microenvironment; Prognosis.

Figure 1

The FMGs interaction network of in ESCC patients. FMGs fatty acid metabolism-related genes, ESCC esophageal squamous cell carcinoma.

Figure 2

ESCC subtypes based on consensus clustering. (A, B) The ESCC patients were divided into cluster 1 and cluster 2 based on the prognostic FMGs. (C) The Kaplan–Meier analysis of the patients among two clusters. (D) A heatmap regarding to the relationships between the clinicopathological features and two clusters. ESCC esophageal squamous cell carcinoma, FMGs fatty acid metabolism-related genes.

Figure 3

Functional enrichment analysis in two ESCC subtypes. (A) GO enrichment analysis of FMGs. (B) KEGG enrichment analysis of FMGs. ESCC esophageal squamous cell carcinoma, FMGs fatty acid metabolism-related genes.

Figure 4

Analysis of the correlation between the ESCC subtypes and immune infiltration, immune functions and immune-related score. (A) Comparison of the ssGSEA scores among the two clusters. (B) The variation in immune functions. (C–E) The results of the correlation analysis between the immune-related score and subtypes. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 5

Construction of the prognostic risk model. The distribution plots of risk scores of ESCC patients in the (A) training, (B) validation, and (C) all cohorts. (D–F) Heatmap showing the expression profiles of the eight FMGs of ESCC patients in training, (B) validation, and (C) all cohorts. ESCC esophageal squamous cell carcinoma, FMGs fatty acid metabolism-related genes.

Figure 6

The Kaplan–Meier survival curves and ROC curves of risk score based on eight FMGs signature. Differences in the OS and predictive ability of risk model for ESCC patients between the high-risk and low-risk groups in the (A) training, (B) test, and (C) all cohorts. ROC receiver operator characteristic, FMGs fatty acid metabolism-related genes, OS overall survival, ESCC esophageal squamous cell carcinoma.

Figure 7

Independent prognostic validation of OS nomogram for ESCC patients. (A, B) Univariate and Multivariate analysis for all cohorts. (C, D) Nomogram and nomogram to predict 1-, 3-, and 5-year OS rates of ESCC patients. (E) ROC curves for clinical characteristics and nomogram. OS overall survival, ESCC esophageal squamous cell carcinoma, ROC receiver operator characteristic.

Figure 8

GSEA of the functional characteristics in the high-risk and low-risk groups. (A, B) GO function annotation among two risk groups. (C, D) The significantly enriched KEGG (https://www.kegg.jp/kegg/kegg1.html) pathways of different risk groups. GSEA gene set enrichment analysis, GO Gene Ontology, KEGG Kyoto Encyclopedia of Genes and Genomes, ESCC esophageal squamous cell carcinoma.

Figure 9

The correlation between the risk model and immune activity in ESCC patients. (A) The expression levels of immune cell infiltration in high-risk and low-risk groups. (B–D) Analysis of the association between the risk model and immune-related score. (E–L) Kaplan–Meier curves of OS in ESCC patients based on immune cells. ESCC esophageal squamous cell carcinoma; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 10

The correlation between the risk model and immune checkpoints in ESCC patients. (A) Heatmap of immune checkpoints with different risk score. (B–H) Correlation between the risk model and immune checkpoints. ESCC esophageal squamous cell carcinoma; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 11

The expression levels of eight signature FMGs in ESCC cells. qRT-PCR analysis results of the expression levels of ALOX12B (A), BAALC (B), DNAJA2 (C), FZD10 (D), MUC4 (E), PDLIM1 (F), PRSS12 (G) and TACSTD2 (H) in HEEC, ECA109 and TE1. qRT-PCR quantitative real-time polymerase chain reaction, FMGs fatty acid metabolism-related genes, ESCC esophageal squamous cell carcinoma, ns not significant; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 12

The functional roles of MUC4 for ECA109 and TE1 cells. (A) The transfection with si-MUC4 verified by qRT-PCR. (B, C) CCK-8 assay showed that knockdown of MUC4 resulted in growth retardation of ECA109 and TE1 cells. (D, E) Cell migration was evaluated with microscope (magnification, × 40). (F, G) The Transwell assay was used to detect the ability of cell invasion (magnification, × 100). *p < 0.05, **p < 0.01, and ***p < 0.001.