Fig 1.
Flowchart of the analysis in the current study.
LUAD gene expression data obtained from TCGA dataset were evaluated via DEG analysis to assess the differential mRNA expression of sphingolipid metabolism-related genes between NSTs and LUAD tissues. GO and KEGG enrichment analyses, as well as PPI network construction were performed to examine the biological functions correlated with the mRNAs and the interactions between the genes. Survival analysis and clinical correlation analysis were performed for pathway-related genes. The two significant genes, B4GALNT1 and CERS4, were further evaluated using Cox regression analyses, and their reproducibility was validated in an independent cohort. Cell-based functional experiments were performed to assess the biological roles of B4GALNT1 and CERS4. (Abbreviations: LUAD, lung adenocarcinoma; TCGA, The Cancer Genome Atlas; DEG, differential gene expression; mRNA, messenger RNA; NST, normal solid tissue; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein–protein interaction.).
Fig 2.
Differentially expressed sphingolipid metabolism-related genes.
(A) Heatmap of significantly differentially expressed sphingolipid metabolism-related genes based on mRNA expression levels between LUAD tissues and NSTs. In the matrix format, rows correspond to individual sphingolipid metabolism-related genes; columns represent distinct tissue samples. Each cell in the matrix displays the relative mRNA expression of a specific gene in a particular tissue. Red and blue indicate higher and lower expression levels, respectively. Samples are organized from NSTs to LUAD tissues based on the standardized expression levels of each gene. (B) Relative mRNA expression changes of various sphingolipid metabolism-related genes; statistical significance set at P < 0.05 (NSTs vs. LUAD tissues). (Abbreviations: LUAD, lung adenocarcinoma; NST, normal solid tissue; mRNA, messenger RNA.).
Fig 3.
Functional enrichment analyses and PPI network construction.
(A) GO enrichment analysis of biological process, cellular component, and molecular function terms. The top five significantly enriched terms from each domain are visualized. (B) Top five significantly enriched KEGG pathways based on enrichment analysis. (C) PPI network constructed with genes involved in the sphingolipid metabolism pathway identified in the KEGG analysis. (D) From the PPI network, genes with high interconnectivity were identified as a core module. Nodes represent genes, and edges indicate known or predicted interactions. (Abbreviations: GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein–protein interaction.).
Fig 4.
Survival analysis of key sphingolipid metabolism-related genes.
(A–D) Kaplan–Meier survival curves for ASAH1, CERS4, B4GALNT1, and CERS6, created using a non-parametric log-rank test. For each gene, patients are stratified into high-expression and low-expression groups based on the optimal cutoff value. The x-axis represents survival time (days); the y-axis indicates the survival probability. Shaded areas represent 95% confidence intervals for each group. (Abbreviations: ASAH1, N-acylsphingosine amidohydrolase 1; CERS4, ceramide synthase 4; B4GALNT1, beta-1,4-N-acetyl-galactosaminyltransferase 1; CERS6, ceramide synthase 6.).
Fig 5.
Expression of B4GALNT1 and CERS4 based on tumor stage and metastatic status.
(A, B) mRNA expression levels of B4GALNT1 and CERS4 across different tumor stages (Stage I, II, III, and IV) in LUAD. (C, D) mRNA expression levels of B4GALNT1 and CERS4 according to metastatic status (M0 vs. M1). Statistical significance assessed using one-way ANOVA for stage comparison and Student’s t-test for metastasis comparison. (Abbreviations: LUAD, lung adenocarcinoma; mRNA, messenger RNA; ANOVA, analysis of variance; B4GALNT1, beta-1,4-N-acetyl-galactosaminyltransferase 1; CERS4, ceramide synthase 4.).
Fig 6.
Prognostic significance of B4GALNT1 and CERS4 expression in LUAD.
(A, B) Cox regression analyses of B4GALNT1 and CERS4 expression in the TCGA-LUAD dataset. (C, D) Kaplan–Meier survival analyses of B4GALNT1 and CERS4 expression in the GSE72094 dataset. (E, F) Cox regression analyses of B4GALNT1 and CERS4 expression in the GSE72094 dataset. (Abbreviations: LUAD, lung adenocarcinoma; HR, hazard ratio; CI, confidence interval; B4GALNT1, beta-1,4-N-acetyl-galactosaminyltransferase 1; CERS4, ceramide synthase 4.).
Fig 7.
Effect of B4GALNT1 knockdown and CERS4 overexpression on A549 cell aggressiveness.
(A) Relative growth at 24 h and 48 h compared to 0 h in siCon and siB4GALNT1 groups. (B) Representative bright-field images of the wound area 0, 24, and 48 h after scratch in siCon and siB4GALNT1 groups. (C) Relative wound closure at 24 h and 48 h compared to 0 h in siCon and siB4GALNT1 groups. (D) Western blot analysis of EMT markers (Snail, Vimentin, N-cadherin, E-cadherin) and B4GALNT1 at 0, 24, and 48 h after siB4GALNT1 treatment; β-actin is the loading control. (E) Relative cell proliferation at 24 h and 48 h compared to 0 h in oe_Con and oe_CERS4 groups. (F) Representative bright-field images of the wound area at 0, 24, and 48 h after scratch in oe_Con and oe_CERS4 groups. (G) Relative wound closure at 24 h and 48 h compared to 0 h in oe_Con and oe_CERS4 groups. (H) Western blot analysis of EMT markers (Snail, Vimentin, N-cadherin) and CERS4 at 0, 24, and 48 h after oe_CERS4-HA transfection; HA and β-Actin were also detected. (Abbreviations: EMT, epithelial–mesenchymal transition; siCon, small interfering RNA control; siB4GALNT1, siRNA targeting B4GALNT1; oe_Con, overexpression control; oe_CERS4, overexpression of CERS4; HA, hemagglutinin tag; B4GALNT1, beta-1,4-N-acetyl-galactosaminyltransferase 1; CERS4, ceramide synthase 4.).