Identification of MAD2L1 as a novel biomarker for hepatoblastoma through bioinformatics and machine learning approaches

Abstract

Objective: This study aims to identify potential biomarkers for Hepatoblastoma (HB) using bioinformatics and machine learning, and to explore their underlying mechanisms of action.

Methods: We analyzed the datasets GSE131329 and GSE133039 to perform differential gene expression analysis. Single-sample gene set enrichment analysis (ssGSEA) and weighted gene co-expression network analysis (WGCNA) were utilized to identify gene modules linked to gene set activity. Protein-protein interaction (PPI) networks were constructed to identify hub genes, while random forest and support vector machine models were employed to screen for key diagnostic genes. Survival and immune infiltration analyses were conducted to assess the prognostic significance of these genes. Additionally, the expression levels, biological functions, and mechanisms of action of the selected genes were validated in HB cells through relevant experimental assays.

Results: We identified 1,377 and 1,216 differentially expressed genes in datasets GSE131329 and GSE133039, respectively. ssGSEA and WGCNA analyses identified 234 genes significantly linked to gene set activity. PPI analysis identified 20 core Hub genes. Machine learning highlighted three key diagnostic genes: CDK1, CCNA2, and MAD2L1. Studies have demonstrated that MAD2L1 is significantly overexpressed in HB and is associated with prognosis. WGCNA revealed that MAD2L1 is enriched in gene sets related to E2F_ TARGETS and G2M_CHECKPOINT. Experimental assays demonstrated that MAD2L1 knockdown significantly inhibits the proliferation, migration, and invasion of HB cell lines, and that MAD2L1 promotes cell cycle progression through the regulation of E2F.

Conclusion: Our study identifies MAD2L1 as a novel potential biomarker for HB, providing new strategies for early diagnosis and targeted therapy in HB.

Keywords: MAD2L1; WGCNA; biomarkers; hepatoblastoma; machine learning.

Figure 1

Difference Analysis and WGCNA in the GSE131329 Dataset. (A) Differential analysis volcano plot; gray indicates upregulated genes, downregulated blue, and orange not statistically significant. (B) Cluster heat map showing gene expression differences. (C) ssGSEA results; p < 0.001 marked ***, p < 0.01 as **, p < 0.05 as *, and p > 0.05 as ns. (D) Sample clustering diagram. (E) Soft threshold determination. (F) Histograms and logarithmic plots. (G) Gene clustering tree. (H) Module correlation heat map. (I) Sample clustering and gene set enrichment heat map. (J) Heat map of gene module-gene set relationships.

Figure 2

Difference Analysis and WGCNA in the GSE133039 Dataset. (A): Differential analysis volcano plot; purple indicates upregulated genes, green downregulated, and gray not statistically significant. (B): Cluster heat map showing gene expression differences. (C): ssGSEA results; p < 0.001 marked ***, p < 0.01 as **, p < 0.05 as *, and p > 0.05 as ns. (D): Sample clustering diagram. (E): Soft threshold determination. (F): Histograms and logarithmic plots. (G): Gene clustering tree. (H): Module correlation heat map. (I): Sample clustering and gene set enrichment heat map. (J): Heat map of gene module-gene set activity relationships.

Figure 3

Screening and analysis results of hub gene. (A) Venn diagram. (B) PPI network analysis reveals the interaction relationship between proteins. (C) Network chart shows the higher degree of gene network. (D) Network diagram of the top 20 Hub genes. (E) The resulting map of GO enrichment analysis. (F) The result map of KEGG enrichment analysis. (G, H) Enrichment analysis network maps show that these Hub genes influence multiple cell functions and potential biological roles.

Figure 4

Machine learning algorithms screen for key genes. (A) RF identified HB biomarkers in GSE131329. (B) Top five genes from RF. (C) SVM-RFE selected seven genes, achieving a minimum error rate of 0.195. (D) Seven genes from SVM-RFE with 0.805 accuracy. (E) Venn diagram. (F–H) ROC curve analysis for key genes. (I) RF identified HB biomarkers in GSE133039. (J) Top five genes from RF. (K) SVM-RFE selected seven genes, with a minimum error rate of 0.19. (L) Seven genes from SVM-RFE, achieving 0.81 accuracy. (M) Venn diagram. (N, O) ROC curve analysis for key genes.

Figure 5

Validation of MAD2L1 as a biomarker of HB. (A) qPCR detected MAD2L1 mRNA levels in HB tumor and adjacent normal tissues. (B) WB analysis showed MAD2L1 protein expression in HB tumors (T) vs. non-cancerous tissues (N). (C) Quantitative analysis of MAD2L1 in HB (n = 6). (D) Kaplan-Meier analysis of MAD2L1 expression and DFS in HB patients. (E) OS analysis of MAD2L1 expression in HB patients. (F, G) MAD2L1 ssGSEA analysis of immune cell infiltration in GSE131329 and GSE133039 datasets. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6

MAD2L1 promotes hepatoblastoma cell proliferation. (A, B) HuH6 and HepG2 cells were transfected with NC or four siRNAs targeting MAD2L1; relative mRNA levels were assessed by RT-qPCR. Knockdown efficiency evaluated by Western blot. (C, D) CCK-8 assay assessed proliferation effects. (E, F) The effect of MAD2L1 knockdown on cell proliferation was explored using a colony formation assay. (G) The effect of MAD2L1 knockdown on PCNA expressions was detected by western blotting. *p < 0.05, **p < 0.01, ***p < 0.001, p ****<0.0001.

Figure 7

MAD2L1 promotes hepatoblastoma cell migration and invasion. (A, B) Transwell assays evaluated invasion capability. (C, D) Scratch assays assessed migration effects. (E) The expression levels of EMT markers were determined by western blotting. *p < 0.05, **p < 0.01, ***p < 0.001, p ****<0.0001.

Figure 8

MAD2L1 activates E2F transcription and regulates cell cycle regulators. (A, B) E2F transcription factor relative luciferase activity in dual-luciferase reporter gene assay. (C, D) The effect of MAD2L1 knockdown on Cyclin A2, E2F3 and Cyclin E1 expressions was detected by western blotting. (E, F) The effect of MAD2L1 knockdown on cell cycle was explored using a flow cytometry assay. *p < 0.05, **p < 0.01, ***p < 0.001, p ****<0.0001.