Proteomic profiling identifies miR-423-5p as a modulator of oncogenic metabolism in HCC

Abstract

Background: Hepatocellular carcinoma (HCC) remains a significant clinical challenge due to limited diagnostic and therapeutic options. Non-coding RNAs (ncRNAs), such as microRNAs (miRNAs), play key roles in cancer biology. Our previous findings showed that miR-423-5p enhances anti-cancer effects on HCC patients treated with sorafenib by promoting autophagy. Here, we investigated the molecular mechanisms underlying miR-423-5p function through a comprehensive proteomic approach.

Methods: We generated an HCC cell line stably overexpressing miR-423-5p via lentiviral transduction. Total proteins were extracted from SNU-387 cells, enzymatically digested into peptides, and subsequently analysed by liquid chromatography-tandem mass spectrometry (LC-MS/M). Raw spectral data were processed and quantified using MaxQuant. Differentially expressed proteins (DEPs) were defined based on fold-change (|log2FC| ≥ 1) and false discovery rate (FDR < 0.05). The full proteomic dataset is available via the ProteomeXchange repository (identifier: PXD064869). Functional enrichment analysis of DEPs were performed using DAVID and Reactome. To assess clinical relevance, predicted and validated miR-423-5p targets were integrated with The Cancer Genome Atlas (TCGA) Liver Hepatocellular Carcinoma (LIHC) dataset using GEPIA platform. Survival analyses were performed using the Kaplan-Meier method.

Results: Proteomic profiling identified 698 DEPs in miR-423-5p-overexpressing cells compared to controls with significant enrichment in metabolic pathways, related to purine/pyrimidine metabolism and gluconeogenesis. Integration with bioinformatic predictions and miRTarBase validation identified 43 DEPs as potential direct targets of miR-423-5p. Among these, seven proteins (ACACA, ANKRD52, DVL3, MCM5, MCM7, RRM2, SPNS1, and SRM) were significantly associated with patient prognosis in the TCGA-LIHC cohort. These targets were downregulated in miR-423-5p-overexpressing cells but upregulated in advanced-stage HCC tissues, suggesting a potential role for miR-423-5p in the regulation of HCC pathogenesis. Stage-specific expression analysis showed increased levels from stage I to III, followed by a decline at stage IV. Notably, we experimentally confirmed miR-423-5p-mediated suppression of MCM7, DVL3, IMPDH1, and SRM (SPEE), supporting their functional involvement in HCC progression.

Conclusion: Overall, our findings support a tumour-suppressive role for miR-423-5p in HCC, mediated by modulation of metabolic pathways and suppression of oncogenic proteins. These results suggest that miR-423-5p and its downstream effectors may serve as promising biomarkers and potential therapeutic targets in HCC.

Highlights: miR-423-5p acts as a tumor suppressor in HCC by targeting key nodes of pro-tumorigenic signalling. miR-423-5p significantly altered metabolic pathways, including purine/pyrimidine metabolism and gluconeogenesis. Seven miR-423-5p targets correlate with poor prognosis in TCGA-LIHC patients and are downregulated in miR-423-5p overexpressing HCC cells. miR-423-5p over-expression induces a significant downregulation of MCM7, DVL3, IMPDH1, SPEE in HCC cell models. miR-423-5p limits tumor metabolic plasticity, suggesting therapeutic potential.

Keywords: Gluconeogenesis; Hepatocellular carcinoma; Nucleotide metabolism; Overall survival; Proteomics; Spermidine synthase; Stage plot; miR-423-5p.

Fig. 1

A Volcano plot showing differentially expressed proteins (DEPs) in miR-423-5p-overexpressing SNU-387 cells compared to empty plasmid (pLKO.1). The top five upregulated and downregulated proteins are highlighted. Red dots represent significantly upregulated proteins (log2FC >0.6), green dots indicate significantly downregulated proteins (log2FC < -0.6), and grey dots denote proteins that did not meet the significance criteria (adjusted p-value ≥ 0.05 or |log2FC| ≤ 0.6). B Heatmap of the top 50 Differentially Expressed Proteins (DEPs) in miR-423-5p-overexpressing SNU-387 cells compared to empty plasmid (pLKO.1)

Fig. 2

Dot plot of enriched Gene Ontology (GO) categories selected for proteins down-regulated by miR-423-5p compared to pLKO.1. A Biological Processes (GO: BP) B Cellular Components (GO: CC). Dot sizes represent the number of downregulated proteins associated with the GO term and dot colors represent the p-value from the over-represented Fisher’s test

Fig. 3

Enriched Gene Ontology (GO) analysis of down-regulated proteins in miR-423-5p overexpressing SNU-387 cells compared to pLKO.1. A Molecular Function (GO: MF) B KEGG pathways Dot sizes represent the number of down-regulated proteins and dot colors indicate significance (p-value, Fisher’s exact test)

Fig. 4

TGs prediction analysis and comparison with DEPs. A Predicted targets of hsa-miR-423-5p identified using four independent target prediction tools: miRDIP (v5.3), mirDB (v6.0), TargetScanHuman (v8.0), and DIANA-microT (v22.1) depicted in the Venn diagram B Comparison between experimentally validated and computationally predicted targets of miR-423-5p and DEPs identified in response to miR-423-5p overexpression in SNU-387 hepatocellular carcinoma (HCC) cell line

Fig. 5

Kaplan -Meier curves for overall and disease-free survival of differentially expressed proteins (DEPs) in miR-423-5p overexpressing SNU-387 cells, showing significant correlation with patients’ outcomes in TCGA-LIHC dataset (log-rank p-value < 0.05)

Fig. 6

Box plots and stage-wise expression plots showing the expression patterns of seven target genes (TGs) identified in miR-423-5p-overexpressing SNU-387 cells. These TGs exhibit significantly altered expression in TCGA-LIHC tumor samples compared to normal liver tissues from the TCGA and GTEx datasets

Fig. 7

Western blot analysis of transduced HCC cell models. A SNU-387 cells lysates were incubated with MCM7, SPEE, IMPDH1 and DVL3. B Hep-3B cells lysates were incubated with MCM7, SPEE, IMPDH1 and DVL3. β-actin served as a loading control. Each experiment was independently repeated at least three times, yielding consistent similar results. Bar graphs represent the band intensities expressed in arbitrary units. Error bars indicate standard deviations (SD). Statistical significance was determined by ANOVA: ∗p < 0.01; ∗∗p < 0.05; ∗∗∗p < 0.005; ∗∗∗∗p < 0.0001