Prognostic 7-SLC-Gene Signature Identified via Weighted Gene Co-Expression Network Analysis for Patients with Hepatocellular Carcinoma.

Background: Solute carrier (SLC) proteins play an important role in tumor metabolism. But SLC-associated genes' prognostic significance in hepatocellular carcinoma (HCC) remained elusive. We identified SLC-related factors and developed an SLC-related classifier to predict and improve HCC prognosis and treatment. Methods: From the TCGA database, corresponding clinical data and mRNA expression profiles of 371 HCC patients were acquired, and those of 231 tumor samples were derived from the ICGC database. Genes associated with clinical features were filtered using weighted gene correlation network analysis (WGCNA). Next, univariate LASSO Cox regression studies developed SLC risk profiles, with the ICGC cohort data being used in validation. Result: Univariate Cox regression analysis revealed that 31 SLC genes (P < 0.05) were related to HCC prognosis. 7 (SLC22A25, SLC2A2, SLC41A3, SLC44A1, SLC48A1, SLC4A2, and SLC9A3R1) of these genes were applied in developing a SLC gene prognosis model. Samples were classified into the low-andhigh-risk groups by the prognostic signature, with those in the high-risk group showing a significantly worse prognosis (P < 0.001 in the TCGA cohort and P=0.0068 in the ICGC cohort). ROC analysis validated the signature's prediction power. In addition, functional analyses showed enrichment of immune-related pathways and different immune status between the two risk groups. Conclusion: The 7-SLC-gene prognostic signature established in this study helped predict the prognosis, and was also correlated with the tumor immune status and infiltration of different immune cells in the tumor microenvironment. The current findings may provide important clinical indications for proposing a novel combination therapy consists of targeted anti-SLC therapy and immunotherapy for HCC patients.

Hypoxia-responsive PPARGC1A/BAMBI/ACSL5 axis promotes progression and resistance to lenvatinib in hepatocellular carcinoma.

Emerging evidence has indicated that peroxisome proliferator-activated receptor-gamma coactivator-1α (PPARGC1A) is involved in hepatocellular carcinoma (HCC). However, its detailed function and up- and downstream mechanisms are incompletely understood. In this study, we confirmed that PPAGC1A is lowly expressed in HCC and is associated with poor prognosis using large-scale public datasets and in-house cohorts. PPAGC1A was found to impair the progression and sensitivity of HCC to lenvatinib. Mechanistically, PPAGC1A repressed bone morphogenetic protein and activin membrane-bound inhibitor (BAMBI) by inhibiting WNT/ß-catenin signaling. BAMBI mediated the function of PPARGC1A and regulated ACSL5 through TGF-ß/SMAD signaling. PPARGC1A/BAMBI regulated ROS production and ferroptosis-related cell death by controlling ACSL5. PPARGC1A/BAMBI/ACSL5 axis was hypoxia-responsive. METTL3 and WTAP silenced PPARGC1A in an m6A-YTHDF2-dependent way under normoxia and hypoxia, respectively. Metformin restored PPARGC1A expression by reducing its m6A modification via inhibiting METTL3. In animal models and patient-derived organoids, consistent functional data of PPARGC1A/BAMBI/ACSL5 were observed. Conclusions: These findings provide new insights into the role of the aberrant PPARGC1A/BAMBI/ACSL5 axis in HCC. And the mechanism of PPARGC1A dysregulation was explained by m6A modification. Metformin may benefit HCC patients with PPARGC1A dysregulation.

IHGA: An interactive web server for large-scale and comprehensive discovery of genes of interest in hepatocellular carcinoma.

Mining gene expression data is valuable for discovering novel biomarkers and therapeutic targets in hepatocellular carcinoma (HCC). Although emerging data mining tools are available for pan-cancer-related gene data analysis, few tools are dedicated to HCC. Moreover, tools specifically designed for HCC have restrictions such as small data scale and limited functionality. Therefore, we developed IHGA, a new interactive web server for discovering genes of interest in HCC on a large-scale and comprehensive basis. Integrative HCC Gene Analysis (IHGA) contains over 100 independent HCC patient-derived datasets (with over 10,000 tissue samples) and more than 90 cell models. IHGA allows users to conduct a series of large-scale and comprehensive analyses and data visualizations based on gene mRNA levels, including expression comparison, correlation analysis, clinical characteristics analysis, survival analysis, immune system interaction analysis, and drug sensitivity analysis. This method notably enhanced the richness of clinical data in IHGA. Additionally, IHGA integrates artificial intelligence (AI)-assisted gene screening based on natural language models. IHGA is free, user-friendly, and can effectively reduce time spent during data collection, organization, and analysis. In conclusion, IHGA is competitive in terms of data scale, data diversity, and functionality. It effectively alleviates the obstacles caused by HCC heterogeneity to data mining work and helps advance research on the molecular mechanisms of HCC.