Multiple-omics analysis of aggrephagy-related cellular patterns and development of an aggrephagy-related signature for hepatocellular carcinoma.

Background: Protein aggrephagy, a selected autophagy process response for degrading protein aggregates, plays a critical role in various cancers. However, its regulatory mechanisms and clinical implications in hepatocellular carcinoma (HCC) remain largely unexplored.

Methods: We integrated bulk RNA-seq data from TCGA and single-cell RNA sequencing (scRNA-seq) data from GEO databases to systematically analyze aggrephagy-related genes (AGGRGs) in HCC. Prognostic aggrephagy-related genes (AGGRGs) were identified through univariate Cox and LASSO regression analyses, followed by the construction of a risk prediction model. Patients were stratified into high- and low-risk groups based on the median risk score. Comparative analyses were performed to assess clinical outcomes, pathway enrichment, and drug sensitivity. Independent risk factors were incorporated a nomogram using univariate and multivariate Cox regression. At the single-cell level, the AGG scores were calculated using AUCell algorithm, and cell interactions and pseudotime trajectory analyses were conducted. Finally, protein levels of key AGGRG was assessed via tissue microarray.

Results: Eight AGGRGs (PFKP, TPX2, UBE2S, GOT2, ST6GALNAC4, ADAM15, G6PD, and KPNA2) were identified as prognostic markers for HCC. The high-risk group exhibited significantly worse survival outcomes, heightened drug resistance, and enrichment in cell cycle, mTORC1 signaling, and reactive oxygen species pathways. Single-cell transcriptomic analysis revealed 11 distinct cell types within the HCC tumor microenvironment (TME), including hepatocytes, T cells, NK cells, macrophages, monocytes, dendritic cells, plasma B cells, mature B cells, mast cells, endothelial cells, and fibroblasts. Hepatocytes exhibited the highest AGGRG scores and were associated with metabolic reprograming, proliferation, and immune evasion. Further subclustering of malignant hepatocytes using inferCNV revealed eight functionally heterogeneous subpopulations with extensive intercellular crosstalk. Trajectory analysis showed G6PD- and CCNB1-expressing subpopulations in early-to-intermediate differentiation states, whereas C3 and ARGs marked terminal differentiation. Notably, G6PD was predominantly expressed in early and mid-stages, while KPNA2, PFKP, and TPX2 were upregulated in advanced tumor states. Immunohistochemical (IHC) validation confirmed significant overexpression of G6PD in HCC tissues compared to adjacent normal tissues.

Conclusion: These findings provide a molecular framework for targeting aggrephagy pathways in HCC treatment strategies.