Exploration of the Role of M2 Macrophages in Hepatocellular Carcinoma: Insights into Disulfidptosis and Cellular Interactions

Background: The interplay between regulated cell death processes, especially disulfidptosis, and the immune microenvironment in hepatocellular carcinoma (HCC) remains poorly understood. Therefore, this study aimed to elucidate these relationships systematically at single-cell resolution. Methods: Multi-omics data from TCGA (n = 315) were integrated with single-cell RNA sequencing (scRNA-seq) data from 21 HCC samples. Disulfidptosis-related subtypes were identified using non-negative matrix factorization (NMF) clustering of co-expressed long non-coding RNAs (lncRNAs). At the single-cell level, tumor cells were stratified into high- and low-disulfidptosis groups based on a 14-gene signature that included SLC7A11 and G6PD. Metabolic reprogramming, immune infiltration, and intercellular communication between high-disulfidptosis tumor cells and M2 macrophages were analyzed. Gene signatures were employed to quantify ferroptosis, and correlations with disulfidptosis were assessed. Key molecular axes, such as GPX4-RAC1, were validated by multiplex immunofluorescence. Results: Three molecular subtypes with distinct prognoses were identified. The disulfidptosis-enriched subtype (subtype 1) demonstrated superior survival and specific metabolic features, including glucose/glutamine enrichment. Single-cell analysis revealed that high-disulfidptosis tumor cells exhibited a paradoxical metabolic state: upregulation of disulfidptosis-associated genes (SLC7A11, RAC1), accompanied by downregulation of traditional energy pathways. These cells actively recruited and engaged M2 macrophages through a bidirectional communication network dominated by the macrophage migration inhibitory factor (MIF) pathway from tumor cells to M2 macrophages, and the GALECTIN/CypA pathways from M2 macrophages to tumor cells. Notably, a significant positive correlation was observed between disulfidptosis and ferroptosis scores, with the strongest association the M2 macrophage population and centered on the core regulatory stage (Stage 2). This correlation was molecularly anchored by a significant co-expression of GPX4 and RAC1 in M2 macrophages, a finding corroborated by spatial protein co-localization. Conclusion: This study delineates a disulfidptosis-associated HCC subtype with a unique metabolic–immune niche. These findings reveal an active, M2 macrophage-involved adaptive network and propose a novel mechanistic link between disulfidptosis and ferroptosis via the GPX4–RAC1 axis within the tumor microenvironment.