TEAD3 + high-risk melanoma cells crosstalk with GAS6 + macrophages via the GAS6-TYRO3 ligand-receptor axis to modulate propionate metabolism and drive melanoma progression

Abstract

Background: Melanoma, a highly heterogeneous malignancy, remains refractory to conventional therapies due to poorly defined molecular and metabolic drivers. Short-chain fatty acid (SCFA) metabolism influences tumor progression, yet its role in melanoma subtypes and clinical outcomes is unclear. This study aims to delineate melanoma subgroups driven by SCFA metabolic dysregulation and identify mechanisms underlying their aggressiveness. Methods: Using non-negative matrix factorization (NMF), we clustered 468 TCGA melanoma samples into six subgroups based on SCFA-related gene sets (GO:0019745, GO:0019746, GO:0006085). Survival, differential expression, and pathway analyses were performed to characterize high-risk subgroups. Key drivers were validated via CRISPR/Cas9, siRNA knockdown, and immunohistochemistry. Single-cell RNA-seq (GSE215120) and spatial transcriptomics elucidated tumor-microenvironment crosstalk. Metabolic profiling, Seahorse assays, and myeloid-specific GAS6 knockout models were employed to dissect mechanisms. Results: NMF clustering revealing a high-risk subtype (Group 6) with dysregulated short-chain fatty acid (SCFA) metabolism and poor survival. Group 6 exhibited upregulation of GLTP and RAPGEFL1, enrichment in melanogenesis, Hippo signaling, and skin/lipid metabolism pathways. Through integrative analysis, TEAD3 emerged as a key risk driver, with high expression correlating with poor prognosis. Functional validation demonstrated that TEAD3 knockout suppressed melanoma proliferation, migration, and epithelial-mesenchymal transition (EMT) in vitro and in vivo. Single-cell RNA sequencing of acral melanoma revealed TEAD3-enriched tumor cells interacting with M2 macrophages via the GAS6-TYRO3 axis. Mechanistically, GAS6 + macrophages exhibited hypermetabolic phenotypes (elevated glycolysis/OXPHOS) that fueled GAS6 secretion. GAS6-TYRO3 signaling in TEAD3 + cells drove tumor aggressiveness by rewiring propionate metabolism, inducing methylmalonic acid accumulation via Mmut upregulation. Targeting this axis in myeloid-specific GAS6 knockout mice enhanced anti-PD-1 efficacy, boosting CD8 + T cell infiltration and survival. Conclusion: We define a TEAD3-driven melanoma subtype reliant on SCFA metabolic reprogramming and M2 macrophage crosstalk. The GAS6-TYRO3 axis and Mmut-mediated methylmalonic acid accumulation represent actionable targets. Combining myeloid-GAS6 ablation with immune checkpoint blockade overcomes therapy resistance, offering a precision strategy for high-risk melanoma.