Melanoma cell-derived LAG-3 enhances CXCL1/8-driven MDSCs recruitment and immune escape via TRIM28-mediated IκBα degradation

Introduction Lymphocyte activation gene 3 (LAG-3) is recognized as an important inhibitory receptor on T cells, and LAG-3-targeted therapy has been approved for advanced melanoma treatment. Recently, LAG-3 has been reported to be expressed on tumor cells. However, the potential effects and underlying mechanisms of tumor cell-derived LAG-3 in mediating tumor growth and immunosuppression remain unclear. Objectives This study aimed to evaluate the role of tumor cell-derived LAG-3 within the tumor microenvironment (TME) and the efficacy of anti-LAG-3 combination therapy against melanoma. Methods The Cancer Genome Atlas, Cancer Cell Line Encyclopedia, single-nuclear RNA sequencing database and tissue microarray analyses were conducted to detect the LAG-3 gene and protein expressions. The biological functions and underlying mechanism of tumor cell-derived LAG-3 in tumor growth and immunosuppression were investigated using phospho-array, RNA-seq, co-immunoprecipitation, flow cytometry, co-culture systems, and various tumor-bearing mouse models. Results Our findings revealed that LAG-3 was present in a subpopulation of cancer cells and clinical melanoma specimens associated with poorer prognosis. High LAG-3 expression on tumor cells not only contributed to the proliferation and metastasis of melanoma cells, but also promoted the recruitment of myeloid-derived suppressor cells (MDSCs) and dysfunctional CD8 + T cells into the TME. Notably, depletion of MDSCs counteracted the pro-tumor activity of tumor cell-derived LAG-3. Mechanistically, LAG-3 overexpression triggered CXCL1 and CXCL8 production of melanoma cells by enhancing TRIM28-mediated IκBα ubiquitination and degradation, which in turn activated the NF-κB pathway and fostered tumor growth and an immunosuppressive microenvironment. Intriguingly, mice bearing LAG-3-overexpressing melanoma cells showed reduced survival but responded to anti-LAG-3 antibody therapy. Furthermore, inhibiting the CXCL1/8-CXCR2 axis sensitized LAG-3-overexpressing melanoma to anti-LAG-3 antibody. Conclusion This study firstly identified LAG-3 as a pro-tumor factor and a predictor for anti-LAG-3 immunotherapy efficacy. Combined CXCL1/8-CXCR2 inhibitor and anti-LAG-3 treatment may represent a promising therapeutic strategy for LAG-3-driven tumor