Abstract
Myeloid-derived suppressor cells (MDSCs) are responsible for antitumor immunodeficiency in tumor-bearing hosts. Primarily, MDSCs are classified into 2 groups: monocytic (M)-MDSCs and polymorphonuclear (PMN)-MDSCs. In most cancers, PMN-MDSCs (CD11b(+) Ly6C(low) Ly6G(+) cells) represent the most abundant MDSC subpopulation. However, the functional and phenotypic heterogeneities of PMN-MDSC remain elusive, which delays clinical therapeutic targeting decisions. In the 4T1 murine tumor model, CD11b(+) Ly6G(low) PMN-MDSCs were sensitive to surgical and pharmacological interventions. By comprehensively analyzing 64 myeloid cell-related surface molecule expression profiles, cell density, nuclear morphology, and immunosuppressive activity, the PMN-MDSC population was further classified as CD11b(+) Ly6G(low) CD205(+) and CD11b(+) Ly6G(high) TLR2(+) subpopulations. The dichotomy of PMN-MDSCs based on CD205 and TLR2 is observed in 4T07 murine tumor models (but not in EMT6). Furthermore, CD11b(+) Ly6G(low) CD205(+) cells massively accumulated at the spleen and liver of tumor-bearing mice, and their abundance correlated with in situ tumor burdens (with or without intervention). Moreover, we demonstrated that CD11b(+) Ly6G(low) CD205(+) cells were sensitive to glucose deficiency and 2-deoxy-d-glucose (2DG) treatment. Glucose transporter 3 (GLUT3) knockdown by siRNA significantly triggered apoptosis and reduced glucose uptake in CD11b(+) Ly6G(low) CD205(+) cells, demonstrating the dependence of CD205(+) PMN-MDSCs survival on both glucose uptake and GLUT3 overexpression. As GLUT3 has been recognized as a target for the rescue of host antitumor immunity, our results further directed the PMN-MDSC subsets into the CD205(+) GLUT3(+) subpopulation as future targeting therapy.