Galectin-9 (Gal-9), a ligand for T cell Ig mucin-3 (Tim-3), induces apoptosis in cluster of differentiation 4 (CD4)(+) Tim-3(+) T helper 1 (T(H)1) cells via the Gal-9-Tim-3 pathway and negatively regulates T(H)1 immunity. In turn, Gal-9 activates dendritic cells (DC) to produce TNF-α, which promotes the T(H)1 response. We investigated the efficacy of Gal-9 against T(H)1-mediated autoimmune diabetes in NOD mice and compared with anti-Tim-3 monoclonal antibody (RMT3-23), which inhibited the binding between Tim-3-Ig and Gal-9 in a solid-phase binding assay. mRNA expression of Gal-9 was prominently induced by the treatment of interferon-γ in MIN6 cells, and Gal-9 was also expressed in the pancreatic β-cells in NOD mice, suggesting Gal-9 may be released from pancreatic β-cells to terminate T(H)1-mediated inflammation. Long-term injection of Gal-9 exhibits preventive efficacy for development of diabetes in NOD mice, but RMT3-23 demonstrated further prominent therapeutic potential compared with Gal-9. Gal-9 induced apoptosis of CD4(+)Tim-3(+) T(H)1 cells at the concentration of 0.2 μM, whereas RMT3-23 failed to enhance the apoptosis of CD4(+)Tim-3(+) T(H)1 cells. In contrast, Gal-9 induced TNF-α production in cultured DC in a dose-dependent manner; however, RMT3-23 inhibited Gal-9-induced TNF-α production in a dose-dependent manner. Although Gal-9 exhibited certain therapeutic potential against autoimmune diabetes by enhancing apoptosis of CD4(+)Tim-3(+) T(H)1 cells, RMT3-23 exhibited prominent therapeutic efficacy by suppressing the TNF-α production and activation of DC. Taken together, the inhibition of the Gal-9-Tim-3 pathway on DC, upstream of T(H)1 response, may be a new target for the treatment of type 1 diabetes.