Vaccination with B16/Gal induces T cell–dependent antitumor resistance. (A) 10⁶ α-GalCer–loaded DCs per mouse were administered to mice. 10⁵ B16 melanoma tumor cells were injected s.c. 2 wk later. Measurement of tumor size was done at the indicated time points (n = 5 per group). Similar results were obtained in two independent experiments. (B) 10⁵ B16/Gal or nonloaded B16 melanoma cells per mouse were administered s.c., and tumor size was followed (n = 5 per group). Similar results were obtained in two independent experiments. (C and D, left) 5 × 10⁵ α-GalCer–loaded B16 or CD1d^hi-B16 tumor cells per mouse were given i.v., and 2 wk later mice were challenged with 10⁵ parental B16 tumors s.c. (n = 10 per group). Similar results were obtained in two independent experiments. (D, right) At 2 wk after vaccination of CD4^−/−, CD8^−/−, and MHC II^−/− mice with 5 × 10⁵ CD1d^hi-B16/Gal i.v., the mice were challenged s.c. with B16 tumor cells (n = 5 per group). Similar results were obtained in two independent experiments. (E) CD19⁺ B cells were isolated with magnetic beads from the spleen, coated with α-GalCer, and injected together with irradiated CD1d^hi-B16 cells (n = 5 per group). Data are means ± SEM. Similar results were obtained in two independent experiments.

Vaccination with live and irradiated tumor/Gal induces antitumor protection against several mouse tumors. (A, left) Immunization experiments were set up as shown; i.e., mice were vaccinated with 5 × 10⁶ of different forms of J558 myeloma cells, challenged at 4 wk with 10⁶ live J558 s.c., and evaluated for their survival (n = 10 per group). Similar results were obtained in two independent experiments. (B) Mice were injected i.v. with 5 × 10⁵ live and irradiated CD1d^hi-B16/Gal (n = 9 and 6 per group, respectively), 2 × 10⁶ CD1d^hi-WEHI-3B/Gal (n = 5 per group), or 5 × 10⁵ CD1d^hi-EL4/Gal (n = 8 and 5 per group, respectively). The mice were challenged s.c. with parental tumor cells (5 × 10⁴ B16, 2 × 10⁵ WEHI-3B, or 2 × 10⁵ EL4) 4 wk later. Tumor sizes were measured at the indicated time points. Similar results were obtained in two independent experiments. (C) Mice were immunized and challenged as indicated in the diagram (n = 5 B16 or EL4; n = 9 tumor/Gal-B16 −EL4). Data are means ± SEM. Similar results were obtained in two independent experiments.

Maturing DCs capture tumor/Gal after i.v. injection. (A–C) 10 million CFSE-labeled CD1d^hi-EL4 with or without loading with α-GalCer were injected into C57BL/6 mice. (A, left) Gated areas indicate subsets of CD11c⁺ DCs, CD8α⁺ DCs, and CD8α⁻ DCs. (right) The frequency of uptake of CFSE⁺ tumor cells by CD11c⁺ splenic DCs was measured by flow cytometry. (B) CD11c cells were enriched by MACS for analysis by confocal microscopy after injection of CFSE-labeled live CD1d^hi-EL4/Gal cells. (i) Tumor fragments (green), (ii) CD86 to detect mature DCs (red), (iii) DAPI nuclear stain (blue), and (iv) a merged image are shown. Bar, 10 μm. The data in A and B are representative of three independent experiments (n = 3 per group). (C) The mean percentage of DCs taking up CFSE-labeled tumor debris at the indicated time points is shown. (D) As in C, but antigen capture by mature DCs at 10 h was quantified in CD1d^hi-B16 ± Gal injected with WT C57BL/6 or Jα18^−/− mice. All data in C and D are means ± SEM obtained from three mice per indicated time point in two independent experiments per each group.

DCs are required for inducing immunity to tumor/Gal. (A) 10 h after injecting CD1d^hi-B16/Gal tumor cells, CD11c⁺ and CD11c⁻ cells were isolated, and 3 × 10⁶ CD11c⁺ and 5 × 10⁶ CD11c⁻ cells were transferred to naive C57BL/6 mice; spleen cells from uninjected mice served as the negative control. 1 wk later, we measured CD8⁺ T cells specific for melanoma antigens (trp2 and gp100) by ELISPOT assay. (B) 1 d after injecting CD1d^hi-B16/Gal i.v. into CD11c-DTR transgenic mice, CD11c⁺ DCs were depleted with DT. 7 d later, we analyzed antigen-specific T cell responses to trp2 or gp100 peptides by ELISPOT assays. All data are means obtained from two independent experiments (n = 5 per group). *, P < 0.05 for CD11c⁺ transferred mice versus others (left) and DT-treated CD11c-DTR mice versus immunized WT mice (right). SFC, spot-forming cells.

Vaccination with B16-GalCer tumor cells i.v. leads to T cell immunity to several defined melanoma differentiation antigens. (A) Mice were injected i.v. with 5 × 10⁵ B16, B16/Gal, CD1d^hi-B16, or CD1d^hi-B16/Gal. To monitor antigen-specific T cell responses 1 and 4 wk later, CD8⁺ T cells were positively selected from the spleen and were cultured with splenic CD11c⁺ DCs from naive mice for 36 h. The DCs had been cultured for 2 h in the presence or absence of 10 μM gp100_25–33, Trp2_180–188, Tyrp_455–463, Tyrp_522–529, or Dct_363–371 peptides. T cells responding to the DCs were detected by ELISPOT assay for IFN-γ production. All data are means ± SEM obtained from three independent experiments with two mice per group. *, P < 0.05 for B16/G versus CD1d^hi-B16/G. (B) As in A, but the immune responses were monitored with intracellular cytokine staining. CD8⁺ T cells were magnetically isolated from the spleen 7 d after i.v. CD1d^hi-B16/Gal. The CD8⁺ T cells were co-cultured for 16 h with splenic CD11c⁺ DCs from naive mice in the presence of brefeldin A, which had been pulsed with 10 μM Trp2_180–188 peptide for 2 h, for the last 10 h. Numbers indicate the percentage of total CD8⁺ cells. The data are representative of three independent experiments (n = 4 per group). (C) Mice were immunized with 5 × 10⁵ CD1d^hi-B16/Gal and depleted NK1.1⁺ cells by injections of 300 μg of anti-NK1.1(PK136) antibody per mouse at days −1, 3, and 5 (before vaccination), or days 3 and 5 (after vaccination). As in A, the TRP2-specific CD8⁺ T cell response was detected by ELISPOT assay for IFN-γ at day 7. Data are means ± SEM obtained from three independent experiments (n = 4 per group). **, P < 0.01.

DCs are less effective than tumor/Gal in presenting OVA antigen when injected via the i.v. route. (A) Mice were adoptively transferred with 2 × 10⁶ CFSE-labeled OT-I cells and were immunized the next day with EL4(OVA) or CD1d^hi-EL4(OVA) ± Gal. 3 d later, the animals were tested for proliferation of transferred OT-I cells as well as CD25 (top). Data are representative of two separate experiments (n = 2 per group). Different strains of C57BL/6 mice (bottom) were adoptively transferred with 2 × 10⁶ CFSE-labeled OT-I cells or OT-II T cells and then challenged with 2 × 10⁶ CD1d^hi-EL4(OVA)/Gal cells i.v. 3 d later, cell proliferation was monitored in the spleen by CFSE dilution (bottom). Naive indicates OT-I or OT-II cell-transferred, but not immunized, mice. The numbers indicate CD25⁺ (top) or CD25⁻ (bottom) differentiated OT-I cells. Data are representative of three separate experiments (n = 3 per group). (B) As in A, but now the immune response was measured in naive and immunized mice using H-2K^b/OVA tetramers 7 d later. Mice were injected i.v. with 2 × 10⁶ BM-derived DCs that had been pulsed with 1 μM OVA_257–264 peptide in the absence or presence of 100 ng/ml α-GalCer. The expansion of OVA-specific CD8⁺ T cells was compared with different forms of 2 × 10⁶ EL4 tumors that were loaded with OVA (see Materials and methods) in the absence or presence of α-GalCer. Data are representative of four separate experiments (n = 4 per group). (C) The expansion of OVA-specific CD8⁺ T cells was shown. Mice were immunized with 100 μg OVA peptide per mouse with or without 1 μg of free α-GalCer i.v. per mouse (reference 20). Data are representative of four separate experiments (n = 4 per group). Gated cells in B and C indicate CD8⁺ K^b/OVA_257–264⁺ double-positive cells using K^b/OVA_257–264 tetramer–PE and CD8-FITC. (D) As in B, but the immune response was measured with an intracellular cytokine staining. 7 d after immunization, spleen cells were cultured with or without OVA_257–264 peptide to stimulate OVA-specific CD8⁺ T cells for 6 h, and T cell production of IFN-γ was measured by intracellular cytokine staining. Data are means ± SEM obtained from five mice per group in three independent experiments. *, P < 0.05 for EL4(OVA)/G or CD1d^hi-EL4(OVA)/G versus the other groups and EL4(OVA)/G versus CD1d^hi-EL4(OVA)/G). (E) Jα18^−/−, CD4^−/−, TAP^−/−, or CD40^−/− mice were used as recipients for CD1d^hi-EL4(OVA)/Gal, and the immune responses were tested by intracellular cytokine staining in the absence or presence of OVA peptide 1 wk later. Data are means ± SEM obtained from five mice per group in three independent experiments. *, P < 0.05 for WT versus Jα18^−/−, TAP^−/−, or CD40^−/−; P > 0.05 for WT versus CD4^−/−.

In vivo maturation of DCs and cross-presentation of α-GalCer in DCs after administration of tumor/Gal. (A) Spleen cells were obtained at 0, 2, 4, and 6 h after administration of CD1d^hi-B16/Gal. Low density spleen cells were stained with anti-CD11c–PE and CD8α-FITC (to identify DCs and their CD8⁺ or CD8⁻ subsets) and with biotinylated isotype control or CD86 mAb, followed by streptavidin-allophycocyanin. (B) As in A, but DCs were tested in three different organs for CD86 expression 6 h after injecting the tumor cells indicated on the left. All data in A, B, and D are representative of two independent experiments (n = 2 per group). (C) Splenic DCs were analyzed by intracellular cytokine staining for IL-12p40 4 h after immunization with tumor/Gal, followed by 4 h of culture in brefeldin A. DCs were identified with CD11c-allophycocyanin and CD8α-FITC and were subsequently fixed and stained with PE-conjugated anti–IL-12p40 mAb. (D and E) DC maturation studies, as in A and C, were performed in WT and Jα18^−/− mice to show the NKT cell dependence. Numbers in C and E indicate the percentage of IL-12–producing cells in CD8⁺ or CD8⁻ subsets of CD11c⁺ DCs, and all data are representative of three independent experiments (n = 3 per group). (F) To evaluate the cross-presentation of α-GalCer derived from tumor/Gal, DCs or non-DCs were isolated by CD11c magnetic beads at 10 h after injecting 5 × 10⁵ CD1d^hi-B16/Gal into WT C57BL/6 or CD1d^−/− mice. 10⁵ APCs per well were co-cultured with 10⁵ liver MNCs per well from C57BL/6 or Jα18^−/− mice for 48 h. Then, the IFN-γ from the supernatants was measured by ELISA. All data are means ± SEM obtained from three independent experiments (n = 5 per group). *, P < 0.05 for WT versus Jα18^−/− for liver MNCs; **, P < 0.01 for WT versus CD1d^−/− for APCs.