Cross-priming of breast cancer specific CTLs. HLA-A*0201+ CTLs were generated as described in Figure 1, except that DCs were loaded with (panel a) HLA-A*0201- T47D or (panels b and c) Hs578T breast cancer cells, and CTL activity was tested against HLA-A*0201+ MCF7 breast cancer cells, HLA-A*0201+ Me275 melanoma cells and NK cell sensitive K562 cells. (a) DCs loaded with killed HLA-A*0201- T47D breast cancer cells induced CTLs that were able to lyse HLA-A*0201+ MCF7 breast cancer cells but not HLA-A*0201+ Me275 melanoma cells or NK cell sensitive K562 cells. The y-axis shows the specific lysis, and the x-axis shows the different E:T ratios. Data from four independent experiments are indicated by color code. A nonparametric Kruskal-Wallis comparison between the killing of three targets at different E:T ratios was conducted. (b) A similar experiment as is summarized in panel a, except CD8+ T cells were primed with DCs loaded with killed Hs578T cells; shown are representatives of three experiments. Values are expressed as average and standard deviation of triplicate wells. (c) CD8+ T cells cultured either with cytokines only (no DCs), with unloaded DCs (UL), or with DCs loaded with killed Hs578T breast cancer cells (Hs578T) were co-cultured for 24 hours with HLA-A*0201+ MCF7 breast cancer cells in triplicates. Survival of MCF7 cells (x-axis) was calculated by Trypan blue exclusion. In each case, shown are the results of one representative experiment from two performed. CTL, cytotoxic T lymphocyte; DC, dendritic cell; E:T, effector:target; NK, natural killer.