Induction of tumor cell apoptosis is accompanied by peritumoral accumulation of DCs and up-regulation of lytic molecules on inflammatory cells. (A and B) Immunofluorescence triple labeling of cryostat sections of IMQ-treated BCC lesions with antipancytokeratin (TRITC), anti–HLA-DR (FITC), and either (A) anti-CD11c (APC) or (B) anti-CD123 (Cy5) revealed a close proximity of both DC populations with tumor cells. Arrows indicate double-positive DCs. (C) Untreated BCCs and (D) BCCs after 2 wk of IMQ treatment were stained with an antipancytokeratin antibody (TRITC) and TUNEL staining (FITC). Note the double-positive tumor cells occurring after 2 wk of IMQ treatment (arrows). The pictures are representative for all evaluated biopsies. The dotted lines point out the margins of BCCs. (E) Identification of IFN-α–producing pDCs in 2-wk–treated BCCs by triple stainings with anti– IFN-α (TRITC), anti–HLA-DR (FITC), and anti-CD123 (Cy5) mAbs. The arrow indicates a triple-positive cell. (F) Quantitative analysis of lytic molecules expressed by inflammatory cells of the peritumoral infiltrate. Immunofluorescence single stainings were performed using antiperforin, anti–granzyme B, anti-Fas ligand, and anti-TRAIL antibodies (all TRITC). Data are given as absolute numbers of positive cells ± SEM for the indicated markers.

IMQ treatment induces the up-regulation of perforin and granzyme B on mDCs in the peritumoral infiltrate. A detailed quantitative analysis of the expression pattern of (A) perforin and (C) granzyme B on leukocytes before, during, and after IMQ treatment was performed by immunofluorescence double stainings with antibodies against the indicated lytic molecules and lineage markers in all seven treated patients. The data are given as absolute numbers of double-positive cells ± SEM. Representative pictures of immunofluorescence triple stainings with anti-CD11c, anti–HLA-DR, and (B) antiperforin or (D) anti–granzyme B mAbs of the peritumoral infiltrate after 2 wk of treatment with IMQ. (E) Triple immunofluorescence labeling of BCCs upon IMQ treatment with anti-CD11c (PE), anti–HLA-DR (APC), and anti-iNOS (FITC) or anti–TNF-α (FITC) mAbs identifies iNOS and TNF-α to be produced by CD11c⁺HLA-DR⁺ cells (arrows).

Secretion of perforin and granzyme B by mDCs upon TLR7 and TLR8 stimulation. (A) Perforin and (B) granzyme B protein levels were determined by ELISA in media conditioned by purified blood–derived mDCs and pDCs after 12 h of stimulation with TLR7 and TLR8 agonists. Media conditioned by 12-h PMA/ionomycin-stimulated T cells were used as positive controls. Data are given as means ± SEM from two independent experiments with 2 × 10⁵ cells in each setting.

Complete regression of superficial BCC after a 6-wk treatment period with IMQ. IMQ topically applied five times a week for a period of 6 wk led to a local inflammatory response, which resulted in a complete clinical and histopathological tumor clearance in all patients treated. The clinical pictures are representative for all patients (n = 7) treated with IMQ.

Upon IMQ treatment, TRAIL R1⁺ BCC islets are surrounded by TRAIL-expressing pDCs. (A) Detailed quantitative in situ analysis of the expression pattern of TRAIL on leukocytes before, during, and after IMQ treatment was performed by immunofluorescence double stainings with an anti-TRAIL mAb and lineage markers in all patients treated. The data are given as absolute numbers of double-positive cells ± SEM. (B) A representative view of anti-CD123 (FITC), anti–HLA-DR (APC), and anti-TRAIL (TRITC) triple stainings of a BCC lesion after 2 wk of IMQ treatment identifies pDCs as TRAIL-expressing cells. Immunofluorescence triple labeling of (C) untreated and (D) IMQ-treated BCCs with antipancytokeratin (TRITC), anti-TRAIL R1 (A488), and anti-CD123 (Cy5) reveals TRAIL R1⁺ BCC cells (C and D) surrounded by CD123⁺ cells (arrows; D).

Peripheral blood–derived DCs express lytic molecules after TLR7/8 stimulation. (A) Characterization of isolated peripheral blood–derived pDCs and mDCs as HLA-DR⁺BDCA-2⁺ and HLA-DR⁺CD11c⁺, respectively, and lineage-negative cell populations by flow cytometry. (B and C) Analysis of intracellular expression levels of perforin and granzyme B of isolated mDCs before and after culture with TLR7/8 agonists by flow cytometry. (B) Representative dot plots of mDCs show double-positive cells for perforin (FITC) and granzyme B (PE) after stimulation with TLR7, TLR8, and TLR7/8 agonists. (C) Quantitative analysis of granzyme B⁺ and perforin⁺ mDCs. Data are given as mean percentages of perforin- or granzyme B–positive mDCs ± SEM from three independent experiments. (D) Freshly isolated and TLR7/8-stimulated mDCs do not display membrane-bound TRAIL (A488). Representative histogram plots from three independent experiments show the anti-TRAIL reactivity (bold line) and the matching isotype control (dashed line). (E and F) Determination of intracellular expression levels of perforin and granzyme B on isolated pDCs before and after stimulation with TLR agonists. (E) Representative dot plots of pDCs show a decrease of granzyme B–positive cells after stimulation with TLR7 and TLR7/8 agonists. (F) Quantitative analysis of pDCs expressing perforin and granzyme B. Data are given as mean percentages of perforin- and granzyme B–positive cells ± SEM from three independent experiments. (G) TLR7/8-stimulated, but not freshly isolated, pDCs display membrane-bound TRAIL (A488). Representative histogram plots from three different experiments show the anti-TRAIL reactivity (bold line) and the matching isotype control (dashed line). Dead cells were excluded by gating out propidium iodide–positive cells in all experiments.

Tumoricidal activity of blood-derived mDCs and pDCs after stimulation with TLR7/8 agonists. Unstimulated, TLR7/8- stimulated, or vehicle-stimulated (A) mDCs and (B) pDCs were cultured with K562 or Jurkat as target cells for a conventional 2-h europium-TDA release assay at the indicated effector/target cell ratios. Data represent the means of triplicate wells ± SEM from two independent experiments. (C and D) For inhibition of different cytotoxicity-inducing pathways, 12-h TLR7/8-stimulated mDCs and pDCs as effector cells were preincubated with 100 nM concanamycin A for 120 min, 5 μg/ml anti-TRAIL for 30 min, 5 μg/ml anti-Fas ligand for 30 min, or 5 μg/ml of an IgG1 isotype for 30 min. Target cells were (C) K562 for TLR7/8-stimulated mDCs and (D) Jurkat for TLR7/8-stimulated pDCs. Cytotoxicity was determined by a 2-h europium-TDA release assay at the indicated effector/target cell ratios. Data represent the means of triplicate wells ± SEM from two independent experiments.