Human blood myeloid and plasmacytoid dendritic cells cross activate each other and synergize in inducing NK cell cytotoxicity

Abstract

Human blood dendritic cells (DCs) hold great potential for use in anticancer immunotherapies. CD1c⁺ myeloid DCs and plasmacytoid DCs (pDCs) have been successfully utilized in clinical vaccination trials against melanoma. We hypothesize that combining both DC subsets in a single vaccine can further improve vaccine efficacy. Here, we have determined the potential synergy between the two subsets in vitro on the level of maturation, cytokine expression, and effector cell induction. Toll-like receptor (TLR) stimulation of CD1c⁺ DCs induced cross-activation of immature pDCs and vice versa. When both subsets were stimulated together using TLR agonists, CD86 expression on pDCs was increased and higher levels of interferon (IFN)-α were produced by DC co-cultures. Although the two subsets did not display any synergistic effect on naive CD4⁺ and CD8⁺ T cell polarization, CD1c⁺ DCs and pDCs were able to complement each other's induction of other immune effector cells. The mere presence of pDCs in DC co-cultures promoted plasma cell differentiation from activated autologous B cells. Similarly, CD1c⁺ DCs, alone or in co-cultures, induced high levels of IFN-γ from allogeneic peripheral blood lymphocytes or activated autologous natural killer (NK) cells. Both CD1c⁺ DCs and pDCs could enhance NK cell cytotoxicity, and interestingly DC co-cultures further enhanced NK cell-mediated killing of an NK-resistant tumor cell line. These results indicate that co-application of human blood DC subsets could render DC-based anticancer vaccines more efficacious.

Keywords: Adaptive immunity; NK cells; cancer immunotherapy; crosstalk; myeloid DCs; plasmacytoid DCs.

Figure 1.

Human CD1c⁺ DCs and pDCs cross activate each other. CD1c⁺ DCs and pDCs were cultured overnight, either separately or together, in the presence of indicated stimuli. Maturation was checked on the specific DC subsets within the co-culture. (A) Relative expression of CD86, CD80, PD-L1, and MHC class I and II on pDCs in geometric MFI normalized to pDCs cultured alone with IL-3. (B) Relative expression of CD86, CD80, PD-L1 and MHC class I and II on CD1c⁺ DCs in geometric MFI normalized to CD1c⁺ DCs cultured alone with GM-CSF. (C) IL-12p70, IFN-α, and TNF-α in supernatants of overnight cultures were analyzed by ELISA. Black bars, pDCs; white bars, CD1c⁺ DCs; gray bars, CD1c⁺ DCs and pDCs. Results are the mean ± SEM of at least five (A, B) or at least three (C) independent experiments. Significance was determined by repeated measures one-way ANOVA, followed by a post-hoc Dunnett's test (*p < 0.05; **p < 0.01; ***p < 0.001). Only conditions that are significantly different when compared with extra control groups (see Fig. S1) are denoted in the graphs.

Figure 2.

Co-cultures of human CD1c⁺ DCs and pDCs induce effector cell responses in a mixed lymphocyte reaction. After CD1c⁺ DCs and/or pDCs were stimulated overnight with indicated stimuli, they were cultured with allogeneic PBLs at a 10:1 PBL:DC ratio. Proliferation was determined at day 4 by [³H]-thymidine incorporation. IFN-γ in supernatants was analyzed at day 3 by ELISA. Black bars, pDCs with PBLs; white bars, CD1c⁺ DCs with PBLs; gray bars, CD1c⁺ DCs and pDCs with PBLs; striped bars, PBLs only. Results are the mean ± SEM of five (first panel) or three (second panel) independent experiments.

Figure 3.

Co-cultures of human CD1c⁺ DCs and pDCs activate and polarize naive CD4⁺ and CD8⁺ T cells. After CD1c⁺ DCs and/or pDCs were stimulated overnight with indicated stimuli, they were cultured with SEB and allogeneic naive CD4⁺ (A) or CD8⁺ (B) T cells at a 4:1 T cell:DC ratio. (A) Resting CD4⁺ T cells were restimulated with PMA/ionomycin in the presence of brefeldin A and intracellular IFN-γ expression was measured by flow cytometry. In parallel, resting CD4⁺ T cells were restimulated with anti-CD3/CD28 beads and after 24 h, IFN-γ and IL-10 in supernatants were analyzed by ELISA. Black bars, pDCs with CD4⁺ T cells; white bars, CD1c⁺ DCs with CD4⁺ T cells; gray bars, CD1c⁺ DCs and pDCs with CD4⁺ T cells. (B) Intracellular granzyme B expression by resting CD8⁺ T cells was measured by flow cytometry. Resting CD8⁺ T cells were restimulated with PMA/ionomycin in the presence of brefeldin A and intracellular IFN-γ and TNF-α expression was measured by flow cytometry. Black bars, pDCs with CD8⁺ T cells; white bars, CD1c⁺ DCs with CD8⁺ T cells; gray bars, CD1c⁺ DCs and pDCs with CD8⁺ T cells. Results are the mean ± SEM of at least five (A, first panel), four (A, second and third panels; B, second and third panels), or three (B, first panel) independent experiments.

Figure 4.

Co-cultures of human CD1c⁺ DCs and pDCs induce plasma cell differentiation. CD1c⁺ DCs and/or pDCs were cultured with autologous B cells in the presence of indicated stimuli. After 7 days, plasma cell differentiation was determined by flow cytometry. (A) Plasma cells were defined as CD38^highCD20^low expressing cells. (B) The percentages of plasma cells in co-cultures with DCs. Black bars, pDCs with B cells; white bars, CD1c⁺ DCs with B cells; gray bars, CD1c⁺ DCs and pDCs with B cells; striped bars, B cells only. Results are the representative or mean ± SEM of three independent experiments. Significance was determined by repeated measures one-way ANOVA, followed by a post-hoc Bonferroni's test (*p < 0.05; **p < 0.01; ***p < 0.001).

Figure 5.

Co-cultures of human CD1c⁺ DCs and pDCs enhance NK cell responses. CD1c⁺ DCs and/or pDCs were cultured overnight with indicated stimuli. (A) Stimulated DCs were subsequently cultured with autologous NK cells at a 1:1 ratio and with IL-2. IFN-γ in 24 h supernatants was analyzed by ELISA. (B–E) Stimulated DCs were cultured with autologous NK cells at a 5:1 NK cell:DC ratio for 24 h. Labeled K562 or Daudi tumor cells were added for the last 4 h of the culture at 5:1 NK cell:target cell ratio. Specific target cell killing was determined by flow cytometry. Black bars, pDCs with NK cells; white bars, CD1c⁺ DCs with NK cells; gray bars, CD1c⁺ DCs and pDCs with NK cells; striped bars, NK cells only. Results are the mean ± SEM of at least four (A, B), five (C), or three (D, E) independent experiments. Significance in (C) was determined by repeated measures one-way ANOVA, followed by a post-hoc Dunnett's test (*p < 0.05; **p < 0.01; ***p < 0.001), and only conditions showing significant differences in all comparisons, are denoted in the graph. Significant differences among experimental conditions in (D) were determined by one-way ANOVA followed by a post-hoc Bonferroni's test (*p < 0.05; **p < 0.01; ***p < 0.001). Significant differences among experimental conditions in (E) were determined by two-tailed Student's t-test on paired samples (*p < 0.05; **p < 0.01; ***p < 0.001).