Short-term cultured, interleukin-15 differentiated dendritic cells have potent immunostimulatory properties

Abstract

Background: Optimization of the current dendritic cell (DC) culture protocol in order to promote the therapeutic efficacy of DC-based immunotherapy is warranted. Alternative differentiation of monocyte-derived DCs using granulocyte macrophage colony-stimulating factor (GM-CSF) and interleukin (IL)-15 has been propagated as an attractive strategy in that regard. The applicability of these so-called IL-15 DCs has not yet been firmly established. We therefore developed a novel pre-clinical approach for the generation of IL-15 DCs with potent immunostimulatory properties.

Methods: Human CD14+ monocytes were differentiated with GM-CSF and IL-15 into immature DCs. Monocyte-derived DCs, conventionally differentiated in the presence of GM-CSF and IL-4, served as control. Subsequent maturation of IL-15 DCs was induced using two clinical grade maturation protocols: (i) a classic combination of pro-inflammatory cytokines (tumor necrosis factor-alpha, IL-1beta, IL-6, prostaglandin E2) and (ii) a Toll-like receptor (TLR)7/8 agonist-based cocktail (R-848, interferon-gamma, TNF-alpha and prostaglandin E2). In addition, both short-term (2-3 days) and long-term (6-7 days) DC culture protocols were compared. The different DC populations were characterized with respect to their phenotypic profile, migratory properties, cytokine production and T cell stimulation capacity.

Results: The use of a TLR7/8 agonist-based cocktail resulted in a more optimal maturation of IL-15 DCs, as reflected by the higher phenotypic expression of CD83 and costimulatory molecules (CD70, CD80, CD86). The functional superiority of TLR7/8-activated IL-15 DCs over conventionally matured IL-15 DCs was evidenced by their (i) higher migratory potential, (ii) advantageous cytokine secretion profile (interferon-gamma, IL-12p70) and (iii) superior capacity to stimulate autologous, antigen-specific T cell responses after passive peptide pulsing. Aside from a less pronounced production of bioactive IL-12p70, short-term versus long-term culture of TLR7/8-activated IL-15 DCs resulted in a migratory profile and T cell stimulation capacity that was in favour of short-term DC culture. In addition, we demonstrate that mRNA electroporation serves as an efficient antigen loading strategy of IL-15 DCs.

Conclusions: Here we show that short-term cultured and TLR7/8-activated IL-15 DCs fulfill all pre-clinical prerequisites of immunostimulatory DCs. The results of the present study might pave the way for the implementation of IL-15 DCs in immunotherapy protocols.

Figure 1

Phenotypic characteristics of immature IL-15 DCs. Immature DCs were analyzed by flow cytometry for the expression of CD1a, CD14, CD56, CD80, CD207 (Langerin) and CD209 (DC-SIGN). The histograms represent the expression of the indicated cell surface antigens (bold-line histograms) and the corresponding isotype controls (grey-filled histograms). The mean ± SEM percentage of positive cells (%) and delta MFI ± SEM (ΔMFI) were calculated as specified in the "Methods" section (n = 3-6). (a) Phenotype of monocyte-derived DCs generated in the presence of GM-CSF + IL-15 and harvested at the immature stage 2-3 days after initiation of the DC culture. (b) Corresponding phenotypic profile of conventional immature DCs, differentiated in the presence of GM-CSF + IL-4.

Figure 2

Phenotypic characteristics of mature IL-15 DCs. Immunophenotypic expression of CD40, CD70, CD80, CD83, CD86 and CD209 (DC-SIGN) by (a) short-term cultured IL-15 DCs, (b) long-term cultured IL-15 DCs and (c) conventional IL-4 DCs after maturation induction with either a classic maturation cocktail (cc-mDC; dashed-line histograms) or a TLR7/8 ligand-containing mixture (TLR-mDC; bold-line histograms). Isotype controls are represented by the grey-filled histograms. A detailed overview of the flow cytometry data of mature DCs is provided in "Additional file 1" (n = 4).

Figure 3

Mannose receptor-mediated endocytosis of FITC-dextran particles. Histogram overlays depicting the in vitro uptake of FITC-dextran molecules by immature DCs, respectively short-term cultured IL-15 DCs (left), long-term cultured IL-15 DCs (middle) and control IL-4 DCs (right). The FITC-dextran endocytosis at 37°C (bold-line histograms) is compared to the non-specific fluorescence at 4°C (dashed-line histograms) and to the autofluorescence from unlabeled samples (grey-filled histograms), as described in "Methods". The uptake of FITC-dextran was quantified as mean ± SEM percentage of FITC-dextran positive cells (%) and as delta MFI ± SEM (ΔMFI), which was calculated by subtracting the MFI value of the non-specific FITC-dextran uptake at 4°C from the MFI value obtained at 37°C (n = 3).

Figure 4

CCR7 expression and migratory capacity. Histogram overlays comparing the CCR7 expression on (a) short-term cultured and (b) long-term cultured IL-15 DCs, either at the immature stage (iDC) or at the mature stage (cc-mDC: + TNF-α, IL1β, IL-6 and PGE₂ for the last 24 hours; TLR-mDC: + R-848, IFN-γ, TNF-α and PGE₂ for the last 24 hours). Bold-line histograms represent the CCR7-specific staining, whereas the corresponding isotype controls are indicated by grey-filled histograms (n = 3). (c) Migration of the indicated DC subsets towards CCL21 in a Transwell chemotaxis assay. The mean ± SEM percentages of migrated cells after 180 min were calculated according to the formula specified in "Methods" (n = 3-6; *, P = 0.01). The values shown in the grey bars represent the cell viabilities of the different DC subsets (mean ± SEM; n = 3-6).

Figure 5

IL-12p70 production following CD40 ligation ("signal-3 assay"). Dendritic cells were differentiated in the presence of GM-CSF + IL-4 for 6 days (control IL-4 DCs), or in the presence of GM-CSF + IL-15 for 2 days (short-term IL-15 DCs) or 6 days (long-term IL-15 DCs). Dendritic cell maturation was induced by addition of two different maturation cocktails 24 hr prior to DC harvest (cc-mDC: TNF-α, IL1β, IL-6 and PGE₂; TLR-mDC: R-848, IFN-γ, TNF-α and PGE₂). Production of the T_h1-polarizing cytokine IL-12p70 was assessed by ELISA after a 24-hr coculture of mDCs and CD40L-expressing 3T3 fibroblasts, as specified in "Methods". Results are shown from 3-9 independent experiments, each symbol expressing the mean of triplicate ELISA values obtained from one individual donor. The horizontal bars represent the mean IL-12p70 production in pg/mL per condition (*, P = 0.02).

Figure 6

Induction of viral antigen-specific CD8⁺ T cell responses. As described previously, short-term and long-term cultured IL-15 DCs were matured using two different maturation cocktails (cc-mDC: TNF-α, IL1β, IL-6 and PGE₂; TLR-mDC: R-848, IFN-γ, TNF-α and PGE₂). Conventionally matured IL-4 DCs were used as a control (control IL-4 DCs). The mDCs were harvested, pulsed with a pool of cytomegalovirus-, Epstein-Barr virus- and influenza a virus (CEF)-derived peptides, and cocultured with autologous PBLs for 7 days. Viral antigen-specific CD8⁺ T cell responses were determined after this 7-day period and a short restimulation with the CEF peptide pool (CEF; filled bars). As specified in "Methods", the antigen-specific production of IFN-γ was assessed using two techniques: (a) ELISA to detect the amount of IFN-γ produced after restimulation (pg/mL) and (b) ICS to determine the % of IFN-γ⁺ CD8⁺ T cells. The non-specific IFN-γ release in response to restimulation with an irrelevant HPV peptide pool is shown (HPV; unfilled bars). Results are expressed as mean ± SEM of three independent experiments (*, P = 0.03; **, P = 0.006; ***, P < 0.001).

Figure 7

mRNA transfectability of mature IL-15 DCs. Monocytes were cultured for 2 days with GM-CSF + IL-15, followed by a 24-hr incubation with a TLR7/8 agonist-based maturation cocktail (TLR-mDC). The resultant mDCs were harvested and electroporated with mRNA encoding the enhanced green fluorescent protein (eGFP). The green dots represent the mean ± SEM percentages of eGFP⁺ cells, as assessed by flow cytometry at different time points post-electroporation (4 hr, 24 hr, 48 hr). The insert shows a representative histogram overlay in which the flow cytometric eGFP expression 4 hr post-electroporation (green line histogram) is compared with the expression in a mock-electroporated negative control (grey-filled histogram). The values below indicate the delta MFI ± SEM of the eGFP expression (ΔMFI) and the mean ± SEM percentage of viable cells (%) at 4 hr, 24 hr and 48 hr following mRNA electrotransfection of IL-15 DCs (n = 5).

Figure 8

Induction of antigen-specific CD8⁺ T cell responses by mRNA-electroporated mature IL-15 DCs. Short-term cultured IL-15 DCs were matured with our TLR7/8 agonist-based maturation cocktail (TLR-mDC), electroporated with mRNA encoding the influenza virus matrix protein M1 and cocultured with autologous PBLs for 6 days. (a) The expansion of M1-tetramer binding CD8⁺ T cells was determined by flow cytometry. The lower dot plot represents the observed percentage of M1-tetramer⁺ CD8⁺ T cells in one representative donor (n = 4; mean ± SEM percentage of M1-tetramer⁺ CD8⁺ T cells: 4.4 ± 2.9). Correct positioning of the M1-tetramer⁺ CD8⁺ gate was defined by the respective negative control, as exemplified in the upper dot plot. (b) Simultaneously, a fraction of the PBL was harvested and stimulated with an irrelevant HLA-A*0201-restricted peptide (CEA) or rechallenged with the immunodominant influenza matrix protein (M1). The mean ± SEM percentage of antigen-specific IFN-γ⁺ CD8⁺ T cells was determined by ICS, as specified in the "Methods" section (n = 4; *, P = 0.03).