Modulation of human memory T-cell function by different antigen-presenting cells.


Dendritic cells (DCs) are the most potent type of antigen-presenting cell (APC) to stimulate T cells [1, 2]. However, the modulation of antigen-specific T-cell responses against a variety of pathogens using different types of APCs, such as DCs and monocytes, has never been systematically compared. We have therefore performed a comprehensive analysis of the quantity and quality of memory CD4 T-cell responses directed against a variety of antigens including peptide-, protein-, viral-, fungaland bacterial-derived antigens following stimulation with antigenloaded DCs or monocytes. We compared antigen-specific memory T-cell responses following stimulation of total blood mononuclear cells versus stimulation with cultured DCs (termed MoDCs) loaded with antigens derived from peptides (Supporting Information Table 1), proteins (adenovirus hexon protein, PPDT, tetanus toxoid), whole virus particles (adenovirus serotype 5), heat-inactivated Gram-positive bacteria (Staphylococcus aureus and Streptococcus pneumoniae), Gram-negative bacteria (Pseudomonas aeruginosa, Klebsiella pneumonia, Salmonella typhi and Escherichia coli), and the two forms (unicellular and multicellular) of eukaryotic pathogens, i.e. Candida albicans yeast and hyphae. The frequency of antigen-specific memory CD4 T cells was determined using polychromatic flow cytometry through the quantification of the CD4 T-cell populations producing TNF-a, IFN-g, IL-2 and IL-17. Previous studies indicated that the detection of these cytokines in CD4 T-cell populations represents a reliable assessment of circulating antigen-specific CD4 T cells [3, 4]. We first performed a series of preliminary experiments showing that (i) DCs did not release cytokines in the presence of brefeldin A, which may interfere with CD4 T-cell responses (Supporting Information Fig. 1A and 2); (ii) unloaded TLR-exposed DCs did not trigger non-specific autologous CD4 T-cell responses (Supporting Information Fig. 1B) and (iii) antigen-exposed DCs did not trigger naı̈ve CD45RA/CCR7 /CD62LCD4 T cells (CD31-positive or negative) to any stimuli tested (Supporting Information Fig. 1C). The flow cytometric cytokine profiles of antigen-specific memory CD4 T cells producing IL-2 and IFN-g are shown in a representative donor (Fig. 1A). The cumulative data were generated from cells isolated from 14 healthy volunteers and 147 antigen-specific memory CD4 T-cell responses were assessed. The evaluation of the CD4 T-cell frequencies was determined by the capacity to produce either individual cytokines (TNF-a or IFN-g or IL-2) (Supporting Information Fig. 2) or any cytokine (total responses; TNF-a and/or IL-2 and/or IFN-g; Fig. 1B). No significant differences in the frequencies of peptidespecific (cytomegalovirus-, Epstein Barr virusand adenovirus-type 5-derived peptides) cytokine-producing CD4 T cells were observed between the two modes of stimulation while significant differences (po0.005) in the frequency (ranging from three to eight-fold) of individual (Supporting Information Fig. 2) or total cytokines (Fig. 1B) producing CD4 T cells were observed following stimulation with virus, protein, Gram-positiveand C. albicans-derived antigens. No significant differences (in the range of onefold) in the percentage of IL-17-producing CD4 T cells were observed and the addition of anti-CD28 antibody during direct PBMC stimulation did not influence CD4 T-cell responses (Supporting Information Fig. 1D). We then determined among the subjects with positive MoDC-stimulated CD4 T-cell responses, the capacity of direct stimulation of blood mononuclear cells to trigger a positive response (positive responses are defined in the Methods–See Supporting Information). Consistent with the results shown in Fig. 1B, the proportion of peptide-specific CD4 T-cell responses scored as positive was not significantly affected. However, 30–60% of positive responses detected following MoDC stimulation with proteinor virus-derived antigens was scored as negative using total blood mononuclear cells (Fig. 1C). The proportion of CD4 T-cell responses to bacteria and C. albicans was unaffected mainly due to their capacity to produce TNF-a. However, both IL-2 and IFN-g T-cell responses were significantly reduced (po0.001) in blood mononuclear cells stimulation (Fig. 1C). Total blood mononuclear cells, freshly isolated monocytes and MoDCs were generated from the same individuals and tested for the stimulation of antigenspecific memory CD4 T-cell responses against a panel of antigens. Representative flow cytometric profiles of antigen-specific cytokine-producing CD4 T cells are shown in Supporting Information Fig. 3. Cumulative data indicated that loaded freshly isolated monocytes showed intermediate potency, i.e. superior to total blood mononuclear cells but inferior to MoDCs (Fig. 1D) in eliciting antigenspecific memory CD4 T-cell responses. When antigen-loaded MoDCs were used, the common change observed in These are not the final page numbers

DOI: 10.1002/eji.201142094

3 Figures and Tables

Cite this paper

@article{Perreau2012ModulationOH, title={Modulation of human memory T-cell function by different antigen-presenting cells.}, author={Matthieu Perreau and Hugh Welles and Alexandre Harari and Thierry F Calandra and Thierry W. Roger and Giuseppe Pantaleo}, journal={European journal of immunology}, year={2012}, volume={42 3}, pages={799-802} }