Cross-presentation of viral antigens in dribbles leads to efficient activation of virus-specific human memory T cells

Abstract

Background: Autophagy regulates innate and adaptive immune responses to pathogens and tumors. We have reported that autophagosomes derived from tumor cells after proteasome inhibition, DRibbles (Defective ribosomal products in blebs), were excellent sources of antigens for efficient cross priming of tumor-specific CD8⁺ T cells, which mediated regression of established tumors in mice. But the activity of DRibbles in human has not been reported.

Methods: DRibbles or cell lysates derived from HEK293T or UbiLT3 cell lines expressing cytomegalovirus (CMV) pp65 protein or transfected with a plasmid encoding dominant HLA-A2 restricted CMV, Epstein-Barr virus (EBV), and Influenza (Flu) epitopes (CEF) were loaded onto human monocytes or PBMCs and the response of human CMV pp65 or CEF antigen-specific CD4⁺ and CD8⁺ memory T cells was detected by intracellular staining. The effect of cytokines (GM-CSF, IL-4, IL-12, TNF-α, IFN-α and IFN-γ) TLR agonists (Lipopolysaccharide, Polyinosinic-polycytidylic acid (poly(I:C), M52-CpG, R848, TLR2 ligand) and CD40 ligand on the cross-presentation of antigens contained in DRibbles or cell lysates was explored.

Results: In this study we showed that purified monocytes, or human PBMCs, loaded with DRibbles isolated from cells expressing CMV or CEF epitopes, could activate CMV- or CEF-specific memory T cells. DRibbles were significantly more efficient at stimulating CD8⁺ memory T cells compared to cell lysates expressing the same antigenic epitopes. We optimized the conditions for T-cell activation and IFN-γ production following direct loading of DRibbles onto PBMCs. We found that the addition of Poly(I:C), CD40 ligand, and GM-CSF to the PBMCs together with DRibbles significantly increased the level of CD8⁺ T cell responses.

Conclusions: DRibbles containing specific viral antigens are an efficient ex vivo activator of human antigen-specific memory T cells specific for those antigens. This function could be enhanced by combining with Poly(I:C), CD40 ligand, and GM-CSF. This study provides proof-of-concept for applying this strategy to activate memory T cells against other antigens, including tumor-specific T cells ex vivo for immunological monitoring and adoptive immunotherapy, and in vivo as vaccines for patients with cancer.

Figures

Figure 1

The kinetics of CD4 and CD8 T-cell activation following exposure to DRibble-pulsed APC. (A,B) PBMCs were separated into monocytes and lymphocytes by Elutra apheresis. Monocytes were pulsed with CEF DRibbles for 6 hours. After that time autologous T cells were added into wells at a ratio of 5:1 (T cells:monocytes). Brefeldin A was added to the culture after the time periods specified in the figure. All cells were harvested at the same time and prepared for flow cytometry. (C,D) PBMCs (monocytes and T cells) were cultured with pp65 DRibbles and then Brefeldin A was added to the culture after the time periods specified in the figure. Cells were harvested at the same time and prepared for flow cytometry.

Figure 2

DRibbles were efficient antigen carriers for the activation of human CD8+ T cells and CD4+ T cells. PBMCs were separated according to their density into monocytes and lymphocytes by Elutra apheresis. DRibbles were collected from HEK 293 T cells that expressed E6E7 protein or CEF protein. Monocytes were loaded with 25ug/ml CEF DRibbles or 25ug/ml control E6E7 DRibbles. After 6 hours, lymphocytes were added. Activation of T cells was assessed by determining the percentage of IFN-γ+CD8+ cells (A) and IFN-γ+CD4+ cells (B) detected by ICS. The mean ± SEM of 3 separate experiments from the same donor are shown. CD8+(C) and CD4+ T cell responses (D) against UbiLT3 pp65 DRibbles or control UbiLT3 GFP DRibbles and CMV pp65 protein were analyzed in frozen-thawed PBMCs from 24 subjects. (E) shows the representative dot plot from one of the donors.

Figure 3

CD8+ T cell response can be enhanced by adding Poly (I:C) and GM CSF/IL-4 to DRibble-pulsed PBMCs. Poly (I:C) and CD40L were added with UbiLT3 pp65 DRibbles (25ug/ml) to rested PBMCs. The mean ± SEM of 3 separate experiments are shown. (A) represents the percentage of IFN-γ producing CD8+ T cell and (B) represents the CD4+ T cell response. (C) PBMCs were cultured with or without GM-CSF and IL-4 for 12 hours, then Poly (I:C) and CD40L were added to PBMCs with UbiLT3 pp65 DRibbles (25ug/ml). The mean ± SEM of 3 separate experiments are shown.

Figure 4

Antigen-specific CD8+ and CD4+ T cells response following stimulation by DRibbles or cell lysates. (A) The ubiquitinated proteins and pp65 proteins contained in the cell lysate and DRibbles were examined by western blot. Lysates and DRibbles extracted from cell lines treated with bortezomib contained more ubiquitinated proteins than those from non-treated cells (left). The pp65 protein was detected in the lysates and DRibbles extracted from the UbiLT3 pp65 cell line. The pp65 DRibbles collected from the bortezomib-treated UbiLT3 pp65 cell line contained more 148Kd pp65 protein compared with the other groups (right). Data are representative of 3 independent experiments. (B,C) PBMCs were stimulated by DRibbles and cell lysates at the following doses: 3ug/ml, 10ug/ml and 25ug/ml. The percentage of IFN-γ+CD8+(B) and IFN-γ+CD4+ cells (C) were calculated by flow cytometry. Percentages of IFN-γ+ T cells are shown as mean ± SEM. Data are representative of results from 3 independent experiments. (D) Dot plots at the antigen dose of 25ug/ml.