Engagement of CD83 ligand induces prolonged expansion of CD8+ T cells and preferential enrichment for antigen specificity

Abstract

Following T-cell receptor and CD28 signaling, CD8+ T cells express a receptor for CD83, a molecule up-regulated on functionally mature dendritic cells. Although this expression pattern suggests that CD83 is involved in adaptive immunity, little is known about its function in the periphery, and the existence of its ligand on T cells is controversial. We demonstrate that the engagement of the CD83 ligand (CD83L) preferentially enriches and significantly amplifies the number of antigen-specific CD8+ T cells. Coengagement of the T-cell receptor, CD28, and CD83L supports priming of naive CD8+ T cells that retain antigen specificity and cytotoxic function for more than 6 months. Therefore, engagement of the CD83L provides a unique signal to activated CD8+ T cells that could be exploited to generate long-lived antigen-specific cytotoxic T cells for the treatment of cancer and infection.

Figure 1.

Construction, structure, and verification of dodecameric form of soluble CD83. (A) A dodecameric form of soluble CD83 (dodCD83) was constructed by fusing the CD83 extracellular portion, Fcγ portion of human IgG₁, and human IgA tailpiece. The primary structure of dodCD83 is shown as a monomeric form. (B) The predicted structure of dodCD83. (C) With or without reducing treatment by dithiothreitol, purified dodCD83 protein was subjected to SDS-PAGE, transferred to a PVDF membrane, and blotted with goat anti-human Ig (H+L) Ab. (D) Purified dodCD83 (100 μg) was loaded onto a size-exclusion HPLC column and fractionated based on size. Purified human IgM, IgG, or gel filtration standard was also run and used as a reference protein. (E) CHO/CD83 cells were preincubated with graded amounts of dodCD83 or control protein and subsequently stained with PE-labeled anti-CD83 mAb. (F) T-cell leukemic lines Jurkat and HPB-ALL and monocytic cell line U937 cells were analyzed for CD83L expression using dimeric and dodecameric of soluble CD83. Cells were incubated with either dodCD83 (open curve) or control dodecameric protein (filled curve), followed by incubation with PE-conjugated secondary antibody.

Figure 2.

Engagement of CD3 and CD28 induces CD83L expression on the surface of T lymphocytes. Peripheral T cells were stimulated and tested for CD83L expression by staining with either dodCD83 (open curve) or control dodecameric protein (filled curve). (A) Purified CD4⁺ and CD8⁺ T cells demonstrated CD83L expression when stimulated by anti-CD3 and anti-CD28, but not when stimulated with anti-CD3 alone. (B) CD83L expression is demonstrated on CD3⁺ T cells stimulated with allogeneic mature DCs. Expression is abrogated by incubation with anti-CD80 and anti-CD86 mAbs but not with isotype controls. (C) Purified CD4⁺ and CD8⁺ T cells demonstrated CD83L expression when stimulated with allogeneic mature DCs, but not when stimulated with immature DCs. (D) Purified CD8⁺ T cells were optimally stimulated with anti-CD3 and anti-CD28 for the indicated time periods and analyzed for CD83 counterreceptor expression using dodCD83. The CD8 dim and CD8 bright populations were defined by electronic gating after staining with anti-CD8 mAb. (E) CD45RA and CD45RO CD8⁺ T cells were optimally stimulated with anti-CD3 and anti-CD28 for 48 hours and analyzed for CD83 counterreceptor expression using dodCD83. Between each experiment, flow cytometry settings were held constant to allow for direct comparison. Since DC-stimulated T cells have higher autofluorescence than mAb-activated T cells, the control staining was shifted slightly to the right for DC-stimulated T cells. Also, with increasing time of stimulation, greater autofluorescence of activated T cells was observed. Figures are representative of multiple donors: in panel A 1 of 4 is shown, and in panels B-E 1 of 2 is shown.

Figure 3.

Expression profiles of A2, CD80, and CD83 molecules on K562-derived stable cell lines coexpressing A2 with either CD80, CD83, or both CD80 and CD83. K562-derived stable transfectants were analyzed by flow cytometry with specific antibody (open curve) and an isotype control (filled curve) coupled to the appropriate chromophore. Note that there is a dim expression of endogenous CD80 on K562 cells.

Figure 4.

CD83L signaling enhances the expansion of antigen-specific CD8⁺ T cells. HLA-A2⁺ CD8⁺ T cells were stimulated by MART1 (A) or Flu (B) peptide-pulsed APCs (APC/A2, APC/A2/CD80, APC/A2/CD83, or APC/A2/CD80/CD83). After 3 rounds of antigen stimulation and IL-2/IL-15 addition in between the stimulations, the cultures were stained with the relevant tetramer to determine percentage of peptide-specific T cells. The number of peptide-specific T cells was determined by calculating the product of the total number of T cells and the percentage of tetramer-staining cells. Representative results from 4 different donors are presented. (A) Compared with APC/A2, large increases in the percentage and number of antigen-specific CD8⁺ T cells were observed when T cells were stimulated by MART1-pulsed APC/A2/CD80 (percentage, P = .017; number, P = .002) and APC/A2/CD80/CD83 (percentage, P = .009; number, P < .001), but not by APC/A2/CD83 (percentage, P = .48; number, P = .30). When compared with APC/A2/CD80, APC/A2/CD80/CD83 generates a significant increase in the percentage and number of MART1-specific T cells (percentage, P = .012; number, P = .003). indicates A2; □, A2/CD80; , A2/CD83; ▪, A2/CD80/CD83. (B) Likewise, when APCs are pulsed with Flu peptide, the percentage and total number of antigen-specific T cells was increased when T cells were stimulated by APC/A2/CD80 (percentage, P = .009; number, P = .01) and APC/A2/CD80/CD83 (percentage, P = .009; number, P = .023), but not by or APC/A2/CD83 (percentage, P = .32; number, P = .32). When compared with APC/A2/CD80, APC/A2/CD80/CD83 generates a significant increase in the percentage and number of Flu-specific T cells (percentage, P = .01; number, P = .035). indicates A2; □, A2/CD80; , A2/CD83; ▪, A2/CD80/CD83.

Figure 5.

Effector function of MART1-specific CD8⁺ T cells. (A) MART1-specific cytotoxic CD8⁺ T cells generated by APC/A2/CD80/CD83 killed target cells in an HLA-A2-specific manner. Target cells (T2 and Jurkat cells transduced with vector or A2) were pulsed with MART1 peptide (▪) or control peptide (•). ¹¹LLFGYPVYV¹⁹ peptide of the TAX of HTLV-I was used as a negative control peptide. (B) Blocking experiments were performed using HLA-A2-specific mAb (BB7.2) and isotype control (mIgG2b). Peptide-pulsed T2 cells were used at an E/T ratio of 30:1. (C) MART1-specific cytotoxic CD8⁺ T cells generated by APC/A2/C80/CD83 secreted IFN-γ in an antigen-specific manner. Error bars represent standard deviation (SD).

Figure 6.

Stimulation with APCs that express CD83 enhances proliferation and inhibits apoptosis, permitting continued expansion of MART1-specific T cells beyond 6 weeks. (A) Following 5 rounds of stimulation with MART1 peptide-pulsed APC/A2/CD80, T-cell lines noted to cease expansion were split and subsequently stimulated with either peptide-pulsed APC/A2/CD80 (▪) or peptide-pulsed APC/A2/CD80/CD83 (○). Between the stimulations, IL-2 and IL-15 were added to the culture. Those T cells stimulated by peptide-pulsed APC/A2/CD80/CD83 demonstrated significant antigen-specific expansion, while those stimulated by peptide-pulsed APC/A2/CD80 did not (P = .006). (B) An increase in MART1-specific T-cell proliferation was determined by tetramer staining and BrdU incorporation. This was consistently observed in both donors tested at weeks 7 and 8. (C) Apoptosis was inhibited as determined by Annexin V staining. This was consistently observed in both donors tested at weeks 7 and 8. The proliferating (B) and apoptotic fractions (C) are shown as a percentage of MART1 tetramer-positive cells. Two experiments with different donors were performed.

Figure 7.

CD83L engagement supports sustained growth of antigen-specific CD8⁺ T cells. MART1-specific CD8⁺ T cells were generated by stimulation with MART1-pulsed APC/A2/CD80/CD83 every 7 to 14 days. The percentage increase or decrease in the number of cells was determined at each stimulation, and a fraction was subsequently stimulated and maintained in culture. Antigen specificity was demonstrated every 3 or 4 rounds of stimulation by tetramer analysis. (A) The predicted total number of CD8⁺ T cells generated over a 191-day culture period is shown. Arrow denotes removal of debris by Ficoll density gradient. (B) PE-conjugated MART1 tetramer staining versus CD8 staining is shown at 4 time points during the culture period. (C) Effector function was determined by cytotoxicity assay on day 190 using peptide-pulsed T2 cells as targets (▪ indicates MART1 peptide; •, HIV pol as control peptide).