Immune tolerance after delivery of dying cells to dendritic cells in situ

Abstract

Peripheral immune tolerance is believed to be induced by the processing and presentation of self-tissues that die during physiologic tissue turnover. To examine the mechanism that mediates tolerance, we injected mice with dying syngeneic TAP(-/-) splenocytes loaded with small amounts of the protein antigen, ovalbumin (OVA). After ingestion and presentation of cell-associated OVA by the CD8(+) subset of dendritic cells in situ, large numbers of antigen-reactive, CD8(+) T cell receptor (TCR) transgenic T lymphocytes were driven into cell cycle, but then the T cells were deleted. The animals were also tolerant to challenge with OVA in complete Freund's adjuvant. An agonistic anti-CD40 monoclonal antibody was then administered together with the OVA-loaded splenocytes, so that the dendritic cells in the recipient mice would mature. In contrast to observations made in the steady state, the antigen-reactive T cells expanded in numbers for 1-2 wk and produced large amounts of interleukin 2 and interferon gamma, while the animals retained responsiveness to antigen rechallenge. The specific tolerance that develops when dendritic cells process self tissues in the steady state should prevent or reduce the development of autoimmunity when dying cells are subsequently processed during infection.

Figure 1.

Spleen DCs in the steady state capture apoptotic cells and present cell-associated OVA to CD8⁺ OT-I T cells in a TAP-dependent manner. (A) Uptake (arrows) by the CD8⁺ subset of CD11c⁺ DCs, 3 h after injection of CFSE-labeled TAP^−/− splenocytes, loaded with OVA during an osmotic shock. (B) As in A, but CD11c⁺ and CD11c⁻ fractions of spleen low density cells from B6 or TAP^−/− were added in graded doses to 2 × 10⁵ purified CD8⁺ OT-I T cells; T cell proliferation was measured by ³H-TdR uptake at 40–50 h. (C) Proliferation 3 × 10⁶ CFSE-labeled, CD8⁺ OT-I T cells (arrow) 2 d after intravenous injection of TAP^−/− splenocytes loaded with either HEL or OVA protein in spleen and subcutaneous lymph nodes. (D) As in C, but proliferation was monitored 3 d after different splenocytes were injected into the indicated recipients (top labels). (E) In response to OVA-loaded splenocytes, CFSE-labeled, OVA-specific OT-I T cells (Vα2Vβ5) proliferate at day 3, but not 2F5 influenza-specific T cells (Vβ11). A–D are representative of at least three experiments and E, two experiments.

Figure 2.

The early response, by surface phenotype and cytokine production, of OT-I T cells proliferating in response to dying cells in the absence or presence of anti-CD40 maturation. (A) Proliferation and surface phenotype of 3 × 10⁶ CFSE-labeled, CD45.1⁺ OT-I T cells 3 d after injection of OVA-loaded TAP^−/− splenocytes ± 100 μg anti-CD40 mAb intraperitoneally. Black and white arrows represent CD62L^high and CD62L^low populations, respectively. (B) As in A, but the day 3 responding splenocytes were restimulated with OVA peptide (SIINFEKL) for 4 h at 37°C with 5 μg/ml BFA. The cells were stained for CD8 and CD45.1 as in A, and intracellular IFN-γ and IL-2 identified with mAb after fixation and permeabilization. (C) Activation of OT-I cells by tissue culture in the absence or presence of restimulation by SIINFEKL pulsed splenocytes for 4 h. The data shown are representative of three experiments.

Figure 3.

Contrasting survival of OT-I T cells responding to DCs presenting OVA-loaded splenocytes in the absence or presence of anti-CD40 maturation. As in Fig. 2, the proportion of responding OT-I T cells (CD45.1⁺ CD8⁺) was measured at 3 d (black) or 10 d (gray) as a percentage of CD8⁺ cells in spleen and subcutaneous lymph nodes (means of 4–5 experiments) or as a proportion of total blood cells (2–3 experiments).

Figure 4.

Tolerance of mice given OVA-loaded TAP^−/− splenocytes to OVA challenge with CFA. (A) 2 d after injection of 0.3 × 10⁶ OT-I T cells, B6 mice were injected with 25 × 10⁶ OVA-loaded or unloaded TAP^−/− splenocytes intravenously. Some mice were given OVA-loaded TAP^−/− splenocytes once (day –6) and others twice (days –6 and 0). 9 d after the last injection of OVA-loaded TAP^−/− splenocytes, the mice were challenged with OVA in CFA, and total numbers of CD8⁺ CD45.1⁺ OT-I cells in draining (brachial) and distal (inguinal) nodes were measured by FACS^® (left) as in Fig. 3. Enriched CD8⁺ T cells using magnetic microbeads^® also were cocultured with the graded doses (inset) of SIINFEKL-pulsed spleen cells after γ-irradiation at 15 Gy. T cell proliferation was measured by ³H-TdR uptake at 38–48 h (middle), and IFN-γ–secreting cells by ELISPOT at 36 h (right). (B) Immunity develops if OVA-loaded splenocytes are presented along with anti-CD40 agonistic antibody, assessed as in B with ELISPOT (right panel) and T cell proliferation (38–48 h; left panel). The data in A and B are representative of three experiments.