Bidirectional interactions between antigen-bearing respiratory tract dendritic cells (DCs) and T cells precede the late phase reaction in experimental asthma: DC activation occurs in the airway mucosa but not in the lung parenchyma

Abstract

The airway mucosal response to allergen in asthma involves influx of activated T helper type 2 cells and eosinophils, transient airflow obstruction, and airways hyperresponsiveness (AHR). The mechanism(s) underlying transient T cell activation during this inflammatory response is unclear. We present evidence that this response is regulated via bidirectional interactions between airway mucosal dendritic cells (AMDC) and T memory cells. After aerosol challenge, resident AMDC acquire antigen and rapidly mature into potent antigen-presenting cells (APCs) after cognate interactions with T memory cells. This process is restricted to dendritic cells (DCs) in the mucosae of the conducting airways, and is not seen in peripheral lung. Within 24 h, antigen-bearing mature DCs disappear from the airway wall, leaving in their wake activated interleukin 2R+ T cells and AHR. Antigen-bearing activated DCs appear in regional lymph nodes at 24 h, suggesting onward migration from the airway. Transient up-regulation of CD86 on AMDC accompanies this process, which can be reproduced by coculture of resting AMDC with T memory cells plus antigen. The APC activity of AMDC can be partially inhibited by anti-CD86, suggesting that CD86 may play an active role in this process and/or is a surrogate for other relevant costimulators. These findings provide a plausible model for local T cell activation at the lesional site in asthma, and for the transient nature of this inflammatory response.

Figure 1.

In situ analysis of respiratory tract cell populations. (A–C) Tracheal epithelium stained for MHC class II. Sections were taken at (A) 0, (B) 2, and (C) 24 h after OVA aerosol challenge of primed animals. Images are projections along the z axis (“top” view) from stacks of 30 optical sections acquired at 2-μm increments at ×100 magnification. (D) Tracheal sections from primed OVA/ALOH animals after OVA aerosol exposure, stained for eosinophils, T cells, and DCs. Cells were quantitated by light (Epithelial DC, T-cells, Eos) confocal microscopy (Submucosal DC). Data are mean ± SD from three to five observations for each cell type. (E) Lung tissue was isolated at the indicated time points after challenge and single cell suspensions were prepared and immunostained. DCs were enumerated by flow cytometry and results were expressed as numbers recovered per gram wet weight (mean ± SD from three experiments).

Figure 2.

AHR after exposure to aerosolized OVA. Sensitized animals (n = 6 per group; ○) together with naive controls (•) were exposed to aerosolized OVA, and 24 h later MCh challenge was performed. The top shows a left shift in the airway dose response curve to MCh contrasted with no change in parenchymal responsiveness (bottom). Differences between test and control groups in this experiment (and in Figs. 3–6) were analyzed by Student's t test. *, P < 0.01; **, P < 0.05.

Figure 3.

Flow cytometric analysis of DCs after OVA aerosol challenge of OVA-primed animals. Data are representative of a series (A, E, and F) or mean ± SE from three to five experiments. (A) Flow cytometric analysis of total tracheal digest cells indicating gating for total DCs (R1) and subregions (R2–R4), based on SSC and MHC class II expression profiles. (B) Surface marker expression (after correction for background staining with isotype control) in total MHC class II⁺ DC populations (R1) in the trachea at 0 (open bars), 2 (solid bars), and 24 h (hatched bars) after challenge (*, P < 0.01 compared with 0 and 24 h). (C) Endocytic activity of R1 tracheal DCs at 2, 24, and 48 h after challenge as determined by 10-min dextran FITC uptake with trypan blue quenching at 37°C minus uptake at 4°C. Activity was significantly reduced relative to unexposed animals (0 h). *, P < 0.04; **, P < 0.01. (D) Relative numeric changes in tracheal DCs within each gating region at 2 (solid bars) and 24 h (hatched bars) after challenge compared with control immunized, nonaerosol challenged animals (open bars). (E and F) Expression of CD86 and MHC class II by R4 tracheal DCs at 2 h after challenge. Nonspecific binding of isotype control (E) versus staining with anti-CD86 on the MHC class II^hi subset. (G) Changes in CD86 expression on tracheal DCs in subregions R2–R4 as designated in A and D. Significance of increase relative to 0 h. *, P < 0.01; **, P < 0.001. (H) Flow cytometry analysis of DCs in PTLN after aerosol challenge of sensitized animals. The gating strategy comprised initial selection of total DCs (as per R1 in A), followed sequentially by gating for high MHC class II and low isotype control binding, and the same gates applied directly to obtain percentage of CD86 expression (see Fig. S5, available at http://www.jem.org/cgi/content/full.jem.20021328/DC1). MHC class II^hi CD86^hi DCs are shown as percentage of total DCs at 0 h compared with animals at 2, 24, and 48 h after challenge. >0 h: *, P < 0.01; **, P < 0.001. (J) Surface marker expressions by R1 DCs in the parenchymal lung at 0 (open bars), 2 (solid bars), and 24 h (hatched bars) after challenge. (K) Endocytic activity of R1 parenchymal lung DCs at 24 h after challenge compared to unexposed animals (t0), determined as per C.

Figure 4.

Antigen presentation by purified airway and lymph node DCs. Purified DCs from challenged animals were used to stimulate a CD4⁺ OVA-specific T cell line for 48 h. T cell stimulation (³H-DNA synthesis as CPM/culture) is shown as mean ± SE from three or more experiments (A and B) or from individual experiments (C). (A) OVA presentation by tracheal DCs isolated from OVA-immune animals 2 (•) and 24 h (▪) after challenge, or from naive animals 2 h (⋄) after challenge. 2 h > 0 h and 24 h: *, P < 0.05–0.01. (B) Total PTLN DCs were purified at 2 (•) or 24 h (□) after OVA aerosol challenge and additionally, sorted into MHC class II^hi (♦) or MHC class II^low (▵) expressing cells (as per Fig. 3) at the 24-h time point. OVA presentation by PTLN DCs is maximal at 24 h after aerosol challenge and was restricted to the MHC class II^hi population of DCs. (C) Tracheal DCs isolated 2 h after challenge were used to stimulate OVA-specific T cells in the absence (□) or presence (▪) of blocking antibody to CD86. In experiments 3 and 4, control cultures contained isotype control mAb versus medium only in control cultures in experiments 1 and 2. Data are normalized against the OVA-specific T cell response (³H-DNA synthesis at 72 h) in the absence of blocking antibody. <control: *, P < 0.05; **, P < 0.01.

Figure 5.

DC–T cell interactions in airway mucosa. (A–C) Confocal microscopy of tracheal sections 2 h after challenge of OVA-immune rats stained for MHC II (green, DCs) and TcRαβ (red, T cells). Dual staining contact points appear yellow. Each panel in the series represents a 1-μm optical section shown in ascending order, where individual DCs can be seen interacting with ≥1 T cells. (D) Individual T cells (n = 50–100) were scored for direct contact with DCs, and numbers in contact were expressed as a percentage of total T cells. Relative numbers of T cells at each time point were determined by factoring in data on absolute cell numbers per unit area of tissue from Fig. 1 D. *, P < 0.05 relative to 0 h, based on mean data from five experiments. (E) IL-2R⁺ T cells were initially enumerated as a percentage of total T cells by flow cytometric analysis of tracheal digests from exposed (solid bars) or control (open bars) animals in five experiments, and finally expressed as numbers of IL-2R⁺ T cells per unit area of epithelium by reference to data in Fig. 1 D, yielding five data points for each group at each time point. Data analysis used means derived from these data points. 24 h > 2 h in OVA-exposed animals: *, P < 0.05. (F) Tracheal DCs from naive animals were cultured alone or in the presence of either control or OVA-immune T cells at a density of 20:1 (T cell/DC). After the addition of 50 μg/ml soluble OVA, CD86 expression was examined (as in Fig. 3) at 0 and 3 or 4 h (data shown are for 3 h and are representative of a series of five experiments). Test culture (solid bar) > controls (open bars). *, P < 0.02 based on comparisons of means derived from five data points for each manipulation. Raw data from an exemplary experiment are illustrated in Fig. S9, available at http://www.jem.org/cgi/content/full.jem.20021328/DC1.

Figure 6.

Responses to OVA aerosol challenge in rats primed for Th1 immunity with OVA/complete Freund's adjuvant. DCs from tracheal digests 2 and 24 h after challenge were immunostained as per Fig. 3. (A) illustrates time-dependent changes in CD80 (▪) and CD86 (□) expression after aerosol exposure. CD80/CD86⁺ DCs are shown as a percentage of total MHC class II⁺ DCs (mean ± SE from three experiments). Controls and gating strategies are as described in Fig. 3. MCh challenge was performed on parallel groups (n = 5) of control (•) and OVA/complete Freund's adjuvant–primed (○) animals. (B) illustrates a left shift in the MCh dose response curve, contrasting with no change in parenchymal responses shown in C. *, P < 0.01; **, P < 0.05.