Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Naive T cells are initiated into the productive phase of their maturational pathway after encountering antigen (Ag) presented by accessory cells. Through accessory cells, the T cell is exposed to antigens that have been processed and presented in association with surface class II molecules of the major histocompatibility complex (MHC) (1). Several cell types have the capacity to present antigen, including dendritic cells, B cells, and macrophages. Additionally, various epithelial cells can express MHC class II molecules and have been reported to be capable of antigen presentation. In both the rat and human systems, colonic epithelial cells have been found capable of inducing T-cell proliferation in response to antigenic stimulation (2, 3). Respiratory epithelial cells have also been reported to express MHC class II molecules and to induce T-cell proliferation (4). However, the role of respiratory epithelial cells in Ag presentation in the upper and lower airways is uncertain (5). The respiratory tract is populated with professional accessory cells such as mucosal dendritic cells, and it is unclear how the different MHC class II Ag-expressing cell types function in the upper and lower airways (6). The ability of respiratory epithelial cells to process and present antigen to T cells may have physiologic importance, since these epithelial cells vastly outnumber all other constituents of the mucosal surface exposed to airborne infectious agents and allergens.

Our laboratory has previously shown that respiratory epithelial cells can stimulate T-cell proliferation in a mixed lymphocyte reaction and can induce receptor-associated tyrosine kinase activation in cocultured T cells (4, 7). We studied the kinetics of Ag uptake and trafficking patterns in two respiratory epithelial cell lines, A549 and BEAS-2B, utilizing confocal and electron microscopy. We also investigated specific cell-surface molecules that play a role in the function of antigen-presenting cells (APC). Nasal epithelial cells (NEC) and airway epithelial cells (AEC) are similar to conventional APC in that they express CD80 and CD86 and functional Fc receptor (FcR), and contain acidic compartments in which class II MHC molecules and processed peptide associate. The site of these cells, juxtaposed to the external environment, may make them important in the regulation of local immune responses.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

NEC and AEC

The NEC and AEC were obtained from nasal turbinates of patients undergoing surgical resection, or from lobar airway segments of patients undergoing thoracotomy, respectively. The patients with nasal turbinate resection had no underlying medical problems and were having the operation for cosmetic reasons. The lobar airway segments were derived from patients undergoing thoracotomy for lung cancer. None of the patients had been previously treated, and all were smokers. Specimens were transported on ice in sterile complete medium (CM) and were processed immediately. The surgical specimens were trimmed and thoroughly rinsed, and the luminal surfaces were exposed. The nasal and airway specimens were then incubated in Dispase (3 mg/ml; Boehringer-Mannheim, Indianapolis, IN) for 30 min at 37°C before gentle scraping with a scalpel blade. The cell suspension was washed in Ham's F-12 medium (Sigma, St. Louis, MO), passed over a 40-µm filter mesh, and examined with light microscopy for viability and cell composition. The viability was greater than 95% as determined by trypan blue dye exclusion, and the cells contained less than 1% lymphoid cell or macrophage contamination (98% anticytokeratin antibody positive) and less than 1% HLe1/Leu M3 positive (Becton- Dickinson, Mountain View, CA) by immunofluorescence. Ciliated epithelium was identified through the beating of cilia. Nonciliated columnar epithelial cells retained their characteristic morphologic appearance and were easily differentiated through light microscopy from erythrocytes and lymphocytes (4, 7). We had 15 sets of cultures, and the yield of epithelial cells varied from 10⁵ to 2 × 10⁶. All of the epithelial cells were cultured at the same time in 96-well round-bottom plates (Linbro, Oxnard, CA).

Cell Lines

The A549, FO1-D, and BEAS-2B cell lines were obtained from the American Type Culture Collection (Rockville, MD), and were grown in F-12 medium supplemented with 10% fetal calf serum (FCS), 1% penicillin and streptomycin, and 1% L-glutamine (Life Technologies, Grand Island, NY), henceforth called CM. For our experiments the cells were grown on glass coverslips (Fisher Scientific, Pittsburgh, PA).

Stimulation with Ag, Mitogen, and Monoclonal Anti-CD3 Antibody

The NEC and AEC were used as accessory cells for Ag, mitogen, and monoclonal anti-CD3 antibody-induced T-cell proliferation. For these experiments, T cells were obtained from MHC- and non-MHC-matched peripheral blood mononuclear cells (PBMC) (from normal blood donors), and were monocyte-depleted by passage through a nylon wool column. The PBMC were incubated on the column for 45 min and eluted. T cells isolated in this manner failed to respond to Ag, phytohemagglutinin (PHA), and monoclonal anti-CD3 antibody. Monocyte-depleted T cells (10⁵) were cocultured with varying concentrations of irradiated (6,000 rads, cesium source) NEC or AEC, tetanus toxoid (TT) (Wyeth Laboratories, Philadelphia, PA) (0.4-4 µg/ml), Candida (0.4-4 µg/ml), PHA (0.01-1 µg/ ml), or the monoclonal anti-CD3 antibody 446 (1 µg/ml) in 0.2 ml of CM in duplicate round-bottom microtiter plates (Linbro) at 37°C in 5% CO₂ in an incubator (8). In some experiments the NEC and AEC were pretreated with monoclonal anti-CD80 and CD86 antibodies (PharMingen, Mountain View, CA) for 30 min at 4°C before coculture. The mitogen- and anti-CD3-stimulated T cells were maintained in culture for 3 d and the TT- and Candida-stimulated T cells were maintained for 5 d. The monoclonal antibody 446 has been previously characterized. It is an antibody directed against the -chain of the CD3 complex (9). This antibody can stimulate T cells by using the FcR on accessory cells to crosslink the Ag receptor. The 446 antibody was purified from culture supernatant on a protein G column. For the last 16 h of culture, the cells were pulsed with 1 µCi of (³H)-thymidine, harvested with a 96-well Tontec harvester, and counted with a Wallac Micro-beta scintillation counter (EG&G Company, Gaithersburg, MD) (10).

Fluorescence Microscopy

TT, keyhole limpet hemocyanin (KLH) (Calbiochem, La Jolla, CA), and ovalbumin (OVA) (Sigma) were labeled by dialysis with fluorescein isothiocyanate (FITC) (5 mg/ml) (Sigma) in 0.5 carbonate buffer (pH 9.5) for 16 h. The mixture was rotated overnight at 4°C, and unbound FITC was removed by passage over a Sephadex G-25 column. The protein content of the column fractions was determined from the ratio of the optical density (OD) of FITC to that of FITC-labeled protein at 280 nm. A ratio of 5:1 (FITC:protein) was considered optimal for these studies. Interferon (IFN)- (100 U/ml for 48 h)- and granulocyte- macrophage colony-stimulating factor (GM-CSF) (100 U/ml for 48 h)-stimulated and unstimulated A549 cells, BEAS-2B cells, NEC and AEC were pulsed with FITC-labeled Ags at 37°C, harvested at various time points, and analyzed with fluorescence microscopy. In some experiments the cells were pretreated with 100 mM of chloroquine for 1 h, whereas other experiments were conducted at 4°C. For the colocalization studies with NEC and AEC, we used a monoclonal anti-human leukocyte antigen (HLA)-DR antibody (Becton-Dickinson). For the colocalization studies with A549 and BEAS-2B cells, we utilized mouse anti-cathepsin D (early endosomal marker) (Accurate Antibodies, Westbury, NY), rabbit polyclonal anti-mannose-6-phosphate receptor (M6PR) (late endosomal marker) (kindly provided by Dr. R. Dunn of the University of Florida), and mouse anti-CD63 (lysosomal marker) (Becton-Dickinson) antibodies. The NEC, AEC, A549, and BEAS-2B cells were incubated with the FITC-labeled Ags for varying periods, after which they were rinsed in phosphate-buffered saline (PBS) and fixed with methanol:acetic acid (3:1) for 5 min at 4°C. Fixed cells were then blocked with PBS containing 5% goat serum and 0.2% Triton-X (Sigma) for 15 min before incubating them with the primary antibody for 1 h. The cells were rinsed three times in PBS prior to incubating them with Texas-red-labeled conjugated donkey antirabbit Ig (Accurate Antibodies). The NEC and AEC were rinsed twice in PBS and cytospin preparations of the cells were made. The coverslips containing the A549 and BEAS-2B cells were then rinsed in PBS and mounted with Immun-mount (Shandon, Pittsburgh, PA) before viewing them with a Leica fluorvent laser confocal microscope (Leica, Deerfield, IL) at a step position of 1 µm on the x-y or x-z axes. To colocalize with-3- (2,4-dinitroanilino)-3' - aminomethyldipropylamine (DAMP), cells were incubated with the FITC-labeled Ags and 30 mM DAMP, and were fixed with 4% paraformaldehyde in PBS for 30 min. The cells were washed twice with 50 mM NH₄Cl for 5 min before permeabilization with 0.2% Triton-X in PBS for 30 min. The cells were then incubated with rabbit polyclonal antidinitrophenol (DNP) antibodies (Molecular Probes, Eugene, OR). After three washes with PBS, the cells were incubated with Texas Red- conjugated donkey antirabbit Ig before being mounted with Immun-mount (Shandon) and viewed with a confocal microscope. Two observers routinely examined 10 separate fields. Mean fluorescence intensities of the internalized FITC-labeled Ags in the cells of the entire field examined were determined with the Leica fluorvert laser confocal microscope by using the NIH Image software package (National Institutes of Health, Bethesda, MD) (10).

Nasal and Airway Explant Cultures

Unstimulated nasal turbinates and airways were pulsed apically with FITC-labeled KLH, OVA, and TT, and were examined with laser confocal microscopy for antigen uptake. In these experiments, 4-mm nasal or airway explants were pulsed on the luminal side with FITC-labeled KLH, OVA, and TT, and were harvested at different time points by flash freezing in liquid nitrogen. The nasal turbinates and airways were cryosectioned into 4-µm sections that were examined by two independent observers as described earlier.

Transmission Electron Microscopy

For the kinetic studies of antigen uptake, gold-labeled 10-nm OVA particles were obtained from Sigma. The A549 and BEAS-2B cells were incubated with gold- labeled OVA for varying periods before being fixed with 2% paraformaldehyde, 2.5% glutaraldehyde, 4 mM CaCl₂, and 2 mM MgCl₂ in 0.1 M sodium cacodylate buffer, pH 7.4, for 45 min. After 45 min of fixation, the cells were rinsed for 45 min in 0.1% cacodylate:3% potassium ferrocyanide (1:1), washed twice in Na maleate, pH 5.15, for 5 min each, and stained with 1% uranyl acetate in Na maleate, pH 6.0, for 30 min. After dehydration in an alcohol series (50%, 75%, 90%, 100%; 10 min each), the cells were further dehydrated in 100% ethanol. Afterwards, the cells were incubated in 100% ethanol:EPON (1:1) for 10 min, 100% EPON for 15 min, and EPON mixture (16 ml EPON, 11 ml of dodecenyl succinic anhydride, 8 ml of nadic methyl anhydride, and 0.48 ml of 2,4,6-tri[dimethylaminomethyl]phenol), with three changes of the mixture over a 2 h incubation period. The cells were then covered with EPON mixture and incubated at 68°C for 48 h before sectioning with a diamond knife mounted on an Ultracut ultramicrotome (Reichert, Valley Cottage, NY). The sections were mounted on copper grids (300 mesh), counterstained with 1% uranyl acetate and 1% lead citrate for 5 min, and viewed with an Hitachi 7000 transmission electron microscope (Hitachi, Danbury, CT) operating at an accelerating voltage of 75 kV. All chemicals used in these procedures were reagent grade and were obtained from Electron Microscopy Sciences (Washington, PA) (10).

Immunofluorescence

The BEAS-2B and A549 cells and the NEC and AEC were stained through indirect methods as previously described, using various monoclonal antibodies (mAbs) (see the subsequent discussion) or with isotype-matched control antibodies followed by the affinity-purified FITC-conjugated F(ab')₂ fragment of goat antimouse Ig (Tago, Burlingame, CA), and were analyzed through flow cytometric gating of live cells (10). The isotype controls were IgM (CD80), IgG1 (CD86), and IgG_2b (4.3). The W6/32 (anti-class I) antibodies were obtained from the American Type Culture Collection, the anti-CD80 and anti-CD86 antibodies were purchased from Pharmingen, and the mAb 4.3 (anti-FcR) was provided by Dr. J. Unkeless (Mount Sinai School of Medicine).

Immune Complexes

Human anti-TT IgG at a concentration of 25 mg/ml (Miles, Inc., Elkhart, IN) was incubated with 8 mg/ml of FITC-TT at 37°C for 1 h. The insoluble immune complexes were removed by centrifugation at 10,000 × g for 1 h, and the concentration of the soluble immune complexes was adjusted at 2 mg/ml.

Isolation of Primary Monocytes

Mononuclear cells obtained from healthy volunteer blood donors were separated from their buffy coats through Ficoll-Hypaque (Pharmacia, Piscataway, NJ) density centrifugation. The cells were then washed three times with sterile PBS and resuspended in RPMI 1640 medium (GIBCO) supplemented with 10% FCS (GIBCO), 2 mM L-glutamine, and 1% penicillin-streptomycin (GIBCO). From 5- to 10-million freshly isolated PBMC were incubated at 37°C in culture flasks and allowed to adhere for 45 min. The nonadherent cells were extensively washed with PBS, harvested with a rubber policeman, and stained with monocyte specific monoclonal anti-CD14 antibody to assess the purity of the preparation.

Reverse Transcription-Polymerase Chain Reaction for CD80 and CD86

RNA was extracted from primary monocytes, FO1-D cells, NEC, and AEC by using the acid guanidine thiocyanate/phenol/chlorochloroform method as described previously (8). Known quantities of RNA were mixed with 1 µg of total cellular RNA and reverse transcribed at 37°C for 60 min in 20 µl of buffer containing 10 mM Tris, pH 8.3; 50 mM KCl₂; 5 mM MgCl₂; 1 mM each of deoxyadenosine triphosphate, deoxycytosine triphosphate, deoxyguanosine triphosphate, and deoxythymidine triphosphate; 20 U of ribonuclease inhibitor (Boehringer Mannheim); and 50 U murine leukemia virus reverse transcriptase (Bethesda Research Laboratories, Gaithersburg, MD). Reactions were stopped by heat inactivation for 10 min at 95°C with 2.5 min of annealing and extension at 65°C. Negative controls were run by omitting RNA from the complementary DNA (cDNA) synthesis and with amplification via the polymerase chain reaction (PCR) (11). PCR products were separated in 2% NuSieve agarose (FMC, Rockland, ME).

Statistical Analysis

Nonparametric statistical analysis was done with the SAS statistical program package (SAS Institute, Cary, NC) and through methods previously described. Data were compared through use of the Mann-Whitney U test. A value of P < 0.05 was considered statistically significant (8).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Accessory Cell Function for Antigens, Mitogens, and Anti-CD3-Induced T-Cell Proliferation

We were interested in assessing respiratory epithelial cell accessory activity for MHC class II molecule- and non-class II molecule-mediated functions. To this end, we studied Ag, mitogen, and anti-CD3-mediated T-cell proliferation, using NEC and AEC as accessory cells. As previously reported, monocyte-depleted PBMC did not support Ag-, PHA-, or anti-CD3-mediated T-cell proliferation as determined from thymidine incorporation (8). T-cell proliferation, however, could be restored by adding either autologous NEC or AEC (Table 1). There was a graded T-cell proliferative response to stimulation with TT, Candida, PHA, and the monoclonal anti-CD3 antibody 446 at different concentrations of NEC and AEC. Because we had previously demonstrated that respiratory epithelial cells can induce the proliferation of CD4⁺ T cells, we first wanted to investigate patterns of Ag trafficking (4).

Colocalization of Ag with HLA-DR

Because respiratory epithelial cells express MHC class II antigens (4), and in our study induced Ag-specific T-cell proliferation in accord with previously reported data (5), we wanted to determine whether the endocytosed Ag followed a class II trafficking pattern. We colocalized Texas red-labeled anti-HLA-DR antibody with FITC-labeled KLH (red + green = yellow) in BEAS-2B and A549 cells, using laser confocal microscopy (Figure 1). Colocalization occurred optimally 90 min after Ag pulsing.

View larger version (43K): [in this window] [in a new window]   Figure 1.   Colocalization of FITC-labeled KLH with HLA-DR in A549 and BEAS-2B cells. A549 and BEAS-2B cells were harvested and pulsed with FITC-labeled KLH, recovered at different time points, and stained cytoplasmically with Texas Red-labeled anti-HLA-DR antibody. The yellow staining represents colocalization. This is representative of an experiment repeated five times.

Ag Trafficking in BEAS-2B and A549 Cells

We first investigated Ag trafficking in the BEAS-2B and A549 cell lines. After both the BEAS-2B and A549 cells were pulsed with FITC-labeled KLH, extensive speckled, punctate, green signals were observed, with a maximal intensity at 60 min, and which persisted for up to 120 min (Figure 2). Similar results were obtained when the BEAS-2B and A549 cells were pulsed with FITC-labeled OVA and TT (data not shown). Two observers routinely examined 10 separate fields. Ag pulsing experiments at 4°C showed no evidence of Ag uptake, suggesting that internalization of Ag by BEAS-2B and A549 cells is an active process (data not shown). We also found that chloroquine pretreatment (100 mM for 1 h) had no effect on the uptake of Ag, suggesting that the uptake occurs via non-receptor- mediated endocytosis (data not shown). Accessory cell function, including the kinetics of Ag uptake, can be upregulated by cytokines including IFN- and GM-CSF. We treated both the BEAS-2B and A549 cells with IFN- (100 U/ml for 48 h) to determine whether both the magnitude and the kinetics of Ag uptake would be affected (Figures 2A and 2B). In contrast to the lack of staining in untreated BEAS-2B and A549 cells, extensive punctate staining was observed at 30 min in both BEAS-2B and A549 cells after pulsing with FITC-labeled KLH. Staining was diminished at 120 min. Comparable results were obtained when GM-CSF (100 U/ml for 48 h)-treated A549 and BEAS-2B cells were pulsed with FITC-labeled OVA (Figures 2A and 2B). To better quantify the difference in Ag uptake in the BEAS-2B and A549 cells after treatment with IFN- and GM-CSF, we determined the mean fluorescence intensity of the endocytosed Ag in the BEAS-2B and A549 cells at all time points tested. There was a significant increase in flourescence intensity of the entire field examined, corresponding to increased Ag uptake, starting at 30 min in the IFN-- and GM-CSF-treated BEAS-2B and A549 cells as compared with untreated cells (P = 0.05); this became maximal at 60 min and was diminished at 120 min (P = 0.01) (Table 2).

View larger version (43K): [in this window] [in a new window]   View larger version (40K): [in this window] [in a new window]   Figure 2.   Ag trafficking in IFN-- and GM-CSF-stimulated and unstimulated A549 and BEAS-2B cells. IFN-- (A) and GM-CSF (B) (100 U/ml for 48 h)-stimulated and unstimulated A549 and BEAS-2B cells were grown on coverslips, pulsed with FITC-labeled KLH, harvested at different time points, and analyzed through laser confocal microscopy. This is representative of an experiment repeated nine times.

Ag Trafficking in NEC and AEC

Having observed Ag uptake that could be upregulated in both the BEAS-2B and A549 cell lines by IFN- and GM-CSF, we sought to validate these findings in isolated primary NEC and AEC and in nasal and airway cell explant cultures. NEC and AEC were isolated from surgical specimens; some were treated with IFN- (100 U/ml for 48 h) and some with GM-CSF (100 U/ml for 48 h), whereas others were maintained without treatment in culture. In accord with the results obtained in the unstimulated BEAS-2B and A549 cells, we observed punctate staining corresponding to endocytosed Ag at 60 min, which persisted for up to 120 min (data not shown). When the NEC and AEC were treated with IFN- (100 U/ml for 48 h), extensive speckled, punctate staining similar to that observed in the BEAS-2B and A549 cells was evident in the treated cells at 15 min, was maximal at 60 min, and was diminished at 120 min (data not shown). Similar results were obtained when the NEC and AEC were treated with GM-CSF (100 U/ml for 48 h) and when the cells were pulsed with FITC-labeled OVA and TT (data not shown). Two observers routinely examined 10 separate fields. Since we demonstrated Ag uptake in BEAS-2B and A549 cells and in NEC and AEC when we pulsed isolated cells of all four types, we wanted to determine whether there was polarity to the trafficking of Ag. We introduced Ag on the luminal side of NEC and AEC in tissue explant cultures to better mimic Ag uptake in vivo. In both the nasal and airway cell explant cultures pulsed with FITC-labeled KLH, staining consistent with Ag uptake was observed at 60 min, confirming the results that we observed in nonpolarized BEAS-2B and A549 cells and in isolated NEC and AEC (Figure 3). Fluorescence signals appeared first in the apical cytoplasm of the nasal and airway explant culture cells at 30 min and in the basal cytoplasm at 60 min, and were present at 120 min. The same results were obtained when the explants were pulsed with FITC-labeled OVA and TT (data not shown).

View larger version (38K): [in this window] [in a new window]   Figure 3.   Ag trafficking in nasal turbinates and airways. Nasal turbinates and airways were pulsed apically on the luminal side of the specimen (lumen) with FITC-labeled KLH, harvested at different time points, sectioned, and examined through laser confocal microscopy. This is representative of an experiment repeated five times.

Intracellular Trafficking of Ag

To more carefully determine the fate of Ag following its internalization in BEAS-2B and A549 cells, we conducted kinetic studies to follow the uptake of Ag labeled with colloidal gold (Au-OVA), using electron microscopy for this. BEAS-2B and A549 cells were pulsed with Au-OVA, incubated at 37°C for varying periods, fixed, embedded, and examined electron microscopically (Figure 4). At 15 min, Au-OVA was evident in early endosomes and in endosomal compartments that resemble the compartment of uncoupling of receptor and ligand. At 30 min after pulsing, multivesicular bodies were labeled with Au-OVA, and by 60 min, Au-OVA was found in primary lysosomes and at 120 min in secondary lysosomes of the BEAS-2B and A549 cells. A problem with electron microscopy in this application is that there may be dissociation of OVA from the gold labeling particles within endosomes, with following of the gold particles rather than of the protein itself. Since electron microscopy could only reveal the vesicular compartments of the BEAS-2B and A549 cells morphologically, we performed immunofluorescence colocalization studies of FITC-labeled Ags with markers for the acidic, endosomal, and lysosomal compartments, in order to more clearly define Ag trafficking in the two types of cells.

View larger version (90K): [in this window] [in a new window]   Figure 4.   Ag trafficking in A549 and BEAS-2B cells as assessed with electron microscopy. The A549 and BEAS-2B cells were pulsed with Au-OVA (arrows) and harvested at different time points. Magnification: ×45,000. The inset is a ×5 magnification of Au-OVA in the A549 and BEAS-2B cells. This is representative of an experiment repeated three times.

Colocalization of Endocytosed Ag

We used Texas Red-conjugated antibody markers for the acidic compartments that occur in both early (cathepsin D) and late (DAMP) endosomes, the late endosomal marker M6PR, and the lysosomal marker CD63 (Figure 5) for our studies in BEAS-2B and A549 cells. Colocalization of FITC-labeled KLH was evident at 60 min in the early compartments, at 90 min in the late endosomes (M6PR), and at 120 min in the lysosomes (CD63) after FITC- labeled KLH pulsing. DAMP (acidic compartments) colocalization was noted at 60 and 90 min. We also colocalized FITC-labeled OVA and FITC-labeled TT in A549 and BEAS-2B cells in early and late endosomes, acid vesicles, and lysosomes (data not shown).

View larger version (60K): [in this window] [in a new window]   Figure 5.   Colocalization of FITC-labeled KLH in A549 and BEAS-2B cells. The A549 and BEAS-2B cells were pulsed with FITC- labeled KLH, harvested after 60 min, and stained with Texas Red-labeled antibodies specific for the various subcellular compartments. Cells were stained for early endosomes (cathepsin D) and for late endosomal (M6PR), lysosomal (CD63), and acidic compartments (DAMP). The yellow staining represents colocalization of antigen in a subcellular compartment. This is representative of an experiment repeated five times.

FcR Expression and Uptake of IgG Immune Complexes

Although IgA is clearly important in host defense in respiratory secretions, the role of IgG is less well defined (12). IgG can penetrate into secretions in the upper and lower respiratory tract by passive diffusion, and its concentration is markedly increased during inflammation (13). Little is known about FcR expression on respiratory epithelial cells, and if present, whether this could serve as a pathway for Ag entry. To determine whether respiratory epithelial cells express FcR, we performed surface immunofluorescence on IFN--stimulated (100 U/ml for 48 h) and unstimulated A549 and BEAS-2B cells. There was some expression of FcR in both the unstimulated A549 and BEAS-2B cells that was upregulated by IFN- (Figure 6). We then investigated the uptake of IgG immune complexes by respiratory epithelial cells by studying the uptake by BEAS-2B and A549 cells of FITC-labeled TT complexed with human polyclonal anti-TT antibody in the form of soluble immune complexes. After pulsing both cell lines with TT-anti-TT complexes, punctate fluorescence staining was observed at 30 min, which persisted until 120 min (Figure 7). In the cells pulsed with FITC-labeled TT alone, uptake was not observed until 60 min, as in the case of KLH and OVA (Figure 7). There was a difference in the mean fluorescence intensities of the endocytosed, FITC-labeled immune complexes and those of FITC-labeled Ag in BEAS-2B and A549 cells within the entire field examined at 30 min (P = 0.05) (Table 3).

View larger version (42K): [in this window] [in a new window]   Figure 6.   FcR expression. Surface immunofluorescence for FcR expression was quantitated with 4.3 anti-FcR antibody on IFN-- stimulated and unstimulated A549 and BEAS-2B cells and flow cytometric analysis, with gating on live cells. The gated population of cells studied is shown (circle) in the left upper panel, and the percentages of cells staining for the negative control and 4.3 antibody are indicated in the right upper panel. The data are representative of an experiment repeated three times.

View larger version (76K): [in this window] [in a new window]   Figure 7.   Immune complex uptake. A549 and BEAS-2B cells were grown on coverslips, pulsed with FITC-labeled TT and FITC-labeled TT-anti-TT soluble IgG immune complexes, harvested at different time points, and analyzed with laser confocal microscopy. The data are representative of an experiment repeated five times.

CD80 and CD86 Expression

Classic APC express CD80 and CD86 (14). Expression of these costimulatory molecules can be upregulated in response to stimulation with cytokines such as IFN- and GM-CSF (15). We wanted to determine whether there was constitutive CD80 and CD86 expression in NEC and AEC, and whether IFN- or GM-CSF could increase this expression. Both NEC and AEC expressed CD80 and CD86 as determined through surface immunofluorescence analysis (Figure 8). We also investigated whether there was any CD80 or CD86 expression in the BEAS-2B and A549 cell lines. BEAS-2B cells were derived from SV-40- transformed nasal epithelial cells, whereas A549 cells represent type II epithelial cells and were established from a lung cancer. As in the case of freshly isolated NEC and AEC, both BEAS-2B and A549 cells expressed CD80 and CD86 (data not shown). We confirmed the presence of CD80 and CD86 in NEC and AEC by measuring the messenger RNA (mRNA) for these two clonal designators with the PCR. Fragments of 605 bp and 332 bp, corresponding to the predicted sizes of CD80 and CD86, respectively, were detected in the NEC and AEC (Figure 9) (11). We next sought to establish a functional role for CD80 and CD86 on NEC and AEC. We chose to study T-cell proliferation in response to stimulation with the monoclonal anti-CD3 antibody 446, with TT, and with Candida, since these Ags signal through the T-cell receptor and more closely represent the in vivo environment. To accomplish this, we attempted to block Ag-induced T-cell proliferation by pretreating NEC and AEC with either anti-CD80 or anti-CD86 antibodies. Inhibition of TT- and Candida-induced T-cell proliferation was observed after pretreatment of the NEC and AEC with either the anti-CD80 or the anti-CD86 antibody (Table 4).

View larger version (39K): [in this window] [in a new window]   Figure 8.   CD80 and CD86 expression on NEC and AEC. NEC and AEC were harvested from surgical specimens. Surface immunofluorescence on the NEC and AEC was measured with FITC-labeled monoclonal anti-CD80 and CD86 antibodies, followed by flow cytometric analysis, with gating on live cells. The gated population of cells studied is shown in the left upper corner (circle) of each panel, and percentages of cells staining for the negative control and anti-CD80 and CD86 antibodies are indicated in the right upper corner. The data are representative of an experiment repeated three times.

View larger version (68K): [in this window] [in a new window]   Figure 9.   CD80 and CD86 mRNA production by NEC and AEC. RNA was extracted and RT-PCR was done with specific primers for CD80 and CD86 mRNA. Fragments corresponding to mRNA for CD80 (605 bp) and CD86 (332 bp) were detected in NEC and AEC but not in the negative control FO1-D cells. Primary monocytes served as the positive controls. The data are representative of an experiment repeated three times.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In order to initiate an immune response, exogenous Ags are internalized, processed into small immunogenic peptides, and presented in the context of MHC class II molecules to CD4⁺ responder T cells (1). In classic APC the expression of the costimulatory molecules CD80 and CD86, and the class II pathway of intracellular trafficking, have been extensively studied (1). In the present study we found that as do classic APC, NEC and AEC express CD80 and CD86 (Figures 8-10) and can serve as accessory cells for both MHC class II- and non-class II-restricted functions (Table 1). A direct comparison of classic APC and respiratory epithelial cells will be part of our future studies. In accord with previous findings in conventional APC, our immunofluorescence, electron-microscopic, and colocalization studies (Figures 2-5) showed that Ag followed a class II endocytic pathway inside the respiratory epithelial cells in our study, and it was also possible to colocalize Ag with HLA-DR (Figure 1). In contrast to the case with passive diffusion, there was no correlation between the size of Ag (OVA = 46,000 D, TT = 70,000 D, and KLH = 1,300,000 D) and the rate of Ag uptake. OVA and KLH have both been used in an animal model of respiratory mucosal immune responses (16). Inhalational challenge in mice and canines with KLH elicits a primary local IgA and systemic IgG antibody response (17, 18). Inhalation induces CD4⁺ T-cell accumulation in the lung, increased IL-5 secretion by the accumulated T-cells, the accumulation of eosinophils, the development of an OVA-specific IgE response, and T-cell-dependent bronchial hyperreactivity (19). These findings closely resemble the response in human allergen inhalation challenge in atopic asthmatic individuals (24).

The kinetics of Ag uptake by the respiratory epithelial cells in our study were dramatically increased by IFN- and GM-CSF (Figure 2). Respiratory epithelial cells express receptors for IFN- and GM-CSF (25) and are in intimate contact with a variety of immunocompetent cells that secrete a number of cytokines, including IFN- and GM-CSF (26). AEC are themselves a source of GM-CSF (26). In monocytes stimulated with either IFN- or GM-CSF, there is increased antigen presentation, including the upregulation of MHC class II molecules, increased cytokine secretion, and enhanced phagocytic capacity (1). These findings may be relevant, since the upper and lower airways are constantly exposed to a wide variety of infectious agents, allergens, and environmental contaminants. The ability of IFN- and GM-CSF to enhance the accessory cell function of respiratory epithelial cells may help in eliminating potential pathogens, or alternatively may help to induce an inflammatory response (27).

The cells of the upper and lower airways that take up soluble Ag after its delivery to the mucosa have not been clearly delineated (2). Local dendritic cells and macrophages are presumed to be involved (28), although alveolar macrophages (AM) downregulate local immune responses against intratracheally administered T-cell-dependent Ags (29, 30). Resident pulmonary AM also actively suppress the APC function of lung dendritic cells in situ (6). Our data clearly suggest that respiratory epithelial cells perform an essential function of APC: the uptake of Ag in a typical processing pathway. AEC and NEC may function to sample and present airborne Ags to local T-cell populations, whereas dendritic cells may be involved in eliciting more systemic immune responses. Our findings, coupled with the previous observations that bronchial epithelial cells express MHC class II Ags and intercellular adhesion molecule-1, that coculture of AEC and T-lymphocytes results in T-lymphocyte activation and proliferation, and that AEC release IL-16, a chemoattractant factor for T lymphocytes, suggest that NEC and AEC participate at several levels in the airway mucosal response to airborne Ags (31, 32).

In further support of respiratory epithelial cells as accessory cells, the BEAS-2B and A549 cells as well as the NEC and AEC in our study were capable of internalizing IgG immune complexes through FcR (Figure 7). There is precedence for FcR expression on epithelial cells. Fc receptors for IgG have been described in human and rat neonatal small-intestinal epithelium (33, 34). Although IgA is actively secreted into the upper and lower airways, the role of IgG is less certain (35). The effect of inflammation on the transfer of IgG antibodies into nasal and lung secretions during viral infection is well documented (36). The kinetics of uptake of IgG complexes by respiratory epithelial cells were greater than for the uptake of soluble Ags, and may promote inflammation. Fc-mediated uptake of Ag by respiratory epithelial cells may be of immunopathologic significance, since it may augment tissue damage by increasing the number of Ag-specific T cells. IgG itself may cause tissue injury in the upper and lower airways by activating complement (13).

Ag presentation by respiratory epithelial cells may be important not only for local mucosal immunity but also for systemic immunization (37). Recent experiments with murine and primate systems have shown that the nasal route of immunization induces effective mucosal and systemic responses (38, 39). Animals immunized intranasally had strong secretory IgA antibody responses, not only in the respiratory tract and mouth (salivary glands), but also in the genitourinary tract (38, 39). The immunization also resulted in IgG and IgA antibodies in the serum (38, 39). This may be potentially important in developing vaccines for sexually transmitted diseases, including infection with human immunodeficiency virus type-1, since lymphocytes that are part of the nasal-associated lymphoid tissue (NALT) migrate to the genitourinary tract (40). On the basis of promising findings obtained in animal model systems, others have initiated similar studies in humans. Influenza vaccines introduced nasally have induced both mucosal and systemic responses comparable to those induced by parenteral injection (41). Similar results have been reported with peptide vaccines generated against bacterial pathogens (42). Intranasal immunization thus appears to have the potential to provide an effective route of vaccination for the induction of a protective response against both viral and bacterial pathogens.

In conclusion, our findings suggest that Ag uptake by respiratory epithelial cells is tightly regulated and may have an important role in controlling immune responses in the upper and lower airways. Upregulation of respiratory epithelial cell accessory function may aid in protective responses against specific pathogens, or in inducing inflammatory responses. An improved understanding of Ag presentation by respiratory epithelial cells may lead to better treatment of chronic inflammatory disorders of the upper and lower respiratory tract, and to the development of vaccines against respiratory infections.