Sensitized mice were injected intraperitoneally (i.p.) with 20 μg of Af (Hollister Stier, Elkhart, IN) and 20 mg Al(OH)₃ (Imject Alum; Pierce, Rockford, IL) in PBS (100 μl) on days 1 and 14, followed by intranasal (i.n.) challenge on day 27 with 25 μl of allergen extract: (12.5 μg Af in 21% glycerol/ PBS). To analyze the kinetics of SP-A protein and mRNA changes BAL fluid and lung tissue were collected before (0 h) and 1, 6, 12, 24, 48 and 72 h after intranasal (i.n.) treatment, using a separate animal group at each time point (n = 6 in each) as described previously 4. These time points were selected on the basis of our previous experience with acute models of locally elicited inflammation in mice 41617. Af sensitized and vehicle challenged mice (n = 8), non sensitized, Af challenged mice (n = 6) and non sensitized, glycerol challenged mice (n = 8) were studied 12 h after treatment. Naïve mice (n = 8) were used to control for the effects of i.p. sensitization and/or glycerol challenge. Lung function, BAL inflammatory cell and cytokine profile and SP-A levels were not significantly different between naïve animals and the groups that received sensitization plus glycerol challenge or glycerol challenge alone (not shown). All experimental subjects in this study were under a protocol approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania.

Lungs were lavaged with sterile saline as previously described 456. SDS-PAGE of the surfactant samples was carried out using NuPAGE 10% Bis-Tris gels (Novex, San Diego, CA). Western blots were performed using rabbit polyclonal anti-SP-A. Each lane was loaded with 10 μg of total protein. To investigate the absolute SP-A content in the BAL in addition to Western blot analysis, we used an ELISA protocol as described previously in detail 5. Total RNA was isolated from lungs and specific SP-A mRNA content was determined by Northern blot analysis as described previously 618.

In vivo measurement of lung function after Af challenge was performed in conscious, unrestrained, spontaneously breathing mice in a whole-body plethysmograph (Buxco Electronics Inc., Troy, NY) as described previously 4. Airway function was quantified using the Enhanced Pause (Penh).

Analysis of total and differential cell count and cytokine profile of the BAL was carried out after lungs were lavaged with sterile saline as previously published 4619. Cytokine levels were measured in the cell free supernatant of the BAL by ELISA using antibodies and recombinant cytokines from PharMingen (San Diego, CA).

Human SP-A was purified from human surfactant collected previously from patients with pulmonary alveolar proteinosis (PAP) as previously published 20. The purified material was analyzed for protein concentration, subjected to Coomassie Blue staining (Figure 4A) and Western blot analysis (Figure 4B) and stored in aliquots at -80°C. The biological activity of the purified SP-A was tested in stimulated human peripheral blood monocytes (PBMC) (Figure 4C).

Mice were sacrificed and spleens were harvested 24 hours after treatment. Spleens were homogenized in 10 mL of sterile PBS and layered over 5 mL of Histopaque (Histopaque-1077, Sigma) followed by density gradient centrifugation at 400 × g for 30 minutes at 18°C. Cells were then washed and cultured at 2 × 10⁶ cells ml^-1 in U-bottom 96-well plates (Nunclon Delta Surface, Nunc, Denmark).

Af and PMA-Ionomycin (PI, Sigma-Aldrich, St. Louis, MO) were used as mitogens for the lymphocyte proliferation assay. Varying amounts of SP-A were added to stimulated lymphocytes. Mouse splenocyte cultures were incubated for 37°C in a humidified atmosphere with 5% CO₂ for 96 h. At the 72 h time point, each well was pulsed with [³H] thymidine (0.2 mCi ml^-1) for an additional 24 h at 37°C. Cells were then harvested on a PHD Harvester (Cambridge Technology, Cambridge, MA). [³H] thymidine incorporation was determined via liquid scintillation counting (Beckman Instruments, New York, NY). Samples were plated in triplicates and the experiments were repeated n = 4 times unless otherwise stated.

Human peripheral blood mononuclear cells were isolated from healthy volunteers using a standard protocol approved by the IRB of the University of Pennsylvania.

The effects of SP-A on lymphocyte proliferation was further assessed using CFSE labeling as previously described 21. Surface staining was performed at the time of harvest using Phycoerythin conjugated anti-mouse CD3, Fluorescin (FITC) conjugated anti-mouse CD4, FITC conjugated rat IgG_2b isotype control (all from Pharmingen) and Allophycocyanin conjugated anti-human CD8 (Caltag, Burlingame, CA). Data were acquired on a four-color, dual laser FACSCalibur (Becton Dickinson, San Jose, CA). For FACS analysis the gated cell populations were divided into four different generations (G1, G2, G3 and G4) according to their CFSE content. Isotype controls were used to distinguish positive and negative populations. The SP-A-treated samples were compared with, and data were expressed as the percentage of the non-treated samples.

Data were expressed as mean ± SEM. Time courses and dose responses were compared using ANOVA. To test differences between individual groups Student's t test assuming equal variances were performed. Correlations were investigated by regression analysis. A p value of < 0.05 was considered as significant. Data were analyzed with the Sigma Stat standard statistical package (Jandel Scientific).

To study the relationship between changes in SP-A levels and the sequence of inflammatory events we used a model of single allergenic airway provocation 4. Western blots were analyzed from the cell-free supernatant of the BAL fluid, using naïve control samples as standards that were run on each gel. Decreases in SP-A levels began at 1 h (88 ± 19% of naïve control) and reached the lowest values 12 h (26 ± 5%) after Af challenge. Recovery was observed 72 h later (135 ± 40 %) (Fig. 1A and 1B). To verify that the alterations in SP-A levels were specific at the 12 h time point, sensitized and Af-challenged mice (26.5 ± 5%) were compared with mice that received vehicle (glycerol) treatment instead of antigen (90.0 ± 19%; p < 0.05 n = 8, Fig 1C).

SP-A protein decreases were not associated with commensurate mRNA changes. Northern blot analysis on RNA extracted from the lung tissue of each group of sensitized mice showed that SP-A mRNA actually increased 12 h after allergen challenge in comparison with the 0 h controls (160 ± 6% vs. 106 ± 16 p < 0.05, n = 5, Fig 1B). The absolute BAL SP-A content (determined by ELISA) was decreased from 378 ± 46 to 239 ± 25 μg/lung (p < 0.05 n = 8) at 12 h and to 222 ± 22 μg/lung (p < 0.05 n = 8) 48 h after Af challenge.

In order to examine the proinflammatory changes that were linked with the greatest fall in SP-A concentrations, we chose to study the 12 h time point. Following allergen provocation there were significant Penh increases 12 h after Af-challenge of sensitized mice (1.91 ± 0.36, n = 8) in comparison with glycerol challenge (0.88 ± 0.06, n = 8, p < 0.01). The glycerol challenge controls were not different from mice that received no sensitization or challenge (Fig 2A). Levels of Penh returned to near normal by 72 h (1.36 ± 0.1; n = 8). The cellular composition of BAL showed that 12 h after intranasal Af challenge there was a significant influx of inflammatory cells predominated by neutrophils (500 × 10³ cells, p < 0.01, n = 8) and eosinophils (110 × 10³ cells p < 0.01, n = 8) in comparison with glycerol challenged controls. These cell numbers were also reduced (1 ± 1 and 31 ± 19, respectively, n = 8) by the 72 h time point.

Similarly to our previously characterized ovalbumin and Af-induced models of allergic airway inflammation 4619, i.p. sensitization and i.n. Af challenge induced a local production of predominantly Th2-type cytokines. In comparison with glycerol challenged controls there was a significant release of IL-4 (559 pg/ml vs. 3 pg/ml p < 0.01), IL-5 (283 pg/ml vs. 10 pg/ml p < 0.001; Figure 2C) and Eotaxin (173 pg/ml vs. 10 pg/ml p < 0.001) but not IFN-γ 12 h after allergen challenge (Figure 2C). Regression analysis using data pooled from the sensitized/Af challenged, sensitized glycerol challenged and non sensitized glycerol challenged mice revealed a significant, negative correlation with each of these cytokines (SP-A vs. IL-4: r = -0.61, p < 0.05; SP-A vs. IL-5: r = -0.82, p = 0.01; SP-A vs. Eotaxin: r = -0.75, p < 0.01). There was no correlation between SP-A and IFN-γ (not shown). Further, by the 72 h time point after allergen challenge IL-4 (8.7 ± 2.7 pg/ml), IL-5 (10.3 ± 7.3 pg/ml) and Eotaxin (9.72 ± 3.5 pg/ml) returned to close to baseline. To verify that the amount of airway SP-A is negatively associated with Th2 cytokine production, we also performed quantitative real-time PCR analysis of IL-4, IL-5 and IFN-γ mRNA. These results showed that mRNA levels for the Th2 cytokines increased approximately 35 fold in comparison with naïve controls 12 h after allergen challenge and returned to control levels by 72 h (Figure 2D).

To test the hypothesis that SP-A exerts an inhibitory effect on Th2 processes, we used a model of in vitro allergen re-challenge of lymphocytes isolated from sensitized and control mice and treated the cells with SP-A purified from human PAP material. Figure 3 shows that following purification and endotoxin depletion the structural integrity (Fig 3A) and immunogenicity (Fig 3B) of SP-A remained intact. Purified, LPS-depleted human SP-A exhibited significant biological activity as assessed by its effects on proliferation of human PBMC (Fig 3C).

Splenic lymphocytes isolated from naïve mice and mice sensitized with Af were re-challenged in vitro with Af in the presence or absence of SP-A. ³H-thymidine uptake was expressed as cpm (counts per minute). The background cpm throughout our experiments ranged between 0 and 16. The SP-A preparation had < 0.07 endotoxin units/ml. This level is consistent with other cell culture SP-A preparations 10. Furthermore, culturing murine lymphocytes in up to twice the amount of LPS (0.14 endotoxin units/ml) had no significant effect on lymphocyte function (Fig 4A). Baseline proliferation of naïve lymphocytes was not affected by SP-A (Figure 4A). It appears that an activated state of the lymphocytes is a prerequisite for inhibition by SP-A since sensitized cells stimulated with Af were markedly inhibited by SP-A in a dose-dependent manner (Fig 4B). Indeed greater antigenic stimulation resulted in greater suppression of proliferation by SP-A at 10 μg/ml (Fig 4C).

Suppression of T cell proliferation was not a result of sequestering Af in the culture because lymphocytes stimulated with PMA (10 ng/ml) and ionomycin (500 ng/ml) were inhibited by SP-A (10 μg/ml) to a similar extent as the Af-stimulated lymphocytes by 66 ± 4% (Figure 4D), suggesting a direct inhibitory action on activated cells.

To further confirm the effects of SP-A on the proliferating cells, we added an intracellular fluorescent dye 5,6-carboxy- fluorescein diacetate succinimidyl ester (CFSE) to the cell cultures. The amount of this dye is halved as the Af-stimulated lymphocytes divide, allowing individual generations to be detected using FACS analysis. Mononuclear cells were gated for positive CD3/CD4 or CD3/CD8 expression and marked as G1 (the highest CFSE content), G2, G3 and G4 population according to decreasing levels of CFSE expression. Cells incubated in the presence of 10 μg/ml of SP-A were compared with equivalent populations of cells stimulated in the absence of SP-A and the results are expressed as a percentage of the SP-A-free (100%) control. In comparison with medium treatment, (represented by the middle line) addition of SP-A increased the proportion of cells that remained in the G1 population in the samples stimulated by Af (Figure 5A). The proportion of the G1 cells in the SP-A treated CD3/CD4 populations were 95%, 118%, and 163% of the medium treated (SP-A-free) controls at 0, 1 and 10 μg/ml Af concentration, respectively. Thus, cells exposed to greater antigen stimulation appeared to be more susceptible to inhibition by SP-A. We did not observe a similar inhibitory effect by SP-A in the CD3/CD8 cells (not shown).

Addition of SP-A (10 μg/ml) to sensitized lymphocytes significantly inhibited IL-4 and IL-5 production both in the Af-induced (Figure 5B) and PMA/ionomycin stimulated (not shown) cultures (p < 0.05, n = 4). SP-A apparently also inhibited IFN-γ levels in the non-stimulated and in the PMA/ionomycin stimulated samples indicating a non-selective activity on cytokine production by stimulated T cells (not shown).