Surfactant protein-A suppresses eosinophil-mediated killing of Mycoplasma pneumoniae in allergic lungs

Abstract

Surfactant protein-A (SP-A) has well-established functions in reducing bacterial and viral infections but its role in chronic lung diseases such as asthma is unclear. Mycoplasma pneumoniae (Mp) frequently colonizes the airways of chronic asthmatics and is thought to contribute to exacerbations of asthma. Our lab has previously reported that during Mp infection of non-allergic airways, SP-A aides in maintaining airway homeostasis by inhibiting an overzealous TNF-alpha mediated response and, in allergic mice, SP-A regulates eosinophilic infiltration and inflammation of the airway. In the current study, we used an in vivo model with wild type (WT) and SP-A(-/-) allergic mice challenged with the model antigen ovalbumin (Ova) that were concurrently infected with Mp (Ova+Mp) to test the hypothesis that SP-A ameliorates Mp-induced stimulation of eosinophils. Thus, SP-A could protect allergic airways from injury due to release of eosinophil inflammatory products. SP-A deficient mice exhibit significant increases in inflammatory cells, mucus production and lung damage during concurrent allergic airway disease and infection (Ova+Mp) as compared to the WT mice of the same treatment group. In contrast, SP-A deficient mice have significantly decreased Mp burden compared to WT mice. The eosinophil specific factor, eosinophil peroxidase (EPO), which has been implicated in pathogen killing and also in epithelial dysfunction due to oxidative damage of resident lung proteins, is enhanced in samples from allergic/infected SP-A(-/-) mice as compared to WT mice. In vitro experiments using purified eosinophils and human SP-A suggest that SP-A limits the release of EPO from Mp-stimulated eosinophils thereby reducing their killing capacity. These findings are the first to demonstrate that although SP-A interferes with eosinophil-mediated biologic clearance of Mp by mediating the interaction of Mp with eosinophils, SP-A simultaneously benefits the airway by limiting inflammation and damage.

Figure 1. Mp burden is decreased in Mp-infected SP-A^−/− allergic mice.

A) WT and SP-A^−/− mice were injected ip with Ova/Alum mixture on days 1 and 14, subject to Ova aerosolization on days 21–23 and instilled with either Mp or saline on day 25. Mp burden was determined in B) BAL by plating dilutions on PPLO agar plates and counting via 10× magnification or in C) lung tissue by RT-PCR for Mp P1-adhesin relative to housekeeper. n = at least 15 mice/group, 3 experiments combined, **p<.01.

Figure 2. Recruitment of inflammatory cells is enhanced in Mp-infected SP-A^−/− allergic mice.

A–F) Cells in the BAL were examined by cell surface labeling, as described in the methods section, and flow cytometry 3 days after Mp infection (day 28 of the model). Macs (macrophages), Ex-Macs (exudative macrophages), IMs (inflammatory monocytes), DCs (dendritic cells), PMNs (neutrophils), Eos (eosinophils). n = at least 12 mice/group, 3 experiments combined, *p<.05, **p<.01.

Figure 3. Eosinophil mediators are increased in the absence of SP-A.

On day 28 of the Ova+Mp model, lungs were harvested and A) IL-5 and B) EAR were assessed by RT-PCR. C) Histochemical staining for EPO positive eosinophils was done 1 day after Mp infection and is representative of 3 experiments, n = 5/group. D) Total EPO activity in BAL and lung tissue from Ova+Mp mice was determined via colorimetric assay and absorbance read at 492 nm. n = combined 3 experiments, *p<.05, **p<.01.

Figure 4. SP-A binds to eosinophils in a Calcium-dependent manner.

Eosinophils were incubated with A) fluorescent labeled SP-A-DY649 (shaded = unstained cells, turquoise = 0.1 µg/ml, pink = 1 µg/ml, purple = 10 µg/ml) in calcium-rich buffer or B) fluorescent labeled IgG in calcium-rich buffer (at the same concentrations used above) or C) fluorescent labeled SP-A in calcium depleted buffer and binding was assessed by flow cytometry. SP-A binding (10 µg/ml) to labeled eosinophils was also observed with confocal microscopy as shown in the insets. DAPI (blue stain) and SP-A-DY649 (pink stain) stained slides were examined at 40× via confocal microscopy. D) SP-A binding of the percentage of highly positive cells (gated at 5×10³) was assayed by flow cytometry (geomean of DY649) after cells were pre-incubated with FC antibody (pink line) versus SP-A alone (purple line).

Figure 5. Eosinophil-mediated Mp killing is attenuated by exogenous SP-A.

A) Purified eosinophils were added to Mp (10∶1) for 1 hr and aliquots were diluted on PPLO agar plates for CFU counts. B) Mp and C) Eosinophils were pre-incubated with SP-A (10 µg/ml) for 30 minutes to allow for binding and centrifuged prior to their addition to the killing assay. D–F) SP-D (0.5–5 µg/ml) or SP-A (5–10 µg/ml) was pre-incubated with eosinophils prior to the addition of Mp MOI 10∶1. EPO was measured in the supernatant after 1 hr of stimulation. G) Mp or Eosinophils were pre-incubated with SP-A (10 µg/ml) for 30 minutes to allow for binding and centrifuged prior to their addition to the EPO stimulation assay *p<.05, **p<.01, n = 3 experiments.

Figure 6. Alteration of surfactant levels during inflammation and infection.

A) SP-A or B) SP-D levels were determined by densitometry of Western blots of BAL samples of untreated, Ova only (day 28 of our model) or Ova+Mp (day 28 of our model). n = 2 experiments with 3–4 mice/group each.

Figure 7. EPO is vital for Mp killing mechanisms in vitro and in vivo.

A) Purified human EPO (0.5 µM) was added to Mp (5×10⁶) and viability was assessed over the course of 1 hr. **p<.01,*p<.05. n = 3 experiments. B) Mice were treated with Ova and Mp (O+Mp) as previously described but 2 hrs prior to Mp infection, some mice were given either vehicle (V) or resorcinol (R). Mice were given boosters after 24 hrs and samples harvested at 72 hrs for Mp burden. *p<.05. n = 2 experiments (8–10 mice/group).

Figure 8. SP-A is protective against Mp-induced allergic pathologies.

A) PAS stained lung sections examined at 10× magnification were B) blindly scored for mucus production with 0 representing no mucus present and 5 being mucus present in greater than 75% of airways and bronchioles. C) H&E stained lung sections shows inflammation and RBCs in alveolar spaces and lymphatics of Ova+Mp treated mice. D) Albumin measured in the BAL as a measure of lung damage and vascular permeability. n = representative of 3 experiments, *p<.05.