Transgenic expression of human S100A12 induces structural airway abnormalities and limited lung inflammation in a mouse model of allergic inflammation

Abstract

Background: The calcium-binding protein S100A12 is highly up-regulated in the serum and sputum of patients with allergic asthma and is suggested to be a biomarker and pathologic mediator of asthma.

Objective: To test the role of S100A12 in mediating airway inflammation in a mouse model of allergic lung inflammation.

Methods: Transgenic (TG) mice that express human S100A12 and wild-type (WT) littermates were sensitized and challenged with ovalbumin (OVA) and assessed for inflammation, lung structure, and function.

Results: Following OVA sensitization and challenge, S100A12 TG mice showed reduced peribronchial and perivascular inflammation, mucus production, and eosinophilia as well as attenuated airway responsiveness to contractile agonist compared with WT sensitized and challenged animals. This is explained, at least in part, by remodelled airways in S100A12 TG mice with thinning of the airway smooth muscle. S100A12 exposure induced Fas expression and activation of caspase 3 in cultured airway smooth muscle cells, suggesting that airway smooth muscle abnormalities observed in S100A12 TG mice may be mediated through myocyte apoptosis.

Conclusion and clinical relevance: S100A12 is one of the most abundant proteins found in the airways of human asthmatics, and it was postulated that S100A12 could mediate the inflammatory process. Our study shows for the first time that TG expression of S100A12 in the lung of mice does not exacerbate lung inflammation in a model of OVA-induced allergic inflammation. We speculate that the high levels of S100/calgranulins found in bronchoalveolar lavage fluid of asthmatics and of OVA-treated TG S100A12 mice do not significantly mediate pulmonary inflammation.

Figure 1

Characterization of S100A12 TG mice. (A) RT-PCR detects S100A12 in the lung, trachea, pulmonary artery, and aorta of TG mice but not WT mice. (B) Expression of endogenous murine S100A8 protein is unaltered in peripheral blood mononuclear cells (PBMC) and lung of TG and WT mice. (C) OVA sensitization and challenge protocol to induce lung inflammation. (D) In response to OVA aerosol challenge, S100A12 is present in the BALF and serum of OVA-sensitized S100A12 TG mice and not in TG-control or WT mice.

Figure 2

Attenuated lung inflammation in TG-OVA mice. (A and B) H&E stained lung section in 1.25× magnification and 20× magnification; scale bar 10 µm. Arrows and arrowheads show peribronchial and perivascular and inflammation, respectively. (C) Quantification of the peribronchial and (D) of perivascular inflammation (* p<0.05). (E) Analysis of inflammatory cells in BALF. TG-control mice have increased macrophage counts compared to WT-control mice (*p<0.05), and TG-OVA mice have less eosinophila than WT-OVA mice (**p<0.01). (F and G) Representative lung sections stained with PAS (arrow marks PAS positive cells; 40× magnification; scale bar, 10µ). (H) Quantification of PAS positive cells (*p<0.01).

Figure 3

S100A12-TG mice produce comparable levels of OVA-specific serum IgG1.

Figure 4

Bronchoconstrictor responsiveness is attenuated in S100A12 TG mice. Respiratory system resistance was measured after bolus i.v. injections of methacholine (* p<0.001 WT-OVA vs.TG-OVA, and # p<0.05 TG-control vs. WT-control).

Figure 5

S100A12-TG mice exhibit lung structural abnormalities. (A) Reduction of airway smooth muscle and dilated and irregularly shaped large airways in S100A12-TG mice is evident compared to WT mice. Upper panel shows H&E stains at 2.5× magnification and the six lower panels inserts show representative images of the proximal (1), mid (2) and distal (3) airway segments at 40× magnification; scale bar, 10 µm. Arrows indicate the airway smooth muscle. (B) Masson trichrome stain showed enhanced peri-vascular fibrosis in S100A12 TG lungs. Hydroxyproline content in the lungs is shown on the right (* p<0.05). (C) Verhoeff van Giessan stain revealed thinning and degradation of elastic fibers in S100A12 TG airway (arrows indicate elastic fibers). Original magnification 40× for the proximal airway, and 100× for the distal airway.

Figure 6

S100A12 modulates chemokine secretion in activated HASMC and is pro-apoptotic. (A) quantitative RT-PCR shows increased mRNA for CXCL10 and CCL-9 in cytokine stimulated HASMC (0.1ng/ml TNF-α and 10ng/ml INF-γ for 6 hours, * p=0.001 vs baseline), that is attenuated in cells expressing S100A12 (p=0.018 and p<0.01 vs mock transfected). Data are expressed as mean ±SEM. (B) ELISA for CXCL10 protein measured in cell culture supernatant of HASMC stimulated as indicated. Data are expressed as mean ±SEM. (C) qRT-PCR shows increased mRNA of Fas, caspase 10 and caspase 3 in HASMC transfected with S100A12, but not mock transfected cells (*p<0.05). (D) Western blot revealed activation of caspase 3 in S100A12 transfected cells compared to mock transfected control cells. Active caspase 3 abundance was further increased by co-stimulation with Fas-clone CH11 antibody (50 ng/ml for 48 hrs, *p< 0.01).