NAD(P)H quinone oxidoreductase 1 regulates neutrophil elastase-induced mucous cell metaplasia

Abstract

Mucous cell metaplasia (MCM) and neutrophil-predominant airway inflammation are pathological features of chronic inflammatory airway diseases. A signature feature of MCM is increased expression of a major respiratory tract mucin, MUC5AC. Neutrophil elastase (NE) upregulates MUC5AC in primary airway epithelial cells by generating reactive oxygen species, and this response is due in part to upregulation of NADPH quinone oxidoreductase 1 (NQO1) activity. Delivery of NE directly to the airway triggers inflammation and MCM and increases synthesis and secretion of MUC5AC protein from airway epithelial cells. We hypothesized that NE-induced MCM is mediated in vivo by NQO1. Male wild-type and Nqo1-null mice (C57BL/6 background) were exposed to human NE (50 μg) or vehicle via oropharyngeal aspiration on days 1, 4, and 7. On days 8 and 11, lung tissues and bronchoalveolar lavage (BAL) samples were obtained and evaluated for MCM, inflammation, and oxidative stress. MCM, inflammation, and production of specific cytokines, granulocyte-macrophage colony-stimulating factor, macrophage inflammatory protein-2, interleukin-4, and interleukin-5 were diminished in NE-treated Nqo1-null mice compared with NE-treated wild-type mice. However, in contrast to the role of NQO1 in vitro, we demonstrate that NE-treated Nqo1-null mice had greater levels of BAL and lung tissue lipid carbonyls and greater BAL iron on day 11, all consistent with increased oxidative stress. NQO1 is required for NE-induced inflammation and MCM. This model system demonstrates that NE-induced MCM directly correlates with inflammation, but not with oxidative stress.

Fig. 1.

Treatment protocol for neutrophil elastase (NE) aspiration. Wild-type (WT) and Nqo1 (NADPH quinone oxidoreductase 1)-deficient C57BL/6 mice received 3 doses of human NE (50 μg) or control vehicle (PBS) intratracheally via oropharyngeal aspiration (days 1, 4, and 7). Bronchoalveolar lavage (BAL) and lungs were harvested at 2 time points (days 8 and 11).

Fig. 2.

Mucous cell metaplasia (MCM) index post-NE in Nqo1-null and WT mice. C57BL/6 WT and Nqo1-null mice treated with NE or control vehicle were euthanized on day 8 or 11, and the left lung was formalin fixed and sectioned for histology. A: Alcian blue (AB)/periodic acid-Schiff (PAS) staining revealed increased MCM after NE treatment (left: WT control mouse on day 11; right: WT NE-treated mouse on day 11, grade 4 MCM index). Images (×20) are characteristic of each treatment condition. B: histological mucus index for WT and Nqo1-null NE and control treated mice. A minimum of 10 representative cross-sectioned airways were graded per animal. The results are represented graphically as means ± SE; n = 10–14 mice per treatment condition per time point. *P < 0.05, NE treatment is significantly different from control treatment. On day 11, post-NE treatment, WT MCM index is significantly greater than Nqo1-null (P = 0.03).

Fig. 3.

BAL cell counts, differentials, and cytokine levels post-NE in Nqo1-null and WT mice. BAL fluid was obtained from NE-treated or control mice using 3–4 ml of saline. Both WT (C57/BL6) and Nqo1-null mice were utilized. A: total leukocyte counts were obtained by hemocytometer. Cells from BAL were cytocentrifuged on slides, fixed, and stained with Wright-Giemsa stain for differential cell count percentages for macrophages, lymphocytes, neutrophils, and eosinophils. Values are means ± SE; n = 10–14 mice per treatment condition per time point. *P < 0.05, NE treatment was significantly different from control treatment. B: polymorphonuclear leukocytes are represented as percentage of total cell count for WT mice vs. Nqo1-null mice. At least 200 cells were counted. Values are means ± SE; n = 9–14 mice per treatment condition per time point. NE-treated Nqo1-null mice had significantly fewer BAL neutrophils on both days 8 and 11 (P = 0.0014 and P = 0.0001, respectively). Nqo1-null animals had significantly lower levels of macrophage inflammatory protein-2 (MIP-2) (day 11) and granulocyte-macrophage colony-stimulating factor (GM-CSF) (day 8) after NE treatment. Values are means ± SE; n = 7–9 mice per treatment condition per time point. C: eosinophils are represented as percentage of total cell count for WT mice vs. Nqo1-null mice. Values are means ± SE; n = 9–14 mice per pretreatment condition per time point. NE-treated Nqo1-null mice had significantly fewer eosinophils on both days 8 and 11 (P = 0.03 and P = 0.001, respectively). Nqo1-null animals had significantly lower levels of interleukin (IL)-4 (day 11) and IL-5 (day 8) after NE treatment. Values are means ± SE; n = 9–14 mice per treatment condition per time point. *P < 0.05, NE treatment is significantly different from control treatment on both days 8 and 11 for both WT and Nqo1-null mice.

Fig. 4.

Reduced-to-oxidized glutathione ratio in BAL post-NE treatment in Nqo1-null and WT mice. Total glutathione and oxidized glutathione were measured in BAL of NE-treated and control mice (Cayman Glutathione Assay Kit). Reduced glutathione levels were calculated as the difference between total and oxidized glutathione. C57BL/6 WT and Nqo1-null mice treated with NE or control vehicle were lavaged with 3–4 ml of saline at day 11. The ratio of reduced to oxidized glutathione levels is represented graphically as means ± SE; n = 8–14 mice per treatment condition per time point. There was no significant difference in the redox ratio between NE-treated WT and NE-treated Nqo1-null mice.

Fig. 5.

Iron levels in BAL post-NE treatment in Nqo1-null and WT mice. C57/BL6 WT and Nqo1-null mice were treated with NE or control vehicle on days 1, 4, and 7, and BAL (3 ml) was obtained on day 8 or 11. Iron levels were measured in the BAL by inductively coupled plasma optical emission spectroscopy, showing an increase in iron after NE treatment. NE-treated Nqo1-null animals had significantly greater iron levels on day 11 compared with WT (P = 0.0147). The results are represented graphically as means ± SE; n = 6–10 mice per treatment condition per time point. *P < 0.05, NE treatment is significantly different from control treatment. Nqo1-null animals at baseline had a trend for higher iron levels in BAL compared with WT (P = 0.0523).

Fig. 6.

Lipid carbonyls in the BAL and lung tissue post-NE treatment in Nqo1-null and WT mice. Following the NE vs. control vehicle oropharyngeal aspiration protocol, BAL (A) was collected, and murine lungs (B) were harvested and snap frozen on day 8 or 11 and evaluated for thiobarbituric acid-reactive products as an indicator of lipid carbonyl production and lipid peroxidation (32, 37). The results are represented graphically as means ± SE; n = 3–5 mice per treatment condition per time point. A: BAL. B: lung. *P < 0.05, NE treatment is significantly different from control treatment. NE-treated Nqo1-null mice had significantly greater levels of lipid carbonyls in both BAL (P = 0.03) and lung (P = 0.0476) on day 11 compared with NE-treated WT mice.