ROS generated by pollen NADPH oxidase provide a signal that augments antigen-induced allergic airway inflammation

Abstract

Pollen exposure induces allergic airway inflammation in sensitized subjects. The role of antigenic pollen proteins in the induction of allergic airway inflammation is well characterized, but the contribution of other constituents in pollen grains to this process is unknown. Here we show that pollen grains and their extracts contain intrinsic NADPH oxidases. The pollen NADPH oxidases rapidly increased the levels of ROS in lung epithelium as well as the amount of oxidized glutathione (GSSG) and 4-hydroxynonenal (4-HNE) in airway-lining fluid. These oxidases, as well as products of oxidative stress (such as GSSG and 4-HNE) generated by these enzymes, induced neutrophil recruitment to the airways independent of the adaptive immune response. Removal of pollen NADPH oxidase activity from the challenge material reduced antigen-induced allergic airway inflammation, the number of mucin-containing cells in airway epithelium, and antigen-specific IgE levels in sensitized mice. Furthermore, challenge with Amb a 1, the major antigen in ragweed pollen extract that does not possess NADPH oxidase activity, induced low-grade allergic airway inflammation. Addition of GSSG or 4-HNE to Amb a 1 challenge material boosted allergic airway inflammation. We propose that oxidative stress generated by pollen NADPH oxidases (signal 1) augments allergic airway inflammation induced by pollen antigen (signal 2).

Figure 1

Pollen grain extracts show NADPH oxidase activity. (A) Reduction of NBT to formazan by allergenic extracts using NBT assay in the presence (+) or absence (–) of NADPH. RWE, RWE^H, and Amb a 1 were tested. X+XO were used as a positive control. ^#P < 0.001; ^##P < 0.0001. (B) pRWE^OX+-induced NBT reduction is inhibited by NADPH oxidase inhibitors DPI, QA, and SOD. The presence or absence of NADPH in the reaction mixture is indicated. (C) NBT reduction by allergenic extracts in situ after nondenaturing PAGE. RWE^H, heat-inactivated oak extract (Oak^H), and heat-inactivated timothy grass extract (Timothy^H) are shown. (D) Detection of p67^phox by Western blot analysis using a rabbit anti-human p67^phox antibody. (E and F) Immunolocalization of p67^phox in ragweed pollens detected by fluorescence microscopy using anti-p67^phox antibody (E) or normal rabbit IgG control (F). Right panels show differential interference contrast images of the same pollens. Magnification, ×600. (G) Kinetics of O₂^•– generation determined by cytochrome c assay. Shown are cytochrome c (filled diamonds); cytochrome c plus pRWE^OX+ (filled squares); cytochrome c plus NADPH (open triangles); cytochrome c plus pRWE^OX+ and NADPH (open squares); and cytochrome c plus pRWE^OX+, NADPH, and SOD (filled triangles).

Figure 2

RWE increases ROS levels in cultured epithelial cells. (A) Effect of inhibitors of NADPH oxidase on the pollen extract–induced increase in ROS levels in A549 cells. The presence or absence of NADPH in the reaction mixture is indicated. PN, pecan tree extract; TG, timothy grass extract. ^#P < 0.001; ^##P < 0.0001. (B) Induction of intracellular ROS by RWE in various epithelial cells and its inhibition by NAC. Shown are RWE (black bars), RWE plus NAC (white bars), and X+XO (gray bars). (C) Effect of inhibitors of NADPH oxidase on the RWE-induced increase in ROS levels in NHBE cells grown in air-liquid interphase. (D) Effect of protease inhibitors on ROS production induced by HDM extracts and RWE in A549 cells. Shown are cells with (white bars) and without (black bars) protease inhibitors.

Figure 3

RWE increases oxidative stress and its markers in lungs. (A and B) Effect of RWE challenge on markers of oxidative stress GSSG (A) and MDA plus 4-HNE (B) in airway-lining fluids. Shown are PBS (white bars) and RWE (black bars; n = 4–6 per group). Results are means ± SEM. **P < 0.01; ^#P < 0.001. (C and D) Change in ROS levels in lung epithelium of mice after ex vivo challenge as determined by DCF fluorescence (C) or NBT reduction (D). Magnification, ×100. AA, ascorbic acid. (E) RWE challenge did not recruit inflammatory cells at 1 hour after challenge in sensitized mice. Shown are PBS (white bars) and RWE (black bars).

Figure 4

ROS induced by NADPH oxidase are required for allergic lung inflammation and mucin production. (A and B) Removal of NADPH oxidase activity (RWE^H and Amb a 1) inhibits allergic airway inflammation, and surrogate O₂^•– generator X+XO reconstitutes RWE^H- and Amb a 1–induced inflammation to the level of RWE. Results are means ± SEM. n = 7–9 mice per group. (C) Th2 cytokine levels in splenocyte supernatants. Splenocytes from RWE-sensitized mice were cultured with PBS (white bars), RWE (black bars), RWE^H (light gray bars), or Amb a 1 (dark gray bars; n = 4–6 per group) for 96 hours, and supernatants were analyzed by ELISA. Results are means ± SEM. **P < 0.01; ^#P < 0.001; ^##P < 0.0001.

Figure 5

Removal of NADPH oxidase activity from RWE decreases accumulation of eosinophils in peribronchial location and mucin-containing cells in airway epithelium. (A) Immunohistochemical staining of lung cryosections shows accumulation of eosinophils in peribronchial regions of lungs. (B) Total eosinophil area in the peribronchial region was quantified by morphometric analysis. (C and D) Periodic acid-Schiff staining for mucin in representative lung sections. Magenta-colored epithelial cells are positive for mucin. Magnification, ×100. (D) Total area of mucin-containing cells in peribronchial area was quantified by morphometric analysis. (E) Coadministration of X+XO with Amb a 1 increases the number of mucus cells in airway epithelium. Magnification, ×100. Results are means ± SEM. n = 3–6 mice per group. ^##P < 0.0001.

Figure 6

Removal or neutralization of intrinsic pollen NADPH oxidase of RWE reduces IgE levels and extent of allergic lung inflammation. (A) pRWE^OX–-specific IgE levels in serum. Mice were sensitized with RWE intraperitoneally and challenged intranasally with PBS or equal protein concentrations of either pRWE^OX– or RWE. pRWE^OX–-specific IgE was quantified by ELISA. (B and C) RWE-sensitized mice were challenged with RWE or RWE plus QA. Adding QA inhibited RWE-induced eosinophil (B) and total inflammatory cell (C) recruitment in the airways. (D) Challenge with active NADPH fraction (pRWE^OX+ plus pRWE^OX–), but not inactive NADPH oxidase fractions (heat-inactivated pRWE^OX+ [pRWE^OX+H] plus pRWE^OX– or pRWE^OX+ plus pRWE^OX– and Tiron), induces eosinophil recruitment in the airways of RWE-sensitized animals. Results are means ± SEM. n = 5–7 mice per group. *P < 0.05; **P < 0.01; ^#P < 0.001.

Figure 7

Impact of signal 1 on airway inflammation. (A) Recruitment of inflammatory cells induced by challenge with pRWE^OX+ and pRWE^OX– in naive mice at 24 hours. (B and C) Effect of GSSG and 4-HNE challenge on recruitment of neutrophils (B) and total inflammatory cells (C) in naive mice at 24 hours. (D) GSSG and 4-HNE induce tyrosine phosphorylation (P) of p38 MAPK in cultured A549 cells in a time-dependent manner. (E and F) GSSG and 4-HNE induce IL-8 production (E) and IL-8 promoter activity (F) in A549 cells. (G and H) GSSG and 4-HNE augment Amb a 1–induced recruitment of eosinophils (G) and total number of inflammatory cells (H) in RWE-sensitized mice.

Figure 8

Presence of signal 2 (antigen) is required to induce allergic airway inflammation. RWE induces significantly higher inflammation in the airways in WT mice (capable of processing antigen [signal 2] and redox activity [signal 1]; black bars) at 72 hours compared to that in SCID mice (capable of processing redox activity [signal 1] only; white bars). *P < 0.05; ^#P < 0.001.