NADPH:Quinone Oxidoreductase 1 Regulates Host Susceptibility to Ozone via Isoprostane Generation*

Background: NQO1 regulates pulmonary susceptibility to ozone. Results: In NQO1-null mice, ozone exposure generates precursors of A2-isoprostane in the lung. A2-isoprostane suppresses ozone-induced IL-8 expression, inhibits NF-κB, and modifies Cys179 in IKK. Conclusion: A2-isoprostane inhibits ozone-induced NF-κB activation via IKK inhibition. Significance: This molecular mechanism explains the paradoxical observation that loss of NQO1 protects from ozone toxicity. NADPH:quinone oxidoreductase 1 (NQO1) is recognized as a major susceptibility gene for ozone-induced pulmonary toxicity. In the absence of NQO1 as can occur by genetic mutation, the human airway is protected from harmful effects of ozone. We recently reported that NQO1-null mice are protected from airway hyperresponsiveness and pulmonary inflammation following ozone exposure. However, NQO1 regenerates intracellular antioxidants and therefore should protect the individual from oxidative stress. To explain this paradox, we tested whether in the absence of NQO1 ozone exposure results in increased generation of A2-isoprostane, a cyclopentenone isoprostane that blunts inflammation. Using GC-MS, we found that NQO1-null mice had greater lung tissue levels of D2- and E2-isoprostanes, the precursors of J2- and A2-isoprostanes, both at base line and following ozone exposure compared with congenic wild-type mice. We confirmed in primary cultures of normal human bronchial epithelial cells that A2-isoprostane inhibited ozone-induced NF-κB activation and IL-8 regulation. Furthermore, we determined that A2-isoprostane covalently modified the active Cys179 domain in inhibitory κB kinase in the presence of ozone in vitro, thus establishing the biochemical basis for A2-isoprostane inhibition of NF-κB. Our results demonstrate that host factors may regulate pulmonary susceptibility to ozone by regulating the generation of A2-isoprostanes in the lung. These observations provide the biochemical basis for the epidemiologic observation that NQO1 regulates pulmonary susceptibility to ozone.

In epidemiologic studies, ozone levels directly correlate with emergency department visits for asthma, school absences, and hospitalization rates (1)(2)(3). Ozone inhalation causes several pulmonary responses including increased cough, chest pain, decreased forced expiratory volume in 1 s, airway hyperreactivity (4), and increased airway inflammation, specifically neutrophilic inflammation (1,5). Ozone triggers these physiologic sequelae via the generation of reactive oxygen species that act as potent intracellular signals.
Ozone generates lipid ozonation products including aldehydes, hydroxyhydroperoxides, and hydrogen peroxide (6) in the airway. Lipid ozonation products and downstream reactive oxygen species activate transcription factors including nuclear factor from B cells (NF-B) 2 (7,8). NF-B activation is essential for pulmonary inflammation following ozone exposure (9,10). Ozone triggers NF-B by activating the IB kinase (IKK) complex, resulting in phosphorylation of IB␣, which is degraded, resulting in the release and nuclear translocation of the NF-B heterodimer, which often includes p65/RelA (11). Ozone exposure of human airway epithelial cells in vitro results in increased production of inflammatory mediators including granulocyte/macrophage colony-stimulating factor, TNF-␣, and IL-8 (12)(13)(14).
An NADPH:quinone oxidoreductase 1 (NQO1) polymorphism (15,16) has been reported in epidemiologic studies to affect pulmonary susceptibility to ozone. NQO1 is highly expressed in airway epithelia (17) and catalyzes the reduction of quinones such as ubiquinone and ␣-tocopherone to regenerate antioxidant capacity and prevent lipid peroxidation of cellular membranes (18,19). One NQO1 single nucleotide polymorphism, 609C3 T (Ser 187 variant) accelerates degradation of NQO1, resulting in functional loss of the enzyme. Importantly, this NQO1 polymorphism, 609 TT (Ser-Ser 187 ), has a protective effect against asthma in children with GSTM1-null genotype and high lifetime ozone exposure (20) and protects subjects from airway obstruction in high ozone environments (16,20,21).
Consistent with this epidemiologic data, we reported that compared with congenic C57BL/6 wild-type mice NQO1-null mice were protected from ozone-induced airway hyperresponsiveness and airway inflammation. Paradoxically, following ozone exposure, NQO1-null mice produced significantly less F 2 -isoprostane, a non-enzymatic peroxidation product of arachidonic acid that is a stable biomarker of oxidant stress (22). These results were unexpected because absence of NQO1 should increase oxidative stress postozone exposure. Therefore, we investigated why there was a difference in F 2 -isoprostane production in wild-type versus NQO1-null mice and whether the difference in isoprostane production affected pulmonary responses to ozone.
Isoprostanes (isoPs) are formed by the non-enzymatic peroxidation of arachidonic acid. They are named according to the prostaglandin isomer closest in structure. F 2 -isoPs (isomers of PGF 2␣ ) are chemically stable molecules that are accepted as a "gold standard" biomarker of endogenous oxidative stress. IsoPs with an E 2 or D 2 ring are also generated in vivo (23) depending upon the reduction-oxidation (redox) status of the cell (24) and spontaneously undergo dehydration to form the cyclopentenone isoPs, known as A 2 -and J 2 -isoPs, respectively. A 2 -/J 2 -isoPs contain a highly reactive ␣,␤-unsaturated carbonyl group on the cyclopentenone ring that readily adducts thiols via Michael addition, which allows these molecules to exert potent biological activities. For example, A 2 -isoPs have been reported to inhibit activation of NF-B, suggesting that they have an anti-inflammatory function (25,26).
In this study, we tested whether NQO1-null mice generated greater levels of E 2 /D 2 -isoPs, the A 2 /J 2 precursors, following ozone exposure compared with control mice. We also tested whether A 2 -isoPs inhibited ozone-triggered NF-B activation and IL-8 expression in normal human bronchial epithelial (NHBE) cells. Our results suggest a novel biochemical mechanism to explain the protection afforded by loss of NQO1 against ozone-induced pulmonary inflammation.

Cell Culture
The primary NHBE cells were harvested from human tracheobronchial tissues from donors obtained from the Lung Transplant Program and the Department of Pathology, Duke University Medical Center. The protocol was approved by the Institutional Review Board for Clinical Investigations, Duke University Medical Center. Cells were plated as described previously on 6-well or 12-well Transwell Clear chambers (Corning, Corning, NY) in a serum-free growth factor-supplemented medium with all-trans-retinoic acid (27,28) and cultured in air-liquid interface (ALI) culture conditions. Experiments were performed on days 10 -14 after the change of the culture condition to ALI. NHBE cells were changed to medium supplemented with retinoic acid for 24 h before ozone exposure. Cells were preincubated with A 2 -isoPs (provided by Dr. Ginger Milne or Cayman Biochemical; 50 nM or 2.5 M) or the equivalent volume of control vehicle (N 2 -purged ethanol) and then co-incubated with ozone or filtered air (FA) (see "Ozone Exposure" below).

Mouse Models
C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME). A breeding colony was established at Duke University from breeding pairs of NQO1-null mice (on a C57BL/6J background) obtained from Dr. Frank Gonzalez at the National Cancer Institute (Bethesda, MD). Male C57BL/6J or NQO1-null mice were used at 6 -8 weeks of age. Experimental protocols were approved by the Institutional Animal Care and Use Committee at Duke University Medical Center and performed in accordance with the standards established by the United States Animal Welfare Act.

Ozone Exposure
Mice were exposed to either ozone (1 ppm) or filtered air for 3 h as reported previously (22). The ozone concentration in the chamber was continuously monitored with an ozone ultraviolet light photometer (Dasibi model 1003AH, Dasibi Environmental Corp., Glendale, CA). Mouse lungs were harvested 24 h after ozone or filtered air exposure, snap frozen, and stored at Ϫ80°C until lipid extraction for E 2 -/D 2 -isoprostane analysis.
NHBE cells were exposed to either FA or 0.4 ppm ozone from 30 min to 5 h in exposure chambers; each gas was provided at 20 liters/min balanced with 5% CO 2 and at 88% relative humidity. Immediately after the exposure, cells were processed for isolation of total RNA, cell lysate protein, or collection of basolateral medium. By lactate dehydrogenase assay, ozone does not cause cytotoxicity at this concentration or duration of exposure. ELISA (LEGEND MAX kit for human IL-8, BioLegend, San Diego, CA) was performed for IL-8 quantitation in basolateral medium diluted 1:5 or 1:10 according to the manufacturer's instructions.

Analysis of E 2 -/D 2 -isoprostanes
D 2 -/E 2 -isoprostanes were analyzed by gas chromatographynegative ion chemical ionization-mass spectrometry (GC-NICI-MS) using a modification of methods described previously for the analysis of PGE 2 and PGD 2 (29). Briefly, 1.5 ng of [ 2 H 4 ]PGE 2 internal standard was initially added to a biological fluid and adjusted to pH 3 with 1 M HCl. The sample was then applied to a C 18 Sep-Pak cartridge, and compounds were eluted with 10 ml of ethyl acetate and evaporated to dryness under nitrogen. Compounds were subsequently derivatized by treatment with a 2% solution of aqueous methoxyamine⅐HCl and extracted with ethyl acetate, and the organic layer was evaporated under nitrogen. Compounds were converted to a pentafluorobenzyl ester and dried under nitrogen, and the residue was reconstituted in 30 l of chloroform and 20 l of methanol and chromatographed on a silica TLC plate to 13 cm in a solvent system of ethyl acetate:methanol (98:2, v/v). The methyl ester of PGF 2␣ and the 0-methyloxime, pentafluorobenzyl ester derivative of PGD 2 (ϳ5 g each) were chromatographed on a separate lane and visualized with 10% phosphomolybdic acid in ethanol by heating. The R f of PGF 2␣ methyl ester in this solvent system is ϳ0. 25, and the R f of the 0-methyloxime, pentafluorobenzyl ester derivative of PGF 2 is ϳ0.60. Compounds migrating in the region 1 cm above the PGF 2 standard to 0.5 cm below the PGF 2 standard were scraped from the TLC plate, extracted with 1 ml of ethyl acetate, and dried under nitrogen. Following TLC purification, compounds were converted to trimethylsilyl ether derivatives and dried under nitrogen. The residue was redissolved for GC-MS analysis in 10 l of undecane that had been stored over a bed of calcium hydride.
GC-NICI-MS was carried out on an Agilent 5973 Inert Mass Selective Detector coupled with an Agilent 6890n Network GC system interfaced with an Agilent computer. The GC was performed using a 15-m, 0.25-mm-film thickness, DB-1701 fused silica capillary column (J&W Scientific, Folsom, CA). The column temperature was programmed from 190 to 300°C at 20°C/min. The major ion generated in the NICI mass spectrum of the pentafluorobenzyl ester, trimethylsilyl ether derivatives of the E 2 /D 2

Total RNA Collection and Real Time Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
RNA was isolated from NHBE cells using TRIzol (Invitrogen) according to the manufacturer's instructions, and IL-8 gene expression was analyzed by quantitative real time RT-PCR on an SDS 7300 machine (Applied Biosystems) in a 25-l reaction that contained 1ϫ TaqMan gene expression assay (mixture of PCR primer set specific for IL-8 and a FAM dye-labeled Taq-Man MGB probe) using universal amplification conditions. The amplification reaction of 18 S rRNA control contained 0.2 M forward primer, 0.2 M reverse primer, and 0.8 M VIC dye-labeled Taq-Man MGB probe. Each sample was amplified in duplicate reactions for both the gene of interest and 18 S rRNA control. The relative gene expression level was calculated by the ⌬⌬CT method, which represents the -fold difference in gene expression corrected for 18 S rRNA control expression, and normalized to the control treated sample (28,30).

Isolation of Cellular Proteins
Ozone-exposed cells were rinsed with ice-cold PBS containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and lysed for 5 min on ice in Cell Lysis Buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM glycerol phosphate, 1 mM Na 3 VO 4 , 1 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride) (Cell Signaling Technology, Beverly, MA) containing both 1ϫ Complete protease (Roche Applied Science) and 1ϫ phosphatase inhibitor mixtures (Sigma). Cultures were then scraped and disrupted mechanically by pipetting and sonication. Samples were centrifuged at 14,000 rpm for 20 min at 4°C using a tabletop microcentrifuge. Total cellular protein concentrations in each supernatant were determined by Bio-Rad DC Protein Assay.

Western Blot Analysis
To determine the activation of the NF-B pathway in NHBE cells exposed to ozone versus filtered air, the phosphorylation of subunit p65 of NF-B was evaluated. Equal amounts of total proteins (60 -100 g/lane) were separated by electrophoresis on a 10% sodium dodecyl sulfate-polyacrylamide gel and then transferred onto nitrocellulose membranes for Western analysis. Nonspecific binding was blocked with 5% milk powder in TBST (20 mM Tris-HCl (pH 7.6), 150 mM NaCl, and 0.1% Tween 20) for 1 h at room temperature followed by incubation with primary rabbit polyclonal antibody specific to phospho-p65 (Ser 536 ) (diluted 1:1000; Cell Signaling Technology) overnight at 4°C. After washing, membranes were incubated with horseradish peroxidase-conjugated secondary antibody (antirabbit IgG HRP-linked antibody, 1:2000; Cell Signaling Technology) for 1 h at room temperature. After two 20-min washes in TBST, immunoreactive protein complexes were detected by enhanced chemiluminescence (GE Healthcare). To confirm equal loading, blots were rehybridized with mouse monoclonal antibody specific to ␤-actin (1:5000; Sigma-Aldrich).

IKK Activity Assay
The IKK enzyme activity was evaluated in the proteins isolated from ozone (OZ)-or FA-exposed cells. The IKK enzyme was selectively immunoprecipitated from 500 g of total cellular protein in a 500-l reaction volume of Cell Lysis Buffer containing both 1ϫ Complete protease and 1ϫ phosphatase inhibitor mixtures using rabbit polyclonal antibodies that recognized IKK␣ (1:250) and IKK␤ (1:50) (Cell Signaling Technology) or rabbit IgG as control. Following overnight immunoprecipitation at 4°C, the immune complexes were then precipitated with a 25-l volume of magnetic Dynabeads protein G (4°C, 4 h) (Invitrogen). Beads were washed twice with lysis buffer and then twice with kinase reaction buffer (25 mM Tris-HCl (pH 7.5), 5 mM ␤-glycerophosphate, 2 mM dithiothreitol (DTT), 0.1 mM Na 3 VO 4 , and 10 mM MgCl 2 ) (Cell Signaling Technology). The IKK activity assay was initiated by the addition of kinase buffer containing 200 M ATP and 1 g of GST-IB␣ as a substrate for the IKK enzyme activity. The kinase reactions were incubated at 30°C for 30 min and then stopped by addition of SDS sample loading buffer. The reaction mixture was boiled for 5 min, and aliquots were resolved by 10% SDS-PAGE followed by Western analysis using antibodies to phospho-IB␣ (Ser 32/36 ) (mouse monoclonal, 1:1000 dilution; Cell Signaling Technology) and horseradish peroxidase-conjugated secondary antibody (sheep anti-mouse IgG HRP-linked antibody, 1:2000; GE Healthcare). After two 20-min washes in TBST, immunoreactive protein complexes were detected by enhanced chemiluminescence. IKK activity was determined by the intensity of phospho-GST-IB␣ (Ser 32/36 ) signals on the autoradiogram. The blots were rehybridized with antibodies against IKK␤ (rabbit polyclonal antibody, 1:1000 dilution; Cell Signaling Technology) followed by incubation with horseradish peroxidase-conjugated secondary antibody (goat anti-rabbit IgG HRP-linked antibody, 1:2000; Cell Signaling Technology), and immunoreactive protein complexes were detected by enhanced chemiluminescence. The band densities of phos- FEBRUARY 15, 2013 • VOLUME 288 • NUMBER 7 JOURNAL OF BIOLOGICAL CHEMISTRY 4683 phorylated IB␣ were normalized to the corresponding immunoprecipitated IKK␤ band densities.

IKK Cys 179 Peptide in Vitro Modification Analysis by Mass Spectrometry
Peptide Preparation-To investigate the modification mechanism of the reactive Cys 179 residue of IKK␤ peptide with isoprostanes, two synthetic 27-amino acid peptides encompassing Glu 172 -Lys 198 of the primary sequence of human IKK␤ containing Cys 179 residue (WT; M r 3009.500) and C179A (MUT; M r 2977.528) were obtained commercially (Invitrogen). These peptides were dissolved in nitrogen-purged N,N-dimethylformamide (5 g/l) and stored at Ϫ80°C. An aliquot of 10-l (50 g; 166 M) was further diluted in PBS to obtain a 100-l reaction volume in a shallow dish and exposed to ozone (0.4 ppm; 30 min) or filtered air in the presence or absence of A 2 -isoprostane (166 M) or as a control F 2 -isoprostane (166 M). Peptides under each of the conditions were frozen in aliquots at 1 mg/ml at Ϫ80°C for mass spectrometry analysis in the Duke Proteomics Core Facility. Upon thawing, 5 l of each peptide was transferred into a separate tube, and 15 l of 0.1% trifluoroacetic acid (TFA) was added to each. Millipore C 18 ZipTips were used according to the manufacturer's protocol to desalt 5 g of each peptide sample. The samples were vacuum-dried and dissolved in 100 l of 2:98: LC-MS Data Collection-Each of the six peptide exposure conditions were analyzed using a nanoAcquity UPLC system coupled to a Synapt G2 HDMS mass spectrometer (Waters Corp., Milford, MA). Approximately 50 ng of peptide material in 1 l was first trapped at 5 l/min for 3 min in 99.9% water with 0.1% (v/v) formic acid on a 20-m ϫ 180-mm Symmetry C 18 column. Separations were then performed on a 75-m ϫ 250-mm column with 1.7-m C 18 Bridged Ethyl Hybrid (BEH) particles (Waters) using a 30-min gradient of 5-40% acetonitrile with 0.1% formic acid at a flow rate of 0.4 l/min and 55°C column temperature. The samples were then subjected to MS/MS acquisition in data-dependent analysis mode using a 0.9-s MS scan followed by MS/MS acquisition on the top three ions with charge greater than 1. MS/MS scans for each ion used an isolation window of ϳ3 Da, a maximum of 3 s per precursor, and dynamic exclusion for 120 s within 1.2 Da.
LC-MS Data Processing-The resulting raw data files were processed using Mascot Distiller, and processed spectra were submitted to the Mascot v2.2 (Matrix Sciences, Inc.) search engine. The data were searched against the current Swiss-Prot database with Homo sapiens taxonomy, semitryptic enzyme specificity, precursor ion mass tolerance of 10 ppm, and product ion tolerance of 0.04 Da. Dynamic peptide modifications allowed were single, double, or triple oxidation on cysteine, cysteine mutation to alanine (Ϫ31.972 Da), and oxidized (334.2144 Da) or reduced (336.230 Da) isoprostane modification of cysteine. Database search results for all conditions are available upon request.

Statistical Analysis
Except for the analysis of A 2 -isoP modification of the IKK peptide, all other experimental results were expressed as mean Ϯ S.E., and differences between mean values were ana-lyzed by one-way analysis of variance with post hoc analysis by the Wilcoxon rank sum test (Statistix 8, Analytical Software, Tallahassee, FL). p values Ͻ0.05 were considered statistically significant.

NQO1-null Mice Had Greater Levels of E 2 -/D 2 -isoPs (A 2 -/J 2 -isoP Precursors) Both at Base Line and Postozone Exposure
Compared with Wild-type Mice-We reported previously that ozone exposure increases bronchoalveolar lavage F 2 -isoP levels, a marker of oxidative injury, in wild-type mice but not in NQO1-null mice (22). In this report, we measured by GC-NICI-MS murine lung tissue levels of E 2 -/D 2 -isoPs in wildtype and NQO1-null mice after filtered air or ozone exposure. There was increased generation of E 2 -/D 2 -isoPs in NQO1-null mice compared with wild-type mice both after filtered air and after ozone exposure (Fig. 1). These results demonstrate that the absence of NQO1 favored the generation of E 2 -/D 2 -isoP in murine lung tissue.
A 2 -isoP Blocked Ozone-induced IL-8 mRNA Expression in Human Airway Epithelial Cells-To determine the effect of ozone exposure on proinflammatory cytokine expression in human airway epithelial cells, we used quantitative RT-PCR to measure IL-8 mRNA expression in NHBE cells cultured at the air-liquid interface and exposed to filtered air or ozone (0.4 ppm) over time (ranging from 30 min to 5 h) ( Fig. 2A). IL-8 mRNA levels were normalized to 18 S rRNA levels and expressed as a percentage of the 30-min filtered air control treatment condition. Ozone exposure increased IL-8 expression in a time-dependent manner with a significant increase in IL-8 expression by 1 h of ozone exposure and a progressive increase in IL-8 expression up to 5 h.
To explore the anti-inflammatory effect of A 2 -isoP in ozone-exposed NHBE cells, we examined the effect of A 2 -isoP on ozone-induced IL-8 gene expression (Fig. 2,  B-E). We first pretreated epithelial cells with two different concentrations of A 2 -isoP (50 nM and 2.5 M) for 1 h and then exposed cells to ozone or filtered air for 2 or 5 h. IL-8 FIGURE 1. Isoprostane quantitation in murine lungs. C57BL/6 wild-type and NQO1-null mice were exposed to filtered air or OZ (1 ppm; 3 h) and then euthanized 24 h later for lung tissue collection. Mouse lungs were snap frozen, stored at Ϫ80°C, and then extracted for E 2 -and D 2 -isoprostane quantification by GC-NICI-MS. E 2 -and D 2 -isoprostane quantitation is presented as ng/g of lung tissue (mean Ϯ S.E. (error bars); n ϭ 13-14 animals/group). *, wild-type OZ significantly different from wild-type FA control, p Ͻ 0.05; #, NQO1-null FA significantly different from wild-type FA control, p Ͻ 0.05.

NQO1 Regulates Host Susceptibility to Ozone via A 2 -isoP
mRNA expression was significantly up-regulated by ozone exposure both at 2 (Fig. 2, B and C) and 5 h (Fig. 2, D and E).

. Time course of ozone-induced IL-8 mRNA expression in NHBE cells in the presence or absence of A 2 -isoprostane.
NHBE cells cultured at ALI were exposed to FA or OZ over time (0.4 ppm; 0.5-5 h), and total RNA was isolated using TRIzol reagent. IL-8 gene expression was determined using quantitative RT-PCR and normalization to 18 S rRNA, and expression was calculated by the ⌬⌬CT method (A). Results were expressed as a percentage of the 30-min filtered air control (mean Ϯ S.E. (error bars); n ϭ 10). *, OZ significantly different from corresponding FA control, p Ͻ 0.05. NHBE cells cultured at ALI were exposed to filtered air (control) or ozone (0.4 ppm; 2 (B and C) or 5 h (D and E)) in the presence or absence of A 2 -isoprostanes (A2) (50 nM or 2.5 M) for 1 h prior to and then during the exposure. Total RNA was isolated using TRIzol. IL-8 gene expression was determined using quantitative RT-PCR and normalization to 18 S rRNA, and expression was compared with control treatment levels by the ⌬⌬CT method (mean Ϯ S.E. (error bars); n ϭ 6 -7). *, significantly different from control, p Ͻ 0.05; #, ozone ϩ A 2 -isoprostanes significantly different from ozone alone, p Ͻ 0.05. FEBRUARY 15, 2013 • VOLUME 288 • NUMBER 7 JOURNAL OF BIOLOGICAL CHEMISTRY 4685 exposed to filtered air or ozone (0.4 ppm) for 5 h in the presence or absence of A 2 -isoP (2.5 M), and basolateral medium was collected for ELISA determination of IL-8 (Fig. 3). Ozone significantly increased IL-8 protein levels in NHBE cells from two different donors, and exposure to A 2 -isoP (2.5 M) blocked this release.

A 2 -isoP Abrogated Ozone-induced NF-B Activation in Normal Human Bronchial Epithelial Cells-NF-B
is a major transcription factor activated by ozone and required for up-regulation of IL-8 expression (7). We examined the effect of A 2 -isoP on ozone activation of the NF-B pathway in NHBE cells using two complementary methods. First, we examined activation/ phosphorylation of one component of the NF-B heterodimer, p65, by Western analysis (Fig. 4), and second, we evaluated the activity of the IKK complex by using the same cell lysates to immunoprecipitate IKK␣ and IKK␤ and then perform a kinase assay to evaluate IKK activity-phosphorylation of substrate IB␣ detected by Western analysis (Fig. 5).
Following NHBE cell exposure to filtered air or ozone (0.4 ppm; 30, 60, or 120 min), total protein lysates were evaluated by Western analysis for p65 phosphorylation (Fig. 4A). The results demonstrated a time-dependent increase in phosphorylation of p65 that was evident at 30 and 60 min but not present at 120 min of ozone exposure (Fig. 4, A and B). No changes in phosphorylation of p65 were observed in cells exposed to filtered air, and ozone exposure did not up-regulate total p65 concentrations in cell lysates.
We next investigated the effect of A 2 -isoP (50 nM and 2.5 M) on ozone-induced NF-B activation by measuring the phosphorylation of p65 both at 30 and 60 min of ozone exposure (Fig. 4, C-F). Ozone-induced p65 phosphorylation in NHBE cells was significantly attenuated by co-incubation with A 2 -isoP at both concentrations and at both 30 (Fig. 4, C and D) and 60 min (Fig. 4, E and F) of ozone exposure. A 2 -isoP alone with filtered air exposures did not affect p65 phosphorylation at either A 2 -isoP concentration or time point.
To further confirm the activation of the NF-B pathway, we analyzed the effect of ozone on IKK activity. IKK␣ and IKK␤ are the catalytic moieties of the IKK complex, and their activity is required for IB␣ phosphorylation, degradation, and release of NF-B for its translocation to the nucleus. To test activation of the IKK complex by ozone and potential inhibition by A 2 -isoP, IKK␣ and IKK␤ were immunoprecipitated from total protein lysates from cells exposed to filtered air or ozone (0.4 ppm; 30 and 60 min), and their activities were determined by in vitro kinase assays with a fusion protein, GST-IB␣, as substrate. This substrate contains the specific phosphorylation sites Ser 32 and Ser 36 and eliminates the possibility of nonspecific phosphorylation at the C terminus of full-length IB␣ protein. Specificity was ensured by using normal rabbit IgG as a negative control, which failed to precipitate IKK␣ or IKK␤. In agreement with results demonstrating phosphorylation of p65, ozone exposure increased IKK activity at 30 and 60 min but not 120 min in NHBE cells (Fig. 5A). There was no change in total IKK expression following ozone exposure.
We then tested IKK activity in the presence or absence of A 2 -isoP following filtered air or ozone exposure (0.4 ppm; 30 and 60 min) of NHBE cells. A 2 -isoP at both 50 nM and 2.5 M significantly inhibited ozone-induced IKK activity at both 30 (Fig. 5, B and C) and 60 min (Fig. 5, D and E). Changes in activity were not due to loss of IKK as there was no change in IKK expression levels in NHBE cells under any treatment condition. These data suggest that co-incubation of A 2 -isoP inhibited IKK activity and subsequent phosphorylation of p65, thereby preventing NF-B nuclear translocation and transcriptional upregulation of IL-8 in ozone-exposed NHBE cells.  (31). To determine whether it is possible that A 2 -isoP inactivates IKK by oxidation/Michael adduction of the sulfhydryl group, a 27-amino acid synthetic IKK␤ peptide Glu 172 -Lys 198 encompassing the Cys 179 residue, was exposed to filtered air or ozone (0.4 ppm; 30 min) in the presence or absence of equimolar amounts of A 2 -isoP or F 2 -isoP. Using nanoscale LC-MS/MS analysis, we evaluated whether any of these treatment conditions chemically modified the Cys 179 residue of the synthetic peptide (Fig. 6). The isotopic distribution of the experimentally obtained spectrum for the intact, Cys 179 -modified peptide precisely matched the theoret-

NQO1 Regulates Host Susceptibility to Ozone via A 2 -isoP
ical spectrum due to A 2 -isoP modification (Fig. 6A). Identification of the modified species with m/z 1115.578 was obtained by MS/MS, and comparison of the MS/MS spectrum of the modified and unmodified (m/z 1004.174) species showed the same peptide backbone (Fig. 6B). When ozone was added, a notable increase in the quantity of cysteic acid (triply oxidized Cys 179 ) was observed with or without the presence of A 2 -isoP (data not shown). Neither the singly or doubly oxidized Cys 179 peptide was observed in this experiment. The peptide was modified by A 2 -isoP in the absence of ozone only to 10% of the ϩozone level and was not modified by F 2 -isoP in the presence or absence of ozone (Fig. 6C), suggesting that this reaction is specific to the A 2 -isoP maleamide in an ozone-catalyzed manner. The Cys residue was required because the mutant peptide C179A was not modified by ozone Ϯ A 2 -isoP (data not shown). The pro-posed structure corresponding to the A 2 -isoP-modified IKK peptide as sequenced in this experiment is shown in Fig. 6D.

Absence of NQO1 Regulates Host Responses to OZ by Increasing Production of A 2 -isoprostane-Epidemiologic studies have
focused attention on several potential candidate gene polymorphisms that influence ozone-activated inflammation and airway hyperresponsiveness. In this report, we present evidence that the absence of one specific host factor, NQO1, that regulates reduction of cellular antioxidants causes a shift in ozonestimulated isoP production, resulting in increased production of precursors for the cyclopentenone isoP A 2 -isoP (Fig. 7). Importantly, published reports demonstrate that there is a balance of F 2 -isoPs to E 2 /D 2 -isoPs and A 2 /J 2 -isoPs formed that is

. Time course of ozone-induced NF-B activation in NHBE cells in the presence and absence of A 2 -isoprostane.
NHBE cells cultured at ALI were exposed to FA or OZ (0.4 ppm; 30, 60, and 120 min), and then total cell lysates were collected. Cell lysate proteins (70 g) were evaluated by Western blot analysis for phosphorylated p65 (p-p65), total p65, and actin as a control for equal loading and transfer (A). The -fold change between the time points was determined by densitometry, and results are presented as a percentage of the corresponding FA control (B). NHBE cells cultured at ALI were exposed to FA or OZ (0.4 ppm; 30 (C and D) or 60 min (E and F)) in the presence or absence of A 2 -isoprostanes (A2) (50 nM or 2.5 M) for 1 h prior to and then during the exposure. At the end of exposure, total cell lysates (60 -100 g) were evaluated by Western blot analysis for phosphorylated p65 (p-p65), total p65, and actin. Results shown are representative of four to six independent experiments. The -fold change between the time points was determined by densitometry, and results are presented as a percentage of the corresponding FA control. Results are summarized graphically (mean Ϯ S.E. (error bars); n ϭ 4 -6). *, significantly different from filter air control, p Ͻ 0.05; #, ozone ϩ A 2 -isoprostanes significantly different from ozone alone, p Ͻ 0.05. FEBRUARY 15, 2013 • VOLUME 288 • NUMBER 7 JOURNAL OF BIOLOGICAL CHEMISTRY 4687 dependent on the relative reducing environment in the cell (32). The presence of NQO1 regenerates dihydroquinones such as ␣-tocopherol that increase the intracellular reducing capacity. Following an oxidative stress in the presence of sufficient levels of antioxidants, ␣-tocopherol, and reduced glutathione, there is a shift toward formation of F 2 -isoPs. In contrast, when cellular antioxidants are depleted following an oxidative stress, there is a shift toward formation of E 2 /D 2 -isoPs and A 2 /J 2 -isoPs. Sup-plementation with reducing agents such as dithiothreitol, cysteine, ␤-mercaptoethanol, and glutathione increases F 2 -isoP production, supporting the importance of cellular redox status to determine the types of isoprostanes generated (24).

NQO1 Regulates Host Susceptibility to Ozone via A 2 -isoP
Previously, we demonstrated that in NQO1-null mouse lung ozone exposure does not increase F 2 -isoP production (22). In this report, we show that ozone exposure of NQO1-null mice generated increased production of E 2 -/D 2 -isoPs, the precursors FIGURE 5. Time course of ozone-induced IKK activity in NHBE cells in the presence and absence of A 2 -isoprostane. NHBE cells cultured at ALI were exposed to FA or OZ (0.4 ppm; 30, 60, or 120 min), and then total cell lysates (500 g of protein) were immunoprecipitated with specific antibodies to IKK enzymes. The resultant immunoprecipitates were assayed for IKK enzyme activity using GST-IB␣ as a substrate (A). Phosphorylation of IB␣ detected by Western analysis was used as a measure of IKK activity (upper panel). The amount of immunoprecipitated IKK enzyme was determined by reprobing the membranes with total anti-IKK␤ antibody (lower panel). Western results are representative of at least two independent time course experiments. NHBE cells cultured at ALI were exposed to filtered air (control) or OZ (0.4 ppm; 30 (B and C) or 60 min (D and E)) in the presence or absence of A 2 -isoprostanes (A2) (50 nM or 2.5 M) for 1 h prior to and then during the exposure. At the end of exposure, total cell lysates (500 g of protein) were immunoprecipitated with specific antibodies to IKK enzyme, and the resultant immunoprecipitates were assayed for IKK enzymatic activity using GST-IB␣ as a substrate. Phosphorylation of IB␣ detected by Western blot analysis was used as a measure of IKK activity. The amount of immunoprecipitated IKK was determined by reprobing the membranes with total anti-IKK␤ antibody. The -fold change between the time points was determined by densitometry, and results are presented as a percentage of the corresponding FA control. Results are summarized graphically from five separate experiments (mean Ϯ S.E. (error bars); n ϭ 5-8). *, significantly different from control, p Ͻ 0.05; #, ozone ϩ A 2 -isoprostanes significantly different from ozone alone, p Ͻ 0.05. Incubation with non-immune polyclonal rabbit IgG failed to immunoprecipitate IKK (A, lane 1, IgG con). CON, control.
of A 2 -/J 2 -isoPs. This mechanism may be generalized to other diseases. For example, patients with Alzheimer disease have increased levels of D 2 -and E 2 -isoprostane compared with F 2 -isoprostane levels (23). D 2 /E 2 -isoPs but not F 2 -isoPs are also increased in settings of traumatic brain injury in both animal models and in human cerebrospinal fluid (33,34). Our report is one of the first to provide evidence for a molecular mechanism explaining an epidemiologic link between gene polymorphisms and host response to ozone. We speculate that other gene ϫ environment interactions responsible for increased susceptibility to ozone may be due to a shift in cellular redox status affecting the generation of F 2versus A 2 -isoPs.
NF-B Is a Redox-sensitive Transcription Factor-Classically, NF-B is a redox-sensitive transcription factor (35), and there is ample evidence that reactive oxygen species including H 2 O 2 and ionizing radiation activate NF-B (36). Ozone exposure (7) and specifically a lipid ozonation product, hydroxyhydroperoxynonane, activate both NF-B and NF-IL6 to up-regulate IL-8 expression in airway epithelial cells (10). NF-B activation by LPS and cytokines requires reactive oxygen species, and TNF␣ and IL-1␤ activate NADPH oxidases that contribute to reactive oxygen species generation and are required for NF-B activation (35). Ozone regulates TNF␣ expression in mice, and this regulation is blocked in NF-B p50-null mice (9).
However, depending on the cell type, the intracellular redox status, and the kinetics of exposure, reactive oxygen species may modify the sulfhydryl group in susceptible cysteine residues of key regulatory enzymes in the NF-B activation pathway and inactivate NF-B (35). The Cys 179 residue in IKK is a critical target for reactive oxygen species that negatively regulate NF-B activity. Acrolein, an ␣,␤-unsaturated aldehyde combustion product, blocks NF-B activation and IL-8 expression in airway epithelial cells (37). Arsenite binds to and modi-  . Hypothetical mechanism of ozone-induced and NQO1-mediated production of isoprostanes. The presence or absence of NQO1 regulates the redox status of epithelial cells. In the presence of an oxidative stress such as ozone, isoprostanes are generated. The isoprostane generated is dependent on the redox status of the cells with F 2 -isoprostane preferred under reducing conditions, whereas A 2 -isoprostane is preferred under oxidizing conditions. In the presence of ozone, A 2 -isoprostane inhibits NF-B activation, resulting in blunted IL-8 up-regulation and decreased neutrophilic inflammation. ROS, reactive oxygen species; KC, keratinocyte chemoattractant; PMN, polymorphonuclear neutrophil. fies the Cys 179 in the activation loop of IKK␤, resulting in IKK inhibition and failure to activate NF-B (38). Several other studies have also shown similar modification of Cys 179 in the IKK␤ activation loop induced by triterpenoid 2-cyano-3,12-dioxooleana-1,9,-dien-28-oic acid (CDDO) and the C-28 methyl ester (CDDO-Me) (39), nimbolide (40), piceatannol (41), and xanthohumol (42). Our report and others demonstrate that cyclopentenone PGA 1 and A 2 -isoP both inhibit inflammation by direct inhibition of IKK activity (25,43) via oxidation of Cys 179 (43). The oxidant modifications of reactive cysteine residues may lead to either a reversible modification such as a sulfenic group, which may react with another cysteine to form a disulfide bond or may be reduced to return to a sulfhydryl group and permit enzyme reactivation. Sulfenic groups may also react with glutathione to form glutathionylation modifications that are also reversible. Importantly, our results demonstrate for the first time that in the presence of ozone A 2 -isoP covalently binds to the sulfhydryl group of IKK Cys 179 .
IsoPs Have Biological Activity-In this report, we demonstrate that A 2 -isoP inhibits ozone-induced inflammation. Furthermore, the increased presence of cyclopentenone precursors in NQO1-null mouse lungs complements our previous report that NQO1-null mice have decreased ozone-induced F 2 -isoP production, IL-8 mRNA or keratinocyte chemoattractant (KC) protein expression, neutrophilic inflammation in the lung, and airway hyperresponsiveness (22). We also demonstrate for the first time that A 2 -isoP has biological activity in normal human airway epithelial cells. To date there have been limited data demonstrating direct biological activity of isoprostanes in murine models or human tissues.
Although F 2 -isoP in biological fluids is widely used as a quantitative index of in vivo oxidative stress, it has also been reported to have biological activity in the lung (44,45). 15-F 2 -isoP is a ligand for thromboxane receptors, prostaglandin F receptors, and other yet uncharacterized receptors, resulting in airway smooth muscle contractility (46). It is not known whether F 2 -isoP stimulates primary human airway epithelial cell signaling or whether it can directly activate NF-B and its downstream cascade.
A 2 -isoP is difficult to isolate and quantify in biological samples and fluids due to its reactivity with sulfhydryl groups, and this property has hampered investigations to determine its biological activities. In this report, we demonstrate that A 2 -isoP at physiologic concentrations (nM) (47) and in concert with OZ blocked NF-B activation in the lung. A 2 -isoP modified a sulfhydryl moiety in Cys 179 within the active loop of IKK, resulting in inhibition of IKK activity. Inhibition of IKK results in failure to phosphorylate and degrade IB␣, impeding ozone activation of NF-B. These data are consistent with previous reports that A 2 -isoP inhibits NF-B activation in LPS-activated macrophages and in human gestational tissues (25,26). Together, published reports and our results suggest the novel concept that A 2 -isoP may be generated under conditions of severe oxidative stress and may function as a "brake" to inhibit sustained inflammation. We propose that this molecular mechanism explains the epidemiologic observation that loss of NQO1 protects the host against ozone-induced lung inflammation.