toxscitoxsciToxicological Sciences1096-09291096-6080Oxford University Press10.1093/toxsci/kfq034SYSTEMS TOXICOLOGYAcute Exposure to Ozone Exacerbates Acetaminophen-Induced Liver Injury in MiceIbrahim AiboDaher*†BirminghamNeil P.†LewandowskiRyan*MaddoxJane F.†‡RothRobert A.†‡GaneyPatricia E.†‡WagnerJames G.*†HarkemaJack R.*1*Department of Pathobiology and Diagnostic Investigation†Center for Integrative Toxicology‡Department of Pharmacology and Toxicology, Michigan State University, East Lansing, Michigan 488241To whom correspondence should be addressed at Department of Pathobiology and Diagnostic Investigation, Michigan State University, 212 Food Safety and Toxicology Bldg., East Lansing, MI 48824. Fax: (517) 353-9902. E-mail: harkemaj@msu.edu.52010122010115126728561020091812010© The Author 2010. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org2010Ozone (O3), an oxidant air pollutant in photochemical smog, principally targets epithelial cells lining the respiratory tract. However, changes in gene expression have also been reported in livers of O3-exposed mice. The principal aim of the present study was to determine if acute exposure to environmentally relevant concentrations of O3 could cause exacerbation of drug-induced liver injury in mice. Overdose with acetaminophen (APAP) is the most common cause of drug-induced liver injury in developed countries. In the present study, we examined the hepatic effects of acute O3 exposure in mice pretreated with a hepatotoxic dose of APAP. C57BL/6 male mice were fasted overnight and then given APAP (300 mg/kg ip) or saline vehicle (0 mg/kg APAP). Two hours later, mice were exposed to 0, 0.25, or 0.5 ppm O3 for 6 h and then sacrificed 9 or 32 h after APAP administration (1 or 24 h after O3 exposure, respectively). Animals euthanized at 32 h were given 5-bromo-2-deoxyuridine 2 h before sacrifice to identify hepatocytes undergoing reparative DNA synthesis. Saline-treated mice exposed to either air or O3 had no liver injury. All APAP-treated mice developed marked centrilobular hepatocellular necrosis that increased in severity with time after APAP exposure. O3 exposure increased the severity of APAP-induced liver injury as indicated by an increase in necrotic hepatic tissue and plasma alanine aminotransferase activity. O3 also caused an increase in neutrophil accumulation in livers of APAP-treated animals. APAP induced a 10-fold increase in the number of bromodeoxyuridine-labeled hepatocytes that was markedly attenuated by O3 exposure. Gene expression analysis 9 h after APAP revealed differential expression of genes involved in inflammation, oxidative stress, and cellular regeneration in mice treated with APAP and O3 compared to APAP or O3 alone, providing some indications of the mechanisms behind the APAP and O3 potentiation. These results suggest that acute exposure to near ambient concentrations of this oxidant air pollutant may exacerbate drug-induced liver injury by delaying hepatic repair.ozoneacetaminophenhepatotoxicitymiceO3 is the principal oxidant pollutant in photochemical smog. Approximately half of the U.S. population lives in areas that persistently exceed the U.S. EPA’s National Ambient Air Quality Standard for this highly reactive and irritant gas (U.S. EPA, 2008). Short- and long-term exposures to high ambient concentrations of O3 have been linked to adverse health outcomes that include increases in both morbidity and mortality from respiratory causes (Bell et al., 2004; Jerrett et al., 2009; Katsouyanni et al., 1995). Though numerous studies in laboratory animals and human subjects have documented the toxic effects of inhaled O3 on the lung, much less is known about its effects on extrapulmonary organs like the liver (U.S. EPA, 2008).Recently, using global gene expression analyses, investigators found that livers of C57BL/6 mice acutely exposed to inhaled O3 had significant downregulation of gene families related to lipid, fatty acid, and carbohydrate metabolisms that were consistent with systemic cachexic responses to exposure (Last et al., 2005). Transcription of several messenger RNAs (mRNAs)–encoding enzymes of xenobiotic metabolism was also decreased in livers of these O3-exposed mice. Since several interferon (IFN)-dependent hepatic genes were downregulated with O3 exposure, the investigators suggested that IFN may act as the signaling molecule between the lung and liver. Interestingly, mice exposed to O3 have prolonged pentobarbital sleeping time (Graham et al., 1981) also suggesting impairment of hepatic drug metabolism.To our knowledge, no studies investigating the potential hepatotoxic interactions of inhaled environmental pollutants and commonly used therapeutic drugs have been reported. In the present study, we investigated the acute effects of inhaled high ambient concentrations of O3 on drug-induced liver injury in mice caused by a widely used antipyretic/analgesic agent, acetaminophen (APAP). APAP is one of the most commonly used nonprescription drugs in the world, and although remarkably safe within therapeutic doses, it has a relatively narrow therapeutic window. Indeed, APAP overdose is a commonly reported cause of liver failure in the United States (Larson et al., 2005). Like in humans, mice receiving an overdose of APAP develop acute liver injury that is characterized pathologically by centrilobular hepatocellular degeneration and necrosis with elevated blood activity of liver enzymes, such as alanine aminotransferase (ALT) (Jemnitz et al., 2008; Tee et al., 1987).Commonly reported risk factors for APAP-induced liver injury include chronic alcohol use as well as the concurrent intake of some medicinal agents (e.g., isoniazid, phenytoin, zidovudine) (McClements et al., 1990; Shriner and Goetz, 1992). Environmental pollutants have also been recognized as risk factors in pulmonary, cardiovascular, and metabolic conditions, such as type II diabetes (Gent et al., 2003; Morris et al., 1995; O’Neill et al., 2005). More recently, it has been reported that inhalation of ambient air particulates promotes systemic and liver oxidative stress in mice (Araujo et al., 2008). Though these risk factors are well documented, the potential interactive effects of inhaled air pollutants, like O3, with APAP have not been investigated.The principal aim of the present study was to determine if acute exposure to environmentally relevant concentrations of O3 could cause exacerbation of ongoing drug-induced liver injury in mice. APAP is the most widely used hepatotoxicant for experimental studies of drug-induced liver injury. A known hepatotoxic dose of APAP in mice was used to ensure centrilobular hepatocellular injury prior to the start of O3 exposure. We report for the first time that a single near ambient exposure to O3 exacerbates APAP-induced hepatic injury in mice, resulting in more severe hepatocellular necrosis and attenuation of early hepatic repair mechanisms.MATERIALS AND METHODSLaboratory animals.Pathogen-free male C57BL/6J mice (8–10 weeks of age; the Jackson Laboratory, Bar Harbor, ME) were used in this study. Mice were housed in polycarbonate cages on heat-treated aspen hardwood bedding (Nepco-Northeastern Product Corp., Warrensburg, NY). Boxes were covered with filter bonnets, and animals were provided free access to food (Harlan Teklad Laboratory Rodents 22/5 diet, Madison, WI) and water. Mice were maintained in Michigan State University (MSU) animal housing facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care and according to the National Institutes of Health guidelines as overseen by the MSU Institutional Animal Care and Use Committee. Rooms were maintained at temperatures of 21°C–24°C and relative humidities of 45–70%, with a 12-h light/dark cycle starting at 7:30 A.M.Experimental protocol: APAP treatment and O3 exposures.Mice were randomly divided into 12 groups, each consisting of six animals. They were given an intraperitoneal (ip) injection of 0 (saline vehicle) or 300 mg/kg APAP (Sigma Chemical Co., St Louis, MO) in 20 ml/kg saline. Animals were fasted 14 h before the administration of APAP at 4:00 A.M.. The dose of 300 mg/kg APAP was chosen based on the results of a pilot study in our laboratory that found mice treated with this dose, but not 150 mg/kg, developed centrilobular hepatocellular necrosis as identified by histopathology with corresponding elevations in plasma ALT activity.Two hours after APAP administration, mice were exposed to 0 (air), 0.25, or 0.5 ppm O3 for 6 h (Fig. 1). Most of the bioactivation of APAP and formation of the toxic APAP metabolite (N-acetyl-p-benzo-quinone imine) in the livers of mice are completed within the first 2 h after administration (Jollow et al., 1973; Muldrew et al., 2002). Therefore, the selected exposure time for O3 was after the bioactivation phase of APAP-induced hepatotoxicity. Mice were killed 9 or 32 h after APAP (1 or 24 h after O3 exposure, respectively). Data analysis in animals given 0.25 ppm was limited to morphological evaluations (liver necrosis and 5-bromo-2-deoxyuridine bromodexyuridine [BrdU] immunostaining) and plasma ALT activity at the later time (32 h) (Fig. 1).FIG. 1.Experimental design. Eight- to 10-week-old C57BL/6 male mice were given 0 (saline) or 300 mg/kg APAP and then exposed to O3 (0, 0.25, or 0.5 ppm) for 6 h. Mice were sacrificed at 9 or 32 h after APAP injection (1 or 24 h after O3 exposure, respectively).Mice were housed individually and exposed to O3 in stainless steel wire cages, whole-body inhalation exposure chambers (H-1000; Lab Products, Maywood, NJ). O3 was generated with an OREC 03V1-O ozonizer (O3 Research and Equipment Corp., AZ) using compressed air as a source of oxygen. Total airflow through the exposure chambers was 250 l/min (15 chamber air changes per hour). The concentration of O3 within chambers was monitored during the exposure using Dasibi 1003 AH ambient air O3 monitors (Dasibi Environmental Corp., Glendale, CA). Two O3 sampling probes were placed in the middle of the ozone chambers, 10–15 cm above cage racks. Airborne concentrations during the inhalation exposures were 0.26 ± 0.02 or 0.53 ± 0.01 ppm (mean ± SEM) for O3 chambers and 0.02 ± 0.009 ppm for air chambers.Animal necropsies and microscopic and biochemical analyses.Two hours before sacrifice, mice euthanized at the 32-h time were given BrdU (50 mg/kg; Fisher Scientific, Fair Lawn, NJ) ip. At the time of necropsy, mice were anesthetized with an ip injection of sodium pentobarbital (50 mg/kg; Fatal Plus; Vortech Pharmaceuticals, Dearborn, MI), the abdominal cavity was opened, and blood was collected from the abdominal vena cava in heparinized tubes (BD Microtainer, Franklin Lakes, NJ). Animals were then killed by exsanguination.The liver was removed from the abdominal cavity, and the left liver lobe was fixed in 10% neutral buffered formalin (Fisher Scientific) for light microscopic examination and morphometric analyses. The caudate liver lobe from each mouse was removed and placed in RNAlater (Qiagen, Valencia, CA) at 4°C for 24 h and then stored at −20°C for gene expression analyses using real-time PCR. The remaining liver lobes were frozen and stored at −80°C for biochemical analysis of inflammatory cytokines, glutathione, and thiobarbituric acid–reactive substances (TBARS).After collection of liver samples, hemidiaphragms were punctured to allow collapse of right and left lung lobes, and the thoracic cavity was opened for the removal of the trachea and heart-lung en bloc. After the trachea was cannulated, the heart-lung block was excised, and the lungs were gently lavaged twice with 0.9 ml of sterile saline. Approximately 75–90% of the intratracheally instilled saline was recovered as bronchoalveolar lavage fluid (BALF) from the lavaged lung lobes and immediately placed on ice until further analysis.Cellular and biochemical analyses of BALF.Total cell counts in the collected BALF from each mouse were determined using a hemocytometer. Cytological slides prepared by centrifugation at 40g for 10 min using a Shandon cytospin 3 (Shandon Scientific, Sewickley, PA) were stained with Diff-Quick (Dade Behring, Newark, DE). Differential counts of neutrophils, eosinophils, macrophages, and lymphocytes were assessed on a total of 200 cells. Remaining BALF was centrifuged at 240g for 15 min to collect the supernatant fraction, which was stored at −80°C for later biochemical analysis.Flow cytometric analyses for inflammatory cytokines.BALF supernatants were assayed for inflammatory cytokines that included interleukin (IL)-1β, tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-γ), IL-6, monocyte chemotactic protein-1 (MCP-1), IL-12, keratinocyte-derived chemokine (KC), and IL-10. Plasma cytokine concentrations for KC, TNF-α, MCP-1, and IL-6 were also determined. All cytokine kits were purchased as Flex Set reagents or as a preconfigured cytometric bead array kit (BD Biosciences, San Jose, CA). Cytokines analysis was performed using an FACSCalibur flow cytometer (BD Biosciences). Briefly, 50 μl of BALF or plasma was added to the antibody-coated bead complexes and incubation buffer. Phycoerythrin-conjugated secondary antibodies were added to form sandwich complexes. After acquisition of sample data using the flow cytometer, cytokine concentrations were calculated based on standard curve data using FCAP Array software (BD Biosciences).Liver tissues designated for similar cytokine analyses were suspended in PBS at 4°C and homogenized on ice. Homogenates were then centrifuged at 17,000g for 10 min at 4°C. Fifty microliters of the resulting supernatant was collected and assayed for IL-6, MCP-1, KC, TNF-α, and IL-10 by flow cytometry as described above.Plasma ALT assay.Blood collected at the time of necropsy was used to evaluate plasma ALT activity spectrophotometrically using Infinity ALT reagents purchased from Thermo Electron Corp. (Louisville, CO).Liver tissue processing for light microscopy and immunohistochemistry.Transverse sections from the middle of the left liver lobe were embedded in paraffin, cut at a thickness of 5 μm, and stained with hematoxylin & eosin (H&E) for routine histopathological examination and morphometric analyses. Other tissue sections were histochemically stained with periodic acid–Schiff (PAS) staining and counterstained with hematoxylin to identify intracellular glycogen.Routine immunohistochemical techniques were used for hepatocellular detection of nuclear BrdU, hepatic infiltration of neutrophils, and hepatocellular expression of hypoxia-inducible factor-1 alpha (HIF-1α). Briefly, liver sections were deparaffinized in xylene and rehydrated through descending grades of ethanol and immersed in 3% hydrogen peroxide to block endogenous peroxides. Sections were incubated with normal sera to inhibit nonspecific proteins (normal horse, rabbit, or goat sera for BrdU, neutrophils, or HIF-1α immunostaining, respectively; Vector Laboratories Inc., Burlingame, CA) followed by specific dilutions of primary antibodies (1:40, monoclonal mouse anti-BrdU antibody, BD Biosciences; 1:2500, monoclonal rat anti-neutrophil antibody, AbD Serotec, Raleigh, NC; and 1:200, polyclonal rabbit anti-HIF-1α, Novus Biologicals, Littleton, CO). Tissue sections were subsequently covered with secondary biotinylated antibodies, and immunostaining was developed with the Vector RTU Elite ABC kit (BrdU and HIF-1α; Vector Laboratories Inc.) or the RTU Phosphatase-labeled Streptavidin kit (neutrophils; Kirkegaard Perry Labs, Gaithersburg, MD) and visualized with Vector Red (neutrophils; Vector Laboratories Inc.) or 3,3′-diaminobenzidine (BrdU or HIF-1α; Sigma Chemicals) chromogens. Slides were counterstained with Gill 2 hematoxylin (Thermo Fisher, Pittsburgh, PA).Morphometric analyses of liver.BrdU-stained and unstained hepatocellular nuclei were counted in 10 medium power fields (×200) for each animal, starting with a randomly selected field and evaluating every third field. The hepatocellular labeling index (LI; % of hepatocytes undergoing DNA synthesis) was determined by counting the number of BrdU-labeled cells divided by the total number of hepatocytes and multiplying by 100.Hepatic neutrophil accumulation was assessed by averaging the numbers of neutrophils in 10 medium power fields (×200) in each slide. Analyzed fields were selected in an unbiased manner with a random start and counting every third field. Neutrophils were identified by positive immunohistochemical staining with the neutrophil-specific antibody and their polymorphologic nuclear profiles.Hepatocellular degeneration/necrosis in sections from the left liver lobe was quantified using standard morphometric methods that were similar to those previously described in detail (Yee et al., 2000). Briefly, H&E-stained liver sections from the left liver lobe were visualized with an Olympus BX-40 light microscope (Olympus Corp., Lake Success, NY) coupled with a 3.3-megapixel digital color camera (Qimaging, Surrey, British Columbia, Canada). Images at a magnification of ×200 were evaluated employing a 168-point lattice grid overlaying fields of hepatic parenchyma to determine (1) the total area of liver analyzed, (2) the area of degenerative/necrotic hepatic parenchyma, and (3) the area of normal parenchyma. The area of each object of interest (e.g., lesion) was calculated using the following expression (Cruz-Orive, 1982):Distance between points was 13 μm. Accordingly, the area represented by each point was 511 μm2. Section from the liver of each mouse was systematically scanned using adjacent nonoverlapping microscopic fields. The first image field analyzed in each section was chosen randomly. Thereafter, every third field was evaluated (∼10–14 fields evaluated/section). The measured fields represented ∼65% of the total area of each liver section. Percent lesion area was estimated based on the following formula:Quantitative real-time reverse transcription-PCR for hepatic gene expression.The caudate liver lobe was isolated and placed in RNAlater (Qiagen) and kept at 4°C for at least 24 h and then transferred to −20°C until processed. Total RNA was extracted using RNeasy Mini Kit according to the manufacturer's instructions (Qiagen). Briefly, tissues were homogenized in RLT buffer containing β-mercaptoethanol with a 5-mm Rotor-Sator Homogenizer (PRO Scientific, Oxford, CT) and centrifuged at 12,000g for 3 min. Samples were then treated with Rnase-free Dnase, Rnase-free buffer, and water on the column for 30 min (Qiagen). Eluted RNA was diluted 1:5 with Rnase-free water and quantified using a GeneQuant Pro spectrophotometer (BioCrom, Cambridge, UK).Reverse transcription (RT) reaction was performed using High Capacity cDNA Reverse Transcription Kit reagents (Applied Biosystems, Foster City, CA) and a GeneAmp PCR System 9700 Thermocycler PE (Applied Biosystems). Each RT reaction was run in 5 μl of sample with 20 μl of cDNA Master Mix prepared according to the manufacturer's protocol (Applied Biosystems).Expression analyses of isolated mRNA were performed by quantitative real-time PCR using individual animal's cDNA with the ABI PRISM 7900 HT Sequence Detection System using Taqman Gene Expression Assay reagents (Applied Biosystems). The cycling parameters were 48°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s followed by 60°C for 1 min. Individual data are reported as fold change of mRNA in experimental samples compared to the saline/air control group. Real-time PCR amplifications were quantified using the comparative Ct method normalized to the mean of two endogenous controls (18S and glyceraldehyde 3-phosphate dehydrogenase). The cycle number at which each amplified product crosses the set threshold represented the Ct value. The amount of target gene normalized to the mean of the endogenous reference genes was calculated by subtracting the endogenous reference Ct from the target gene Ct (ΔCt). Relative mRNA expression was calculated by subtracting the mean ΔCt of the treated samples from the ΔCt of the control samples (saline treated and air exposed) (ΔΔCt). The absolute values of the comparative expression level (fold change) was then calculated by using the formula: Fold change = 2−ΔΔCt.Glutathione assay.To determine hepatic concentrations of oxidized glutathione (GSSG) and total glutathione (reduced plus oxidized GSH and GSSG, respectively), median lobes of the liver (preserved at −80°C) were homogenized in cold buffer (0.4M 2-(N-morpholino)ethanesulfonic acid, 0.1M phosphate, and 2mM EDTA, pH 6). Homogenates were centrifuged at 10,000g for 15 min at 4°C, and the supernatants were collected and deproteinated. For deproteination, an equal volume of tissue sample and metaphosphoric acid were mixed and centrifuged. The supernatant was carefully pipetted and mixed with triethanolamine. The total glutathione concentration was then assayed as recommended by the manufacturer (Cayman Chemical Co., Ann Arbor, MI). GSSG concentration was determined after derivatization of GSH with 2-vinylpyridine. Sample absorbance was determined at 405 nm, and the total or oxidized glutathione concentrations in liver homogenates was assessed by comparison of absorption to standard curves.TBARS assay.Lipid peroxidation in the liver was estimated using a commercially available kit according to the manufacturer's recommendations and malonaldehyde as a standard (TBARS kit; Cayman Chemical Co.). Liver tissue was homogenized on ice in RIPA Buffer and Proteases Inhibitor (Thermo Scientific, Rockford, IL). Homogenates were centrifuged at 1600g, and the supernatant was collected and used to detect malonaldehyde and TBARS adducts in acidic conditions and under high temperature (100°C). Absorbance was measured at 530 nm.Statistical analyses.Data were reported as mean ± SEM. Differences among groups were analyzed by a one- or two-way ANOVA followed by Student-Newman-Keuls post hoc test. When normality or variance equality failed, a Kruskal-Wallis ranked test was conducted. All analyses were performed using SigmaStat software (SigmaStat; Jandel Scientific, San Rafael, CA). Significance was assigned to p values less than or equal to 0.05.RESULTSNo unscheduled deaths occurred in any groups of this study, and all mice were sacrificed at the scheduled posttreatment times described above. No mortality was observed in any group of APAP, O3, or APAP/O3 mice held for 96 h prior to sacrifice (data not shown).Inflammatory Responses Reflected in BALFSaline treatment/O3 exposure (O3-alone or SAL/O3 group) did not cause changes in BALF total inflammatory cell number at any time compared to saline treatment/air exposure (controls or SAL/air group) (Figs. 2A and 2B). As compared to control mice, animals that were administered APAP and were exposed to either air (APAP/air) or O3 (APAP/0.25 ppm O3 or APAP/0.5 ppm O3) had a time-dependent statistical increase in the number of total inflammatory cells in the BALF at 9 and 32 h after APAP administration (Figs. 2A and 2B). Though not statistically significant, there was a trend for greater total inflammatory cells in the lungs of the APAP and 0.5 ppm O3-coexposed mice compared to APAP alone (Figs. 2A and 2B). No difference in total inflammatory cell number was detected between APAP-alone and APAP/0.25 ppm O3 groups at 32 h (Fig. 2B).FIG. 2.Inflammatory cell accumulation in BALF of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air), 0.25 (32 h only), or 0.5 ppm O3 for 6 h. Nine (A, C, and E) or 32 h (B, D, and F) after APAP administration, mice were sacrificed and lungs were lavaged with saline as described in detail in the text. The numbers of inflammatory cells per milliliter in the recovered BALF are graphically presented as total inflammatory cells (A and B), neutrophils (C and D), and macrophages (E and F). Bars represent group means ± SEM (n = 6). a, significantly different from saline/air group; b, significantly different from saline/0.5 ppm O3 group; and c, significantly different from saline/0.25 ppm O3 (p ≤ 0.05).Pulmonary inflammatory cell responses, as reflected in the BALF, were due to increases in alveolar macrophages and/or neutrophils (Figs. 2C–F). O3 alone did not cause changes in neutrophil or macrophage number in BALF at any time (Figs. 2C–F). At 9 and 32 h after APAP treatment, mice exposed to APAP alone or with O3 had significant increases of alveolar macrophages (Figs. 2E and 2F). This response was somewhat greater at 32 h compared to 9 h. On the other hand, the number of neutrophils in BALF was not affected by APAP at 9 h (Figs. 2C and 2D). At 32 h, only APAP/O3-coexposed mice had marked increases in neutrophil numbers in BALF (Fig. 2D), indicating a potentiating effect.BALF supernatants were analyzed by flow cytometry using cytometric bead array technology for exposure-induced changes in several inflammatory cytokines (IL-1, TNF-α, IL-10, IFN-γ, IL-6, MCP-1, IL-12, and KC). Most of these cytokines were not significantly changed at either examined time (data not shown), with the exception of IL-6 and MCP-1. APAP or O3 alone did not cause an increase in IL-6 at 9 h and only a minimal increase occurred at 32 h with the APAP treatment (Figs. 3A and 3B). APAP/O3 coexposure resulted in a significant increase of IL-6 in BALF compared to either substance alone at 9 h but not at 32 h after APAP administration (Figs. 3A and 3B). At the early time, MCP-1 was undetectable in the BALF of mice from any of the groups (Fig. 3C). O3 exposure caused a slight, but significant, elevation of MCP-1 at 32 h after APAP treatment (Fig. 3D). MCP-1 in BALF was similarly and significantly elevated in APAP alone- or APAP/O3-coexposed mice 32 h after APAP (Fig. 3D).FIG. 3.IL-6 (A and B) and MCP-1 (C and D) protein concentrations in the BALF of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine or 32 h after APAP administration, mice were sacrificed and lungs were lavaged with saline as described in detail in the text. The amount of IL-6 and MCP-1 in the recovered BALF are graphically presented. Bars represent group means ± SEM (n = 6). a, significantly different from saline/0.5 ppm O3 group; b, significantly different from APAP/air group; and c, significantly different from saline/air group (p ≤ 0.05).Inflammatory Cytokine Concentrations in PlasmaExposure-related changes in plasma cytokines were restricted to IL-6, MCP-1, and KC. Changes in plasma IL-6 reflected those in BALF with only APAP/O3 coexposure inducing significant elevation in IL-6 concentration and only at 9 h after treatment (Figs. 4A and 4B). O3 alone–exposed mice had no plasma MCP-1 change at any time compared to controls. APAP alone caused significant elevation of plasma MCP-1 concentration 9 h after treatment that was not observed in the APAP/O3 coexposure group (Fig. 4C). At 32 h, APAP alone– and APAP/O3-coexposed mice had similar increases in plasma MCP-1 compared to controls (Fig. 4D).FIG. 4.IL-6 (A and B), MCP-1 (C and D), and KC (E and F) protein concentrations in plasma of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine or 32 h after APAP administration, animals were sacrificed and blood collected and analyzed as described in detail in the text. Bars represent group means ± SEM (n = 6). a, significantly different from saline/0.5 ppm O3 group; b, significantly different from APAP/air group; and c, significantly different from saline/air group (p ≤ 0.05).O3 exposure did not change the plasma concentrations of KC, a neutrophil chemokine (Figs. 4E and 4F). KC was significantly elevated only in APAP/O3-coexposed mice at 9 h after APAP treatment (Fig. 4E). At the later time, plasma KC returned to control levels in APAP/O3-coexposed mice, but there was a significant elevation in KC concentration in the plasma of mice given APAP alone (Fig. 4F).Histopathology and Morphometric Assessment of Liver InjurySaline-treated mice exposed to either air or O3 had no hepatic histopathology at either time postexposure (Figs. 5A and 5B). All APAP-treated mice developed hepatic centrilobular necrosis at 9 (data not shown) and 32 h (Figs. 5C and 5D). This drug-induced liver lesion increased in severity with time after APAP administration. Inhalation exposures to either 0.25 or 0.5 ppm O3 markedly increased the APAP-induced centrilobular necrosis at 32 h (Fig. 5D) but not at 9 h after APAP administration (24 and 1 h after O3 exposure, respectively). At the later time, APAP- and O3-coexposed mice had expanded areas of centrilobular necrosis compared to APAP alone–treated mice. These expanded areas were rimmed by a distinctive layer of enlarged hepatocytes with highly vacuolated cytoplasm and pyknotic nuclei. This one to two cell layer of hepatocytes undergoing ballooning degeneration separated the conspicuous centrilobular areas of coagulative necrosis from the normal midzonal and periportal hepatocytes. Morphometrically, APAP-treated mice exposed to 0.25 or 0.5 ppm O3 had 1.46 or 1.62 times increase, respectively, in hepatocellular necrosis compared to APAP alone–treated mice (Fig. 5E).FIG. 5.Hepatic histopathology/morphometry: liver damage induced by APAP and/or O3 exposure 32 h after APAP. Light photomicrographs of liver sections from mice treated with saline/air (A), saline/0.5 ppm O3 (B), APAP/air (C), or APAP/0.5 ppm O3 (D). All tissue sections are stained with H&E. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (filtered air), 0.25, or 0.5 ppm O3 for 6 h. Thirty-two hours after APAP administration, mice were euthanized and liver tissues were processed for light microscopy. No histopathology is evident in the livers of control or O3-exposed mice (A and B, respectively). Centrilobular hepatocellular necrosis (solid arrow) surrounding the central vein (CV) is present in the liver of APAP/air mouse (C). Increased amount of hepatocellular necrosis (solid arrow) circumscribed with hepatocytes undergoing ballooning vacuolar degeneration (stippled arrow) is present in the liver section of the APAP/0.5 ppm O3 mouse (D). Graphic representation of the morphometric determinations of the amounts of hepatocellular injury is presented in (E). Bars represent the group means ± SEM (n = 6). a, significantly different from APAP/Air group (p ≤ 0.05).No changes in plasma ALT activity, a marker of hepatocellular injury in circulating blood, were detected in O3 alone–exposed mice as compared to control mice (Figs. 6A and 6B). As expected, plasma ALT activity was significantly elevated in APAP-treated mice at 9 or 32 h after administration (Figs. 6A and 6B). At the early time, no significant differences in ALT activity were observed between APAP alone– and APAP/O3-coexposed mice (Fig. 6A). At 32 h after APAP injection, however, ALT activity was significantly greater in APAP/O3-coexposed mice as compared to APAP/air-exposed mice (Fig. 6B). This finding was reflected in differences in the extent of the centrilobular lesions between these groups. As mentioned previously, no mortality was observed in any group up to 96 h after APAP administration. By 96 h, APAP-induced hepatic lesions were markedly attenuated (minimal severity) or completely resolved, and plasma ALT concentrations had returned to normal in both the APAP/air-exposed and the APAP/O3-coexposed mice (data not shown).FIG. 6.Plasma ALT activity in APAP and/or O3-exposed mice 9 h (A) and 32 h (B) after APAP. Nine or 32 h after APAP administration, animals were sacrificed and plasma collected and analyzed for ALT activity as described in detail in the text. Bars represent group means ± SEM (n = 6). a, significantly different from saline/air group; b, significantly different from saline/0.25 ppm O3 group; c, significantly different from saline/0.5 ppm O3; and d, significantly different from APAP/air group (p ≤ 0.05).No neutrophilic accumulation was observed in livers of O3 alone–exposed mice at either time postexposure (Figs. 7B, 7E, and 7F). In APAP-treated mice, neutrophilic accumulation was observed predominantly within the areas of hepatocellular degeneration and necrosis (Figs. 7C and 7D). APAP/air and APAP/O3 groups had similar numbers of neutrophils in the liver as determined by morphometric analyses at 9-h posttreatment (Fig. 7E). At 32 h, the numbers of neutrophils in the livers of APAP/O3 mice were not significantly greater compared to APAP/air mice (Fig. 7F).FIG. 7.Liver neutrophil infiltration in APAP- and/or O3-exposed mice after APAP. Light photomicrographs of liver sections from mice treated 32 h earlier with saline/air (A), saline/0.5 ppm O3 (B), APAP/air (C) and APAP/0.5 ppm O3 (D). Tissue sections were immunohistochemically stained for infiltrating neutrophils (red chromagen; arrows) and counterstained with hematoxylin as described in detail in the text. Morphometric determinations of the numeric cell density of neutrophils in the hepatic parenchyma are graphically presented in (E and F). Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Bars in (E and F) represent group means ± SEM (n = 6). a, significantly different from saline/air group and b, significantly different from saline/O3 group (p ≤ 0.05); CV, central vein.Hepatocellular Regeneration and HypoxiaBrdU was administered to mice euthanized 32 h after APAP to identify hepatocytes undergoing DNA synthesis (S phase of the cell cycle). O3 exposure alone had no effect on hepatocellular BrdU incorporation (Figs. 8A and 8E). APAP treatment caused a marked increase of BrdU immunopositive nuclei, which was dose dependently reduced by coexposure with O3 (Figs. 8C–E). O3 (0.25 and 0.5 ppm) completely blocked the APAP-induced increase in BrdU incorporation in the liver (Fig. 8E).FIG. 8.Hepatocellular proliferation (DNA synthesis) in APAP- and/or O3-exposed mice 32 h after APAP. Light photomicrographs of liver sections from mice treated with saline/air (A), saline/0.5 ppm O3 (B), APAP/air (C), and APAP/0.5 ppm O3 (D). Tissues were immunohistochemically stained for nuclear incorporation of BrdU (brown chromagen; arrows) in hepatocytes undergoing DNA synthesis (cells in S phase of cell cycle). Morphometric determinations of the LI (%) of BrdU-labeled hepatocytes is graphically presented in (E). Mice were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air), 0.25, or 0.5 ppm O3 for 6 h. Two hours before sacrifice, mice were injected ip with BrdU. In (E), bars represent group means ± SE (n = 6). a, significantly different from saline/air group and b, significantly different from APAP/O3 groups (p ≤ 0.05). PV, portal vein; CV, central vein.At 32 h, no change in glycogen or HIF-1α staining was seen in mice exposed to O3 alone compared to controls (Figs. 9A1–2 and 9B1–2). At the same time, necrotic areas in the livers of APAP-treated mice were surrounded by a one to two cell thick layer of glycogen-depleted hepatocytes (Fig. 9A3). Interestingly, in APAP/O3-coexposed mice, the ballooning degeneration of hepatocytes was located in this layer of glycogen depletion and appeared to be the targeted tissue for the O3-induced expansion of APAP-induced liver injury (Fig. 9A4). APAP hepatotoxicity was accompanied by an increase in hepatocellular HIF-1α, a key transcription factor that mediates cellular response to hypoxia (Pouyssegur et al., 2006; Semenza, 2003). HIF-1α accumulation in APAP-treated mice was consistently found in the cytoplasm and less frequently in the nucleus of hepatocytes located in glycogen-depleted areas (junction of centrilobular necrotic zone and healthy parenchyma) (Fig. 9B3). In the APAP/O3 group, few hepatocellular nuclei, located primarily at the periphery of the necrotic zone, had HIF-1α accumulation (Fig. 9B4).FIG. 9.Intracellular glycogen (A) and HIF-1α (B) staining 32 h after APAP. Light photomicrographs of liver sections from mice given saline/air (A1 and B1), saline/0.5 ppm O3 (A2 and B2), APAP/air (A3 and B3), and APAP/0.5 ppm O3 (A4 and B4). Mice were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. In (A), all tissue sections were histochemically stained with PAS for glycogen (purple stain) as described in detail in the text. In (B), all tissue sections were immunohistochemically stained for HIF-α (dark brown chromagen) as described in detail in the text. nH, normal hepatocytes; HH, hypertrophic hepatocytes circumscribing areas of hepatocellular necrosis (asterisk); black arrows in (A) indicate hepatocellular vacuolar degeneration of HH; arrows in (B) indicate cytoplasmic or nuclear localization of HIF-α (solid and stippled arrows, respectively); CV, central vein. In APAP-treated mice exposed to air or O3 (A3 and A4, respectively), there is loss of PAS-stained glycogen in the areas of centrilobular necrosis as well as in hypertrophic hepatocytes (HH). In (A4), some of the HH are undergoing vacuolar degeneration. In (B), no HIF-α is present in liver sections (B1 and B2) from saline/air control mouse and saline/0.5 ppm O3 mouse, respectively. In the liver section from the APAP/air mouse (B3), HH circumscribing the areas of centrilobular necrosis (asterisk) contain cytoplasmic and/or nuclear HIF-α (solid arrow and stippled arrow, respectively). In the liver section from the APAP/O3 mouse, there is a loss of HIF-α in the HH undergoing vacuolar degeneration and necrosis.Relative Gene Expression and Protein Concentration of Inflammatory Cytokines in the LiverO3 exposure alone did not cause relative gene expression changes in chemokines KC or MIP-2 in the livers at any time postexposure (Figs. 10A–D). This correlated with the lack of neutrophil liver accumulation in these mice. APAP treatment or APAP/O3 coexposure caused significant increases in relative expression of KC and MIP-2 genes 9 h after APAP (Figs. 10A and 10C). KC protein concentration was also elevated in APAP/air and APAP/O3 groups at both 9 and 32 h after APAP (Figs. 11A and 11B). At 9 h, no differences were observed in mRNA expression of KC between APAP/air and APAP/O3 groups (Fig. 10A). At the same time, APAP/O3-coexposed mice had approximately three times the MIP-2 mRNA expression of the APAP-alone group (Fig. 10C). Relative gene expression of these chemokines declined to levels similar to those of controls at 32 h after APAP (Figs. 10B and 10D).FIG. 10.KC (A and B), MIP-2 (C and D), and MCP-1 (E and F) gene expression in livers of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine or 32 h after APAP administration, mice were sacrificed and liver samples were analyzed by quantitative real-time RT-PCR as described in detail in the text. Bars represent the group means ± SEM (n = 6) of the fold change in mRNA expression relative to that of the saline/air control group. a, significantly different from saline/air group; b, significantly different from saline/0.5 ppm O3 group; and c, significantly different from APAP/air group (p ≤ 0.05).FIG. 11.KC (A and B), MCP-1 (C and D), and IL-6 (E and F) protein concentrations in livers of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine (A, C, and E) or 32 h (B, D, and F) after APAP administration, mice were sacrificed and liver samples were analyzed as described in detail in the text for protein concentrations. Bars represent group means ± SEM (n = 6). a, significantly different from saline/air group; b, significantly different from saline/0.5 ppm O3 group; and c, significantly different from APAP/Air group (p ≤ 0.05).O3 exposure alone had minimal effects on relative mRNA expression or protein concentration of MCP-1 in the liver at any time (Figs. 10E and 10F and Figs. 11C and 11D). APAP treatment alone caused a significant increase in MCP-1 relative expression or protein concentration over control levels 9 h after its administration (Figs. 10E and 11C). In comparison, 0.5 ppm O3 exposure caused a more than fivefold reduction in APAP-induced increase in the mRNA or protein concentration of MCP-1 (Figs. 10E and 11C). At 32 h, both APAP/air and APAP/O3 groups had significant increases in MCP-1 mRNA expression and protein concentration (Figs. 10F and 11D).O3 exposure had no effect on IL-6 or plasminogen activator inhibitor-1 (PAI-1) relative expression at any time (Figs. 12A–D). APAP treatment on the other hand resulted in significantly elevated hepatic IL-6 and PAI-1 mRNA at 9 or 32 h after its administration (Figs. 12A and 12B). O3 coexposure tended to reduce the increases in IL-6 and PAI-1 mRNA caused by APAP, but this reduction reached statistical significance at the early time only (Figs. 12A and 12B). In agreement with these gene expression data, IL-6 protein was increased in the liver by APAP, but not APAP/O3, at 9 h only after APAP treatment (Figs. 11E and F).FIG. 12.IL-6 (A and B) and PAI-1 (C and D) genes expression in livers of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine (A and C) or 32 h (B and D) after APAP administration, mice were sacrificed and liver samples were analyzed by quantitative real-time RT-PCR as described in detail in the text. Bars represent the group means ± SEM (n = 6) of the fold change in mRNA expression relative to that of the saline/air control group. Data are expressed as mean ± SE (n = 6). a, significantly different from saline/air group; b, significantly different from APAP/air group; and c, significantly different from saline/O3 group (p ≤ 0.05).Regeneration-Related Gene Expression in Liver TissueO3 or APAP alone caused an increase in liver mRNA expression of the cyclin-dependent kinase inhibitor P21, at the early time postexposure (Fig. 13A). At the same time, APAP/O3 coexposure resulted in significantly greater expression of P21 mRNA expression compared to either APAP or O3 (Fig. 13A). At 9 h, suppressor of cytokine signaling 3 (SOCS3) expression was decreased by O3 exposure but increased by APAP treatment (Fig. 13C). APAP/O3 group had no statistically significant increase in expression of SOCS3 as compared to APAP-alone group (Fig. 13C). At 32 h, APAP-alone and APAP/O3 mice had similar increases in the expression of P21 and SOCS3 as compared to control mice (Figs. 13B and 13D).FIG. 13.P21 (A and B) and SOCS3 (C and D) genes expression in APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine (A and C) or 32 h (B and D) after APAP administration, mice were sacrificed and liver samples were analyzed by quantitative real-time RT-PCR as described in detail in the text. Bars represent the group means ± SEM (n = 6) of the fold change in mRNA expression relative to that of the saline/air control group. a, significantly different from saline/air group; b, significantly different from saline/0.5 ppm O3 group; and c, significantly different from APAP/air group (p ≤ 0.05).Liver Oxidative Damage (Antioxidant Genes, Glutathione, and TBARS Assays)Heme oxygenase-1 (HO-1), metallothionein-1 (MT-1), and the catalytic subunit of glutamate-cysteine ligase (GCLC) were evaluated as markers of oxidative stress. For all three of these antioxidant genes, mRNA expression was significantly elevated with APAP treatment, whereas only MT-1 was increased with O3 exposure 9 h after APAP (Figs. 14A, 14C, and 14E). At 9 h, APAP/O3 coexposure resulted in significant increase of MT-1 expression above APAP or O3 levels (Fig. 14A). At the later time (32 h), expression of these genes declined in APAP/air or APAP/O3 groups as compared to the early time, and APAP/O3-coexposed mice had less or comparable mRNA expression compared to APAP/air-treated mice (Figs. 14B, 14D, and 14F).FIG. 14.MT-1 (A and B), HO-1 (C and D), and GCLC (E and F) gene expression in livers of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine (A, C, and E) or 32 h (B, D, and F) after APAP administration, mice were sacrificed and liver samples were analyzed by quantitative real-time RT-PCR as described in detail in the text. Bars represent the group means ± SEM (n = 6) of the fold change in mRNA expression relative to that of the saline/air control group. a, significantly different from saline/air group; b, significantly different from saline/0.5 ppm O3 group; and c, significantly different from APAP/air group (p ≤ 0.05).To explore further O3 exacerbation of APAP-induced liver toxicity and the role of oxidative stress, we evaluated concentrations of GSH and GSSG. At 9 h, total glutathione concentration was greater in APAP-alone and APAP/O3 exposure groups than in control animals (Fig. 15A). At the same time, O3-exposed mice had less total glutathione concentration in the liver compared to control mice (Fig. 15A). Interestingly, O3-alone and APAP/O3 groups had more GSSG than control and APAP-alone groups, respectively (Fig. 15B).FIG. 15.Total (A) or oxidized (B) glutathione and TBARS (C and D) concentrations in livers of APAP- and/or O3-exposed mice. Animals were given 0 (saline) or 300 mg/kg APAP ip and 2 h later exposed to 0 (air) or 0.5 ppm O3 for 6 h. Nine (A, B, and C) or 32 h (D) after APAP administration, mice were sacrificed and liver samples were processed for analytical determination of glutathione (GSH and GSSG) and TBARS concentrations by standard assays described in detail in the text. Bars represent the group means ± SEM (n = 6). a, significantly different from saline/air group; b, significantly different from saline/0.5 ppm O3 group; and c, significantly different from APAP/air group (p ≤ 0.05).APAP-induced changes in the levels of GSH and GSSG were time dependent, as have been previously reported by others. It has been shown that APAP treatment in mice, at a similar dose used in our study, causes depletion of hepatic GSH within the first hour after administration (Muldrew et al., 2002) but GSH recovers to pretreatment levels by 2 h (Saito et al., 2009). Therefore, the APAP-induced elevations in hepatic GSH observed in the present study were probably a rebound effect from an early depletion of GSH.We also evaluated lipid peroxidation levels in the livers using TBARS assay. Mice coexposed to APAP and O3 had greater concentrations of TBARS at the early time postexposure relative to APAP alone–treated mice (Fig. 15C). O3 alone–exposed mice had no significant increases in the concentration of TBARS compared to controls (Fig. 15C). By 32 h, TBARS concentrations in APAP-alone and APAP/O3 groups were less than that of their respective control groups (Fig. 15D). These results also indicate that APAP has a time-dependent effect on TBARS in the livers of mice. It has previously been shown in mice that APAP treatment increases hepatic TBARS levels that peak at around 2–3 h after administration (Amimoto et al., 1995). At 6-h posttreatment, the concentration of TBARS decline from peak levels. The early time in our study (9-h post-APAP) was probably on the recovery side of this response, accordingly. As to the later time point, it is likely that the antioxidant mechanisms induced by APAP (as suggested by gene expression data above) were probably responsible for the decrease in the levels of TBARS in the livers of APAP-treated groups.DISCUSSIONTo our knowledge, this is the first study to examine the pulmonary and systemic effects of APAP/O3 coexposure in laboratory animals. Acute exposure to 0.25 or 0.5 ppm O3 alone did not cause pulmonary inflammation in the mice of our study, as evidenced by BALF analysis. In contrast, APAP treatment alone did cause acute inflammation of the lung, and this drug-induced pulmonary response was exacerbated by O3 coexposure. The most remarkable finding, however, was that a single 6-h inhalation exposure to O3 resulted in exacerbation of APAP-induced liver injury. After a hepatotoxic dose of APAP, exposure of mice to 0.5 ppm O3 resulted in a 60% increase in hepatocellular necrosis and an 80% decrease in hepatocellular regeneration as compared to mice treated with APAP alone. How a single acute exposure of O3 caused such marked enhancement of this drug-induced liver injury in mice is unknown.It is unlikely that this potentiation of APAP-induced liver injury caused by O3 exposure was due to changes in APAP metabolism since most, if not all, of the hepatic bioactivation of APAP and formation of the toxic APAP metabolite would have occurred in the first couple of hours after administration (Jollow et al., 1973; Muldrew et al., 2002). Furthermore, others have reported that hepatic levels of several isoforms of cytochrome P450s, including those involved in bioactivation of this drug, are unchanged or downregulated, rather than upregulated, after inhalation exposure to O3 (Last et al., 2005).It is also not likely that O3 caused direct injury to the liver since it is one of the most reactive chemicals known, and others have demonstrated that when inhaled it reacts quickly with airway surface lining fluid and is converted to secondary lipid ozonation products (Miller, 1995; Pryor, 1992; Pryor et al., 1995). These secondary products (e.g., aldehydes, hydroxyhydroperoxides etc) are thought to be the principal toxicants responsible for O3’s toxicity to epithelial cells lining the airway surfaces. Therefore, O3 could not have entered the systemic circulation to be transported from the lung to the liver resulting in direct hepatotoxicity.Though the present study was not designed to definitively determine how acute inhalation exposure to O3 caused exacerbation of APAP-induced liver injury, the results of our biochemical and molecular analyses suggest some plausible hypotheses that will have to be addressed in future studies.One possible hypothesis is that the O3-induced enhancement of APAP hepatotoxicity is due in part to increased oxidative stress in the liver (i.e., more injurious oxidant-free radicals than protective antioxidants). Goldstein et al. (1978) suggested that extrapulmonary effects of O3 are related to lipid oxidation products, particularly malonaldehyde released after the interaction of O3 with airway epithelial cell membrane fatty acids. Interestingly, oxidative stress has also been suggested to play a key role in the progression of APAP-induced liver injury, specifically through induction of mitochondrial permeability transition pore formation (Jaeschke et al., 2003). It is plausible that the systemic increase in oxidative stress induced by inhaled O3 may have been partially responsible for the exacerbation of both APAP-induced liver injury observed in our study.In the present study, APAP/O3 coexposure resulted in significant expression of the oxidative stress–responsive genes, MT-1, HO-1, and GCLC in the liver. We also found that O3 exposure in either APAP- or saline-treated mice caused significant increases of hepatic GSSG, another indicator of oxidative stress (i.e., oxidation of GSH). In addition, lipid peroxidation, an indicator of oxidant-induced cellular injury, was biochemically evident in the livers of O3-exposed mice.The greatest measured response in antioxidant gene expression in the liver of APAP/O3-coexposed mice was MT-1. Metallothioneins, including MT-1, are cysteine-rich proteins with various protective roles, including antioxidant properties (Kang, 2006). Both APAP and O3 have been shown to induce increases of these proteins in the liver and lung of mice, respectively (Johnston et al., 1999; Wormser and Calp, 1988). In addition, mice lacking MT-1 and MT-2 are more sensitive to APAP or O3 toxicity compared to their wild-type counterparts (Inoue et al., 2008; Liu et al., 1999). Last et al. (2005) also reported that mice exposed to O3 exhibited greater hepatic expression of MT-1 compared to air-exposed animals.The antioxidant effects of MTs have been ascribed to their abundant cysteine moieties and direct scavenging properties of phenoxyl or hydroxyl radicals and superoxide anions as demonstrated by in vitro studies (Schwarz et al., 1994). Alternatively, the effect of MTs could be due to antioxidant properties of zinc released from MTs by reactive oxygen species (Powell, 2000; Schwarz et al., 1994). Owing to these roles, it is probable that the greater induction of MT-1 in the combined APAP and O3 coexposure was in response to enhanced oxidative stress.With immunohistochemical analysis, we found that mice given APAP had conspicuous accumulation of HIF-1α in hepatocytes immediately surrounding the centrilobular areas of necrosis. These same hepatocytes had concurrent loss of cytoplasmic glycogen as demonstrated by PAS histochemistry. It was these HIF-1α–overexpressing glycogen-depleted hepatocytes at the edge of the drug-induced necrotic regions that appeared microscopically to be further altered (e.g., ballooning degeneration and necrosis) by acute coexposure to O3. It is known that APAP directly targets mitochondria, inhibiting oxidative phosphorylation and compromising ATP synthesis with subsequent induction of glycolysis (Kon et al., 2004). APAP also induces HIF-1α accumulation in a hypoxia-unrelated oxidative stress–dependent fashion (James et al., 2006). In addition, HIF-1α induction results in decreased oxidative phosphorylation and stimulation of glycolysis in the liver of mice (Denko, 2008). Thus, glycogen depletion in the perinecrotic hepatocytes in APAP-treated mice might have been mediated by HIF-1α in our study.O3 inhalation is also known to alter cardiopulmonary function by increasing breathing frequency and pulmonary resistance and decreasing tidal volume, forced vital capacity, heart rate, and mean arterial pressure (U.S. EPA, 2008). Though speculative, it is possible that changes in cardiopulmonary function during O3 exposure could have compromised oxygen delivery to the APAP-induced areas of hepatic injury leading to additional tissue hypoxia and further injury. Interestingly, other studies in mice have recently demonstrated that chronic intermittent hypoxia can cause lipid peroxidation in the liver and exacerbate APAP-induced hepatic injury (Savransky et al., 2007). It is therefore reasonable that the APAP-induced loss of glycogen and increased HIF-1α in these hepatocytes might have made these cells more susceptible to O3 toxicity. Additional studies designed to examine pulmonary and cardiovascular function as well as blood gas parameters in mice treated with APAP and exposed to O3 will be needed to adequately investigate this possible pathogenesis.In the present study, PAI-1 expression was increased in the livers of APAP-treated mice similar to that reported by others (Bajt et al., 2008; Ganey et al., 2007; Reilly et al., 2001). In contrast, APAP/O3-coexposed mice in our study had markedly less PAI-1 liver expression as compared to mice given APAP alone. PAI-1 inhibits plasminogen activators involved in the formation of plasmin and as such inhibits fibrinolysis in mice (Bajt et al., 2008). In mice deficient in PAI-1, APAP caused greater plasma ALT activity, hepatocellular necrosis, and reduced hepatocellular regeneration. Like these PAI-1–deficient animals, the APAP/O3-coexposed mice in our study had greater hepatocellular injury and impaired hepatocellular repair (i.e., reduction in BrdU LI). One explanation for the role of PAI-1 in APAP/O3 cotoxicity could be that reduced PAI-1 led to early fibrinolysis in these animals, which then resulted in an ischemia/reperfusion-like mechanism.It has been reported that P21 mRNA was increased in APAP-treated PAI-1–deficient mice (Bajt et al., 2008). In our study, APAP/O3-coexposed mice had reduced expression of PAI-1 compared to APAP alone and an increase in the expression of P21 mRNA. P21 halts the cell cycle in the G1 phase by inhibiting the activity of cyclinE/cdk2 complexes (Weinberg and Denning, 2002). Therefore, it is possible that the enhanced liver pathology in APAP/O3-coexposed mice might be due in part to an O3-induced reduction in PAI-1, which in turn caused an increase in P21 that led to impairment of hepatocellular regeneration.As mentioned above, the results of our study clearly demonstrated that O3 exposure significantly impaired reparative hepatocellular regeneration in APAP-treated mice. This suggested to us that the O3 enhancement of APAP-induced liver injury may be due in part to reduced repair mechanisms. Mehendale and collaborators have proposed that the severity of acute chemical–induced liver toxicity is strongly dependent upon tissue repair processes (Soni et al., 1999). They have shown that cotreatment with small doses of hepatotoxicants (e.g., chlordecone and carbon tetrachloride, CCl4) can cause synergistic toxicity by inhibition of tissue repair (Soni and Mehendale, 1998). The differential expression of several genes in the livers of APAP/O3-coexposed mice in our study suggests some possible mechanisms by which O3 exposure might have compromised the hepatocellular regeneration after APAP-induced injury. For example, IL-6 gene and protein expression in the livers of APAP/O3-coexposed mice was significantly reduced compared to that in mice treated with APAP alone. This correlated with the marked reduction in hepatocellular BrdU labeling (reduced DNA synthesis). IL-6 is known to be an essential protein for the initial phases of hepatocellular regeneration, transitioning cells from the G0 to the G1 phase of the cell cycle (Fausto et al., 2006; Taub, 2004).Cressman et al. (1996) have shown that after partial hepatectomy, mice deficient in IL-6 had greater hepatocellular injury and reduced reparative regeneration as compared to IL-6–sufficient hepatectomized mice. IL-6–deficient mice treated with CCl4 had greater hepatocellular damage and decreased number of hepatocytes in the S phase of the cell cycle as compared to IL-6–sufficient mice (Kovalovich et al., 2000). Pretreatment of IL-6–deficient mice with IL-6 significantly reduced CCl4-induced liver injury and restored the reparative induction of DNA synthesis. We found that O3 inhalation caused impairment of hepatocellular repair following APAP-induced injury. Though the downregulation of hepatic IL-6 expression may be responsible for impaired liver regeneration 32 h after APAP, there may be other mechanisms involved.Other potential candidates responsible for the impaired regeneration in APAP/O3-coexposed mice are P21 and MCP-1. In APAP/O3-coexposed mice, expression of P21 was increased relative to mice given APAP or O3 alone. Activation of P21 after DNA damage is known to delay or arrest the cell cycle (Garner and Raj, 2008). APAP treatment or O3 exposure causes DNA damage in the liver or lung, respectively (Bornholdt et al., 2002; Hongslo et al., 1994; Ito et al., 2005; Ray et al., 1990). In the present study, mice treated with APAP or exposed to O3 alone had greater hepatic P21 expression than saline-treated and air-exposed control mice. APAP/O3-coexposed mice had even greater P21 expression compared to mice receiving only one of these chemical agents. Though the level of DNA damage in the liver was not measured in this study, the marked increase in P21 expression in the livers of coexposed mice might have been due to increased DNA damage, which is known to lead to hepatic cell death (Corcoran and Ray, 1992).Another interesting finding in this study was the effect of APAP and O3 on the expression of the inflammatory chemokine MCP-1 in the liver. APAP treatment caused a significant increase in the expression of this chemokine, but O3 exposure caused a marked reduction in MCP-1 expression. This reduction was associated with enhanced liver toxicity and defective hepatocellular regeneration responses. MCP-1 has been shown to be involved in cell regeneration and tissue repair in various tissues after different types of induced cell injury (Kim et al., 2003; Low et al., 2001; Shireman et al., 2007). Mice lacking MCP-1 or the receptor for MCP-1 have been reported to be more (Hogaboam et al., 2000) or similarly (Dambach et al., 2002) sensitive to APAP-induced hepatotoxicity compared to their wild-type counterparts. Interestingly, mice lacking the receptor for MCP-1 also had decreased reparative DNA synthesis in a murine model of arterial injury (Kim et al., 2003). It is not known how a reduction of MCP-1 could lead to impaired cell regeneration, but the role of MCP-1 in the O3 enhancement of APAP-induced hepatotoxicity should be explored in future studies.In conclusion, we found that a single 6-h inhalation exposure of mice to high ambient concentrations of O3 caused marked enhancement of APAP-induced hepatotoxicity in mice. The present study was not designed to determine the underlying mechanism(s) responsible for this observed systemic effect caused by this common oxidant air pollutant. Several biochemical and molecular markers of oxidative stress were elevated in the livers of APAP/O3-coexposed mice compared to mice that received only APAP or O3 alone. In addition, we found that concurrent with the enhancement of hepatotoxicity, O3 also caused a marked attenuation of normal increases in DNA synthesis necessary for hepatocellular regeneration and repair in response to chemical-induced liver injury. This finding suggests a possible role for impaired cellular regeneration and enhanced toxicity in the livers of coexposed mice. Though it is unclear how inhalation of this highly reactive gas could enhance chemical-induced liver injury, several endogenous hepatic proteins such as IL-6, HIF-1α, PAI-1, P21, and MCP-1 have been identified as potentially playing important roles in this effect.These results in mice also suggest that exposures to high ambient concentrations of O3 and other air pollutants (e.g., particulate matter) may pose a risk to people with preexisting chemical-induced and other liver diseases. Further studies are needed not only to understand the biological mechanisms underlying this systemic effect of inhaled O3 but also to determine the smallest doses of O3 and APAP at which these synergistic responses in the liver occur. In addition, epidemiological studies investigating the potential interactive effects of pharmaceutical agents and air pollutants on both hepatic and pulmonary disease appear to be warranted, especially in light of recent reports that headaches are commonly reported in association with increased concentrations of air pollutants (Larrieu et al., 2009) and that APAP and other nonsteroidal, over-the-counter pain medications, are commonly taken by the general public.FUNDINGMSU Respiratory Research Initiative, National Institutes of Health (R01 ES011617 to J.R.H, R01 ES004139 to R.A.R and P.E.G).We thank Dr Yogesh Saini and Dr John LaPres for helping us with the HIF-1α immunohistochemistry and Lori Bramble for technical assistance with animal necropsies and flow cytometry analyses.AmimotoTMatsuraTKoyamaSYNakanishiTYamadaKKajiyamaGAcetaminophen-induced hepatic injury in mice: the role of lipid peroxidation and effects of pretreatment with coenzyme Q10 and alpha-tocopherolFree Radic. Biol. Med.199519169176AraujoJABarajasBKleinmanMWangXBennettBJGongKWNavabMHarkemaJSioutasCLusisAJAmbient particulate pollutants in the ultrafine range promote early atherosclerosis and systemic oxidative stressCirc. Res.2008102589596BajtMLYanHMFarhoodAJaeschkeHPlasminogen activator inhibitor-1 limits liver injury and facilitates regeneration after acetaminophen overdoseToxicol. Sci.2008104419427BellMLMcDermottAZegerSLSametJMDominiciFOzone and short-term mortality in 95 US urban communities, 1987–2000JAMA200429223722378BornholdtJDybdahlMVogelUHansenMLoftSWallinHInhalation of ozone induces DNA strand breaks and inflammation in miceMutat. Res.20025206371CorcoranGBRaySDThe role of the nucleus and other compartments in toxic cell death produced by alkylating hepatotoxicantsToxicol. Appl. Pharmacol.1992113167183CressmanDEGreenbaumLEDeAngelisRACilibertoGFurthEETaubRLiver failure and defective regeneration in interleukin-6-deficient miceScience199627413791383Cruz-OriveLMThe use of quadrants and test systems in stereology, including magnification correctionsJ. Microsc.198012589102DambachDMWatsonLMGrayKRDurhamSKLaskinDLRole of CCR2 in macrophage migration into the liver during acetaminophen-induced hepatotoxicity in the mouseHepatology20023510931103DenkoNCHypoxia, HIF1 and glucose metabolism in the solid tumourNat. Rev. Cancer20088705713FaustoNCampbellJSRiehleKJLiver regenerationHepatology200643S45S53GaneyPELuyendykJPNewportSWEagleTMMaddoxJFMackmanNRothRARole of the coagulation system in acetaminophen-induced hepatotoxicity in miceHepatology20074611771186GarnerERajKProtective mechanisms of p53-p21-pRb proteins against DNA damage-induced cell deathCell Cycle20087277282GentJFTricheEWHolfordTRBelangerKBrackenMBBeckettWSLeadererBPAssociation of low-level ozone and fine particles with respiratory symptoms in children with asthmaJAMA200329018591867GoldsteinBDThe pulmonary and extrapulmonary effects of ozoneCiba Found. Symp.197865295319GrahamJAMenzelDBMillerFJIllingJWGardnerDEInfluence of ozone on pentobarbital-induced sleeping time in mice, rats, and hamstersToxicol. Appl. Pharmacol.1981616473HogaboamCMBone-LarsonCLSteinhauserMLMatsukawaAGoslingJBoringLCharoIFSimpsonKJLukacsNWKunkelSLExaggerated hepatic injury due to acetaminophen challenge in mice lacking C-C chemokine receptor 2Am. J. Pathol.200015612451252HongsloJKSmithCVBrunborgGSoderlundEJHolmeJAGenotoxicity of paracetamol in mice and ratsMutagenesis1994993100InoueKTakanoHKaewamatawongTShimadaASuzukiJYanagisawaRTasakaSIshizakaASatohMRole of metallothionein in lung inflammation induced by ozone exposure in miceFree Radic. Biol. Med.20084517141722ItoKInoueSHirakuYKawanishiSMechanism of site-specific DNA damage induced by ozoneMutat. Res.20055856070JaeschkeHKnightTRBajtMLThe role of oxidant stress and reactive nitrogen species in acetaminophen hepatotoxicityToxicol. Lett.2003144279288JamesLPDonahowerBBurkeASMcCulloughSHinsonJAInduction of the nuclear factor HIF-1alpha in acetaminophen toxicity: evidence for oxidative stressBiochem. Biophys. Res. Commun.2006343171176JemnitzKVeresZMonostoryKKoboriLVereczkeyLInterspecies differences in acetaminophen sensitivity of human, rat, and mouse primary hepatocytesToxicol. In Vitro200822961967JerrettMBurnettRTPopeCA3rdItoKThurstonGKrewskiDShiYCalleEThunMLong-term ozone exposure and mortalityN. Engl. J. Med.200936010851095JohnstonCJStrippBRReynoldsSDAvissarNEReedCKFinkelsteinJNInflammatory and antioxidant gene expression in C57BL/6J mice after lethal and sublethal ozone exposuresExp. Lung Res.1999258197JollowDJMitchellJRPotterWZDavisDCGilletteJRBrodieBBAcetaminophen-induced hepatic necrosis. II. Role of covalent binding in vivoJ. Pharmacol. Exp. Ther.1973187195202KangYJMetallothionein redox cycle and functionExp. Biol. Med.200623114591467KatsouyanniKZmirouDSpixCSunyerJSchoutenJPPonkaAAndersonHRLe MoullecYWojtyniakBVigottiMAShort-term effects of air pollution on health: a European approach using epidemiological time-series data. The APHEA project: background, objectives, designEur. Respir. J.1995810301038KimWJChereshnevIGazdoiuMFallonJTRollinsBJTaubmanMBMCP-1 deficiency is associated with reduced intimal hyperplasia after arterial injuryBiochem. Biophys. Res. Commun.2003310936942KonKKimJSJaeschkeHLemastersJJMitochondrial permeability transition in acetaminophen-induced necrosis and apoptosis of cultured mouse hepatocytesHepatology20044011701179KovalovichKDeAngelisRALiWFurthEECilibertoGTaubRIncreased toxin-induced liver injury and fibrosis in interleukin-6-deficient miceHepatology200031149159LarrieuSLefrancAGaultGChatignouxECouvyFJouvesBFilleulLAre the short-term effects of air pollution restricted to cardiorespiratory diseases?Am. J. Epidemiol.200916912011208LarsonAMPolsonJFontanaRJDavernTJLalaniEHynanLSReischJSSchiodtFVOstapowiczGShakilAOAcetaminophen-induced acute liver failure: results of a United States multicenter, prospective studyHepatology20054213641372LastJAGohilKMathraniVCKenyonNJSystemic responses to inhaled ozone in mice: cachexia and down-regulation of liver xenobiotic metabolizing genesToxicol. Appl. Pharmacol.2005208117126LiuJLiuYHartleyDKlaassenCDShehin-JohnsonSELucasACohenSDMetallothionein-I/II knockout mice are sensitive to acetaminophen-induced hepatotoxicityJ. Pharmacol. Exp. Ther.1999289580586LowQEDrugeaIADuffnerLAQuinnDGCookDNRollinsBJKovacsEJDiPietroLAWound healing in MIP-1alpha(-/-) and MCP-1(-/-) miceAm. J. Pathol.2001159457463McClementsBMHylandMCallenderMEBlairTLManagement of paracetamol poisoning complicated by enzyme induction due to alcohol or drugsLancet19903351526MillerFJUptake and fate of ozone in the respiratory tractToxicol. Lett.199582–83277285MorrisRDNaumovaENMunasingheRLAmbient air pollution and hospitalization for congestive heart failure among elderly people in seven large US citiesAm. J. Public Health19958513611365MuldrewKLJamesLPCoopLMcCulloughSSHendricksonHPHinsonJAMayeuxPRDetermination of acetaminophen-protein adducts in mouse liver and serum and human serum after hepatotoxic doses of acetaminophen using high-performance liquid chromatography with electrochemical detectionDrug Metab. Dispos.200230446451O’NeillMSVevesAZanobettiASarnatJAGoldDREconomidesPAHortonESSchwartzJDiabetes enhances vulnerability to particulate air pollution-associated impairment in vascular reactivity and endothelial functionCirculation200511129132920PouyssegurJDayanFMazureNMHypoxia signalling in cancer and approaches to enforce tumour regressionNature2006441437443PowellSRThe antioxidant properties of zincJ. Nutr.20001301447S1454SPryorWAHow far does ozone penetrate into the pulmonary air/tissue boundary before it reacts?Free Radic. Biol. Med.1992128388PryorWASquadritoGLFriedmanMThe cascade mechanism to explain ozone toxicity: the role of lipid ozonation productsFree Radic. Biol. Med.199519935941RaySDSorgeCLRaucyJLCorcoranGBEarly loss of large genomic DNA in vivo with accumulation of Ca2+ in the nucleus during acetaminophen-induced liver injuryToxicol. Appl. Pharmacol.1990106346351ReillyTPBourdiMBradyJNPise-MasisonCARadonovichMFGeorgeJWPohlLRExpression profiling of acetaminophen liver toxicity in mice using microarray technologyBiochem. Biophys. Res. Commun.2001282321328SaitoCZwingmannCJaeschkeHNovel mechanisms of protection against acetaminophen hepatotoxicity in mice by glutathione and N-acetylcysteineHepatology200951246254SavranskyVNanayakkaraAViveroALiJBevansSSmithPLTorbensonMSPolotskyVYChronic intermittent hypoxia predisposes to liver injuryHepatology20074510071013SchwarzMALazoJSYalowichJCReynoldsIKaganVETyurinVKimYMWatkinsSCPittBRCytoplasmic metallothionein overexpression protects NIH 3T3 cells from tert-butyl hydroperoxide toxicityJ. Biol. Chem.19942691523815243SemenzaGLTargeting HIF-1 for cancer therapyNat. Rev. Cancer20033721732ShiremanPKContreras-ShannonVOchoaOKariaBPMichalekJEMcManusLMMCP-1 deficiency causes altered inflammation with impaired skeletal muscle regenerationJ. Leukoc. Biol.200781775785ShrinerKGoetzMBSevere hepatotoxicity in a patient receiving both acetaminophen and zidovudineAm. J. Med.1992939496SoniMGMehendaleHMRole of tissue repair in toxicologic interactions among hepatotoxic organics Environ. Health Perspect.1998106Suppl. 613071317SoniMGRamaiahSKMumtazMMClewellHMehendaleHMToxicant-inflicted injury and stimulated tissue repair are opposing toxicodynamic forces in predictive toxicologyRegul. Toxicol. Pharmacol.199929165174TaubRLiver regeneration: from myth to mechanismNat. Rev. Mol. Cell. Biol.20045836847TeeLBDaviesDSSeddonCEBoobisARSpecies differences in the hepatotoxicity of paracetamol are due to differences in the rate of conversion to its cytotoxic metaboliteBiochem. Pharmacol.19873610411052U.S. Environmental Protection Agency (U.S. EPA)Air Quality Criteria for O3 and Related Photochemical Oxidants (Final)2008EPA 600/R-05/004-aF-cF, Research Triangle Park, NCWeinbergWCDenningMFP21Waf1 control of epithelial cell cycle and cell fateCrit. Rev. Oral. Biol. Med.200213453464WormserUCalpDIncreased levels of hepatic metallothionein in rat and mouse after injection of acetaminophenToxicology198853323329YeeSBKinserSHillDABartonCCHotchkissJAHarkemaJRGaneyPERothRASynergistic hepatotoxicity from coexposure to bacterial endotoxin and the pyrrolizidine alkaloid monocrotalineToxicol. Appl. Pharmacol.2000166173185