Late Intervention with a Myeloperoxidase Inhibitor Stops Progression of Experimental Chronic Obstructive Pulmonary Disease

Transcription

1 Late Intervention with a Myeloperoxidase Inhibitor Stops Progression of Experimental Chronic Obstructive Pulmonary Disease Andrew Churg 1, Caroline V. Marshall 2, Don D. Sin 3, Sarah Bolton 2, Steven Zhou 1, Katherine Thain 3, Elaine B. Cadogan 2, Justine Maltby 2, Matthew G. Soars 2, Philip R. Mallinder 2, and Joanne L. Wright 1 1 Department of Pathology and UBC James Hogg Research Centre, Institute for Heart and Lung Health, and 3 Division of Respiratory Medicine, Department of Medicine, University of British Columbia, Vancouver, British Columbia, Canada; and 2 AstraZeneca Research and Development, Charnwood, Loughborough, United Kingdom Rationale: Inflammation and oxidative stress are linked to the deleterious effects of cigarette smoke in producing chronic obstructive pulmonary disease (COPD). Myeloperoxidase (MPO), a neutrophil and macrophage product, is important in bacterial killing, but also drives inflammatory reactions and tissue oxidation. Objectives: To determine the role of MPO in COPD. Methods: We treated guinea pigs with a 2-thioxanthine MPO inhibitor, AZ1, in a 6-month cigarette smoke exposure model, with one group receiving compound from Smoking Day 1 and another group treated after 3 months of smoke exposure. Results: At 6 months both treatments abolished smoke-induced increases in lavage inflammatory cells, largely ameliorated physiological changes, andpreventedorstoppedprogression ofmorphologicemphysema and small airway remodeling. Cigarette smoke caused a marked increase in immunohistochemical staining for the myeloperoxidasegenerated protein oxidation marker dityrosine, and this effect was considerably decreased with both treatment arms. Serum 8-isoprostane, another marker of oxidative stress, showed similar trends. Both treatments also prevented muscularization of the small intrapulmonary arteries, but only partially ameliorated smoke-induced pulmonary hypertension. Acutely, AZ1 prevented smoke-induced increases in expression of cytokine mediators and nuclear factor-kb binding. Conclusions: WeconcludethatanMPOinhibitorisabletostopprogression of emphysema and small airway remodeling and to partially protect against pulmonary hypertension, even when treatment starts relatively late in the course of long-term smoke exposure, suggesting thatinhibition ofmpo maybeanovelandusefultherapeutictreatment for COPD. Protectionappears to relate to inhibitionof oxidative damage and down-regulation of the smoke-induced inflammatory response. Keywords: cigarette smoke; chronic obstructive pulmonary disease; emphysema; small airway remodeling; myeloperoxidase (Received in original form March 14, 2011; accepted in final form October 2, 2011) Supported by grant from the Canadian Institutes of Health Research and a collaborative research agreement with AstraZeneca R&D. D.D.S. holds a Canada Research Chair in COPD and is a Senior Scholar with the Michael Smith Foundation for Health Research. Author Contributions: Conception, experimental design, and writing of manuscript: A.C., C.V.M., P.R.M., J.L.W., and D.D.S.; data acquisition and analysis: A.C., J.L.W., E.B.C., J.M., M.G.S., S.Z., and K.T.; immunohistochemistry: S.B.; statistical analysis: D.D.S. Correspondence and requests for reprints should be addressed to Andrew Churg, M.D., Department of Pathology, University of British Columbia, 2211 Wesbrook Mall, Vancouver, BC, V6T 2B5 Canada. achurg@interchange.ubc.ca This article has an online supplement, which is available from this issue s table of contents at Am J Respir Crit Care Med Vol 185, Iss. 1, pp 34 43, Jan 1, 2012 Copyright ª 2012 by the American Thoracic Society Originally Published in Press as DOI: /rccm OC on October 20, 2011 Internet address: AT A GLANCE COMMENTARY Scientific Knowledge on the Subject Cigarette smoke causes a neutrophil and macrophage influx in the lung. Myeloperoxidase is a neutrophil and macrophage product known to produce oxidative damage and cause inflammation in the lungs. What This Study Adds to the Field A myeloperoxidase inhibitor stopped progression of emphysema and small airway remodeling in an animal model, even when started late in the course of a 6-month smoke exposure. Conceivably, myeloperoxidase inhibition may be a means of treating COPD. Chronic obstructive pulmonary disease (COPD) is now the fourth leading cause of death, accounting for some 100,000 deaths per year in North America (1, 2). COPD encompasses four anatomic/ clinical entities: emphysema, small airway remodeling, cigarette smoke induced pulmonary hypertension, and chronic bronchitis. It is becoming increasingly clear that these lesions are largely independent effects of smoke exposure, and that any one, or any combination, can be seen in a given patient. The pathogenesis of COPD is uncertain and is an area of considerable dispute, but in some part relates both to ongoing smoke-evoked inflammation and also to oxidative damage in the lower respiratory tract. In experimental models antiinflammatory and antioxidant treatments have been shown to completely or partially prevent emphysema (reviewed in Reference 3). Whether inflammation and oxidative damage play a role in small airway remodeling and pulmonary hypertension is unclear. Myeloperoxidase (MPO) is a neutrophil and macrophage product with a major function of bacterial killing via the generation of the powerful oxidant, hypochlorous acid. Because of its ability to generate oxidants, MPO can also produce many kinds of oxidative tissue damage. In addition, MPO has a number of other biological effects, particularly the induction and prolongation of inflammatory reactions (Reference 4; and see DISCUS- SION). Neutrophils from cigarette smokers contain higher levels of MPO than those from nonsmokers (5). Cigarette smoke consistently evokes a neutrophil and macrophage influx into the lower respiratory tract, and MPO levels are elevated in the sputum of patients with COPD (6), along with 3-chlorotyrosine, a specific product of MPO activity (7). These observations suggest that MPO inhibition might be a potential target for therapeutic interventions (7).

2 Churg, Marshall, Sin, et al.: MPO Inhibition and COPD 35 Figure 1. (A) Lavage neutrophil (PMN) and (B) macrophage counts. Both treatments protect against smoke-induced increases in lavage inflammatory cells. Columns and error bars represent means and SD. In this study we have used both a prophylactic model (Day 1 of smoke) and a therapeutic intervention model (after 3 mo of smoke) to examine the effects of an MPO inhibitor, AZ1, on the development of smoke-induced COPD in the guinea pig. METHODS MPO Inhibitor AZ1 (3-[[(2S)-tetrahydrofuran-2-yl]methyl]-2-thioxo-7H-purin-6-one) is a potent mechanism-related 2-thioxanthine irreversible inhibitor of MPO developed by AstraZeneca. Further details about AZ1 are provided in the online supplement, and a description of the mechanisms of 2-thioxanthine inhibition as well as their effects in other disease models is available in Reference 8. Animals Chronic study. Four groups of seven female Hartley guinea pigs were used: control; smoke only; smoke with AZ1, starting from Day 1 of smoke exposure (prophylactic group); and smoke with AZ1 starting after 3 months of smoke exposure (therapeutic group). AZ1 was given in an oral dose of 10 mg/kg twice per day, with the first dose 1 hour before smoke exposure, and the second dose 6 hours after smoke exposure. Animals were exposed to smoke for 5 days/week for 6 months and killed 24 hours after the last cigarette. Details are provided in the online supplement. Pulmonary function and vascular pressure measurements. Pulmonary arterial pressures as well as pulmonary function tests were measured at the 6-month killing time. Details are provided in the online supplement. Tissue and blood sampling. After physiological measurements, blood and serum were collected, the lungs were removed, and the left lower lobe was inflated with 0.6 g% agarose at 25 cm H 2 O pressure according to the method of Halbower and colleagues (9). The right lower lobe was lavaged with 2-ml aliquots of saline for a total of 10 ml of lavage fluid; the lavage fluid was used for cell counts. Details are provided in the online supplement. Morphometric Measurements Details are provided in the online supplement. Airspace size. Histologic sections were cut at a thickness of 5 mm and stained with hematoxylin and eosin. Using Image-Pro (Media Cybernetics, Silver Spring, MD), the mean linear intercept (Lm) and surface-to-volume ratio were determined. Small airway morphometry. Slides were stained with Picrosirius red and all rounded noncartilaginous airways were photographed. Using Image-Pro, we measured internal and external bronchiolar diameters. Wall area and collagen content were calculated and the data were normalized to the perimeter of the basement membrane. The mean value for all airways in each animal was determined and used for statistical analysis. Vascular Muscularization Vascular muscularization was determined by smooth muscle actin immunostaining and visual scoring. Immunohistochemical Examination of Oxidant Damage Immunohistochemical staining for dityrosine was performed with antidityrosine antibody (clone 1C3, 5 mg/ml; Cosmo Bio Co Ltd, Tokyo, Japan). Details are provided in the online supplement. Measurement of 8-Isoprostanes 8-Isoprostane levels were determined in terminal plasma with an 8-isoprostane enzyme immunoassay (#516351; Cayman Chemical, Cambridge, UK). Details are provided in the online supplement. Acute Study for Nuclear Factor-kB Activation, Proinflammatory Cytokine Gene Expression, and Myeloperoxidase Activity Three groups of five animals were created: control, smoke only, and smoke with AZ1 given once orally at a dose of 10 mg/kg 1 hour before smoke. Animals were killed immediately after smoke exposure. Details are provided in the online supplement. Statistics Details are described in the online supplement. RESULTS Lavage Inflammatory Cells Figure 1 shows levels of bronchoalveolar lavage neutrophils and macrophages at the 6-month killing point. Both prophylactic

3 36 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Figure 2. Immunohistochemical staining for the oxidative stress marker dityrosine. Smoke exposure greatly increases dityrosine staining, and there is an approximately 60% decrease in staining levels with both treatments. (A) Isotype control serum; (B) control; (C) smoke; (D) prophylactic treatment from Day 1; (E) therapeutic treatment starting at 3 months of smoke exposure. Scale bars: 100 mm. Bottom right: Comparisons above columns are between the smoke group and other groups; comparisons below the columns are between the control group and other groups. Data represent means and SD; values represent staining area per length of alveolar wall. and therapeutic treatments completely abolished smoke-mediated increases in neutrophil counts and reduced smoked-mediated increases in macrophages by about 80%. Neither smoke nor drug had any effect on lavage eosinophil numbers (see Figure E2 in the online supplement). treatment with AZ1 on the concentrations of 8-isoprostane in plasma. Smoke exposure caused a statistically nonsignificant increase in 8-isoprostane levels compared with the air-exposed Immunohistochemical Staining for the Oxidative Stress Marker Dityrosine Dityrosine can be formed by the direct action of MPO (10; and see DISCUSSION). Figure 2 shows staining for dityrosine. Cigarette smoke produced a dramatic increase in staining (P, compared with control), largely in alveolar type 2 cells and alveolar macrophages, with weaker staining of alveolar walls, and this was reduced by about 60% with both treatments (each P, compared with smoke). Serum Isoprostane Levels 8-Isoprostane has been previously proposed as a marker of antioxidant deficiency and oxidative stress, and elevated levels have been reported in cigarette smokers (11). Figure 3 shows the effects of 6 months of cigarette smoke exposure and Figure 3. Box plots showing serum 8-isoprostane levels. There is a trend toward increased levels in the smoke-exposed animals and decreased levels with both treatment groups, but only the difference between the smoke and 3-month treatment groups is statistically significant.

4 Churg, Marshall, Sin, et al.: MPO Inhibition and COPD 37 Figure 4. Left: Flow volume curves. Right: Pressure volume curves. Smoke significantly lowers flows at all volumes. This effect is completely prevented by the 6-month prophylactic treatment and partially prevented by the 3-month therapeutic treatment. Smoke shifts the pressure volume curve upward and to the left, and this effect is completely prevented by the 6-month prophylactic treatment and partially prevented by the 3-month therapeutic treatment. Left: Smoke versus control, P ¼ 0.01; smoke versus 6-month prophylactic treatment, P ¼ 0.03; smoke versus 3-month therapeutic treatment, P ¼ Right: Smoke versus control, P ¼ 0.01; smoke versus 6-month prophylactic treatment, P ¼ 0.003; smoke versus 3-month therapeutic treatment, P ¼ control group. Treatment with AZ1 caused a statistically significant decrease (P ¼ 0.025) in 8-isoprostane concentration in the therapeutically treated animals compared with the 6-month smoke-exposed animals and there was also a reduction in the prophylactically treated animals compared with the 6-month smoke group, although this was not significant. Pulmonary Function Figure 4 shows flow volume (F V) and pressure volume (P V) curves. Smoke exposure shifted the flow volume curve downward, with a slope that was significantly reduced compared with that of the control group (P ¼ 0.038). Guinea pigs that were treated with the MPO inhibitor for 3 months (therapeutic group) demonstrated flow volume curves that were similar to those of the control group (P ¼ 0.920) but significantly enhanced compared with those of the smoke-exposed group (P ¼ 0.048). Guinea pigs that were treated with the MPO inhibitor for 6 months (prophylactic group) demonstrated flow volume curves that were similar to those of the control group (P ¼ 0.846) and a trend toward enhanced curves compared with those of the smoke-exposed group (P ¼ 0.059). These data were consistent with the static lung volumes (see Table 1). In the smoke group the pressure volume curve was shifted upward compared with the control group (P ¼ 0.01), whereas the prophylactic group curve was not significantly different from that of the control group and was shifted down compared with the smokers (P ¼ 0.003). However, in the therapeutic group the P V curve showed some upward shift at high lung volumes, but was not significantly different from the smoke group. Table 1 summarizes other functional measurements. Smoking induced a significant increase in residual volume (RV) and in the ratio of RV to total lung capacity (TLC). Treatment with the myeloperoxidase inhibitor for 3 or 6 months significantly reduced the RV/TLC ratio, suggesting less gas trapping in these animals. Resistance was more than doubled in the smoke group, but with the prophylactic strategy resistance nearly normalized. Resistance was also decreased by about 60% in the therapeutic treatment group, but this difference was not significantly different from the smoke group. Pulmonary Artery Pressures Figure 5 shows pulmonary artery pressures. Smoke produced an approximately 50% increase in systolic pulmonary artery pressure compared with control and this increase was reduced by about 50% from the smoke values in both treatment groups. Diastolic pressures were also significantly increased in the smoke group but were not significantly reduced with either treatment arm. TABLE 1. COMPARISON OF LUNG FUNCTION PARAMETERS Control Smoke Prophylactic Intervention PEF, ml/s FVC, ml RV, ml * RV to TLC, % TLC, ml Airway resistance, cm H 2 O/ml/s FEV 25 75,ml * * * FEV 0.2, ml Definition of abbreviations: FEV ¼ mean forced expiratory flow during the middle half of the FVC; RV ¼ residual volume. Boldface entries indicate significant differences. * P, 0.05, compared with the smoke group. y P, 0.01, compared with the smoke group. z P, , compared with the smoke group.

5 38 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL by about 70%, but there was no significant increase over control values in both treatment groups. Figure 5. Pulmonary artery pressures (left, systolic; right, diastolic). Smoke significantly increases both pressures. Both treatment protocols partially lower systolic pressures from the smoke level but do not significantly lower diastolic pressures. Columns and error bars represent means and SD. Con ¼ control; d ¼ diastolic; NS ¼ not significant; s ¼ systolic; Smk ¼ smoke. Measures of Emphysema and Small Airway Remodeling Smoke produced an approximately 28% increase in mean airspace size (Lm), a measure of emphysema (Figures 6 and 7), whereas no increase in Lm was seen with both treatments. Similar effects were seen for surface-to-volume ratio (Figure 7). Analysis of bronchiolar wall area per unit perimeter (Figures 8 and 9) showed a 50% increase in the smoke-exposed animals and again no increase with both treatments; similar results were seen when small airway remodeling was measured as collagen per unit perimeter, or wall thickness (Figure 9). Measures of Vascular Remodeling Figure 10 shows small arterial remodeling as measured by smooth muscle actin staining. Smoke increased muscularization Mechanisms behind Suppression of Smoke-induced Inflammation To examine the mechanisms behind suppression of smokeinduced inflammation, animals were exposed to air, smoke, or smoke with a single dose of AZ1 given 1 hour before smoke exposure, and then killed immediately after smoke exposure. Nuclear factor (NF)-kB was evaluated by electrophoretic mobility shift assay and showed that smoke exposure caused increased NF-kB binding, which was partially prevented by treatment with AZ1 (Figure 11). Whole lung gene expression analysis for four acute proinflammatory mediators, tumor necrosis factor (TNF)-a, IL-1b, IL-8, and CCL2 (chemokine [C-C motif] ligand-2; also known as monocyte chemotactic protein- 1), showed that all were elevated by smoke exposure and that these elevations were completely prevented by AZ1 (Figure 12). Smoke also acutely increased whole lung myeloperoxidase activity, and this increase was prevented by treatment with AZ1 (Figure 13). DISCUSSION In this article we have shown that a novel MPO inhibitor, AZ1, completely prevents or stops progression (depending on when treatment is started) of smoke-induced morphologic increases in mean airspace size, a measure of emphysema; measures of small airway remodeling; and small vessel muscularization, a parameter that we have previously shown correlates with pulmonary hypertension. The inhibitor largely prevents or stops progression of smoke-induced physiological changes as well, although it only partially reverses pulmonary hypertension. This conclusion is true whether the animals are given the MPO inhibitor from Day 1 as a prophylactic treatment or as a therapeutic treatment started after 3 months of smoke inhalation. We have previously shown that there is minimal emphysema at 3 months in our chronic guinea pig model (12). We did not include a 3-month killing point in this study because subsequent Figure 6. Representative images of lung parenchyma. Smoke increases the mean airspace size (i.e., produces emphysema) and treatment prevents or stops progression of this effect.

6 Churg, Marshall, Sin, et al.: MPO Inhibition and COPD 39 Figure 7. Measures of airspace size (left, meanairspacesize;right, surface-to-volume ratio). Smoke increases mean airspace size (Lm) and decreases the surface-to-volume ratio (Sv) and treatment prevents or stops progression of this effect. Columns and error bars represent means and SD. lung growth increases airspace size and thus obscures minor degrees of emphysema (13); and may have similar effects on small airway remodeling; that is, reversal of the minor morphologic abnormalities present at 3 months cannot be detected by 6 months. The same may apply to physiological measures, which also suffer from greater animal-to-animal intrinsic variation. However, we are not claiming to reverse emphysema or small airway remodeling with the therapeutic intervention in this model; rather, we believe that AZ1 probably prevents disease when given prophylactically and acts to stop progression of disease when given starting at 3 months. The actions of MPO are complex, but many of them are potentially important in cigarette smoke induced disease. Cigarette smoke consistently evokes a neutrophil and macrophage inflammatory response in the lower respiratory tract (3), and MPO is a neutrophil and macrophage product that can catalyze the generation of a variety of powerful oxidants. In the presence of hydrogen peroxide MPO is able, depending on available substrates, to produce hypochlorous acid (HOCl), singlet oxygen, hydroxyl radical, ozone, chloramines, chlorine, and reactive nitrogen species (reviewed in Reference 4). Although these oxidants are, teleologically, designed to kill infectious agents, they are nonspecific in their actions and can oxidize lung proteins, DNA bases, and lipids. One of the substances that MPO can directly oxidize is tyrosine, forming the tyrosyl radical, which in turn can oxidize proteins and lipids, and this process can be detected by measuring dityrosine (10). In the present study smoke markedly increased dityrosine as measured by immunohistochemistry, and both treatments provided substantial protection. Although dityrosine is not completely specific as a marker of MPO activity, our findings here do support a role of MPO in the generation of oxidative stress in smokers, and a protective effect of AZ1 against oxidant stress. Our data on serum 8-isoprostane showed more variability but an overall similar effect. One might ask why, given the high concentration of oxidants that exist in cigarette smoke (14, 15), addition of oxidants from the actions of MPO would make a difference. One possible reason is that the HOCl normally produced by the catalytic action of MPO is an extraordinarily powerful oxidant. An additional possible explanation is that smoke tar produces hydrogen peroxide (14, 15), and MPO released from neutrophils and macrophages could use this hydrogen peroxide as a substrate to generate a variety of oxidants; in effect the smoke would amplify the potentially deleterious effects of MPO released from inflammatory cells. Of note, unlike most free radicals, the radicals in smoke tar are very long-lived (hours) (14, 15), so that the oxidant attack mediated by MPO would last well beyond the actual process of smoking a cigarette There is considerable evidence that cigarette smoke induced COPD is mediated at least in part by oxidants (reviewed in References 16 18), although the issue of what oxidant damage actually does in terms of the specific anatomic changes of COPD is uncertain. It is clear that boosting antioxidant defenses, for example, by transgenic overexpression of human copper Figure 8. Picrosirius red staining of small airways. Smoke causes small airway remodeling visible as a thickened airway wall, and treatment prevents or stops progression of this effect.

7 40 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Figure 10. Muscularization score of small arteries adjacent to alveolar ducts (see the online supplement for a description of the measurement). Smoking increases muscularization and treatment largely prevents or stops progression of this effect. Columns and error bars represent means and SD. Figure 9. Morphometric measures of small airway remodeling. Smoke increases wall thickness (top right), collagen per unit perimeter (top left), and wall area per unit perimeter (bottom), and treatment prevents or stops progression of these effects. Columns and error bars represent means and SD. zinc superoxide dismutase (CuZnSOD) (19) or extracellular superoxide dismutase (ECSOD) or use of an ECSOD mimetic (20), leads to protection against cigarette smoke induced emphysema in mice; this is also true of treatments with the antioxidants curcumin (21) and mate tea (22). Conversely, targeted deletion of Nrf2 (23), a controller of the antioxidant responses, or Rtp801, a suppressor of mtor (mammalian target of rapamycin) signaling (24), or ECSOD (20) increases emphysema in smoke-exposed mice. Yao and colleagues (20) have suggested that this protective effect reflects decreased oxidative fragmentation of matrix proteins, and this may be true. However, in these various models increases in oxidant protection are consistently associated with decreases in smoke-induced lower respiratory tract inflammation, whereas decreases in oxidant protection cause increases in inflammation after smoke exposure, so that a major effect of oxidant attack, at least in terms of emphysema, may also be to boost inflammatory cell derived proteolytic degradation of the alveolar wall matrix (3, 25). On the other hand, oxidants can also inactivate matrix metalloproteinase (MMP)-12, and mice that lack the NAPDH oxidase component gp91 phox develop spontaneous emphysema that can be prevented by concomitant MMP-12 knockout (26); thus in some circumstances oxidants appear to be protective. There is also evidence that MPO can directly up-regulate the inflammatory response, and that these properties do not all depend on its catalytic activity or generation of oxidants (reviewed in Referenced 4). MPO can evoke the release of proinflammatory cytokines such as TNF-a from macrophages, and MPO is internalized by endothelial cells that then release IL-6, IL-8, and reactive oxygen species. Some of these properties reflect MPO-induced activation of NF-kB (4). Further, MPO directly activates neutrophils in an autocrine fashion, leading to increased degranulation as well as adhesion of neutrophils to endothelial cells and increased neutrophil infiltration into tissues. HOCl derived from MPO activity plays a role in inflammation because it can cause formation of advanced glycation end products (AGEs), which activate the receptor for advanced glycation end products (RAGE), leading to release of both profibrotic and proinflammatory mediators (27, 28), and expression of RAGE itself is increased by cigarette smoke (28). In the present study treatment with AZ1 abolished smokeinduced increases in lavage inflammatory cells, and also prevented NF-kB activation as well as up-regulation of proinflammatory cytokine gene expression, suggesting that MPO is at least one of the forces driving inflammation in this model. These results are in a broad sense similar to those found with other lung inflammatory models in which knockout of MPO reduced LPSinduced inflammation (29) and also asbestos-induced inflammation (30) in mice. It should be noted that the effects of MPO inhibition or knockout as reported in the literature are contradictory and confusing, because some studies find that removing MPO from the

8 Churg, Marshall, Sin, et al.: MPO Inhibition and COPD 41 Figure 11. Left: Nuclear factor (NF)-kB binding as demonstrated by electrophoretic mobility shift assay in an acute experiment with killing immediately after smoke exposure. Smoke increases nuclear binding, and this is partially prevented by AZ1. Right-hand lanes show incubation with excess cold probe. Right: Densitometry values. Columns and error bars represent means and SD. system improves disease and some find that it worsens disease. This topic has been reviewed in detail by van der Veen and colleagues (4), who suggest that, as a general proposition, the absence of MPO is beneficial in models in which there is acute neutrophil-driven inflammation, but deleterious in models that depend on an active T-cell immune response. However, this may be an oversimplification because in smoke exposure COPD models evidence exists for both processes, and in fact knockout of CD8 1 T cells and knockout of neutrophil elastase both protect against emphysema (31, 32). We have previously shown that smoke-induced muscularization in the small intrapulmonary arteries develops within 1 month of starting smoke exposure in the guinea pig model (33), and that this finding correlates with increases in pulmonary artery pressure. In the present experiments both treatments completely reversed this muscularization, but only partially reversed the pulmonary artery pressure changes. One of the actions of MPO is to act as a sink for nitrous oxide (NO), converting it to reactive nitrogen species that can oxidatively damage tissues (4). Adequate levels of NO are also necessary for vasorelaxation, and we have previously shown (34) that smoke exposure decreases production of NO by intrapulmonary vascular endothelial cells, a process that can lead to vasoconstriction and increased pulmonary artery pressure. The present data suggest that MPO also plays a role in this process, and, in fact, serum MPO levels have been shown to correlate with endothelial dysfunction (35). Decreased NO scavenging may be one of the reasons that AZ1 decreases smoke-induced increases in pulmonary artery pressure, although it appears to have separate effects on smooth muscle proliferation. The relevance of animal models of smoke-induced COPD has been questioned (36) because interventions such as phosphodiesterase type 4 inhibition with roflumilast (37) or knockout of TNF-a receptors (38) provide complete protection against emphysema (and airway remodeling in the TNF receptor knockout mice), whereas the effects of roflumilast or TNF-a antagonists in humans Figure 12. Gene expression of proinflammatory cytokines from the acute study with killing immediately after smoke exposure. Smoke significantly increased expression of all four cytokines, and this effect is completely prevented by treatment with AZ1. Columns and error bars represent means and SD. CCL2 ¼ chemokine (C-C motif) ligand-2 (CCL2), also known as monocyte chemotactic protein-1; TNF-a ¼ tumor necrosis factor-a.

9 42 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL Figure 13. Whole lung myeloperoxidase (MPO) activity from the acute study with killing immediately after smoke exposure. Smoke increases MPO activity, and this effect is prevented by AZ1. Columns and error bars represent means and SD. with COPD are modest to nonexistent (36, 39) and roflumilast appears to be most useful in the prevention of exacerbations. We suggest that this problem may arise in part from the way in which the animal model experiments are typically set up, namely, that treatment (exogenous or via genetic manipulation) is usually given in a prophylactic fashion from Smoking Day 1 (reviewed in Reference 40). There is increasing evidence that the processes that drive emphysema, and possibly small airway remodeling, change over time. This effect has been seen using global microarray analysis of lung gene expression in rats exposed to cigarette smoke and killed at various times (41), and also by examining the effects of various drugs on suppression of inflammation, a situation in which some compounds that are effective in acute experiments are not effective if administered at the same dose after several months of smoke exposure (42). Similar effects have been reported when drugs have been used to attempt to prevent the various lesions of COPD. For example, Nakanishi and colleagues (43) showed that clarithromycin started on Smoking Day 1 at a dose of 50 mg/kg in mice prevented emphysema, but a dose of 100 mg/kg was required to prevent emphysema if the treatment was started at 3 months. Ou and colleagues (44) claimed that treatment with simvastatin from Day 1 of smoke prevented small airway remodeling in a rat model, but in a more recent study examining the effects of simvastatin started after 3 months of smoke exposure in guinea pigs (34), we found that simvastatin progressively reduced smoke-mediated increases in pulmonary hypertension and prevented emphysema, but had no effect at all on small airway remodeling. It would have been of interest to measure levels of MMP-12, because MMP-12 appears to be the matrix metalloproteinase most consistently associated with emphysema, at least in animal models (reviewed in Reference 45). Unfortunately, the antibody that we used previously for this purpose is no longer available and we have been unable to find a commercial cross-reacting antibody. Arguably we might have included an air plus AZ1 group in this study, and it is possible that AZ1 adversely affects normal processes in the lung. However, because treatment with AZ1 in animals exposed to smoke leads to lavage inflammatory cell, physiological (at least for the prophylactic treatment), and morphometric findings that are identical to control, we do not believe that AZ1 by itself affects any of these parameters. One of the potential beneficial points of 2-thioxanthine inhibitors of MPO is that they are irreversible inhibitors (so-called suicide inhibitors) that interact with MPO in such a fashion as to oxidatively destroy it (see Reference 8 for a detailed description). Use of irreversible inhibitors is probably crucial in developing effective therapies for COPD, because short-acting or reversible inhibitors tend to rapidly cleared, leaving proteolytic enzymes free to operate (46). Translation of these findings to human COPD is not straightforward, the more so as human treatment trials frequently include many ex-smokers because they have the most severe COPD. Our data imply that the same intervention at different time points can have quite different effects. They also raise a question concerning whether there is an optimal time for a given intervention, and that, as suggested by Decramer and Cooper (47), relatively early intervention will be crucial to treating human COPD. Because treatment of humans will never occur from Smoking Day 1, we suggest that animal models of chronic smoke exposure are more likely to be relevant if treatment is started well into the smoking regimen, rather than on Day 1. If our hypotheses are correct, then the current experiments indicate that AZ1 may be useful for the treatment of COPD in humans. Author disclosures are available with the text of this article at Acknowledgment: The authors thank Anh Johansson and Carl Johan Arewång for their role in the first synthesis of AZ1 and Jeff Stonehouse for generating sufficient AZ1 to carry out this study. The authors also thank Carolyn Jones, Sue Smith, and James Awford for providing technical assistance, and Paul Rugman, Shaun Hawley, Kerstin Thim, Steve Jordan, and Tomas Drmota for useful discussions in preparing this manuscript. References 1. Jemal A, Ward E, Hao Y, Thun M. Trends in the leading causes of death in the United States, JAMA 2005;294: Mannino DM, Braman S. The epidemiology and economics of chronic obstructive pulmonary disease. Proc Am Thor Soc 2007;4: Churg A, Cosio M, Wright JL. Mechanisms of cigarette smoke induced COPD: insights from animal models. Am J Physiol Lung Cell Mol Physiol 2008;294:L612 L van der Veen BS, de Winther MP, Heeringa P. Myeloperoxidase: molecular mechanisms of action and their relevance to human health and disease. Antioxid Redox Signal 2009;11: Bridges RB, Fu MC, Rehm SR. Increased neutrophil myeloperoxidase activity associated with cigarette smoking. Eur J Respir Dis 1985;67: Keatings VM, Barnes PJ. Granulocyte activation markers in induced sputum: comparison between chronic obstructive pulmonary disease, asthma, and normal subjects. Am J Respir Crit Care Med 1997;155: O Donnell C, Newbold P, White P, Thong B, Stone H, Stockley RA. 3-Chlorotyrosine in sputum of COPD patients: relationship with airway inflammation. COPD 2010;7: Tidén A-K, Sjögren T, Svensson M, Bernlind M, Senthilmohan R, Auchère F, Norman H, Markgren P-O, Gustavsson G, Schmidt S, et al. 2-Thioxanthines are mechanism-based inactivators of myeloperoxidase that block oxidative stress during inflammation. JBiolChem(in press) 9. Halbower AC, Mason RJ, Abman SH, Tuder RM. Agarose infiltration improves morphology of cryostat sections of lung. Lab Invest 1994;71: Heinecke JW. Tyrosyl radical production by myeloperoxidase: a phagocyte pathway for lipid peroxidation and dityrosine cross-linking of proteins. Toxicology 2002;177: Morrow JD, Frei B, Longmire AW, Gaziano JM, Lynch SM, Shyr Y, Strauss WE, Oates JA, Roberts LJ II. Increase in circulating products of lipid peroxidation (F2-isoprostanes) in smokers: smoking as a cause of oxidative damage. N Engl J Med 1995;332: Wright JL, Churg A. Cigarette smoke causes physiologic and morphologic changes of emphysema in the guinea pig. Am Rev Respir Dis 1990;142: Wright JL, Sun J-P. The effect of smoking cessation on pulmonary and cardiovascular function and structure. J Appl Physiol 1994;76: Church DF, Pryor WA. Free-radical chemistry of cigarette smoke and its toxicological implications. Environ Health Perspect 1985; 64:

10 Churg, Marshall, Sin, et al.: MPO Inhibition and COPD Nakayama T, Church DF, Pryor WA. Quantitative analysis of the hydrogen peroxide formed in aqueous cigarette tar extracts. Free Radic Biol Med 1989;7: MacNee W, Rahman I. Is oxidative stress central to the pathogenesis of chronic obstructive pulmonary disease? Trends Mol Med 2001;7: Rahman I. Antioxidant therapeutic advances in COPD. Ther Adv Respir Dis 2008;2: Cantin AM. Cellular response to cigarette smoke and oxidants: adapting to survive. Proc Am Thorac Soc 2010;7: Foronjy RF, Mirochnitchenko O, Propokenko O, Lemaitre V, Jia Y, Inouye M, Okada Y, D Armiento JM. Superoxide dismutase expression attenuates cigarette smoke or elastase-generated emphysema in mice. Am J Respir Crit Care Med 2006;173: Yao H, Arunachalam G, Hwang JW, Chung S, Sundar IK, Kinnula VL, Crapo JD, Rahman I. Extracellular superoxide dismutase protects against pulmonary emphysema by attenuating oxidative fragmentation of ECM. Proc Natl Acad Sci USA 2010;107: Suzuki M, Betsuyaku T, Ito Y, Nagai K, Odajima N, Moriyama C, Nasuhara Y, Nishimura M. Curcumin attenuates elastase- and cigarette smoke induced pulmonary emphysema in mice. Am J Physiol Lung Cell Mol Physiol 2009;296:L614 L Lanzetti M, Lopes AA, Ferreira TS, de Moura RS, Resende AC, Porto LC, Valenca SS. Mate tea ameliorates emphysema in cigarette smoke exposed mice. Exp Lung Res 2011;37: Rangasamy T, Cho CY, Thimmulappa RK, Zhen L, Srisuma SS, Kensler TW, Yamamoto M, Petrache I, Tuder RM, Biswal S. Genetic ablation of Nrf2 enhances susceptibility to cigarette smoke induced emphysema in mice. J Clin Invest 2004;114: Yoshida T, Mett I, Bhunia AK, Bowman J, Perez M, Zhang L, Gandjeva A, Zhen L, Chukwueke U, Mao T, et al. Rtp801, a suppressor of mtor signaling, is an essential mediator of cigarette smoke induced pulmonary injury and emphysema. Nat Med 2010;16: Yao H, Yang SR, Kode A, Rajendrasozhan S, Caito S, Adenuga D, Henry R, Edirisinghe I, Rahman I. Redox regulation of lung inflammation: role of NADPH oxidase and NF-kB signalling. Biochem Soc Trans 2007;35: Kassim SY, Fu X, Liles WC, Shapiro SD, Parks WC, Heinecke JW. NADPH oxidase restrains the matrix metalloproteinase activity of macrophages. J Biol Chem 2005;280: Anderson MM, Requena JR, Crowley JR, Thorpe SR, Heinecke JW. The myeloperoxidase system of human phagocytes generates N e - (carboxymethyl)lysine on proteins: a mechanism for producing advanced glycation end products at sites of inflammation. J Clin Invest 1999;104: Reynolds PR, Kasteler SD, Cosio MG, Sturrock A, Huecksteadt T, Hoidal RAGE Jr. Developmental expression and positive feedback regulation by Egr-1 during cigarette smoke exposure in pulmonary epithelial cells. Am J Physiol Lung Cell Mol Physiol 2008;294:L1094 L Haegens A, Heeringa P, van Suylen RJ, Steele C, Aratani Y, O Donoghue RJ, Mutsaers SE, Mossman BT, Wouters EF, Vernooy JH. Myeloperoxidase deficiency attenuates lipopolysaccharideinduced acute lung inflammation and subsequent cytokine and chemokine production. J Immunol 2009;182: Haegens A, van der Vliet A, Butnor KJ, Heintz N, Taatjes D, Hemenway D, Vacek P, Freeman BA, Hazen SL, Brennan ML, et al. Asbestosinduced lung inflammation and epithelial cell proliferation are altered in myeloperoxidase-null mice. Cancer Res 2005;65: Maeno T, Houghton AM, Quintero PA, Grumelli S, Owen CA, Shapiro SD. CD8 1 T cells are required for inflammation and destruction in cigarette smoke induced emphysema in mice. J Immunol 2007l;178: Shapiro SD, Goldstein NM, Houghton AM, Kobayashi DK, Kelley D, Belaaouaj A. Neutrophil elastase contributes to cigarette smoke induced emphysema in mice. Am J Pathol 2003;163: Wright JL, Churg A. Effect of long-term cigarette smoke exposure on pulmonary vascular structure and function in the guinea pig. Exp Lung Res 1991;17: Wright JL, Zhou S, Preobrazhenska O, Marshall C, Sin DD, Laher I, Golbidi S, Churg A. Statin reverses smoke-induced pulmonary hypertension and prevents emphysema but not airway remodeling. Am J Respir Crit Care Med 2011;183: Vita JA, Brennan ML, Gokce N, Mann SA, Goormastic M, Shishehbor MH, Penn MS, Keaney JF Jr, Hazen SL. Serum myeloperoxidase levels independently predict endothelial dysfunction in humans. Circulation 2004;110: Barnes PJ. Unexpected failure of ani-tumor necrosis factor therapy in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2007;175: Martorana PA, Beume R, Lucattelli M, Wollin L, Lungarella G. Roflumilast fully prevents emphysema in mice chronically exposed to cigarette smoke. Am J Respir Crit Care Med 2005;172: Churg A, Wang RD, Tai H, Wang X, Xie C, Wright JL. Tumor necrosis factor-a drives 70% of cigarette smoke induced emphysema in the mouse. Am J Respir Crit Care Med 2004;170: Fabbri LM, Calverley P, Izquierdo-Alonso JL, Bundschuh DS, Brose M, Martinez FJ, Rabe KF. Roflumilast in moderate-to-severe chronic obstructive pulmonary disease treated with longacting bronchodilators: two randomised clinical trials. Lancet 2009;374: Churg A, Sin DD, Wright JL. Everything prevents emphysema: are animal models of cigarette smoke induced COPD any use? Am J Respir Cell Mol Biol (In press) 41. Stevenson CS, Docx C, Webster R, Battram C, Hynx D, Giddings J, Cooper PR, Chakravarty P, Rahman I, Marwick JA, et al. Comprehensive gene expression profiling of rat lung reveals distinct acute and chronic responses to cigarette smoke inhalation. Am J Physiol 2007; 293:L1183 L Wan WY, Morris A, Kinnear G, Pearce W, Mok J, Wyss D, Stevenson CS. Pharmacological characterisation of anti-inflammatory compounds in acute and chronic mouse models of cigarette smoke induced inflammation. Respir Res 2010;11: Nakanishi Y, Kobayashi D, Asano Y, Sakurai T, Kashimura M, Okuyama S, Yoneda Y, Shapiro SD, Takayama K. Clarithromycin prevents smoke-induced emphysema in mice. Am J Respir Crit Care Med 2009;179: Ou X-M, Wen F-Q, Uhal BD, Feng YL, Huang XY, Wang T, Wang K, Liu DS, Wang X, Chen L. Simvastatin attenuates experimental small airway remodelling in rats. Respirology 2009;14: Churg A, Zhou S, Wright JL. Matrix metalloproteases in emphysema. Eur Respir J (In press) 46. Snider GL, Lucey EC, Stone PJ. Pitfalls in antiprotease therapy of emphysema. Am J Respir Crit Care Med 1994;150:S131 S Decramer M, Cooper CB. Treatment of COPD: the sooner the better? Thorax 2010;65: