© 2003 American Thoracic Society
A Catalytic Antioxidant Attenuates Alveolar Structural Remodeling in Bronchopulmonary DysplasiaDepartments of Medicine, National Jewish Medical and Research Center; Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado; Department of Pediatrics, University of Texas Health Sciences Center; Southwest Foundation for Biomedical Research, San Antonio, Texas; Department of Pathology, Harvard Medical School; Departments of Pathology, Children's Hospital; and Department of Pathology, Brigham and Women's Hospital, Boston, Massachusetts Correspondence and requests for reprints should be addressed to Ling-Yi L. Chang, Ph.D., National Jewish Medical and Research Center, 1400 Jackson Street, K708, Denver, CO 80206. E-mail: changl{at}njc.org
Superoxide anion and other oxygen-free radicals have been implicated in the pathogenesis of bronchopulmonary dysplasia. We tested the hypothesis that a catalytic antioxidant metalloporphyrin AEOL 10113 can protect against hyperoxia-induced lung injury using a fetal baboon model of bronchopulmonary dysplasia. Fetal baboons were delivered by hysterotomy at 140 days of gestation (term = 185 days) and given 100% oxygen for 10 days. Morphometric analysis of alveolar structure showed that fetal baboons on 100% oxygen alone had increased parenchymal mast cells and eosinophils, increased alveolar tissue volume and septal thickness, and decreased alveolar surface area compared with animals given oxygen as needed. Treatment with AEOL 10113 (continuous intravenous infusion) during 100% oxygen exposure partially reversed these oxygen-induced changes. Hyperoxia increased the number of neuroendocrine cells in the peripheral lung, which was preceded by increased levels of urine bombesin-like peptide at 48 hours of age. AEOL 10113 inhibited the hyperoxia-induced increases in urine bombesin-like peptide and numbers of neuroendocrine cells. An increasing trend in oxygenation index over time was observed in the 100% oxygen group but not the mimetic-treated group. These results suggest that AEOL 10113 might reduce the risk of pulmonary oxygen toxicity in prematurely born infants.
Key Words: metalloporphyrin • morphometry • bombesin-like peptide • neuroendocrine cells
The pathogenesis of bronchopulmonary dysplasia (BPD) is thought to be multifactorial, with prematurity, barotrauma, inflammation, and pulmonary O2 toxicity playing important roles (1). Evidence suggests that an oxidant/antioxidant imbalance exists in lungs that are at risk for BPD. Higher concentrations of lipid peroxidation metabolites such as F2-isoprostanes have been found in premature infants (2, 3). Infants who develop BPD have elevated endothelin-1 in tracheal aspirates (4), known to prime both neutrophils and macrophages to produce more superoxide (5). Preterm babies have lower levels of retinoic acid (6), which has been shown to suppress both superoxide and hydrogen peroxide formation in stimulated neutrophils and macrophages (7). Observations made with preterm animal and human lungs suggest that premature lungs might be deficient in antioxidant enzymes (8–10). In addition, inflammatory mediators that are sensitive to redox regulation are increased in the tracheal aspirates of premature infants (11, 12). The premise of this study is that because increased oxidative stress appears to exist in BPD and oxidants have been postulated to cause lung injury, antioxidant augmentation may be beneficial for this neonatal disease. Treatment of hyperoxic newborn piglets with human recombinant copper zinc superoxide dismutase (SOD) lowered inflammatory cell counts, elastase activity, and albumin concentration in their tracheal aspirates (13). A clinical trial with exogenous bovine copper–zinc SOD reported a decreased incidence of both clinical and radiographic manifestations of BPD (14). In this study, we used a manganic porphyrin that has potent SOD activity in a baboon model of BPD. The catalytic antioxidant metalloporphyrins have been shown to be chemically stable and have high metal affinity and fast rate constants with superoxide. In addition, these compounds have been shown to catalyze the release of O2 from hydrogen peroxide, scavenge peroxynitrite, and inhibit lipid peroxidation (15). Manganese-containing porphyrins have been shown to protect cells against a variety of oxidative stress–inducing agents in vitro, including heat shock, xanthine/xanthine oxidase, paraquat, and lipopolysaccharide (15). Metalloporphyrins have also been shown to provide protection in vivo against paraquat-induced lung injury (16, 17), endotoxin shock (18), and cardiomyopathy in manganese SOD knockout mice (19). Preterm fetal baboons delivered by caesarean section at 140 days of gestation (term = 185 days) and maintained on 100% O2 for 10 days develop clinical and pathologic features that resemble severe BPD (20, 21). We report here that the administration of the catalytic antioxidant AEOL 10113 (Incara Pharmaceuticals Corporation, Research Triangle Park, NC) to premature baboons inhibits alveolar structural modifications and inflammation induced by 100% oxygen; reduces urine levels of bombesin-like peptide (BLP), a proinflammatory peptide and bronchoconstrictor; and reduces the numbers of pulmonary neuroendocrine cells, mast cells, and eosinophils.
Animals All animal care procedures were performed in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals. Protocols were reviewed and approved by the Animal Care Committee of the Southwest Foundation for Biomedical Research in San Antonio, where all animal studies were performed. Fetal baboons were delivered by hysterotomy at 75% of gestation or 140 ± 2 days (21). Standardized procedures were adapted for the care of each premature baboon as previously described (22). Animals were given either 100% O2 continuously or oxygen pro re nata (PRN O2). Because AEOL 10113 has a serum half-life of 0.5 to 1 hour in mice, the antioxidant was administered to 100% O2–exposed baboons by continuous intravenous infusion at a dose of 0.5 mg/kg/day using an infusion rate of 0.1 cc/hour. A preliminary 2-week toxicity study in mice using continuous infusion of the drug (by miniosmotic pumps) was performed. No toxic effect was found by pathology (liver and kidney) and serum levels of alkaline phosphatase, aspartate aminotransferase, alanine aminotransferase, total bilirubin, urea nitrogen, and creatinine at 5 mg/kg/day. Because of the immature and precious nature of the neonatal baboons used in this study, a dose 10-fold lower was chosen. Baboons were supported in a neonatal intensive care unit for 10 days. Details on sample collection are given in the online supplement.
Pharmacokinetics
Morphometry and Pathology
Western Blots
Urine BLP
Statistical Analysis
Pharmacokinetics Figure 1 shows the chemical structure and catalytic activities of AEOL 10113, chemical name of manganese (III) Meso-tetrakis-(N-methylpyridinium-2-yl) porphyrin (28). Serum concentrations of AEOL 10113 in fetal baboons reached an average of 180 ng/ml within 24 hours and remained stable at approximately 250 ng/ml during the first 5 days after delivery. It increased slightly after the 6th day to 320 ng/ml and reached 390 ng/ml by the 10th day (Figure 2) . AEOL 10113 levels in lung, liver, and kidney at autopsy were 331 ± 92 ng/g, 7,229 ± 1,724 ng/g, and 2,441 ± 533 ng/g, respectively. The heart contained approximately 60 ng of AEOL 10113 per gram of tissue. Traces of the drug were detected in the brain (frontal cortex) of only one of the seven AEOL 10113–treated animals, suggesting that the blood/brain barrier remained intact for the most part during the hyperoxic exposure. Examination of hematoxylin and eosin–stained and Pentachrome-stained sections of liver and kidney revealed no detectable tissue injuries.
Antioxidant Enzymes Total lung protein for the three groups of baboons were 50 ± 6.4, 53.2 ± 5.2, and 51.6 ± 3.2 mg/g lung for PRN, 100% O2, and 100% O2 + AEOL 10113 groups, respectively. Neither hyperoxia nor AEOL 10113 appeared to affect the overall protein synthesis in the lung. Protein expressions of manganese SOD and extracellular SOD were measured to see whether the regulation of these antioxidant enzymes was affected in hyperoxia-stressed lungs by treatment with a catalytic antioxidant. Analysis of the immunoblots for the two antioxidant enzymes revealed that exposure to 100% O2 for 10 days did not significantly affect the protein expression of manganese SOD but inhibited the expression of extracellular SOD (Figure 3) . Treatment of 100% O2–ventilated animals with AEOL 10113 had no effect on O2-induced inhibition of extracellular SOD. Manganese SOD expression in antioxidant-treated baboon lungs was not different from those of PRN O2 or 100% O2 animals.
Physiology There was no difference in the birth weights of the animals in the three experimental groups (PRN O2, 534 ± 18 g; 100% O2, 527 ± 14 g; 100% O2 + AEOL 10113, 567 ± 20 g). The clinical course of the animals in either the PRN O2, the 100% O2, or the 100% O2 + AEOL 10113 groups was similar. Table 1 lists the key ventilation and blood–gas parameters. To compare the status of oxygenation in the three groups of animals, the OI, defined by the formula here, was calculated: OI = fraction of inspired oxygen x mean airway pressure x 100 ÷ PO2. A normal OI is low, in the range of one to two. Data from 20 PRN O2 animals as well as 21 100% O2–exposed animals were pooled for comparison with the 7 100% O2 + AEOL 10113–treated fetal baboons. Typically, the OI of animals in all groups started above 10 immediately after delivery and then decreased to below 8 by 8 to 12 hours after delivery. There was no difference in OI during the first 12 hours of life between the three groups of animals. Figure 4 compares the OIs of the three groups of animals for Days 6 to 10. The 100% O2 group showed a significant positive trend (p < 0.0001) of increasing OI over time. The trend was not observed in the other two groups. There were significant mean differences at Day 9 (p < 0.011) and Day 10 (p = 0.0001) between the 100% O2 and PRN O2 groups.
Histology Evaluation Pathologic changes similar to those previously described for this baboon model of BPD were observed. The characteristic overinflated airspaces (22) associated with BPD were observed in animals exposed to 100% O2, and to a lesser extent, in the PRN O2 and AEOL 10113–treated animals. Areas of massive atelectasis, pulmonary interstitial emphysema, and honeycomb lung associated with severe interstitial fibrosis were prominent in the 100% O2–exposed animals without antioxidant treatment. Figures 5A–5C show representative low magnification views of baboon lungs from the three groups of animals. Extensive alveolar wall thickening was found most frequently in the 100% O2–ventilated animals. Trichrome stain revealed collagen deposition mostly in peribronchiolar and perivascular spaces. Diffuse staining was also seen in the interstitium of thickened alveolar septa (Figure 5E); 100% O2 + AEOL 10113–treated animals, in general, displayed less collagen staining than nontreated 100% O2 control subjects (Figure 5F). In the bronchi of 100% O2–ventilated baboons, changes included epithelial hyperplasia, squamous metaplasia, and mucous plugging of small airways (20). The incidents of mucous plugging were reduced in AEOL 10113–treated animals compared with those with 100% O2 alone.
Morphometry Marked changes were found in the alveolar region of the baboon lungs exposed to 100% O2 (Table 2) . Exposure to 100% O2 significantly increased the total volume of alveolar tissue when compared with the PRN O2 animals (56%). Correspondingly, the mean alveolar surface area of the 100% O2–exposed animals was reduced to 72% of that of the PRN O2 animals. The thickness of the alveolar septum was increased by 58% in the 100% O2–exposed animals compared with the PRN O2 animals. The alveolar structure of the animals ventilated with 100% O2 + AEOL 10113 was significantly different from that of animals exposed to 100% O2 alone. The total volume of alveolar tissue and the thickness of the alveolar septum in AEOL 10113–treated fetal lungs were markedly lower than that of the 100% O2–exposed group. In addition, the mean alveolar surface area was greater than that found in the 100% O2 animals.
The mean volumes of the right lower lobes of the three groups of animals were similar (Table 2). Total tissue volume of the airway and vessel compartments, the surface area of the conducting airways and the blood vessels, and the thickness of the airway or vessel walls were also similar among the three groups of fetal baboons studied (data not shown). Reduced alveolar injury in 100% O2 + AEOL 10113–treated fetal baboons in relationship to animals exposed to 100% O2 was also indicated by a significant reduction in the proportion of the right lower lung involved in overinflation, fibrosis, or severe atelectasis (Figure 6A) . Although 40% of the right lower lobes from 100% O2 animals were found to display collapsed alveolar septa or fibrosis, only 14% of the right lower lobes from 100% O2 + AEOL 10113–treated animals showed this pathology.
Inflammation The extent of inflammatory cell infiltration into the airway lumen and alveolar airspaces was greatly reduced in the fetal baboon lungs of 100% O2 + AEOL 10113 animals (Figure 6B) when compared with animals given 100% O2. Examination of the airspace cells in semithin plastic sections of the lung revealed that the majority of the airspace inflammatory cells were macrophages. The percentage of neutrophils in Day 6 tracheal aspirates, necropsy lavage, and the number of neutrophils per square millimeter of lung parenchyma were not affected by treatment with AEOL 10113. Counting of eosinophils and mast cells on lung sections showed increases of both types of inflammatory cells in the parenchyma of 100% O2 animals. AEOL 10113 treatment reduced the cell counts to the level observed in PRN O2 animals (Figure 7) . We found no statistical significance of lung inducible nitric oxide synthase protein expressions among 140-day gestation fetal baboons, 140-day gestation plus 10-day PRN-exposed baboons, 140-day gestation plus 10-day 100% O2–exposed fetal baboons, or AEOL 10113–treated fetal baboons.
Neuroendocrine Cells and BLP Hyperplasia of the neuroendocrine cells was observed in the small airways and parenchyma of fetal baboon lungs exposed to 100% O2 (Figure 8A) . Only an occasional neuroendocrine cell cluster was found in the lungs of preterm baboons treated with 100% O2 + AEOL 10113. Quantitation of the parenchymal neuroendocrine cells on lung sections (cells/mm2 of tissue area) showed a significant decrease of labeled cells in the parenchymal region of 100% O2 + AEOL 10113 animals (Figure 8B) when compared with animals exposed to 100% O2. The decrease in pulmonary neuroendocrine cells at 10 days was preceded by a reduced urine concentration of BLP at 24–48 hours of age. A significant rise in BLP levels at Day 2 after birth (mean twofold increase) has been shown to be associated with the development of BPD in baboons. We also expressed urine BLP levels in the Day 2 samples as the percentage change from the normalized mean urine BLP values in samples collected during Day 1 (Figures 8C and 8D). The sharp percentage increase in urine BLP levels seen in 100% O2–exposed animals was inhibited by the AEOL 10113 treatment.
In this study, a manganese-containing porphyrin with high SOD activity was used to test the benefit of antioxidant treatment in a baboon model of BPD with exposure to 100% O2. It should be noted that this model of BPD resembles the BPD described by Northway and colleagues where inflammation, fibrosis, and smooth muscle hypertrophy in the airways resulted from mechanical injury and oxygen exposure (29). With advances in neonatal care, including the use of antenatal steroid and surfactant and reduced use of high-oxygen tensions, BPD occurs today primarily in severely underweight infants with abnormal lung development as the main consequence (30). The dose of AEOL 10113 given to the fetal baboons was low, being 1/10 of the dose with which no toxicity was observed in mice. Pharmacokinetic analysis of the serum samples from AEOL 10113–treated fetal baboons showed that the serum concentration of the drug quickly equilibrated during the first day and remained stable during Days 2 through 5. The increase of AEOL 10113 in serum during the second half of the oxygen exposure and high concentrations of the drug in both the kidney and the liver at necropsy suggest a slowing down of its removal from the circulation by excretion and/or metabolism. Although accumulation of the drug may indicate impairment of liver and/or kidney function, no overt liver or kidney injury was observed in hematoxylin and eosin or Pentachrome sections after 10 days of treatment at the current dose. The effect of chronic dosing in fetal baboons remains to be determined. Ten days of 100% O2 exposure of these neonatal animals led to structural remodeling, primarily in the alveolar region of the lung. The total thickness of the walls of either large or medium-sized vessels or airways was not changed by hyperoxia. Because of considerable variation in the magnitude of alveolar pathology among the individual fetal baboons in the PRN O2 group in this study, there was no significant difference between the means of alveolar septal thickness and total alveolar surface area of the PRN O2 and 100% O2 groups. However, the volume of total alveolar tissue was significantly greater in the 100% O2 group. AEOL 10113–treated animals demonstrated thinner septa and greater alveolar surface area than animals exposed to 100% O2 without the antioxidant. The protective effect of the catalytic antioxidant may have resulted from improved recruitment of alveoli, since the fraction of alveolar tissue involved in severe atelectasis and overinflation was reduced in drug-treated fetal baboon lungs. It is also possible that AEOL 10113 may protect against hyperoxia-induced inhibition of alveolarization. In addition, the amount of fibrosis in the alveolar interstitium of 100% O2 + AEOL 10113–treated fetal baboon lungs also appeared to be decreased.
The severity of inflammation was also reduced in the airway lumen, alveolar airspaces, and parenchyma. The ability of AEOL 10113 to inhibit inflammatory cell recruitment appeared to be most pronounced for mast cells and eosinophils but not neutrophils. Taken together, these analyses suggest that the AEOL 10113 treatment inhibits specific inflammatory cell responses in this baboon model of BPD. Decreased recruitment of inflammatory cells could possibly reduce alveolar damage by oxygen free radicals or other toxic leukocyte products, such as mast cell tryptase and eosinophil cationic protein (31, 32). The mechanism for the inhibition of leukocyte recruitment by AEOL 10113 is unknown. We speculate that AEOL 10113 may inhibit the activation of the transcription factor nuclear factor- Exposure of fetal baboons to 100% O2 caused a decrease in the lung expression of extracellular SOD. The loss of this enzyme could be an important contributor to lung injury. Although treatment with AEOL 10113 did not alter the downregulation of extracellular SOD, we speculate that AEOL 10113 might partition in tissue in a manner similar to extracellular SOD because it carries a charge of five (28). AEOL 10113 could protect against alveolar injury by effectively replacing the lost extracellular SOD in the extracellular milieu, where the presence of antioxidants to protect against inflammatory cells is important (35).
BLP is a family of neuropeptides associated with morphogenesis, growth, and maturation (36, 37). Gastrin-releasing peptide is the major known BLP associated with lung development in mice, rats, and humans (38). Pulmonary neuroendocrine cells stained positive for BLP have been shown to increase after oxygen exposures in infants with hyaline membrane disease as well as BPD (39). Recently, Sunday and colleagues reported that BLP promotes the development of BPD in fetal baboons (40). In this study, we observed that urine BLP levels are greatly reduced in 100% O2 + AEOL 10113–treated fetal baboons at 1 to 2 days of age. At autopsy, the number of pulmonary neuroendocrine cells in antioxidant-treated fetal baboon lungs was markedly decreased. It appears likely that the decrease in urine BLP levels may be the result of reduced BLP secretion and/or reduced numbers of pulmonary neuroendocrine cells, which are the source of BLP in the lungs. BLP has been reported to stimulate fibroblast migration in culture (41). BLP-positive cells, found in the vicinity of fibrotic lesions, increased in rat lungs with experimental asbestosis (42). We postulate that BLP might be involved in the interstitial fibrosis commonly associated with lung injuries. Treatment of O2-exposed fetal baboons with the blocking monoclonal BLP antibody abrogated the development of pulmonary fibrosis (40). The reduced production of BLP in the lungs of 100% O2 + AEOL 10113–treated animals could contribute to the inhibitory effect of the SOD mimetic on septal thickening. The mechanism by which AEOL 10113 prevented the hyperoxia-induced neuroendocrine cell hyperplasia is unknown but may be linked to redox regulation of cytokines such as tumor necrosis factor- Although the mean OIs of the AEOL 10113–treated baboons were not statistically different from those of the 100% O2 group (p = 0.0617), the trend of increasing OI over time in 100% O2–treated baboons is not found in the antioxidant-treated group. Because of the high cost and limited availability of premature baboons, we were unable to study a significantly large number of fetal baboons or to extend the time frame of the study to reveal a significant difference in OI between control and drug-treated baboons. Nevertheless, our results suggest stabilization of the OI in drug-treated animals. This study shows that AEOL 10113 inhibited 100% O2–induced alveolar structural remodeling, demonstrating a therapeutic effect of the antioxidant against oxygen toxicity. The benefit of the antioxidant treatment for present day BPD where lower oxygen tensions are used and the primary pathology is one of arrested lung development is less certain (30). However, hyperoxia has been shown to inhibit lung growth (44, 45). Alveolar development in SOD knockout mice was observed in mice kept in room air and 50% O2 (46). Recent findings of increased oxidative stress (2–7) and low levels of antioxidants in fetal lungs (47, 48) suggest that redox balance could play a role in the arrest of lung growth in BPD, and that antioxidant augmentation could also be of benefit in this situation. Concern has been raised in recent years about the safety of antioxidant treatment for BPD due to inhibition of cell growth by some antioxidants (49). AEOL 10113 is being tested in a 125-day gestation baboon model of BPD maintained with PRN O2 for 14 days. Preliminary results on this model of BPD have been reported to show that treatment with the antioxidant enhanced the development of secondary septa in PRN O2-treated animals, suggesting that AEOL 10113 does not inhibit lung growth (50). BPD is a chronic lung disease of the premature infant. Adolescents and young adults who suffered from BPD in infancy have persistent abnormalities in pulmonary function (51) consistent with irreversible structural changes. Morphometric studies of the lungs of young children who died with BPD showed severe alveolar structural changes with decreased alveolar internal surface area (52). Structural alterations in the lungs resulting from BPD can be an important factor for long-term consequences. Preventing permanent structural changes from occurring is likely to result in the improvement of lung functions in later life. We conclude that antioxidant treatment with low molecular weight catalytic antioxidants, therefore, is a potential therapy for the prevention of BPD in premature infants.
The authors thank all of the personnel at the Baboon Resource Center and its director, Dr. Jacqueline Coalson. They express special gratitude to Ms. Vicki Winter for her seamless coordination of the fetal baboon tissue resources and assistance to requests for information. They also thank Dr. Gail Deutsch for her assistance on fetal inflammatory cell identification and reading of the pathology. The authors are grateful for technical assistance by Mr. Joseph Rice and Ms. Karen Dockstader.
Supported in part by grants from Incara Pharmaceuticals Corporation (L.C., J.D.C., and B.J.D.) and by National Institutes of Health grants U01 HL63397 (L.C., J.D.C., and B.J.D.), U0152636 (B.A.Y.), U01 HL52638 (M.S. and M.E.S.), and National Institutes of Health P51RR13986 (Southwest National Primate Research Center). This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form March 20, 2002; accepted in final form September 5, 2002
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
-actin.
B and the subsequent induction of chemokines and cytokines involved in the inflammatory cascade. We have recently demonstrated that treatment with AEOL 10113 inhibited ovalbumin-induced vascular adhesion molecule-1 expression in mice (
