Advertisement
Extracellular superoxide dismutase in the airways of transgenic mice reduces inflammation and attenuates lung toxicity following hyperoxia
Rodney J. Folz,1,2 Amir M. Abushamaa,1 and Hagir B. Suliman1
1Department of Medicine, and2Department of Cell Biology, Division of Pulmonary and Critical Care Medicine, Duke University Medical Center, Durham, North Carolina 27710, USA
Address correspondence to: Rodney J. Folz, Box 2620, Room 331 MSRB, Duke University Medical Center, Durham, North Carolina 27710, USA. Phone: (919) 684-3539; Fax: (919) 681-8936; E-mail: rodney.folz@duke.edu
Find articles by Folz, R. in: JCI | PubMed | Google Scholar
1Department of Medicine, and2Department of Cell Biology, Division of Pulmonary and Critical Care Medicine, Duke University Medical Center, Durham, North Carolina 27710, USA
Address correspondence to: Rodney J. Folz, Box 2620, Room 331 MSRB, Duke University Medical Center, Durham, North Carolina 27710, USA. Phone: (919) 684-3539; Fax: (919) 681-8936; E-mail: rodney.folz@duke.edu
Find articles by Abushamaa, A. in: JCI | PubMed | Google Scholar
1Department of Medicine, and2Department of Cell Biology, Division of Pulmonary and Critical Care Medicine, Duke University Medical Center, Durham, North Carolina 27710, USA
Address correspondence to: Rodney J. Folz, Box 2620, Room 331 MSRB, Duke University Medical Center, Durham, North Carolina 27710, USA. Phone: (919) 684-3539; Fax: (919) 681-8936; E-mail: rodney.folz@duke.edu
Find articles by Suliman, H. in: JCI | PubMed | Google Scholar
Published April 1, 1999 - More info
J Clin Invest. 1999;103(7):1055–1066. https://doi.org/10.1172/JCI3816.
© 1999 The American Society for Clinical Investigation
Received: April 23, 1998; Accepted: February 11, 1999
-
Methods
Generation of hEC-SOD Tg mice. The human surfactant protein-C (SP-C), t-intron transgene expression vector was kindly obtained from J.R. Wispe (Children’s Hospital Medical Center, Cincinnati, Ohio, USA). The construction of this vector has been described previously (10). Briefly, this vector contains the 3.7-kb promoter region of the SP-C gene, including the TATAA box, and directs transgene expression to alveolar type II and nonciliated distal bronchial epithelial cells (18). A full-length hEC-SOD cDNA (1.4 kb) encoding an 18–amino acid signal peptide as well as the 222–amino acid mature polypeptide (19) was ligated into the EcoRI site of the expression vector polylinker (Figure 1). Plasmid DNA containing the sense-strand transgene vector was purified, linearized by restriction endonuclease digestion using NotI and NdeI, and an ∼6-kb linear fragment (containing the SP-C promoter, hEC-SOD transgene, and SV40 t-intron and poly A signal) was isolated after agarose gel electrophoresis. Purified linearized DNA at 2 μg/ml in 40 mM Tris·acetate, 2 mM Na2EDTA·2H2O (pH 8.5) was injected into the pronuclei of fertilized eggs isolated from mice [(B6×SJL)F1 × (B6×SJL)F1] obtained from The Jackson Laboratory (Bar Harbor, Maine, USA). Eggs surviving microinjection were implanted into the oviducts of pseudopregnant foster mothers (B6×SJL) F1.
Figure 1Schematic diagram of the strategy used to overexpress hEC-SOD in the mouse lung. The human surfactant protein-C, t-intron Tg expression vector was used to construct a full-length hEC-SOD expression vector. Tg offspring were identified by using PCR to amplify a unique fragment of the hEC-SOD gene (172 bp) from mouse tail genomic DNA. The ethidium bromide–stained agarose gel shows PCR products of Tg positive mice (+) and their negative littermates (–). hEC-SOD, human extracellular superoxide dismutase; Tg, transgenic.
Tg founders were bred into the strain B6C3 to obtain sixth-generation animals and were identified by PCR analysis of tail DNA using the forward primer EC1 (5′-AGACACCTTCCACTCTGAGG-3′) and the reverse primer EC2 (5′-GTTTCGGTACAAATGGAGGC-3′). The PCR was performed using a Perkin-Elmer 9600 thermal cycler and a final sample volume of 25 μl under the following conditions: EC1 and EC2 (10 pmol each), tail DNA (350 ng), 200 μM of dATP, dTTP, dCTP, dGTP, 2.5 mM of MgCl2, Taq polymerase (0.625 U), and 1× PCR buffer (50 mM KCL, 10 mM Tris-HCl [pH 8.3]). After heating the sample to 94°C for 3 min, the sample underwent 32 cycles of three-step PCR (94°C for 40 s, followed by 57°C for 30 s, followed by 72°C for 45 s).
Measurement of lung antioxidant levels. For EC-SOD determinations, tissues from three mice in each group were weighed and homogenized in 10 volumes of ice-cold 50 mM potassium phosphate buffer (pH 7.4) with 0.3 M KBr, and a set of antiproteolytic agents (0.5 mM phenyl methylsulfonyl fluoride, 3 mM diethylentriaminepentaacetic acid, 90 mg of aprotinin per liter, 10 mg of pepstatin per liter, 10 mg of chymostatin per liter, and 10 mg of leupeptin per liter). The homogenate was then sonicated, and separation of EC-SOD from CuZn-SOD and Mn-SOD was accomplished by passage over a concanavalin A–Sepharose column as described previously (20). We have found the recovery of EC-SOD to be >98%. The EC-SOD activity was measured by inhibition of cytochrome c reduction at pH 10 as described (21). One unit of SOD activity is defined as the enzyme activity needed to inhibit a 50% reduction in cytochrome c.
The activities of CuZn-SOD and Mn-SOD were determined by measuring the inhibition of xanthine plus xanthine oxidase–mediated cytochrome c reduction at pH 7.8 (21). To distinguish the contributions of CuZn-SOD and Mn-SOD to the total superoxide dismutase activity, 1 mM KCN was used to inhibit the activity of CuZn-SOD.
For catalase activity, the tissues were homogenized in 1.0% Triton X-100 using a variable-speed tissue tearer (Brinkmann Instruments Inc., Westbury, New York, USA) at 20,000 g for 1 min on ice. The resulting homogenate was centrifuged at 10,000 g at 4°C for 20 min. The enzyme activity was determined by the method described by Aebi (22). Briefly, the supernatant was diluted with 1.5 volumes of the assay buffer containing 50 mM K2HPO4 and 50 mM NaH2PO4 at pH 7.0. The reaction was initiated by adding 1.0 ml of 30 mM H2O2, and the change in absorbance at 240 nm was measured at 25°C for 1 min. A portion of the remaining sample was used for protein determination. The specific activity of catalase is defined as micromoles of H2O2 consumption per minute per milligram of tissue protein. Calculation of H2O2 concentration was made using a millimolar extinction coefficient for H2O2 of 0.0394 (22).
The glutathione peroxidase (GPx) activities in tissue homogenates and bronchoalveolar lavage fluid (BALF) were determined by the indirect, coupled test procedure (23). Briefly, the oxidized glutathione (GSSG) produced during GPx enzyme reaction was immediately reduced by the exogenous addition of nicotinamide adenine dinucleotide phosphate (NADPH) and glutathione reductase. Therefore, the rate of NADPH consumption was monitored as a measurement for the rate of GSSG formation. The enzyme reaction was conducted in a buffer containing 20 mM potassium phosphate (pH 7.0), 0.6 mM EDTA, 0.15 mM NADPH, 4 U of glutathione reductase (Sigma Chemical Co., St. Louis, Missouri, USA), 2 mM reduced glutathione (GSH), 1 mM sodium azide, and 0.1 mM H2O2 at 25°C. The rate of decrease in absorption of NADPH at 340 nm was measured. The GPx activity is defined by nanomoles of NADPH consumption per minute per milligram of tissue protein at 2 mM GSH. Consumption of NADPH was calculated using a millimolar extinction coefficient for NADPH of 6.22 (23).
Immunohistochemistry. The distribution of hEC-SOD expression in the Tg mice lungs was evaluated with light microscopic immunohistochemistry using a biotin/streptavidin–horseradish peroxidase labeling technique. Briefly, paraffin sections were rehydrated with phosphate buffer containing 0.1% Tween-20 (PBS-T) and blocked by immersion in 2.0% normal goat serum or 2.5% nonfat milk. For EC-SOD immunostaining, the tissue sections were incubated with a 1:1,000 dilution of the primary antibody (WT474) for 2 h at 37°C. The sections were washed with PBS-T before incubation with the goat anti–rabbit IgG secondary antibody for 1 h at 37°C, washed again in PBS-T, and detected with peroxidase using the Vector standard kit as recommended by the manufacturer (Vector Laboratories, Burlingame, California, USA). Preimmune rabbit serum and preabsorbed antibodies, prepared by passing the WT474 antibody through the hEC-SOD peptide–bound column, were used as controls for the EC-SOD immunostaining.
Neutrophil depletion. hEC-SOD Tg and Wt littermate mice (6–8 weeks old) were randomly assigned into four groups (two Tg groups and two Wt groups containing 18 mice in each group). One group each of Tg and Wt animals were neutrophil-depleted (ND) by administering polyclonal rabbit anti–mouse neutrophil antibody (24) (Accurate Chemical & Scientific Corp., Westbury, New York, USA). The anti-neutrophil antibody was administered by intraperitoneal injections (0.2 ml of 1:15 dilution in PBS) for three consecutive doses every 12 h, followed by a fourth dose after 50 h. Concurrently, mice from the third and fourth groups were injected with vehicle only (PBS) and used as controls for the neutropenic animals. These mice were exposed to hyperoxia on the fourth day after confirming agranulocytosis by Wright Giemsa–stained buffy coat from peripheral blood smears and total leukocyte counts. The effective neutropenia was 80–86%, with the total neutrophil count decreasing from a range of 302–540 cell/mm3 in the PBS-treated animals to 54–80 cell/mm3 in the anti-neutrophil antibody–treated group (P < 0.05).
Exposure to normobaric hyperoxia. Wt and Tg mice strain 32 (6–8 weeks old) were randomly assigned and housed in a small-animal, airtight plexiglas chamber. Both male and female mice were equally represented in each experimental cohort. These mice were exposed continuously to hyperoxic conditions using an oxygen flow rate of 10 l/min. The concentration of O2 in the chambers was monitored continuously with a Servomax O2 analyzer (model 572; Sybron, Norwood, Massachusetts, USA) and maintained to >99%, except for 10 min/day when the cages were opened to room air during periods of cleaning and general maintenance. The CO2 concentration was held below 0.1%, the temperature was controlled between 25°C and 26°C, and the humidity was <40%. The mice were provided ad libitum access to food and water and were observed continuously for signs of discomfort and O2 toxicity. Mice used in this experiment were sacrificed at 48, 72, and 84 h after exposure. Control Tg and Wt mice were kept in identical chambers and exposed only to room air. ND Tg and Wt animals, as well as their vehicle-only treated controls, were exposed to >99% oxygen for only 72 hours as already described here. In addition, a similar set of ND animals were kept in identical chambers exposed only to room air.
Bronchoalveolar lavage. Mice were anesthetized by intraperitoneal injection of pentobarbital (50 μg/g). After the onset of adequate anesthesia, the trachea was exposed and a 20-gauge needle was used to puncture the trachea and was replaced with a 21-gauge blunt needle. A silk ligature was fastened around the trachea to secure the needle. Bronchoalveolar lavage (BAL) was performed with 1 ml PBS using a 1-ml syringe and repeated once. The collected fluid was immediately processed as follows. BALF (300 μl) was cytocentrifuged and stained with a leukostat stain kit (Fisher Scientific Co., Pittsburgh, Pennsylvania, USA). A minimum of 300 cells were subsequently counted to obtain a differential cell count. Cellular BALF (30–50 μl) was stained with trypan blue, and the total cell count was manually obtained using a Neubauer hemocytometer (Reichert, Buffalo, New York, USA). The remaining BALF was centrifuged at 1,800 g for 5 min at 4°C, and the supernatant was decanted and recentrifuged a second time at 12,000 g for 10 min at 4°C. The supernatant was collected and saved at –80°C for further analysis, and the pelleted cells were snap frozen in liquid nitrogen and stored at –80°C.
Lung edema or wet/dry weight ratio. After euthanasia of mice, the lungs were surgically dissected free from the heart, trachea, and main bronchi and removed. Each lung was blotted dry, weighed, and then dried to a constant weight by placing the lung specimen in a vacuum oven at 60°C for 48 h. The ratio of the wet lung to the dry lung was calculated to assess tissue edema.
Lung permeability. The permeability and/or cellular injury of the lung was determined by measuring the BALF lactate dehydrogenase (LDH) spectrophotometrically using the LDH assay kit (Sigma Chemical Co.) and total protein using the BCA assay (Pierce Chemical Co., Rockford, Illinois, USA) according to directions by the manufacturers.
Lipid peroxidation as measured by malondialdehyde concentration. Lipid peroxidation in the form of malondialdehyde (MDA) was measured according to the procedure of Wong et al. (25). Briefly, 100 μl of BALF or lung homogenate sample was treated with 150 μl of 0.44 M phosphoric acid and 50 μl of 0.6% thiobarbituric acid. The mixture was heated at 100°C in a heating block for 1 h. Samples were then kept on ice until HPLC injection, when they were protein-precipitated and neutralized by adding 300 μl of a 1 M NaOH/methanol (1:11) solution. Then 50 μl of the supernatant liquid was injected into a Supelcosil LC-18 column (15-cm long, 4.0-mm internal diameter, 5 μm) using an HPLC system (model 1100; Hewlett-Packard, Palo Alto, California, USA). The mobile phase was 40% methanol/60% 50 mM potassium phosphate buffer (pH 6.8). The flow rate was 1 ml/min. Detection was accomplished by a fluorescence detector (model 1046A; Hewlett-Packard) set at λεx = 525 nm and λεm = 550 nm. Retention time of the MDA derivative was 2.8 min. The assay is linear from 30 nM to 326 nM MDA (sensitivity is 0.5 pmol per injection).
Glutathione. Total and oxidized glutathione was measured according to the procedure described by Liu and Kehrer (26). BALF or lung homogenate (100 μl) were treated with 50 μl of 0.1 M Tris-HCl (pH 8.5) and 100 μl of 25 mM DTT. After keeping the samples on ice for 30 min, the protein component was precipitated by adding 250 μl of 3.75% 5-sulfosalicylic acid followed by centrifugation at 3,500 g for 2 min. A 200-μl aliquot of the supernatant liquid was then treated with 200 μl of o-phthaldialdehyde solution (50 mg o-phthaldialdehyde were dissolved in 0.5 ml methanol, and the final volume was made up to 10 ml with 0.4 M potassium borate [pH 9.9]) for 2 min at room temperature. Then 200 μl of 250 mM sodium phosphate (pH 7.0) was added, and 100 μl of sample was injected into HPLC. For GSSG, the same procedure was applied, except for the following modifications: BAL samples were first treated with 200 mM N-ethylmaleimide (20 mM final concentration) to block free GSH. Lung tissue was homogenized in 5% 5-sulfosalicylic acid containing 20 mM NEM and 0.5 mM EDTA.
After derivitization, the samples were injected onto a Supelcosil LC-18 column (15-cm long, 4.0-mm internal diameter, 5 μm) and eluted using a mobile phase containing 7.5% methanol/92.5% 0.15 M sodium acetate buffer (pH 7.0). Fluorescence detection was performed using λεx = 340 nm and λεm = 420 nm. The peak retention time was 3.8 min. The assay is linear from 25 nM to 1 μM GSH (sensitivity is 0.1 pmol per injection).
Histopathology. The lungs from six mice in each group and at each time point were fixed in situ for microscopy by intratracheal instillation of 4% paraformaldehyde in PBS at 20 cm H2O pressure for 10 min and subsequently removed from the thorax and immersed in additional fixative for 2 h. These tissues were stored in 2% paraformaldehyde solution at 4°C until they were paraffin-embedded, sectioned, and stained with hematoxylin and eosin.
Histopathologic evidence for lung injury was scored on a scale from 0 (normal) to 4 (most severe) essentially as described by Arkovitz et al. (27). Parameters assessed were the degree of vascular congestion, hemorrhagic edema, thickening of alveolar septum (edema), hyaline membrane formation in the distal airway, and neutrophil accumulation.
Cytokine measurements. Measurement of murine macrophage inflammatory protein-2 (MIP-2) in the BALF was performed by ELISA (sensitivity: 2.0 pg/ml; R&D Systems Inc., Minneapolis, Minnesota, USA) following the manufacturer’s instructions. Murine tumor necrosis factor-α (TNF-α) solid-phase sandwich ELISA (sensitivity: 3.0 pg/ml; BioSource International, Camarillo, California, USA) was used to measure TNF-α in mice BALF following the manufacturer’s instructions.
Intercellular adhesion molecule-1 (ICAM-1) mRNA levels were determined as follows. Total cytoplasmic RNA was extracted from mice lung using the Trizol Total RNA isolation kit (GIBCO BRL, Gaithersburg, Maryland, USA). RNA concentration was determined spectrophotometrically at 260 nm. Total RNA (1 μg) from each sample was reverse-transcribed into scDNA, using oligo (dT) as a primer. PCR amplification was carried out in a thermal cycler for 25 cycles for glyceraldehyde phosphate dehydrogenase (GAPDH) and 27 cycles for mouse ICAM-1 (the number of cycles were determined by titration as the acceptable cycles that can amplify visible products on ethidium bromide–stained gels during the exponential phase of the PCR). The cycling parameters routinely used were: denaturation at 94°C for 30 s, annealing at 56°C for GAPDH or 59°C for ICAM-1 for 45 s, and extension at 72°C for 45 s. Primers specific for mouse GAPDH (GAPDHF, 5′-AACTTTGGCATTGTGGAAGG-3′; GAPDHR, 5′-ACACATTGGGGGTAGGAACA-3′) and ICAM-1 (mICAM-1F, 5′-CGATCTTCCAGCTACCATCC-3′; mICAM-1R, 5′-GTTAAGGTCCTCTGCGTCTC-3′) were used to amplify a 223-bp and 282-bp fragment of mouse GAPDH and ICAM-1, respectively. The amplified products (10 μl) were electrophoresed in 1.5% agarose gels, stained with ethidium bromide, and viewed under ultraviolet light. Quantification of the amplification product was obtained by densitometric scanning of the ethidium bromide–stained gels using ImageQuant (Version 1.1; Molecular Dynamics, Cambridge, Massachusetts, USA). The GAPDH signals were used to control for variation in the efficiency of RNA extraction, RT, and PCR. Levels of ICAM-1 mRNA (after normalization to GAPDH expression) were compared in the Tg vs. Wt mice at 48, 72, and 84 hours of hyperoxia exposure.
Statistical analysis. Data are expressed as the mean ± SEM. Statistical analysis was performed on an IBM-compatible personal computer with commercially available software (SigmaStat 2.03; Jandel Scientific Software, San Rafael, California, USA). Unless otherwise indicated, a one- or two-way ANOVA followed by Fisher’s least significant difference test for multiple comparisons was used. A difference was considered statistically significant at P < 0.05.
-
Results
Characterization of the hEC-SOD Tg mice. The strategy used for creating Tg mice overexpressing hEC-SOD in the mouse lung is shown in Figure 1. We initially identified five founder strains by PCR analysis of tail DNA, and all subsequent work was performed using founder strain 32, whereas line 15 was used to rule out the possibility of a gene-insertion effect. Additionally, the hEC-SOD Tg mice appeared healthy throughout this study and showed no signs of accelerated senescence when compared with their Wt littermates.
The hEC-SOD transgene protein was only detected in the lungs of PCR-positive Tg mice and not in liver, heart, kidney, brain, skeletal muscle, small intestine, and blood, indicating a highly tissue-specific pattern of transgene expression (data not shown). The EC-SOD activity in the BALF and lung tissue of Tg mice was significantly higher than that of their Wt littermates (Table 1). In BALF, there was an approximate 2.9-fold increase, whereas in the lung tissue, there was a threefold increase in EC-SOD enzyme activity (Table 1). The increased expression of the human transgene did not measurably affect the basal expression of endogenous mouse lung EC-SOD (Figure 2). Additionally, no effect was observed in the expression of other BALF and/or lung tissue antioxidants — namely, Mn-SOD, CuZn-SOD, GPx, catalase, and GSH (Table 1).
Figure 2Western blot analysis of human and mouse EC-SOD from the lungs of wild-type (Wt) and hEC-SOD transgenic (Tg) mice. Lung protein homogenates from Wt and Tg mice were electrophoresed in 12.5% SDS–polyacrylamide gel. The separated proteins were transferred to Immobilon-P and probed with the anti–hEC-SOD antibodies (WT474) or with anti–mouse EC-SOD (SS618) polyclonal antibodies. (a) Western blot of total lung protein (10 μg) probed with WT474. Lanes 1, 2, and 3 reflect lung homogenate from three independent Tg mice, whereas lanes 4, 5, and 6 are from Wt littermate controls. (b) An identical Western blot as in a was probed with SS618. Of note, hEC-SOD typically shows two bands with the lower-molecular-weight species representing a truncated COOH-terminus. Mouse EC-SOD typically shows three bands, with the lower two also representing COOH-terminal truncations (14).
Table 1Antioxidant content of BALF and lung tissue from wild-type and hEC-SOD transgenic mice under normal conditions while breathing room air
Immunohistochemical staining of mouse lung tissues using anti–hEC-SOD antibodies showed expression of the transgene only in type II alveolar epithelial cells (Figure 3a) and in nonciliated bronchial epithelial lining cells (Figure 3b). No staining was observed in the sections that were stained with the preimmune serum or the preabsorbed WT474 antibodies (data not shown).
Figure 3Immunohistochemical staining of mouse lung tissues using rabbit anti–hEC-SOD antibodies. (a) Tg mouse lung demonstrating staining of type II cells with anti–hEC-SOD antisera. (b) Tg mouse showing hEC-SOD staining of the nonciliated bronchoepithelial lining cells. (c) Wt mouse lung did not stain with the hEC-SOD antibody. (d) Normal Wt mouse bronchoepithelial lining cells did not stain with anti–hEC-SOD antibody. ×225.
Hyperoxia exposure and mortality. Exposure to >99% oxygen was lethal to mice, especially after 84 hours. Hyperoxia was lethal to 40% of the exposed Wt mice (12/30) and 11.6% of the Tg mice (5/43). Using the rate and proportion z-test, the mortality rate was significantly higher in the Wt mice (P = 0.011).
BAL cell counts. There was no difference in total cell counts between Wt and Tg mice after 48 hours of hyperoxia (Figure 4a). However, the total cell count increased significantly by 72 hours in the Wt mice when compared with the Tg mice (4.5 × 105 ± 0.42 vs. 2.5 × 105 ± 0.29; P < 0.05) and remained elevated even at 84 hours (5.6 × 105 ± 0.54 vs. 3.6 × 105 ± 0.38; P < 0.05). This increase in the total BAL cell count was mainly due to the disproportionate recruitment and influx of the polymorphonuclear cells (PMNs) (Figure 4b). The Wt mice showed an approximate 4.8-fold increase in PMNs at 72 hours and a 4.7-fold increase at 84 hours over that of Tg littermates (P < 0.05; Figure 4b). ND significantly (P < 0.05) suppressed neutrophil migration into the alveolar spaces in hyperoxic mice, and overexpression of hEC-SOD further suppressed PMN BALF counts (Figure 4c).
Figure 4Total cell and PMN counts in the BALF from wild-type (Wt) and transgenic (Tg) mice exposed to hyperoxia. (a) Total cell counts at 48, 72, and 84 h of exposure to >99% oxygen. (b) PMN cell counts at 48, 72, and 84 h of exposure to >99% oxygen. Both total cell and PMN counts significantly increased with time of exposure. (c) Tg and Wt animals were neutrophil-depleted (ND) or not (placebo), and were subsequently exposed to >99% oxygen or room air (RA) for 72 h followed by BALF analysis of total PMN cell counts. The Wt oxygen-exposed groups showed significant increases in PMN cell counts when compared with the Tg oxygen-exposed groups. In addition, in Wt animals, neutrophil depletion alone significantly reduced the BALF PMN cell counts vs. the oxygen–placebo (O-placebo) group. Values are expressed as mean ± SEM (n = 6). *P < 0.05 between groups; **P < 0.01 between groups. BALF, bronchoalveolar lavage fluid; PMN, polymorphonuclear neutrophil.
Markers of lung injury and edema. The amount of acute lung damage elicited by hyperoxia was measured by an increase in BALF total protein and BALF LDH. Hyperoxia exposure resulted in significant increases with time of exposure for both parameters tested (Figure 5, a and b). However, the EC-SOD Tg mouse strain showed a marked attenuation of these lung injury parameters, at both 72 and 84 hours, when compared with their Wt littermates (P < 0.05). ND in Wt mice significantly lessened hyperoxic damage to the lung by 72 hours of exposure as assessed by the total protein and LDH in the BALF (Figure 5, c and d). ND in the Tg mice showed a trend to decrease further acute lung damage, but this difference was not found to be significant (Figure 5, c and d).
Figure 5Markers of lung injury. (a) Total protein in the BALF from mice exposed to hyperoxia at 48, 72, and 84 h. The total protein concentration increased significantly in the wild-type (Wt) group at 72 h. (b) LDH activity in the BALF from mice exposed to hyperoxia at 48, 72, and 84 h. LDH increased with time and was significantly higher in Wt mice. (c) The total protein concentration in the BALF of neutrophil-depleted (ND) animals was significantly reduced compared with non-ND oxygen-exposed mice. For a given oxygen exposure group, overexpressing hEC-SOD further reduced the BALF total protein. (d) LDH activity was decreased significantly in the ND mice. Values are expressed as mean ± SEM. (n = 6). *P < 0.05 Wt vs. Tg group; **P < 0.01 Wt vs. Tg group. LDH, lactate dehydrogenase.
We used the lung wet/dry ratio in an attempt to probe pulmonary capillary permeability differences between the two groups of mice. The length of hyperoxia exposure correlated with an increase in pulmonary edema fluid as measured by the wet/dry ratio (Figure 6a). The edema was significantly higher in the Wt mice at 84 hours when compared with EC-SOD Tg littermates (P < 0.05). Furthermore, ND decreased the lung permeability significantly (P < 0.05) in the Wt mice. However, no significant difference was noted between the ND and non-ND Tg mice (Figure 6b).
Figure 6Lung edema. (a) Lung wet/dry ratios at 48, 72, and 84 h of exposure to hyperoxia. A significant increase in the wet weight of wild-type (Wt) lungs was observed at 84 h of hyperoxic exposure. (b) After 72 h of hyperoxia exposure, neutrophil depletion and overexpression of hEC-SOD decreased lung wet/dry ratio significantly in the Wt vs. transgenic (Tg) group (*P < 0.05). Values are expressed as mean ± SEM (n = 6).
Specificity of EC-SOD expression in Tg mice strains. Studies were undertaken to determine whether the tolerance of Tg animals to hyperoxia was the direct result of overexpressing hEC-SOD in the mouse lung or simply due to the transgene’s random insertion into the mouse genome. A second mouse Tg founder line, line 15, was partly characterized and shown to have similar levels of hEC-SOD in the mouse lung as was found in line 32. Six animals each from line 15, line 32, and Wt were exposed to >99% O2 for 72 hours, and the development of acute lung injury was determined as measured by the total protein and LDH in the BALF. We found Tg line 15 and line 32 to have similar levels of protection from hyperoxia, as judged by BALF protein, and both were significantly less than Wt (288 ± 182, 220 ± 101, 633 ± 218 μg/ml, respectively; P < 0.05). Parallel findings were also noted for LDH levels in BALF (153 ± 35, 132 ± 42, 314 ± 63 U/l, respectively; P < 0.05). We interpret these findings to mean that the tolerance to hyperoxia in the hEC-SOD Tg mice is the result of overexpression of the transgene and is not due to local gene effects of inserting DNA into the mouse genome.
Lung antioxidants. At each of the time points indicated, the total SOD activity was significantly higher in BALF and lung homogenate in Tg vs. Wt mice (Figure 7, a and b) and was mainly due to increases in the hEC-SOD isoenzyme (data not shown). Although the total SOD activity showed a trend to increasing with time of hyperoxia exposure in the Wt animals, this was not significant (Figure 7, a and b).
Figure 7Total SOD in the lung and BALF from mice exposed to hyperoxia. (a) SOD activity (U/ml BALF) in the BALF of the transgenic (Tg), but not wild-type (Wt), mice increased significantly with time. There was a trend to increasing with time in Wt, but this was not significant. (b) The total SOD activity (U/mg lung protein) in the lung homogenate of both Wt and Tg mice showed a trend to increasing with time of oxygen exposure, but this was not significant. At the three time points tested, Tg mice showed significantly higher enzyme activities compared with Wt controls. Values are expressed as mean ± SEM (n = 6). *P < 0.05 Tg vs. Wt.
The total glutathione content in the lung tissue (Wt: 9.7 ± 1.5, 14.7 ± 1.3, 17.5 ± 1.4 μmol/mg protein; Tg: 10.4 ± 2.0, 14.5 ± 1.0, and 18.0 ± 1.0 μmol/mg protein at 48, 72, and 84 hours, respectively) and the BALF (Wt: 4.8 ± 1.1, 6.3 ± 1.0, 8.6 ± 0.8 nmol/ml; Tg: 4.6 ± 1.0, 4.9 ± 0.6, and 7.0 ± 0.9 nmol/ml at 48, 72, and 84 hours, respectively) increased significantly with time of hyperoxia exposure but did not differ between the two groups of mice.
The level of catalase activity was significantly increased (P < 0.05) in the hyperoxia-exposed Tg mice, most prominently during the 72-hour adaptation period, and began to decrease by 84 hours in the two groups of mice (Figure 8a). The GPx activity in BALF increased significantly (P < 0.05) with time of exposure in both groups of animals but showed significantly higher levels (P < 0.05) in Tg mice at 84 hours compared with Wt littermates (Figure 8b). There was no discernible pattern of GPx activity in the total lung preparations in the two groups of mice (Figure 8c).
Figure 8Antioxidant enzymes levels after hyperoxia exposure. (a) Catalase activity was measured in the total lung homogenate from mice exposed to hyperoxia. The enzyme activity increased significantly with time in transgenic (Tg) and wild-type (Wt) mice. (b) GPx activity was measured in the BALF of mice after exposure to hyperoxia. The enzyme increased significantly by 84 h of exposure. (c) GPx activity in the total lung homogenate from mice after exposure to hyperoxia. Values are expressed as mean ± SEM (n = 6). *P < 0.05 Tg vs. Wt. GPx, glutathione peroxidase.
Oxidation of glutathione and lipids. The percentage of the GSSG was significantly increased in the lung tissues of Wt and Tg mice by 84 hours of exposure (Wt: 19.6 ± 2.4, 53.0 ± 4.5, 71.0 ± 6.1; Tg: 21.0 ± 2.1, 66 ± 8.4, 86.0 ± 9.5 at 48, 72, and 84 hours, respectively). In the BALF, the percentage of oxidized GSSG also increased significantly with time of exposure (Wt: 66.2 ± 4.7, 75.2 ± 7.7, 82.1 ± 7.0; Tg: 45.3 ± 7.0, 69.6 ± 13.5, 85.5 ± 5.8 at 48, 72, and 84 hours, respectively). At 72 hours of hyperoxia, ND independently and significantly reduced the GSSG levels in BALF (Figure 9a).
Figure 9Markers of oxidative stress in the BALF after 72 h of hyperoxia exposure. (a) The level of oxidized GSH was significantly elevated in all groups after 72 h of >99% oxygen exposure, particularly in the oxygen–placebo (O-placebo) wild-type (Wt) mice in which the level of oxidized glutathione (GSSG) exceeds the GSSG levels in the other groups of mice. (b) The level of lipid peroxides measured as malondialdehyde was significantly attenuated in the neutropenic animals and hEC-SOD transgenic (Tg) mice. Values are expressed as mean ± SEM (n = 6). *P < 0.05 Tg vs. Wt, and neutrophil-depleted (ND) vs. controls; **P < 0.01 Tg vs. Wt, and ND vs. controls. GSH, reduced glutathione.
Exposure to hyperoxia resulted in enhanced lipid peroxidation, as measured by levels of MDA in both the BALF and lung tissue and increased with the time of exposure. In the Wt BALF, MDA levels measured 43.5 ± 4.1, 305.6 ± 41.5, and 417.2 ± 49.8 pmol/ml at 48, 72, and 84 hours of exposure, respectively, whereas BALF, MDA levels in Tg mice measured 19 ± 1.3, 123.7 ± 14, and 396.6 ± 33.7 pmol/ml. The difference between Wt and Tg animals was significant (P < 0.05) at 72 hours of exposure. When total lung MDA levels were measured, they showed an increase with time of exposure in the two groups but no significant differences between them (Wt: 9.4 ± 1.4, 16.7 ± 2.5, 21.2 ± 2.4 pmol/mg protein at 48, 72, and 84 hours; Tg: 8.8 ± 10, 20.5 ± 1.9, 24.7 ± 1.6 pmol/mg protein at 48, 72, and 84 hours, respectively). After 72 hours of hyperoxia exposure, there was a significant attenuation of BALF lipid peroxidation in ND Wt mice when compared with non-ND animals (Figure 9b). A further attenuation of BALF MDA levels was seen in ND Tg mice (Figure 9b).
Effect of hyperoxia on cytokine release in the BALF. The levels of TNF-α and MIP-2 in the BALF from ND and non-ND Wt and Tg mice were measured after exposure to either 72 hours of >99% O2 or to room air. With room air exposure, the levels of TNF-α and MIP-2 in the BALF of Wt and Tg mice did not differ from each other and were near the limits of detection of our assays (Figure 10, a and b). The two cytokines levels were significantly increased in the hyperoxia-exposed Wt animals, whereas these responses were blunted in the ND mice and even further so in the Tg animals.
Figure 10Inflammatory cytokine levels in BALF after 72 h of hyperoxia exposure. (a) The level of TNF-α measured in the BALF after 72 h of >99% oxygen exposure increased significantly in all hyperoxic (O) groups. In the oxygen-exposed groups, both neutrophil depletion and overexpressing hEC-SOD resulted in significantly lower TNF-α. The levels of TNF-α in the room air–exposed mice were below the limits of detection. (b) The levels of MIP-2 were measured in BALF after 72 h of >99% oxygen exposure and demonstrated a pattern similar to TNF-α. The levels of MIP-2 in the BALF of room air–exposed mice were very low and typically below the limits of detection. Values are expressed as mean ± SEM (n = 6). *P < 0.05 transgenic (Tg) vs. wild-type (Wt), and neutrophil-depleted (ND) vs. controls; **P < 0.01 Tg vs. Wt, and ND vs. controls. MIP-2, macrophage inflammatory protein-2; TNF-α, tumor necrosis factor-α.
ICAM-1 expression. To characterize further the mechanisms that underlie low neutrophil influx in the Tg mice lungs in response to hyperoxia, ICAM-1 mRNA levels were determined by RT-PCR analysis. The amount of GAPDH mRNA transcripts was used as a control for the RT-PCR reactions. The level of ICAM-1 mRNA was significantly different between the two groups after hyperoxia exposure (Figure 11). Hyperoxia induced a twofold increase in ICAM-1 mRNA in the lungs of Wt mice within 84 hours of exposure compared with Tg animals. The level of ICAM-1 mRNA was not significantly different in the two groups of animals at 48 hours of exposure, whereas 72 hours of hyperoxia elicits an increase in ICAM-1 mRNA in the lungs of both groups of animals.
Figure 11Intercellular adhesion molecule-1 (ICAM-1) mRNA levels in the lung. After exposure to 48, 72, and 84 h of >99% hyperoxia, total RNA was extracted from both wild-type (Wt) and transgenic (Tg) mouse lungs and subjected to reverse transcriptase–PCR with primers specific for ICAM-1 and glyceraldehyde phosphate dehydrogenase (GAPDH) as outlined in the text. Aliquots of each PCR reaction were electrophoresed on agarose gels, stained with ethidium bromide, and signal intensity determined. Four animals were used at each time point. The top graph shows the raw agarose gel, and the bottom graph shows the level of ICAM-1 mRNA after normalization to GAPDH mRNA levels. *P < 0.05.
Histopathology. At each time point (48, 72, and 84 hours), the lungs of both Wt and Tg mice were fixed, sectioned, stained, and examined for evidence of lung injury. Using a previously described scoring scheme (27), pathological changes were recorded based on the degree of pulmonary edema, hyaline membrane formation, alveolar septal thickening, and inflammatory cell infiltration. Evidence of vascular injury was detected in the Wt mice as early as after 48 hours of hyperoxic exposure (data not shown). After 72 hours of hyperoxia exposure, the majority of lungs examined from Wt mice showed evidence for decreased pulmonary compliance as indicated by increased edema, thickening of the alveolar septal walls, hyaline membrane formation, vascular and interstitial congestion, hemorrhage, and diffuse inflammatory cell infiltration (Figure 12a), whereas mild histopathologic changes were seen in the Tg mice (Figure 12b). Increased microscopic lung damage was also observed in Tg mice after 84 hours of hyperoxia exposure; however, it was less severe than the injury seen in the Wt animals (data not shown).
Figure 12Light microscopy of paraffin-embedded and hematoxylin and eosin–stained section of representative lungs from mice exposed to hyperoxia. (a) Representative lung sections from Wt mice exposed to 72 h of >99% oxygen showing hyaline membrane formation (open arrows), alveolar septal thickening, and cellular infiltration composed mainly of macrophages and PMNs (filled arrows). (b) Representative lung sections from Tg mice exposed to 72 h of >99% oxygen showing modest alveolar septal thickness and fewer numbers of alveolar macrophages (filled arrows) and PMNs. Original magnification: ×132.
Copyright © 2025 American Society for Clinical Investigation
ISSN: 0021-9738 (print), 1558-8238 (online)