Introduction

Top Abstract Introduction Materials and methods Results Discussion References

The use of supplemental oxygen continues to be an important therapy in the management of premature infants. Despite its necessary use, oxygen in increased concentrations and after prolonged use may cause lung damage (1). Although the exact mechanisms of toxicity are still not fully understood, increased production of reactive oxygen species (ROS) and low levels of cellular antioxidants provide the most plausible explanation for injury in underdeveloped lungs exposed to high concentrations of oxygen (4- 7). ROS are detoxified by a network of antioxidant systems, but during oxidative stresses the production of ROS may exceed the capacity of these cellular defenses, causing oxidative damage to proteins, lipids, and DNA, which can cause cell and tissue dysfunction and lead to cell death (8).

The glutathione (GSH)-dependent antioxidant enzyme system provides cellular protection against ROS, particularly hydrogen peroxide and certain organic hydroperoxides (such as t-butyl hydroperoxide [t-BuOOH]), by using selenium-dependent and independent peroxidases to reduce hydrogen peroxide or lipid peroxides to water or the respective alcohols, with the concurrent oxidation of GSH to glutathione disulfide (GSSG) (11). The flavoenzyme glutathione reductase (GR) catalyzes the reduction of most of the GSSG formed back to GSH at the expense of nicotinamide adenine dinucleotide phosphate (NADPH) oxidation. The physiological significance of regenerating GSH from GSSG by GR is indicated by studies demonstrating that cells with attenuated GR activities are more susceptible to oxidative stresses (15). Although it does not necessarily follow that increased GR activities would enhance antioxidant defense function significantly, we have observed greater resistance to oxidant stresses in Chinese hamster ovary (CHO) cells with genetically enhanced GR activities, particularly if the mitochondrial GR activities are augmented (19). As ROS have been implicated as pathogenic agents in many disease states, including oxygen toxicity, determining the specific mechanisms governing the protective effect of GR overexpression may be crucial to understanding cellular responses in pulmonary oxygen toxicity and other diseases and toxicities mediated in part by oxidant stresses.

High plasma concentrations of GSSG and arterial-venous differences in plasma GSSG concentrations across the lungs in premature infants indicate measurable and potentially significantly oxidant stresses in these individuals (23). Increased production of ROS, deficient antioxidant activities, or a combination of both probably contribute to cellular release of GSSG by immature lungs, and enhancing concentrations of GR in lung cells might prove to be particularly advantageous in this patient population. The present studies were designed to determine whether adenovirus-mediated gene transfer of leader sequence GR (LGR) could increase GR activities in H441 cells, which are derived from lung Clara cells, and whether such cells would be more resistant to t-BuOOH-induced oxidant stresses. We also investigated the effect of GR expression on cellular GSH/GSSG ratios and GSSG export in response to t-BuOOH as an experimental model of oxidant stresses. We have previously observed protection of CHO cells from oxidant challenge by lipofectamine-mediated transfection and augmentation of GR activities, both in transient transfections and in stably transfected cell lines (21). The goals of the present studies were to determine whether adenoviral vectors could be used to enhance cellular GR activities and to determine whether increased GR activities would provide measurable increases in antioxidant function in a human pulmonary-derived cell line. These two questions are critical steps toward the possible use of these strategies in human therapeutic interventions, such as in prematurely born infants. Cells transfected with LGR had increased mitochondrial and cytoplasmic GR activities and demonstrated greater resistance to t-BuOOH cytotoxicity than did cells transfected with control genes or in control cells not infected with virus. These data support the working hypothesis that augmentation of GR activities in lung-derived cells by adenovirus-mediated transfer of LGR can afford greater resistance to oxidant stresses.

	Materials and Methods

Top Abstract Introduction Materials and methods Results Discussion References

Materials

H441 cells, a transformed human lung cell line with characteristics of Clara cells, were obtained from American Type Culture Collection (Manassas, VA). Media (RPMI 1640) used in all experiments were purchased from GIBCO BRL (Gaithersburg, MD). Reagents for protein assays were obtained from Bio-Rad Laboratories (Richmond, CA). All other reagents were obtained from Sigma Chemical Company (St. Louis, MO).

Vector Construction

Details on the cloning of human GR cDNA and addition of the mitochondrial targeting signal (MTS) from the human manganese superoxide dismutase (MnSOD) gene have been described previously (19). Using the GR cDNA clone with a functional MTS, we constructed an adenoviral vector from the replication-defective human adenovirus type 5 (Ad5). Briefly, human GR cDNA containing a functional MTS from the human MnSOD gene (LGR) was inserted into the E1 region of the shuttle plasmid pE1sp1B (Microbix Biosystems, Inc., Toronto, Canada). The transformed shuttle plasmid DNA and the circular Ad5 genome plasmid pBHGE3 (Microbix Biosystems, Inc.) were cotransfected into Ad5-transformed human (293) cells to generate recombinant adenovirus AdLGR. Viral propagation, purification, and titration were carried out according to standard methods (24). To provide experimental controls, we inserted either MnSOD in the human superoxide dismutase antisense orientation (DOS), or ₁-antitrypsin (₁-AT) into identically constructed adenoviral vectors.

Cell Transfections and Exposure to t-BuOOH

H441 cells (1 × 10⁶) were plated in 100-mm tissue culture dishes and incubated in medium with 5% fetal bovine serum (FBS) at 37°C. After 72 h, the cells were counted (average of 3 × 10⁶ cells per plate), the media were changed, and the cells were exposed to adenoviral vectors containing either LGR or DOS expression cassettes at a calculated multiplicity of infection (MOI). Twenty-four hours after transfection, t-BuOOH was added to the incubation media. Medium samples were obtained at 0, 5, 12, 15, 17, and 20 h for measurement of lactate dehydrogenase (LDH) activities. At the time of harvest (20 h after t-BuOOH), the cell monolayer was washed with ice-cold phosphate-buffered saline (PBS; GIBCO BRL), and the cellular material collected by scraping in 700 µl 0.2 M KPO₄; 1 mM ethylenediaminetetraacetic acid (EDTA), pH 8.0; and 0.2% Triton X-100. After collection, the cell samples were incubated on ice for 1 h and centrifuged for 5 min in a microcentrifuge. LDH and GR activities and protein concentrations were measured from the resulting supernatants. Subcellular fractionations for measurements of mitochondrial and cytosolic GR activities were obtained by differential centrifugation.

For cellular GSH and GSSG measurements, cells were harvested 1 h after t-BuOOH exposure. Cells were collected by scraping in 0.2 M phosphate buffer and 1 mM EDTA, pH 7.5, and divided into two microcentrifuge tubes. Ten microliters of 3 M N-ethylmaleimide (NEM) were added immediately to one tube (for GSSG assay). Both samples were incubated with 0.2% Triton X-100 on ice for 1 h and centrifuged for 5 min, and the supernatants were assayed.

GR Assay

Fifty-seven microliters of supernatant were mixed with 193 µl of assay mixture (0.1 M Tris-HCl, pH 8.0; 7 mM GSSG; and 0.225 mM NADPH) in one well of a 96-well microtiter plate, and the rate of consumption of NADPH was measured as the change in optical density per minute at 340 nm. In this reaction 1 U of GR is defined as the oxidation of 1 µmol of NADPH per minute at 25°C (25).

LDH Assay

Fifteen microliters of cellular supernatants, appropriately diluted with 1× PBS to measure the activity within the standard curve, or 50 µl of media were mixed with 200 µl of a reaction mixture (216 µM NADPH in 60 mM K-phosphate; 0.72 mM pyruvate, pH 7.5). LDH activities were determined by the rates of NADPH oxidation measured as the changes in optical density at 340 nm (25).

GSH + GSSG Assay

Total glutathione (tGSH = GSH + 2GSSG) and GSSG concentrations were determined by an adaptation of the method described by Tietze (26). For tGSH determination, 30 µl of cellular supernatants, appropriately diluted with 1× PBS to measure the activity within the standard curve, or 200 µl of media were mixed with 50 µl of reaction mixture 1 (18.75 µl of 0.1 M phosphate buffer; 5 mM EDTA, pH 7.5; 25 µl of 10 mM 5,5'-dithio-bis[2-nitrobenzoic acid]; and 6.25 µl of GR [0.5 U/0.1 ml]) and 50 µl of reaction mixture 2 (37.5 µl of 0.1 M phosphate buffer; 5 mM EDTA, pH 7.5; 12.5 µl of 0.225 mM NADPH). The tGSH concentrations were determined from measurements of rates of change in absorbance at 410 nm and calculated from experimentally derived standard curves. Similarly, for GSSG determinations 30 µl of Sep-Pak (Waters Corp., Milford, MA) eluent of cellular supernatants were mixed with 50 µl of reaction mixture 1 and 50 µl of reaction mixture 2, and GSSG concentrations were determined from measurements of rates of change in absorbance at 410 nm and calculated from experimentally derived standard curves. The GSH and GSSG concentrations are expressed per milligram of protein.

Protein Assay

The protein concentrations of the supernatants were measured using a modification of the Bradford assay (27). Twenty-five microliters of cellular supernatant, diluted with distilled water as needed, were mixed with 200 µl of dye (Bio-Rad reagent diluted 1:5) and the absorbances at 600 nm were measured with a microtiter plate reader (Dynatech Laboratories, Chantilly, VA). The protein concentrations were calculated using experimentally derived standard curves constructed with bovine serum albumin.

Statistics

All data are expressed as means ± SEM. Data were analyzed with one- or two-way analysis of variance (ANOVA), as indicated. When the two-way ANOVA determined a statistically significant difference or an interaction, a subsequent one-way ANOVA was carried out with appropriate post hoc testing. All analyses were performed using SPSS for Windows (SPSS version 7.0, Chicago, IL). Statistical significance was attributed to P  0.05.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

Increasing doses of at least 5 MOI resulted in dose-dependent increases in GR activities through 50 MOI but no further increases in GR activities were observed with 100 MOI (Figure 1). Total LDH activities per cell plate were diminished by exposure to virus at doses of 20 MOI or greater, suggesting toxic effects of virus manifested by inhibition of cell growth above this level of viral exposure (Figure 2). On the basis of these initial results, a viral dose of 10 MOI was used in studies of t-BuOOH exposure.

View larger version (47K): [in this window] [in a new window]   Figure 1.   Effect of viral dose (MOI) on enhancement of cellular GR activities. H441 cells were transfected with increasing doses of adenovirus containing LGR and were incubated in medium with 5% FBS at 37°C for 24 h. Cells were harvested and GR activities were measured. Data are means ± SEM, n = 3 per group. Data were analyzed by one-way ANOVA followed by the Student-Newman-Keuls test. There was a dose-dependent effect of viral MOI on cellular GR activities up to 50 MOI of virus. Bars not accompanied by the same letter are significantly different (P < 0.01).

View larger version (81K): [in this window] [in a new window]   Figure 2.   Total LDH activities per cell plate in cells transfected with increasing doses of adenovirus containing LGR. H441 cells were transfected with increasing doses of adenovirus containing LGR and incubated in medium with 5% FBS at 37°C for 24 h. Cells were harvested and LDH activities were measured. Data are means ± SEM, n = 3 per group. Data were analyzed by one-way ANOVA followed by the Student-Newman-Keuls test. Total LDH activities per cell plate were decreased by exposure to doses of at least 20 MOI of virus; * for 0, 1, 5, and 10 MOI versus 20, 50, and 100 MOI (P < 0.05).

In H441 cells transfected with adenovirus containing LGR at 10 MOI, total cellular GR activities were approximately 11-fold higher, cytosolic GR activities were approximately 7-fold higher, and mitochondrial GR activities were approximately 10-fold higher than in nontransfected control cells or in cells transfected with adenovirus containing DOS (Figure 3). Transfections of cells with vectors containing the cDNA for human ₁-AT similarly gave no changes in cellular or compartmental GR activities (data not shown).

View larger version (34K): [in this window] [in a new window]   Figure 3.   Cellular and compartmental GR activities in H441 cells after transfection with adenovirus containing LGR or DOS. Control (CON) cells were not exposed to virus. Transfected and control cells (3 × 106 per plate) were incubated in medium with 5% FBS at 37°C for 24 h. Cells were harvested and GR activities were measured. Cell homogenates were fractionated by differential centrifugation to provide mitochondrial and cytosolic preparations. Data are means ± SEM, n = 4 per group. Data were analyzed by one-way ANOVA followed by the Student-Newman- Keuls test. Cells transfected with LGR showed significantly greater GR activities (*P < 0.001) than did cells exposed to virus containing DOS or nontransfected control cells. *P < 0.001 LGR versus DOS and CON.

Microscopic examination of the cells 20 h after exposure to t-BuOOH revealed some degree of rounding of the cells exposed to 400 µM t-BuOOH, but no obvious differences between cells exposed to 0 or 300 µM t-BuOOH were observed. Although a dose-dependent effect of t-BuOOH on total LDH (LDH in the cells plus that in the media) activities was observed, at each dose of t-BuOOH there were no differences in total LDH activities among the three cell lines. Dose- and time-dependent increases in media LDH activities were observed in all cell groups exposed to t-BuOOH (Figure 4). Media LDH activities began to increase between 5 and 12 h and increased through 20 h in all three cell types. However, at all time points greater than 5 h, the media LDH activities in cell plates exposed to 400 µM t-BuOOH were lower in cells transfected with LGR than in cells transfected with DOS or in nontransfected control cells. The media LDH activities from cells treated with 0 or 300 µM t-BuOOH were not different. Cells transfected with LGR showed lower percentages of LDH release after exposure to 400 µM of t-BuOOH than did cells transfected with DOS or nontransfected control cells (Figure 5). No differences in percentages of LDH release were observed either within or between cell groups exposed to 0 or 300 µM of t-BuOOH. Cells exposed to 0 or 300 µM of t-BuOOH showed no differences either in the percentages of LDH release or in the media LDH activities over time, suggesting that the toxic effects of 300 µM of t-BuOOH were negligible in this study. However, the data from cells exposed to 400 µM of t-BuOOH indicate that cells transfected with LGR were more resistant to t-BuOOH-induced oxidant stresses than were the other cells studied as experimental controls for GR enhancement and for viral exposure.

View larger version (14K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   Figure 4.   LDH activities in the incubation media after exposure of cells to 0 (A), 300 (B), or 400 (C) µM t-BuOOH. Transfected (LGR and DOS) and control (CON) cells (3 × 106 per plate) were incubated in medium with 5% FBS at 37°C. After 24 h, cells were exposed to 0, 300, or 400 µM t-BuOOH and incubated for an additional 20 h. At 0, 5, 12, 15, 17, and 20 h, medium samples were assayed for LDH activities. Data are means ± SEM, n = 4. Data were analyzed by two-way ANOVA with a subsequent one-way ANOVA, followed by post hoc testing to determine differences at each dose of t-BuOOH during specific time points. Cells transfected with LGR released less LDH activity to the incubation medium when exposed to 400 µM of t-BuOOH than did DOS or control cells at 12 h, 15 h, 17 h, and 20 h. *P < 0.001 at 12 to 20 h.

View larger version (33K): [in this window] [in a new window]   Figure 5.   Fractional LDH release by H441 cells exposed to t-BuOOH. Transfected (LGR and DOS) and control (CON) cells (3 × 106 per plate) were incubated in medium with 5% FBS at 37°C. After 24 h, cells were exposed to 0, 300, or 400 µM t-BuOOH and incubated for an additional 20 h. Cells were harvested and LDH activities were measured. Percentages of LDH release were calculated as total medium LDH/(total medium LDH + total cell LDH). Data are means ± SEM, n = 4. Data were analyzed by two-way ANOVA with a subsequent one-way ANOVA, followed by appropriate post hoc testing to determine differences at each dose of t-BuOOH. No differences in percentages of LDH release were observed between groups exposed to 0 or 300 µM t-BuOOH. At 400 µM of t-BuOOH the percentages of LDH release by the LGR cells were lower than those by the CON or DOS cells. *P < 0.001.

Figure 6 shows the GSH/GSSG ratios in transfected and control cells exposed to t-BuOOH. Microscopic examination of the cells 1 h after addition of t-BuOOH to the incubation media revealed no obvious visual differences between cells exposed to 0 or 400 µM t-BuOOH. Interestingly, both groups transfected with adenovirus had lower resting GSH/GSSG ratios than did nontransfected control cells. All had similar cellular GSH concentrations, but both viral transfected groups had higher cellular GSSG concentrations (data not shown). However, cells transfected with LGR and exposed to t-BuOOH retained higher GSH/GSSG ratios than did similarly exposed cells transfected with DOS or nontransfected control cells. A combination of lower GSSG and greater GSH concentrations in cells transfected with LGR accounted for this difference (data not shown).

View larger version (47K): [in this window] [in a new window]   Figure 6.   GSH/GSSG ratios in transfected and control cells exposed to t-BuOOH. Transfected (LGR and DOS) and nontransfected control (CON) cells (3 × 106 per plate) were incubated in medium with 5% FBS at 37°C. After 24 h, cells were exposed to 0 or 400 µM of t-BuOOH and incubated for 1 h. Cells were harvested and GSH/GSSG ratios were measured. Data are means ± SEM, n = 4. Data were analyzed by two-way ANOVA with a subsequent one-way ANOVA, followed by post hoc testing to determine differences at each dose of t-BuOOH. At 0 t-BuOOH, both cell groups transfected with the virus (LGR and DOS) had lower resting GSH/GSSG ratios than did control cells not exposed to the virus (**P < 0.05). Following exposure to 400 µM of t-BuOOH, cells transfected with LGR retained higher GSH/ GSSG ratios than comparably treated DOS or nontransfected control cells (*P < 0.05).

The addition of t-BuOOH to the incubation media induced increases in medium concentrations of GSH + GSSG, probably reflecting increased cellular efflux of GSSG (Figure 7). Cells transfected with LGR and exposed to t-BuOOH had lower medium GSH + GSSG concentrations than did cells transfected with DOS or nontransfected control cells. In cells not exposed to t-BuOOH the medium [GSH + GSSG] concentrations also were lower in cells transfected with LGR.

View larger version (44K): [in this window] [in a new window]   Figure 7.   Medium GSH + GSSG in transfected cells exposed to t-BuOOH. Transfected (LGR and DOS) and nontransfected control (CON) cells (3 × 106 per plate) were incubated in medium with 5% FBS at 37°C. After 24 h, cells were exposed to 0 or 400 µM of t-BuOOH and then incubated for 1 h. GSH + GSSG concentrations were measured in medium samples. Data are means ± SEM, n = 4. Data were analyzed by two-way ANOVA with a subsequent one-way ANOVA, followed by post hoc testing to determine differences at each dose of t-BuOOH. The medium GSH + GSSG concentrations were lower at 0 µM of t-BuOOH in cells transfected with LGR than in DOS or nontransfected control cells (*P < 0.001). At 400 µM of t-BuOOH, the medium GSH + GSSG concentrations were lower in cells transfected with LGR than in DOS or nontransfected control cells (**P < 0.001).

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

Adenovirus-mediated gene transfer of LGR increased cellular GR activities in H441 cells in a dose-dependent manner through 50 MOI, with no further increase at higher doses. Lower total LDH activities per cell plate in cells exposed to 20 MOI or greater (Figure 2) indicate an inhibition of cell growth at these doses. On the basis of these results, a viral dose of 10 MOI was chosen for use in studies of t-BuOOH exposure. Even though 10 MOI did not alter cell growth, this dose of virus did cause lower cell GSH/ GSSG ratios (Figure 6), independent of t-BuOOH exposure, suggesting virus-dependent effects on cell metabolism. Nevertheless, H441 cells transfected with LGR were more resistant to t-BuOOH-induced LDH release than were nontransfected cells or cells transfected with a control gene that did not increase cellular GR activities. These data support the working hypothesis that augmentation of GR activities by adenovirus-mediated gene transfer of LGR can afford greater resistance to oxidant stresses.

Two complementary, but separable, basic mechanisms can be envisioned for the cytoprotective effects of increased GR activities. Insufficient GR capacity could lead to depletion of cellular GSH availability to levels that compromise reduction of t-BuOOH, and cellular damage might be mediated by GSSG-independent processes via Fenton-type chemistry, such as direct oxidation of critical biomolecules, leading to lipid peroxidation, oxidation of proteins to products other than disulfides (protein carbonyls and similar products), or oxidation of nucleic acids and subsequent cell death (28).

Alternatively, GSSG accumulation and decreases in GSH/GSSG ratios, secondary to increased formation of GSSG relative to cellular capacity to reduce GSSG back to GSH, or to export GSSG and synthesize GSH in replacement, could lead to increased S-thiolation of protein thiols (33, 34). If loss of specific protein thiols leads to altered function in a protein critical for cell viability, this form of protein oxidation could lead to cell injury and cell death. In the present study, we found that 1 h after exposure to t-BuOOH, cellular GSH/GSSG ratios decreased, primarily through increasing cellular GSSG contents, and medium GSH + GSSG levels were elevated, presumably through increased cellular efflux of GSSG. In association with a protective effect on t-BuOOH-mediated cell death, cells transfected with LGR and exposed to t-BuOOH retained higher GSH/GSSG ratios and exhibited lower cellular export of GSH + GSSG into the cell media than did nontransfected cells or cells transfected with a control gene and similarly exposed to t-BuOOH.

GSSG is reduced to GSH by GR at the expense of NADPH oxidation. Cellular supplies of NADPH could become limiting in the reduction of GSSG in some circumstances, but with the present studies we observed increased cellular GR activities and greater resistance to t-BuOOH cytotoxicity in the LGR cells that is best explained as resulting from the enhanced ability of cells transfected with LGR to reduce GSSG during the peroxide challenge. Increases in NADPH production might further enhance the efficacy of the GR transfection strategies, but our data clearly demonstrate that significant cellular antioxidant protection can be attained by enhanced GR activities alone. Although S-thiolations of certain proteins under conditions of oxidant stresses have been reported (35, 36), the physiological relevance and consequences of the observed alterations have not been established. In addition, studies of pathologically relevant oxidant stresses in vivo, such as hyperoxic lung injury (28) or diquat-induced hepatic necrosis (37), have not indicated significant effects of these oxidant stresses on protein S-thiolation. Cellular compartmentalization and molecular specificity (38) might obscure detection of otherwise important alterations, and increasingly specific methods of analysis need to be applied to testing equally specific working hypotheses.

The hypothesis that protein thiol/disulfide shifts, especially in the mitochondria, can contribute to pathologically significant processes is supported strongly by the numerous recent studies on the mitochondrial permeability transition pore (39, 40). Petronilli and colleagues (41) have shown that this contains a critical thiol/disulfide pair that modulates the facility with which this pore opens, allowing free permeability of substrates less than 1500 D molecular weight. This sudden increase in permeability can lead to collapse of critical concentration gradients and has been associated with a rapid onset of cell death (42, 43).

Although many cells can supplement the capacity of the GR system for maintaining high GSH/GSSG ratios by active export of GSSG (44), mitochondria appear not to have a similar capability (45) and rely more heavily on reduction of GSSG for sustaining this important redox tone (46). In agreement with the evidence for a particularly critical role of the thiol/disulfide status of a specific mitochondrial protein, we have observed much greater enhancement of cellular resistance to oxidant challenge by CHO cells by stable transfection with the LGR cDNA than we observed with the construct lacking the MTS, which produced comparable increases in total cellular GR activities but did not increase mitochondrial GR activities (20). In the present studies of virus-mediated transfection, we did not investigate the effect of similar transfections with the GR cDNA lacking the MTS, but focused on the strategy indicated by the data available as more likely to be effective. We did study the effects of transfection with the DOS and ₁-AT constructs as important controls for the effects of viral infection.

Despite the addition of functional mitochondrial leader sequences to the GR cDNA used for transfection, increased GR activities were observed in both the mitochondrial and cytosolic fractions (Figure 3). Although imperfect separation of the cellular compartments probably contributes to the elevation in cytosolic GR activities, our previous studies indicate that this effect is minimal. The LGR clone that we used has two functional start codons, one at the 5' end of the MTS and the other at the 5' end of the human GR cDNA. Transcription initiated from the second start site would not yield a protein that contains the MTS, which might account for the increase observed in cytosolic GR activities achieved with LGR. In addition, the mitochondrial protein transport system might become limiting in the process of importing the LGR protein containing the MTS. The relative contributions of these two mechanisms, and perhaps of other mechanisms not yet recognized, will require direct experimental investigation.

In conclusion, adenovirus-mediated gene transfer of LGR significantly increased the GR activities in transfected H441 cells, and elevated mitochondrial GR activities were obtained by appending a synthetic MTS 5' to the human GR cDNA. Although viral exposures decreased cell GSH/GSSG ratios without exposure to t-BuOOH, cells transfected with LGR showed significantly greater resistance to t-BuOOH than did cells not exposed to virus or cells exposed similarly to adenovirus containing control vector constructs. Adenoviral vectors are promising candidates for intermittent, lung-targeted gene therapy (49). Because adenoviruses naturally infect pulmonary epithelial cells, high intercellular concentrations can be achieved and the transgenes carried by the viruses can be expressed at high levels (50). In addition, the transgene expression is transient because adenoviruses usually do not integrate their DNA into the genomes of the host cells. Thus, adenovirus-mediated LGR transfection provides a useful tool for the study of oxidant-mediated cell injury and may provide useful approaches to therapies for treatment of oxidant-induced injuries, such as pulmonary oxygen toxicity.