Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Environmental nitrogen dioxide (^·NO₂) is formed by the combustion of fossil fuels and can occur in both indoor and outdoor air. Motor vehicles and emissions from power plants and fossil fuel-burning industries comprise the major producers of ^·NO₂ in outdoor air (1, 2). Ambient levels of ^·NO₂ vary with traffic densities and meteorologic conditions. In the Los Angeles area the 1990 annual arithmetic concentrations ranged from 0.02 to 0.056 parts per million (ppm) (1, 2). ^·NO₂ concentrations in indoor air often exceed those present in outdoor air. Gas cooking stoves and kerosene heaters represent major sources of indoor ^·NO₂ emissions, and concentrations as high as 0.5 to 0.6 ppm can occur in the vicinity of the source, with peak concentrations exceeding 2 ppm (1).

In addition to being an air pollutant, ^·NO₂ can also be formed in lung during inflammation. For example, neutrophils and eosinophils contain myeloperoxidase or eosinophil peroxidase that can catalyze a reaction requiring the substrates, hydrogen peroxide (H₂O₂) and nitrite, to produce ^·NO₂ (3). Further, decomposition of peroxynitrite (ONOO) can also give rise to the formation of ^·NO₂ (3). The reaction of ^·NO₂ with a tyrosyl radical is manifested by the accumulation of the stable end product 3-nitrotyrosine, which is present in a variety of inflammatory diseases (for review, see Ref. 3).

Due to its high reactivity, ^·NO₂ has the potential to adversely affect airways and aggravate pre-existing lung disease. The epithelial cell of the lung is the primary target of inhaled ^·NO₂. Consequently, in animals, acute exposure to concentrations of ^·NO₂ ranging from 5 to 20 ppm causes morphologic changes characterized by shedding of the epithelium into the airways. In addition, type II epithelial cell proliferation, inflammation, and pulmonary edema are found after exposure to ^·NO₂ (7, 8). In patients with asthma, the inhalation of ^·NO₂ enhances airway responsiveness after an allergen challenge (9, 10). Similarly, ONOO has also been shown to cause epithelial injury and airway hyperreactivity in guinea pigs (11), illustrating that both nitrating agents have the potential to aggravate allergic airway disease. A recent study of California schoolchildren with asthma revealed that among a number of air pollutants measured, ^·NO₂ concentrations are the strongest predictor of lower respiratory tract complaints (12). This study further illustrates that ^·NO₂ may be an important risk factor in the aggravation of asthma in children (1). However, these studies evaluate environmental exposure to airborne ^·NO₂, and do not consider endogenously produced ^·NO₂, which may also contribute to pathologic or clinical features associated with inflammatory diseases.

The underlying mechanisms by which ^·NO₂ damages pulmonary epithelium and how these interactions contribute to disease are not well understood. One mechanism by which ^·NO₂ may be involved in disease is through the chemical modification of cellular proteins, including the formation of 3-nitrotyrosine. Although tyrosine nitration can adversely affect the function of proteins, such as manganese superoxide dismutase (MnSOD) (13) and surfactant protein A (14), the exact role of 3-nitrotyrosine-containing proteins present during inflammation in the lung remains to be determined (3). Alternatively, ^·NO₂-induced lipid peroxidation may be responsible for the tissue damage associated with this free-radical gas (2).

The goal of the present study is to establish an exposure system to study ^·NO₂-induced molecular responses in cultured pulmonary epithelial cells. We demonstrate herein that acute exposure of type II epithelial cells to ^·NO₂ causes 3-nitrotyrosine formation and cell death. Death occurs in a cell density-dependent fashion and is extensive in log-phase cells or cultures that are wounded before exposure. In contrast, confluent cells are relatively resistant to ^·NO₂-induced injury. Pure ONOO, or the ONOO generator 3-morpholinosydnonimine (SIN-1), also causes death in a density-dependent fashion. Interestingly, death associated with nitrating oxidants is decreased when cells are pretreated with cycloheximide (CHX) or actinomycin D (Act D), suggesting that messenger RNA (mRNA) synthesis and new protein expression are required for cell death. Lastly, the ability of ^·NO₂ to induce death is dependent on its reactivity in close proximity to the cells or the production of an unstable intermediate, inasmuch as ^·NO₂-oxidized culture medium fails to elicit injury. Our findings suggest that actively dividing or migrating cells are uniquely susceptible to the damaging effects of ^·NO₂, which could have important ramifications for wound repair.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

A line of spontaneously transformed rat lung alveolar type II epithelial (RLE) cells (15) was kindly provided by Dr. Kevin Driscoll (Proctor & Gamble, Cincinnati, OH) and propagated in Dulbecco's modified Eagle's medium/F12 medium containing penicillin/ streptomycin, supplemented with 7% newborn bovine serum (NBS) (GIBCO BRL, Grand Island, NY). To ensure log-phase growth or confluence at the time of exposure to ^·NO₂, RLE cells were plated at known densities before exposure. At 1 h before exposure, growth medium was switched to medium containing 0.5% NBS. In selected experiments, the murine alveolar type II epithelial cell line (C10) (16) was used to confirm cell death. C10 cells were propagated in CMRL1066 supplemented with L-glutamine, penicillin/ streptomycin, and 10% fetal bovine serum (GIBCO BRL). To model epithelial wounding in vitro, cells were grown to confluence on glass coverslips and a wound was created by scraping a section of the glass coverslip with a rubber policeman. Cells were then allowed to repopulate the wounded area for 16 h before exposure to ^·NO₂. CHX (50 µg/ml; Sigma, St. Louis, MO) or Act D (1 µg/ml; Sigma) were used in selected experiments to inhibit protein and mRNA synthesis, respectively. Tetramethylammonium ONOO was provided by Dr. Wim Koppenol (ETCH, Zurich, Switzerland) and the concentration was determined using its molar extinction coefficient ( 302 = 1,670 M¹ cm¹). SIN-1 was obtained from Calbiochem (La Jolla, CA) and used at a concentration of 1 mM.

Gas-Phase Exposure to ^·NO₂

Exposure to ^·NO₂ was performed inside a sealed, stainless-steel cell culture cabinet modified to provide a single-pass flow-through exposure to controlled levels of ^·NO₂. ^·NO₂ was diluted from a tank containing 1,000 ppm ^·NO₂ in nitrogen (Messer MG Industries, Morrisville, PA). The flow rate of ^·NO₂ was metered through a two-staged regulator equipped with a nonreactive rotameter. This NO₂-nitrogen stream was diluted with humidified air generated by a compressor. The diluted air was then metered through a high flow rate (approximately 1 cubic foot/min [cfm]) rotameter and passed through an activated carbon filter followed by a bubbler filled with sterile distilled water. The bubbler was maintained within a water bath at approximately 52 to 62°C to obtain sufficient water vapor pressure to assure > 95% relative humidity within the cell-culture exposure chamber to prevent the evaporation of cell culture medium. The combined ^·NO₂ flow stream and dilution air were mixed together within a heated mixing "T" to assure a homogenous mixture without moisture precipitation before introduction directly into the exposure chamber. A Sievers nitric oxide (^·NO) analyzer equipped with a ^·NO₂ thermal converter was used to measure ^·NO₂ in the gas phase according to manufacturer's instructions (Sievers Instruments, Boulder, CO). A sample from the incubator was analyzed every 15 s and allowed continuous monitoring of gas-phase ^·NO₂ throughout the exposure period. Before each experiment, the instrument was calibrated with ^·NO calibration gas (47.6 ppm; Messer MG Industries). The exhaust from the exposure chamber was passed through a stainless-steel exhaust duct to release the gases through a negative pressure line to the top of the building. During actual exposure, approximately 25 to 30 liters per minute (1 cfm) were passed through the exposure chamber at a concentration of 5 ppm. This range of flow rate was essential to maintain a reasonable chamber response time (typically 15 to 20 min rise time to achieve steady-state conditions). Because ^·NO may occur as a contaminant of ^·NO₂ gas, its concentration was measured during the experiments by bypassing the ^·NO₂ converter and found to be less than 50 parts per billion (< 0.005%). Exposure of RLE cells to ^·NO₂ was performed on orbital rotating platforms (16 rpm; Boekel, Feasterville, PA), which lifted the cells out of their culture medium and allowed exposure to gas-phase ^·NO₂ in 50% of the culture at any given time. This partial immersion protocol was chosen because it avoids drying of cultures and allows continuous exposures for up to 4 h. Rocking cells on an orbital rotating platform under identical conditions in the absence of ^·NO₂ was used to provide air-exposed controls.

Assessment of Reaction of ^·NO₂ with Lung Epithelial Cells

To evaluate the extent that the exposure protocol resulted in ^·NO₂ reactivity with the cell culture medium, media samples were assessed for the formation of nitrite using the Griess assay (17).

Detection of Nitrotyrosine Residues by Confocal Microscopy

To evaluate the reactivity of ^·NO₂ with the cells, we measured the formation of 3-nitrotyrosine residues as a marker of exposure to a nitrating agent (3, 18). In brief, cells were fixed in 3% paraformaldehyde, permeabilized with 0.1% Triton-X100, blocked in phosphate-buffered saline (PBS) containing 1% bovine serum albumin (PBS/BSA), and incubated for 1 h with a rabbit polyclonal anti-3-nitrotyrosine antibody (Upstate Biotechnology, Lake Placid, NY) (2 µg/ml PBS). After 3 washes with PBS/BSA, cells contained on glass coverslips were incubated with a 488-Oregon Green- conjugated secondary antibody (Molecular Probes, Eugene, OR) and then analyzed by confocal scanning laser microscopy (Bio-Rad, Hercules, CA). To assess the staining specificity, the nitrotyrosine antibody was reacted with a 10-fold weight excess of free 3-nitrotyrosine for 16 h at 4°C before addition to the coverslips. Alternatively, 3-nitrotyrosine was converted to aminotyrosine using sodium hydrosulfite, as described elsewhere (19).

Assessment of Oxidative Stress

Immediately after a 4-h exposure of RLE cells to 5 ppm ^·NO₂, cells were incubated with 10 µM carboxydichlorodihydrofluorescein diacetate (dichlorofluorescein diacetate [DCF]; Molecular Probes) for 30 min. Subsequently, cells were trypsinized briefly, pelleted by centrifugation at 1,200 rpm for 10 min, and resuspended in Hanks' balanced salt solution. Approximately 10,000 cells were immediately analyzed by flow cytometry to determine DCF oxidation (20).

Assessment of Cell Death

Cell death was assessed using a number of different techniques that provide an indication of apoptosis. Flow cytometric analysis was performed on adherent cells using propidium iodide (PI) to evaluate the DNA content using a laser scanning cytometer (CompuCyte, Cambridge, MA). Apoptosis was defined as the fraction of cells with a sub-G₀/G₁ DNA content (18). In addition, a second technique was used which employed an antibody that specifically recognizes single-stranded DNA (ssDNA) (APOPTAG) as per manufacturer's recommendations (Alexis, San Diego, CA). Lastly, the 3'-hydroxy ends of nicked DNA were enzymatically labeled with digoxygenin-uridine triphosphate using terminal transferase (TUNEL) (Boehringer, Indianapolis, IN). In both in situ techniques, nuclei were counterstained with 15 µg/ml of PI for 30 min before mounting coverslips on glass slides using vectashield medium (Vector, Burlingame, CA). The percentage of labeled cells was determined by laser scanning cytometry (LSC) of five random fields with an average of 2,000 cells counted per coverslip. Cells with a maximum green pixel intensity of > 400,000 units were used to identify either APOPTAG- or TUNEL-positive nuclei and then expressed as a percent of the total nuclei. Apoptosis assessed by LSC was confirmed using classical flow cytometry to assess the number of cells with a sub-G₀/G₁ DNA content using the DNA stain, PI. Total cell counts were determined using a hemocytometer. Lactate dehydrogenase (LDH) was assessed in the medium as an indicator of cellular injury using a commercial assay kit (Promega, Madison, WI) and results were expressed as percent of maximal release.

Statistical Analysis

Results were analyzed with analysis of variance (ANOVA) using Student-Newman-Keuls procedure to correct for multiple comparisons. Experiments were minimally repeated two times and the data are expressed as means ± standard error of the mean (SEM).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Reactivity of ^·NO₂ with RLE Cells

Figure 1A illustrates the design of the gas-phase ^·NO₂ exposure system. Using this system, cultured cells were exposed to known concentrations of ^·NO₂ for 4 h while rotating on a platform. As is shown in Figure 1B, relatively stable concentrations of either 1 or 5 ppm ^·NO₂ were maintained in the incubator over the course of the experiments. To evaluate whether the partial immersion protocol allowed for reaction of cells with ^·NO₂, or a nitrogen species with similar reactivity, the formation of 3-nitrotyrosine was assessed using immunocytochemistry. After exposure of cultured cells to 5 ppm of ^·NO₂ for 4 h, 3-nitrotyrosine immunoreactivity was considerably increased over air controls (Figure 1C, panels a and b). In control experiments, immunofluorescence was blocked by preabsorption of the primary antibody with a 10-fold weight excess of free nitrotyrosine (Figure 1C, panel c), or by the chemical conversion of 3-nitrotyrosine to aminotyrosine (Figure 1C, panel d), illustrating the specificity of the antibody.

View larger version (23K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   Figure 1.   (A) Schematic diagram of the ·NO2 gas-phase exposure system (see MATERIALS AND METHODS). (B) Representative traces from experiments using either 1 or 5 ppm of ·NO2 during an exposure time of 4 h. (C) Formation of 3-nitrotyrosines in cells exposed to 5 ppm of ·NO2 for 4 h. Immediately after cessation of exposure, cells were removed from the incubator, fixed, and incubated with antibodies to visualize 3-nitrotyrosine immunoreactivity by confocal microscopy. a: sham control; b: ·NO2-exposed RLE cells; c: preabsorption of the 3-nitrotyrosine antibody with a 10-fold weight excess of free nitrotyrosine; d: chemical conversion of 3-nitrotyrosine to aminotyrosine. Panels c and d are staining controls used to verify the specificity of the 3-nitrotyrosine antibody.

Analysis of the nitrite content in the cell culture medium revealed marked increases after exposure to ^·NO₂. An average of 759 ± 75 µM (mean ± SEM of nine experiments) of nitrite was observed in the medium. These findings suggest that the partial immersion protocol limited the direct reaction of ^·NO₂ with cells.

Induction of Cell Death after Exposure to ^·NO₂ or Related Reactive Nitrogen Species

Exposure to oxidants is associated with death in a variety of cell types, including RLE cells (18). Further, epithelial shedding is observed in lungs of animals exposed to high concentrations of ^·NO₂ (2) and may reflect apoptotic cell death. Therefore, we used an in situ technique employing an antibody that recognizes ssDNA (APOPTAG) as an indication of cell death in lung epithelial cells exposed to ^·NO₂. Using this criterion, comparative analyses of log-phase and confluent cultures revealed that APOPTAG-positive occurred preferentially in log phase exposed to ^·NO₂. Figure 2A illustrates that the APOPTAG-positive cells were found in log-phase cultures 44 h after cessation of the 4-h exposure period to 5 ppm ^·NO₂. As shown, there was a relative lack of labeled cells at 44 h in confluent cultures exposed to 5 ppm for 4 h. Note that the nuclear morphology of APOPTAG-positive cells is condensed, consistent with apoptotic cell death. Quantitation of these results in Figure 2B revealed that at the 4- to 72-h time points, virtually 100% of the log-phase cells remaining on the coverslips were APOPTAG-positive, in contrast to confluent cultures, which were relatively resistant to cell death by ^·NO₂. Similar observations were obtained using the TUNEL assay to assess apoptosis or using flow cytometry to assess the fraction of cells with a hypodiploid DNA content (data not shown). As expected, the occurrence of death in log-phase cultures exposed to ^·NO₂ was accompanied by decreases in total numbers of attached plus floating cells compared with air-exposed controls. No differences in total cell counts were detected in confluent cultures exposed to ^·NO₂ (Figure 2C). To further assess cellular injury in log-phase cells exposed to ^·NO₂, we measured LDH release in the medium immediately after cessation of the 4-h exposure period and 4 h thereafter. Results in Figure 2D show that modest increases in LDH release occurred immediately after cessation of exposure, with more striking increases at the later time point. These results demonstrate that the cell membranes were compromised in log-phase cells exposed to ^·NO₂, indicative of cellular necrosis or cellular necrosis secondary to apoptosis.

View larger version (32K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Figure 2.   Assessment of death and total cell numbers in log-phase or confluent cells exposed to 5 ppm of ·NO2 for 4 h. Cells were assessed at 4, 24, 48, or 72 h after initiation of exposure. (A) Visualization of death in cells 48 h after exposure using an antibody that recognizes ssDNA (green). Nuclei were counterstained with PI (red). As indicated, virtually all of the log-phase cells remaining on the dish after exposure to ·NO2 stained positive for APOPTAG. In addition, note the condensed appearance of nuclei in ·NO2-treated cells compared with air controls. Only a small fraction of labeled cells was present in confluent cultures exposed to ·NO2. (B) Quantitation of death assessed by ssDNA technique in log-phase or confluent cultures exposed to 5 ppm of ·NO2 for 4 h. Results are expressed as means ± SEM. *P < 0.05 compared with sham controls, ANOVA. (C) Assessment of total cell numbers 24, 48, or 72 h after initiation of a 4-h exposure to 5 ppm of ·NO2. Results are expressed as means ± SEM. *P < 0.05 compared with sham controls, ANOVA. (D) Assessment of LDH released in the medium of log-phase cells immediately after cessation of a 4-h exposure to 5 ppm of ·NO2 or air, or 4 h thereafter. Results are expressed as means ± SEM. *P < 0.05 compared with sham controls, ANOVA.

ONOO decomposes to ^·NO₂ under physiologic conditions and causes the formation of 3-nitrotyrosine residues (3). To determine whether the cell death observed in log-phase cultures occurred after exposure to agents with similar reactivity to ^·NO₂, RLE cells were exposed to tetramethyl ammonium ONOO or the ONOO generator SIN-1 for the assessment of cell death. As shown in Figure 3A, only low percentages of APOPTAG-labeled cells were found in confluent cultures exposed to ONOO or SIN-1. The lack of cell death in confluent cells exposed to ONOO was previously described by our laboratory (18). Therefore, log-phase cells exposed to ^·NO₂, ONOO, or SIN-1 were uniquely susceptible to death. To determine whether the preferential death of log-phase cells exposed to nitrating oxidants occurred in a different cell type, we exposed C10 cells plated at different densities to SIN-1. Results (Figure 3B) demonstrated that APOPTAG-labeling occurred to the greatest extent in the lowest cell density compared with higher densities, which were resistant. Collectively, these findings illustrate that ^·NO₂ or related nitrating oxidants caused death in lung epithelial cells in a density-dependent manner.

View larger version (26K): [in this window] [in a new window]   Figure 3.   (A) Selective induction of death by ONOO or the ONOO generator SIN-1 in log-phase RLE cells. Log-phase or confluent RLE cells were exposed to 1 mM ONOO or SIN-1 and the percentage of TUNEL-positive cells was assessed by counting 500 cells. Results are expressed as means ± SEM. *P < 0.05 compared with sham controls, ANOVA. (B) Death by RNS occurs in a density-dependent manner. Death was evaluated using ssDNA technique in C10 cells after exposure to 1 mM SIN-1 for 16 h. For SIN-1-treated cells, a statistically significant inverse trend for apoptosis in relationship to cell density was demonstrated by one-way ANOVA (P < 0.001).

The preferential induction of apoptosis by nitrating oxidants in log-phase cells may be due to differences in oxidative stress by nitrating agents. Oxidation of DCF was used as a general indicator of oxidative stress in cells, because DCF can be oxidized by a number of oxidants, including H₂O₂ or ONOO (21). Immediately after cessation of the 4-h exposure period, DCF oxidation (Figure 4) was increased in ^·NO₂-treated cells compared with air-exposed sham controls. Importantly, increased oxidation of DCF was found in both log-phase and confluent cells exposed to ^·NO₂, suggesting that the induction of death in log-phase cells exposed to nitrating oxidants was not due to enhanced oxidative stress per se.

View larger version (10K): [in this window] [in a new window]   Figure 4.   Increased sensitivity of log-phase cells to death is not due to differences in oxidative stress. DCF oxidation was measured in RLE cells after exposure to 5 ppm ·NO2 for 4 h. No statistically significant differences were found between ·NO2-treated log-phase or confluent cells. Results are expressed as means ± SEM. *P < 0.05 compared with sham controls, ANOVA.

Because apoptosis of dividing cells by nitrating oxidants may interfere with epithelial regeneration after injury, we modeled this in vitro by growing cells to confluence on glass coverslips, creating a wound, and subsequently allowing the cells to repopulate for 16 h. Cell cultures were then treated with ^·NO₂ or SIN-1 and cell death was assessed using the TUNEL assay. As is illustrated in Figure 5A, TUNEL reactivity was apparent after ^·NO₂ or SIN-1 exposure, and occurred preferentially in the cells located along the leading edge of the wound. Importantly, 3-nitrotyrosine reactivity was detected in ^·NO₂-exposed cells and occurred both in the confluent area and along the edge of the wound. However, as seen in Figure 5B, panel a, selected cells in the wounded area displayed higher 3-nitrotyrosine immunoreactivity compared with cells in the confluent region of the coverslip, suggesting that migrating or dividing cells may have been unique targets for nitration. SIN-1 caused tyrosine nitration in RLE cells which was prevented by the simultaneous incubation with SOD because this prevented the formation of ONOO and generated H₂O₂ and ^·NO (18). We next assessed whether the addition of SOD altered the TUNEL reactivity associated with SIN-1. As shown in Figure 6, TUNEL reactivity, which was limited to cells in the wounded area after exposure to SIN-1, was abolished in presence of SOD. In contrast, cells treated with SIN-1 in presence of SOD were TUNEL- reactive throughout the dish. These results demonstrated that nitrating oxidants preferentially caused death in migrating or dividing cells, whereas H₂O₂ and ^·NO injured cells independently of their growth state (18).

View larger version (42K): [in this window] [in a new window]   View larger version (86K): [in this window] [in a new window]   Figure 5.   (A) ·NO2-induced or SIN-1-induced TUNEL reactivity occurs in cells migrating into a wound. A confluent culture of RLE cells was wounded with a rubber policeman and cells were allowed to repopulate the wound for 16 h before exposure to 5 ppm of ·NO2 for 4 h or 1 mM SIN-1 for 16 h. TUNEL-positive cells are present along the leading edge of the wound, as opposed to confluent areas within the same dish. (B) 3-nitrotyrosine reactivity is present in both the confluent (b) and wounded (a) areas of the coverslip. Arrow indicates the marked 3-nitrotyrosine reactivity of selected cells along the leading edge of the wound.

View larger version (57K): [in this window] [in a new window]   Figure 6.   SOD alters the patterns of TUNEL reactivity observed in cells treated with SIN-1. A wound was created in a confluent culture and cells were allowed to repopulate for 16 h before treatment with 1 mM SIN-1, 1,000 U/ml SOD, or SIN-1 in the presence of SOD for an additional 16 h. Cell death was assessed using the TUNEL assay.

Immediately after cessation of exposure to ^·NO₂, cell death was not present but reached 80 to 100% over the next 4 h (Figure 2B). This delayed response suggested that new gene expression or protein synthesis may be required. To test this, cells were pretreated with CHX to block protein synthesis. Pretreatment with CHX almost completely blocked TUNEL reactivity associated with ^·NO₂ (Figure 7A). Similarly, Act D, an inhibitor of mRNA synthesis, or CHX reduced TUNEL reactivity caused by SIN-1 (Figure 7B). These results suggest that the production of a death-inducing protein or molecule may be responsible for death of cells exposed to reactive nitrogen species (RNS). Importantly, these studies suggest that ^·NO₂-induced death may occur in a regulated fashion, likely involving intracellular signaling pathways.

View larger version (14K): [in this window] [in a new window]   Figure 7.   (A) Inhibition of protein synthesis blocks TUNEL reactivity in RLE cells exposed to ·NO2. Log-phase cultures were pretreated with 50 µg/ml CHX for 1 h and subsequently exposed to 5 ppm of ·NO2 for 4 h. Cells were removed from the incubator and death was assessed immediately and then at 4 and at 20 h using the TUNEL assay, and quantitated by laser scanning cytometry. (B) Act D or CHX reduced TUNEL reactivity by SIN-1. Results are expressed as means ± SEM. *P < 0.05 compared with controls and P < 0.05 compared with SIN-1-treated, ANOVA.

^·NO₂ is highly reactive with proteins and lipids, and lipid peroxidation is thought to be a major factor in the ^·NO₂-induced toxic effects (22). Medium containing serum also contains lipids and proteins that may be oxidized or nitrated. Because the nitrite levels from the media of ^·NO₂-exposed cells reached approximately 750 µM, it is possible that the oxidation products generated may have contributed to the observed cellular injury. Therefore, medium alone was exposed to 5 ppm of ^·NO₂ for 4 h, transferred to log-phase cells, and subsequently incubated for an additional 24 h. Results shown in Figure 8 demonstrated that ^·NO₂-exposed medium did not induce death in log-phase cells. In addition, exposing cells to ^·NO₂ in the presence of overlying medium by not rotating the dishes prevented tyrosine nitration and cell death (data not shown). Collectively, our results indicate that the interaction of ^·NO₂ in close proximity to the cells and/or the production of a labile, short-lived intermediate were necessary for causing cell death.

View larger version (11K): [in this window] [in a new window]   Figure 8.   Stable oxidation of medium components by ·NO2 is not responsible for ·NO2-induced death in log-phase cells. Cell culture medium, in absence of cells, was exposed to 5 ppm of ·NO2 for 4 h and transferred to log-phase cultures, and death was assessed at 24 and 48 h using the TUNEL assay. Results are expressed as means ± SEM. *P < 0.05 compared with air controls, ANOVA.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Inhalation of high concentrations (> 5 ppm) of ^·NO₂ is associated with acute epithelial damage (8), whereas the lower concentrations that exist in urban areas or in an indoor environment can aggravate pre-existing lung disease, including asthma (1, 12). Inasmuch as exposure to ^·NO₂ occurs both as a result of inhalation of polluted air and during inflammation (3), this RNS may play a major role in pulmonary damage associated with these events. To study the mechanisms responsible for injury by ^·NO₂, an in vitro gas-phase exposure system using pure ^·NO₂ was developed. An orbital rotating platform was used in the gas-phase exposures to prevent drying of cells and to allow the reaction of ^·NO₂ with the cell cultures. Additionally, the use of an orbital rotating platform allowed the exposure of cell cultures to gas-phase ^·NO₂ for ~ 50% of the time with minimal interference of overlaying medium that can react with available ^·NO₂ and prevent reaction with cells. After exposure to 5 ppm of ^·NO₂ an average of 750 µM of nitrite existed in the medium, indicating that some of the ^·NO₂ reacted with the medium and limited reactivity with the cells. Importantly, the presence of nitrite in the medium indicates that the cells are exposed to lower-than-expected concentrations of ^·NO₂. Although this may be a limitation of the system, the exposure protocol may be analogous to the in vivo situation where epithelial cells are covered in a thin layer of extracellular lining fluid, which can prevent the reaction of ^·NO₂ in the lung (23). Despite this limitation, the evaluation of 3-nitrotyrosine residues by immunofluorescence reveals marked reactivity and illustrates the direct exposure of cells to ^·NO₂ or another related nitrating species.

Death was induced in both rat and mouse alveolar type II epithelial cells after exposure to ^·NO₂ or related nitrating oxidants. After a 4-h exposure to 5 ppm of ^·NO₂, death was observed beginning at 4 h after cessation of the exposure. A previous report demonstrated caspase-3-mediated apoptosis in HL-60 cells exposed to ONOO, illustrating that apoptosis occurs after exposure to nitrating oxidants (24). Further, log-phase cultures or cells at the leading edge of a wound are uniquely sensitive to apoptosis by RNS compared with quiescent or contact-inhibited cultures. Previously, we reported the induction of apoptosis in confluent cultures exposed to H₂O₂ or various donors of ^·NO, but not after exposure to ONOO or SIN-1 (18). In the present study, we did not observe death of confluent cultures after exposure to ONOO or SIN-1, confirming our previous observations. Density-dependent cell death was also found in Madin-Darby canine kidney cells exposed to hyperoxia. In contrast to our findings, hyperoxia caused increases in apoptosis or LDH release in near-confluent cells but not in subconfluent cells (25). These observations indicate that different cell types may respond differently to distinct reactive oxygen or nitrogen species. However, our observations demonstrate that cell culture conditions should be considered when assessing the toxic effects of nitrating oxidants, because they markedly affect the outcome of exposure. Confluent cells are also affected by ^·NO₂, as evidenced by the marked DCF oxidation, which occurs to a similar extent in log-phase or confluent cells. Therefore, oxidative stress per se does not explain the death-inducing effects of ^·NO₂ in log-phase cells. At present, the significance of DCF oxidation is unclear. Inasmuch as many reactive intermediates can oxidize DCF directly or indirectly, including superoxide, H₂O₂, or ONOO (21), the ultimate molecule responsible for oxidation in log-phase or confluent cells remains to be elucidated. This may be particularly important in light of the fact that confluent cells did not die, whereas log-phase cells were uniquely susceptible to ^·NO₂-induced cell death. Knowledge of the "effector" oxidants produced under these circumstances may be important in further understanding the mechanisms of injury.

Given that ^·NO₂ is a highly reactive gas, and only the actively dividing cells are susceptible to death, this suggests that the targets for ^·NO₂-dependent nitration may be distinct in actively dividing or migrating cells, and that these events may control cell death. The observations in Figure 5B (panels a and b) lend support to this hypothesis, where marked tyrosine nitration was observed in select wounded cells exposed to ^·NO₂ whereas less immunoreactivity was apparent in the confluent area. On the basis of these results, we anticipate that during epithelial proliferation or regeneration, processes that occur in various models of injury or asthma (26), exposure to a nitrating species may interfere with the normal restitution. Similarly, children may be uniquely susceptible to the damaging effects of nitrating oxidants because their lungs are still developing (12, 29).

Currently, the mechanisms that confer sensitivity of log-phase cultures to ^·NO₂ are unknown. Dividing cells contain cell-surface receptors, which may be sensitive to oxidative modification by nitrating agents. In this regard, work in our laboratory has demonstrated the elevated expression of the epidermal growth factor receptor in dividing cultures of mesothelial cells (30), a receptor that can be dimerized by ONOO (31). Because cell-surface receptors and cytoskeletal proteins can mediate signaling events, it is possible that oxidative modification of these proteins may occur when exposed to ^·NO₂ and that these events may mediate a death response.

The identity of a death effector produced in ^·NO₂-exposed cells which may be responsible for the triggering of death remains to be identified. However, a number of known inducers of apoptosis exist, including tumor necrosis factor (TNF)- and Fas ligand, that are known to induce apoptosis by binding to the respective receptors (32). The expression of TNF- and Fas ligand are upregulated in the lung in various models of injury (27, 33, 34). These ligands are therefore likely candidates in a model of cell death induced by ^·NO₂. In this regard, DNA-damaging agents have been shown to cause an increase in Fas ligand expression and apoptotic cell death in T lymphocytes (35), indicating that this pathway may be important in oxidant-mediated apoptosis.

We have described an in vitro system that allows the exposure of lung epithelial cells to gas-phase ^·NO₂. Using this system, we demonstrated specific reactivity of ^·NO₂ within the cell, evidenced by the formation of protein nitrotyrosines. The sensitivity of dividing or migrating cells to ^·NO₂, evidenced by cell death, may point to an important role for this nitrating species in the prevention of epithelial restitution after pulmonary injury and interference with lung development.