INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitrogen dioxide (NO₂) is a common air pollutant that is produced during various combustion processes (1), including the burning of fossil fuels for heating and power generation and in motor vehicles. High concentrations of NO₂ are encountered in certain industries and in homes with gas stoves, where 24-h averages may reach 0.5 ppm and peak concentrations 1 to 2 ppm (2).

Previous epidemiologic studies and data indicate that environmental exposure to NO₂ increases both the morbidity of pulmonary diseases and susceptibility to airway infections. NO₂ may also cause worsening of obstructive lung diseases (1, 3, 4), and there are indications that asthmatic individuals may be more susceptible to the effects of NO₂ than are healthy subjects (5, 6).

The pattern of bronchial inflammation following single, short-term exposures to occupational and workroom concentrations of NO₂ has previously been characterized. Studies employing bronchoalveolar lavage (BAL) have demonstrated dose-dependent and time-dependent increases in mast cells and lymphocytes, which resolve within 72 h (7, 8). The recent practice of analyzing the first BAL aliquot separately as a bronchial washing (BW) has revealed an increased number of neutrophils in the proximal airways after exposure to NO₂ (9, 10).

Markers of inflammation such as albumin, fibronectin, hyaluronan, angiotensin converting enzyme (ACE), ₂-microglobulin, leukocyte elastase, lactate dehydrogenase (LDH), total protein, leukotriene B₄ (LTB₄), prostaglandin E₂ (PGE₂), thromboxane A₂ (TXA₂), interleukin-8 (IL-8) and tumor necrosis factor- (TNF-) have been assayed and found to be unaffected by a single exposure to NO₂ in normal subjects (1, 10, 12), although exposure of human bronchial epithelial cells to 0.4 to 0.8 ppm NO₂ for 6 h in vitro induces synthesis of proinflammatory cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-8, and TNF- (13). Becker and colleagues have reported an increase in ₁-antitrypsin and in plasminogen activator activity in the bronchial fraction as well as a decrease in certain macrophage functions in BALF obtained 16 h after exposure to 2 ppm NO₂ for 4 h (10). In another study, exposure to 1 ppm NO₂ for 3 h induced a small increase in thromboxane B₂ (TXB₂) in normal subjects, and more pronounced changes in prostaglandins and TXB₂ in asthmatic subjects (14). Recently, we have shown that NO₂ modifies the protective antioxidant defense in the respiratory-tract lining fluid (15).

Information gained from BAL provides an indirect reflection of events in the walls of the bronchi, bronchioles, and alveoli. To obtain direct evidence of such events, mucosal biopsies are required, but to date, only one study has reported the effects of NO₂, and found no significant changes in cell numbers in biopsy specimens following exposure to 1 ppm NO₂ for 3 h (16).

The aim of the present study was to measure the time course of inflammatory changes induced in the airways of normal healthy human subjects by exposure to 2 ppm NO₂ for 2 h. This NO₂ concentration is the exposure limit for an 8 h workshift in most European countries, and a level that is occasionally found in the indoor environment in certain industries, mines, and ferries.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Thirty healthy, nonsmoking volunteers (18 males and 12 females; mean age 25 yr, age range 20 to 30 yr) without any history of asthma or other pulmonary disease participated in the study. All had negative skin prick tests and normal lung function. None had a history of airway infection for at least 6 wk prior to the first NO₂ exposure or during the study. During the study, subjects were not allowed to take any nonsteroidal antiinflammatory drugs (NSAIDs), aspirin, vitamin C (ascorbic acid), or vitamin E (-tocopherol).

Study Design

The subjects were exposed to air or air containing 2 ppm NO₂ in an environmental chamber according to a previously described standard protocol (9, 15). During the exposure, light exercise (75 W) on a bicycle ergometer was alternated with rest in 15-min periods. Total exposure time was 4 h. The two exposures occurred in random order and were separated from one another by at least 3 wk. The exposures were single blind in that the subjects were blinded to the presence or absence of NO₂ whereas the chamber operator knew whether or not NO₂ was used. Flexible fiberoptic bronchoscopy with BW and BAL was performed either 1.5 or 6 h after the air/NO₂ exposure in the two groups of subjects. Biopsy procedures after the first exposure were randomized to either the right or the left side, with the lavages done on the contralateral side. The opposite sides were used during the second bronchoscopy. All analyses were performed on randomly coded samples, enabling totally blind analyses and statistics. Informed consent was obtained from the subjects, and the study was approved by the local Ethics Committee of the University of Umeå.

NO₂ Exposure

The exposures to NO₂ took place at the Medical Division of National Institute for Working Life in Umeå, according to principles used in studies since 1985 (7, 15); the NO₂ exposure technique has been described elsewhere (15).

Bronchoscopies

Fiberoptic bronchoscopy was performed as previously described (15, 17). At each bronchoscopy, three endobronchial mucosal biopsy specimens were taken with fenestrated forceps (FB-21C; Olympus, Tokyo, Japan) either from the anterior aspect of the main carina and the subcarinae of the 3rd and 4th generation airways of the right side, or from the posterior aspect of the main carina and the corresponding subcarinae on the left side. Different locations were used during the two bronchoscopies in order to avoid biopsy artefacts at former biopsy sites. BW was performed with 2 × 20 ml and BAL with 3 × 60 ml sterile phosphate buffered saline (PBS), pH 7.3, at 37° C. The lavage fluid was infused with the tip of the bronchoscope carefully wedged in a lingula-lobe bronchus (when biopsies were taken on the right side) or a middle-lobe bronchus (when biopsies were taken on the left side), and was gently suctioned back into a siliconized container placed in ice water.

The chilled lavage fluid was filtered through a nylon filter (pore diameter 100 µm; Syntab Product AB, Malmö, Sweden) and centrifuged at 400 × g for 15 min. The supernatants were separated from the cell pellets, immediately analyzed for albumin and protein, and stored at 70° C for subsequent cytokine and mediator analyses. The cell pellets were resuspended in PBS to a concentration of 10⁶ cells × ml¹. The total number of cells in the lavage fluid was counted in a Bürker chamber. Cytocentrifuged specimens with 5 × 10⁴ nonepithelial cells per slide were prepared using a Cytospin 3 (Shandon Southern Instruments Inc., Sewickley, PA) at 1,000 rpm (96 × g) for 5 min. Cell differential counts were made on slides stained according to the May-Grünwald-Giemsa technique, and 400 cells per slide were counted. Mast cells were counted in at least 10 visual fields at ×160 magnification on slides stained with acid toluidine blue and counterstained with Mayer's acid hematoxylin.

A FACScan flow cytometer (Becton-Dickinson AB, Stockholm, Sweden) was used to determine lymphocyte subsets in duplicate. The following antibodies were used: CD3⁺ T-cells, CD4⁺ T-helper cells, CD8⁺ T-cytotoxic cells, CD19⁺ B-cells, CD25⁺ IL-2 receptor, CD16⁺/ CD56⁺ natural killer (NK) cells, CD45RO⁺ memory T-cells, and CD69⁺ T-cell activation marker (Becton-Dickinson AB). Albumin and total protein were measured with assays from Boehringer Mannheim (Mannheim, Germany) in an autoanalyzer at the Department of Clinical Chemistry of the University Hospital of Northern Sweden at Umeå. IL-8, soluble E-selectin and soluble intercellular adhesion molecule-1 (ICAM-1) were measured with commercially available enzyme-linked immunosorbent assay (ELISA-)kits (R&D Systems, Inc., Minneapolis, MN). Total IgA was assayed with a sandwich ELISA. Briefly, microtiter plates were coated overnight at 4° C with sheep antihuman IgA (Binding Site, Birmingham, UK) at 100 µl per well, diluted 1/1,000 in bicarbonate buffer at pH 9.6. Wells were washed three times with PBS containing 0.05% Tween-20. BW and BALF samples of 100 µl per well, diluted 1/50, and standard myeloma IgA in the concentration range of 2 to 1,000 ng/ml at 100 µl per well, diluted in PBS containing 0.05% Tween-20 and 1% bovine serum albumin (BSA), were added and the microtiter plates incubated for 90 min at 37° C. Plates were again washed three times and 100 µl per well of peroxidase-conjugated rabbit antihuman IgA (Dako Ltd, High Wycombe, UK) at 1/1,000 dilution was added. After incubation for 90 min at 37° C and three washings of the plates, 100 µl per well of o-phenylene diamine (0.4 mg/ml in PBS containing 0.05% H₂O₂) was added. Plates were incubated for 15 min and the reaction was stopped by adding 50 µl of 2 M H₂SO₄ prior to reading at 490 nm on a microtiter plate reader (Dynatech, Billingshurst, UK).

The fluid recovered from the first and second 20-ml BWs was analyzed separately, whereas that recovered from the 3 × 60 ml BAL was pooled. Cell differential counts, total protein, albumin, soluble ICAM-1, and soluble E-selectin concentrations were determined in the first BW and in the BALF.

IL-8 was measured in the first BW, whereas flow cytometry was performed only on the BALF, since the cell number in the first BW was too small to allow for this processing. The second BW was used for analyses of antioxidants and lipid peroxidation, and the results are presented elsewhere (15).

Processing and Immunostaining of Glycol Methacrylate-embedded Sections

The tissue obtained from biopsy was placed in ice-cooled acetone containing the protease inhibitors phenylmethylsulfonyl fluoride (PMSF) (2 nM) and iodoacetamide (2 nM), and was processed into glycol methacrylate (GMA) resin as previously described (18). The blocks were then stored in airtight containers at 20° C until used for immunostaining. Sections were cut at 2-µm thickness, floated onto ammonia water (1:500), and then picked onto 0.01% poly-L-lysine glass slides, where they were allowed to dry at room temperature for 1 to 4 h.

Endogenous peroxidases were blocked with a solution of 0.1% sodium azide and 0.3% hydrogen peroxide, followed by three rinses in Tris-buffered saline (TBS) adjusted to pH 7.6. Undiluted blocking medium consisting of Dulbecco's minimal essential medium (DMEM) containing 10% fetal calf serum (FCS) and 1% BSA was then applied for 30 min and drained, and primary monoclonal antibodies (mAbs) (Table 1) were added. The sections were incubated with the antibodies overnight at room temperature. After rinsing, biotinylated rabbit antimouse IgG F_ab (Dako) was applied to the section for 2 h followed by the streptavidin-biotin-horseradish peroxidase complex (Dako) for a further 2 h. After rinsing of the sections in TBS, stained areas were developed with aminoethyl carbazole in acetate buffer (pH 5.2) and hydrogen peroxide to yield a red color. The sections were then counterstained with Mayer's hematoxylin.

Quantification of Adhesion Molecules and Inflammatory Cells

Quantification of endothelial adhesion molecules in the submucosal blood vessels was done by expressing the number of vessels staining with specific antiadhesion mAb as a percentage of the total vessel complement revealed by staining with the panendothelial mAb, as previously described by Montefort (19). The number of cells stained with each mAb was counted separately in the epithelium and in the submucosa (50 µm beneath the basement membrane), and the total number of positive cells was expressed as cells/mm of epithelium and cells/mm² of submucosa, respectively. Quantification was done with a light microscope, and the length of the basement membrane and the area of the submucosa were measured with a calibrated graticule. Areas including smooth muscle, glands, large blood vessels, and torn or folded tissue within the sections were not included.

Blood Samples

Peripheral blood samples were drawn prior to the bronchoscopies. Analyses of blood cells, and differential counts were performed with an autoanalyzer in the Department of Clinical Chemistry of the University Hospital of Northern Sweden, according to clinical routine.

Statistics

Statistical analyses were done with the SPSS version 6.1.2 (SPSS, Inc., Chicago, IL) for Windows. Wilcoxon's nonparametric signed rank test for paired observations was used. A value of < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Immunohistochemistry

In both the 1.5-h and 6-h groups, 13 paired sets of biopsy specimens were available for analysis. Typical staining patterns were seen, with ring staining for CD3⁺, CD4⁺, and CD8⁺ cells, and cytoplasmic staining for mast-cell tryptase and neutrophil elastase. As compared with air exposure, no changes were found in the number of inflammatory cells in the epithelium or in the submucosa at either the 1.5-h or 6-h time points after NO₂ exposure. A significant decrease in the expression of E-selectin was found in the vascular endothelium 1.5 h after NO₂ exposure (p < 0.05), but no other changes were seen in adhesion-molecule expression at either time point (Table 2).

Cell Parameters in BW and BALF

The median BW and BALF recoveries after air exposure in the 1.5-h group were 7.0 ml (range: 6.0 to 7.0) and 135 ml (range: 125 to 140), and in the 6-h group were 8.0 ml (range: 7.0 to 8.5) and 135 ml (range: 120 to 140), respectively. The amounts of fluid recovered after the NO₂ exposures did not differ significantly.

In BW, no changes were detected in total cell numbers or differential counts after 1.5 h, but at 6 h after exposure to NO₂, both the percentage and the total number of neutrophils were increased in the proximal lavage fluid as compared with the control day (p < 0.05 and p < 0.01, respectively) (Table 3).

In BALF, no significant changes were found in total cell numbers or differential counts at either the 1.5-h or 6-h exposure time-points (Table 4). However, flow cytometry revealed an increase in the percentage of CD45RO⁺ lymphocytes (p < 0.05) in BALF 1.5 h after NO₂ exposure. At 6 h after exposure, the percentage and total number of lymphocytes, as well as of CD4⁺ and CD8⁺ T-cells, were unchanged, but increases were found in the percentages of both CD19⁺ (B-cells) (p < 0.05) and activated CD69⁺ cells (p < 0.05). A significant increase was also detected in the percentage of CD16⁺/CD56⁺ NK-cells (p < 0.05) (Table 5).

Soluble Mediators, Cytokines, and Soluble Adhesion Molecules

An increase in IL-8 (p < 0.05) was found in BW at 1.5 h after NO₂ exposure, at which time a reduction in total protein and a trend toward a decrease in albumin were found. The concentration of soluble ICAM-1 was unchanged at 1.5 h (Table 6). At 6 h after NO₂ exposure there was no difference in IL-8 levels, but reductions were found in both soluble ICAM-1 (p < 0.05) and albumin (p < 0.01), with a trend toward a decrease in total protein (Table 6). Because of a lack of lavage fluid, IgA was measured in only 13 paired samples after 1.5 h and in only seven paired samples at the 6-h time point. In BW, no changes were found in IgA at 1.5 h but a significant decrease was seen in IgA at 6 h (p < 0.05) (Table 6). No changes were found in any of the soluble components of BALF that were measured (Table 6). Soluble E-selectin was not detectable at any time point (data not shown).

Blood Parameters

The total cell numbers and differential counts were unchanged at 1.5 h after exposure to NO₂, but at the 6-h time point, an increase was found in both total leukocyte numbers (median: 5.3/mm³; range: 4.7 to 6.9/mm³, versus median: 6.0/ mm³; median 5.6 to 7.8/mm³) (p < 0.05) and lymphocyte counts (median: 1.8/mm³; range: 1.6 to 2.0/mm³, versus median: 2.0/mm³; median 1.8 to 2.2/mm³) (p < 0.05). No changes were seen in either neutrophil or monocyte numbers.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that exposure of healthy subjects to NO₂ induces a neutrophilic airway inflammation as seen in BW, in association with enhanced IL-8 secretion but without signs of inflammatory-cell recruitment in mucosal biopsies from the proximal airways.

The subjects were exposed to an NO₂ concentration of 2 ppm for 4 h, and bronchoscopy was done at either 1.5 h or 6 h after exposure in order to analyze the time course of airway inflammation. The lavage fluids were collected in two parts, as a bronchial wash (BW) mainly reflecting the airways proximal to the alveoli, including the terminal bronchioles, and as BALF, reflecting the more peripheral bronchoalveolar regions (9, 20). The findings in the present study confirm the recently published finding that exposure to NO₂ induces neutrophilic inflammation in the bronchi of healthy humans (9, 10). The fact that no significant changes were found in the BALF samples indicates that the major target site for the action of NO₂ is probably the terminal bronchioles, as suggested in previous human and animal studies (21). In contrast to the neutrophilic inflammation seen in BW, no signs of inflammatory-cell recruitment or endothelial adhesion-molecule upregulation were detected in the mucosal biopsy specimens at either 1.5 h or 6 h after NO₂ exposure. These findings are consistent with those previously reported by Magnussen and colleagues (16), and provide further evidence that the effects of NO₂ on the mucosa are more pronounced in the smaller airways, probably in the terminal bronchioles, where the major transit of inflammatory cells from the mucosa into the air spaces is thought to take place. This has been shown in animal studies, in which this more peripheral airway tissue can be investigated (22). To reach this area in humans, investigations with transbronchial biopsies would be required.

One of the most interesting findings of the present study is the early (1.5 h) increase in IL-8 in BW. IL-8, a C-X-C chemokine, is a very potent chemoattractant for neutrophils (23). The increase in IL-8 at 1.5 h after exposure was followed by a neutrophilia seen at the 6-h sampling point, at which time IL-8 had returned to control levels. It is therefore probable that the recruitment of neutrophils into BW at 6 h was caused by IL-8. An increased concentration of IL-8 in airway lavage has not previously been reported in humans after exposure to NO₂. In an earlier in vitro study, Devalia and colleagues (13) reported increased IL-8 synthesis in epithelial cells after exposure to 0.4 ppm NO₂ for 6 h. In vivo, the situation is more complex, since the pulmonary epithelium is protected by the antioxidant screen present in the epithelial lining fluid (24). Antioxidants have the potential to modify the effects of the free radical NO₂ on the airway epithelium. We have already reported that ascorbic acid and uric acid are depleted at 1.5 h after NO₂ exposure, indicating a significant compromise in the antioxidant defense screen (15). It has been suggested that the main mechanism of the pulmonary toxicity of NO₂ is the initiation of lipid peroxidation in cell membranes, which results in the structural and functional alteration of various molecules (25, 26). Of note however, is that no evidence of lipid peroxidation was observed in this study, as determined by malondialdehyde formation (15). This finding does not, however, rule out the possibility of more subtle cell-membrane responses. Oxidant stress has been shown to stimulate IL-8 production in a variety of cell lines in vitro (27), and another in vitro study of respiratory syncytial virus (RSV)-infected airway epithelium has also supported a direct role for reduction-oxidation in the production of IL-8 in the epithelium (28). It has also been suggested that oxidant stress can activate the gene transcription factor nuclear factor-B (NF-B) (29), which is active in regulating the expression of a variety of genes, including that for IL-8 (30). Hence, a change in redox state, induced by exposure to NO₂ (15), could result in NF-B activation and subsequent IL-8 production in the airway epithelium. It remains unclear how IL-8 production is initiated and which cells are the source of the IL-8 released after NO₂ exposure. Alveolar macrophages and airway epithelial cells have both been shown to produce IL-8 in vitro (31).

Previous chamber studies using BAL to elucidate the effects of NO₂ found changes in the number of lymphocytes and mast cells in BALF (9, 10). The present study employed a higher concentration of NO₂ and a longer exposure time, and found no changes in lymphocyte or mast-cell numbers. However, Becker and associates also failed to detect any significant changes in lymphocyte numbers in a study with the same NO₂ concentration and exposure duration as used in the present study (10). Hence, it is probable that the exposure duration and amount of inhaled NO₂ influence the lymphocyte response. Small but statistically significant increases were observed in the percentages of CD45RO⁺ T-lymphocytes at 1.5 h and of B-cells and NK-cells at 6 h after NO₂ exposure.

The decreases in total protein and albumin in BW at both 1.5 h and 6 h are difficult to interpret, since similar responses to NO₂ have not been reported previously. It is noteworthy that total protein, IgA, and sICAM-1 all showed proportionate decreases. Becker and associates (10), in a study with a design similar to ours, found a nonsignificant trend toward a decrease in protein levels in their proximal lavages after NO₂ exposure. In another study, exposure to hydrogen fluoride significantly reduced protein as well as albumin levels in BALF (32). A number of possible mechanisms can be proposed that either alone or in combination could be responsible for these changes in protein and albumin concentrations. The first of these is increased clearance of proteins at the site of airway inflammation after NO₂ exposure; second, but less likely, is that NO₂ could affect the airway epithelium, resulting in decreased permeability with less leakage of proteins from the blood vessels into the air spaces; a third possibility is that proteins could have been degraded by the action of metalloproteinases, which has been shown for ICAM-1 (33). Given the known oxidative properties of NO₂ (34), it seems likely that the changes in proteins in BW after NO₂ exposure were due to oxidative degradation. More work will be needed to clarify the mechanisms behind these effects of NO₂ on the human airways.

It is interesting to note that these results are similar to those obtained after experimental exposure to ozone (O₃), another important oxidative air pollutant. Ozone also induces an increase in IL-8 in proximal airway lavage (35) and a neutrophilic inflammation in the airways of healthy humans (35, 36). However, the two pollutants show differential effects on BALF protein. The increased leakage of protein into the airways after ozone exposure is believed to indicate increased vascular permeability in the lower airways (35, 36), whereas no obvious explanation is at hand for the decreased protein levels seen after NO₂ exposure.

In conclusion, exposure of healthy subjects to 2 ppm NO₂ for 4 h induced a neutrophilic inflammation in the airways that was detectable in BW but not in BALF at 6 h after the end of the exposure. The increase in neutrophils was most likely due to enhanced IL-8 secretion, which was seen in BW at 1.5 h after exposure. The mechanisms behind this increase in IL-8, and its cellular origin, are still unclear, but it is plausible that increased oxidant stress during exposure to NO₂ is involved. At both time points after NO₂ exposure, histologic examination of mucosal biopsies from the proximal airways revealed no evidence of cellular inflammation or upregulation of adhesion molecules on the vascular endothelium, indicating that the major site of inflammation following exposure to NO₂ is in the more peripheral airways (i.e., the terminal bronchioles).