Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

In recent years, there has been a progressive increase in air pollution that is characterized by high concentrations of atmospheric hydrocarbons, oxides of nitrogen (NO_x), ozone (O₃), and respirable particulate matter (PM₁₀), resulting primarily from increased use of liquid petroleum and gas in the transport and manufacturing industries and domestic settings. Epidemiologic studies have reported that there is a clear association between episodes of air pollution and impaired lung function, cough, and infections of the lower respiratory tract (1). Studies of increased levels of PM₁₀ have reported that these are associated with emergency room visits for asthma, worsening peak flow, inhaler usage, and respiratory symptoms in asthmatics, with a lag effect of 2-4 d (4, 5). More recently, Pope and colleagues have reported that cardiopulmonary and lung cancer mortality, in a cohort of over half a million adults residing in 151 metropolitan areas in the U.S. between 1982 and 1989, were associated with fine particulate air pollution (6). Recently, Anderson and colleagues have estimated the relative risks for adverse health effects due to daily variations in air pollution in London between April 1987 and March 1992; they suggested that increased concentrations of black smoke (reflecting respirable particles) on the previous day were significantly associated with all cause mortality and that this effect was greater in the warm season (7).

Although the mechanisms underlying respiratory morbidity due to PM₁₀ are not clear, it is thought that the fine particles (those with aerodynamic diameters equal to or less than 2.5 µm) are of greatest concern to health since they can be breathed most deeply into the lung, where they are likely to be more toxic than the larger particles (6, 8). Studies in animals have suggested that diesel exhaust particles (DEP)-induced lung injury may be a result of increased production of active oxygen radicals (9, 10). There is increasing evidence that airway epithelial cells, the primary target for air pollutants, may play an important role in the etiology of airways disease because they can express and synthesize a large variety of proinflammatory cytokines which directly or indirectly influence the activity of eosinophils, neutrophils, mast cells, macrophages, and lymphocytes (11, 12). We have recently shown that exposure of human bronchial epithelial cells (HBEC) to nitrogen dioxide (NO₂) (13, 14) significantly increases the release of interleukin (IL)-8, tumor necrosis factor alpha, and granulocyte-macrophage colony stimulating factor (GM-CSF) in vitro.

To date, however, there is little or no information on the pathologic changes induced by DEP in the human airways, nor the mechanisms underlying these changes. In this study we have cultured HBEC to confluence and investigated the effect of purified DEP and/or filtered DEP solution on the permeability and ciliary beat frequency (CBF) of these cultures. Additionally, we have studied the effect of purified DEP and/or filtered DEP solution on the release of proinflammatory cytokines and adhesion molecules from these cells.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

All chemicals and reagents were of tissue culture grade and, unless stated otherwise, were obtained from the Sigma Chemical Co. (Poole, UK).

Bronchial Tissue

Bronchial tissue was obtained from nine patients who underwent lobectomy or pneumonectomy for lung cancer at St. Bartholomew's Hospital, London. All these patients were heavy smokers, with no known allergies to any of the common allergens, and a mean age of 53 yr (range = 28- 64 yr). After resection, only tissue that appeared macroscopically free of tumor and deemed to be "normal" by the operating surgeon was placed into ice-cold Medium 199 (Northumbria Biologicals Ltd., Cramlington, UK) and brought back to the laboratory for processing for tissue culture, within 0.5 to 1 h of resection.

Isolation, Culture, and Identification of HBEC

HBEC were cultured by an explant cell culture technique we have developed in our laboratory and described fully elsewhere (15). Briefly, the epithelium was carefully dissected away from the underlying tissue and cut further into smaller sections approximately 1-2 mm³ in size. All the sections were "sterilized" by gently washing three times in prewarmed and pregassed Medium 199 containing 1% (vol/vol) antibiotics/antimycotic solution (Sigma); 2-3 sections were then explanted into either untreated 35-mm-diameter Falcon^® "Primaria^TM" plastic culture dishes (Becton Dickinson Ltd., Oxford, UK) or into 9-mm-diameter Falcon^® Cell Culture Inserts, with 0.45-µm pore size microporous membranes (Becton Dickinson Ltd.). The explants were incubated at 37°C in 5% CO₂ in air atmosphere in freshly prepared and micropore filter-sterilized culture medium containing 2.5 ml Nu-serum IV (Universal Biologicals Ltd., Gloucestershire, UK), 250 µg bovine pancreatic insulin, 250 µg human transferrin, 36 µg hydrocortisone, 1 mg L-glutamine, and 1 ml antibiotics/antimycotic solution, in 100 ml Medium 199. Explants cultured in dishes were incubated in 2.0 ml culture medium and observed for cell outgrowth over a period of 2-3 wk, until the cells had grown to confluence. For explants incubated in cell culture inserts, 0.5 ml and 0.4 ml medium was added into the insert and the insert well, respectively, and treated the same as the cultures established in dishes.

The purity and identity of the cells were checked in randomly selected cultures and were confirmed by (1) light microscopy, (2) electron microscopy, and (3) indirect immunoperoxidase staining techniques, using specific monoclonal antibodies directed toward cytokeratin and specific ciliated epithelial cell antigens (11).

Preparation of DEP and Activated Charcoal Suspensions and Filtered DEP Solution

DEP (mass median aerodynamic diameter = 0.4 µm, geometric standard deviation = 0.195 µm and 1.42 µm) were purified and characterized as previously described (9). Briefly, exhaust generated from a diesel engine was introduced into a stainless-steel dilution tunnel and the DEP were collected on a glass-fiber filter situated in a constant-volume sampler placed at the end of the tunnel. The diameter of the particles was measured using a low-pressure Andersen Air Sampler and then investigated further by scanning electron microscopy to determine the general shapes of the particles collected.

The purified DEP were suspended in serum-free supplemented medium (SF-1 medium) at concentrations of 10-100 µg/ml prior to use in further investigations. The SF-1 medium was prepared by mixing 0.5 ml SF-1 serum-free supplement (Northumbria Biologicals Ltd.) and 1.0 ml antibiotics/antimycotic solution in 100 ml Medium 199. Similarly, activated charcoal (250-350 mesh; Sigma) was suspended in SF-1 medium at concentrations of 50-100 µg/ml, and used as negative controls.

In a separate set of experiments, a DEP suspension of 50 µg/ml prepared as above was allowed to stand in the tissue culture incubator at 37°C in 5% CO₂ in air atmosphere for 24 h and, after gentle agitation, was filtered through a Minisart^® micropore (0.2 µm) filter (Sartorius Ltd, Epsom, UK) to obtain a clear solution (filtered DEP solution). The effect of this filtered DEP solution was then investigated the same as the DEP suspensions were.

Analysis of Polyaromatic Hydrocarbons Present in Purified DEP and Filtered DEP Solution

Polyaromatic hydrocarbons (PAHs) present in purified DEP and filtered DEP solution were analyzed with a reversed phase-high performance liquid chromatography (RP-HPLC) system. Prior to analysis, 20 mg DEP were suspended in 200 ml phosphate-buffered saline solution, to give a concentration of 100 µg/ml DEP suspension. After mixing by vigorous vortexing for 3 min, 100 ml of the DEP suspension was then mixed three times with 20-ml aliquots of dichloromethane to extract the PAHs adsorbed onto the DEP. The dichloromethane was blown to dryness under a gentle stream of nitrogen and the residue dissolved in 1.0 ml methanol. This concentrated sample was filtered through a 0.2-µm-pore-size membrane filter (Advantec Toyo, Tokyo, Japan) and then analyzed by RP-HPLC.

In a separate experiment, 100 ml of the 100 µg/ml DEP suspension was incubated for 24 h at 37°C, and then filtered through a 0.2-µm-pore-size filter. The filtered DEP solution was then extracted with dichloromethane and treated further as above.

Samples (100 µl each) of the extracted DEP were injected via a Rheodyne 8125 injection valve fitted with a 10-µl sample loop onto a 15 cm × 4.6 mm (internal diameter) Shin-Pack CLC-ODS column (Shimadzu Co. Ltd., Kyoto, Japan), and chromatography was performed using an acetonitrile/water (3:1, vol/vol) mobile phase. The mobile phase was pumped at a constant flow rate of 1 ml/min by an LC-6A model HPLC pump (Shimadzu) connected to degassing unit DG-660 (GL-Science Co. Ltd., Tokyo, Japan), and the eluted PAHs were detected by an RF-550 spectrofluorometric detector (Shimadzu) fitted with a 12-µl flow-through cell. The spectrofluorometric detector was programmed to read at excitation/emission wavelengths 297 nm/353 nm between 0 and 8.1 min, 349 nm/444 nm between 8.1 and 9.45 min, and 321 nm/390 nm from 9.45 min onward, to detect phenanthrene, fluoranthene, and pyrene, respectively. The eluted PAH peaks were recorded and evaluated by a Chromatopac C-R1A recorder-integrator (Shimadzu).

The concentrations of phenanthrene, fluoranthene, and pyrene in the samples were calculated from standard curves prepared and analyzed under similar conditions, each time the samples were analyzed.

Effect of DEP on Permeability of Confluent Cultures of HBEC

Changes in permeability of HBEC cultures were investigated by measuring changes in both the electrical resistance and movement of ¹⁴C-bovine serum albumin (¹⁴C-BSA) (Amersham International Plc., Amersham, UK) across epithelial cell cultures established in cell culture inserts. Prior to investigation, HBEC were grown to confluence and the explants removed. The epithelial cells in the inserts were allowed to grow over the area of the culture insert membrane that was left barren on removal of the explant, and when completely confluent were used further. Immediately prior to exposure to DEP or activated charcoal, the cultures were washed gently with fresh culture medium and then incubated for 30 min in the presence of 0.025 µCi ¹⁴C-BSA. At the end of this incubation the medium in each insert well was collected and analyzed for total radioactivity by liquid scintillation counting in a Beckman LS6500 scintillation counter (Beckman-RIIC Ltd, High Wycombe, UK). When the total radioactivity passing through the culture was found to be less than 0.5% of the total ¹⁴C-BSA added to the inserts at the beginning of the experiment, the culture was deemed to be fully confluent and the experiment proceeded further. In this manner, sets of at least six separate cultures, each from different individuals, were exposed to different concentrations of DEP ranging from 0-100 µg/ ml or 100 µg/ml activated charcoal. On addition of DEP or activated charcoal to each culture, electrical resistance was measured immediately and then again after 1, 2, 4, 6, and 24 h, using an EVOM^TM microvolt-ohm meter (World Precision Instruments, Owslebury, UK).

Immediately following measurement of the electrical resistance at each specific time point during incubation, the medium was collected from each insert well. Fresh medium was replaced into the insert well and the culture was incubated further at 37°C in 5% CO₂ in air atmosphere. Aliquots (25 µl) of each sample collected were analyzed for radioactivity and, after correction for the total amount of radioactivity passing across the epithelial cultures at each time point, expressed as a percentage of the total added to the culture at the beginning of the experiment.

Effect of DEP on CBF

In contrast to the studies of epithelial permeability, cultures established on tissue culture dishes were used for these studies. CBF was measured by a modification of the analog contrast enhancement technique using the Reece Scientific PCX On Screen Measurement System (Brian Reece Scientific Instruments, Newbury, UK), as we have described before (15). Sets of at least six cultures each from different individuals were investigated as above. Prior to treatment, each culture was equilibrated for 2 min at room temperature and monitored for the "baseline" CBF in at least six randomly chosen areas of ciliated cells in the culture. The position of each of these six areas was noted accurately by means of a horizontal and a vertical Vernier scale on the microscope stage, and the culture was then incubated in either (1) 2 ml culture medium containing a specific concentration of DEP ranging from 0-100 µg/ml, (2) 2 ml of filtered DEP solution, or (3) 2 ml culture medium containing activated charcoal at a concentration of 100 µg/ml. At 2, 4, 6, and 24 h of incubation the culture was equilibrated for 2 min at room temperature and monitored for CBF in the same areas as at the beginning of the experiment. The CBF in each culture was calculated as the mean of six areas within the culture at baseline and at each time point during incubation, and expressed as percent of attenuation from baseline.

Effect of DEP on Release of IL-8, GM-CSF, and Soluble Intercellular Adhesion Molecule-1 from HBEC

Confluent cultures (2- to 3-wk-old) of HBEC established on cell culture inserts were used for these experiments, as described above. Prior to exposure to DEP or activated charcoal, all the cultures were equilibrated by incubation for 24 h in SF-1 medium. Following this initial incubation, the cultures were gently washed three times with fresh prewarmed and pregassed SF-1 medium; sets of at least six cultures each from different individuals were then exposed for 24 h to either 0-100 µg/ml DEP, the filtered DEP solution, or activated charcoal at a concentration of 50 µg/ml (in preliminary studies a concentration of 50 µg/ml DEP was found to be optimal in inducing the release of mediators from HBEC cultures). At the end of exposure, the medium was collected from each culture and the cells were gently washed with 0.2 ml fresh ice-cold Medium 199. The wash was pooled with the culture medium and stored at 70°C, until analyzed for IL-8, GM-CSF, and soluble intercellular adhesion molecule-1 (sICAM-1), using commercially available ELISA kits (R&D Systems, Abingdon, UK). The culture membrane was detached from the insert and immersed in 0.25 ml 1 M NaOH solution, prior to storing at 20°C until analyzed for cellular protein, according to the method of Lowry and associates (16). All results were expressed as pg mediator/µg cellular protein.

Statistical Analysis

All data were tested for distortion and then analyzed further using nonparametric statistical tests. First, the Kruskal- Wallis one-way analysis of variance by ranks was applied to test for significance of any differences between all the sets of treated cultures. This was followed by further analysis with the Mann-Whitney U test to assess differences between individual treatment groups. Bonferroni's correction was employed for multiple comparisons and all results were expressed as median and interquartile (Q1 and Q3) values. All statistical tests were performed using a standard computer package (Minitab Data Analysis Software Release 6.1.1, 1987; Minitab Inc., State College, PA) and values of P < 0.05 were considered to be significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

PAHs Adsorbed to DEP

Table 1 shows the concentrations of phenanthrene, fluoranthene, and pyrene detected in unfiltered and filtered solutions of 100 µg/ml DEP. RP-HPLC analysis demonstrated that the 100 µg/ml DEP suspension and 100 µg/ml filtered DEP solution contained 16.26 and 1.44 ng/ml phenanthrene, 3.65 and 0.18 ng/ml fluoranthene, and 2.53 and 0.15 ng/ml pyrene, respectively.

Effect of DEP on the Permeability of HBEC Cultures

Light microscopic examination of epithelial cell cultures incubated in the presence of DEP or activated charcoal for 24 h demonstrated that this agent did not lead to any gross cellular damage or detachment of the cells at any studied concentration between 10 and 100 µg/ml.

Measurement of the electrical resistance across the epithelial cultures demonstrated that this was significantly increased, in a dose- and time-dependent manner, by incubation of the cells in the presence of DEP (Figure 1). The electrical resistance was found to be significantly increased by 35% over baseline value after 1 h incubation in the presence of 10 µg/ml DEP and was maximal (increased by more than 200% over the baseline value) after 6 h incubation with 100 µg/ml DEP, when compared with incubation in the absence of DEP or presence of 100 µg/ml activated charcoal. Similarly, incubation of HBEC cultures in the presence of filtered DEP solution significantly increased the electrical resistance of the cultures when compared with incubation in the absence of DEP or presence of 100 µg/ml activated charcoal. Analysis of the changes in the electrical resistance induced by incubation of HBEC cultures in either filtered DEP solution or unfiltered solution of 50 µg/ml DEP demonstrated that both solutions led to similar increases in the electrical resistance of these cultures at each time point and that these were maximal after 24 h (median = 177, Q1 = 55, Q3 = 254 for filtered DEP solution; median = 186, Q1 = 99, Q3 = 270 for unfiltered DEP solution). In contrast, the electrical resistance was not significantly altered at any time during incubation, either in the untreated control cultures or in cultures exposed to 100 µg/ml activated charcoal (Figure 1).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Effect of 24 h exposure to 0-100 µg/ml DEP and filtered solution of 50 µg/ml DEP on the electrical resistance (ER) of human bronchial epithelial cell cultures in vitro. Results are calculated as percent change from baseline in each culture and expressed as median and interquartile values (n = 9 at each time point; *P < 0.05 and **P < 0.002 versus 0 µg/ml DEP at all time points).

Evaluation of the effect of DEP on the movement of ¹⁴C-BSA from the serosal to the basal aspects of HBEC cultures demonstrated that although this increased progressively from a median value of 0 to 1.3%, over a period of 0-24 h incubation it was not significantly altered at any time during incubation by any concentration of DEP investigated, when compared with the untreated control cultures (Figure 2).

View larger version (12K): [in this window] [in a new window]   Figure 2.   Effect of 24 h exposure to 0-100 µg/ml DEP on the movement of 14C-BSA from the serosal to the basal aspects of human bronchial epithelial cell cultures in vitro. Results are calculated as percentage of total 14C-BSA in the culture well at each time point and expressed as median and interquartile values (n = 9 at each time point).

Effect of DEP on CBF

The CBF of HBEC cultures was not significantly altered from the baseline at any time during incubation in either the untreated control cultures or cultures exposed to 100 µg/ml activated charcoal. In contrast, incubation of HBEC cultures in the presence of DEP progressively attenuated the CBF of the cultures, in a time- and dose-dependent manner (Figure 3). Exposure to 10 µg/ml DEP attenuated the CBF by 40% (Q1 = 19, Q3 = 46) from baseline after 24 h incubation. Similarly, exposure to 50 µg/ml DEP, filtered DEP solution, or 100 µg/ml DEP attenuated the CBF of these cells by 51% (Q1 = 49, Q3 = 56), 33% (Q1 = 26, Q3 = 36), and 73% (Q1 = 65, Q3 = 83), respectively, from baseline after 24 h incubation. Statistical analysis of the DEP-induced-changes in the CBF demonstrated that these were significant after 4 h incubation in cultures exposed to 50 µg/ml DEP, and after 2 h incubation in cultures exposed to 100 µg/ml DEP, when compared with either untreated control cultures or cultures treated with 100 µg/ml activated charcoal.

View larger version (15K): [in this window] [in a new window]   Figure 3.   Effect of 24 h exposure to 0-100 µg/ml DEP and filtered solution of 50 µg/ml DEP on the CBF of HBEC cultures in vitro. Results are calculated as percent change from baseline in each culture and expressed as median and interquartile values (n = 6 at each time point; *P < 0.05 and **P < 0.002 versus 0 µg/ml DEP at all time points).

Effect of DEP on Release of IL-8, GM-CSF, and sICAM-1

Analysis of IL-8 released into the culture medium demonstrated that this was increased by exposure of HBEC cultures to DEP after 24 h incubation (Figure 4). Although 10 µg/ml DEP released 15.3 pg IL-8/µg cellular protein (Q1 = 3.7 and Q3 = 29.9 pg IL-8/µg cellular protein), this was not significantly different from 12.9 pg IL-8/µg cellular protein (Q1 = 5.3 and Q3 = 14.9 pg IL-8/µg cellular protein) released by untreated control cultures or 11.9 pg IL-8/µg cellular protein (Q1 = 9.6 and Q3 = 13.0 pg IL-8/µg cellular protein) released by cultures treated with 50 µg/ml activated charcoal. In contrast, exposure to 50 and 100 µg/ml DEP released significantly greater quantities of 41.6 pg IL-8/µg cellular protein (Q1 = 25.7 and Q3 = 48.7 pg IL-8/µg cellular protein, P < 0.05) and 44.3 pg IL-8/µg cellular protein (Q1 = 14.5 and Q3 = 60.7 pg IL-8/µg cellular protein; P < 0.05), when compared with incubation in the absence of DEP or presence of 50 µg/ml activated charcoal. Analysis of cultures exposed to filtered DEP solution demonstrated that this led to an even greater, nearly 10-fold, release of 114.9 pg IL-8/µg cellular protein (Q1 = 92.1 and Q3 = 143.5 pg IL-8/µg cellular protein) when compared with untreated control cultures (P < 0.001) or cultures treated with 50 µg/ml activated charcoal (P < 0.002). Release of IL-8 from HBEC cultures exposed to 50 µg/ml activated charcoal, however, was not significantly increased when compared with release from untreated control cultures (Figure 4).

View larger version (9K): [in this window] [in a new window]   Figure 4.   Effect of 24 h exposure to 0-100 µg/ml DEP and filtered solution of 50 µg/ml DEP on the release of IL-8 from HBEC cultures in vitro. Results are expressed as median and individual values (*P < 0.02 and **P < 0.01 versus 0 µg/ml DEP).

Studies of GM-CSF demonstrated that the release of this mediator was also increased by exposure of the cultures to DEP after 24 h incubation. Although 10 and 100 µg/ml DEP appeared to increase the release of this cytokine from HBEC, statistical analysis demonstrated that neither concentration induced significantly greater release of GM-CSF when compared with either untreated control cells or cultures treated with 50 µg/ml activated charcoal (Figure 5). In contrast, 50 µg/ml DEP significantly increased the release of GM-CSF from a control value of 0.03 pg GM-CSF/µg cellular protein (Q1 = 0.02 and Q3 = 0.04 pg GM-CSF/µg cellular protein) to 0.06 pg GM-CSF/ µg cellular protein (Q1 = 0.04 and Q3 = 0.09 pg GM-CSF/ µg cellular protein; P < 0.05). Similarly, 50 µg/ml-induced GM-CSF was significantly higher when compared with cultures treated with 50 µg/ml activated charcoal (median = 0.04, Q1 = 0.02, and Q3 = 0.04 pg GM-CSF/µg cellular protein; P < 0.02). Similarly, filtered DEP solution led to significantly increased release of 0.2 pg GM-CSF/µg cellular protein (Q1 = 0.14 and Q3 = 0.24 pg GM-CSF/µg cellular protein) when compared with untreated control cultures (P < 0.006) or cultures treated with 50 µg/ml activated charcoal (P < 0.006). In contrast, incubation of the HBEC with 50 µg/ml activated charcoal did not significantly increase the release of GM-CSF when compared with untreated control cultures (Figure 5).

View larger version (9K): [in this window] [in a new window]   Figure 5.   Effect of 24 h exposure to 0-100 µg/ml DEP and filtered solution of 50 µg/ml DEP on the release of GM-CSF from HBEC cultures in vitro. Results are expressed as median and individual values (*P < 0.03 and **P < 0.006 versus 0 µg/ml DEP).

Analysis of sICAM-1 demonstrated that neither 10 nor 100 µg/ml DEP significantly increased the release of this mediator when compared with untreated control cultures (Figure 6). However, as was shown to be the case for release of GM-CSF, the release of this adhesion molecule was also significantly increased by exposure to 50 µg/ml DEP after 24 h incubation (median = 12.5 pg sICAM-1/µg cellular protein, Q1 = 11.36 and Q3 = 14.9 pg sICAM-1/ µg cellular protein) when compared with untreated control cultures (median = 7.2 pg sICAM-1/µg cellular protein, Q1 = 5.6 and Q3 = 8.8 pg sICAM-1/µg cellular protein; P < 0.01) or cultures treated with 50 µg/ml activated charcoal (median = 9.4 pg sICAM-1/µg cellular protein, Q1 = 7.9 and Q3 = 10.6 pg sICAM-1/µg cellular protein; P < 0.05). Analysis of cultures exposed to filtered DEP solution demonstrated that this also led to significantly increased release of 15.2 pg sICAM-1/µg cellular protein (Q1 = 12.4 and Q3 = 22.3 pg sICAM-1/µg cellular protein; P < 0.005) when compared with untreated control cultures or cultures treated with 50 µg/ml activated charcoal. In contrast, incubation of the HBEC with 50 µg/ml activated charcoal did not significantly alter the release of sICAM-1 when compared with release from untreated control cultures (Figure 6).

View larger version (9K): [in this window] [in a new window]   Figure 6.   Effect of 24 h exposure to 0-100 µg/ml DEP and filtered solution of 50 µg/ml DEP on the release of sICAM-1 from HBEC cultures in vitro. Results are expressed as median and individual values (*P < 0.01 and **P < 0.005 versus 0 µg/ml DEP).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

These studies have demonstrated that incubation of HBEC either in culture medium containing DEP suspended at concentrations of 10-100 µg/ml or in a filtered DEP solution for 24 h leads to an attenuation of ciliary activity of HBEC and release of proinflammatory cytokines, such as IL-8, GM-CSF, and the adhesion molecule sICAM-1, in vitro. This is in marked contrast to incubation of HBEC in the presence of activated charcoal, which did not lead to any alteration in the activity of these cells. To our knowledge, this is the first report of the effect of DEP on HBEC, and is of particular significance due to the predominance and strategic positioning of these cells in the airways, where they may play an important modulatory role in the development of airways disease (17, 18). In view of our observation that DEP induces significant release from HBEC of proinflammatory mediators which can influence the activity of neutrophils and eosinophils, it is tempting to hypothesize that DEP-induced inflammation of the airways is a consequence of inflammatory mediators released primarily from the airway epithelial cells.

Although the in vitro model used in these studies is limited in that all the naturally occurring defense mechanisms are not present and that the "condition" of the particles may not be representative of freshly generated particles, the findings of the present study give an indication of the putative underlying mechanisms and largely complement the findings of several human and animal studies. Whereas it is possible that under exceptional circumstances, such as phagocytosis, the size of particles may be an important determinant in the activation of cells, in the present study this is unlikely to have been the case because preparations of both DEP and activated charcoal aggregated to form complexes of much larger size than the original size of the individual particles when suspended in the culture medium. Also, our observation that filtered DEP solution, but not activated charcoal, was able to induce changes in electrical resistance, attenuate the ciliary beat frequency, and increase the release of inflammatory mediators from HBEC, suggests that these effects were likely to be a result of compounds adsorbed onto the carbonaceous core of DEP, rather than the size of the DEP. Indeed, our finding that PAHs such as phenanthrene, fluoranthene, and pyrene can be released on suspension of these particles into medium is in accordance with the finding of Barfknecht and colleagues, who have reported that phenanthrenes constitute a major component of the polyaromatic hydrocarbons adsorbed to DEP (19).

Our finding, however, that DEP leads to increased release of inflammatory cytokines from HBEC, is in accordance with the findings of Diaz-Sanchez and Saxon (20), who investigated the effect of DEP on the synthesis of cytokines in the nasal mucosa of subjects allergic to ragweed allergen. These authors found that intranasal challenge with 0.3 mg purified DEP led to increased and readily detectable levels of mRNA for IL-2, -4, -5, -6, -10, and -13 and interferon gamma (IFN-) in the nasal mucosal cells of these individuals 18 h after challenge. Additionally, these authors found that although the levels of mRNAs for IL-4, -5, -6, -10, and -13 were increased even further when allergen and DEP challenge were applied simultaneously, the levels of IL-2 were unchanged and the levels of IFN- mRNA were decreased, suggesting that DEP upregulated the activity of TH2-like lymphocytes. More recently, Tsien and associates have investigated the effect of incubating human B cells with either PAHs extracted from DEP or purified phenanthrene, and found that these compounds lead to significant production of IgE by the B cells in vitro (21). Furthermore, these authors found that the PAH extract and phenanthrene significantly increased the transcription of mRNA coding for IgE in these cells.

Although our studies of epithelial permeability suggest that DEP does not directly alter the permeability of HBEC in vitro, as demonstrated by a lack of effect on the passage of ¹⁴C-BSA across these cultures and an increase (rather than a decrease) in the electrical resistance of these cultures, our finding that DEP attenuate the ciliary activity of HBEC in a dose-dependent manner suggests that these compounds can indeed lead to adverse cellular functional changes, and complements the findings of others. Several animal studies have reported that DEP lead to significantly increased generation of superoxide and hydroxyl radicals and inhibition of the activities of superoxide dismutase, glutathione peroxidase, and catalase in epithelial lining fluid, and have suggested that the increased oxygen radical load leads to increased lung injury and mortality (9, 10, 22, 23). Ichinose and colleagues have investigated the effects of intratracheal instillation of DEP in mice and reported that this results in dose-dependent intra-alveolar hemorrhage, perivascular edema, and bronchiolar cell hypertrophy 18-24 h after DEP instillation (24). Additionally, these authors found that DEP also led to an influx of neutrophils into alveolar spaces. More recently, these authors have reported that intratracheal instillation (once a week for 16 wk) of DEP in mice led to marked infiltration of inflammatory cells, proliferation of goblet cells, increased mucus secretion, respiratory resistance, and airway constriction (10). Furthermore, these authors found that eosinophils in the submucosa of the proximal bronchi and medium bronchioles were increased 8-fold by the DEP, and that this eosinophilic infiltration of the airways was attenuated by pretreatment of the animals with polyethyleneglycol-conjugated superoxide dismutase.

Ulfvarson and colleagues investigated the effect of exposure to diluted diesel exhaust in rabbits and found that this leads to moderate inflammation in the larger and smaller bronchi, as indicated by increased numbers of eosinophils and mononuclear and polymorphonuclear cells in the submucosa and the epithelium (25). Similarly, Rudell and colleagues investigated the effect of exposure to diesel exhaust on inflammatory changes in the lungs of nonasthmatic subjects and found that this led to a significant increase in the numbers of neutrophils in the bronchoalveolar lavage fluid 18 h after exposure (26).

In conclusion, the results of the present study suggest that DEP, which constitute an important fraction of respirable particulate pollutants, may modulate airway disease by adversely influencing the activity of airway epithelial cells that form the first line of cellular defense toward inhaled irritants.