Inhalation of Nitrogen Dioxide Fails to Reduce the Activity of Human Lung Alpha-1-Proteinase lnhlbltor'?

DAVID A. JOHNSON, MARK W. FRAMPTON, R. STEVE WINTERS, PAUL E. MORROW, and MARK J. UTELL Introduction

Nitrogen dioxide (N0 2 ) is a major air pollutant that has been shown to produce emphysematouslike lesions in the lungs of animals (1-4). Although the pathogenic processes resulting in these lesions remain obscure, the mechanisms involved may be similar to those thought to cause emphysema in smokers. The hallmark of emphysema is the degradation of the elastin-rich alveoli, which is thought to result from uncontrolled proteolysis by endogenous elastases. Human at-proteinase inhibitor (at-PI) is primarily responsible for controlling the activity of neutrophil elastase (5), and genetic deficiency of this inhibitor in humans has been shown to result in the early onset of emphysema (6). Oxidants in cigarette smoke wereshown to inactivate at-PI (7), and chemical oxidation of the methionine residue within the active site of the at-PI molecule greatly reduces the inhibitory activity of at-PI (8). Such data have led to the hypothesis that the inhalation of oxidants could result in emphysema by causing a lung-localized deficiency of active at-PI. The literature contains reports supporting and refuting this hypothesis (9) with regard to cigarette smoke. Little data are available examining whether exposure to oxidant air pollutants alters the activity of at-PI. In vitro exposure of pure at-PI to 75 ppm N02 or to 1 mM H 202 did not result in decreased elastase inhibitory activity,whereas inhibitory activity was diminished when a mixture of at-PI and H 202 was exposed to increasing amounts of N02 (10). In an in vivo study of humans, N02 reportedly caused a 45070 decrease in the functional activity of at-PI in bronchoalveolar lavage fluids (BALF) 3 h after exposure to 3 or 4 ppm for 3 h, with intermittent exercise (11). These results suggest that N02 exposure may compromise the activity of at-PI in the lung, which could allow neutrophil elastase to degrade alveolar elastin. Tofurther examine the apparent reduc758

SUMMARY Healthy, nonsmoking human volunteers were exposed to environmentally relevant concentrations of N0 2 followed by bronchoalveolar lavage(BAL)to study whether N0 2 exposure decreases the functional activity of alpha-l-proteinase inhibitor (a1-PI) in the lung. Two3-h exposure protocols with intermittent exercise were employed and BAL was performed 3.5 h after exposure. The first exposure protocol with nine subjects involved three 2-ppm "peaks" with a 0.05 ppm background, whereas the second protocol with 15 subjects was a continuous exposure to 1.5 ppm N0 2 • All subjects were randomly exposed to either air or N0 2 , with at least a 2-wk interval between treatments, and the BAL fluids obtained after air exposure served as the controls. The BAL fluids were analyzed for a1-PI elastase inhibitory activity, the immunologic concentration of a1-PI, total protein, and albumin. The ratio of a1-PI activity to its immunologic concentration was taken as the functional activity of a1-PI, and possible changes in the amount of a1-PI in the lung were assessed by examining the ratio of the immunologic concentration of a1-PI to total protein. Neither of the N0 2 exposure protocols resulted in a decrease in the functional activity of a1-PI, nor were there alterations in the immunologic levels of a1-PI. These data suggest that short-term exposures to low levels of N0 2 do not result in a lung-localized deficiency of active a1-PI, which has been hypothesized to be a contributing factor in the pathogenesis of emphysema. AM REV RESPIR DIS 1990; 142:758-762

tion of at-PI activity after inhalation of N02 reported by Mohsenin and Gee (11),

we determined functional and immunologic levels of at-PI in BALF using two environmentally relevant patterns of N02 delivery and total exposure. We also incorporated a randomized exposure to filtered air to serve as a control for each subject. Methods Subjects All subjects in these studies werehealthy, nonsmoking volunteers 19 to 37 yr of age who met the following requirements: no pulmonary disease by history and physical examination, no present or past history of smoking, absence of upper respiratory illness for at least 6 wk prior to study, normal pulmonary function tests, and absence of airway hyperreactivity in response to inhaled carbachol (12). Informed consent was obtained from all subjects, and the studies were approved by the Committee on Investigations Involving Human Subjects at the University of Rochester.

N02 Exposure Exposures wereperformed in a 45-m3 environmental chamber in the Clinical Research Center at the University of Rochester. Characteristics of this facility have been described previously (13). Pulmonary function testing equipment, bicycle ergometer, and subject

monitoring equipment werehoused within the main exposure room. For these studies, temperature and relative humidity within the chamber were maintained at 22 ± 10 C and 30 ± 5070, respectively. Nitrogen dioxide generation and monitoring have been described previously (14). A high rate of atmosphere turnover within the chamber (approximately 0..3 air changes per minute) enabled N0 2 levels to reach 90070 or greater of steady state value within 4 min. Concentra(Received in originalform December 29, 1989 and in revised form April 6, 1990) 1 From the Department of Biochemistry, James H. Quillen College of Medicine, East Tennessee State University, Johnson City, Tennessee, and the Pulmonary and Critical Care Unit, University of Rochester School of Medicine and Dentistry, Rochester, New York. 2 Research described in this report was conducted under contract to the Health Effects Institute (HEI), an organization jointly funded by the U.S. Environmental Protection Agency (EPA) (Assistance Agreement X-812059) and automotive manufacturers. The contents of this report do not necessarily reflect the views of the HEI, nor do they necessarily reflect the policies of the EPA or of the automotive manufacturers. Additional support for this work was provided by Grant No. ROI-ES-02679 from the National Institutes of Health. 3 Correspondence and requests for reprints should be addressed to David A. J ohnson, Ph.D., Department of Biochemistry, College of Medicine, East Tennessee State University, Johnson City, TN 37614-0002.

NITROGEN DIOXIDE EXPOSURE AND HUMAN WNG Q·l·PROTEINASE INHIBITOR

tions of N0 2 at the 3-foot and 6-foot levels within the chamber varied by no more than 5% of the mean. Nitrogen dioxide levelswere monitored continuously using the EPA Reference Standard NO/NO x Analyzer (Model 8840;Monitor Labs, San Diego, CAl, calibrated by standardized, National Bureau of Standards-traceable N0 2 and NO gases, and verified by a colorimetric method (15).

Exposure Protocols Each subject wasexposed for 3 h to either N02 or clean filtered air, the exposure being administered in a double-blind randomized fashion. The alternate exposure (air or N0 2 ) followed the first exposure by at least 2 wk. The first or "peaks" protocol consisted of a baseline level of 0.05 ppm N0 2 interspersed with three 15-min"peak" levelsofN0 2 at 2.0 ppm. The second or "1.5 ppm" protocol consisted of continuous exposure to 1.5 ppm N0 2 for 3 h. All subjects exercised for 10min, approximately every 30 min, on the bicycle ergometer at a previously determined level sufficient to quadruple resting ventilation. During the "peaks" exposures, the lO-min exercise periods occurred during the higher N0 2 concentrations, thus maximizing total intake. Minute ventilation was monitored at rest and during exerciseusing inductive plethysmography; figure I summarizes the exposure protocols. Bronchoalveolar Lavage Three and one-half hours after the exposure, bronchoalveolar lavage (BAL) was performed as described previously (14). In brief, after premedication with atropine and administration of aerosolized 2% lidocaine to the upper airway,a flexible fiberoptic bronchoscope (Pentax FB-19H)was inserted orally and gent1y wedged in a subsegmental airway of the inferior segment of the lingula. Three 50-ml aliquots of sterile normal saline were sequentially instilled and immediately withdrawn under gentle suction; the position of the bronchoscope was not altered until the lavage procedure was complete. Lavage return averaged 65010 of volume instilled. The same lingular subsegment was entered for each of the two lavages for each subject. The fluid retrieved was immediately trans-

2 ppm NO, PEAKS n •

9

n =

15

0 .05 ppm

ppm

NO

10 MIN . EXERCISE PERIODS

o

1

2

TIME (hrs)

3

ported to the laboratory for processing, where it was filtered through four layers of sterile gauze to removedebris and mucus, before centrifugation at 500 g for 20 min to remove cells. The supernatant fluid was then stored at - 70° C for subsequent analysis. The storage intervals of the samples were 2.3 yr to 3 yr for the "peaks" protocol and 0.72 yr to 1.1 yr for the 1.5 ppm continuous protocol. Total cell counts were obtained from a dilution of the resuspended cell pellet using a hema cytometer, viability was determined by exclusion of trypan blue, and differential cellcounts were performed on cytospin smears stained with Diff-Quikf (American Scientific Products, McGraw Park, IL).

759

wellto stop the reaction, and the absorbance was read at 405 nm in a microtiter plate reader after blanking the instrument on wellscontaining only buffer and substrate. Elastase control wells contained elastase, buffer, and substrate, but no BALF. Each lavage fluid sample was assayed in triplicate, and the results were averaged. Because at -PI and elastase react 1:1 on a molar basis, the amount of active at-PI was calculated from the amount of elastase inhibited when compared with the controls. Immunoassay ofa,.PI. The immunologic concentration of at-PI in the HALF samples was quantified in an enzyme-linked immunoassay of BALF proteins bound to nitrocellulose. We have shown (18)that the sensitiviLavage Fluid Determinations ty, specificity, and quantitation of such asElastase inhibition assays. Active at-PI con- says are improved by pretreatment of the centrations were determined by elastase inhi- protein samples with sodium dodecyl sulfate bition assays performed in 96-well flat-bottom (SDS) and boiling for 2 to 3 min. On the bamicrotiter plates. When assaying the BALF sis of these findings, samples and standards from the 2 ppm "peaks" protocol, the sample were diluted to 20 ug/rnl of total protein, and volume added to each well was 75 or 100 Ill. SDS was added to a final concentration of However, the lower inhibitory activity of the 10 ug/ml. Pure human at -PI was used as the 1.5 ppm protocol samples required using 150 standard in these assays. After boiling, the IIIof sample in each well. The volume of each cooled samples and standards were blotted well was then adjusted to 150 IIIwhen assay- on nitrocellulose in a Bio-Dot 96-well apparaing the 2 ppm "peaks" samples and 175 III tus with 3 mm wells (Bio-Rad Laboratories, when assaying the 1.5 ppm protocol samples Richmond, CAl. The nitrocellulose memwith an appropriate concentration of buffer brane was blocked with 1% BSA for 4 hand to give a final assay buffer concentration of soaked in a 1:1,000 dilution of the primary 0.1M HEPES, I mM CaCh, 0.01% Tween@ 20 anti-human aI-PI antibody (Atlantic Antiat pH 7.5 containing 50 IIg HSA/ml. In both bodies , Stillwater, MN) for 8 h. After washassays the amount of remaining elastase ac- ing, the membrane was incubated overnight tivity ranged from 30 to 70%, which is known in a 1:2,000dilution of peroxidase-linked goat to yield accurate measurements of inhibitory antirabbit IgG (Bio-Rad, Richmond, CA; activity. blotting grade). Detection was accomplished Chromatographically purified porcine pan- by reaction of the bound peroxidase with creatic elastase (PPE) (Elastin Products, H 202 and 4-chloro-I-naphthol, which yieldOwensville, MO) was dissolved in the 0.1 M ed a blue, insoluble product where the at-PI HEPES buffer to a concentration of 10.99 antigen was bound. Because the amount of nmol/ml (285 ug/ml) of active enzyme based enzyme in each spot was proportional to the on inhibition titration assays with 100% ac- amount of at-PI bound, the intensity of the tive human at-PI (16). This stock solution was dots could be used to quantitate the at-PI. diluted to 55 or 58 pmol/ml in 0.1 M HEPES Standards of pure at-PI were blotted onto assay buffer. Elastase control and sample wells each membrane along with samples . Quantieach received 50 IIIof the freshly diluted PPE tation was achieved by photographing the desolution. In the assays of the 2 ppm "peaks" veloped nitrocellulose membrane followed by samples , each well contained 2.75 pmol of densitometric scanning of the photographic elastase, whereas in assays of the 1.5ppm pro- negative. Computer-integrated areas from the tocol samples each well contained 2.9 pmol scans of the standards yielded standard curves of PPE. The mixture was allowed to incu- from which the concentration of samples were bate for 90 min for the enzyme-inhibitor reac- read based on their areas. Functional activity ofa,.PI. The functional tion to reach equilibrium. Remaining active elastase after incubation activity of at-PI was determined by dividing with the 2 ppm "peaks" protocol samples was the concentration of active at-PI in the BALF measured by adding to each microtiter well (ug/ml) by its immunologic concentration 50 IIIof the Succinyl-(Ala).-p-nitroanalide (17) (ug/ml), Theoretically, this ratio, which is resubstrate (Sigma Chemical, St. Louis, MO), ferred to as ACTIVE/IMMUNO, should which was freshlydiluted from a 2O-mM stock equal 1.0 for fully active at-PI. in dimethylsulfoxide into the 0.1 M HEPES Total protein. Total protein was measured assay buffer. The volume of substrate added by the method of Lowry and coworkers (19) was 25 III for the inhibition assays of the 1.5 using crystalline bovine serum albumin as the ppm protocol samples, but the final substrate standard. Albumin concentrations were meaconcentration was0.4 mM in both assays. The sured using a modified enzyme-linked imfinal assay volume for the 2 ppm "peaks" sam- munosorbent assay (ELISA), employing apples was 200 III,and for the 1.5ppm protocol propriate commercially obtained polyclonal samples it was 250 Ill. After 30 to 45 min, antibodies (Organon Teknika-Cappel, Mal10IIIof glacial acetic acid were added to each vern, PAl. In brief, the samples to be tested

760

JOHNSON, FRAMPTON, WINTERS, MORROW, AND UTELL

were absorbed onto 96-well microtiter plates. After extensive washing and blocking with heat-inactivated sheep serum, the samples were incubated with the appropriate primary antibody. After additional washes, the plate was then incubated with a 1:1,000 dilution of goat antimouse antibody conjugated with biotin. The amount of antibody bound was detected by addition of avidin-conjugated alkaline phosphatase and its substrate, p-nitrophenyl-phosphate. Change in optical density was detected on a Bio Teck ELISA plate reader at 405 nm ;

Statistical Analyses All determinations on BAL flu id were performed concurrently on both the air and NO] exposure samples for each subject, and the investigators performing the analyses were blinded to the exposure. Data were analyzed by comparing the data from NO] exposures with the air-exposed controls, using the paired t test . This test was chosen because the same subjects were used for controls and exposures. The 95070 confidence level (p < 0.05) was taken as the significance limit. Results

The first protocol involved the exposure of nine nonsmoking volunteers (sixmale, three female with a mean age of 25 yr) to either air or 0.05 ppm N0 2 with three I5-min N0 2 "peaks" at 2.0 ppm. The second protocol involved 15 subjects (12 male, three female with a mean age of 24 yr) who were exposed continuously to 1.5 ppm N0 2 • All exposures were 3 h in duration, and the subjects exercised intermittently to increase the ventilation rate approximately fourfold. Each subject was exposed to both air and N0 2 , separated by at least 2 wk. The average time between exposures was 21.5 days (range, 14 to 75 days) for the 1.5 ppm protocol and 47 days (range, 14 to 138 days) for the "peaks" protocol. There were no statistical differences in the concentration of protein or numbers of cells in the two BAL samples irrespective of whether the exposure was air or N0 2 • No differences were seen between air and N0 2 exposure for total fluid recovery or cell concentration (table 1). Similarly, no differences in cell differential counts were observed (0/0 alveolar macrophages: 93.3 ± 1.2after air versus 93.6 ± 1.2 after 1.5 ppm N0 2 exposure; 95.1 ± 1.3 after air versus 94.7 ± 0.7 after 2.0 ppm "peaks" exposure) . The findings from analysis of at-PI immunologic concentration and activity in BAL fluid are shown in table 1.No significant differences in the concentration or the activity of acPI wereseen between air and N02 exposure with either exposure protocol. Since the principal measure of

TABLE 1 ALPHA-1·PROTEINASE INHIBITOR IN BALF SAMPLES FROM NO,·EXPOSED SUBJECTS' BALF Volume (m/)

BALF Cells (x 70 s/m/)

Albumin (Ilg/m/)

Total Protein (Ilg/m/)

Active a.-PI (Ilg/m/)

Immuno a.·PI (Ilg/m/)

"Peaks," n = 9 Air NO,

106.2 ± 12 103.9 ± 18

1.42 ± 0.20 1.26 ± 0.13

80.7 ± 9.4 77.8 ± 5.6

128.6 ± 56 126.2 ± 40

0.81 ± 0.08 0.80 ± 0.08

1.18 ± 0.63 1.02 ± 0.28

"1.5 ppm ," n = 15 Air NO,

113.2 ± 14 115.9 ± 8

1.38 ± 0.20 1.27 ± 0.14

112.2 ± 12.8 102.5 ± 9.1

133.5 ± 46 124.8 ± 34

0.42 ± 0.06 0.36 ± 0.05

1.26 ± 0.93 0.89 ± 0.60

Treatment

• The data presented are mean :!: SEM for each 9rouP. The same subjects were exposed to air or NO, in a random fashion with a 2-wk period between exposures.

interest wasthe ratio of activeto immunologic at-PI, the ACTIVE/IMMUNO data from both protocols are shown in figure 2. Although a ratio of 1.0 indicates fully active at-PI, it averaged 0.82 for both air and N0 2 with the 2 ppm "peaks" protocol. These ratios were lower for the 1.5ppm protocol, with means of 0.41 for the air exposures and 0.55 for the N0 2 exposures. Afford and coworkers (20) have recently reported that the activity of at-PI in BALF decreases with time, even when frozen. The present results do not support their findings since the "peaks" protocol samples were stored for three times as long as the 1.5 ppm protocol samples. Irrespective of why these ACTIVE/IMMUNO ratios were less than 1.0, the conclusion that N0 2 had no effect on the ratios relative to air-exposed controls is clear. In the 1.5 ppm protocol, which had the lowest ratios, the mean values for the N0 2-exposed samples were actually higher than the air-exposed samples. Although it might appear that the differences in the means result from a few higher values in the N0 2-exposed group, it is worth noting that the ACTIVE/IMMUNO ratios were higher 0

1 .4 D

Z

1.2

::s

1.0

::l

~

Ul

E=:

U

-e

• S·s

0 .8 0.6

>

•

0.4 0 .2

• • •

9

,.s •••

•

•

1S

D •

for 10 of the 15 subjects after N0 2 exposure relative to the control values with air. The possibility that the immunologic level of at-PI might be changed by exposure was examined by analyzing the immunologic concentration of at-PI in BALF relative to the concentration of protein. These data are presented in figure 3A, where the immunologic concentration of at-PI relative to the total protein concentration is given as IMMUNOI PRarEIN. Similar results were obtained with both the 2 ppm "peaks" and the 1.5 ppm protocols, and there were no significant differences between air and NOrexposed samples. To confirm that the at-PI was less active in both the air and NOrexposed samples from the 1.5 ppm protocol, the ratios of active at-PI concentra-

•• • •

• •••

.:1

ZW •1•

••• ••

0.0 AIR

NO ,

AIR l.S

NO, p pm

Fig. 2. The functional activity of a,-PI in BALF,expressed as the ratio of the concentration of active inhibitor to its immunologic concentration (ACTlVEIIMMUNO) are shown in a scatter graph . The AIR and NO, columns represent the data from the same subjects after differ· ent exposures. The two different exposure protocols are designated as 2 ppm "peaks"and 1.5ppm , as described. Horizontal bars are drawn at the means for each group.

2 ppm · pe aks·

1.5

ppm

Fig. 3. A. Bar graphs show the immunologic concen tration of a,-PI relative to the total protein concentration (IMMUNO/PRarEIN). Error bars represent the standard error of the mean. B. Bar graphs show the concentration of active a,-PI relative to the concentration of total protein (ACTIVEIPRarEIN) in the BALF. Error bars lndicate the standard error of the mean .

NITROGEN DIOXIDE EXPOSURE AND HUMAN WNG a-1-PROTEINASE INHIBITOR

tions to total protein concentrations were examined, and these ratios are referred to as ACTIVE/PROTEIN. These data, along with those from the 2 ppm "peaks" protocol, are shown in figure 3B. Although these ratios were lower for the 1.5 ppm protocol, there were no significant differences between air and N0 2 exposures within each protocol. Discussion

One possible mechanism of injury by inhaled oxidant pollutants such as N0 2 is alteration in protective enzyme systems. One of the most important of these is at-PI, which provides the predominant elastase inhibitory function in pulmonary tissue (21). The clinical observation that genetic deficiency of at-PI results in emphysema has served as a confirmatory experiment of nature. Environmental influences that decrease the function or availability of this enzyme might permit unopposed proteolysis of lung tissue. In addition, at-PI and other proteinase inhibitors may modulate pulmonary immune responses (22, 23). The recent report by Mohsenin and Gee (11) that exposure to N0 2 at 3 or 4 ppm decreases functional activity of at-PI in normal volunteers fuels the concern that chronic exposures to ambient levels of this pollutant could contribute to the development of emphysema or alter host defense against infection. In two separate protocols, we observed no effect of N0 2 exposure on the functional activity of at-PI. These findings contrast with those of Mohsenin and Gee (11), who examined activity in BALF from subjects exposed to 3 or 4 ppm N02. As in the present studies, subjects exercised intermittently during exposure, and BAL was performed 3.5 to 4 h after exposure. These investigators found that at-PI functional activity (as a percentage of theoretical activity) decreased 45070 in the BALF from N0 2-exposed subjects relative to a separate group of air-exposed subjects. The use of higher levels of N02 in the Mohsenin and Gee exposures may in part explain the differing observations. We calculated the product of N0 2 concentration x time of exposure x minute ventilation for the present studies as well as for the Mohsenin and Gee exposures in order to estimate the total intake of N0 2. The subjects of Mohsenin and Gee exposed to 3 ppm N0 2 inhaled an estimated 15.8 mg, and subjects exposed to 4 ppm inhaled 23.5 mg of N0 2 during the exposure period. In contrast, subjects in our 2.0 ppm "peaks" protocol inhaled an estimated 5.1 mg, and sub-

jects in the 1.5 ppm protocol inhaled 8.2 mg of N0 2. However, in the present studies there was in fact not even a trend toward loss of activity of at-PI in response to N0 2 exposure; in the 1.5 ppm N02exposures enzyme functional activity was slightly higher in relation to immunologic concentration when compared with air exposure (figure 2). The difference in findings between the present studies and those of Mohsenin and Gee may also be explained by the potential variability in measurements of at-PI activity. Considerable variability in the activity of at-PI in BALF relative to its immunologic concentration has been reported by other investigators (24-27), with mean values ranging from 39 to 125070. Indeed, in the present studies we observed an unexpected difference in activity of at-PI between the two exposure groups. Although immunologic concentrations were comparable in the two protocols, activity of at-PI in the 1.5-ppm protocol after both air and N0 2exposure was approximately half that observed in the "peaks" exposure protocol (table 1; p < 0.0001, for differences between protocols after air exposure). The reason for this variability is unknown, but it may be related to differences in technical aspects of sample handling. The use of unconcentrated BALF samples, as in the present study, may have contributed to the variability of our data relative to the data of others using concentrated samples. The use of unconcentrated samples was expected to reduce variability since BALF concentration by ultrafiltration, as performed in the Mohsenin and Gee study, has been shown to result in variable recovery of at-PI protein and activity (28). Another difference between our study and those of other groups is that we used a dot-blot immunoassay to measure the total amount of at-PI, whereas others, including Mohsenin and Gee (11), used radial immunodiffusion to quantitate at-PI. These difficulties, as well as the potential for variability between different groups of subjects, emphasize the importance of using each subject as his or her own control and performing assays on BALF obtained after pollutant and control exposures simultaneously and in a blinded fashion. It is not clear from the report by Mohsenin and Gee whether BALF samples obtained after air and N0 2 exposure were analyzed concurrently. In conclusion, we found no evidence that short-term, low-level N0 2inhalation results in a lung-localized deficiency of active at-PI in healthy volunteers. Our

761

failure to detect any inactivation of at-PI is consistent with the finding of Dooley and Pryor (10) that bubbling solutions of pure at-PI with 75 ppm of N0 2 for 2 h did not inactivate the inhibitor unless H 20 2 was included in the solution. In contrast, exposure to 0.60 ppm N0 2 for 3 h has been found to alter immunologic levels of another protease inhibitor, a2-macroglobulin in BALF; however, little effect was observed after inhalation of 1.5 ppm N0 2 (29). Clearly, further studies are needed to determine whether repeated N0 2exposures alter the inhibition of protease activity in the human lung. Acknowledgment Appreciation is expressed for the technical assistance of Mrs. Diane Earwood, F. Raymond Gibb, and Donna Speers, and the help of Mrs. Raymonde Cox in preparing the manuscript.

References 1. Freeman G, Crane SC, Stephens RJ, Furiosi NJ. Pathogenesis of the nitrogen dioxide-induced lesion in the rat lung: a review and presentation of new observations. Am Rev Respir Dis 1968; 98: 429-43. 2. Kleinerman J, Ip MPC. Effects of N0 2 on elastin and collagen content in the lung. Arch Environ Health 1979; 34:228-32. 3. Glasgow JE, Pietra GG, Abrams WR, Blank JB, Oppenheim DM, Weinbaum G. Neutrophil recruitment and degranulation during induction of emphysema in the rat by nitrogen dioxide. Am Rev Respir Dis 1987; 135:1129-36. 4. Evans MJ. Oxidant gases. Environ Health Perspect 1984; 55:85-95. 5. Beatty K, Bieth J, Travis J. Kinetics of association of serine proteinases with native and oxidized u-l-proteinase inhibitor and a-l-antichymotrypsin. J BioI Chern 1980; 255:3931-4. 6. Laurell CB, Eriksson S. The electrophoretic u-l-globulin pattern of serum in n-l-antitrypsin deficiency. Scand J Clin Lab Invest 1963; 15:132-40. 7. Carp H, Janoff A. Possible mechanisms of emphysema in smokers: in vitro suppression of serum elastase-inhibitory capacity by fresh cigarette smoke and its prevention by antioxidants. Am Rev Respir Dis 1978; 118:617-21. 8. Johnson D, Travis J. The oxidative inactivation of human u-l-proteinase inhibitor. J BioI Chern 1979; 254:4022-6. 9. Janoff A, Chan S-K. Is ai-proteinase inhibitor inactivated by smoking? Science 1984; 224: 755-6. 10. Dooley MM, Pryor WA. Free radical pathology: inactivation of human a-I-proteinase inhibitor byproducts from the reaction of nitrogen dioxide with hydrogen peroxide and the etiology of emphysema. Biochem Biophys Res Commun 1982; 106:981-7. 11. Mohsenin V, Gee JBL. Acute effect of nitrogen dioxide exposure on the functional activity of ai-proteinase inhibitor in the bronchoalveolar lavage of normal subjects. Am Rev Respir Dis 1987; 136:646-50. 12. Utell MJ, Swinburne AJ, Hyde RW, Speers DM, Gibb FR, Morrow PE. Airway reactivity to nitrates in normal and mild asthmatic subjects. J Appl Physiol 1979; 46:189-96. 13. Utell MJ, Morrow PE, Hyde RW,Schreck RM.

762 Exposure chamber for studies of pollutant gases and aerosols in human subjects: design considerations. J Aerosol Sci 1984; 15:219-21. 14. Frampton MW, Smeglin AM, Roberts NJ Jr, Finkelstein IN, Morrow PE, Utell NJ. Nitrogen dioxide exposure in VIVO and human alveolar macrophages inactivation of influenza virus in vitro. Environ Res 1989; 48:179-92. 15. Hartwell TD, Clayton CA, Decker CE, Hunt PN. Comparability of nine methods for monitoring N0 2 in ambient air. Research Triangle Park, NC: Environmental Protection Agency, 1974. EPA Publication No. 650/4-72-012, 11-85. 16. Travis J, Johnson D. Human at-proteinase inhibitor. Methods Enzymol 1981; 80:754-71. 17. Bieth J, Spieth B, Wermuth CG. The synthesis and analytical use of a highly sensitive and convenient substrate of elastase. Biochem Med 1974; 11:350-7. 18. Smith CE, Musich PR, Johnson DA. Sodium dodecyl sulfate enhancement of quantitative immunoenzyme dot-blot assays on nitrocellulose. Anal Biochem 1989; 177:212-9. 19. Lowry OH, Rosebrough HJ, Farr AL, Ran-

JOHNSON, FRAMPTON, WINTERS, MORROW, AND UTELL

dall RJ. Protein measurement with the Folin phenol reagent. J BioI Chern 1951; 193:265-75. 20. Afford SC, Burnett D, Campbell EJ, Cury JO, Stockley RA. The assessment of alpha-l-proteinase inhibitor form and function in lung lavage fluid from healthy subjects. BioI Chern Hoppe Seyler 1988; 165:1065-74. 21. Gadek JE, Fells GA, Zimmerman RL, Rennard SI, Crystal RG. Antielastases of the human alveolar structures: implications for the proteaseantiprotease theory of emphysema. J Clin Invest 1981; 68:889-98. 22. Ades EW, Hinson A, Chapuis-Cellier C, Arnaud P. Modulation of the immune response by plasma proteinase inhibitors. I. Alphas-macroglobulin and alpha-antitrypsin inhibit natural killing and antibody-dependent cell-mediated cytotoxicity. Scand J Immunol 1982; 15:109-13. 23. James K. Alpha, macroglobulin and its possible importance in immune systems. Trends Biochern Sci 1980; 5:43-7. 24. Bridges RB, Kimmel OA, Wyatt RJ, Rehm SR. Serum antiproteases in smokers and nonsmokers. Am Rev Respir Ois 1985; 132:1162-9.

25. Lellouch J, Claude J-R, Martin J -P, Orssaud G, Zaoui D, Bieth JG. Smoking does not reduce the functional activity of serum alpha-l-proteinase inhibitor. Am Rev Respir Dis 1985; 132:818-20. 26. Abboud RT, Fera T, Richter A, Tabona MZ, J ohal S. Acute effect of smoking on the functional activity of alpha-l-protease inhibitor in bronchoalveolar lavage fluid. Am Rev Respir Dis 1985; 131:79-85. 27. Stone PJ, Calore JO, McGowan SE, Bernardo J, Snider GL, Franzblau C. Functional at-proteinase inhibitor in the lower respiratory tract of cigarette smokers is not decreased. Science 1983; 221: 1187-9. 28. Afford SC, Stockley RA, Kramps JA, Dijkman JH, Burnett O. Concentration of bronchoalveolar lavage by ultrafiltration; evidence of differential protein loss and functional inactivation of proteinase inhibitors. Anal Biochem 1985; 151:125-30. 29. Frampton MW, Finkelstein IN, Roberts NJ Jr, Morrow PE, Utell MJ. Effects of nitrogen dioxide exposure on bronchoalveolar lavage proteins in humans. Am J Respir Cell Mol BioI 1989; 1:499-505.