Mechanisms of mineral dust-induced emphysema.

Mineral dust exposure can result in emphysema and chronic airflow obstruction. We postulated that dust-induced emphysema has a pathogenesis similar to that in cigarette smoke-induced emphysema, namely, excess release of proteolytic enzymes from dust-evoked inflammatory cells, and inactivation of alpha-1-antitrypsin (A1AT) by dust-catalyzed formation of oxidants. To test this theory we examined the antiproteolytic activity of A1AT exposed to quartz in vitro and found that it was decreased in a dose-response fashion. Catalase prevented this effect, which suggested that it was mediated by quartz-generated hydrogen peroxide. We also showed that a variety of dusts could oxidize methionine to methionine sulfoxide in vitro, using either pure amino acid or whole protein. The relative order of activity was coal > quartz > titanium dioxide. Lastly, we used a new high-performance liquid chromatography technique to demonstrate that quartz, coal, and titanium dioxide produced connective tissue breakdown in rat lungs, as determined by the appearance of desmosine and hydroxyproline in lavage fluid after dust instillation. On a particle-for-particle basis, the order of dust potency was similar to that for methionine oxidation. Connective tissue breakdown was associated with elevations of both polymorphonuclear leukocytes and macrophages in lavage fluid, and it is unclear whether one or both of these types of inflammatory cell mediates this process. These observations support our theory that dust-induced emphysema and smoke-induced emphysema occur through similar mechanisms.


Introduction
Occupational exposure to mineral dusts, particularly coal and silica, is associated with the development of emphysema (1)(2)(3)(4)(5)(6) and chronic airflow obstruction. Although the anatomic presence of dust-induced emphysema is established, the mechanisms of emphysema formation are uncertain.
The protease-antiprotease theory of cigarette smoke-induced emphysema (7,8) suggests that in smokers, emphysema develops as a result of excess unopposed proteolytic activity in the lung. This excess proteolytic activity has two sources: a) release of proteolytic enzymes, particularly neutrophil elastase, from smoke-evoked inflamtnatory cells, with subsequent degradation of the connective tissue matrix of the lung; and b) inactivation of the major anti-proteolytic agent in the lung, alpha-1-antitrypsin (AlAT), by oxidants in the smoke.
Because mineral dusts also evoke a more or less intense and prolonged inflammatory response (9), and because many mineral dusts can catalyze the formation of active oxygen species (AOS) in aqueous media (10,11), we propose that dustinduced emphysema arises on a very similar basis, namely, release of large amounts of proteolytic enzymes from dust-evoked inflammatory cells, and oxidative inactivation of AIAT. This question has not been This paper is based on a presentatiorl at The Sixth International Meeting on the Toxicology of Natural and Man-Made Fibrous and Non-Fibrous Particles held 15-18 September 1996 (12) showed that there were increases in lavage proteolytic activity from rats given quartz or titanium dioxide, but Hannothiaux et al. (13) found only small and inconsistent increases in lavage elastase activity from monkeys administered quartz. Rom (14) reported increased levels of elastase complexed to AlAT in lavages from coal miners.
In this paper we report a series of experiments designed to explicitly test our The antiproteolytic activity of AlAT was determined in vitro by the SLAPN assay using porcine pancreatic elastase (PPE) (16) and levels of hydrogen peroxide generated by quartz by a modification of Sigma method P8250 (16). Methionine sulfoxide levels were assayed in vitro by high-performance liquid chromatography (HPLC) (17). When used with pure amino acids, the solutions were analyzed directly; when used with whole proteins, the proteins were first hydrolyzed in potassium hydroxide (17).
For analysis of lavage desmosine (DES) and hydroxyproline (HP), lavage was performed with distilled water to avoid analytical problems caused by high salt concentrations. The lavageate was hydrolyzed in 6 M hydrochloric acid. Lavage DES and HP analyses were carried out using HPLC methods (15).

Results and Discussion
Oxidative Inactivadon ofAlAT by Quar Figure 1 shows the effects of a 1-hr exposure to quartz on the antiproteolytic potency of pure human AlAT in vitro. Quartz inactivated AIAT in a dose-response fashion, and the degree of inactivation was greater for freshly ground quartz compared to quartz used directly as received (aged quartz). In Figure 2, a single concentration of AIAT was used and catalase or no catalase added to the quartz-AlAT mixture to investigate the role of hydrogen peroxide. Catalase protected AlAT against the effects of all quartz concentrations and against both fresh and aged quartz. Hydrogen peroxide assay showed that quartz liberated hydrogen peroxide into the medium, with greater amounts of hydrogen peroxide produced by fresh compared to aged quartz. Reagent hydrogen peroxide also inactivated AIAT and this effect could be similarly reduced by inclusion of catalase in the incubation mixture (data not shown).
To determine whether quartz interfered with formation of the AlAT-PPE complex, AlAT was treated with quartz or hydrogen peroxide, then mixed with PPE. The supernatants from this mixture were analyzed by sodium dodecyl sulfate gel electrophesis and Western blots prepared using anti-AlAT. This approach showed that the intensity of the band representing the AlAT-PPE complex was diminished by both quartz and hydrogen peroxide treatment.
These observations indicate that quartz can inactivate AIAT and decrease the formation of AIAT-PPE complexes by a mechanism that appears to be mediated largely through hydrogen peroxide formation. It should be noted that the concentration of quartz required is fairly high and that the system by definition ignores the Figure 2. Catalase provides protection against silicainduced inactivation of AlAT in an elastase inhibitory capacity assay similar to that shown in Figure 1. However, for this experiment only one concentration of AlAT was used. *, inhibitory capacity is significantly reduced by quartz compared to no dust treatment (equal to bars labeled AlAT). Catalase protects against all quartz concentrations. Abbreviations: A20, 20 mg/ml aged quartz; A200, 200 mg/ml aged quartz; F20, 20 mg/ml freshly ground quartz; F200, 200 mg/ml freshly round quartz. Reproduced from Zay et al. (15), by permission.
protective effects of proteins that might adsorb to the dust in vivo. Nonetheless, the local concentration of mineral dusts and subsequent levels ofAOS formation in vivo conceivably could be quite high, and the current observations suggest that any dust that can generate AOS in solution may have the same effect.

Oxidation ofMedbionine Residues by Mineral Dusts
As noted in the introduction, it is believed that cigarette smoke oxidatively inactivates AlAT in the lungs of smokers (7,8,18); the specific mechanism involved appears to be oxidation of methionine residues that are crucial to the protective function of AIAT (18). Proof of this proposition has been difficult, although Carp et al. (18) did report the presence of increased methionine sulfoxide (the oxidation product of methionine) levels in the lavage of smokers. Evans and Pryor (19) addressed this issue by exposing AIAT in vitro to aqueous extracts of smoke tar. They found that smoke tar was able to oxidize methionine to methionine sulfoxide when methionine was supplied as pure amino acids or in AIAT, and that the process appeared to be mediated by hydrogen peroxide.
We performed a series of similar in vitro experiments using the dusts listed in "Materials and Methods" and either pure methionine or whole protein (human AlAT or human albumin). All the dusts listed were able to oxidize pure methionine (Figure 3), as was reagent hydrogen peroxide. This process was dust dose dependent and could be reversed to a great extent by catalase and to a variable extent by scavengers of hydroxyl radical. Coal, silica, and amosite asbestos were also able to oxidize methionine residues in whole proteins (data not shown). The relative order of potency was coal > silica > amosite.
These mineral dust findings are quite similar to those of Evans and Pryor (19) with cigarette smoke and suggest that mineral dusts can potentially inactivate AlAT by oxidizing methionine. The process again appears to be mediated through hydrogen peroxide and less certainly through hydroxyl radical. It is interesting that the relative potency of various dusts appears to parallel their ability to produce emphysema, and it should also be noted that if one uses the mean size data to estimate the relative number of particles present, the differences on a particle-forparticle basis among coal, silica, and titanium dioxide are much more marked than Figure 3 would imply.

Dust-induced Connective lissue Brk wn
By definition, connective tissue breakdown must occur during the genesis of emphysema. In cigarette smokers the demonstration of connective tissue breakdown has been controversial; older studies failed to find increased levels of urinary DES (a marker of elastin breakdown) in smokers compared to nonsmokers. Newer studies, however, which daim to use more sensitive techniques, have reported increased urinary DES levels in smokers (20).
To investigate this issue with dusts, we administered a single intratracheal dose of quartz, coal, or titanium dioxide to Environmental Health Perspectives * Vol 105, Supplement 5 * September 1997 Sprague-Dawley rats, then analyzed DES and HP in lavage fluid at various times after dust administration. One lung was used for the chemical analyses and the other for lavage inflammatory cell counts.
We found that all three dusts produced a dose-response increase in DES and HP at 24 hr, with considerably greater elevations after quartz and coal administration than after titanium dioxide administration. Of particular interest was the time course of events when a single 30-mg dose of dust was administered and animals sacrificed up to 21 days later. Lavage DES returned to control levels within 7 days ( Figure 4A) and lavage HP within 14 days. Lavage polymorphonuclear leukocyte(s) (PMN) counts were briefly elevated but dropped rapidly, whereas lavage macrophage(s) (MAC) counts remained elevated over the 21-day period. With quartz and coal the pattern was quite different: both lavage DES and HP remained markedly elevated over 21 days with no evidence of a decrease over time ( Figure  4B). With quartz, both PMN and MAC counts also remained elevated, whereas with coal the PMN count was briefly elevated and returned to almost control level by 7 days, while the MAC count remained elevated.
These findings are the first report of connective tissue breakdown after mineral dust exposure. It is important to bear in mind that the doses of dust used here are high and that intratracheal instillation can produce enormous local particle concentrations that would never occur during inhalation exposures. Similarly, the intensity of the inflammatory infiltrate is probably much higher than would be seen with inhalation of dust. Thus the level of connective tissue breakdown seen here may be quite artifactual. Nonetheless, even after inhalation exposures mineral dusts induce a marked inflammatory infiltrate, and thus our system, albeit exaggerated, may serve as a reasonable model for dustinduced connective tissue breakdown.
Different dusts behave quite differently. Titanium dioxide is usually regarded as a prototypical nuisance dust. In our system sufficiendy high doses of titanium dioxide produce connective breakdown, but this effect is short lived. Whether this finding applies to humans at any reasonable exposure level is unclear. More important are the findings that coal and silica, dusts that produce emphysema in humans, cause much greater levels of connective tissue breakdown, which persist for long periods of time. It is important to note that, as is true of methionine oxidation, on a particlefor-particle number basis, the effects of coal and quartz are much greater than those of titanium dioxide in producing connective tissue breakdown.
With all three dusts, connective tissue breakdown is accompanied by a fairly intense inflammatory infiltrate, the usual response to any instilled or inhaled dust. However, the correlation of connective tissue breakdown with inflammatory cell counts is unclear. In the silica group PMN levels and MAC levels remain elevated over the entire time course of the experiment, whereas with both coal and titanium dioxide PMN  lime, days Figure 4. Lavage desmosine levels up to 21 days after a single intratracheal instillation of 30 mg (A) titanium dioxide or (8) silica. *, significantly greater than control. Note that DES levels rapidly return to control after titanium dioxide but remain elevated after silica treatment. Reproduced from Li et al. (14), by permission.
remarkably similar. Interpretation of these findings is difficult but it is possible that some other factor (degree of oxidant generation by dusts and inflammatory cells, actual levels of protease release by inflammatory cells) differs among the dusts. We are currently evaluating this possibility.