asbestosis and silicosis

8/21/2019 Asbestosis and Silicosis

1/15

Am J Respir Crit Care Med Vol 157. pp 16661680, 1998

St a t e o f t he Art

Mechanisms in the Pathogenesis ofAsbestosis and Silicosis

BROOKE T. MOSSMAN and ANDREW CHURG

Departments of Pathology, University of Vermont College of Medicine, Burlington, Vermont; and University of

British Columbia, Vancouver, British Columbia, Canada

CONTENTS

1. Clinical and P at hologic Features of Asbestosisand Silicosis

2. Chemical a nd P hysical P roperties of Asbestos and SilicaImportant in Ca usation of Lung D isease

3. The R elationship of Pa rticle Burden to D isease inAsbestosis and Silicosis

4. Cellular a nd Mo lecular M echanisms of A sbestosis

and Silicosis5. Approa ches to P revention and Treatment o f Fibrosis6. Summary a nd C onclusions

Interstitial pulmonary fibrosis caused by the inhalation of as-bestos fibers or silica particles continues to be an importantcause of interstitial lung disease. Although more stringentcontrol of asbestos in the workplace and decreasing industrialuse has contributed t o declines in the preva lence of asbestosisin the United States (1), new cases continue to be identified.Similarly, silicosis is seen among sandblasters, undergroundminers, foundry and quarry workers, and in other dust-exposedtra des (2). B oth diseases, which may ha ve relatively long la-tency periods, are ob served in the clinic today , usually as a re-

sult of high occupational exposures in the past, and they areproblematic in that treatment with corticosteroids and immu-nosuppressants, the usual approaches to therapy for fibroticlung disease, is ineffective (3).

Asbestos and silica are complex, nat urally occurring miner-als that are chemically and physically distinct. Moreover, thepatho logy of asbestosis and silcosis is dissimilar. H owever, t hepatho genesis of these lesions and the major changes in pulmo-nary architecture, namely, the laying down of collagen in aninterstitial location, appear to be similar to many of the fea-tures seen in idiopathic pulmonary fibrosis (IPF). Like IPFand representative animal models of IPF such as bleomycininstillation (4), both asbestosis and silicosis are characterizedby a persistent inflammatory response and generation of pro-

inflammatory and profibrotic mediators.Although asbestosis and silicosis have been studied in-tensely by basic and clinical research scientists, little is knownabout the crucial cellular mechanisms that initiate and drivethe processes of inflammation and fibrogenesis. Many labora-

tories have developed animal and in vitro

models of a sbestosisand silicosis to elucidate the cellular events and properties ofminerals importa nt in disease causat ion. Ot hers have exploredconfounding factors contributing to particulate-induced cellinjury a s well as cellular a nd molecular d efense mechanisms inresponse to these minerals. This information ha s been used tomodulate inflammation and fibrosis in expermental animalmodels in attempts to develop more effective treatment regi-mens for pulmonary fibro ses.

This review will briefly address the clinical and pathologicfeatures of asbestosis and silicosis, and consider the mineral-ogic features of asbestos and silica that may be important indisease causation along w ith confounding facto rs such as coex-posures to smoking an d/or other minera l dusts. The relat ion-ship of particle number, type, and size to disease patterns willbe reviewed. We will then summarize data published withinthe past 5 yr on cellular a nd molecular mechanisms of a sbesto-sis and silicosis and preventive approa ches to t hese diseases inexperimental animal models. Lastly, we emphasize in ourS

UMMARY

AN D

C

ONCLUSIONS

the common mediators and celltypes affected in the pa thogenesis of bo th mineral-related a ndother forms of pulmonary fibrosis and plausible interrelation-ships between the development of fibrosis and lung cancer, a

disease linked to o ccupat ional exposures to a sbestos and pos-sibly to exposures to silica (1, 5).

1. CLINICAL AND PATHOLOGIC FEATURES OFASBESTOSIS AND SILICOSIS

Asbestosis

Asbestosis is defined as bilat eral diffuse interstitial fibrosis ofthe lungs caused by the inha lation o f a sbestos fibers (6). C lini-cally, asbestosis is very similar t o IP F. Most pa tients with w ell-established asbestosis present with shortness of breath a nd drycough, and physical examination typically reveals dry rales atthe bases on inspiration. The usual functional changes in thefully developed case are a restrictive defect and decreaseddiffusing capacity (610). However, it should be noted that

pathologic examination may reveal mild cases that do nothave ob vious function changes. The typical ra diographic find-ing is a lower zone reticulonodular infiltrate on plain films. Inwell-established asbestosis, the CT features appear to be verysimilar if not identical to t hose seen in usual interstitial pneu-monia, i.e., peripheral ba nds, lines, thickened interlobular septa,and honeycombing, with d isease most severe at the lung bases(11, 12).

The clinical, physiologic, and radiologic findings of asbesto-sis are not in any way specific, and they can be seen in diffuseinterstitial fibrosis of other causes, particularly usual intersti-tial pneumonia (idiopa thic interstitial fibro sis), except tha t pa -tients with asbestosis always have a history of heavy occupa-

(

Received in original form July 28, 1997 and in revised form November 17, 19 97

)

Supported by Grants from NHLBI (HL-39469), NIEHS/NHLBI (09213), and NIEHS(ES-O6499) to BTM and Grants (MT6097, MT8051, and MT7820) from the

Medical Research Council of Canada to AC.

Correspondence and requests for reprints should be addressed to Dr. Brooke T.Mossman, Department of Pathology, University of Vermont, College of Medi-

cine, Burlington, VT 05405. E-mail: [email protected]


2/15

State of the Art 1667

tional asbestos exposure. The presence of benign asbestos-induced pleural disease (plaques or diffuse pleural fibrosis) ishelpful in suggesting the correct diagnosis. Specific criteria forthe clinical diagnosis of asbestosis have been published (6, 8, 13).

The gross pathologic picture of asbestosis is that of d iffuseinterstitial fibrosis most marked in the lower zones, with theworst d isease generally seen closest to the pleura, a nd with rel-ative sparing of the central portions of the lung. Honeycomb-ing is common in adva nced cases. D isease is alwa ys bilatera l.B enign asbestos-induced pleural disease may also be present,but it is not asbestosis. The microscopic diagnosis of asbestosisrequires two findings: diffuse interstitial fibrosis, which in ad-vanced cases is identical to that seen in usual interstitial pneu-monia, a nd the presence of asbestos bodies (fibers of asbestosto w hich the lung has add ed an iron-protein coating) in micro-scopic sections 5

m thick (14, 15).Epidemiologic studies indicate very clearly that the devel-

opment of asbestosis requires heavy exposure to asbestos andprovide strong evidence that t here is a threshold fiber dose be-low which asbestosis is not seen; this dose appears to be, at aminimum, in the ran ge o f a pproxima tely 25 to 100 fiber/ml/yr(7, 1619). Thus, asbestosis is usually seen in workers whohave had many years of high level exposure, for example, as-bestos miners and millers, asbestos textile wo rkers, and a sbes-

tos insulators. Asbestosis can also be produced by very highexposure of relatively short dura tion such as in shipyard w ork-ers employed for a few yea rs inside ship compartment s duringand aft er the Second World War, where exposure levels some-times reached hundreds of fibers per ml of a ir.

The time from first exposure to the a ppeara nce of asbesto-sis, i.e., the latency period, is inversely proportional to expo-sure level. Becklake (20) noted in a report in 1938 that theaverage latency was only 5.2 yr, reflecting extremely high ex-posures in the past, whereas in more recent series the meanyears of exposure varied from 12.6 to 20.2 yr. Progressivelydownward regulation of permitted exposure levels shows astrong decade by decade correlation with increasing latencyperiods (7). These observations emphasize the idea that asbes-

tosis does not appea r until a t hreshold exposure level has beenreached, and t hat t he lower the exposure, the longer it takes toreach this threshold in workplace settings. However, thisstatement does not imply that any exposure produces somedegree of subclinical fibrosis; simple histologic observationsindicate quite clearly that most workers with asbestos expo-sure, as well as members of the general population w ho inhaleasbestos fibers from ambient a ir, show no evidence of fibrosis,presumably because the normal pulmonary defense mecha-nisms are able to remove inhaled dusts until the exposurelevel becomes relatively high. The same comments apply tothe effects of silica.

An ad ditional fa ctor that pla ys a role in the pathogenesis ofasbestosis in humans is cigarette smoke. There is considerableevidence from rad iographic studies that smoking increases theat ta ck and/or the progression rat e for a sbestosis (7, 2123).Experimental studies suggest that this phenomenon is medi-ated by smoke-induced increases in fiber retention (

see

Sec-tion 3).

Silicosis

Silicosis is disease produced by inhalation of one of the formsof crystalline silica, most commonly quartz (

see

Section 2 formineral subtype definitions). There a re several d ifferent clini-cal a nd pa thologic va rieties of silicosis, including simple (nod-ular) silicosis, acute silicosis (silicoproteinosis), complicatedpneumoconiosis (progressive massive fibrosis), and true dif-fuse inte rstitia l fibro sis (24, 25). This review will concern itself

only with simple silicosis since this is the condition that hasgenerally been studied in experimental mo dels.

Simple silicosis is usually diagnosed by a combination of asuitable exposure history, for example, sandblasting, quarry-ing, stone dressing, refractory manufacture, or foundry work,and the finding of fine nodules on plain chest film or C T scan.D isease tends to be upper zonal, but the lower zones may beinvolved in severe cases. B y definition the nodules of simplesilicosis are no grea ter t han 1 cm in maximum diameter; la rgernodules are classified as complicated pneumoconiosis (25).

Many patients with simple silicosis are asymptomatic.Cough may be present, possibly reflecting irritation of tra-cheal and bronchial nerves by silicotic nodules (25). Shortnessof breath is not common. In many patients with simple silico-sis, pulmonary function is normal or shows only a minor de-crease in vital capacity; however, there is increasing evidencethat silica exposure can be associated w ith changes of chronicairflow obstruction (2528).

G ross pathologic examination o f lungs with simple silicosisshows discrete nodules that are extremely hard and vary incolor from grey to blue to green if the exposure is to rela tivelypure silica, but they ma y be black (when silicosis develops in acoal miner) or red (when silicosis develops in a hematiteminer). Microscopically early lesions are characterized by

nodular to stellate aggregates of dust-laden macrophages ar-ranged around a collagenous central region. With time, thecentral collagen becomes distinctly whorled and the relativenumber of inflamma tory cells around the periphery decreases.In so-called mixed dust pneumoconioses (i.e., diseases causedby a combination o f silica plus another d ust such as iron oxide,or a silicate), the nodules tend to retain their stellate contourand the central collagen shows less of the whorled arrange-ment seen in ordinary silicosis. Examination with polarizedlight usually reveals birefringent part icles, most of which, con-trary to the usual description in textbooks, are silicates ratherthan silica; the lat ter a re only wea kly birefringent (24).

As is true o f a sbestosis, the lat ency period for simple silico-sis is inversely proportional to exposure levels and is generally

quite long. E xtremely high levels of silica exposure can resultin the ra pid appea rance of silicoproteinosis or a ccelerated sili-cosis. As opposed to asbestosis, where the factors that governthe appea rance of disease appear to be primarily fiber dose, fi-ber type, cigarette smoking habit and, perhaps, fiber size, thedevelopment of silicosis is influenced by d ose and type o f sil-ica, but apparently not by smoking. Moreover, coexposure toother dusts has no apparent effect on the incidence of asbesto-sis but tends to markedly decrease the incidence of silicosis(

see

Section 2).

2. CHEMICAL AND PHYSICAL PROPERTIES OFASBESTOS AND SILICA IMPORTANT INCAUSATION OF LUNG DISEASE

Asbestos is a group of crystalline 1:1 layer hyd rated silicate fi-bers that are classfied into six types based on different chemi-cal and physical features (29). These include chrysotile[Mg

6

Si

4

O

10

(OH)

8

], the most common and economically im-portant asbestos in the Northern Hemisphere, and the am-phiboles: crocidolite [Na

2

(Fe

3

)

2

(Fe

2

)

3

Si

8

O

22

(OH)

2

], amosite[(Fe,Mg)

7

Si

8

O

22

(OH)

2

], anthophyllite [(Mg,Fe)

7

Si

8

O

22

(OH)

2

],tremolite [Ca

2

Mg

5

Si

8

O

22

(OH)

2

], and actinolite [Ca

2

(Mg,Fe)

5

-Si

8

O

22

(OH)

2

]. The morphology of chrysotile, a pliable curly fi-ber, is shown in Figure 1a in comparison with the rodlike fiber,crocidolite (Figure 1b and c). In contrast, silica (Figure 1dshows cristobalite) is a compact particle with a less than 3:1length to diameter ratio. The geometry and dimensions of these


3/15

1668

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 157 1998

minerals may govern their deposition and clearance kinetics,biologic reactivity, and dissolution in the lung, but chemical andsurface propert ies, including sorption, oxidat ion/reduction reac-tions, and charge, also play important roles in biopersistence,cellular responses, and pathogenicity (29, 30).

Like a sbestos, silica is a group of natura lly occurring miner-als, but these exist both in noncrystalline (amorphous) andcrystalline forms (31). D usts composed of a morphous silicas

have not been implicated in human disease and will not beconsidered further in this review. The seven recognized crys-talline silicas, comprising silicon and oxygen (SiO

2

) with traceamounts of Al, Fe, Mn, Mg, Ca, and Na in their structures, in-clude quartz, cristobalite, moganite, tridymite, melanphlogite,coesite, a nd stishovite. Ro cks such as granite, shale, and sa nd-stones contain a s much as 67% quart z, the most common crys-ta lline silica; t hus, mining, blasting, a nd construction a ctivities

Figu re 1. Morphology of asbestos fibers and silica particles (arrows) as visualized by scanning electron mi-croscopy. (a) Note attempted phagocytosis of long chrysotile fiber by a tracheal epithelial cell in vitro. (b)Rodlike bundle of crocidolite asbestos fibers. (c) Phagocytosis of long crocidolite asbestos fiber by a noncil-iated tracheal epithelial cell in vitro. (d) Nonfibrous particles of cristobalite silica.


4/15


may result in significant exposures. Cristobalite is encoun-tered in the ceramic, refractory, and diatomaceous earth in-dustries where processing of the crude materials involvesheating t o high tempera tures (25).

Workers exposed t o a sbestos or silica freq uently encounterdifferent t ypes of these and other dusts, which may b e dissimi-lar in composition from source to source. In addition, manyepidemiologic and experimental studies fail to specify or char-acterize the source and properties of minerals used.* Thus, de-fining the exact properties of minerals importa nt in the causa-tion of asbestosis or silica-induced lung diseases is problematicand controversial (3133). Moreover, recent studies havedemonstrated that the toxicity and pathogenicity of asbestosand silica may be dependent on a combination o f physical/me-chanical and chemical propert ies.

Mineral surfaces are dynamic and complex and may bemodified in the lung after adsorption of proteins and othermacromolecules or uptake by cells. Elements such as magne-sium from chrysotile (34) or iron from crocidolite or amositeasbestos (35) may be leached or mobilized intracellularly orextracellularly, thus mediating the toxicity of these fibersthrough surface charge or redox reactions (the latter topic isreviewed extensively below). Magnesium initially is leachedfrom chrysotile fibers (36), thus changing their surface charge

from positive to negative and reducing their toxic potential(37). However, silica release profiles govern and are predic-tive of fiber residence times in lung, an estimated 9 mo for achrysotile fiber with a diameter of 1

m (36). Clearance anddissolution kinetics for amphibole fibers are considerablyslower than those for chrysotile in the lung (36, 38, 39), andthese data might explain why chrysotile fibers are fewer, incomparison with amphibole types of asbestos, in the lungs ofworkers exposed primarily to chrysotile (

see

Section 3).Surface chemistry may also govern the pathogencity of sil-

ica dusts. Freshly fractured silica, generated during abrasiveblasting, is more toxic to alveolar macrophages (AMs) thanaged silica (40), presumably beca use of its increased redo x po-tential, a s its fresh surface is highly reactive with hyd rogen, ox-

ygen, carbon, and sometimes nitrogen (30). Crushing silicayields Si

and SiO

radicals on the cleava ge planes of this min-eral that also react with water to produce the damaging hy-droxyl radical (

OH ) (40). U sing silica polymorphs of thesame size dimensions and surface area, Wiessner and col-leagues (41) showed correlations between increased toxicity(lysis of red bloo d cells), inflammation, a nd fibrogenicity aft erintratracheal injection into mice, of quartz, cristobalite, andtridymite, in contrast to coesite. B ecause the low ato mic pack-ing density of t he fibrogenic minerals was the only varia ble de-lineating them from the high packing density of the nonfibro-genic coesite, they concluded that the pathogenicity of silicaincreases as the atomic packing density of the surface struc-ture decreases. Unfo rtunat ely, the active surface sites on as-bestos and silica part icles remain undefined.

A further problem in understanding the relationship be-tween silica exposure and disease is the tendency of otherminerals, particularly clay components, to adhere to and evenchemically integrat e with the surface of silica part icles. For ex-ample, Tourmann and Kaufmann (42) reported that only 1 to2% of q uartz particles recovered fro m the lungs of coal minershad bare (silica) surfaces. It has been proposed that this pro-

cess, sometimes labeled occlusion of the silica surface, is re-sponsible for the relatively low incidence of silicosis seen incoal miners and other workers with coexposure to silica andanot her dust (43). B y contra st, silicosis is relat ively commonand often quite severe in workers who are exposed to freshlyfractured silica surfaces, for example, sandbla sters (25).

Although it is clear that the surface characteristics andcomposition of asbestos and silica are linked to bioreactivityas w ell as to biopersistence in lung, structural f eat ures, includ-ing the size dimensions of inhaled minerals, part icularly a sbes-tos fibers, are also critical in the development of disease. Ex-perimentally, long thin fibers are more fibrogenic than shortfibers (i.e.,

5

m in length) in both intratracheal (4446) andinhalation studies (47), and chronic rodent inhalation experi-ments using short chrysotile fibers have not yielded fibrotic le-sions (48). L ong fibers a re more po tent inducers of cell prolif-eration (44), cell injury, and inflammation (49), and oxidantrelease from AMs (50), thus providing a mechanistic frame-work for their increased fibrogenicity (

see

Sections 3 and 4).However, in humans, correlations between fiber size and dis-ease are difficult to pro ve, possibly because of confounding bycigarette smoking, which tends to increase the retention ofshort fibers (

see

Section 3). The correlation betw een silica pa r-ticle size and disease is similarly uncertain (

see

Section 3).

3. THE RELATIONSHIP OF PARTICLE BURDEN TODISEASE IN ASBESTOSIS AND SILICOSIS

Asbestosis

Fiber burden studies of human lung have shed light on thepathogenesis of asbestosis, and in particular have shown thatthere is a relationship between high retained fiber concentra-tion and the development of asbestosis. Nonetheless, suchstudies still leave unanswered many questions about the exactrelationship of fiber number, type, and size to the appearanceof asbestosis. Moreover, there are some apparent contradic-tions between human a nd animal studies.

The reader should be aware that one of the peculiarities of

fiber burden studies on human material is the considerablelaboratory-to-laboratory variation in absolute fiber counts ob-tained, even when several laboratories analyze portions of thesame specimen (51). Part of this variation relates to differentinstrumentation with different resolution limits, part to differ-ent counting rules (some laboratories count all fibers, somecount only long fibers), and part to poorly defined idiosyn-cratic factors. Thus, to understand the relationship betweenthe various asbestos-related diseases and fiber burden, pat-terns must be sought within the data collected by each labora-tory as comparison of absolute numbers from laboratory tolabora tory is meaningless.

A second important issue in the interpretation of fiber bur-den studies is the relative lack of biopersistence of chrysotilecompared with amphibole asbestos. B oth animal and humanstudies show t hat continuing exposure to a mphiboles results incontinuously increasing amphibole fiber levels recoverablefrom t he lung, whereas continuing exposure to chrysotile is as-sociated with a negligible increase in chrysotile fiber burdenover time (reviewed in reference 38). The estimated half-lifefor amphibole fibers, while difficult to estimate accurately be-cause of the type of laboratory-to-laboratory variations men-tioned above, appears to be on the order of decades, whereasthat of chrysotile appears to be around a few months (38, 52).For this reason, virtually all studies of human fiber burdensshow a predo minance of amphiboles, even with nominal expo-sure only to chrysotile (38) (

see

below

). B iopersistence is prob-ably responsible for the much greater tendency of amosite

* Throughout this review, the terms silica and asbestos are used for sim-

plicity, but they may reflect certain mineralogical species having different patho-genic potentials. The reader is referred to the original references for sources and

characterization of minerals.


5/15

1670


or crocidolite-induced asbestosis to progress, compared withchrysotile-induced asbestosis (reviewed in reference 7).

Intralaboratory comparisons of fiber concentration and as-bestos-induced disease from three different laboratories arepresented in Tables 1 to 3. With the exception of cases of me-sothelioma in Quebec chrysotile miners and millers (Table 1),the lungs of wo rkers with asbesto sis consistently show the high-

est average fiber burdens of any of the asbestos-induced dis-eases. B ecause of the labora tory-to-labora tory varia tions men-tioned ab ove, it is difficult to be certain of how much greater t hetypical concentrations seen in asbestosis are compared withother diseases, and there is also considerable variation amongfiber types. G ibbs and P ooley (53), presenting data largely fromU .K. end products users, reported a rithmetic mean based rat iosof the fiber concentrations found in workers with asbestosiscompared with pleural mesothelioma of about 20:1 for croci-dolite and 4.5:1 for amosite (Table 2), whereas C hurg and V eda l(54), presenting data as geometric means, found a ratio of about12:1 for amosite in shipyard workers and insulators (Table 1),and R oggli and colleagues (55) found a ra tio of media n valuesof a bout 10:1 for uncoated fibers, most of which were appar-ently amosite (Table 3). The amphibole data are at least rea-sonably consistent and serve to emphasize both the require-ment for high fiber burdens for the development of asbestosisand also the relative sensitivity to amosite and crocidolite fi-bers of the pleura compared with the parenchyma.

The chrysotile data are considerably more complicated. Inend-products users, G ibbs and Po oley (53) (Table 2) reportedan asbestosis:mesothelioma ratio of about 1.5:1 for chrysotileand also showed that the chrysotile values were considerablyabove the general population background. On the other hand,we (54) reported approximately equal concentrations of chryso-tile in shipyard workers and insulators with asbestosis or me-sothelioma, but these values were not above general popu-lation background, a common observation in those with fairlyremote occupationa l exposure. B ut in Queb ec chrysotile min-

ers and millers from the Thetford Mines region, we (56) (Ta-ble 1) observed that mesothelioma and asbestosis occurred atabo ut the same, very high, fiber burden. This was also true o fthe tremolite fibers that a re a component of the chrysotile ore.Like all amphiboles, tremolite fibers accumulate in lung andwere in fact present in greater concentrations than were thechrysotile fibers (Table 1). Thus, although induction of asbes-tosis by chrysotile clearly requires very high pulmonary fiberconcentrations, the parenchyma and pleura appear to showequal sensitivity to chrysotile (and its contaminant tremolite)in terms of the development of mesothelioma. A schematicsummary of the relationship between fiber burden a nd diseasepatt erns in persons with a mphibole versus chrysotile exposureis shown in Tab le 4.

Fiber burden studies also indicate tha t there is a correlationbetween pathologic severity of asbestosis and increasing bur-den of asbestos bodies (which are largely markers of amphib-ole exposure) or uncoated amosite and crocidolite fibers (re-viewed in reference 57). Aga in, at least for end-products users,there is less consistency to the chrysotile data. For example,Wagner and colleagues (58) found no correlation betweenchrysotile fiber burden and asbestosis grade in East Londonfacto ry workers, although G reen and colleagues (59) did re-port a correlation for both chrysotile and tremolite in chryso-tile textile workers, and we found a similar correlation forchrysotile miners and millers (60).

One additional piece of information that comes out of fiberburden studies is the apparently greater fibrogenic potential

of amosite (and crocidolite) compared with chrysotile and itscontaminant , tremolite, o n a fiber-for-fiber basis. This can beseen by comparing the absolute fiber concentrations in chryso-tile miners and millers versus shipyard workers and insulatorsin Table 1.

There are few available human data on the relationship offiber size measures and asbestosis, and these data are difficultto reconcile with animal studies. It is important to note tha t, ingeneral, animal data suggest that long fibers are cleared bymucociliary or macrophage-mediated transport much moreslowly than are short fibers, and that above a certain fiberlength (probably 16 to 20

m), fiber cleara nce is severely com-promised (39, 61).

As described in Sect ion 2, D avis an d colleagues (46, 47)and A damson and B owd en (44, 62) both observed tha t long fi-bers were considerably more fibrogenic than short (

5

m)

TABLE 1

GEOMETRIC MEAN FIBERS CONCENTRATION BY DISEASE INSHIPYARD WORKERS AND INSULATORS WITH HEAVY AMOSITE

EXPOSURE OR IN CHRYSOTILE MINERS AND MILLERS*

Group

Fibers

10

6

/g Dry Lung

Shipyard

Workers/Insulators Chrysotile Miners/Millers

Amosite Chrysotile Tremolite

Exposed, no disease 0.7 2.0 9.0Pleural plaques 1.4 15 75

Mesothelioma 0.9 34 180

Asbestosis 10 30 140

* From References 54 and 56.

TABLE 2

MEAN ASBESTOS FIBER CONCENTRATION BY TYPE AND DISEASE*

Group

Millions of Fibers/g Dry Lung

Chrysotile Amosite Crocidolite

Controls

2.89.3 0.090.93 0.141.00

Pleural mesotheliomas 45 103 53

Peritoneal mesotheliomas 75 100 304

Asbestosis 69 450 1,100

* From Reference 53.

General population, from various reports from this laboratory.

TABLE 3

MEDIAN BURDEN OF UNCOATED FIBERS BY DISEASE*

Group Fibers

10

3

/g

Pleural plaques 2.2

Mesothelioma 67

Asbestosis 690

* From Reference 55.

TABLE 4

RELATIVE FIBER BURDEN ASSOCIATED WITH SPECIFIC DISEASES

Chrysotile (including tremolite) burden

General population

urban areas

* Plaques

Asbestosis or mesothelioma

Amosite or crocidolite burden

General population

Plaques

Mesothelioma

Asbestosis

* Indicates approximately 1 order of magnitude in fiber burden.


6/15


fibers in rats and mice. But in humans, we w ere unable toshow that lungs with asbestosis had longer fibers than lungswith other types of asbestos-induced disease (60, 63) and, infact, we found an inverse correlation between fibrosis gradeand mean fiber length for either amosite or chrysotile andtremolite in the two cohorts described above. Coin and col-leagues (39) have suggested that this observation may reflectinterference of subsequent fiber clearance in any area wherelong fibers accumulate and start to produce local fibrosis, sothat over time there is secondary a ccumulation of short fibers.

The issue of fiber size a nd disea se in humans is also compli-cated by the fact that, historically, most asbestos-exposedworkers smoked and, as indicated previously, there are goodradiographic data to indicate that smoking increases the at-ta ck/progression ra te o f a sbestosis. This phenomenon ca nbe reproduced in animal models (64) and may be mediatedin part by smoke-induced interference with fiber clearance,thus producing an increased effective fiber dose to the lung.Smoke-induced increases in fiber uptake by pulmonary epi-thelial cells may occur (65) with resulting increases in celldamage and cytokine production. One of the peculiarities ofcigarette smoke exposure is that it increases the retention ofshort fibers much more than the retention of long fibers inboth animals and humans (65, 66) so that even if short fibers

are only weakly fibrogenic on a fiber-for-fiber basis, smokersmay be increasing their short fiber load to the point that itplays a role in fibrogenesis. Thus, the issue of fiber length a ndasbestosis in humans remains unsettled, but it is proba bly im-porta nt in the development of disease, especially in smokers.

The results summarized in Table 4 indicate that asbestosisis clearly a fiber dose-driven disease but, nonetheless, only afraction of a ny cohort exposed to a fibro genic dose of asbestosdevelops asbestosis. It has been proposed that person-to-per-son variat ions in either fiber deposition or fiber cleara nce mayaccount fo r this phenomenon. B egin and colleagues (67, 68)observed that, when equal doses of chrysotile were adminis-tered to sheep, roentgenographic evidence of asbestosis aswell as lavage evidence of a n active alveolitis developed in the

sheep that had the lowest fiber clearance (measured by actualanalysis of fiber burden), and that the differences in fiberclearance preceded the appeara nce of asbestosis. B ecklakeand colleagues (69) showed that roentgenographic asbestosisappea red to be more prevalent in workers who were short a ndhad narrow er chests, findings which these investigat ors att rib-uted to (unknown) patterns of underlying lung structure, butwhich might relate to narrower airways in those with smallerlungs.

Silicosis

A number of studies have investigated t he relationship of pul-mona ry silica burd en an d silicosis. This issue is complicated bya variety of factors, including type of silica particle involvedand coexposures to ot her types of minerals.

Nagelschmidt (70) summarized much of the literature onthe question of total silica burden and resulting pathologic re-action pa ttern. A s is true for a sbestosis, there is a rela tionshipbetw een increasing weight of silica ret ained in the lung and in-creasing pathologic grade of silicosis. However, in contradis-tinction to a sbestosis, the presence or absence of other mineralsis important in determining whether or not the patient devel-ops silicosis. This is illustra ted in Figure 2. When there is expo-sure to relatively pure silica, as in gold miners or foundryworkers, then total retained silica loads of 1 to 3 g are suffi-cient to produce silicosis. On the other hand, when there isconcomitant exposure to another (relatively nonfibrogenic)dust such as is seen in coal miners or hemat ite miners, then the

same weight of silica produces very little silicosis. This phe-nomenon presumably reflects the adsorption of o ther dusts ortheir constituents onto the silica surface with consequent de-creases in silica toxicity (Section 2), and has important impli-cations for the prevention of silicosis.

As has been made clear for asbestos, different mineralogictypes of silica have different pathogenic potential, althoughthe exact relationship between fibrosis and exposure to partic-ular silica polymorphs is somewhat confused. The experimen-tal data consistently show that tridymite, cristobalite, andquartz are more fibrogenic than amorphous silica and coesite(41, 71, 72). Simple intratracheal injections of equal masses ofdust suggests that the order of potency is tridymite

cristo-balite

quartz. However, Wiessner and colleagues (41) ob-served that, if the doses were adjusted to produce equal sur-face areas, then both the hemolytic and the fibrogenicpotential of these three dusts was equal. A further importantquestion arises from the observation that freshly fracturedquartz produces considerably greater quantities of active oxy-gen species and greater inflammatory reactions than doesaged quartz (40), and whether this phenomenon applies totridymite and cristob alite is not clear. Unfo rtunat ely, theredoes not appear to be human data that would allow one tocompare t he potencies of these silica polymorphs.

The relationship between silica particle size and disease isalso unsettled. King and colleagues (73) claimed that, usingparticles of the same composition and equal surface, particles

in the 1 to 2

m size ranges were more fibrogenic than smalleror la rger part icles. Ho wever, Wiessner and colleagues (74) ob-served that relatively large particles (

5

m) were more fib-rogenic than 1

m particles. Finally, it should be noted thatvirtually no informat ion exists about the relationship of differ-ent types or sizes of silica pa rticles to t he cellular a nd molecu-lar events discussed in Section 4.

4. CELLULAR AND MOLECULAR MECHANISMS OFASBESTOSIS AND SILICOSIS

The pathology of human and rodent lung tissues after inhala-tion of asbestos or silica has afforded knowledge of the cell

Figur e 2. Average weight (g) of total dust and quartz in lungs ofworkers with advanced forms of fibrosis. Reproduced with permis-sion from Reference 70.


7/15

1672


types affected in lung after d eposition of minerals in the bron-chiolar-alveolar duct region as well the chronology of eventsleading to the development of fibrosis (reviewed in references75 and 76). In addition, experimental animal models and invitro

approaches have allowed elucidation of many of the mo-lecular a nd functional changes in AM s, bronchiolar a nd alveo-lar epithelial cells, fibroblasts, and other cell types that mayparticipate in the patho genesis of asbestosis and silicosis.

B oth inflamma tion and f ibrosis as well as expression ofgenes linked to cell proliferation and antioxidant defense oc-cur in a dose-related fashion after inhalation exposures to as-bestos (77). Whereas lower intensity exposures to either as-bestos (77) or silica (78) evoke reversible inflammatory lesionscharacterized by foca l aggregations of mineral-laden AM s andmaintena nce of the normal a rchitecture of the lung, higher ex-posures elicit intense and protracted inflammatory changes,cell proliferation in va rious compart ments of the lung, and ex-cessive deposition of collagen and other extracellular matrixcomponents by mesenchymal cells. The AM is viewed a s a piv-ota l cell type in fibrogenesis in both lung defense and elab ora-tion of grow th fa ctors and oxidant s (reviewed in references 79and 80). In a ddition, va rious cell types of t he immune system,including neutrophils (78, 81), T-lympho cytes (29, 82), and mastcells (83, 84) accumulate in B AL and /or inter stitia l regions in

rodents exposed to a sbestos or silica a nd are implicated in thedevelopment of fibrosis. Recent studies suggest cross-ta lkbetween mast cells and neutrophils in a murine model of sili-cosis in that silica instillation induces less severe lung lesionsin mast-cell-deficient mice w hen compared with mast-cell-intactmice (84). Moreover, ado ptive tra nsfer of bo ne-marrow -derivedcultured mast cells from mast-cell-intact littermates into defi-cient mice results in both increased neutro phils in bronchoa lve-olar lava ge (B AL ) and more severe pulmonary lesions. A mul-tiplicity o f interactions between these effector cells and targetcell types of injury, including bronchiolar a nd a lveolar epithe-lial cells and fibroblasts, may govern the pathogenesis and pro-gression of disease.

Injury to the a lveolar type I epithelial cell is regarded as a n

early event in fibrogenesis followed b y hyperplasia and hyper-trophy (85) of type II epithelial cells. Increases in epithelialcell proliferation initially may be crucial to repair a nd regener-at ion, and, if unchecked, fibrogenesis and carcinogenesis. Forthese reasons, and because hyperproliferation of mesenchy-mal cells is also a hallmark of the fibrot ic lesion, asbestos- andsilica-induced cell proliferat ion has b een studied in a numberof in vivo

and in vitro

models. Inhalation of asbestos at highairborne concentrations (i.e.,

7 mg/m

3

air) by rodents fortime points of up to 20 d causes early and reversible increasesin uptake of 5

-bromod eoxyuridine (B rdU ), an indication ofcells in the S phase of the cell cycle, in bronchiolar epithelialcells, and in the alveolar duct region and interstitial compart-ments (8688). These proliferative changes are not observedafter exposure to nonfibrogenic dusts or glass fibers, and pat-terns differ from the more protracted increases in Brd U incor-poration observed in mesothelial cells after exposure to am-phibole types of asbestos by inhalation (86) or intratrachealinstillation (89). Conceivably, proliferation may occur intiallyat sites of accumulation of inhaled minerals, but later at distalsites where particles or fibers are translocated over time. Al-ternat ively, mitogenic cytokines may med iate signaling events,leading to cell replication at sites physically remote from fi-bers (89, 90).

The initiation of proliferation in epithelial cells and fibro-blasts by asbestos or silica may occur aft er upregulation of theearly response protooncogenes, c-

fos

, c-

jun

, and c-

myc

(77, 9193). c-

fos

and c-

jun

encode proteins of the Fos and Jun family

that can dimerize to form activator protein-1 (AP-1), a tran-scription factor that binds to the D NA of the promoter regionof a number of intermediate genes governing inflammation,proliferation, and apoptosis (reviewed in reference 94). Mes-sage levels of both c-

jun

and ornithine decarboxylase (

odc

), agene with an AP-1 site in its promoter region that encodes akey enzyme in the b iosynthesis of polya mines, are increased inrat lung homogenates after inhalation of asbestos (77). Asfunctional ramifications of c-

jun

(95) and odc

(96) o verexpres-sion are increased cell proliferation and transformation invitro

, upregulation of t hese genes in lung may be critical to thepathogenesis of both pulmonary fibrosis and lung cancer (

see

below

).Increased expression of early response genes and protein

products is also linked to the development of apoptosis, orprogrammed cell death, in a number of cell types. In this re-gard, we and others have demonstrated the development ofapoptosis by asbestos and silica in mesothelial cells (97, 98)and AMs (99), respectively, a phenomenon that is not ob-served aft er exposure of cells in vitro

to a number of nonfibro-genic dusts. Moreover, striking increases in apoptosis are ob-served in bronchiolar and alveolar type II epithelial cells afterexposure of ra ts to asbestos and cigarett e smoke in contrast toits rare occurrence in normal rat lungs (D avis et al .

, unpub-

lished o bservations). The physiologic consequences of apopt o-sis in epithelial cells of the lung are speculative, but a recentstudy suggests that apoptosis is a major pathway responsiblefor the remova l of proliferating alveolar type I I epithelial cellsin acute lung injury (100). Others hypothesize that apoptosisof bronchial and alveolar epithelial cells in the bleomycin in-stillation model of fibrosis is linked to the development o f dis-ease as apoptosis intially peaks and is sustained during theprogression of fibrosis. In addition, bot h apopt osis and f ibrosisare blocked in this model after administration of corticoster-oids (101).

A dramatic and persistent inflammatory response is ob-served in animal models of a sbestosis and silicosis that ma y belinked to the development of cell injury, proliferation, a popto-

sis, and fibrogenesis. The importance of reactive oxygen spe-cies (R OS) in mediating t hese events is presently under inves-tigation in a number of laboratories. Various types of asbestosand silica can spontaneously catalyze the formation of ROS inaqueous solutions or after uptake by cells. It has been knownfor several years that the surface iron(II) or leachable iron(II)on mineral surfaces reduces molecular oxygen to superoxideanion, which then dismutates to hydrogen peroxide. In thepresence of asbestos or silica, hyd rogen peroxide and superox-ide react via a Fenton-like reaction driven by iron to form thepotent hydroxyl radical in vitro

(reviewed in reference 102).Evidence suggests that these reactions also occur in rodentlungs after intratracheal instillation of silica, but not the inertdust, titanium dioxide (103). In addition, iron may be mobi-lized from a sbestos fibers (104), and silica a nd a sbestos fibersmay adsorb iron extracellularly in t he lung or pleura (105, 106).

Another pathwa y of free radical generation by asbestos orsilica occurs via an oxidative burst when fibers are phago-cytized by AMs or other cell types, including alveolar epithe-lial cells and fibroblasts (reviewed in reference 107). Longer,more fibrogenic fibers of asbestos cause a frustrated, ineffec-tive phagocytosis and more protracted elevations in release ofROS (50, 108). In addition, activated inflammatory cells suchas AM s recovered by B AL may release increased amounts ofoxidants spontaneo usly during the progression o f d isease (109),an observation that might explain the prevalence of oxidizedB AL proteins in patients w ith interstitial lung diseases (110).

Superoxide can also react rapidly with nitric oxide to form


8/15


peroxynitrite, an agent that oxidizes and nitrates macromole-cules. Recent studies also indicate that elaboration of reactivenitrogen species (RNS) is increased in AMs recovered fromrats exposed by inhalation to fibrogenic concentrations of as-bestos (111). These data and in vitro

studies showing increasednitrate and nitrite production from AMs exposed to asbestos(112) indica te a p lausible ro le of R NS in toxicity a nd/or cellsignaling pathways affecting lung epithelial and other celltypes (113). Asbestos (111) and silica (114) cause increasedsteady-state mRNA levels of inducible nitric oxide synthase(iNOS) in AMs in vitro

and in lung homogenates after in-tratracheal injection of particles, respectively, suggesting thatthis enzyme is induced in AM s and po ssibly other cell types inlung after exposure to fibrogenic dusts.

Although R OS have been viewed classically as injurious tomany cell types via modification of macromolecules such asD NA, the eff ects of oxidants on cells are complex and dose-related. Oxidants generated by fibrogenic dusts or cigarettesmoke may induce uptake of a variety of particle types (re-viewed in reference 107), lipid peroxidation (115, 116), stimu-lation of cell-signaling cascades and transcription factors (93,117, 118), and release of cytokines such as tumor necrosis fac-tor-alpha (TNF-

) (119). That these interre lated oxida nt-inducedevents are important in inflammation and fibrogenesis is sub-

stantia ted b y successful at tempts to inhibit lung disease in ro-dent inhalation and intratracheal injection models using anti-oxidants and antibo dies to TNF (

see

Section 5).A number of gro wth fa ctors and cytokines have been impli-

cated in clinical studies and in animal mo dels of asbestosis andsilicosis (Table 5). There is no doubt that fibrogenic mineralsinduce elaboration of a number of cytokines in vitroin BA Lfluids and in the lung, but it is still uncertain which, if any, a reessential to the development of fibrosis. D ata also indicat ethat multiple feedback mechanisms governing cell signalingand gene expression of various importa nt cytokines may exist.Elegant work has established by immunohistochemistry thatisoforms of transforming growth fa ctor-beta (TG F-) are in-creased at sites of developing asbestotic lesions (120) and in

silica-induced granulomas (121123). This growth factor mayhave multiple roles in fibrogenesis as it is a potent chemoat-tractant for monocytes and neutrophils and upregulates genesinvolved in collagen and fibronectin biosynthesis. Paradoxi-

cally, it can act as an inhibitor of inflammation and epithelialcell proliferation, but its production by AMs and alveolar epi-thelial cells may serve as a mitogenic stimulus for fibrob lasts.

Cro ss-ta lk between TG F- and platelet-derived growthfactor (PD G F) may be important in chemotaxis and fibroblastproliferation (124). L ike TG F-, various PD G F isoforms aremitogenic for mesenchymal cells, but they induce chemota xisdifferentially in rat lung fibroblasts in vitro(125, 126). The up-regulation by asbestos of PD G F-AA a nd PD G F- receptorsin these cell types suggests an auto crine pathwa y in add ition tothe documented ro le of A Ms in PD G F production (127, 128).Insulinlike growth fa ctor (IG F-1) also is produced by A Ms ex-posed to silica (129), thus providing a possible additional stim-ulus for fibrob last proliferation in the lung.

U pregulation of epidermal growth factor (EG F) and trans-forming growth factor-alpha (TG F-), which binds to theE G F-receptor, at sites of asbestos (130) or silica deposition(131) may also be key to mitogenesis of pulmonary epithelialcells and fibrobla sts. E G F-like activity has been demonstrat edin B AL fluids in experimental silicosis (132), and detection ofincreased amounts of the extracellular domain of the EG F-receptor has been observed in the serum of asbestotic patients(133). The increased bio synthesis of TG F- (130) and E G F-Rprotein (Zanella et al., unpublished observations) as well as

auto phosphorylation of t he EG F-receptor by asbestos (117)are plausible mechanisms leading to availability and stimula-tion of cell-signaling pathwa ys by E G F amd TG F-.

Inflamma tory media tors such as TNF and various interleu-kins (IL-1, IL-8) are under intense investigation in mineral-exposed persons (134136) and experimental models of fibro-sis as they appear to have a role in the initiation of disease.Most compelling are data showing that neutralization of TNFwith anti-TNF antibodies or soluble TNF receptors (137) andIL -1 receptor anta gonists (138) prevent a nd a meliorate exper-imental silicosis. Elevated mRNA expression and release ofIL -1 and TNF occur in AMs from pa tients with a sbestosis andIPF, and both cytokines cause upregulation of collagen and fi-bronectin gene expression in normal human fibroblasts in

vitro (134). Moreover, investigation of TNF and IL-1 expres-sion in animal mod els of silicosis and asbestosis show early in-creases in these inflammatory mediators that precede frankinflammation and fibrosis (139). A vast number of inflamma-tory and immune responses are stimulated by IL-1 and TNF,including T- and B -cell lymphocyte proliferat ion and activa -tion, increases in ara chidonic acid metabo lism, oxidative burstand degra nulation of inflammat ory cells, and expression of ad-hesion molecules (reviewed in reference 140). Thus, multipleroles of these cytokines are implicated in fibro sis and other in-flammatory diseases.

Silica and asbestos-induced hydrolysis of pho sphoinositidesand activation of the arachidonic acid cascade are also impli-cated in regulation of TNF (141, 142). For example, inhibitionof the lipoxygenase pat hwa y decreases secretion of TNF fromsilica-exposed macrophages, wherea s add ition of LTB4stimu-lates its release (143). Silica also directly increases TNF genetranscription and production in macrophagelike cells in partby upregulating the TNF promoter (144).

A novel group of proinflammatory cytokines known aschemokines, which includes IL-8, macrophage inflammatoryprotein 2 (MIP-2), macrophage inflammatory protein 1and, and monocyte chemoattractant proteins 1, 2, and 3 (MCP-1,2,3) has been implicated as key contributors to the inflam-matory response in bleomycin-induced fibrosis (4), asbestosis(145, 146), and silicosis (145, 147). These chemokines can beproduced by a variety of cell types, including immune andnonimmune cells and exhibit chemotactic activity for cells

TABLE 5

PUTATIVE MEDIATORS OF INFLAMMATORY AND FIBROGENICRESPONSES TO SILICA AND ASBESTOS

Factors Silicosis Asbestosis

Reactive oxygen species (ROS) *

Reactive nitrogen species (RNS)

Arachidonic acid and lipid metabolites

Platelet-derived growth factors (PDGF) Transforming growth factor-(TGF-)

Transforming growth factor-(TGF-)

Epidermal growth factor (EGF)

Insulinlike growth factor (IGF-1)

Interleukin-1 (IL-1)

Tumor necrosis factor-(TNF-)

Interleukin-8 (IL-8)

Macrophage inflammatory proteins-1and 2

(MIP-1, MIP-2)

Monocyte chemoattractant protein-1 (MCP-1)

Cytokine-induced neutrophil chemoattractant

(CINC)

* Plus sign indicates positive association in experimental models and/or clinical studies.


9/15

1674 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 157 1998

such a s neutrophils, monocytes, lymphocytes, and eosinophils.The demonstrat ion that anti-MIP -2 antiserum att enuates neu-trophil infiltration associated with exposure of rats to silicasupports the concept that these chemokines are intrinisic toinflammat ion (147). R ecent work in t hese experimental mod-els also indicates that TNF mediates the MIP-1(4) and MIP-2pathw ays as pa ssive immunization o f mice against TNF mark-edly attenuates increased steady-state mRNA levels of MIP-2in response to silica (145). The exciting demonstration that bron-chiolar and alveolar epithelial cells, in addition to AMs, syn-thesize MIP -2, which is upregulated aft er add ition of TNF, as-bestos, or silica to rat alveolar type II epithelial cells in vitro(147), supports the concept that pulmonary epithelial cells aremediators as well as targets of inflammation.

A variety of cell types conventionally have been regardedas key pa rticipants in the inflammat ory process. The AM his-torically has been viewed a s a two-edged sword in both lungdefense and injury elicited by fibrogenic dusts, and aggrega-tion at sites of mineral deposition is thought to play a role inproliferation of alveolar type II epithelial cells, phagocytosisand cleara nce of part icles, and fibrogenesis (reviewed in refer-ences 79 and 80). Co mpelling da ta show tha t ma st cells are es-sential for the development of silica-induced pulmonary in-flammation (84), and they have historically been viewed as

key pla yers in a sbestosis (83). Last ly, T-lymphocyt es (148) andneutrophils (149, 150) may be protective or injurious, respec-tively, in asbestos-induced cell damage and inflammation.Thus, communication via elaboration of chemokines or cyto-kines by these cell types and their interactions with epithelialcells and fibrobla sts may govern the eventual outcomes of cellinjury and proliferation in response to pathogenic minerals(Figure 3).

5. APPROACHES TO PREVENTION ANDTREATMENT OF FIBROSIS

Regulating occupational exposures to minerals and removalof symptomatic persons from the workplace are importantmeasures to prevent or ameliorate mineral-induced lung dis-ease. However, little advancement has occurred in effectivetherapeutic strategies for patients. As emphasized in a NIHworkshop report (151), IP F, silicosis, and asbestosis have tra -ditionally been treated with corticosteroids or immunosup-

pressants, wit h discouraging results in terms of both mo rbidityand mortality. These data, and information gleaned frommechanistic studies described above, provide an impetus forthe development of novel approaches for prevention andtreatment of mineral-induced inflammation and fibrosis. Re-cent work ha s focused, using animal models of disease, on: (1)administration of antioxidants or iron chelators, (2) inhibitionof TNF and I L-1 production or receptor interactions, (3) inhi-bition of phospholipases, and (4) modification of mineral sur-face properties.

Early elevations in mRNA levels, immunoreactive protein,and activity of antioxidant enzymes in both alveolar type IIepithelial cells (152) and lung homogenates (153) from ro-dents after inhalation of silica or asbestos indicated that oxi-

dant stress responses might be linked to lung defense fromminerals. U pregulation of ant ioxidant enzymes such as man-ganese-containing superoxide dismutase (MnSOD ) by asbes-tos or silica (153, 154) suggested that dusts generating oxidantswere importa nt in inducing MnSOD expression. Moreover,overexpression of MnSO D , achieved after tra nsfection of tra-cheal epithelial cells, prevented asbestos-induced toxicity, anobservation supporting the hypothesis that this antioxidantenzyme is linked to cell defense from mineral dusts (155).

Figur e 3. A hypothetical schema of events occurring in lung after exposure to pathogenic mineral dusts.AM alveolar macrophage; IM interstitial macrophage; F fibroblast; ROSreactive oxygen species;RNS reactive nitrogen species; TNF tumor necrosis factor; IL-1 interleukin-1; MIP macrophageinflammatory proteins; MCPmacrophage chemotactic proteins.


10/15


Since superoxide dismutase converts superoxide to hydrogenperoxide, a distal candidat e in the cascade of free ra dical reac-tions, we explored the use of cata lase as a preventive approa chto a sbestosis in a rodent inhalation model of d isease (156). Inthese experiments, the half-life of catalase in lung and serumwas enhanced by coupling the enzyme to polyethylene glycol(PE G ), and continuous administration over the 20 d of inhala-tion necessary for development of fibrosis was achieved viasubcutaneously implanted osmotic pumps. Results show thatboth inflammation and fibrosis, as measured by a number ofquantitative biochemical and morphologic end points in lungtissue and B AL , can be inhibited by cat alase in a dose-depen-dent fa shion (156). R ecently, phytic acid, an iron chelator, wa salso administered to rats at the time of intratracheal instilla-tion of asbestos (157). These experiments also demonstratedsignificant inhibition of a sbestos-associated inflamma tion a ndfibrosis using an approach that inhibited oxidant production.Although generation of ROS by dusts intrinsically or via theinflammatory process is clearly important in initiation ofdisease, whether or not antioxidant administration can be oftherapeutic value after fibrosis has developed needs furtherinvestigation. A n intriguing report showing tha t inherited glu-ta thione-S-transfera se deficiency is a risk factor for a sbestosis(158) suggests that cellular glutathione levels may also be a

source of antioxidant prot ection against mineral dusts.Links between oxidant and TNF production after mineral

exposures have been indicated in a number of studies, but re-cent work demonstrates that elaboration of R OS may be a pri-mary and necessary event in TNF production, inflammation,and fibrosis (119). In an instillation model of silicosis, pre-treatment of rats with the free radical scavenger, N-ter-butyl-alpha-phenylnitrone, inhibited production of ROS and eleva-tions in TNF mRNA levels in AMs as well as histologic evi-dence of fibrosis. These investigat ors also show ed inhibition o fsilicosis in animals treated with an anti-TNF antibody, thusconfirming successful approaches by others using either anti-bodies to TNF or human recombina nt soluble TNF receptorsto ameliorate silica- and bleomycin-induced fibrosis (137, 138).

Continuous administration of an interleukin receptor antago-nist (IL-1ra) also results in both prevention and reduction offibrosis in these models when given after fibrosis has devel-oped (138, 159). The natural diterpene compound, acanthoicacid, which reduces both TNF and IL -1 production from AM sas well as oxidant production, suppresses both gra nuloma for-mation and fibrosis after instillation of silica (160), providingfurther evidence that anti-inflammatory compounds that in-hibit both TNF and oxidant production may be pot entially an-tifibrotic in patients.

Elevated intracellular and extracellular phospholipids oc-cur during the development of experimental silicosis, but thephysiologic significance of these increases are unclear (161).Phospholipase inhibition is a potential mechanism to explainthe phospholipidosis induced by amiodarone, a cationic drugthat inhibits phospholipase activity in lung (162). A recentstudy reports that administration of amiodarone before andafter intratracheal instillation of silica into rats causes signifi-cant reductions in inflammat ion, lung dama ge, and pulmonaryfibrosis (163). These investigators propose that inhibition oflysosomal phospholipases by amiodarone prevents digestionof the norma lly protective surfacta nt coat ing of silica particles,thus preventing cell dama ge by biorea ctive inhaled silica.

Additional strategies have been used to modify the surfaceproperties and inflammatory potential of silica. For example,pretreat ment with a luminum lacta te ha s been successful in in-hibiting qua rtz-induced increases in neutrophils and release ofproteolytic enzymes from A Ms in B AL (164), and some inves-

tigators report amelioration of toxicity and increased mobili-zation of qua rtz from t he lungs in experimental a nimals treatedwith a luminum (165168). H owever, o thers have found that theprotective effects of a luminum are only tempora ry (169). Me-tallic aluminum and a lumina ha ve been used as preventive agentsin humans exposed to silica dust and appea r to red uce the inci-dence of new ca ses of silicosis, but t hey do not have a ny effecton estab lished lesions (170).

Po lyvinylpyridine-N-oxide (PVP NO) is thought to d etoxifysilica by shielding SiOH groups on the mineral surface, andwhen used alone or in combination with tetrandine, an herbaldrug used to treat pulmonary fibrosis in patients, causesboth inhibition of collagen mRNA levels and its degradationin lung (171, 172). There is some evidence that PVPNO de-creases radiographic progression of silicotic lesions in humans,but d isease progression recommences if the P VP NO is discon-tinued (173). C onceivably, mineral surfaces are modified a fterinhalation and deposition within the lung, but they may re-main bioreactive during the development and progression ofdisease.

6. S UMM ARY AND CONCLUSIONS

It is clear that both asbestosis and silicosis are f iber/particle

dose-related diseases. A sbestosis in particular req uires a veryhigh fiber burden for its development, and it appears that, ona fiber-for-fiber basis, amphibole forms of asbestos are morefibrogenic than chrysotile, perhaps reflecting the differing bio-persistence of the two fiber types. The relative potencies ofthe three clearly fibrogenic forms of crystalline silica (quartz,cristobalite, and tridymite) are less well established, althoughcertainly much greater t han coesite or a morphous silica.

Whether every single asbestos fiber or silica particle thatremains in the lung actua lly elicits production of cytokines andoxidants, or whether there is a minimum number of fibers/par-ticles required to initiate or sustain these processes is un-known, but in either case it is clear from epidemiologic, fiberburden, and experimental studies that the lung is able to deal

with a considerab le number of fibers and particles without de-tectable molecular or pathogenic events or the developmentof fibrosis (77, 81).

The complexity of various types of asbestos and silicamakes it difficult to d issect the chemical and physical feat uresof t hese mineral dusts, which are intrinsic to the initiation a nddevelopment of fibrosis after considerable exposures. How-ever, it is clear that their surface chemistry is important indriving oxidant production and possibly other deleterious re-actions in the lung that are linked to the advent of inflamma-tion a nd fibrogenesis. Size dimensions of minerals also governcellular reactions such as frustrated phagocyto sis and prolifer-ation. A t overload concentrations historically a ssociatedwith the development of asbestosis and silicosis in the work-place, durability and impaired clearance of mineral dusts mayexplain their chronic effects over time a nd progression of lungdisease in pat ients removed from the wo rkplace.

D espite the different histologic presentat ions of asbestosisand silicosis, several commonalities exist in the developmentof these diseases. B oth minerals stimulate pro duction of oxi-dant s, chemokines, and cytokines from a number o f cell types(Table 5). Several of t hese factors may act a lone or in concertto cause chemotaxis, cell injury, proliferat ion, and synthesis ofcollagen. Mechanistic studies suggest that multiple cell typesand signaling events are involved in t he pat hogenesis of a sbes-tosis and silicosis. Moreover, individual cell types may haveseveral roles in the disease process. For example, the a lveolartype II epithelial cell, regarded historically as a key target cell


11/15


in initial injury by inhaled part icles, now appea rs to be impor-tant in both defense from lung damage as well as elaborationof chemokines and cytokines.

A hypo thetical schema of events occurring in the lung af terinhalation of fibrogenic mineral dusts is shown in Figure 3.The diagram depicts possible interrelationships between theelaboration of oxidants, TNF and IL-1 production, and thedevelopment of cellular events contributing to pulmonary fi-brosis as reviewed abo ve. Various feedback loops and cross-ta lk betw een cell-signaling events may occur, making thera-peutic approaches difficult and complex. Nonetheless, it hasproven possible in some experiments to use the type o f infor-mation show n in Table 5 and Figure 3 to a meliorate the d evel-opment of fibrosis.

The limitations of our current understanding must bestressed. B oth Table 5 and Figure 3, taken a t fa ce value, wouldlead t o the conclusion that asbestosis and silicosis are identicaldiseases. That t hey are not indicates the need to ref ine our un-derstanding of the role of oxidants, chemokines, and cytokines.Are some of the mediators listed important with one dust andmerely bystanders with the other, or are the different pathol-ogy a nd geogra phic distribution o f silicosis and asbestosis a re-flection of specific mineral deposition and clearance patternsthat affect mediator production? This issue needs further study

if new approa ches to thera py are to be effective. Co ntinued in-vestigation of the mechanisms of mineral-induced pulmonaryfibrosis, which appear similar to those involved in IPF or thebleomycin model, is certainly warranted, a nd may lead to trea t-ments for IP F as well as trea tments for dust-induced disease.

One last issue, which is not addressed in this review but isextremely important in clinical disease, is the relationship be-tween mineral-dust-induced fibrosis and lung ca ncer. D iffusepulmonary fibrosis of any cause appears to be a precursor ofcarcinoma, a nd this phenomenon is well described in pat ientswith I PF o r diffuse interstitial fibrosis associated w ith collagenvascular disease (174, 175). In both humans and experimentalanimals, there is a strong association between the presence ofasbestosis and an increased risk of lung cancer (176, 177). In-

deed, in experimental animals, there is a close spacial relation-ship between asbestos-induced fibrosis and lung cancer (176).Whether an increased lung cancer risk also exists in the ab-sence of asbestosis is a matt er of considerab le deba te (177). Inrats, crystalline silica is clearly a lung carcinogen and tumorsagain arise in close spacial relationship to silicotic nodules,whereas strains of mice and hamsters that do not develop fi-brotic lesions after exposure to silica also do not show an in-creased incidence of cancers (178). The International Agencyfor Research on Cancer (IARC) recently classified crystallinesilica as a definite human carcinogen (179). This classificationis controversial, but the association between silica exposureand lung cancer in humans appears to be on much firmerground in t hose with silicosis than in those w ithout (5, 25). E x-actly w hat mechanisms of f ibrosis or fibrogenesis are critical inthe initiation of transformation are unknown, but even with-out an understa nding of these events, the importa nce of avoid-ing or decreasing dust-induced fibrot ic reactions is quite clear.

Acknowledgment: The writers thank Laurie Sabens for preparation of themanuscript. Because of space limitations, the authors have emphasized sci-entific papers published within the past 7 yr for this review, but they ac-knowledge the excellent work in this field historically by a number of inves-tigators.

References

1. Mossman, B . T., and J. B . L. G ee. 1989. Medical progress: asbestos-

related diseases. N. E ngl. J. M ed.320:17211730.

2. American Thoracic Society. 1997. Adverse eff ects of cyrsta lline silica

exposure (ATS statement). Am. J. Respir . Crit. Care M ed.155:761

765.

3. Schwart z, M. I. 1993. Interstitial pulmonary fibrosis. I nW. N. Kelley,

editor. Textbook of Internal Medicine. J. P . Lippincott C o., P hiladel-

phia . 19021905.

4. Smith, R. E., R . M. Strieter, S. H. P han, and S. L. K unkel. 1996. C-C

chemokines: novel mediators of the profibrotic inflammatory re-

sponse to bleomycin challenge. Am. J. Respir. Cell M ol. Bio l.15:693

702.

5. Weill, H ., and J . C . McD onald. 1996. E xposure to crystalline silica and

risk of lung cancer: the epidemiological evidence. Thorax51:97102.

6. American Thoracic Society. 1986. Dia gnosis of nonma lignant diseasesrelated to asbestos. Am. Rev. Respir . D is.134:363368.

7. B rowne, K. 1994. Asbestos-related disorders. I nW. R. Parkes, editor.

Occupational Lung D isorders, 3rd ed. B utterworth-Heinemann,

Ltd., Oxford. 411504.

8. Murphy, R. L . J., E. A. G aensler, B. G . Ferris, M. Fitzgerald, N. Solli-

day , and W. Morrisey. 1978. D iagnosis of asbestosis : observations

from a longitudinal survey of shipyard pipe coverers. Am. J. Med.65:

488498.

9. Medical Advisory Pa nel to the Asbestos International Association.

1982. Criteria for the diagnosis of asbestosis and considerations in

the attribution of lung cancer and mesothelioma to asbestos expo-

sure. Int. A rch. Occup. Environ. H ealth49:357361.

10. Huuskonen, M. S. 1978. Clinical features, morta lity and survival of pa-

tients with asbestosis. Scand. J. Work E nvir on. Health 4:265274.

11. Staples, C. A. 1992. Co mputed tomography in the evaluation of benign

asbestos-related disease. Radiol. Clin. North A m.30:11911207.

12. Muller, N. L., a nd R . R . Miller. 1990. Co mputed tomography of chronicdiffuse infiltrative lung disease. Am. Rev. Respir . D is. 142:12061215,

14401448.

13. Ha rber, P., and J. Smitherman. 1991. Asbestosis: diagnostic dilution. J.

Occup. Med.33:786793.

14. Craighead, J. E., J. L. Ab raham, A. Churg, F. H. Y. G reen, J. Kleiner-

man, P. C. Pratt, T. A. Seemayer, V. Vallyathan, and H. Weill. 1982.

The pathology of asbestos-associated diseases of the lungs and pleu-

ral cavities: diagnostic criteria and proposed grading schema. Arch.

Pathol. Lab. M ed.106:544596.

15. Hinson, K. F. W., H. O tto, I. Webster, and C . E . Ro ssiter. 1973. Criteria

for the diagnosis and grading of asbestosis. I nP. B ogovski, J . C. G il-

son, V. Timbrell, and J. C. Wagner, editors. B iological Ef fects of As-

bestos. P ergamon P ress, Oxford. 5457.

16. D upres, J. S., J. F. Mustard, and R . J. U ffen. 1984. Report of the R oyal

Commission on matters of health and safety arising from the use of

asbestos in Onta rio. Queens Printer for Ontario, Toronto.

17. Weill, H. 1994. Asbestos cement manufa cturing. A nn. Occup. Hyg.38:

533538.

18. Jacobsson, K., U. Stromberg, M. Albin, H. Welinder, and L. Ha gmar.

1995. Radiological changes in asbestos cement workers. Occup. En-

viron. M ed. 52:2027.

19. Coo kson, W., N. de Klerk, A. Musk, B. Armstrong, J. G lancy, and M.

Hobbs. 1986. The natural history of asbestosis in former crocidolite

workers of Wittenoom. Am. Rev. Respir . D is.133:994998.

20. B ecklake, M. R. 1983. Occupationa l lung disease: past record and fu-

ture trend using the asbestos case as an example. Cli n. Invest. Med.6:

3053l7.

21. Weiss, W. 1984. Ciga rette smoke, a sbestos, and small irregular opaci-

ties. Am. Rev. Respir . D is.130:293301.

22. B arnhart, S., M. Thornquist, G . S. Omenn, G . G oodman, P. Feigl, and

L. Rosenstock. 1990. The degree of roentgenographic parenchymal

opacities attributable to smoking among asbestos-exposed subjects.

Am. Rev. Respir . D is.141:11021106.23. B lanc, P. D., J . A. G olden, G. G amsu, D. R . Aberle, and W. M. Gold.

1988. A sbestos exposure-cigarette smoking interactio ns amo ng ship-

yard workers. J.A.M .A.259:370373.

24. G ibbs, A. R., and J . C. Wagner. 1988. D iseases due to silica. I n A.

Churg and F. H. Y. G reen, editors. Pa thology of Occupational Lung

D iseases. Igaku-Shoin Medica l P ublishers, New Yo rk. 155176.

25. Weill, H., R . N. Jones, a nd W. R. P arkes. 1994. Silicosis and related dis-

eases. I nW. R . Pa rkes, editor. O ccupational Lung D isorders, 3rd ed.

B utterwo rths, Lo ndon. 285339.

26. B ecklake, M. R. 1989. Occupationa l exposures: evidence for a causal

association with chronic obstructive pulmonary disease. Am. Rev.

Respir . D is.140:S85S91.

27. Oxman, A. D., D . C. F. Muir, H. S. Shannon, S. R. Stock, E. Hnizdo,

and H. J. Lange. 1993. Occupational dust exposure and chronic ob-

structive pulmonary disease: a systematic overview of the evidence.


12/15


Am. Rev. Respir . Di s.148:3848.

28. B egin, R., G . Ostiguy, A. Ca ntin, and D. B ergeron. 1988. Lung function

in silica-exposed w orkers. Chest94:539545.

29. Veblen, D. R ., and A. G . Wylie. 1993. Mineralogy of amphiboles and

1:1 layer silicates. I nG . D. G uthrie, Jr. and B . T. Mossman, editors.

Hea lth Eff ects of Mineral Dusts. Reviews in Mineralogy, Vol. 28.

Mineralogical So ciety of A merica, Washington, D C. 61138.

30. Ho chella, M. D., Jr. 1993. Surface chemistry, structure, and reactivity of

hazardous mineral dust. I n G . D. G uthrie, Jr. and B. T. Mossman,

editors. H ealth Ef fects of Mineral D usts. R eviews in Mineralogy,

Vol. 28. Mineralogical Society o f America, Washington, D C. 275

308.31. Ro ss, M. P., R . P. Nolan, A. M. L anger, and W. C. C ooper. 1993.

Hea lth effects of mineral dusts other than asbestos. I nG . D. G uthrie,

Jr. and B . T. Mossman, editors. Health Eff ects of Mineral D usts. Re-

views in Mineralogy, Vol. 28. Mineralogical Society of America,

Washingt on, D C . 361409.

32. Mossman, B . T., J. B ignon, M. Corn, A. Seaton, and J. B . L. G ee. 1990.

Asbestos: scientific developments and implications for public policy.

Science247:294301.

33. Churg, A. 1993. Asbestos-related disease in the workplace and enviro n-

ment: con troversial issues.Monogr. Pathol.36:5477.

34. Jaurand, M. C., A. G audichet, S. Halpern, and J. Bignon. 1984. In vitro

biodegradation of chrysotile fibres by alveolar macrophages and me-

sothelial cells in culture: comparison with a pH effect. Br . J. Ind. M ed.

41:389395.

35. Ha rdy, J. A ., and A. E . Aust. 1995. Iron in asbestos chemistry and car-

cinogenicity. Chem. Rev.95:97118.

36. H ume, L. A., and J. D . Rimstidt. 1992. The biodurability of chrysotileasbestos. Am. Mineral.77:11251128.

37. Light, W. G ., and E . T. Wei. 1977. Surface charge a nd asbestos toxicity.

Nature265:537539.

38. Churg, A. 1994. Depo sition and clearance of chrysotile asbestos. Ann.

Occup. Hyg.38:625633.

39. Coin, P. G ., V. L. Roggli, and A. R. B rody. 1992. D eposition, clearance,

and translocation of chrysotile asbestos from peripheral and central

regions of the lung. Environ. Res.58:97116.

40. Vallyathan, V., X. Shi, N. S. Da lal, W. Irr, and V . Ca stranova. 1988.

G eneratio n of free radicals from freshly fractured silica dust: poten-

tial role in acute silica-induced injury. Am. Rev. Respir. D is.138:1213

1219.

41. Wiessner, J. H., J. D . Henderson, Jr., P. G . Sohnle, N. S. Mandel, and

G . S. Ma ndel. 1988. The effect o f crystal structure on mo use lung in-

flammation a nd fibrosis. Am. Rev. Respir . D is.138:445450.

42. Tourmann, J.-L., and R . Kaufma nn. 1994. B iopersistence of the min-

eral mat ter of co al mine dusts in silicotic human lungs: is there a pref-

erential release of iron? Enviro n. Health Perspect.102(Suppl. 5):265

268.

43. Wallace, W. E., M. J. K eane, J. C . Ha rrison, J. W. Stephens, P. S.

B rower, R. L . G rayson, and M. D . Attfield. 1995. Surface properties

of silica in mixed dusts. I nV. Castranova, V. Vallyathan, and W. E.

Wallace, editors. Silica and Silica-induced D isease. CRC Press, Bo ca

R at on, FL . 107120.

44. Adamson, I. Y . R., and D . H. B owden. 1987. Response of mouse lung

to crocidolite asbestos: II. Pulmonary fibrosis after long fibres. J.

Pathol.152:109117.

45. Lema ire, I. 1985. An assessment of th e fibrogenic potential of very

short 4-T30 chrysotile by intratracheal instillation in rats. Environ.

Res.36:314326.

46. D avis, J. M., and A. Jones. 1988. Comparison of the pathogenicity of

long and short fibres of chrysotile asbestos in rats. Br . J. Exp. Pathol.

69:717737.47. D avis, J . M. G ., J . Addison, R. E . Bo lton, K. D onaldson, A. D . Jones,

and T. Smith. 1986. The pat hogenicity of long vs. short fibre samples

of amosite asbestos administered to rats by inhalation and intraperi-

toneal injection. Br. J. Exp. Pathol.67:415430.

48. Platek, S., D. G roth, C. U lrich, L. Stettler, M. Finnell, and M. Stoll.

1985. Chronic inhalation of short asbestos fibres. Fundam. A ppl.

Toxicol.5:327340.

49. Donaldson, K., G. M. B rown, D. M. Brown, R. E. Bolton, and J. M. G .

D avis. 1989. Infla mmation genera ting potential of long and short f i-

bre amosite asbestos samples. Br . J. Ind. M ed.46:271276.

50. Mossman, B. T., K. Ha nsen, J. Marsh, M. E. B rew, S. Hills, M. Berg-

eron, a nd J . Pe truska. 1989. Mechan isms of f iber-induced superoxide

(O 2) release from alveolar macrophages (AM) and induction of su-

peroxide dismutase (SOD ) in the lungs of rats inhaling crocidolite. I n

J. B ignon, J. Peto , and R . Saracci, editors. Mineral Fibres in the Non-

Occupational Environment. I.A.R.C . (I.A.R .C. P ublication No. 90),

Lyon, France. 8197.

51. G ylseth, B., A. Churg, J. M. G. D avis, N. Johnson, A. Morgan, G .

Mowe, A. Roggers, and V. Roggli. 1985. Analysis of asbestos fibers

and asbestos bodies in human lung tissue samples: an international

laboratory trial. Scand. J. Work E nvir on. Health11:107110.

52. Churg, A., a nd J. W. Wright. 1994. Persistence of natura l mineral fibers

in human lungs: an overview. Enviro n. Health Perspect. 102(Suppl.

5):229233.

53. G ibbs, A. R., and F. D . Poo ley. 1996. Analysis and interpretation of in-

organic mineral partcles in lung tissues. Thorax31:327334.

54. Churg, A., and S. Veda l. 1994. Fiber burden and patterns of asbestos-related disease in workers with heavy mixed amosite and chrysotile

exposure. Am. J. Respir . Crit. Care M ed.150:663669.

55. Roggli, V. L., P. C. Prat t, and A. R . Brody. 1986. Asbestos content of

lung tissue in asbestos associated disease: a study of 110 cases. Br . J.

Ind. M ed.43:1829.

56. Churg, A., J . L. Wright, and S. Veda l. 1993. Fiber burden and patterns

of d isease in chrysotile miners and millers. Am. Rev. Respir . D is.148:

2531.

57. Churg, A . 1988. Non-neoplastic asbestos-induced disease. I nA. Churg

and F. H. Y . G reen, editors. Pa thology of Occupational Lung Dis-

eases. Iga ku-Shoin M edical P ublishers, New Yo rk. 279326.

58. Wagner, J. C., M. L. Newhouse, B. Co rrin, C. E. R . Rossiter, and D. M.

G riffiths. 1988. Correlation bet ween fibre content o f the lung and

disease in East L ondon factory workers.Br. J. Ind. M ed.45:305308.

59. G reen, F. H. Y., R . Ha lrye, V. Vallyathan, J. D ement, J. Pooley, and R.

Althouse. 1987. Pulmonary fibrosis and a sbestos exposure in chryso-

tile asbestos textile workers. I n J. C. Wagner, editor. The B iologicalEffects of Chrysotile. J. B. Lippincott, Philadelphia, PA. 5968.

60. Churg, A., J. L. Wright, L. D ePaoli, and B . Wiggs. 1989. Mineralogic

correlates of fibrosis in chrysotile miners and millers. Am. Rev. Re-

spir . Di s.139:891896.

61. Morgan, A. 1980. E ffect of length on the clearance of fibres from the

lung and on bo dy formation. I nJ. C . Wagner, editor. Biological Ef-

fects of M ineral Fibers. I.A.R .C., L yon, Fra nce. 329336.

62. Adamson, I. Y. R ., and D . H. B owden. 1987. Response of mouse lung

to crocidolite asbestos: I. Minimal fibrotic reaction to short fibres. J.

Pathol.152:99107.

63. Churg, A., J . Wright, B . Wiggs, and L. D ePaoli. 1990. Mineralogic pa-

rameters related to amosite asbestos induced fibrosis in man. Am.

Rev. Respir . D is.142:13311336.

64. Tron, V., J. L . Wright, N. Ha rrison, B. Wiggs, and A. Churg. 1987. Cig-

arette smoke makes airway and early parenchymal asbestos-induced

lung disease worse in the guinea pig. Am. Rev. Respir. D is. 136:271

275.

65. Churg, A., J. L. Wright, J. H obson, and B . Stevens. 1992. Ef fects of cig-

arette smoke on the clearance of long and short asbestos fibers from

the lung. Int. J. Exp. Pathol.73:287297.

66. Churg, A., and B . Stevens. 1995. E nhanced retention of asbestos fibers

in the airways of human smokers. Am. J. Respir . Crit. Care M ed.151:

14091413.

67. B egin, R., and P. Sebastien. 1989. Excessive accumulatio n of asbestos

fibre in the bronchoalveolar space may be a marker of individual sus-

ceptibility to developing asbestosis: experimental evidence. Br. J.

Ind. M ed.46:853855.

68. B egin, R., A. C antin, and P . Sebastien. 1990. Chrysotile asbestos expo-

sures can produce an alveolitis with limited fibrosing activity in a

subset of high fibre reta iner sheep. Eu r. Respir . J.3:8190.

69. B ecklake, M. R., B. Toyota, M. Stewart, R. H anson, and J. Janley.

1983. Lung structure a s a risk facto r in adverse pulmona ry responses

to asbestos exposure. Am. Rev. Respir . D is.128:385388.70. Nagelschmidt, G . 1960. The relation betw een lung dust and lung pa-

thology in pneumoconiosis. Br. J. Ind. M ed.17:247259.

71. Za idi, S. H., E. J . King, C. V. H arrison, and G . Nagelschmidt. 1956. Fib-

rogenic activity of different forms of free silica. A rch. Ind. H ealth13:

112121.

72. King, E. J., G . P. Moha nty, C. V. H arrison, and G . Nagelschmidt. 1953.

The action of different forms of pure silica on the lungs of rats. Br. J.

Ind. M ed.10:917.

73. King, E. J., G . P. Moha nty, C. V. H arrison, and G . Nagelschmidt. 1953.

The action of flint of variable size injected at constant weight and

constant surface into the lungs of rats. Br . J. Ind. M ed.10:7692.

74. Wiessner, J. H ., N. S. Mandel, P. G . Sohnle, and G . S. Mandel. 1989. Ef -

fect of particle size on quartz-induced hemolysis and on lung inflam-

mation and fibrosis. Exp . Lung Res.15:801812.

75. Ka mp, D . W., and S. A. Weitzman. 1997. Asbestosis: clinical spectrum


13/15


and pathogenic mechanisms. Proc. Soc. Exp. Bio l. M ed.214:1226.

76. Craighead, J. E ., J. Kleinerman, and J . L. Abraha m. 1988. Diseases as-

sociated with exposure to silica and nonfibrous silicate minerals.

A rch. Pathol. L ab. M ed.112:673790.

77. Quinlan, T. R., J. P . Marsh, Y. M. W. Janssen, K. O. Leslie, D. H emen-

way , P. Va cek, and B . T. Mossman. 1994. D ose responsive increases

in pulmonary fibrosis after inhalation of asbestos. Am. J. Respir . Crit.

Care Med.150:200206.

78. Velan, G . M., R. K. K umar, and D . D. C ohen. 1993. Pulmonary inflam-

mation and fibrosis following subacute inhalational exposure to sil-

ica: determinants of progression. Pathology25:282290.

79. Lehnert, B . E . 1993. D efense mechanisms against inhaled particles andassociated particle-cell interactions. I nG . D. G uthrie, Jr. and B. T.

Mossman, editors. Health E ffects of Mineral D usts. R eviews in Min-

eralogy, Vol. 28. Mineralogical Society of America, Washington,

D C . 427470.

80. Ob erdorster, G . 1994. Macro phage-associated responses to chrysotile.

Ann. Occup. Hyg.38:601615.

81. Quinlan, T. R., K . A. B eruBe, J. P . Marsh, Y. M. W. Janssen, P. Taishi,

K. O . Leslie, D . H emenway, P. T. OShaughnessy, P. Vacek, and

B . T. Mossman. 1995. Pat terns of inflammatio n, cell proliferation

and related gene expression in lung after inhalation of chrysotile as-

bestos. Am. J. Pathol.147:728739.

82. D avis, G . S., K. O. Leslie, J. E. Schwarz, L. M. Pfeiffer, L. Hill-

Eubanks, and D. Hemenway. 1993. Altered patterns of lung lympho-

cyte a ccumulation in silicosis in cytokine-sufficient (C 3H/HeN) and

cytokine-deficient (C3H/H eJ-LP Sd) mice. Chest103(Suppl. 2):120S

121S.

83. Wagner, M., J . C. Wagner, and M. G riffiths. 1994. Mast cells and inha-lation of asbestos in rats. Thorax39:539544.

84. Suzuki, N., T. H oriuchi, K. O hta, M. Ya maguchi, T. U eda, H . Tak-

izawa, K. Hirai, J. Shiga, K. Ito, and T. Miyamoto. 1993. Mast cells

are essential for t he full development of silica-induced pulmonary in-

flammation: a study with mast cell-deficient mice. Am. J. Respir . Cell

Mol. Biol.9:475483.

85. Lesur, O., T. B ouhadiba, B . Melloni, A. Cantin, J. A. Whitsett, and R .

B egin. 1995. Altera tions of surfactant lipid turnover in silicosis: evi-

dence of a role for surfactant-associated protein A (SP-A). I nt. J.

Exp. Pathol.76:287298.

86. B eruBe, K. A., T. R. Quinlan, G . Moulton, D. H emenway, P. OShaugh-

nessy, P. Vacek, and B . T. Mossman. 1996. Compa rative proliferat ive

and histopathologic changes in rat lungs after inhalation of chrysotile

or crocidolite asbestos. Toxicol. Appl. Pharmacol.137:6774.

87. D ixon, D., A. D . Bow ser, A. Badgett, J . K. Haseman, and A. R. B rody.

1995. Incorporat ion of bromodeo xyuridine (B rdU ) in the bronchi-

olar-alveolar regions of the lungs following two inhalation exposures

to chrysot ile asbestos in strain A /J mice. J. Environ. Pathol. Toxicol.

Oncol.14:205213.

88. D onaldson, K., D. M. B rown, B. G . Miller, and A. R . B rody. 1995. B ro-

modeoxyuridine (BR D U ) uptake in the lungs of rats inhaling am-

osite asbestos or vitreous fibers at equal airborne fibre concentra-

tions. Exp. T oxicol. Pathol.47:207211.

89. Sekhon, H., J. Wright, and A. C hurg. 1995. Effects of cigarette smoke

and asbestos on airway, vascular and mesothelial cell proliferation.

I nt. J. Exp. Pathol.76:411418.

90. Janssen, Y. M. W., K. E . D riscoll, B . How ard, T. R. Q uinlan, M. Tread-

well, A. B archow sky, and B . T. Mossman. 1997. Asbestos causes

translocation of p65 protein and increases NF-B DNA binding ac-

tivity in rat lung epithelial and pleural mesothelial cells. Am. J.

Pathol.151:389401.

91. H eintz, N. H., Y. M. Ja nssen, and B . T. Mossman. 1993. Persistent in-

duction of c-fo sand c-junexpression by asbestos. Proc. Natl. A cad.Sci. U .S.A .90:32993303.

92. Janssen, Y. M. W., N. H. Heintz, J. P. Marsh, P. J. A. B orm, and B. T.

Mossman. 1994. Induction o f c-fosand c-jun protooncogenes in tar-

get cells of the lung and pleura by carcinogenic fibers. Am. J. Respir .

Cell M ol. Biol.11:522530.

93. Janssen, Y. M. W., A. B archowsky, M. Treadwell, K. E. D riscoll, and

B . T. Mo ssman. 1995. A sbestos induces nuclear fa ctor B (NF-B )

D NA-binding activity and NF-B -dependent gene expression in tra-

cheal epithelial cells. Proc. Natl. Acad. Sci. U.S.A.92:84588462.

94. Angel, P., a nd M. Ka rin. 1991. The role of Jun, Fos and the AP -1 com-

plex in cell proliferation and transformation. Biochim. B iophys. A cta

1072:129157.

95. Timblin, C. R ., Y. M . W. Janssen, and B . T. Mossman. 1995. Transcrip-

tional activation of the proto-oncogene c-jun, by asbestos and H 2O 2is directly related to increased proliferation and transformation of

trachea l ep

asbestosis and silicosis

Documents