- Email: [email protected]
Fibronectin and Procollagen 3 Levels in Bronchoalveolar Lavage of Asbestos-Exposed Human Subjects and Sheep
Fibronectin and Procollagen 3 Levels in Bronchoalveolar Lavage of AsbestosExposed Human Subjects and Sheep* R. Begin, M.D., F.G.G.P.; M. Martel, B.Sc.; Y Desmarais, B.Sc.; G. Drapeau, M.Sc.; R. Boileau, M.D., F.G.G.P.; M. Rola-Pleszczynski, M.D.; and S. Masse, M.D.
To evaluate the potential interest oflevels ofSbronectin and procollagen 3 in bronchoalveolar lavage 8uid as markers of 6brogenic activity, we characterized the time course of changes in 6bronectin and procollagen 3 levels in the tracheal lobe of sheep exposed to nonSbrogenic and 6brogenic materials. We correlated these observations with those ofbronchoalveolar lavage in long-term asbestos workers in various stages of disease activity. Following studies before exposure, the tracheal lobe of three groups of 24 sheep were exposed once to 100 mI of phosphate-buffered saline solution (PBS~ to 100 mg of latex beads in 100 mI of PBS, or to 100 mg of chrysotile Sbers in 100 mI of PBS. Bronchoalveolar lavages were obtained at 0, 1, 2, 4, 8, and 12 months after exposure, and four or Sve sheep per group were 1dIIed after each lavage for histopathologic analysis. Fibronectin in bronchoaIveolar lavage 8uid increased significantly only in the asbestos-exposed sheep to values two to
three times above controls or latex-exposed sheep and remained elevated during the 12 months ofthe study. Levels of procollagen 3 in bronchoalveolar lavage 8uid were increased signi6cantly only during the first two months following exposure in the asbestos-exposed sheep only. In the asbestos workers without disease, levels of 6bronectin and procoUagen 3 in bronchoalveolar lavage 8uid were comparable to controls, but these levels were signi6cantly elevated in those with asbestos-associated a1veolitis or asbestosis. This study documents that the measurement of levels of 6bronectin and procollagen 3 in bronchoalveolar lavage 8uid assesses 6brogenic activity of alveolitis and should be useful to predict its progression in a 6brotic process. In asbestos workers the potential use of these markers is primarily related to early detection of asbestos-induced pulmonary injury.
Until recently, the assessment of 6brogenic activity of human interstitial pulmonary diseases was limited to the histomorphologic and immunohistochemical analyses of limited samples of pulmonary tissues. l -3 Through bronchoalveolar lavage of diseased areas, it has been documented that human pulmonary diseases with histologic evidence of 6brogenic activity were associated in the bronchoalveolar lavage fluid with increased levels of fibronectin and procollagen 3 peptides, molecules implicated in the biochemical process of pulmonary fibrosis,4-8 and thus of potential interest as markers of fibrogenic activity; however, these observations were limited to the individuals with chronic disease. The time course of these changes in bronchoalveolar lavage fluid have not been studied in spite of increasing interest in the biologic monitoring of humans exposed to environmental materials or therapeutic drugs with known pulmonary toxic effects. In the present study, we characterized the time course of these changes in fibronectin and procollagen 3 levels in bronchoalveolar lavage fluid from the sheep
tracheal lobe exposed to nonfibrogenic and fibrogenic materials. These observations were correlated with those of bronchoalveolar lavage in long-term asbestos workers in various stages of disease activity. The data clearly document that fibronectin and procollagen 3 levels in bronchoalveolar lavage fluid are increased early in alveolitis with fibrogenic activity, but not in those without, which should contribute to further refine our clinical understanding of disease activity in the chronic inflammatory pulmonary disorders.
*From the Unite de Recherche Pulmonaire, Centre Hospitalier Universitaire, Universite de Sherbrooke, Sherbrooke, Quebec, Canada. Supported by I.R.S.S.l: and I.R.D.A. Quebec and M.R.C. Canada. Manuscript received April 10; revision accepted August 12. Reprint requests: Dr: Begin, CHUS, Sherbrooke, Quebec, Canada ]lH 5N4
MATERIALS AND METHODS
Animals Seventy-two male sheep weighing between 25 and 45 kg were used in this study. They were prepared and accustomed to the pulmonary techniques as previously reported. ~13 Experimental Design The flock was divided into three groups of 24 sheep. Following studies before exposure (control studies), the sheep tracheal lobe was exposed once to 100 ml of phosphate-buffered saline solution (PBS) (control group), to 100 mg of0.1.., latex beads (Sigma Chemical Co) in 100 ml of PBS (latex group), or to 100 mg of UICC. Canadian chrysotile asbestos fibers in 100 ml of PBS (asbestos group). These asbestos fibers were relatively uniform and well characterized, 14 with 92 percent less than 0.25.., in diameter and 20.., in length. Exposure ofthe tracheal lobe was carried out via bronchoscopic catheterization of the tracheal bronchus and slow infusion ofthe suspension into the lobe. The animals were studied prior to exposure and at 1, 2, 4, 8, and 12 months after exposure by bronchoalveolar lavage and by CHEST I 89 121 FEBRUARY. 1986
237
histopathologic methods (four or five sheep per group killed each time).
Subsets of Asbestos Workers On the basis ofcriteria reported in detail,15-19 the asbestos workers were divided into three categories ofdisease: (1) group A consisted of seven workers without evidence of asbestos-induced pulmonary damage; they did not have the diagnostic criteria for asbestosis, and their 87Ga pulmonary uptake and pulmonary pressure-volume curve were within normallimits;m.21 (2) group B was eight asbestos workers with evidence of asbestos-induced alveolitis documented at biopsy in three of three15-19 in whom the biopsy was obtained; they did not have the diagnostic criteria for asbestosis but had enhanced 87Ga pulmonary uptake and rigid pulmonary pressure-volume curve; and (3) group C was 19 asbestos workers who met the diagnostic criteria for asbestosis. The clinical data and bronchoalveolar lavage cellularity of these patients are summarized in Figure 1.
Human Controls Twenty-one manual workers without exposure to asbestos were tested within the period. They were matched for age, height, sex, smoking, and work habits to the asbestos workers; seven were current or ex-smokers, and 14 were lifetime nonsmokers. The mean age ofcontrol subjects was 57 ± 5 years (range, 50 to 70 years). None of them was exposed to environmental dust at risk of pneumoconiosis. In this control population, total cells in bronchoalveolar lavage fluid averaged 171 ± 24 X 1()3/ml, with macrophages at 159 ± 22 x 1()3/ml, and neutrophils at 0.86 ± 0.30 x 1()3/ml.
Asbestos Workers
Bronchoalveolar Lavage and Fluid Analysis
The 34 asbestos workers were between the ages of42 and 70 years (mean, 58±4 years). All had been exposed to Canadian chrysotile asbestos only in the mines and mills of the eastern townships of Quebec for an average of33 ± 5 years (range, 24 to 42 years). Twentyeight were nonsmokers for more than two years, and the six others were either current or recent ex-smokers. Many of these patients were included in our studies ofaiJway function in asbestos workers, 15 or computerized tomographic scanning study16 and were among 226 asbestos workers in a study of clinical features of the associated alveolitis. 17 In this present study, we specifically report on fibronectin and procollagen 3 in bronchoalveolar lavage fluid from these 34 workers as we relate these findings to serial measurements of these markers in the sheep model of asbestosis.
Most ofthe techniques in the procedures and analyses ofbronchoalveolar lavage have been previously described.8-lJ.18 The effiuent from lavage was passed through four layers ofcheesecloth to remove mucus, and the cells were pelletized by centrifugation. Cells were counted in a hemocytometer, and cell viability was determined by the trypan blue exclusion technique. Cytocentrifuge smears served to identify the cellular populations recovered with the WrightGiemsa and naphthyl acetate esterase stains. 12 In the supernatant, albumin and fibronectin were measured by the immunochemical methods of Killingsworth and Savory,13 using a laser nephelometer instrument (Behring LN modular system). For sheep albumin, specific antiserum raised in rabbit were obtained commercially (Cappell Lab. Inc.). Sheep fibronectin from bronchoalveolar lavage 8uid was purified by affinity column followed by chromatography and antisheep fibronectin antibodies prepared in rabbits as described.... 25 Pr0c01lagen 3 from bronchoalveolar lavage fluid was measured as type 3 procollagen N-terminaI peptides by radioimmunoassay, as reported by Low et al, 8 based on the method originally
Clinical Evaluation All patients underwent a medical history, complete physical examination, standard high-kilovoltage posteroanterior, lateral, and oblique chest films, detailed tests of pulmonary function, and 87Ga lung scan as previously reported. 15-19
.ms
CLINICAL DATA ON ASBESTOS G.-87
LUNG P-V CURVE
~
I
I
til
2.'
JI
I' "
JI
I
, •• c
t •
t •
..
..
..
•• IC
••• C
I A' C
lAIC
LUNG LAVAGE CELLtuRlTY
TOTAL CELL 1 M 10
i ·
~
~
••• C
c:1IH10
40
DIFFUSION CAPACITY
J
J
J I.'
'.1
••
VITAL CAPACITY
TOTAL LWIC CAPACITY
RALEB 1.1
II
I •••
t.
t ••
~
LlMG a.TAICE
I.'
12.
PARENCHYItAL IPACITY
-6
TOTAL CELL 480
1
M
10
-3
/...
MACROPHAGE
"ACROPHAGE
NEUTROPHIL
S
1 M 10
S
100
480
-3
/...
11
NEUTROPHIL 5.0
1
30
350
7S
3.0
20
240
50
240
3.0
10
120
120
1.5
". I C
I. I C
10
-3
/ ...
4.5
12
lAIC
M
• AI C
lAIC
FIGURE 1. Clinical, functional, radiographic and bronchoaIveolar lavage cellularity data of our patients. In group N (normal controls), there were 21 manual workers without asbestos exposure. Group ~ was seven asbestos workers without any evidence of asbestos-induced pulmonary damage; group B was eIght workers with increased mCa pulmonary uptake and rigid pulmonary pressure-volume (P-V) curve without criteria for established asbestosis; group C was 19 asbestos workers who met diagnostic criteria for asbestosis. 15-18 mGa pulmonary uptake, parenchymal opacities, and rales were scored as previously reported.15-18 Data on pulmonary function are reported as percent predicted of Bates et aI. 1O 238
SAL of Asbestos-expose Human Subjects and Sheep (Begin et aI)
gen 3 did not change. In the latex-exposed sheep, we have previously reported a clear increase in total cells from lavage at month 1 to eight times the control value, lJ,13 and at month 12, this had returned to the level of saline-exposed sheep. From that point and after; there was no significant difference in levels of macrophages, neutrophils, fibronectin, and procollagen 3 between latex-exposed and saline-exposed sheep. In the asbestos-exposed sheep, total cells were clearly increased through the experiment; and one year after exposure, total cells in bronchoalveolar lavage fluid remained at 150 percent of controls, macrophages at 150 percent of controls, and neutrophils at 285 percent of controls. Fibronectin in bronchoalveolar lavage fluid was significantly elevated only in the asbestos-exposed sheep, and the levels increased throughout the experiment. Procollagen 3 in lavage fluid increased significantly early and returned to control levels at month 4 after exposure (Fig 2). Albumin was increased significantly only after month 4, and it paralleled the second increase in fibronectin (data not shown). Pulmonary Histopathology. In the saline-exposed
described by Rohde et al. 1S All results were expressed per milliliter ofbronchoalveolar lavage fluid. All values ofhumoral components of lavage fluid were also analyzed in terms of the ratio to the albumin content of bronchoalveolar lavage fluid supernatant. Histopathology
At months 1, 2, 4, 8, and 12 ofthe study, four sheep in each group were killed and the lungs removed from the thoracic cavity. The tracheal lobe was identified, and nine samples of the lobe were obtained each time for microscopic examination. The pulmonary samples were processed as routinely done for human pulmonary tissue. Statistical Analysis
All results are expressed as mean ± SEe The data were tested by Students t-test for differences between groups. A value of p < 0.05 was considered significant in this study. RESULTS
Sheep Bronchoalveolar Lavage. The results of analysis of bronchoalveolar lavage fluid are presented in Figure 2. In the saline-exposed sheep, total cells from lavage averaged 250,OOO/ml and did not vary significantly in the course of the experiment. Similarly, the levels of macrophages, neutrophils, fibronectin, and procolla-4
TOTM. CELL, INIO •••
-t
MACROPHAGE,
120
to
-4
INIO •••
-t
-4
NEUTR.-etIL. INIO •••
-t
U
.-tr
U
*
30 1
15
.5
12
12
12 ....TH
....TH
"ONTH ~
TOT-. CELL, iNiO
40
75
30
50
20
25
10
12
12
-I
120
PROCOLLAC£N I I I.
12
....TH
tIONTH
....TH F I.cJNECT IN. ua•• '
NEUTROPHIL. • 0# TOT-. CELL
MCROPHACE • • 0# TOT-. CELL 100
na •• 1
-t
1.0
*'tt
to
LO
••0
2.0
30
....TH
FIGURE 2. Analyses ofbronchoalveolar lavage fluid in sheep model. Sheep were exposed once at one to seven days after control lavage (month 0) either to 100 ml ofPBS (solid squares), to 100 mg ofO.1lA-latex beads in 100 ml of PBS (open squares), or to 100 mg of U. I. C. C. chrysotile in 100 ml of PBS (triangles). Star indicates p
CHEST I 89 I 2 I FEBRUARY, 1986
239
FIGURE 3. Latex-exposed tracheal lobe at month 4, showing early pneumonia-like reaction with preservation of airway structures (hematoxylin-eosin. original magnification x 160).
sheep, the histopathology of the tracheal lobe remained nonnal during the experiment. In the latexexposed sheep at month 1, there was a pneumonia-like infiltrate of the tracheal lobe;li the infiltrate was predominantly alveolar and was largely composed of macrophages without distortion of the pulmonary architecture. At month 4, the infiltrate had regressed, and areas of residual cell accumulation constituted less than 5 percent of the total lobe (Fig 3). After month 4, the histopathology of the latex-exposed tracheal lobes had returned to nonnal (Fig 4). In the asbestos-exposed sheep at month 1, there was also a pneumonialike infiltrate composed of macrophages and neutrophils, with considerable distortion of the architecture of the lung and airway;li at month 4, the infiltrate had considerably regressed to some 20 percent ofthe initial extent; it was located predominantly in the peribronchiolar and endobronchiolar areas, with persistent severe distortion of the small airways (Fig 5). At month 8, the infiltrate continued to regress but peribronchiolar and endobronchiolar fibrosis could be detected; at month 12, the infiltrate was at 5 to 10 percent of the initial extent, but peribronchiolar and endobronchiolar fibrosis were severely distorting the pulmonary architecture (Fig 6). Human Subjects
Clinical Data. On the basis of clinical, radiologic, and functional data, 19 of the 34 subjects were recog240
FIGURE 4. Latex-exposed tracheal lobe at month 12. showing nearly complete resolution of process seen in Figure 3 (hematoxylin-eosin, original magnification x 63).
FIGURE 5. Representative histopathologic findings in asbestosexposed tracheal lobes at month 4. Early pneumonia-like reaction is associated with severe structural distortion of most small airways (hematoxylin-eosin, original magnification x 160). SAl 01 Asbesto8-expoee Human Subjects end ShMp (BegIn et eI)
PROCOLLAGEN/A
FIBRONECTIN/A ugh,g 100
3.2
ng/lltg
1
2.4
75
....
....
0.8
~
!
....
~ ~
''"" ........ NAD C
NAB C
FIGURE 7. Levels of fibronectin and procollagen 3 in human bronchoalveolar lavage Huid reported as ratios to albumin (A). Groups N, A, B, and C are as in Figure 1.
nary uptake (group A), the 6bronectinlalbumin ratio
was 33.1j.Lg/mg±6.0j.Lg/mg, and the procollagen 31 albumin ratio was 1.03±0.24 ng/mg. In group B, the 6bronectin albumin ratio was 45.7j.Lg/mg ± 34j.Lg/mg
(not signi6cant), and the procollagen 3/albumin ratio
was 2.3 ± 0.8 ng/mg (p<0.05 vs controls and group A).
FIGURE 6. Histopathologic findings in asbestos-exposed tracheal lobe at month 12, demonstrating less intense in8ammatory process than at month 4 (Fig 5), but lesions of small airways did not regress and became fibrotic (hematoxylin-eosin, original magnification x
63).
nized as having asbestosis. In comparison with the other 15 workers and the controls, this group had identical age and smoking habits. They had a signi6cantly higher rale score, higher score of radiographic pulmonary opacities, lower vital capacity, lower total lung capacity (1'1£), lower diffusing capacity for carbon monoxide, higher 67Ga pulmonary uptake, and more rigid pulmonary pressure-volume curve (Fig 1) (p < 0.05 in all). The workers of group B differed (by de6nition) from group A only in terms of rigidity ofthe pulmonary pressure-volume curve and increased 67Ga pulmonary uptake. Eight had a roentgenogram scored at 0/0 or Oil; none of them had bilateral rales. The clinical data in the asbestos workers ofgroups A, B, and C are essentially in keeping with our previous work. Bronchoalveolar Lavage. In Figure 7, we present the results ofanalysis of fibronectin and procollagen in bronchoalveolar lavage fluid. In the controls, the ratio of 6bronectinlalbumin in bronchoalveolar lavage fluid was 33.7j.Lg/mg ± 6.4j.Lg/mg, and the ratio of procollagen 3/albumin was 1.22±0.22 ng/mg. In the asbestos workers without asbestosis and with normal pulmonary pressure-volume curve and normal 67Ga pulmo-
In the workers with asbestosis, the fibronectinl albumin level was at 7l.4j.Lg/mg±32j.Lg/mg and was significantly higher than controls and group A (p<0.05); the procollagen 3/albumin ratio was at 2.0±0.8ng/mg and was comparable to levels in group B (p<0.05 vs controls and group A). Analysis ofthe data excluding smokers did not signi6cantly change the results. DISCUSSION
The present report extends earlier investigations on mechanisms of pulmonary injury associated with the presence of respirable particles in the bronchoalveolar milieu. In the animal model, this study documents that asbestos-associated alveolitis which progresses to peribronchiolar and endobronchiolar 6brosis (the fundamentallesion ofasbestosis) is associated with signi6cant early increase in the level of procollagen 3 in bronchoalveolar lavage fluid and a sustained increase in the 6bronectin level, which do not occur in the non6brosing and regressive latex-associated alveolitis. In asbestos workers with clinical and lavage evidences ofalveolitis, levels of procollagen 3 and 6bronectin are similarly increased in bronchoalveolar lavage fluid, which should improve our understanding of disease activity in asbestos workers. In the sheep tracheal lobe, this study documents that after a single high-dose exPosure to asbestos, there are severe derangements ofthe bronchoalveolar milieu and activation ofa chronic sustained alveolar, peribronchiolar and endobronchiolar inflammatory process which evolves to 6brosis. On bronchoalveolar lavage the increased macrophage and neutrophil cellularity is associated with increased 6bronectin, a glycoprotein CHEST I 89 I 2 I FEBRUARY, 1986
241
produced locally by the macrophage, 5 known to be chemotactic and an attachment factor for fibroblasts and a stimulant for fibroblast replication in association with another macrophage-derived fibroblast growth factor (MDFGF).4-6 In the asbestos-associated alveolitis in sheep, we have reported enhanced production of M D FG p7 and macrophage-derived neutrophil chemotactic factor (MDNCF). J8 Because these three macrophage-derived factors (fibronectin, MDFGF, and MDNCF) are currently thought to have major roles in the pathogenesis of pulmonary fibrosis,4-6 our data from sheep suggest that these factors conmbute to maintain the alveolitis on a long-term basis and lead to its progression to asbestosis. The mechanisms which activate the macrophage to produce these factors are likely related to the prolonged retention of asbestos fibers in the bronchoalveolar milieu, as suggested by our previous study ofalveolar clearance ofchrysotile in the sheep model. 29 In the model, we have shown that one year after cessation of exposure, 11 percent of the fibers shorter than 8,.." 32 percent of the fibers with length in the range of 8,.., to 20,.." and 50 percent of the fibers longer than 20,.., were still recoverable by pulmonary lavage.' The persistence of these fibers in the bronchoalveolar milieu may constitute the determinant factor for maintenance of activated macrophages and alveolitis. In the assessment of fibrogenic activity of alveolitis, the quantitation of type 3 collagen compared to type 1 in lung biopsies and in postmortem samples from patients with fibrosing alveolitis documented that an increased ratio of type 3 to type 1 collagen was associated with an earlier fibrotic process which responded better to therapy,l-3 which would be of interest in the staging of activity of the fibrosing alveolitis; howevex; this method necessitates lung biopsies and thus is of limited usefulness clinically. The recent development by Rohde et alie ofa radioimmunoassay of procollagen 3 N -terminal peptide allowed Low et al8to demonstrate increased levels of procollagen 3 in bronchoalveolar lavage fluid from patients with idiopathic pulmonary fibrosis and in patients with sarcoidosis, with the highest values in those with idiopathic pulmonary fibrosis, suggesting increased secretion of type 3 collagen. In the sheep model, we document a similar increase in procollagen 3 in lavage fluid, which occurs during the 6rst two months after asbestos exposure, at which time early peribronchiolar fibrosis could be observed in pulmonary tissues. The absence ofsimilar changes in the latex-associated alveolitis strengthen the contention of its specificity for a fibrosing process; its increased levels occurring early in the fibrotic process are in keeping with immunohistochemical data. 1-3 Thus, the data ofLow et al8and our data suggest that measurement of levels of procollagen 3 in bronchoalveolar lavage fluid may be a useful method of 242
assessing fibrogenic activity of alveolitis. In the study of asbestos workers, we have documented that in those with asbestosis, the alveolitis is characterized by a Significant increase in total cells, macrophages, and neutrophils in the bronchoalveolar lavage fluid, which confirms the previous report of Bignon et al. 30 Furthermore, in workers with asbestosis, 6bronectin and procollogen 3 in lavage fluid were significantly elevated, which parallels previous results in patients with idiopathic pulmonary fibrosis. 4-6.8 These data from bronchoalveolar lavage are also particularly useful in demonstrating that in the workers who did not meet the usual diagnostic criteria for asbestosis, those with enhanced e7Ga pulmonary uptake and rigid pulmonary pressure-volume curve (group B) had Significantly higher levels of total cells, macrophages, neutrophils, fibronectin, and procollagen 3 in bronchoalveolar lavage fluid than workers with similar asbestos exposures but with a normal 67Ga scan and pulmonary pressure-volume curve. These data from bronchoalveolar lavage suggest that the alveolitis of asbestos workers from group B has all the features to expect its progression to asbestosis, which has been our observation in 12 of the 16 workers ofour original report. 18 In conclusion, this study documents that the measurement of levels of 6bronectin and procollagen 3 in bronchoalveolar lavage fluid assesses 6brogenic activity of alveolitis and can predict its progression in a fibrotic process. The potential use of these markers of fibrogenic activity in asbestos workers is primarily related to early detection of asbestos-induced pulmonary injury in workers with uncertain disease activity. In the reversible interstitial pulmonary disease, these markers of fibrogenic activity should be of interest as indicators to stage the disease and to predict its potential reversibility by therapy. ACKNOWLEDGMENT: M. Geoffroy, S. Gouin, and S. provided assistance. REFERENCES
P~loquin
1 Bateman ED, Turne~Warwick M, Adelmann BC. Immunohistochemical study ofcollagen types in human foetal lung and fibrotic lung disease. Thorax 1981; 36:645-53 2 Bateman ED, Thrne~Warwick M, Haslam PL, Adelmann-Grill BC. Cryptogenic fibrosing alveolitis: prediction of fibrogenic activity from immunohistochemical studies of collagen types in lung biopsy specimens. Thorax 1983; 38:93-101 3 Kirk JME, Heard BE, Kerr I, 1bme~Warwick M, Laurent GJ. Quantitation of types I and III collagen in biopsy lung samples from patients with cryptogenic fibrosing alveolitis. Coli ReI Res 1984; 4:169-82 4 Rennard SI, Crystal RG. Fibronectin in human bronchopulmonary lavage fluid: elevation in patients with interstitial lung disease. J Coo Invest 1981; 69:113-22 5 Rennard SI, Hunningbake G~ Bitterman PB, Crystal RG. Production of fibronectin by the human alveolar macrophage: mechanism for the recruitment of fibroblasts to sites of tissue injury in interstitial lung disease. Proc Natl Acad Sci 1981; 78:7147-51 SAL of Asbestos-expose Human Subjects and Sheep (Begin st aI)
6 Bitterman PB, Rennard SI, Adelberg S, Crystal RG. Role of fibronectin as a growth factor for fibroblasts. J Cell BioI 1983; 97:1925-32 7 Kirk JME, Bateman ED, Haslam PL, Laurent GJ, TurnerWarwick M. Serum type III procollagen peptide concentration in cryptogenic fibrosing alveolitis and its clinical relevance. Thorax 1984; 39:726-32 8 Low RB, Cutroneo KR, Davis GS, Giancola MS. Lavage type III procollagen N-terminal peptides in human pulmonary fibrosis and sarcoidosis. Lab Invest 1983; 48:755-59 9 Begin R, Rola-Pleszczynski M, Sirois ~ Masse S, Nadeau D, Bureau MA. Sequential analysis ofthe bronchoalveolar milieu in conscious sheep. J Appl Physioll981; 50:665-71 10 Begin R, Masse S, Bureau MA. Morphology and function of the airways in early asbestosis ofthe sheep. Am Rev Respir Dis 1982; 126:870-76 11 Begin R, Rola-Pleszczynski M, Masse S, Lemaire I, Sirois ~ Boctor M, et ale Asbestos-induced lung injury in the sheep model: the initial alveolitis. Environ Res 1983; 30:195-210 12 Begin R, Rola-Pleszczynski M, Masse S, Berthiaume Y, Drapeau G. Rapid in vivo assessment of toxicity of respirable particles: a concise report. Res Commun Chem Pathol Pharmacol 1983; 40:267-78 13 Begin R, Rola-Pleszczynski M, Masse S, Berthiaume Y, Drapeau G. Selective exposure and analysis of the sheep tracheal lobe to study rapidly in vivo the toxicity of respirable particles. Environ Res 1985; 36:389-404 14 Itmbrell ~ Gibson JC, Webster I. UICC reference samples of asbestos. Int J Cancer 1978; 3:406-08 15 Begin R, Cantin A, Berthiaume ~ Boileau R, Peloquin S, Masse S. Airway function in lifetime nonsmoking older asbestos workers. Am J Med 1983; 75:631-38 16 Begin R, Boctor M, Bergeron D, Cantin A, Berthiaume Y, Peloquin S, et ale Radiographic assessment ofpleuro-pulmonary disease in asbestos workers: postero-anterior, four view films and computed tomograms of the thorax. Br J Industr Med 1984; 41:373-83 17 Begin R, Cantin A, Berthiaume Y, Boileau R, Bisson G, Lamoureux G, et ale Clinical features to stage the alveolitis in asbestos workers. Am J Industr Med 1985; 8:402-17 18 Begin R, Cantin A, Drapeau G, Lamoureux G, Boctor M, Masse S, et ale Pulmonary uptake of gallium-67 in asbestos exposed
humans and sheep. Am Rev Respir Dis 1983; 127:623-30 19 Bisson G, Drapeau G, Lamoureux G, Cantin A, Rola-Pleszczynski M, Begin R. Computer based quantitative analysis of gallium-67 uptake in normal and diseased lungs. Chest 1983; 84:513-17 20 Bates D~ Macklem PI: Christie ~ The normal lung: physiology and methods ofstudy. In: Respiratory function in disease. Philadelphia: WB Saunders, 1971:11-95; 276-80 21 Begin R, Renzetti AD, Watanabe S, Bigler AB. Flow and age dependence of airway closure and dynamic compliance. J Appl Physiol 1974; 38:199-207 22 Rola-Pleszczynski M, Sirois ~ Begin R. Cellular and humoral components ofbronchoalveolar lavage in the sheep. Lung 1981; 159:91-9 23 Killingsworth LM, Savory J. Manual nephelometric methods for immunochemical determination of immunoglobulins IgG, IgA and IgM in human serum. Clin Chem 1972; 18:335-39 24 Engvall E, Ruoslahti E. Binding of soluble form of fibroblast surface protein, fibronectin, to collagen. Int J Cancer 1977; 20:1-5 25 Rennard SI, Berg R, Martin CR, Foidart JM, Robey PC. Enzyme-linked immunoassay for connective tissue components. Anal Biochem 1980; 104:205-14 26 Rohde H, Vargas L, Eckhart E, Kalbfleish H, Bruguera M, TImpl R. Radioimmunoassay for type III procollagen peptide and its application to human liver disease. Eur J Clin Invest 1979; 9:451-59 27 Lemaire I, Rola-Pleszczynski M, Begin R. Asbestos exposure enhances the release offibroblast growth factor by sheep alveolar macrophages. J Reticuloendothelial Soc 1983; 33:275-85 28 Rola-Pleszczynski M, Gouin S, Begin R. Asbestos-induced lung inBammation: role of local macrophage derived chemotactic factors in accumulation ofneutrophils in the lungs. Inflammation 1984; 8:53-62 29 Sebastien ~ Begin R. Alveolar clearance of chrysotile in the sheep model. lLO 6th Int Pneumoconiosis Conf Monogr 1984; 2:1010-20 30 Bignon J, Atassi K, Jaurand MC, Yamine J, Kaplan H, Geslin ~ et ale Etude cytologique et biochimique du liquide de lavage bronchoalveolaire (LBA) dans la fibrose pulmonaire idiopathique et rasbestose. Rev Franc Mal Respir 1978; 6:353-58
CHEST I 89 121 FEBRUARY. 1988
243
To evaluate the potential interest oflevels ofSbronectin and procollagen 3 in bronchoalveolar lavage 8uid as markers of 6brogenic activity, we characterized the time course of changes in 6bronectin and procollagen 3 levels in the tracheal lobe of sheep exposed to nonSbrogenic and 6brogenic materials. We correlated these observations with those ofbronchoalveolar lavage in long-term asbestos workers in various stages of disease activity. Following studies before exposure, the tracheal lobe of three groups of 24 sheep were exposed once to 100 mI of phosphate-buffered saline solution (PBS~ to 100 mg of latex beads in 100 mI of PBS, or to 100 mg of chrysotile Sbers in 100 mI of PBS. Bronchoalveolar lavages were obtained at 0, 1, 2, 4, 8, and 12 months after exposure, and four or Sve sheep per group were 1dIIed after each lavage for histopathologic analysis. Fibronectin in bronchoaIveolar lavage 8uid increased significantly only in the asbestos-exposed sheep to values two to
three times above controls or latex-exposed sheep and remained elevated during the 12 months ofthe study. Levels of procollagen 3 in bronchoalveolar lavage 8uid were increased signi6cantly only during the first two months following exposure in the asbestos-exposed sheep only. In the asbestos workers without disease, levels of 6bronectin and procoUagen 3 in bronchoalveolar lavage 8uid were comparable to controls, but these levels were signi6cantly elevated in those with asbestos-associated a1veolitis or asbestosis. This study documents that the measurement of levels of 6bronectin and procollagen 3 in bronchoalveolar lavage 8uid assesses 6brogenic activity of alveolitis and should be useful to predict its progression in a 6brotic process. In asbestos workers the potential use of these markers is primarily related to early detection of asbestos-induced pulmonary injury.
Until recently, the assessment of 6brogenic activity of human interstitial pulmonary diseases was limited to the histomorphologic and immunohistochemical analyses of limited samples of pulmonary tissues. l -3 Through bronchoalveolar lavage of diseased areas, it has been documented that human pulmonary diseases with histologic evidence of 6brogenic activity were associated in the bronchoalveolar lavage fluid with increased levels of fibronectin and procollagen 3 peptides, molecules implicated in the biochemical process of pulmonary fibrosis,4-8 and thus of potential interest as markers of fibrogenic activity; however, these observations were limited to the individuals with chronic disease. The time course of these changes in bronchoalveolar lavage fluid have not been studied in spite of increasing interest in the biologic monitoring of humans exposed to environmental materials or therapeutic drugs with known pulmonary toxic effects. In the present study, we characterized the time course of these changes in fibronectin and procollagen 3 levels in bronchoalveolar lavage fluid from the sheep
tracheal lobe exposed to nonfibrogenic and fibrogenic materials. These observations were correlated with those of bronchoalveolar lavage in long-term asbestos workers in various stages of disease activity. The data clearly document that fibronectin and procollagen 3 levels in bronchoalveolar lavage fluid are increased early in alveolitis with fibrogenic activity, but not in those without, which should contribute to further refine our clinical understanding of disease activity in the chronic inflammatory pulmonary disorders.
*From the Unite de Recherche Pulmonaire, Centre Hospitalier Universitaire, Universite de Sherbrooke, Sherbrooke, Quebec, Canada. Supported by I.R.S.S.l: and I.R.D.A. Quebec and M.R.C. Canada. Manuscript received April 10; revision accepted August 12. Reprint requests: Dr: Begin, CHUS, Sherbrooke, Quebec, Canada ]lH 5N4
MATERIALS AND METHODS
Animals Seventy-two male sheep weighing between 25 and 45 kg were used in this study. They were prepared and accustomed to the pulmonary techniques as previously reported. ~13 Experimental Design The flock was divided into three groups of 24 sheep. Following studies before exposure (control studies), the sheep tracheal lobe was exposed once to 100 ml of phosphate-buffered saline solution (PBS) (control group), to 100 mg of0.1.., latex beads (Sigma Chemical Co) in 100 ml of PBS (latex group), or to 100 mg of UICC. Canadian chrysotile asbestos fibers in 100 ml of PBS (asbestos group). These asbestos fibers were relatively uniform and well characterized, 14 with 92 percent less than 0.25.., in diameter and 20.., in length. Exposure ofthe tracheal lobe was carried out via bronchoscopic catheterization of the tracheal bronchus and slow infusion ofthe suspension into the lobe. The animals were studied prior to exposure and at 1, 2, 4, 8, and 12 months after exposure by bronchoalveolar lavage and by CHEST I 89 121 FEBRUARY. 1986
237
histopathologic methods (four or five sheep per group killed each time).
Subsets of Asbestos Workers On the basis ofcriteria reported in detail,15-19 the asbestos workers were divided into three categories ofdisease: (1) group A consisted of seven workers without evidence of asbestos-induced pulmonary damage; they did not have the diagnostic criteria for asbestosis, and their 87Ga pulmonary uptake and pulmonary pressure-volume curve were within normallimits;m.21 (2) group B was eight asbestos workers with evidence of asbestos-induced alveolitis documented at biopsy in three of three15-19 in whom the biopsy was obtained; they did not have the diagnostic criteria for asbestosis but had enhanced 87Ga pulmonary uptake and rigid pulmonary pressure-volume curve; and (3) group C was 19 asbestos workers who met the diagnostic criteria for asbestosis. The clinical data and bronchoalveolar lavage cellularity of these patients are summarized in Figure 1.
Human Controls Twenty-one manual workers without exposure to asbestos were tested within the period. They were matched for age, height, sex, smoking, and work habits to the asbestos workers; seven were current or ex-smokers, and 14 were lifetime nonsmokers. The mean age ofcontrol subjects was 57 ± 5 years (range, 50 to 70 years). None of them was exposed to environmental dust at risk of pneumoconiosis. In this control population, total cells in bronchoalveolar lavage fluid averaged 171 ± 24 X 1()3/ml, with macrophages at 159 ± 22 x 1()3/ml, and neutrophils at 0.86 ± 0.30 x 1()3/ml.
Asbestos Workers
Bronchoalveolar Lavage and Fluid Analysis
The 34 asbestos workers were between the ages of42 and 70 years (mean, 58±4 years). All had been exposed to Canadian chrysotile asbestos only in the mines and mills of the eastern townships of Quebec for an average of33 ± 5 years (range, 24 to 42 years). Twentyeight were nonsmokers for more than two years, and the six others were either current or recent ex-smokers. Many of these patients were included in our studies ofaiJway function in asbestos workers, 15 or computerized tomographic scanning study16 and were among 226 asbestos workers in a study of clinical features of the associated alveolitis. 17 In this present study, we specifically report on fibronectin and procollagen 3 in bronchoalveolar lavage fluid from these 34 workers as we relate these findings to serial measurements of these markers in the sheep model of asbestosis.
Most ofthe techniques in the procedures and analyses ofbronchoalveolar lavage have been previously described.8-lJ.18 The effiuent from lavage was passed through four layers ofcheesecloth to remove mucus, and the cells were pelletized by centrifugation. Cells were counted in a hemocytometer, and cell viability was determined by the trypan blue exclusion technique. Cytocentrifuge smears served to identify the cellular populations recovered with the WrightGiemsa and naphthyl acetate esterase stains. 12 In the supernatant, albumin and fibronectin were measured by the immunochemical methods of Killingsworth and Savory,13 using a laser nephelometer instrument (Behring LN modular system). For sheep albumin, specific antiserum raised in rabbit were obtained commercially (Cappell Lab. Inc.). Sheep fibronectin from bronchoalveolar lavage 8uid was purified by affinity column followed by chromatography and antisheep fibronectin antibodies prepared in rabbits as described.... 25 Pr0c01lagen 3 from bronchoalveolar lavage fluid was measured as type 3 procollagen N-terminaI peptides by radioimmunoassay, as reported by Low et al, 8 based on the method originally
Clinical Evaluation All patients underwent a medical history, complete physical examination, standard high-kilovoltage posteroanterior, lateral, and oblique chest films, detailed tests of pulmonary function, and 87Ga lung scan as previously reported. 15-19
.ms
CLINICAL DATA ON ASBESTOS G.-87
LUNG P-V CURVE
~
I
I
til
2.'
JI
I' "
JI
I
, •• c
t •
t •
..
..
..
•• IC
••• C
I A' C
lAIC
LUNG LAVAGE CELLtuRlTY
TOTAL CELL 1 M 10
i ·
~
~
••• C
c:1IH10
40
DIFFUSION CAPACITY
J
J
J I.'
'.1
••
VITAL CAPACITY
TOTAL LWIC CAPACITY
RALEB 1.1
II
I •••
t.
t ••
~
LlMG a.TAICE
I.'
12.
PARENCHYItAL IPACITY
-6
TOTAL CELL 480
1
M
10
-3
/...
MACROPHAGE
"ACROPHAGE
NEUTROPHIL
S
1 M 10
S
100
480
-3
/...
11
NEUTROPHIL 5.0
1
30
350
7S
3.0
20
240
50
240
3.0
10
120
120
1.5
". I C
I. I C
10
-3
/ ...
4.5
12
lAIC
M
• AI C
lAIC
FIGURE 1. Clinical, functional, radiographic and bronchoaIveolar lavage cellularity data of our patients. In group N (normal controls), there were 21 manual workers without asbestos exposure. Group ~ was seven asbestos workers without any evidence of asbestos-induced pulmonary damage; group B was eIght workers with increased mCa pulmonary uptake and rigid pulmonary pressure-volume (P-V) curve without criteria for established asbestosis; group C was 19 asbestos workers who met diagnostic criteria for asbestosis. 15-18 mGa pulmonary uptake, parenchymal opacities, and rales were scored as previously reported.15-18 Data on pulmonary function are reported as percent predicted of Bates et aI. 1O 238
SAL of Asbestos-expose Human Subjects and Sheep (Begin et aI)
gen 3 did not change. In the latex-exposed sheep, we have previously reported a clear increase in total cells from lavage at month 1 to eight times the control value, lJ,13 and at month 12, this had returned to the level of saline-exposed sheep. From that point and after; there was no significant difference in levels of macrophages, neutrophils, fibronectin, and procollagen 3 between latex-exposed and saline-exposed sheep. In the asbestos-exposed sheep, total cells were clearly increased through the experiment; and one year after exposure, total cells in bronchoalveolar lavage fluid remained at 150 percent of controls, macrophages at 150 percent of controls, and neutrophils at 285 percent of controls. Fibronectin in bronchoalveolar lavage fluid was significantly elevated only in the asbestos-exposed sheep, and the levels increased throughout the experiment. Procollagen 3 in lavage fluid increased significantly early and returned to control levels at month 4 after exposure (Fig 2). Albumin was increased significantly only after month 4, and it paralleled the second increase in fibronectin (data not shown). Pulmonary Histopathology. In the saline-exposed
described by Rohde et al. 1S All results were expressed per milliliter ofbronchoalveolar lavage fluid. All values ofhumoral components of lavage fluid were also analyzed in terms of the ratio to the albumin content of bronchoalveolar lavage fluid supernatant. Histopathology
At months 1, 2, 4, 8, and 12 ofthe study, four sheep in each group were killed and the lungs removed from the thoracic cavity. The tracheal lobe was identified, and nine samples of the lobe were obtained each time for microscopic examination. The pulmonary samples were processed as routinely done for human pulmonary tissue. Statistical Analysis
All results are expressed as mean ± SEe The data were tested by Students t-test for differences between groups. A value of p < 0.05 was considered significant in this study. RESULTS
Sheep Bronchoalveolar Lavage. The results of analysis of bronchoalveolar lavage fluid are presented in Figure 2. In the saline-exposed sheep, total cells from lavage averaged 250,OOO/ml and did not vary significantly in the course of the experiment. Similarly, the levels of macrophages, neutrophils, fibronectin, and procolla-4
TOTM. CELL, INIO •••
-t
MACROPHAGE,
120
to
-4
INIO •••
-t
-4
NEUTR.-etIL. INIO •••
-t
U
.-tr
U
*
30 1
15
.5
12
12
12 ....TH
....TH
"ONTH ~
TOT-. CELL, iNiO
40
75
30
50
20
25
10
12
12
-I
120
PROCOLLAC£N I I I.
12
....TH
tIONTH
....TH F I.cJNECT IN. ua•• '
NEUTROPHIL. • 0# TOT-. CELL
MCROPHACE • • 0# TOT-. CELL 100
na •• 1
-t
1.0
*'tt
to
LO
••0
2.0
30
....TH
FIGURE 2. Analyses ofbronchoalveolar lavage fluid in sheep model. Sheep were exposed once at one to seven days after control lavage (month 0) either to 100 ml ofPBS (solid squares), to 100 mg ofO.1lA-latex beads in 100 ml of PBS (open squares), or to 100 mg of U. I. C. C. chrysotile in 100 ml of PBS (triangles). Star indicates p
CHEST I 89 I 2 I FEBRUARY, 1986
239
FIGURE 3. Latex-exposed tracheal lobe at month 4, showing early pneumonia-like reaction with preservation of airway structures (hematoxylin-eosin. original magnification x 160).
sheep, the histopathology of the tracheal lobe remained nonnal during the experiment. In the latexexposed sheep at month 1, there was a pneumonia-like infiltrate of the tracheal lobe;li the infiltrate was predominantly alveolar and was largely composed of macrophages without distortion of the pulmonary architecture. At month 4, the infiltrate had regressed, and areas of residual cell accumulation constituted less than 5 percent of the total lobe (Fig 3). After month 4, the histopathology of the latex-exposed tracheal lobes had returned to nonnal (Fig 4). In the asbestos-exposed sheep at month 1, there was also a pneumonialike infiltrate composed of macrophages and neutrophils, with considerable distortion of the architecture of the lung and airway;li at month 4, the infiltrate had considerably regressed to some 20 percent ofthe initial extent; it was located predominantly in the peribronchiolar and endobronchiolar areas, with persistent severe distortion of the small airways (Fig 5). At month 8, the infiltrate continued to regress but peribronchiolar and endobronchiolar fibrosis could be detected; at month 12, the infiltrate was at 5 to 10 percent of the initial extent, but peribronchiolar and endobronchiolar fibrosis were severely distorting the pulmonary architecture (Fig 6). Human Subjects
Clinical Data. On the basis of clinical, radiologic, and functional data, 19 of the 34 subjects were recog240
FIGURE 4. Latex-exposed tracheal lobe at month 12. showing nearly complete resolution of process seen in Figure 3 (hematoxylin-eosin, original magnification x 63).
FIGURE 5. Representative histopathologic findings in asbestosexposed tracheal lobes at month 4. Early pneumonia-like reaction is associated with severe structural distortion of most small airways (hematoxylin-eosin, original magnification x 160). SAl 01 Asbesto8-expoee Human Subjects end ShMp (BegIn et eI)
PROCOLLAGEN/A
FIBRONECTIN/A ugh,g 100
3.2
ng/lltg
1
2.4
75
....
....
0.8
~
!
....
~ ~
''"" ........ NAD C
NAB C
FIGURE 7. Levels of fibronectin and procollagen 3 in human bronchoalveolar lavage Huid reported as ratios to albumin (A). Groups N, A, B, and C are as in Figure 1.
nary uptake (group A), the 6bronectinlalbumin ratio
was 33.1j.Lg/mg±6.0j.Lg/mg, and the procollagen 31 albumin ratio was 1.03±0.24 ng/mg. In group B, the 6bronectin albumin ratio was 45.7j.Lg/mg ± 34j.Lg/mg
(not signi6cant), and the procollagen 3/albumin ratio
was 2.3 ± 0.8 ng/mg (p<0.05 vs controls and group A).
FIGURE 6. Histopathologic findings in asbestos-exposed tracheal lobe at month 12, demonstrating less intense in8ammatory process than at month 4 (Fig 5), but lesions of small airways did not regress and became fibrotic (hematoxylin-eosin, original magnification x
63).
nized as having asbestosis. In comparison with the other 15 workers and the controls, this group had identical age and smoking habits. They had a signi6cantly higher rale score, higher score of radiographic pulmonary opacities, lower vital capacity, lower total lung capacity (1'1£), lower diffusing capacity for carbon monoxide, higher 67Ga pulmonary uptake, and more rigid pulmonary pressure-volume curve (Fig 1) (p < 0.05 in all). The workers of group B differed (by de6nition) from group A only in terms of rigidity ofthe pulmonary pressure-volume curve and increased 67Ga pulmonary uptake. Eight had a roentgenogram scored at 0/0 or Oil; none of them had bilateral rales. The clinical data in the asbestos workers ofgroups A, B, and C are essentially in keeping with our previous work. Bronchoalveolar Lavage. In Figure 7, we present the results ofanalysis of fibronectin and procollagen in bronchoalveolar lavage fluid. In the controls, the ratio of 6bronectinlalbumin in bronchoalveolar lavage fluid was 33.7j.Lg/mg ± 6.4j.Lg/mg, and the ratio of procollagen 3/albumin was 1.22±0.22 ng/mg. In the asbestos workers without asbestosis and with normal pulmonary pressure-volume curve and normal 67Ga pulmo-
In the workers with asbestosis, the fibronectinl albumin level was at 7l.4j.Lg/mg±32j.Lg/mg and was significantly higher than controls and group A (p<0.05); the procollagen 3/albumin ratio was at 2.0±0.8ng/mg and was comparable to levels in group B (p<0.05 vs controls and group A). Analysis ofthe data excluding smokers did not signi6cantly change the results. DISCUSSION
The present report extends earlier investigations on mechanisms of pulmonary injury associated with the presence of respirable particles in the bronchoalveolar milieu. In the animal model, this study documents that asbestos-associated alveolitis which progresses to peribronchiolar and endobronchiolar 6brosis (the fundamentallesion ofasbestosis) is associated with signi6cant early increase in the level of procollagen 3 in bronchoalveolar lavage fluid and a sustained increase in the 6bronectin level, which do not occur in the non6brosing and regressive latex-associated alveolitis. In asbestos workers with clinical and lavage evidences ofalveolitis, levels of procollagen 3 and 6bronectin are similarly increased in bronchoalveolar lavage fluid, which should improve our understanding of disease activity in asbestos workers. In the sheep tracheal lobe, this study documents that after a single high-dose exPosure to asbestos, there are severe derangements ofthe bronchoalveolar milieu and activation ofa chronic sustained alveolar, peribronchiolar and endobronchiolar inflammatory process which evolves to 6brosis. On bronchoalveolar lavage the increased macrophage and neutrophil cellularity is associated with increased 6bronectin, a glycoprotein CHEST I 89 I 2 I FEBRUARY, 1986
241
produced locally by the macrophage, 5 known to be chemotactic and an attachment factor for fibroblasts and a stimulant for fibroblast replication in association with another macrophage-derived fibroblast growth factor (MDFGF).4-6 In the asbestos-associated alveolitis in sheep, we have reported enhanced production of M D FG p7 and macrophage-derived neutrophil chemotactic factor (MDNCF). J8 Because these three macrophage-derived factors (fibronectin, MDFGF, and MDNCF) are currently thought to have major roles in the pathogenesis of pulmonary fibrosis,4-6 our data from sheep suggest that these factors conmbute to maintain the alveolitis on a long-term basis and lead to its progression to asbestosis. The mechanisms which activate the macrophage to produce these factors are likely related to the prolonged retention of asbestos fibers in the bronchoalveolar milieu, as suggested by our previous study ofalveolar clearance ofchrysotile in the sheep model. 29 In the model, we have shown that one year after cessation of exposure, 11 percent of the fibers shorter than 8,.." 32 percent of the fibers with length in the range of 8,.., to 20,.." and 50 percent of the fibers longer than 20,.., were still recoverable by pulmonary lavage.' The persistence of these fibers in the bronchoalveolar milieu may constitute the determinant factor for maintenance of activated macrophages and alveolitis. In the assessment of fibrogenic activity of alveolitis, the quantitation of type 3 collagen compared to type 1 in lung biopsies and in postmortem samples from patients with fibrosing alveolitis documented that an increased ratio of type 3 to type 1 collagen was associated with an earlier fibrotic process which responded better to therapy,l-3 which would be of interest in the staging of activity of the fibrosing alveolitis; howevex; this method necessitates lung biopsies and thus is of limited usefulness clinically. The recent development by Rohde et alie ofa radioimmunoassay of procollagen 3 N -terminal peptide allowed Low et al8to demonstrate increased levels of procollagen 3 in bronchoalveolar lavage fluid from patients with idiopathic pulmonary fibrosis and in patients with sarcoidosis, with the highest values in those with idiopathic pulmonary fibrosis, suggesting increased secretion of type 3 collagen. In the sheep model, we document a similar increase in procollagen 3 in lavage fluid, which occurs during the 6rst two months after asbestos exposure, at which time early peribronchiolar fibrosis could be observed in pulmonary tissues. The absence ofsimilar changes in the latex-associated alveolitis strengthen the contention of its specificity for a fibrosing process; its increased levels occurring early in the fibrotic process are in keeping with immunohistochemical data. 1-3 Thus, the data ofLow et al8and our data suggest that measurement of levels of procollagen 3 in bronchoalveolar lavage fluid may be a useful method of 242
assessing fibrogenic activity of alveolitis. In the study of asbestos workers, we have documented that in those with asbestosis, the alveolitis is characterized by a Significant increase in total cells, macrophages, and neutrophils in the bronchoalveolar lavage fluid, which confirms the previous report of Bignon et al. 30 Furthermore, in workers with asbestosis, 6bronectin and procollogen 3 in lavage fluid were significantly elevated, which parallels previous results in patients with idiopathic pulmonary fibrosis. 4-6.8 These data from bronchoalveolar lavage are also particularly useful in demonstrating that in the workers who did not meet the usual diagnostic criteria for asbestosis, those with enhanced e7Ga pulmonary uptake and rigid pulmonary pressure-volume curve (group B) had Significantly higher levels of total cells, macrophages, neutrophils, fibronectin, and procollagen 3 in bronchoalveolar lavage fluid than workers with similar asbestos exposures but with a normal 67Ga scan and pulmonary pressure-volume curve. These data from bronchoalveolar lavage suggest that the alveolitis of asbestos workers from group B has all the features to expect its progression to asbestosis, which has been our observation in 12 of the 16 workers ofour original report. 18 In conclusion, this study documents that the measurement of levels of 6bronectin and procollagen 3 in bronchoalveolar lavage fluid assesses 6brogenic activity of alveolitis and can predict its progression in a fibrotic process. The potential use of these markers of fibrogenic activity in asbestos workers is primarily related to early detection of asbestos-induced pulmonary injury in workers with uncertain disease activity. In the reversible interstitial pulmonary disease, these markers of fibrogenic activity should be of interest as indicators to stage the disease and to predict its potential reversibility by therapy. ACKNOWLEDGMENT: M. Geoffroy, S. Gouin, and S. provided assistance. REFERENCES
P~loquin
1 Bateman ED, Turne~Warwick M, Adelmann BC. Immunohistochemical study ofcollagen types in human foetal lung and fibrotic lung disease. Thorax 1981; 36:645-53 2 Bateman ED, Thrne~Warwick M, Haslam PL, Adelmann-Grill BC. Cryptogenic fibrosing alveolitis: prediction of fibrogenic activity from immunohistochemical studies of collagen types in lung biopsy specimens. Thorax 1983; 38:93-101 3 Kirk JME, Heard BE, Kerr I, 1bme~Warwick M, Laurent GJ. Quantitation of types I and III collagen in biopsy lung samples from patients with cryptogenic fibrosing alveolitis. Coli ReI Res 1984; 4:169-82 4 Rennard SI, Crystal RG. Fibronectin in human bronchopulmonary lavage fluid: elevation in patients with interstitial lung disease. J Coo Invest 1981; 69:113-22 5 Rennard SI, Hunningbake G~ Bitterman PB, Crystal RG. Production of fibronectin by the human alveolar macrophage: mechanism for the recruitment of fibroblasts to sites of tissue injury in interstitial lung disease. Proc Natl Acad Sci 1981; 78:7147-51 SAL of Asbestos-expose Human Subjects and Sheep (Begin st aI)
6 Bitterman PB, Rennard SI, Adelberg S, Crystal RG. Role of fibronectin as a growth factor for fibroblasts. J Cell BioI 1983; 97:1925-32 7 Kirk JME, Bateman ED, Haslam PL, Laurent GJ, TurnerWarwick M. Serum type III procollagen peptide concentration in cryptogenic fibrosing alveolitis and its clinical relevance. Thorax 1984; 39:726-32 8 Low RB, Cutroneo KR, Davis GS, Giancola MS. Lavage type III procollagen N-terminal peptides in human pulmonary fibrosis and sarcoidosis. Lab Invest 1983; 48:755-59 9 Begin R, Rola-Pleszczynski M, Sirois ~ Masse S, Nadeau D, Bureau MA. Sequential analysis ofthe bronchoalveolar milieu in conscious sheep. J Appl Physioll981; 50:665-71 10 Begin R, Masse S, Bureau MA. Morphology and function of the airways in early asbestosis ofthe sheep. Am Rev Respir Dis 1982; 126:870-76 11 Begin R, Rola-Pleszczynski M, Masse S, Lemaire I, Sirois ~ Boctor M, et ale Asbestos-induced lung injury in the sheep model: the initial alveolitis. Environ Res 1983; 30:195-210 12 Begin R, Rola-Pleszczynski M, Masse S, Berthiaume Y, Drapeau G. Rapid in vivo assessment of toxicity of respirable particles: a concise report. Res Commun Chem Pathol Pharmacol 1983; 40:267-78 13 Begin R, Rola-Pleszczynski M, Masse S, Berthiaume Y, Drapeau G. Selective exposure and analysis of the sheep tracheal lobe to study rapidly in vivo the toxicity of respirable particles. Environ Res 1985; 36:389-404 14 Itmbrell ~ Gibson JC, Webster I. UICC reference samples of asbestos. Int J Cancer 1978; 3:406-08 15 Begin R, Cantin A, Berthiaume ~ Boileau R, Peloquin S, Masse S. Airway function in lifetime nonsmoking older asbestos workers. Am J Med 1983; 75:631-38 16 Begin R, Boctor M, Bergeron D, Cantin A, Berthiaume Y, Peloquin S, et ale Radiographic assessment ofpleuro-pulmonary disease in asbestos workers: postero-anterior, four view films and computed tomograms of the thorax. Br J Industr Med 1984; 41:373-83 17 Begin R, Cantin A, Berthiaume Y, Boileau R, Bisson G, Lamoureux G, et ale Clinical features to stage the alveolitis in asbestos workers. Am J Industr Med 1985; 8:402-17 18 Begin R, Cantin A, Drapeau G, Lamoureux G, Boctor M, Masse S, et ale Pulmonary uptake of gallium-67 in asbestos exposed
humans and sheep. Am Rev Respir Dis 1983; 127:623-30 19 Bisson G, Drapeau G, Lamoureux G, Cantin A, Rola-Pleszczynski M, Begin R. Computer based quantitative analysis of gallium-67 uptake in normal and diseased lungs. Chest 1983; 84:513-17 20 Bates D~ Macklem PI: Christie ~ The normal lung: physiology and methods ofstudy. In: Respiratory function in disease. Philadelphia: WB Saunders, 1971:11-95; 276-80 21 Begin R, Renzetti AD, Watanabe S, Bigler AB. Flow and age dependence of airway closure and dynamic compliance. J Appl Physiol 1974; 38:199-207 22 Rola-Pleszczynski M, Sirois ~ Begin R. Cellular and humoral components ofbronchoalveolar lavage in the sheep. Lung 1981; 159:91-9 23 Killingsworth LM, Savory J. Manual nephelometric methods for immunochemical determination of immunoglobulins IgG, IgA and IgM in human serum. Clin Chem 1972; 18:335-39 24 Engvall E, Ruoslahti E. Binding of soluble form of fibroblast surface protein, fibronectin, to collagen. Int J Cancer 1977; 20:1-5 25 Rennard SI, Berg R, Martin CR, Foidart JM, Robey PC. Enzyme-linked immunoassay for connective tissue components. Anal Biochem 1980; 104:205-14 26 Rohde H, Vargas L, Eckhart E, Kalbfleish H, Bruguera M, TImpl R. Radioimmunoassay for type III procollagen peptide and its application to human liver disease. Eur J Clin Invest 1979; 9:451-59 27 Lemaire I, Rola-Pleszczynski M, Begin R. Asbestos exposure enhances the release offibroblast growth factor by sheep alveolar macrophages. J Reticuloendothelial Soc 1983; 33:275-85 28 Rola-Pleszczynski M, Gouin S, Begin R. Asbestos-induced lung inBammation: role of local macrophage derived chemotactic factors in accumulation ofneutrophils in the lungs. Inflammation 1984; 8:53-62 29 Sebastien ~ Begin R. Alveolar clearance of chrysotile in the sheep model. lLO 6th Int Pneumoconiosis Conf Monogr 1984; 2:1010-20 30 Bignon J, Atassi K, Jaurand MC, Yamine J, Kaplan H, Geslin ~ et ale Etude cytologique et biochimique du liquide de lavage bronchoalveolaire (LBA) dans la fibrose pulmonaire idiopathique et rasbestose. Rev Franc Mal Respir 1978; 6:353-58
CHEST I 89 121 FEBRUARY. 1988
243