Toxicology, 60 (1990) 53--61 Elsevier Scientific Publishers Ireland Ltd.
In vitro assays to predict the pathogenicity of mineral fibers B r o o k e T. M o s s m a n a n d A n n M. Sesko Department of Pathology, University of Vermont College of Medicine, Burlington, VT, 05405 (U.S.A.) (Received September 7th, 1989; accepted September 12th, 1989)
Summary A number of mineral dusts are associated with the development of inflammation in lung and pulmonary interstitial fibrosis. In an effort to find alternative approaches to animal testing, cells (rat alveolar macrophages, hamster tracheal epithelial ceils, rat lung fibroblasts) of the respiratory tract have been evaluated for cytotoxic and metabolic changes after exposure to fibers (defined as a greater than 3 : I length to diameter ratio) and particles. In all bioassays, fibrous materials provoked greater biological responses in cells in comparison to non-fibrous, chemically similar particulates at identical concentrations. For example, release o f superoxide (O:~) from alveolar macrophages (AMs) was increased (in comparison to that observed in untreated cells) after exposure to the fibers, crocidolite asbestos, erionite, Code 100 fiberglass and sepiolite. Riebeckite, mordenite and glass particles elicited minimal generation of (~2 at similar concentrations of dusts. In hamster tracheal epithelial (HTE) cells, fibers such as chrysotile asbestos, crocidolite asbestos, and Code 100 fiberglass caused increased release of 5~Chromium in comparison to the particles antigorite, riebeckite and glass. Another area of exploration is the measurement of collagen and non-collagen protein in a cell line (RL-82) of rat lung fibroblasts as an indication of the fibrogenic potential of minerals. Crocidolite asbestos caused an increase in the ratio of cell-associated collagen to non-collagen protein in these cell types whereas glass beads did not affect biosynthesis o f collagen. Results suggest that a battery of in vitro assays can be used to screen the capability of minerals to elicit cell damage and inflammatory changes in the respiratory tract.
Key words." Asbestos; Alveolar macrophages; Lung fibroblasts; Tracheobronchial epithelium; Superoxide; Active oxygen species
Introduction Occupational exposure to certain mineral dusts causes chronic bronchitis and pulmonary interstitial fibrosis in man including kaolinosis, silicosis, coal workers pneumoconiosis and asbestosis [1]. Mineral fibers such as asbestos and erionite also increase the risk of development of malignant mesothelioma and bronchogenic carcinoma, especially in smokers [2]. With the increasing commercial needs for substitutes for asbestos fibers, a number of man-made and other naturally occurring mineral fibers currently are being assessed for their pathogenic 0300-483X/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
53
potential in laboratory animals. However, these long-term studies can be timeconsuming and expensive. Moreover, the sophisticated technology required for inhalation experiments and difficulties involved in generation of respirable fibers often preclude dose-response studies and definition of " s a f e " concentrations of materials. In an effort to predict the pathogenic potential of mineral dusts in man, several in vitro assays have been developed over the past 2 decades. Initially, suspensions of red blood cells (RBCs) from several species were evaluated for hemolysis after addition of minerals [3]. The preliminary results of these experiments were encouraging in that fibrogenic minerals such as chrysotile asbestos and silica were more hemolytic than a number of 'nuisance' or inert dusts including titanium dioxide, ferric oxide and glass [4]. However, other minerals, namely crocidolite and other amphibole types of asbestos, were only weakly hemolytic but documented fibrogenic and carcinogenic agents in man. False 'positives' also were encountered in these studies. For example, montmorillonite (bentonite), a positively charged particle, was more hemolytic than chrysotile asbestos at identical concentrations [5]. Further work showed that surface charge (i.e. zeta potential) of particulates was a predominate factor in the hemolysis bioassay - - minerals (i.e. chrysotile asbestos) possessing a strongly positive surface charge were more hemolytic than others such as crocidolite asbestos which possesses a neutral or slightly negative charge [6]. The development of tissue culture techniques and cell lines in the 1970s allowed evaluation of the cytotoxic potential of mineral dusts in a variety of cell types [7]. Several important observations were made in these studies. First, although certain cell lines were more sensitive to fibers and particles in comparison to cells of other origins, chrysotile asbestos again was more toxic in most bioassays when compared to crocidolite and other amphiboles at identical concentrations. Thus, surface charge seemed to be an important determinant of membrane damage in both RBCs and mammalian ceils in general. Second, when fibers of various dimension were assessed comparatively, longer, thinner fibers proved more cytotoxic than shorter fibers, a phenomenon which correlated with their carcinogenic potential after intrapleural or intraperitoneal injection into rodents [8--10]. Lastly, fibers at non-lethal concentrations elicited proliferative and metabolic changes in some cell types in vitro [7]. During the past several years, this laboratory has developed techniques useful in understanding how fibers interact with cells of the respiratory tract. Such mechanisms of interaction may be important in the causation of disease [reviewed in 11] To determine whether fibrous geometry influences biological responses in vitro, we have examined fibers and their non-fibrous, chemically similar analogs comparatively in cells of the respiratory tract. We focus here on O F release from AMs in response to particulates, a phenomenon predictive of the inflammatory capability of fibers and release of 5~Chromium from HTE cells as an indication of the membranolytic potential of fibers. Increased biosynthesis of collagen after exposure of rat lung fibroblasts to asbestos also is presented as a tool to predict the fibrogenic potential of minerals.
54
TABLE I FIBERS A N D PARTICLES USED IN IN VITRO ASSAYS Fibers
Particles
Crocidolite [Na20 • Fe203. FeO • 8 SiO 2" H~O] a Chrysotile [3MgO. 2SiO 2- 2H20] a Erionite [Na 2" K s" C a ' Mg)4.~ (AI9Si~vOTt)'27H20] Code 100 fiberglass [SiO2] a Sepiolite [Mg, (Si205)J' OH- 6H:O]
Riebeckite [like crocidolite] Antigorite [like chrysotile] Mordenite [Ca Na 2" K2)4" AlsSi40 09( 28H20 ] Glass beads [SiO2]
aMain elemental composition. Other trace elements may be present.
Materials and methods
Minerals Fibers and respective particles that were assessed in Of studies in AMs and 51Chromium release assays in HTE ceils are listed in Table I. Both crocidolite and chrysotile asbestos were International Union Against Cancer (U.I.C.C.) reference samples. Code 100 fibers were obtained from Manville Corp., Denver, CO. Erionite (Rome, Oregon) was a gift from the MRC Pneumoconiosis Unit, Penarth, Wales. Sepiolite and mordenite were purchased from Minerals Research, Clarkstown, NY. Riebeckite, antigorite (both from Wards Scientific Establishment, Rochester, NY) and glass beads (Particle Information Service, Inc., Kingston, WA) also were employed in these studies. All preparations of dusts were characterized for purity and size distributions as reported previously [12,13].
Isolation o f AMs Two hundred 250-g Fischer 344 male rats were obtained from N.I.H. and anesthetized with phenobarbital before lavage as described in a prior publication [14]. Ceils from lavage were centrifuged at 900 x g, and differential cell counts were performed on the cell pellet to ascertain the purity of the preparation. AMs (> 98o70 of cell yield from BAL) were plated into 12-well dishes at 5 x 105 cells/ well in 1 ml of Medium 199 containing 10°70 heat-inactivated fetal bovine serum (FBS) (GIBCO, Grand Island, NY). After 2 h at 37°C, the medium was decanted, and adherent cells exposed to particulate before detection of O~.
Assay for generation of 0 5 Fibers and particles previously resuspended in complete Medium 199 were added to cells at concentrations ranging from 2.5 to 25 /ag/cm 2 surface area dish in a reaction mixture containing 1 ml of Earle's Balanced Salt Solution (GIBCO, Grand Island, NY) and 40/am cytochrome c (Type VI, horse heart; Sigma Chemical Co., St. Louis, MO). After 1 h at 37°C, the supernatant was removed from
55
control and dust-exposed cells, centrifuged at 8000 g to remove debris, and superoxide dismutase (SOD)-inhibitable reduction of cytochrome c measured spectrophotometrically at 550 nm. Cell-free reaction mixtures also were incubated with and without asbestos. Results, calculated as/amol reduced cytochrome c/5 x 105 cells/h using a 0.029 mM extinction coefficient for cytochrome c, were analyzed by the Dunnett's procedure [15].
Cultures of H T E cells The H T E diploid cell line, derived from the tracheal epithelium of a female golden Syrian hamster [16], was used in 5tChromium release assays. Cells were passaged routinely in H a m ' s F12 medium containing 10O7o FBS (GIBCO, Grand Island, NY).
Assay for release o f 51Chromium At confluency, cells were removed by trysinization and plated at 5 x 10 4 cells/well in 12-well dishes containing 1 ml H a m ' s F12 medium with 2°/0 FBS. At confluency, cells were labeled with 5tChromium (New England Nuclear, Wilmington, DE); 5/aCi/ml medium for 1 h, and then chased with unlabeled medium for an additional hour. Dusts then were added at concentrations ranging from 0.01 to 20 g g / c m 2 surface area dish, and cultures incubated at 3 7 o c for 3 h. Controls for each group consisted of untreated cells. At the end of the incubation period, both medium and solubilized cells, rinsed previously with buffered saline, were assayed for release of 5~Chromium. The release index then was calculated as (cpm medium)/[total counts (cpm medium + cpm cells)]. Regression of 5~Chromium released upon concentration of particulate applied was tested for statistical significance by A N O V A methodology [17].
Maintenance o f rat lung fibroblasts (RL-82 line) A diploid fibroblast line from the lung homogenate of a Fischer 344 rat was obtained from Dr. Marlene Absher, Department of Medicine, University of Vermont. Cells were maintained to passage 20 in Minimal Essential Medium (MEM, GIBCO, Grand Island, NY) containing 10°70 FBS. When cells approached confluency on plates, crocidolite asbestos at 1.0, 2.5 and 5.0 /ag/cm 2 dish or glass beads at 5.0/ag/cm 2 dish were added for 24-, 48- and 72-h intervals.
Assays to determine collagen and non-collagen protein synthesis To determine whether asbestos elicited increases in cell layer-associated collagen in RL-82 cells, 10-4 M ascorbate was added to all plates at 24-h intervals after exposure to dusts. [3H]proline (New England Nuclear, Wilmington, DE; 10 ~Ci/ml medium) then was added for 2 h. Cells were scraped from dishes using a rubber policeman, homogenized, and boiled for 10 min to inactivate proteases. Samples then were assayed in the presence and absence of purified bacterial collagenase (Advance Biofactures Corp; ~ 40 U / a s s a y tube) for determination of the ratio of collagen to non-collagen protein [18]. Data were analyzed by the Student's t-test adjusting for multiple comparisons between groups.
56
t500~
A.
~C.
1000 "6 o
"5 o~ 500
100
Croc. Croc. Rieb. 2.5 5.0 25
Code Sap. 100 Beads 5.0 25 25
Erion. ~rion. Mord. Mord 5.0 12.5 25 2.5
Fig. 1. Generation of superoxide (O:2) from rat AMs exposed to selected fibers and non-fibrous dusts for 1 h. Shaded and unshaded bars represent separate experiments. *Increased (P < 0.01) in comparison to untreated controls. + = increased (P < 0.05) in comparison to untreated controls. Modified with permission from Hansen and Mossman [13].
Results
Release of 05 from AMs exposed to fibrous and non-fibrous dusts Untreated plates of rat AMs generated approximately 0.172 __. 0.071 /amol reduced cytochrome c/h~5 x 105 cells (Mean of 3 experiments, N = 3 wells/ group/experiment). In comparison to untreated cultures, the fibers, crocidolite asbestos, Code 100 fiberglass and erionite caused significant dose-related increases in production of Of (Fig. 1). Five- to ten-fold greater concentrations of the particles riebeckite, glass beads, and mordenite, were required to elicit comparable responses. Sepiolite, a mixture of short (~< 2/am) fibers and particles, also caused release of Of at high concentrations (25/ag/cm 2 dish)
Release of 5~Chromium from HTE cells At concentrations o f i> 1 /ag/cm 2 dish, crocidolite asbestos, chrysotile asbestos and Code 100 fiberglass caused significant dosage-dependent increases in 5~Chromium release in HTE cells. In contrast, their non-fibrous analogs - - riebeckite, antigorite and glass beads, respectively - - did not cause enhanced release of 5~Chromium in comparison to untreated cultures.
Collagen and non-collagen protein synthesis in RL-82 cells After 24 h, no changes in the ratio of collagen to non-collagen proteins were observed in crocidolite-exposed rat lung fibroblasts. However, at 48 h, a significant (P < 0.05) increase in the ratio of collagen to non-collagen protein was
57
100-
90-
o,oo,oo,,,e
0~0
ehrysotile
e~e
antigorite
T
70
In vitro assays to predict the pathogenicity of mineral fibers B r o o k e T. M o s s m a n a n d A n n M. Sesko Department of Pathology, University of Vermont College of Medicine, Burlington, VT, 05405 (U.S.A.) (Received September 7th, 1989; accepted September 12th, 1989)
Summary A number of mineral dusts are associated with the development of inflammation in lung and pulmonary interstitial fibrosis. In an effort to find alternative approaches to animal testing, cells (rat alveolar macrophages, hamster tracheal epithelial ceils, rat lung fibroblasts) of the respiratory tract have been evaluated for cytotoxic and metabolic changes after exposure to fibers (defined as a greater than 3 : I length to diameter ratio) and particles. In all bioassays, fibrous materials provoked greater biological responses in cells in comparison to non-fibrous, chemically similar particulates at identical concentrations. For example, release o f superoxide (O:~) from alveolar macrophages (AMs) was increased (in comparison to that observed in untreated cells) after exposure to the fibers, crocidolite asbestos, erionite, Code 100 fiberglass and sepiolite. Riebeckite, mordenite and glass particles elicited minimal generation of (~2 at similar concentrations of dusts. In hamster tracheal epithelial (HTE) cells, fibers such as chrysotile asbestos, crocidolite asbestos, and Code 100 fiberglass caused increased release of 5~Chromium in comparison to the particles antigorite, riebeckite and glass. Another area of exploration is the measurement of collagen and non-collagen protein in a cell line (RL-82) of rat lung fibroblasts as an indication of the fibrogenic potential of minerals. Crocidolite asbestos caused an increase in the ratio of cell-associated collagen to non-collagen protein in these cell types whereas glass beads did not affect biosynthesis o f collagen. Results suggest that a battery of in vitro assays can be used to screen the capability of minerals to elicit cell damage and inflammatory changes in the respiratory tract.
Key words." Asbestos; Alveolar macrophages; Lung fibroblasts; Tracheobronchial epithelium; Superoxide; Active oxygen species
Introduction Occupational exposure to certain mineral dusts causes chronic bronchitis and pulmonary interstitial fibrosis in man including kaolinosis, silicosis, coal workers pneumoconiosis and asbestosis [1]. Mineral fibers such as asbestos and erionite also increase the risk of development of malignant mesothelioma and bronchogenic carcinoma, especially in smokers [2]. With the increasing commercial needs for substitutes for asbestos fibers, a number of man-made and other naturally occurring mineral fibers currently are being assessed for their pathogenic 0300-483X/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
53
potential in laboratory animals. However, these long-term studies can be timeconsuming and expensive. Moreover, the sophisticated technology required for inhalation experiments and difficulties involved in generation of respirable fibers often preclude dose-response studies and definition of " s a f e " concentrations of materials. In an effort to predict the pathogenic potential of mineral dusts in man, several in vitro assays have been developed over the past 2 decades. Initially, suspensions of red blood cells (RBCs) from several species were evaluated for hemolysis after addition of minerals [3]. The preliminary results of these experiments were encouraging in that fibrogenic minerals such as chrysotile asbestos and silica were more hemolytic than a number of 'nuisance' or inert dusts including titanium dioxide, ferric oxide and glass [4]. However, other minerals, namely crocidolite and other amphibole types of asbestos, were only weakly hemolytic but documented fibrogenic and carcinogenic agents in man. False 'positives' also were encountered in these studies. For example, montmorillonite (bentonite), a positively charged particle, was more hemolytic than chrysotile asbestos at identical concentrations [5]. Further work showed that surface charge (i.e. zeta potential) of particulates was a predominate factor in the hemolysis bioassay - - minerals (i.e. chrysotile asbestos) possessing a strongly positive surface charge were more hemolytic than others such as crocidolite asbestos which possesses a neutral or slightly negative charge [6]. The development of tissue culture techniques and cell lines in the 1970s allowed evaluation of the cytotoxic potential of mineral dusts in a variety of cell types [7]. Several important observations were made in these studies. First, although certain cell lines were more sensitive to fibers and particles in comparison to cells of other origins, chrysotile asbestos again was more toxic in most bioassays when compared to crocidolite and other amphiboles at identical concentrations. Thus, surface charge seemed to be an important determinant of membrane damage in both RBCs and mammalian ceils in general. Second, when fibers of various dimension were assessed comparatively, longer, thinner fibers proved more cytotoxic than shorter fibers, a phenomenon which correlated with their carcinogenic potential after intrapleural or intraperitoneal injection into rodents [8--10]. Lastly, fibers at non-lethal concentrations elicited proliferative and metabolic changes in some cell types in vitro [7]. During the past several years, this laboratory has developed techniques useful in understanding how fibers interact with cells of the respiratory tract. Such mechanisms of interaction may be important in the causation of disease [reviewed in 11] To determine whether fibrous geometry influences biological responses in vitro, we have examined fibers and their non-fibrous, chemically similar analogs comparatively in cells of the respiratory tract. We focus here on O F release from AMs in response to particulates, a phenomenon predictive of the inflammatory capability of fibers and release of 5~Chromium from HTE cells as an indication of the membranolytic potential of fibers. Increased biosynthesis of collagen after exposure of rat lung fibroblasts to asbestos also is presented as a tool to predict the fibrogenic potential of minerals.
54
TABLE I FIBERS A N D PARTICLES USED IN IN VITRO ASSAYS Fibers
Particles
Crocidolite [Na20 • Fe203. FeO • 8 SiO 2" H~O] a Chrysotile [3MgO. 2SiO 2- 2H20] a Erionite [Na 2" K s" C a ' Mg)4.~ (AI9Si~vOTt)'27H20] Code 100 fiberglass [SiO2] a Sepiolite [Mg, (Si205)J' OH- 6H:O]
Riebeckite [like crocidolite] Antigorite [like chrysotile] Mordenite [Ca Na 2" K2)4" AlsSi40 09( 28H20 ] Glass beads [SiO2]
aMain elemental composition. Other trace elements may be present.
Materials and methods
Minerals Fibers and respective particles that were assessed in Of studies in AMs and 51Chromium release assays in HTE ceils are listed in Table I. Both crocidolite and chrysotile asbestos were International Union Against Cancer (U.I.C.C.) reference samples. Code 100 fibers were obtained from Manville Corp., Denver, CO. Erionite (Rome, Oregon) was a gift from the MRC Pneumoconiosis Unit, Penarth, Wales. Sepiolite and mordenite were purchased from Minerals Research, Clarkstown, NY. Riebeckite, antigorite (both from Wards Scientific Establishment, Rochester, NY) and glass beads (Particle Information Service, Inc., Kingston, WA) also were employed in these studies. All preparations of dusts were characterized for purity and size distributions as reported previously [12,13].
Isolation o f AMs Two hundred 250-g Fischer 344 male rats were obtained from N.I.H. and anesthetized with phenobarbital before lavage as described in a prior publication [14]. Ceils from lavage were centrifuged at 900 x g, and differential cell counts were performed on the cell pellet to ascertain the purity of the preparation. AMs (> 98o70 of cell yield from BAL) were plated into 12-well dishes at 5 x 105 cells/ well in 1 ml of Medium 199 containing 10°70 heat-inactivated fetal bovine serum (FBS) (GIBCO, Grand Island, NY). After 2 h at 37°C, the medium was decanted, and adherent cells exposed to particulate before detection of O~.
Assay for generation of 0 5 Fibers and particles previously resuspended in complete Medium 199 were added to cells at concentrations ranging from 2.5 to 25 /ag/cm 2 surface area dish in a reaction mixture containing 1 ml of Earle's Balanced Salt Solution (GIBCO, Grand Island, NY) and 40/am cytochrome c (Type VI, horse heart; Sigma Chemical Co., St. Louis, MO). After 1 h at 37°C, the supernatant was removed from
55
control and dust-exposed cells, centrifuged at 8000 g to remove debris, and superoxide dismutase (SOD)-inhibitable reduction of cytochrome c measured spectrophotometrically at 550 nm. Cell-free reaction mixtures also were incubated with and without asbestos. Results, calculated as/amol reduced cytochrome c/5 x 105 cells/h using a 0.029 mM extinction coefficient for cytochrome c, were analyzed by the Dunnett's procedure [15].
Cultures of H T E cells The H T E diploid cell line, derived from the tracheal epithelium of a female golden Syrian hamster [16], was used in 5tChromium release assays. Cells were passaged routinely in H a m ' s F12 medium containing 10O7o FBS (GIBCO, Grand Island, NY).
Assay for release o f 51Chromium At confluency, cells were removed by trysinization and plated at 5 x 10 4 cells/well in 12-well dishes containing 1 ml H a m ' s F12 medium with 2°/0 FBS. At confluency, cells were labeled with 5tChromium (New England Nuclear, Wilmington, DE); 5/aCi/ml medium for 1 h, and then chased with unlabeled medium for an additional hour. Dusts then were added at concentrations ranging from 0.01 to 20 g g / c m 2 surface area dish, and cultures incubated at 3 7 o c for 3 h. Controls for each group consisted of untreated cells. At the end of the incubation period, both medium and solubilized cells, rinsed previously with buffered saline, were assayed for release of 5~Chromium. The release index then was calculated as (cpm medium)/[total counts (cpm medium + cpm cells)]. Regression of 5~Chromium released upon concentration of particulate applied was tested for statistical significance by A N O V A methodology [17].
Maintenance o f rat lung fibroblasts (RL-82 line) A diploid fibroblast line from the lung homogenate of a Fischer 344 rat was obtained from Dr. Marlene Absher, Department of Medicine, University of Vermont. Cells were maintained to passage 20 in Minimal Essential Medium (MEM, GIBCO, Grand Island, NY) containing 10°70 FBS. When cells approached confluency on plates, crocidolite asbestos at 1.0, 2.5 and 5.0 /ag/cm 2 dish or glass beads at 5.0/ag/cm 2 dish were added for 24-, 48- and 72-h intervals.
Assays to determine collagen and non-collagen protein synthesis To determine whether asbestos elicited increases in cell layer-associated collagen in RL-82 cells, 10-4 M ascorbate was added to all plates at 24-h intervals after exposure to dusts. [3H]proline (New England Nuclear, Wilmington, DE; 10 ~Ci/ml medium) then was added for 2 h. Cells were scraped from dishes using a rubber policeman, homogenized, and boiled for 10 min to inactivate proteases. Samples then were assayed in the presence and absence of purified bacterial collagenase (Advance Biofactures Corp; ~ 40 U / a s s a y tube) for determination of the ratio of collagen to non-collagen protein [18]. Data were analyzed by the Student's t-test adjusting for multiple comparisons between groups.
56
t500~
A.
~C.
1000 "6 o
"5 o~ 500
100
Croc. Croc. Rieb. 2.5 5.0 25
Code Sap. 100 Beads 5.0 25 25
Erion. ~rion. Mord. Mord 5.0 12.5 25 2.5
Fig. 1. Generation of superoxide (O:2) from rat AMs exposed to selected fibers and non-fibrous dusts for 1 h. Shaded and unshaded bars represent separate experiments. *Increased (P < 0.01) in comparison to untreated controls. + = increased (P < 0.05) in comparison to untreated controls. Modified with permission from Hansen and Mossman [13].
Results
Release of 05 from AMs exposed to fibrous and non-fibrous dusts Untreated plates of rat AMs generated approximately 0.172 __. 0.071 /amol reduced cytochrome c/h~5 x 105 cells (Mean of 3 experiments, N = 3 wells/ group/experiment). In comparison to untreated cultures, the fibers, crocidolite asbestos, Code 100 fiberglass and erionite caused significant dose-related increases in production of Of (Fig. 1). Five- to ten-fold greater concentrations of the particles riebeckite, glass beads, and mordenite, were required to elicit comparable responses. Sepiolite, a mixture of short (~< 2/am) fibers and particles, also caused release of Of at high concentrations (25/ag/cm 2 dish)
Release of 5~Chromium from HTE cells At concentrations o f i> 1 /ag/cm 2 dish, crocidolite asbestos, chrysotile asbestos and Code 100 fiberglass caused significant dosage-dependent increases in 5~Chromium release in HTE cells. In contrast, their non-fibrous analogs - - riebeckite, antigorite and glass beads, respectively - - did not cause enhanced release of 5~Chromium in comparison to untreated cultures.
Collagen and non-collagen protein synthesis in RL-82 cells After 24 h, no changes in the ratio of collagen to non-collagen proteins were observed in crocidolite-exposed rat lung fibroblasts. However, at 48 h, a significant (P < 0.05) increase in the ratio of collagen to non-collagen protein was
57
100-
90-
o,oo,oo,,,e
0~0
ehrysotile
e~e
antigorite
T
70
