Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure Induces Cytotoxicity and Proliferative Activity of Type I1 Pneumocytes Oliview Lesur, AndrC M. Cantin, A. Keith Tanswell, Boris Melloni, Jean-FranCois Beaulieu, and Raymond Bkgin
ABSTRACT: The contribution of the type 11 pneumocyte to the pathogenesis of silicosis is largely tmknown. Prominent featwes of silicosis are hypqlasia and hypertrophy of type 11 epithelial cells, often accompanied by phospholipid accumulation in the lung. The biologic regulation of these events is poorly understood. This study addresses the question of a direct e f f t of silica on type II pneumocytes, since direct contact of the inhaled silica dust can occur in vivo. Type 11 cells were isolated from j k a l rat lungs and their epithelial specificity was verfied. Experiments were performed on 2nd passage monolayers in 2%serum. Repair, replication, and growth activity was evaluated by the incorporation o f f Mthymidine. Cytotoxicity was measured by quantitating the release of ["qadenine and expressed as a cytotoxicity index (Cr). T p e 11cell proliferation and cytotoxicity were evaluated for the mineral dusts silica (SiOj, aluminum-treated silica (SiOfilK), and titanium (TiOJ. Ofthese mineral dusts, only low concentrations of silica increased type 11 cell [3H]thymidine incorporation (silica 2.5 pg/mL: 52% above control, P < .Or; silica 20 pg/mL: >7% above control, P < .O2), In addition, silica increased the cell number signijicantly, although to a lesser degree. Exposure of the type 11epithelial cells to silica dust for 24 h resulted in dosedependent cytotoxicity (silica 10 pg/mL, CI = 9.1%, P < .OOO2; 250 pg/mL, CI 45.1%, P < .OOOI). SiOJI completeely szdppressed these proliferation and cytotoxicity effects, which were then similar to those of the inert dust, KO,. These data suggest that direct exposure and contact of the type 11pneumocytes to lowdose silica dust initiated repair, replication, and growth activity, while exposure to higher silica concentrations resulted in marked cytotoxicity, Both the repair, replication, and growth and the cytotoxic responses of the type II epithelial cells to silica exposure are related to the surface properties of silica.
-
From the Unite' de Recherche Pulmonaire and Dkpartement dxnatomie et Biologie Cellulaire, Universite' de Sherbrooke, Sherbrooke, Quebec, Canada; and Hospitalfor Sick Children, University of Toronto, Toronto, Ontario, Canada. Address all correspondence to Raymond Be'gin, MD, Unite' de Recherche Pulmonaire, CHUS, Sberbrooke, Quebec, Canada JlH IN4. Received 25 January 1991; accepted 16 June 1991.
Experimental Lung Research 18:173-190 (1992) Copyright 0 1992 by Hemisphere Publishing Corporation
173
174
0. Lesur et al.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
INTRODUCTION Type I1 epithelial cell alterations have been observed in inflammatory fibrotic lung diseases [l, 21. Recent advances in the isolation procedures and culture conditions maintaining the type I1 pneumocyte in a differentiated state in vitro [3] have permitted major progress in the understanding of type 11 alveolar epithelial cell physiology. It has been documented that the type I1 epithelial cell, in addition to producing surfactant, acts as a stem cell to repair injured type I epithelium [4, 51. Type I1 cells can also synthesize a large variety of factors, such as matrix components [6], components of complement [7], cytokines [8], and other effector molecules. These factors are implicated in inflammatory and fibrogenic reactions. In experimental silicosis, type 11 epithelial cell hyperplasia and hypertrophy are well established [9-111, but the biologic regulation of this enhanced activity is poorly understood. In the present study, we address the question of a direct effect of silica particles on type I1 epithelial cells, utilizing an in vitro cell culture previously described by Tanswell et al. [12-141. This study could be relevant to in vivo conditions, because type I1 epithelial cells are located in part at the alveolar surface of the lung where direct contact of the inhaled dust with the epithelial cells could occur.
MATERIALS AND METHODS Isolation and Culture of Type I1 Epithelial Cells Pneumocytes from fetal rats were obtained with the minimally modified method described by Post and Torday [13]. Briefly, timed gestation SpragueDawley rats were killed by chloroform excess on day 20 of gestation (term of 22 d). The fetuses were removed following intrauterine decapitation. The lungs were minced into small pieces (1-2 mm3),and the tissue was stirred in HBSS containing 0.05% trypsin, 0.02% EDTA, and 50 pg/mL DNase (Sigma, St. Louis, MO) until dispersion. After neutralization of the proteinase activity by fetal bovine serum, FBS (Gibco Canada Inc., Burlington, Ontario), the cell suspension was filtered through a 100-pm mesh nylon bolting cloth (BSH Thompson, Mt. Royal, Quebec) and pelleted at 3009 for 10 min. The cell pellet was resuspended in MEM with Earle’s salts, 50 pg/mL gentamycin, 2.5 g/mL amphotericin By and 1% nonessential amino acids containing 10% heat-inactivated FBS at 56°C for 30 min (FBS HI), and plated in 25-cm’ culture flasks (Falcon Inc., Lincoln Park, NJ). After 60 min, the nonadherent cells were removed and allowed to stand for 60 min at 37OC before being resuspended in 1 mL fresh MEM with 10% FBS HI. This cell suspension was applied as 50-pL aliquots to l-cm3 gelatin sponges (Gelfoam, Upjohn, Don Mills, Ontario) previously saturated with
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure and Type I1 Cells In Vitro
175
medium. After 1 h, the culture flasks containing the sponges (10/flask) had MEM added to float the sponges, which were left to incubate at 37OC overnight. The next day the sponges were removed and transferred to plastic centrifugation tubes and dissolved with 1% collagenase, 0.02% EDTA in HBSS. Following the addition of FBS (lO%/volume), the cells were resuspended in fresh MEM with 10% FBS and incubated at 37OC for 15 min to allow any residual fibroblasts to attach. The nonadherent cells were grown to confluence with MEM containing 10% FBS HI. The cells of the monolayer were split with 0.05'/0 trypsin: 0.02% EDTA and replaced for another 15-min differential adherence step. These cells were seeded at a density of 15 x 103cell/well in 24-well plates with medium, in which L-valine was replaced by D-vahe to prevent fibroblast growth. Isolation, culture, and all experiments were done in a gas atmosphere of 1% 0,, 5% CO,. Before each experiment, cells were incubated with MEM D-vahe containing only 2% FBS HI for 48 h. Characterization of Type I1 Cells in the Culture System Morpbologic studies. The proliferation of fetal type I1 pneumocytes was assessed by a synchroneous study of the ['Hlthymidine incorporation and cell counting. At the second passage, the cells were seeded in 24 multiwell plates at the initial low density of 2.5 x lo4 cells/well in MEM D-Val with 10% FBS HI. Media were changed every 2 days. Tritiated thymidine incorporation was determined as described below. Cell counting was done with an electronic counter (Coulter counter ZM, Coultronics, Margency, France) after trypsinization, passage through an 18-gauge needle. Cell identification was assessed by light microscopy, after tannic acid polychrome staining, according to the method of Mason et al. [15]. In addition, the ultrastructural appearance of fetal rat lungs and 15-day-old monolayers were explored and compared by transmission EM (Philips 300 ME, Philips Electronic Equipment Ltd, Toronto, Ontario). Immunobistochemical studies. At 48 h after the first passage, type I1 cells were washed in phosphate-buffered saline (PBS) and fixed for 10 min in methanol cooled at -20°C. After two rinses in PBS, cells were permeabilized in PBS containing 0.1% triton, washed twice with PBS, and blocked in PBS containing bovine serum albumin (BSA) for 30 min. Cells were then incubated with monoclonal anti-cytokeratin-19 (Amersham Canada, Oakville, Ontario) and anti-vimentin (Boehringer Mannheirn Canada, DorVal, Quebec) diluted 1 : l O and 15, respectively, in PBS/BSA. Control cells were incubated with either PBS/BSA alone or PBS/BSA containing an unrelated immunoglobulin (monoclonal anti-a-smooth muscle actin; Sigma, St. Louis, MO). After three washes in PBS, cells were incubated with fluorescein isothiocyanate (FITC) conjugated sheep antimouse IgG (Boehringer
176
0. Lesur et al.
Mannheim Canada, Dorval, Quebec) diluted 1:25 in PBS/BSA. After 1 h incubation, the cells were washed three times in PBS, mounted in glycerol, and viewed in a Leitz Orthoplan microscope equipped for epifluorescence.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Biochemical studies. Phospholipids were studied by the analysis of [ l - 1 4 C ] a ~ eacid t i ~ incorporation (specific activity 54 mCi/mmol) (Amersham Can Ltd, Oakville, Ontario) into the cells. The [1-"Clacetic acid (2.5 pCi/well) was added to MEM D - d i n e and 2% FBS H I on subconfluent cells for 24 h at 37OC in gaseous conditions described above. The monolayers were then washed three times with 0.5 mL PBS and removed from the plates with 0.5 mL trypsin 0.05%: EDTA 0.02%. Phospholipids of each sample were extracted according to the method of Bligh and Dyer [16]. The lower hydrophobic layer was divided in two equal parts, for an individual phospholipid separated by TLC on LK5D Silica gel 105 A plates (Whatman Inc., Clifton, NJ), using chloroform/methanol/water/triethylamine (30:34:8:35) and for disaturated phosphatidylcholine (DSPC) determination by the osmium tetroxide method [17]. After the solvent migration, each sample was compared with phospholipid standards identified by ninhydrine staining, and scraped into a vial containing 5 mL scintillation fluid (Ready Safe, Beckman Inc., Montreal, Quebec). The vials were then left in the dark overnight and counted in dpm using a quenching curve previously established on a liquid scintillation counter (LKB Wallac 1215 Rack Beta, Beckman Inc., Montreal, Quebec).
Mineral Dust Assays The effects of three mineral dusts potentially able to influence the type I1 cell were assessed on the in vitro system described above. The three mineral particle dusts were of respirable size ( < 5 prn diameter), and consisted of the following: titanium dioxide (TiO,) (Kronos Inc., Paris, France), silicon dioxide (SiO,) (Minusil-5, PGS Pennsylvania Glass Sand Corp, Pittsburg, PA), and aluminum treated silica (Si0,Al) [18]. Each dust was prepared in an initial suspension of 500 pg/mL, and vortexed before using in successive dilutions and multiwell seeding. To test the effects of these mineral dusts on the type I1 epithelial cells, the following procedures were carried out.
tH]7%ymidine incovoration assays. The assays were carried out by counting [methyl-)H]thymidine incorporation into DNA (specific activity 5 Ci/mmol) (Amersham Can Ltd, Oakville, Ontario). All tests were carried out in MEM D-vahe and 2% FBS HI and in triplicate, at several concentrations (range 2.5 to 500 pg/mL) of the minerals. During the 2nd passage, following a new differential adherence period, the type I1 cells were seeded in 24-well tissue culture plates at about 3 x lo4 cells per well in MEM D-vahe with
177
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure and Type I1 Cells In Vitro
10% FBS HI and incubated at 37°C in standard gaseous atmosphere. Media and atmosphere were changed every 2nd until subconfluence was reached. Then, [3H]thymidine(0.5 pCi/well) in MEM D-vahe with 2% FBS HI and with the mineral of interest for each test was applied. Twenty-four hours later, the cell cultures were rinsed twice with cold PBS and twice with iced 10% TCA. A drop (50 pL) of ethanol-ether (3:l) was put into each well and plates were allowed to air dry, Then 0.6 mL of 0.2N sodium hydroxide (NaOH) was added for 15 min; 0.5 mL of the resulting solution was removed from wells and placed into corresponding scintillation vials containing 5 mL of Ready Safe and 50 pL glacial acetic acid for counting on a dpm quenching curve previously established, after a 2-h dark period. The SiO, doses were used in a large range of concentrations (0.6-500 pg/mL), while the two other dusts (SiO,Al, TiO,) were studied in the range of 10-500 pg/mL, the latter dusts being inactive. To look for a possible autocrine regulation of the type I1 cell growth activity, conditioned media from type I1 cells exposed for 24 h to 2.5-50 pg/mL of the dusts were added to fresh unstimulated cells, and additional [3H]thymidineincorporation assays were done.
Cell counts. To evaluate changes in cell number, type I1 pneumocytes were also seeded in large plastic plates (9.6 cm', Falcon Inc., Lincoln Park, NJ) at a density of 5 x 105/plate, After 24 h incubation in MEM 10% FBS H1,the cells were exposed to five different conditions (MEM D-vahe with 2% FBS HI, silica: 2.5, 20, and 500 pg/mL, and insulin 10 pg/mL in MEM D-vahe with 10% FBS HI) for 48 h. Type I1 cells were counted on a hemacytometer after trypsinization in a small volume; the numeration was converted into the total number of celldplate and then as a percentage of the negative control. The hemacytometric method was preferred here for cell counting because quartz particles in samples can produce false electronic count readings. Cytotoxicity tests. The [8-14C]adenine (specific activity 55 mCi/mmol) (Amersham Can Ltd, Oakville, Ontario) was added at 0.05 pCi/well on subconfluent monolayers in MEM D-valine with 2% FBS HI. After 24 h of incubation, the culture media were removed with 3 x 0.5-mL PBS washes to discard any residual extracellular isotope. The three mineral dusts (SiO,, SiO,Al, TiO,) (range 10-500 pg/mL) were added in MEM D-vahe with 2% FBS HI in triplicate and separate experiments. After 24 h incubation, the supernatants were withdrawn, followed by 2 x 0.5 mL PBS washes, and the leftover materials were transferred into scintillation vials containing 5 mL of Ready Safe. The dpm measurement was done after 12 h in the dark by a liquid scintillation counter on a quenching curve previously established. For each condition (dust type, dust dose), cell injury was evaluated in terms of a dpm cytotoxicity index (CI) expressed as (A-B)/(C-B)x 100, where A
-
178
0. Lesur et al.
released into the media of the test sample, B dpm released from the control cells (background release), and C = dpm released from cells removed by trypsin 0.05%: EDTA 0.02%.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Statistical Analysis Data are expressed as means f SEM. StatisticaI analysis was performed using a variance analysis for repeated measures (ANOVA) with an LY value of 0.05 and a fl value of 0.20 (power of 0.80). A Dunnett t test for comparison with control data was used for each assay. To evaluate association between parameters of cell activity and dust doses, we used the Spearman correlation procedure [19]. RESULTS Characterization of Type I1 Epithelial Cells The results of the proliferation kinetic study over a %day period are presented in Fig, 1. We found a time-related increase in cell number, and the doubling time of the type I1 cell population at this density seeding was 4 d, which was paralleled by [3H]thymidine incorporation (Fig. 1). The purity of the type I1 epithelial cell population at the 2nd passage
2
0
4
6
8
10
Days from plating Figure 1 G r o w h of fetal rat type I1 pneumocyte over 9 d in culture, assessed by tritiated thymidine incorporation rate and cell counting. The assays were done under the in vitro conditions as used for fetal type II cell growth (i.e., MEM D Val with 10% FBS HI, 1% 0 2 : 5% COJ. All points are mean f SD of triplicate samples. Cell counting curve; 0,thymidine incorporation curve. The doubling time of the cell population was reached at 4 d (---).
+,
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure and Type I1 Cells In Vitro
179
Figure 2 Upper Light microscopy (upper panel) of a type I1 pneumocyte monolayer (15 d) stained with rannic acid-polychrome. Bar = 100 pm. Lower: A type I1 pneumocyte loaded with rounded dense black bodies (tannic aid-polychrome staining). Bar = 1 pm.
(15 f 1 d) was 93 -t 3%; the cell viability, tested for each experiment, was 95 f 2% (trypan blue dye exclusion test). Three reference methods were used for the characterization: (I) Light microscopy after tannic acid polychrome staining [15], which is specific by its pattern of distribution into rounded lamellar bodies (Fig. 2). Other cells, especially residual fibroblast contaminating cells, show a more spindle-shaped morphology and a shadow-like perinuclear distribution of tannic acid. (2) Transmission electronic microscopy, where the ultrastructural appearance of fe-
180
0. Lesur
et al.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
tal type I1 cells at 15 d of culture showed typical patterns (glycogen pools, dilated rough endoplasmic reticulum, microvillosities, prominent Golgi apparatus, many large mitochondria, and dense bodies), which is representative of the active metabolism of the type I1 cells [14,20]. Different kinds of osmiophilic structures were observed, from the dense immature basal granules to the multivesicular bodies with a mixture of lamellae and homogeneous material and also some more mature form of lamellar bodies (Fig. 3). (3) Immunocytochemical staining for Cytokeratin 19, a cytoskeletal intermediate filament specific for epithelial cells and the absence of staining for Vimentin, a cytoskeletal intermediate filament specific for the mesenchymal cells [21] (Fig. 4). The profile of [14C]acetate incorporation into each major cellular phospholipid is presented in Table 1 for type I1 cells and fibroblasts. The results were in agreement with the literature [13,22], and reproduced the relatively distinctive disaturated phosphatidylcholine percentages between epithelial cells and fibroblasts. In addition, quartz exposure below 50 pg/rnL for 24 h did not significantly alter the [14C]acetate incorporation into type 11 cells (not shown). These morphological, immunohistochemical, and biochemical observations documented the differentiate state of the type I1 epithelial cells in our in vitro culture system.
Mineral Dust Assays The direct exposure of type I1 cells to the silica mineral dust produced a significant stimulation of [3H]thymidine incorporation (Fig. 5) between 52% (P < .05)and 57% ( P < .02) above control for 2.5 and 20 pg/mL of SiO,, respectively. The inert dusts, Si0,Al and TiO,, did not affect the rate of the radionucleotide integration. Higher doses of the three tested dusts reduced [3H]thymidine incorporation, The enhancements of ['Hlthymidine incorporation induced by silica contact were paralleled by cell number increases (Fig. 6). The cytotoxicity index (Fig. 7) demonstrated the toxic activity of silica on type I1 cells in a dose-dependent manner, with a maximum of 45.1 f 2.5% (P < .OOOl) at 250 pg/mL of SiO,. Low doses (
Silica Exposure Induces Cytotoxicity and Proliferative Activity of Type I1 Pneumocytes Oliview Lesur, AndrC M. Cantin, A. Keith Tanswell, Boris Melloni, Jean-FranCois Beaulieu, and Raymond Bkgin
ABSTRACT: The contribution of the type 11 pneumocyte to the pathogenesis of silicosis is largely tmknown. Prominent featwes of silicosis are hypqlasia and hypertrophy of type 11 epithelial cells, often accompanied by phospholipid accumulation in the lung. The biologic regulation of these events is poorly understood. This study addresses the question of a direct e f f t of silica on type II pneumocytes, since direct contact of the inhaled silica dust can occur in vivo. Type 11 cells were isolated from j k a l rat lungs and their epithelial specificity was verfied. Experiments were performed on 2nd passage monolayers in 2%serum. Repair, replication, and growth activity was evaluated by the incorporation o f f Mthymidine. Cytotoxicity was measured by quantitating the release of ["qadenine and expressed as a cytotoxicity index (Cr). T p e 11cell proliferation and cytotoxicity were evaluated for the mineral dusts silica (SiOj, aluminum-treated silica (SiOfilK), and titanium (TiOJ. Ofthese mineral dusts, only low concentrations of silica increased type 11 cell [3H]thymidine incorporation (silica 2.5 pg/mL: 52% above control, P < .Or; silica 20 pg/mL: >7% above control, P < .O2), In addition, silica increased the cell number signijicantly, although to a lesser degree. Exposure of the type 11epithelial cells to silica dust for 24 h resulted in dosedependent cytotoxicity (silica 10 pg/mL, CI = 9.1%, P < .OOO2; 250 pg/mL, CI 45.1%, P < .OOOI). SiOJI completeely szdppressed these proliferation and cytotoxicity effects, which were then similar to those of the inert dust, KO,. These data suggest that direct exposure and contact of the type 11pneumocytes to lowdose silica dust initiated repair, replication, and growth activity, while exposure to higher silica concentrations resulted in marked cytotoxicity, Both the repair, replication, and growth and the cytotoxic responses of the type II epithelial cells to silica exposure are related to the surface properties of silica.
-
From the Unite' de Recherche Pulmonaire and Dkpartement dxnatomie et Biologie Cellulaire, Universite' de Sherbrooke, Sherbrooke, Quebec, Canada; and Hospitalfor Sick Children, University of Toronto, Toronto, Ontario, Canada. Address all correspondence to Raymond Be'gin, MD, Unite' de Recherche Pulmonaire, CHUS, Sberbrooke, Quebec, Canada JlH IN4. Received 25 January 1991; accepted 16 June 1991.
Experimental Lung Research 18:173-190 (1992) Copyright 0 1992 by Hemisphere Publishing Corporation
173
174
0. Lesur et al.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
INTRODUCTION Type I1 epithelial cell alterations have been observed in inflammatory fibrotic lung diseases [l, 21. Recent advances in the isolation procedures and culture conditions maintaining the type I1 pneumocyte in a differentiated state in vitro [3] have permitted major progress in the understanding of type 11 alveolar epithelial cell physiology. It has been documented that the type I1 epithelial cell, in addition to producing surfactant, acts as a stem cell to repair injured type I epithelium [4, 51. Type I1 cells can also synthesize a large variety of factors, such as matrix components [6], components of complement [7], cytokines [8], and other effector molecules. These factors are implicated in inflammatory and fibrogenic reactions. In experimental silicosis, type 11 epithelial cell hyperplasia and hypertrophy are well established [9-111, but the biologic regulation of this enhanced activity is poorly understood. In the present study, we address the question of a direct effect of silica particles on type I1 epithelial cells, utilizing an in vitro cell culture previously described by Tanswell et al. [12-141. This study could be relevant to in vivo conditions, because type I1 epithelial cells are located in part at the alveolar surface of the lung where direct contact of the inhaled dust with the epithelial cells could occur.
MATERIALS AND METHODS Isolation and Culture of Type I1 Epithelial Cells Pneumocytes from fetal rats were obtained with the minimally modified method described by Post and Torday [13]. Briefly, timed gestation SpragueDawley rats were killed by chloroform excess on day 20 of gestation (term of 22 d). The fetuses were removed following intrauterine decapitation. The lungs were minced into small pieces (1-2 mm3),and the tissue was stirred in HBSS containing 0.05% trypsin, 0.02% EDTA, and 50 pg/mL DNase (Sigma, St. Louis, MO) until dispersion. After neutralization of the proteinase activity by fetal bovine serum, FBS (Gibco Canada Inc., Burlington, Ontario), the cell suspension was filtered through a 100-pm mesh nylon bolting cloth (BSH Thompson, Mt. Royal, Quebec) and pelleted at 3009 for 10 min. The cell pellet was resuspended in MEM with Earle’s salts, 50 pg/mL gentamycin, 2.5 g/mL amphotericin By and 1% nonessential amino acids containing 10% heat-inactivated FBS at 56°C for 30 min (FBS HI), and plated in 25-cm’ culture flasks (Falcon Inc., Lincoln Park, NJ). After 60 min, the nonadherent cells were removed and allowed to stand for 60 min at 37OC before being resuspended in 1 mL fresh MEM with 10% FBS HI. This cell suspension was applied as 50-pL aliquots to l-cm3 gelatin sponges (Gelfoam, Upjohn, Don Mills, Ontario) previously saturated with
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure and Type I1 Cells In Vitro
175
medium. After 1 h, the culture flasks containing the sponges (10/flask) had MEM added to float the sponges, which were left to incubate at 37OC overnight. The next day the sponges were removed and transferred to plastic centrifugation tubes and dissolved with 1% collagenase, 0.02% EDTA in HBSS. Following the addition of FBS (lO%/volume), the cells were resuspended in fresh MEM with 10% FBS and incubated at 37OC for 15 min to allow any residual fibroblasts to attach. The nonadherent cells were grown to confluence with MEM containing 10% FBS HI. The cells of the monolayer were split with 0.05'/0 trypsin: 0.02% EDTA and replaced for another 15-min differential adherence step. These cells were seeded at a density of 15 x 103cell/well in 24-well plates with medium, in which L-valine was replaced by D-vahe to prevent fibroblast growth. Isolation, culture, and all experiments were done in a gas atmosphere of 1% 0,, 5% CO,. Before each experiment, cells were incubated with MEM D-vahe containing only 2% FBS HI for 48 h. Characterization of Type I1 Cells in the Culture System Morpbologic studies. The proliferation of fetal type I1 pneumocytes was assessed by a synchroneous study of the ['Hlthymidine incorporation and cell counting. At the second passage, the cells were seeded in 24 multiwell plates at the initial low density of 2.5 x lo4 cells/well in MEM D-Val with 10% FBS HI. Media were changed every 2 days. Tritiated thymidine incorporation was determined as described below. Cell counting was done with an electronic counter (Coulter counter ZM, Coultronics, Margency, France) after trypsinization, passage through an 18-gauge needle. Cell identification was assessed by light microscopy, after tannic acid polychrome staining, according to the method of Mason et al. [15]. In addition, the ultrastructural appearance of fetal rat lungs and 15-day-old monolayers were explored and compared by transmission EM (Philips 300 ME, Philips Electronic Equipment Ltd, Toronto, Ontario). Immunobistochemical studies. At 48 h after the first passage, type I1 cells were washed in phosphate-buffered saline (PBS) and fixed for 10 min in methanol cooled at -20°C. After two rinses in PBS, cells were permeabilized in PBS containing 0.1% triton, washed twice with PBS, and blocked in PBS containing bovine serum albumin (BSA) for 30 min. Cells were then incubated with monoclonal anti-cytokeratin-19 (Amersham Canada, Oakville, Ontario) and anti-vimentin (Boehringer Mannheirn Canada, DorVal, Quebec) diluted 1 : l O and 15, respectively, in PBS/BSA. Control cells were incubated with either PBS/BSA alone or PBS/BSA containing an unrelated immunoglobulin (monoclonal anti-a-smooth muscle actin; Sigma, St. Louis, MO). After three washes in PBS, cells were incubated with fluorescein isothiocyanate (FITC) conjugated sheep antimouse IgG (Boehringer
176
0. Lesur et al.
Mannheim Canada, Dorval, Quebec) diluted 1:25 in PBS/BSA. After 1 h incubation, the cells were washed three times in PBS, mounted in glycerol, and viewed in a Leitz Orthoplan microscope equipped for epifluorescence.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Biochemical studies. Phospholipids were studied by the analysis of [ l - 1 4 C ] a ~ eacid t i ~ incorporation (specific activity 54 mCi/mmol) (Amersham Can Ltd, Oakville, Ontario) into the cells. The [1-"Clacetic acid (2.5 pCi/well) was added to MEM D - d i n e and 2% FBS H I on subconfluent cells for 24 h at 37OC in gaseous conditions described above. The monolayers were then washed three times with 0.5 mL PBS and removed from the plates with 0.5 mL trypsin 0.05%: EDTA 0.02%. Phospholipids of each sample were extracted according to the method of Bligh and Dyer [16]. The lower hydrophobic layer was divided in two equal parts, for an individual phospholipid separated by TLC on LK5D Silica gel 105 A plates (Whatman Inc., Clifton, NJ), using chloroform/methanol/water/triethylamine (30:34:8:35) and for disaturated phosphatidylcholine (DSPC) determination by the osmium tetroxide method [17]. After the solvent migration, each sample was compared with phospholipid standards identified by ninhydrine staining, and scraped into a vial containing 5 mL scintillation fluid (Ready Safe, Beckman Inc., Montreal, Quebec). The vials were then left in the dark overnight and counted in dpm using a quenching curve previously established on a liquid scintillation counter (LKB Wallac 1215 Rack Beta, Beckman Inc., Montreal, Quebec).
Mineral Dust Assays The effects of three mineral dusts potentially able to influence the type I1 cell were assessed on the in vitro system described above. The three mineral particle dusts were of respirable size ( < 5 prn diameter), and consisted of the following: titanium dioxide (TiO,) (Kronos Inc., Paris, France), silicon dioxide (SiO,) (Minusil-5, PGS Pennsylvania Glass Sand Corp, Pittsburg, PA), and aluminum treated silica (Si0,Al) [18]. Each dust was prepared in an initial suspension of 500 pg/mL, and vortexed before using in successive dilutions and multiwell seeding. To test the effects of these mineral dusts on the type I1 epithelial cells, the following procedures were carried out.
tH]7%ymidine incovoration assays. The assays were carried out by counting [methyl-)H]thymidine incorporation into DNA (specific activity 5 Ci/mmol) (Amersham Can Ltd, Oakville, Ontario). All tests were carried out in MEM D-vahe and 2% FBS HI and in triplicate, at several concentrations (range 2.5 to 500 pg/mL) of the minerals. During the 2nd passage, following a new differential adherence period, the type I1 cells were seeded in 24-well tissue culture plates at about 3 x lo4 cells per well in MEM D-vahe with
177
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure and Type I1 Cells In Vitro
10% FBS HI and incubated at 37°C in standard gaseous atmosphere. Media and atmosphere were changed every 2nd until subconfluence was reached. Then, [3H]thymidine(0.5 pCi/well) in MEM D-vahe with 2% FBS HI and with the mineral of interest for each test was applied. Twenty-four hours later, the cell cultures were rinsed twice with cold PBS and twice with iced 10% TCA. A drop (50 pL) of ethanol-ether (3:l) was put into each well and plates were allowed to air dry, Then 0.6 mL of 0.2N sodium hydroxide (NaOH) was added for 15 min; 0.5 mL of the resulting solution was removed from wells and placed into corresponding scintillation vials containing 5 mL of Ready Safe and 50 pL glacial acetic acid for counting on a dpm quenching curve previously established, after a 2-h dark period. The SiO, doses were used in a large range of concentrations (0.6-500 pg/mL), while the two other dusts (SiO,Al, TiO,) were studied in the range of 10-500 pg/mL, the latter dusts being inactive. To look for a possible autocrine regulation of the type I1 cell growth activity, conditioned media from type I1 cells exposed for 24 h to 2.5-50 pg/mL of the dusts were added to fresh unstimulated cells, and additional [3H]thymidineincorporation assays were done.
Cell counts. To evaluate changes in cell number, type I1 pneumocytes were also seeded in large plastic plates (9.6 cm', Falcon Inc., Lincoln Park, NJ) at a density of 5 x 105/plate, After 24 h incubation in MEM 10% FBS H1,the cells were exposed to five different conditions (MEM D-vahe with 2% FBS HI, silica: 2.5, 20, and 500 pg/mL, and insulin 10 pg/mL in MEM D-vahe with 10% FBS HI) for 48 h. Type I1 cells were counted on a hemacytometer after trypsinization in a small volume; the numeration was converted into the total number of celldplate and then as a percentage of the negative control. The hemacytometric method was preferred here for cell counting because quartz particles in samples can produce false electronic count readings. Cytotoxicity tests. The [8-14C]adenine (specific activity 55 mCi/mmol) (Amersham Can Ltd, Oakville, Ontario) was added at 0.05 pCi/well on subconfluent monolayers in MEM D-valine with 2% FBS HI. After 24 h of incubation, the culture media were removed with 3 x 0.5-mL PBS washes to discard any residual extracellular isotope. The three mineral dusts (SiO,, SiO,Al, TiO,) (range 10-500 pg/mL) were added in MEM D-vahe with 2% FBS HI in triplicate and separate experiments. After 24 h incubation, the supernatants were withdrawn, followed by 2 x 0.5 mL PBS washes, and the leftover materials were transferred into scintillation vials containing 5 mL of Ready Safe. The dpm measurement was done after 12 h in the dark by a liquid scintillation counter on a quenching curve previously established. For each condition (dust type, dust dose), cell injury was evaluated in terms of a dpm cytotoxicity index (CI) expressed as (A-B)/(C-B)x 100, where A
-
178
0. Lesur et al.
released into the media of the test sample, B dpm released from the control cells (background release), and C = dpm released from cells removed by trypsin 0.05%: EDTA 0.02%.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Statistical Analysis Data are expressed as means f SEM. StatisticaI analysis was performed using a variance analysis for repeated measures (ANOVA) with an LY value of 0.05 and a fl value of 0.20 (power of 0.80). A Dunnett t test for comparison with control data was used for each assay. To evaluate association between parameters of cell activity and dust doses, we used the Spearman correlation procedure [19]. RESULTS Characterization of Type I1 Epithelial Cells The results of the proliferation kinetic study over a %day period are presented in Fig, 1. We found a time-related increase in cell number, and the doubling time of the type I1 cell population at this density seeding was 4 d, which was paralleled by [3H]thymidine incorporation (Fig. 1). The purity of the type I1 epithelial cell population at the 2nd passage
2
0
4
6
8
10
Days from plating Figure 1 G r o w h of fetal rat type I1 pneumocyte over 9 d in culture, assessed by tritiated thymidine incorporation rate and cell counting. The assays were done under the in vitro conditions as used for fetal type II cell growth (i.e., MEM D Val with 10% FBS HI, 1% 0 2 : 5% COJ. All points are mean f SD of triplicate samples. Cell counting curve; 0,thymidine incorporation curve. The doubling time of the cell population was reached at 4 d (---).
+,
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
Silica Exposure and Type I1 Cells In Vitro
179
Figure 2 Upper Light microscopy (upper panel) of a type I1 pneumocyte monolayer (15 d) stained with rannic acid-polychrome. Bar = 100 pm. Lower: A type I1 pneumocyte loaded with rounded dense black bodies (tannic aid-polychrome staining). Bar = 1 pm.
(15 f 1 d) was 93 -t 3%; the cell viability, tested for each experiment, was 95 f 2% (trypan blue dye exclusion test). Three reference methods were used for the characterization: (I) Light microscopy after tannic acid polychrome staining [15], which is specific by its pattern of distribution into rounded lamellar bodies (Fig. 2). Other cells, especially residual fibroblast contaminating cells, show a more spindle-shaped morphology and a shadow-like perinuclear distribution of tannic acid. (2) Transmission electronic microscopy, where the ultrastructural appearance of fe-
180
0. Lesur
et al.
Exp Lung Res Downloaded from informahealthcare.com by V U L Periodicals Rec on 12/28/14 For personal use only.
tal type I1 cells at 15 d of culture showed typical patterns (glycogen pools, dilated rough endoplasmic reticulum, microvillosities, prominent Golgi apparatus, many large mitochondria, and dense bodies), which is representative of the active metabolism of the type I1 cells [14,20]. Different kinds of osmiophilic structures were observed, from the dense immature basal granules to the multivesicular bodies with a mixture of lamellae and homogeneous material and also some more mature form of lamellar bodies (Fig. 3). (3) Immunocytochemical staining for Cytokeratin 19, a cytoskeletal intermediate filament specific for epithelial cells and the absence of staining for Vimentin, a cytoskeletal intermediate filament specific for the mesenchymal cells [21] (Fig. 4). The profile of [14C]acetate incorporation into each major cellular phospholipid is presented in Table 1 for type I1 cells and fibroblasts. The results were in agreement with the literature [13,22], and reproduced the relatively distinctive disaturated phosphatidylcholine percentages between epithelial cells and fibroblasts. In addition, quartz exposure below 50 pg/rnL for 24 h did not significantly alter the [14C]acetate incorporation into type 11 cells (not shown). These morphological, immunohistochemical, and biochemical observations documented the differentiate state of the type I1 epithelial cells in our in vitro culture system.
Mineral Dust Assays The direct exposure of type I1 cells to the silica mineral dust produced a significant stimulation of [3H]thymidine incorporation (Fig. 5) between 52% (P < .05)and 57% ( P < .02) above control for 2.5 and 20 pg/mL of SiO,, respectively. The inert dusts, Si0,Al and TiO,, did not affect the rate of the radionucleotide integration. Higher doses of the three tested dusts reduced [3H]thymidine incorporation, The enhancements of ['Hlthymidine incorporation induced by silica contact were paralleled by cell number increases (Fig. 6). The cytotoxicity index (Fig. 7) demonstrated the toxic activity of silica on type I1 cells in a dose-dependent manner, with a maximum of 45.1 f 2.5% (P < .OOOl) at 250 pg/mL of SiO,. Low doses (
