The Response of Guinea Pig Airway Epithelial Cells and Alveolar Macrophages to Environmental Stress D. Scott Cohen, Elise Palmer, William J. Welch, and Dean Sheppard Lung Biology Center, Department of Medicine, University of California, San Francisco, California
Cells lining the respiratory tract form an interface between the organism and the external environment and are repeatedly exposed to physical, chemical, and metabolic stresses. We examined the response of cultured guinea pig tracheal epithelial cells and alveolar macrophages to various forms of stress, including clinically and environmentally relevant metabolic stresses such as ozone and acid exposure. Classic stress treatments such as heat shock and sodium arsenite treatment induced the synthesis of 28, 32, 72, 73, 90, and 110 kD stress proteins similar to those observed in other cell types. In contrast, no significant changes in the pattern of protein synthesis were detected after exposure to ambient concentrations of ozone, although ozone exposure caused significant cytotoxicity to both cell types. Another potent oxidant, hydrogen peroxide, similarly did not induce appreciable stress protein synthesis. However, surface acidification of tracheal epithelial cells and alveolar macrophages caused the induction of 72 and 78 kD stress proteins. While stress proteins may playa role in the response of respiratory cells to certain injuries such as hyperthermia and surface acidification, they may not be important in the defense against ozone or other forms of oxidative injury.
Cells lining the respiratory tract are situated at an air-tissue interface where the organism encounters the external environment. These cells represent the organism's "first line of defense" against inhaled or aspirated particles, organisms, and pollutants such as ozone (1, 2). Alterations in gene expression following environmental stress in respiratory epithelial cells and alveolar macrophages have not been well characterized. Most organisms defend themselves against adverse changes in their environment by altering their patterns of gene expression. One remarkably well-conserved cellular response to abrupt changes in local environmental conditions consists of the increased and selective synthesis of a small group of proteins referred to as heat shock or stress proteins. The transient synthesis of this family of proteins after heat shock treatment or after other physical, chemical, and metabolic insults has been observed with only minor variations in most cell types studied to date (3-5). Stress proteins perform important metabolic functions in the unstressed cell and appear to function in a protective or reparative manner after stress (6-11). Environmental stresses (Received in original form June 27, 1990 and in revised form November 27, 1990) Address correspondence to: Dean Sheppard, M.D., Lung Biology Center, University of California, San Francisco, Box 0854, San Francisco, CA 94143-0854. Abbreviations: heat shock proteins, hsp; lactate dehydrogenase, LDH; phosphate-buffered saline, PBS; sodium dodecyl sulfate polyacrylamide gel electrophoresis, SDS-PAGE. Am. J. Respir. Cell Mol. BioI. Vol. 5. pp. 133-143, 1991
relevant to specific cell types, such as the effects of ozone 01 acid exposure on respiratory cells, have not been studied in the context of the stress response. Studies in both animals and humans have demonstrated significant functional and structural respiratory tract abnormalities after exposure to ambient concentrations of ozone (2, 12-14). The effects of ozone on specific lung cell types have been difficult to study due to difficulties in establishing an in vitro exposure system that mimics in vivo conditions. Previous in vitro studies reported minimal or no cytotoxic effects of ambient concentrations of ozone on cultured epithelial cells and alveolar macrophages (15-17). In contrast, a recent in vivo study in human subjects described increased lactate dehydrogenase (LDH) release into bronchoalveolar lavagefluid after exposure to as little as 0.08 parts per million (ppm) of ozone, suggestive of ozone-induced cytotoxicity (18, 19). Although the basis for this difference between in vitro and in vivo findings is unclear, previous in vitro exposure systems utilized rocker platforms or tilted rotating plates that interposed a layer of medium between the cells and gaseous ozone for all or part of the exposure period. This liquid barrier may have resulted in decreased contact between the ozone gas and the cultured cells under study. Airway epithelial cells and alveolar macrophages are frequently exposed to acidic conditions due to the aspiration of acidic gastric contents into the respiratory tract during deep sleep (20). Although acid aspiration has been associated with severe lung injury and significant mortality in certain clinical situations (21,22), most people do not develop lung injury after nocturnal acid aspiration. Inhalation of acid aerosols into the respiratory tract as a result of acidic air pol-
134
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
lution has also been associated with significant respiratory morbidity and mortality (23, 24). The effects of surface acidification on gene expression in respiratory cells have not been characterized. In the present study, we examined the response of two respiratory cell types, alveolar macrophages and tracheal epithelial cells, to various forms of environmental stress. Utilizing an exposure system designed to simulate in vivo exposure conditions, we examined the patterns of protein expression in cultured tracheal epithelial cells and alveolar macrophages after heat shock, sodium arsenite, hydrogen peroxide, and acid exposure as well as after ambient concentrations of ozone.
Materials and Methods Epithelial Cell Culture Male Hartley outbred guinea pigs weighing 500 to 1,000 g were obtained from a caesarean-originated, barrier-sustained colony (Charles River Breeding Laboratories, Stoneridge, NY). After anesthetization with pentobarbital (390 mg/kg) and exsanguination via cardiac puncture, the trachea and mainstem bronchi were removed, incised longitudinally, and incubated at 37° C in Hanks' balanced salt solution containing 0.1% type XIV protease from Streptomyces griseus (Sigma Chemical Co., S1. Louis, MO) as described by Adler and co-workers (25). The airway epithelium was removed by gentle scraping with a sterile no. 10 scalpel blade, disaggregated by repeated pipetting through a sterile pasteur pipette, and pelleted at 200 x g in a bench-top centrifuge. The epithelial cells were washed twice in Ham's F-12 medium (UCSF Cell Culture Facility, San Francisco, CA) before resuspension in supplemented Ham's F-12 medium and plating on collagen-treated filter supports (Transwell-COL, 0.45-J"m pore size, 24.5-mm diameter; Costar, Cambridge, MA) at a density of 5.0 x 105 cells/well. Fresh medium was added above and below the cell cultures 24 h after plating and subsequently at 48- to 72-h intervals. Growth medium consisted of Ham's F-12 medium supplemented with transferrin (5 J.tg/rnl; GIBCO, Grand Island, NY), insulin (5 J"g/ml, bovine; GIBCO), cholera toxin (40 ng/rnl; Sigma), epidermal growth factor (15 ng/rnl, mouse SUbmaxillary gland; Sigma), hydrocortisone (360 J.tg/ml; Sigma), hypothalamus extract (12 J.tg/rnl, bovine, ECGS; Collaborative Research, Bedford, MA), retinoic acid (30 J.tg/rnl; Sigma), penicillin (100 J.tg/rnl; UCSF Cell Culture Facility), streptomycin (100 J.tg/rnl; UCSF Cell Culture Facility), fungizone (0.5 J.tg/ml; USCF Cell Culture Facility), and fetal calf serum (5% vol/vol; UCSF Cell Culture Facility). Airway epithelial cell cultures continued to proliferate for 7 to 10 days. Confluent monolayers were studied between days 10 and 15. Electron micrographs of cell monolayers demonstrated intercellular tight junctions and immunofluorescent staining demonstrated the presence of keratin, consistent with a population of differentiated epithelial cells. Cell cultures were washed and placed in unsupplemented growth medium immediately prior to stress treatments. Alveolar Macrophage Cell Culture Male Hartley outbread guinea pigs weighing 500 to 1,000 g were obtained from a caesarean-originated, barrier-sustained
colony (Charles River Breeding Laboratories). After anesthetization with pentobarbital (390 mg/kg), a tracheostomy was placed and total lung lavage using five 9-ml aliquots of 3r C Ca/Mg/endotoxin-free Hanks' balanced salt solution was performed as described by Holt (26). A typical lavage yielded 1.0 to 1.5 X 107 cells, comprised of 90 to 95 % macrophages. Granulocytes were removed using discontinuous Percoll gradient centrifugation. After being washed and resuspended in endotoxin-free RPMI 1640 medium (UCSF Cell Culture Facility), purified macrophages were plated on collagen-treated filter supports (Transwell-COL, 0.45-J.tm pore size, 24.5-mm diameter; Costar) at a density of 1.0 to 1.5 X 105 cells/well. The cell cultures were incubated at 37° C (in a humidified 95% air/5% CO 2 atmosphere) for 1 h to allow adherence. Cell cultures were washed and placed in unsupplemented growth medium immediately prior to stress treatments. Heat Shock, Arsenite, and Hydrogen Peroxide Exposures For heat shock experiments, cell culture dishes were placed in 37°, 42°, or 43° C waterbaths within incubators at the same temperatures for 45 min, returned to 37° C, then metabolically labeled with radiolabeled amino acids. Cultured cells were exposed to sodium arsenite (Sigma) 0 to 150 J.tM for 2 h or to hydrogen peroxide (Sigma) 10-5 to 10-1 M for 1 h at 37° C. After the exposure periods, cells were washed and returned to growth medium for metabolic labeling. Cells were maintained in unsupplemented growth medium during exposures and subsequent radiolabeling. Surface Acidification The apical surfaces of tracheal epithelial cells were exposed for 4 h to 0.5 rnl of 0.05 M 2-(N-morpholino)ethanesulfonic acid buffer in sodium chloride solution (final osmolarity, 300 mOsm) with the pH adjusted to 5.0,6.0, or 7.0by the addition of 1 N hydrochloric acid or 1 N sodium hydroxide (with normal growth medium in the basal compartment). Immediately after the exposure period, the cells were washed and returned to growth medium prior to metabolic labeling. For alveolar macrophage experiments, 0.05 M 2-(N-morpholino)ethanesulfonic acid buffer was prepared in RPMI medium with the pH adjusted tb 5.3, 5.8, 6.3, 6.8, or 7.20. After a 2-h exposure period, the alveolar macrophages were washed and returned to growth medium prior to metabolic labeling. Measurements of pH performed on the various media before and after exposure periods confirmed that a consistent degree of acidity was maintained throughout the exposure period. Ozone Exposure Tracheal epithelial cells and alveolar macrophages cultured on microporous filter supports were exposed to air or ozone after removal of the overlying medium to allow direct contact between the cell surface and the exposure atmosphere. Culture dishes were placed in two 27 x 22 x 17 em 'Ieflon'"lined acrylic exposure chambers in separate incubators. The exposure temperature was strictly maintained at 37° C (confirmed by thermistor probes inside both exposure chambers). An ultraviolet light source (Jelight Co., Laguna Hills, CA) was used to generate small quantities of ozone from a 95 % air/5 % CO2 gas mixture at low flow rates which was
Cohen, Palmer, Welch et al.: Stress Response of Epithelial Cells and Alveolar Macrophages
B
A
116 -
68-
43 -
ter metabolic labeling for 1.5 to 3 h, the medium was removed and the cells were washed with cold phosphatebuffered saline (PBS). The cells were then solubilized in Laemmli sample buffer, treated with nuclease, boiled at 100° C for 3 min, and stored at -70° C (27).
37° 42°
370 42° 43°
.. .. ~
116 -
68 -
135
..
..
43 -
One-dimensional and Two-dimensional Gel Electrophoresis One-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using a 12.5% polyacrylamide gel was performed as described by Blattler and colleagues (28). Two-dimensional gel electrophoresis using isoelectric focusing in the first dimension (60% pH 5 to 7/40% pH 6 to 8 or 60% pH 5 to 7/30% pH 3 to 10 ampholenes) followed by SDS-PAGE in the second dimension (12.5% polyacryl-
Figure 1. Heat shock induces stress protein synthesis in respiratory cells. Tracheal epithelial cells (A) and alveolar macrophages (B) were incubated at 37°, 42°, or 43° C for 45 min and radiolabeled with pSSlmethionine during treatment and subsequently for 2 h at 37° C. After labeling, the cells were harvested and the radiolabeled proteins analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Shown are the autoradiograms ofthe gels with molecular mass markers indicated at the left of each autoradiogram. The major heat shock-induced proteins of no, 90, 73, and 72 kD are indicated (in descending order) by arrowheads to the right of each autoradiogram.
B diluted with a metered flow of humidified 95 % air/5 % CO 2 to achieve a final flow rate of 1.5 liters/min into the ozone exposure chamber. All ozone containing gas mixtures flowed through Teflon tubing. The ozone concentration inside the exposure chamber was continuously monitored with an ultraviolet ozone analyzer (model no. lO03AH; Dasibi Environmental Corp., Glendale, CA). Control cells were exposed in an analogous manner to a humidified 95 % air/5 % CO 2 gas mixture at a rate of 1.5 liters/min. In addition, control cells from the same harvest were maintained in a standard humidified 37° C incubator. All ozone exposures in this study lasted 1 h. Metabolic Labeling Studies For metabolic labeling, cells to be stressed by heat shock or ozone exposure were washed twice apically and basally with Ham's F-12 medium lacking methionine (tracheal epithelial cells) or RPMI 1640 medium lacking methionine (alveolar macrophages) and labeled with (35S]methionine (Tran35Slabel, sp act = 1,200 Ci/mmol; ICN Biochemicals, Irvine, CA) 50 to 100 }-I-Ci/dish immediately prior to stress treatment. Cells stressed by other methods (or ozone-exposed cells undergoing delayed pulse labeling) were returned to growth medium after stress treatment, washed twice apically and basally with growth medium lacking methionine, and then labeled with (3sS]methionine (50 to 100 }-I-Ci/dish). Af-
Figure 2. Two-dimensional gel analysis ofthe heat shock-induced proteins in airway epithelial cells. Tracheal epithelial cells were incubated at 37° C (A) or 43 ° C (B) for 45 min and radio labeled with [35S1methionine during treatment and subsequently for 2 h at 37° C. After labeling, the cells were harvested and the radio labeled proteins analyzed by two-dimensional gel electrophoresis (acidic end of the gels is to the left). Shown are regions of the fluorographed gels illustrating the major heat shock-induced proteins. The individual stress proteins are indicated by a letter to the lower left and designated as: a, no kD; b, 90 kD (both isoforms are indicated); c, 73 kD; d, 72 kD (multiple isoforms are indicated); and e, 28 kD. The position of actin is indicated by an unlabeled arrowhead.
136
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
B
A 0 0.5 5.0 25 75 1 ~O
lARS)
lARS) 0
0.5
10
25
50 100
Figure 3. Sodium arsenite induction of stress proteins in respiratory cells. Tracheal epithelial cells (A) and alveolar macrophages (B) were incubated in 0 to 150 JlM sodium arsenite for 2 h, washed, and radiolabeled with pSS]methionine for 3 hat 3r C. After labeling, the cells were harvested and the radiolabeled proteins analyzed by SDS-PAGE. Shown are the autoradiograms of the gels with molecular mass markers indicated at the left of each autoradiogram. The major arsenite-induced proteins of 110,90, 73,72, and 32 kD are indicated (in descending order) by arrowheads to the right of each autoradiogram.
116 -
68 -
.
43 -
.
29-
amide) was performed as described by Garrels (29). The gels were fiuorographed (30) and exposed to Kodak XAR 5 film at -70 0 C. Western Immunoblotting Equal amounts of cell lysate total protein were resolved by SDS-PAGE (12.5% polyacrylamide), electrophoretically transferred to nitrocellulose membranes (Schleicher and Schuell, Keene, NH), and probed with monoclonal antibodies to hsp 70 proteins (N27 or C92) (4, 11)as previously described (31, 32). Visualization of bound antibody was done by alkaline phosphatase-mediated color development (33). Cytotoxicity Assays To screen for stress-induced cytotoxicity, LDH release was measured spectrophotometrically (LD-L Kit; Sigma) from a 0.5-rnl surface wash obtained by incubating the apical cell surface with 0.5 rnl of Ham's F-12 medium (tracheal epithelial cells) or RPMI 1640 medium (alveolar macrophages) for 5 min immediately after stress treatment. The LDH released was divided by the maximal releasable LDH (after cell lysis with 2% Triton) to obtain the %LDH release (34). "Cr release assays were subsequently performed to more definitively examine the cytotoxic effects of various stress treatments. Trachea) epithelial cells and alveolar macrophages were labeled with 5JCr (sodium (51Cr]chromate, sp act =; 400 Ci/g; New England Nuclear, Boston, MA) 50 jkCi/dish for 1.5 h then washed 3 times apically and basally with growth medium before stress treatment. After ozone or air exposures, "Cr release was measured in a gamma counter from a 0.5-rnl surface wash obtained by incubating the apical cell surface with 0.5 ml of growth medium for 5 min immediately after stress treatment. All ozone or air exposure condi-
tions were repeated in triplicate. After metabolic and thermal stress treatments, SICr release was measured in a gamma counter from the combined apical and basal media present during stress treatment. All metabolic and thermal stress measurements were repeated in duplicate. Specific SICr release (%) was calculated using the formula f(E - S)/(T S)] x 100, where E = cpm of SICr released under experimental conditions, S = cpm of 51Cr released spontaneously, and T = total releasable cpm of 51Cr after cell lysis with 2 % Triton (35). Statistical Methods Data are expressed as mean ± SEM. Differences between control and experimental data groups were assessed using Student's two-sample t test (unpaired) for two-sided alternatives in which P < 0.05 was defined as statistically significant.
Results Cultured tracheal epithelial cells and alveolar macrophages were initially treated with heat shock and sodium arsenite, classic inducers of the stress response. Heat shock treatment resulted in increased synthesis of proteins with apparent masses of72, 73, 90, and 110 kD in both epithelial cells and alveolar macrophages (Figure 1). To facilitate higher resolution of the induced proteins, the cell lysates were also analyzed by two-dimensional gel electrophoresis (isoelectric focusing in the first dimension followed by SDS-PAGE in the second dimension). Tracheal epithelial cells exhibited modest synthesis of most of the heat shock proteins (hsp) at 37° C, as observed in other cell types (Figure 2). Following heat shock, increased synthesis of the constitutive hsp 73,90,
Cohen, Palmer, Welch et al.: Stress Response of Epithelial Cells and Alveolar Macrophages
137
other cell types, hsp 72 consisted of multiple isoforms. Twodimensional gel analysis of heat shock-treated alveolar macrophages demonstrated very similar patterns of protein synthesis (data not shown). Cultured tracheal epithelial cells and alveolar macrophages were exposed to sodium arsenite in concentrations ranging from 0 to 150 JLM. At lower concentrations of arsenite, synthesis of a 32 kD protein was dramatically induced, with maximal induction at 5.0 JLM in epithelial cells and at 0.5 JLM in alveolar macrophages (Figure 3). At higher concentrations of arsenite, there was significant induction of 72, 73, 90, and 110 kD proteins, with maximal induction at 75 JLM in epithelial cells and at 10 JLM in alveolar macrophages. Analysis of alveolar macrophage cell lysates by two-dimensional gel electrophoresis demonstrated the constitutive expression of most of the stress proteins (Figure 4). Sodium arsenite treatment significantly increased the synthesis of the constitutive stress proteins and dramatically induced the synthesis of the 32 and 72 kD proteins. Western immunoblotting with monoclonal antibodies specific for hsp 72 confirmed that the 72 kD arsenite-induced protein was hsp 72. Two-dimensional gel analysis of sodium arsenitetreated tracheal epithelial cells demonstrated very similar patterns of protein synthesis (data not shown). To detect stress-induced cytotoxicity, LDH release assays were performed. Heat shock treatment resulted in mild LDH release (7 % for 43 0 C-exposed cells and 3 % for 370 C cells), whereas sodium arsenite exposure resulted in significant LDH release only after exposure to very high concentrations (26% for 150 JLM-exposed cells and 3% for nonexposed cells). To more definitively examine the cytotoxic effects of these and other stress treatments, s'Cr release assays were performed. As demonstrated in Table 1, heat shock treatment resulted in 4 to 6 % specific "Cr release in both cell types whereas exposure to sodium arsenite at concentrations ~ 100 JLM resulted to 0 to 2 % specific 51Cr release. Despite varying degress of cytotoxicity after heat shock and arsenite treatment, both epithelial cells and alveolar macrophages responded to these metabolic stresses with induction of stress protein synthesis. To determine if similar cellular responses occur after exposure to more clinically relevant stressors, we next examined the response of respiratory cells to ozone. There was significant LDH release after 0.20 ppm, and to a lesser extent, 0.05 ppm ozone exposures of both cultured tracheal epithelial cells and alveolar macrophages. SlCr release assays demonstrated a statistically significant and concentrationdependent increase in specific "Cr release after 0.05 ppm and 0.20 ppm ozone exposures in both tracheal epithelial cells (5.4% in ozone-exposed cells versus 2.1% in nonex-
Figure 4. Two-dimensional gel analysis of arsenite-induced proteins in alveolar macrophages. Alveolar macrophages were incubated in 0 fLM (A) or 5 fLM (B) of sodium arsenite for 2 h, washed, and radiolabeled with pSS]methionine for 3 h at 3r C. After labeling, the cells were harvested and the radiolabeled proteins analyzed by two-dimensional gel electrophoresis (acidic end of the gels is to the left). Shown are regions ofthe fluorographed gels illustrating the major arsenite-induced proteins. The individual stress proteins are indicated by a letter to the lower left and designated as: a, 110 kD; b, 90 kD (both isoforms are indicated); c, 73 kD; d, 72 kD (multiple isoforms are indicated); and e, 32 kD. The position of actin is indicated by an unlabeled arrowhead. Asterisk designates additional basic isoforms of the 72 kD stress protein. and 110 was observed. Increased synthesis of a 28 kD heat shock protein, not obvious on SDS-PAGE, was also detected. Synthesis of the highly stress inducible form of hsp 70, referred to here as hsp 72, appeared minimal at 3r C but was highly induced after heat shock. As observed in
TABLE I
Stress-induced airway epithelial cell and alveolar macrophage cytotoxicity Specific 51Cr Release (%) Heat Shock CO C)
Epithelial cells Alveolar macrophages
Arsenite (JlM)
Hydrogen Peroxide (M)
42
43
5
25
100
10- 5
5.6 4.4
5.5 4.3
0 1.2
0.5 2.4
0.9 2.0
0.8 0.3
Acid (pH)
Ozone (ppm)
10- 3
10- 1
6.0
5.0
0.05
0.20
0.6 2.1
18.5 5.3
0 0
0 36.9
5.4 14.6
19.2 18.4
138
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
Figure 5. Ambient concentra25 A. B. tions of ozone cause dose-depen. t $ dent respiratory cell cytotoxicity,'.:,"':"."'., _
~... 'I"... ,.,
..
Cells lining the respiratory tract form an interface between the organism and the external environment and are repeatedly exposed to physical, chemical, and metabolic stresses. We examined the response of cultured guinea pig tracheal epithelial cells and alveolar macrophages to various forms of stress, including clinically and environmentally relevant metabolic stresses such as ozone and acid exposure. Classic stress treatments such as heat shock and sodium arsenite treatment induced the synthesis of 28, 32, 72, 73, 90, and 110 kD stress proteins similar to those observed in other cell types. In contrast, no significant changes in the pattern of protein synthesis were detected after exposure to ambient concentrations of ozone, although ozone exposure caused significant cytotoxicity to both cell types. Another potent oxidant, hydrogen peroxide, similarly did not induce appreciable stress protein synthesis. However, surface acidification of tracheal epithelial cells and alveolar macrophages caused the induction of 72 and 78 kD stress proteins. While stress proteins may playa role in the response of respiratory cells to certain injuries such as hyperthermia and surface acidification, they may not be important in the defense against ozone or other forms of oxidative injury.
Cells lining the respiratory tract are situated at an air-tissue interface where the organism encounters the external environment. These cells represent the organism's "first line of defense" against inhaled or aspirated particles, organisms, and pollutants such as ozone (1, 2). Alterations in gene expression following environmental stress in respiratory epithelial cells and alveolar macrophages have not been well characterized. Most organisms defend themselves against adverse changes in their environment by altering their patterns of gene expression. One remarkably well-conserved cellular response to abrupt changes in local environmental conditions consists of the increased and selective synthesis of a small group of proteins referred to as heat shock or stress proteins. The transient synthesis of this family of proteins after heat shock treatment or after other physical, chemical, and metabolic insults has been observed with only minor variations in most cell types studied to date (3-5). Stress proteins perform important metabolic functions in the unstressed cell and appear to function in a protective or reparative manner after stress (6-11). Environmental stresses (Received in original form June 27, 1990 and in revised form November 27, 1990) Address correspondence to: Dean Sheppard, M.D., Lung Biology Center, University of California, San Francisco, Box 0854, San Francisco, CA 94143-0854. Abbreviations: heat shock proteins, hsp; lactate dehydrogenase, LDH; phosphate-buffered saline, PBS; sodium dodecyl sulfate polyacrylamide gel electrophoresis, SDS-PAGE. Am. J. Respir. Cell Mol. BioI. Vol. 5. pp. 133-143, 1991
relevant to specific cell types, such as the effects of ozone 01 acid exposure on respiratory cells, have not been studied in the context of the stress response. Studies in both animals and humans have demonstrated significant functional and structural respiratory tract abnormalities after exposure to ambient concentrations of ozone (2, 12-14). The effects of ozone on specific lung cell types have been difficult to study due to difficulties in establishing an in vitro exposure system that mimics in vivo conditions. Previous in vitro studies reported minimal or no cytotoxic effects of ambient concentrations of ozone on cultured epithelial cells and alveolar macrophages (15-17). In contrast, a recent in vivo study in human subjects described increased lactate dehydrogenase (LDH) release into bronchoalveolar lavagefluid after exposure to as little as 0.08 parts per million (ppm) of ozone, suggestive of ozone-induced cytotoxicity (18, 19). Although the basis for this difference between in vitro and in vivo findings is unclear, previous in vitro exposure systems utilized rocker platforms or tilted rotating plates that interposed a layer of medium between the cells and gaseous ozone for all or part of the exposure period. This liquid barrier may have resulted in decreased contact between the ozone gas and the cultured cells under study. Airway epithelial cells and alveolar macrophages are frequently exposed to acidic conditions due to the aspiration of acidic gastric contents into the respiratory tract during deep sleep (20). Although acid aspiration has been associated with severe lung injury and significant mortality in certain clinical situations (21,22), most people do not develop lung injury after nocturnal acid aspiration. Inhalation of acid aerosols into the respiratory tract as a result of acidic air pol-
134
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
lution has also been associated with significant respiratory morbidity and mortality (23, 24). The effects of surface acidification on gene expression in respiratory cells have not been characterized. In the present study, we examined the response of two respiratory cell types, alveolar macrophages and tracheal epithelial cells, to various forms of environmental stress. Utilizing an exposure system designed to simulate in vivo exposure conditions, we examined the patterns of protein expression in cultured tracheal epithelial cells and alveolar macrophages after heat shock, sodium arsenite, hydrogen peroxide, and acid exposure as well as after ambient concentrations of ozone.
Materials and Methods Epithelial Cell Culture Male Hartley outbred guinea pigs weighing 500 to 1,000 g were obtained from a caesarean-originated, barrier-sustained colony (Charles River Breeding Laboratories, Stoneridge, NY). After anesthetization with pentobarbital (390 mg/kg) and exsanguination via cardiac puncture, the trachea and mainstem bronchi were removed, incised longitudinally, and incubated at 37° C in Hanks' balanced salt solution containing 0.1% type XIV protease from Streptomyces griseus (Sigma Chemical Co., S1. Louis, MO) as described by Adler and co-workers (25). The airway epithelium was removed by gentle scraping with a sterile no. 10 scalpel blade, disaggregated by repeated pipetting through a sterile pasteur pipette, and pelleted at 200 x g in a bench-top centrifuge. The epithelial cells were washed twice in Ham's F-12 medium (UCSF Cell Culture Facility, San Francisco, CA) before resuspension in supplemented Ham's F-12 medium and plating on collagen-treated filter supports (Transwell-COL, 0.45-J"m pore size, 24.5-mm diameter; Costar, Cambridge, MA) at a density of 5.0 x 105 cells/well. Fresh medium was added above and below the cell cultures 24 h after plating and subsequently at 48- to 72-h intervals. Growth medium consisted of Ham's F-12 medium supplemented with transferrin (5 J.tg/rnl; GIBCO, Grand Island, NY), insulin (5 J"g/ml, bovine; GIBCO), cholera toxin (40 ng/rnl; Sigma), epidermal growth factor (15 ng/rnl, mouse SUbmaxillary gland; Sigma), hydrocortisone (360 J.tg/ml; Sigma), hypothalamus extract (12 J.tg/rnl, bovine, ECGS; Collaborative Research, Bedford, MA), retinoic acid (30 J.tg/rnl; Sigma), penicillin (100 J.tg/rnl; UCSF Cell Culture Facility), streptomycin (100 J.tg/rnl; UCSF Cell Culture Facility), fungizone (0.5 J.tg/ml; USCF Cell Culture Facility), and fetal calf serum (5% vol/vol; UCSF Cell Culture Facility). Airway epithelial cell cultures continued to proliferate for 7 to 10 days. Confluent monolayers were studied between days 10 and 15. Electron micrographs of cell monolayers demonstrated intercellular tight junctions and immunofluorescent staining demonstrated the presence of keratin, consistent with a population of differentiated epithelial cells. Cell cultures were washed and placed in unsupplemented growth medium immediately prior to stress treatments. Alveolar Macrophage Cell Culture Male Hartley outbread guinea pigs weighing 500 to 1,000 g were obtained from a caesarean-originated, barrier-sustained
colony (Charles River Breeding Laboratories). After anesthetization with pentobarbital (390 mg/kg), a tracheostomy was placed and total lung lavage using five 9-ml aliquots of 3r C Ca/Mg/endotoxin-free Hanks' balanced salt solution was performed as described by Holt (26). A typical lavage yielded 1.0 to 1.5 X 107 cells, comprised of 90 to 95 % macrophages. Granulocytes were removed using discontinuous Percoll gradient centrifugation. After being washed and resuspended in endotoxin-free RPMI 1640 medium (UCSF Cell Culture Facility), purified macrophages were plated on collagen-treated filter supports (Transwell-COL, 0.45-J.tm pore size, 24.5-mm diameter; Costar) at a density of 1.0 to 1.5 X 105 cells/well. The cell cultures were incubated at 37° C (in a humidified 95% air/5% CO 2 atmosphere) for 1 h to allow adherence. Cell cultures were washed and placed in unsupplemented growth medium immediately prior to stress treatments. Heat Shock, Arsenite, and Hydrogen Peroxide Exposures For heat shock experiments, cell culture dishes were placed in 37°, 42°, or 43° C waterbaths within incubators at the same temperatures for 45 min, returned to 37° C, then metabolically labeled with radiolabeled amino acids. Cultured cells were exposed to sodium arsenite (Sigma) 0 to 150 J.tM for 2 h or to hydrogen peroxide (Sigma) 10-5 to 10-1 M for 1 h at 37° C. After the exposure periods, cells were washed and returned to growth medium for metabolic labeling. Cells were maintained in unsupplemented growth medium during exposures and subsequent radiolabeling. Surface Acidification The apical surfaces of tracheal epithelial cells were exposed for 4 h to 0.5 rnl of 0.05 M 2-(N-morpholino)ethanesulfonic acid buffer in sodium chloride solution (final osmolarity, 300 mOsm) with the pH adjusted to 5.0,6.0, or 7.0by the addition of 1 N hydrochloric acid or 1 N sodium hydroxide (with normal growth medium in the basal compartment). Immediately after the exposure period, the cells were washed and returned to growth medium prior to metabolic labeling. For alveolar macrophage experiments, 0.05 M 2-(N-morpholino)ethanesulfonic acid buffer was prepared in RPMI medium with the pH adjusted tb 5.3, 5.8, 6.3, 6.8, or 7.20. After a 2-h exposure period, the alveolar macrophages were washed and returned to growth medium prior to metabolic labeling. Measurements of pH performed on the various media before and after exposure periods confirmed that a consistent degree of acidity was maintained throughout the exposure period. Ozone Exposure Tracheal epithelial cells and alveolar macrophages cultured on microporous filter supports were exposed to air or ozone after removal of the overlying medium to allow direct contact between the cell surface and the exposure atmosphere. Culture dishes were placed in two 27 x 22 x 17 em 'Ieflon'"lined acrylic exposure chambers in separate incubators. The exposure temperature was strictly maintained at 37° C (confirmed by thermistor probes inside both exposure chambers). An ultraviolet light source (Jelight Co., Laguna Hills, CA) was used to generate small quantities of ozone from a 95 % air/5 % CO2 gas mixture at low flow rates which was
Cohen, Palmer, Welch et al.: Stress Response of Epithelial Cells and Alveolar Macrophages
B
A
116 -
68-
43 -
ter metabolic labeling for 1.5 to 3 h, the medium was removed and the cells were washed with cold phosphatebuffered saline (PBS). The cells were then solubilized in Laemmli sample buffer, treated with nuclease, boiled at 100° C for 3 min, and stored at -70° C (27).
37° 42°
370 42° 43°
.. .. ~
116 -
68 -
135
..
..
43 -
One-dimensional and Two-dimensional Gel Electrophoresis One-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using a 12.5% polyacrylamide gel was performed as described by Blattler and colleagues (28). Two-dimensional gel electrophoresis using isoelectric focusing in the first dimension (60% pH 5 to 7/40% pH 6 to 8 or 60% pH 5 to 7/30% pH 3 to 10 ampholenes) followed by SDS-PAGE in the second dimension (12.5% polyacryl-
Figure 1. Heat shock induces stress protein synthesis in respiratory cells. Tracheal epithelial cells (A) and alveolar macrophages (B) were incubated at 37°, 42°, or 43° C for 45 min and radiolabeled with pSSlmethionine during treatment and subsequently for 2 h at 37° C. After labeling, the cells were harvested and the radiolabeled proteins analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Shown are the autoradiograms ofthe gels with molecular mass markers indicated at the left of each autoradiogram. The major heat shock-induced proteins of no, 90, 73, and 72 kD are indicated (in descending order) by arrowheads to the right of each autoradiogram.
B diluted with a metered flow of humidified 95 % air/5 % CO 2 to achieve a final flow rate of 1.5 liters/min into the ozone exposure chamber. All ozone containing gas mixtures flowed through Teflon tubing. The ozone concentration inside the exposure chamber was continuously monitored with an ultraviolet ozone analyzer (model no. lO03AH; Dasibi Environmental Corp., Glendale, CA). Control cells were exposed in an analogous manner to a humidified 95 % air/5 % CO 2 gas mixture at a rate of 1.5 liters/min. In addition, control cells from the same harvest were maintained in a standard humidified 37° C incubator. All ozone exposures in this study lasted 1 h. Metabolic Labeling Studies For metabolic labeling, cells to be stressed by heat shock or ozone exposure were washed twice apically and basally with Ham's F-12 medium lacking methionine (tracheal epithelial cells) or RPMI 1640 medium lacking methionine (alveolar macrophages) and labeled with (35S]methionine (Tran35Slabel, sp act = 1,200 Ci/mmol; ICN Biochemicals, Irvine, CA) 50 to 100 }-I-Ci/dish immediately prior to stress treatment. Cells stressed by other methods (or ozone-exposed cells undergoing delayed pulse labeling) were returned to growth medium after stress treatment, washed twice apically and basally with growth medium lacking methionine, and then labeled with (3sS]methionine (50 to 100 }-I-Ci/dish). Af-
Figure 2. Two-dimensional gel analysis ofthe heat shock-induced proteins in airway epithelial cells. Tracheal epithelial cells were incubated at 37° C (A) or 43 ° C (B) for 45 min and radio labeled with [35S1methionine during treatment and subsequently for 2 h at 37° C. After labeling, the cells were harvested and the radio labeled proteins analyzed by two-dimensional gel electrophoresis (acidic end of the gels is to the left). Shown are regions of the fluorographed gels illustrating the major heat shock-induced proteins. The individual stress proteins are indicated by a letter to the lower left and designated as: a, no kD; b, 90 kD (both isoforms are indicated); c, 73 kD; d, 72 kD (multiple isoforms are indicated); and e, 28 kD. The position of actin is indicated by an unlabeled arrowhead.
136
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
B
A 0 0.5 5.0 25 75 1 ~O
lARS)
lARS) 0
0.5
10
25
50 100
Figure 3. Sodium arsenite induction of stress proteins in respiratory cells. Tracheal epithelial cells (A) and alveolar macrophages (B) were incubated in 0 to 150 JlM sodium arsenite for 2 h, washed, and radiolabeled with pSS]methionine for 3 hat 3r C. After labeling, the cells were harvested and the radiolabeled proteins analyzed by SDS-PAGE. Shown are the autoradiograms of the gels with molecular mass markers indicated at the left of each autoradiogram. The major arsenite-induced proteins of 110,90, 73,72, and 32 kD are indicated (in descending order) by arrowheads to the right of each autoradiogram.
116 -
68 -
.
43 -
.
29-
amide) was performed as described by Garrels (29). The gels were fiuorographed (30) and exposed to Kodak XAR 5 film at -70 0 C. Western Immunoblotting Equal amounts of cell lysate total protein were resolved by SDS-PAGE (12.5% polyacrylamide), electrophoretically transferred to nitrocellulose membranes (Schleicher and Schuell, Keene, NH), and probed with monoclonal antibodies to hsp 70 proteins (N27 or C92) (4, 11)as previously described (31, 32). Visualization of bound antibody was done by alkaline phosphatase-mediated color development (33). Cytotoxicity Assays To screen for stress-induced cytotoxicity, LDH release was measured spectrophotometrically (LD-L Kit; Sigma) from a 0.5-rnl surface wash obtained by incubating the apical cell surface with 0.5 rnl of Ham's F-12 medium (tracheal epithelial cells) or RPMI 1640 medium (alveolar macrophages) for 5 min immediately after stress treatment. The LDH released was divided by the maximal releasable LDH (after cell lysis with 2% Triton) to obtain the %LDH release (34). "Cr release assays were subsequently performed to more definitively examine the cytotoxic effects of various stress treatments. Trachea) epithelial cells and alveolar macrophages were labeled with 5JCr (sodium (51Cr]chromate, sp act =; 400 Ci/g; New England Nuclear, Boston, MA) 50 jkCi/dish for 1.5 h then washed 3 times apically and basally with growth medium before stress treatment. After ozone or air exposures, "Cr release was measured in a gamma counter from a 0.5-rnl surface wash obtained by incubating the apical cell surface with 0.5 ml of growth medium for 5 min immediately after stress treatment. All ozone or air exposure condi-
tions were repeated in triplicate. After metabolic and thermal stress treatments, SICr release was measured in a gamma counter from the combined apical and basal media present during stress treatment. All metabolic and thermal stress measurements were repeated in duplicate. Specific SICr release (%) was calculated using the formula f(E - S)/(T S)] x 100, where E = cpm of SICr released under experimental conditions, S = cpm of 51Cr released spontaneously, and T = total releasable cpm of 51Cr after cell lysis with 2 % Triton (35). Statistical Methods Data are expressed as mean ± SEM. Differences between control and experimental data groups were assessed using Student's two-sample t test (unpaired) for two-sided alternatives in which P < 0.05 was defined as statistically significant.
Results Cultured tracheal epithelial cells and alveolar macrophages were initially treated with heat shock and sodium arsenite, classic inducers of the stress response. Heat shock treatment resulted in increased synthesis of proteins with apparent masses of72, 73, 90, and 110 kD in both epithelial cells and alveolar macrophages (Figure 1). To facilitate higher resolution of the induced proteins, the cell lysates were also analyzed by two-dimensional gel electrophoresis (isoelectric focusing in the first dimension followed by SDS-PAGE in the second dimension). Tracheal epithelial cells exhibited modest synthesis of most of the heat shock proteins (hsp) at 37° C, as observed in other cell types (Figure 2). Following heat shock, increased synthesis of the constitutive hsp 73,90,
Cohen, Palmer, Welch et al.: Stress Response of Epithelial Cells and Alveolar Macrophages
137
other cell types, hsp 72 consisted of multiple isoforms. Twodimensional gel analysis of heat shock-treated alveolar macrophages demonstrated very similar patterns of protein synthesis (data not shown). Cultured tracheal epithelial cells and alveolar macrophages were exposed to sodium arsenite in concentrations ranging from 0 to 150 JLM. At lower concentrations of arsenite, synthesis of a 32 kD protein was dramatically induced, with maximal induction at 5.0 JLM in epithelial cells and at 0.5 JLM in alveolar macrophages (Figure 3). At higher concentrations of arsenite, there was significant induction of 72, 73, 90, and 110 kD proteins, with maximal induction at 75 JLM in epithelial cells and at 10 JLM in alveolar macrophages. Analysis of alveolar macrophage cell lysates by two-dimensional gel electrophoresis demonstrated the constitutive expression of most of the stress proteins (Figure 4). Sodium arsenite treatment significantly increased the synthesis of the constitutive stress proteins and dramatically induced the synthesis of the 32 and 72 kD proteins. Western immunoblotting with monoclonal antibodies specific for hsp 72 confirmed that the 72 kD arsenite-induced protein was hsp 72. Two-dimensional gel analysis of sodium arsenitetreated tracheal epithelial cells demonstrated very similar patterns of protein synthesis (data not shown). To detect stress-induced cytotoxicity, LDH release assays were performed. Heat shock treatment resulted in mild LDH release (7 % for 43 0 C-exposed cells and 3 % for 370 C cells), whereas sodium arsenite exposure resulted in significant LDH release only after exposure to very high concentrations (26% for 150 JLM-exposed cells and 3% for nonexposed cells). To more definitively examine the cytotoxic effects of these and other stress treatments, s'Cr release assays were performed. As demonstrated in Table 1, heat shock treatment resulted in 4 to 6 % specific "Cr release in both cell types whereas exposure to sodium arsenite at concentrations ~ 100 JLM resulted to 0 to 2 % specific 51Cr release. Despite varying degress of cytotoxicity after heat shock and arsenite treatment, both epithelial cells and alveolar macrophages responded to these metabolic stresses with induction of stress protein synthesis. To determine if similar cellular responses occur after exposure to more clinically relevant stressors, we next examined the response of respiratory cells to ozone. There was significant LDH release after 0.20 ppm, and to a lesser extent, 0.05 ppm ozone exposures of both cultured tracheal epithelial cells and alveolar macrophages. SlCr release assays demonstrated a statistically significant and concentrationdependent increase in specific "Cr release after 0.05 ppm and 0.20 ppm ozone exposures in both tracheal epithelial cells (5.4% in ozone-exposed cells versus 2.1% in nonex-
Figure 4. Two-dimensional gel analysis of arsenite-induced proteins in alveolar macrophages. Alveolar macrophages were incubated in 0 fLM (A) or 5 fLM (B) of sodium arsenite for 2 h, washed, and radiolabeled with pSS]methionine for 3 h at 3r C. After labeling, the cells were harvested and the radiolabeled proteins analyzed by two-dimensional gel electrophoresis (acidic end of the gels is to the left). Shown are regions ofthe fluorographed gels illustrating the major arsenite-induced proteins. The individual stress proteins are indicated by a letter to the lower left and designated as: a, 110 kD; b, 90 kD (both isoforms are indicated); c, 73 kD; d, 72 kD (multiple isoforms are indicated); and e, 32 kD. The position of actin is indicated by an unlabeled arrowhead. Asterisk designates additional basic isoforms of the 72 kD stress protein. and 110 was observed. Increased synthesis of a 28 kD heat shock protein, not obvious on SDS-PAGE, was also detected. Synthesis of the highly stress inducible form of hsp 70, referred to here as hsp 72, appeared minimal at 3r C but was highly induced after heat shock. As observed in
TABLE I
Stress-induced airway epithelial cell and alveolar macrophage cytotoxicity Specific 51Cr Release (%) Heat Shock CO C)
Epithelial cells Alveolar macrophages
Arsenite (JlM)
Hydrogen Peroxide (M)
42
43
5
25
100
10- 5
5.6 4.4
5.5 4.3
0 1.2
0.5 2.4
0.9 2.0
0.8 0.3
Acid (pH)
Ozone (ppm)
10- 3
10- 1
6.0
5.0
0.05
0.20
0.6 2.1
18.5 5.3
0 0
0 36.9
5.4 14.6
19.2 18.4
138
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 5 1991
Figure 5. Ambient concentra25 A. B. tions of ozone cause dose-depen. t $ dent respiratory cell cytotoxicity
~... 'I"... ,.,
..
