Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Idiopathic pulmonary fibrosis (IPF) is an interstitial lung disorder characterized by the presence of alveolar macrophages, neutrophils, and eosinophils within the alveolar structures (1). Inflammatory cells obtained by bronchoalveolar lavage (BAL) from patients with IPF spontaneously release high levels of superoxide and H₂O₂ (2). Furthermore, although the glutathione (GSH) concentrations in the normal alveolar epithelial lining fluid (ELF) are high, the levels are deficient in patients with IPF, thus creating a marked oxidant and antioxidant imbalance that is thought to enhance alveolar epithelial cell damage (3, 4).

Glutathione is an essential antioxidant tripeptide present in most eukaryotic cells (5). Because of its sulfhydryl group, GSH is a versatile molecule capable of protecting cells against toxic oxidants and xenobiotics. Export of GSH synthesized by lung cells such as the epithelial Type II cell is thought to be the major source of ELF GSH (3, 6). The mechanisms by which ELF GSH decreases in patients with IPF are unknown. It is unlikely that ELF GSH deficiency in IPF simply results from its consumption by the chronic oxidant burden present at the alveolar surface. Cigarette smokers also have a chronic alveolar oxidant burden, but in contrast to patients with IPF, their ELF GSH is increased (3). In addition, it has recently been shown that superoxide and H₂O₂ generated through redox cycling of 2,3-dimethoxy-1,4-naphthoquinone (DMNQ) increase rather than decrease lung alveolar epithelial-cell GSH, -glutamylcysteine synthetase (-GCS) activity, and the -GCS heavy-subunit (-GCShs) protein level and gene transcription (7). These observations strongly suggest that mechanisms other than oxidant-mediated GSH consumption are important in the regulation of GSH concentrations at the alveolar surface of the IPF lung.

One of the characteristics of patients with IPF is the high level of expression by alveolar epithelial cells of inflammatory cytokines, particularly transforming growth factor-₁ (TGF₁) (8). The family of TGF- includes several isoforms (TGF-_1-3), which are associated with both normal tissue repair and fibrosis (11, 12). The alveolar epithelial cells of patients with IPF express high levels of the gene encoding TGF-₁, particularly in the areas of lung adjacent to those with fibrosis (9, 10). TGF-₁ has been shown to act as a pro-oxidant molecule in endothelial cells by increasing the cellular release of H₂O₂ (13). TGF-₁ also induces a marked decrease in endothelial-cell GSH, but the two effects do not occur simultaneously, and appear to be independent of each other (14). Cellular GSH depletion may occur through several mechanisms. Acute depletion occurs in the presence of an excessive oxidant burden (17); however, the long-term effect of a chronic oxidant burden is an adaptive increase in GSH (18). Since IPF is a chronic disorder, and since glutathione disulfide (GSSG) is not increased in IPF ELF, oxidant- mediated GSH depletion is not likely to account for ELF GSH depletion in IPF. A second potential mechanism of ELF GSH depletion could be through increased GSH catabolism by the enzyme -glutamyl transpeptidase (-GT). However, in vitro studies have shown that -GT has the opposite effect on lung epithelial cells. Overexpression of lung alveolar epithelial-cell -GT results in increased GSH synthesis by enhancing epithelial-cell uptake of GSH precursor amino acids (19). Additionally, it is known that -GCS plays an essential role in regulating steady-state levels of cellular GSH. Mice treated with the potent and specific -GCS inhibitor L-buthionine-(S,R)-sulfoximine (BSO) develop severe lung-cell GSH depletion (20). In the current study, we examined the effect of TGF-₁ on the GSH synthetic pathway, and particularly on the rate-limiting enzyme -GCS, of the alveolar epithelial cell line A549.

On the basis of the clinical observations that patients with IPF have: (1) decreased GSH at their alveolar surface; and (2) markedly increased expression of TGF-₁ in their alveolar epithelial cells, we hypothesized that TGF-₁ decreases alveolar epithelial-cell GSH levels. Our studies indicate that TGF-₁ downregulates expression of the rate-limiting enzyme -GCShs in the GSH synthetic pathway of the human alveolar epithelial cell line A549. The relevance of these findings to TGF--related pathologic conditions is discussed in the following sections.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Cell Culture

The A549 pulmonary epithelial cell line was obtained from the American Type Culture Collection (CCL 185; Rockville, MD). A549 cells were grown in 100-mm Petri dishes (Falcon Labware, Becton Dickinson Inc., Lincoln Park, NJ) or in 24-well culture plates (Linbro Chemical Co., New Haven, CT) in RPMI medium supplemented with 10% fetal bovine serum (FBS) and 2 mM glutamine (complete medium) in 5% CO₂ at 37°C.

Cytotoxicity Assays

To determine the effect of TGF-₁ on A549-cell susceptibility to H₂O₂-mediated injury, a cytotoxicity assay was developed. The A549 cells were plated at 5 × 10⁴ per well in 24-well culture plates with or without TGF-₁ or BSO (200 µM) for 72 h in 5% CO₂ at 37°C. The cells were labeled with 0.05 µCi / well (8-¹⁴C)adenine (specific activity: 1.96 GBq/ mmol, 53 mCi/mmol; Amersham Life Science, Oakville, ON, Canada) (21). After three washes, ¹⁴C-labeled A549 cells were incubated in the presence or absence of different concentrations of H₂O₂ (0 to 2 mM) in 0.5 ml of Earle's balanced salt solution (EBSS) for 7 h in 5% CO₂ at 37°C. The amount of ¹⁴C released in the supernatant was then quantitated. Results are expressed as a cytotoxicity index (CI) determined with the formula: CI = 100 × (A  B)/ (C   B), where A = dpm of test sample, B = dpm of spontaneous release in EBSS alone, and C = dpm of 1% Triton-X treated cells as previously described (22).

Effect of TGF-₁ on Total Cellular Glutathione (GSH + GSSG)

To determine the effect of TGF-₁ on cellular GSH + GSSG, cells were plated at 5 × 10⁴ cells/well in 24-well culture plates in complete medium for 1 h to allow for adherence. The medium was then removed and replaced with fresh medium containing acid-activated TGF-₁ at the specified concentrations for different times. GSH + GSSG was determined as previously described (3).

Effect of TGF-₁ on Antioxidant Enzyme Activities

To study the effect of TGF-₁ on the activity of antioxidant enzymes, we grew cells in 100-mm Petri dishes with or without TGF-₁ (2 ng/ml) in 5% CO₂ at 37°C for 72 h. The cells were then washed twice with phosphate buffered saline (PBS) and incubated for 5 min with PBS containing 0.1% trypsin and 1 mM ethylenediamine tetraacetic acid (EDTA) (Sigma Chemical Co., St. Louis, MO) at 37°C. The cells were suspended in complete medium to inactivate trypsin and were then washed twice with PBS. The pellet was resuspended in 0.1 M Tris, pH 8. An aliquot of the cellular suspension was removed and diluted for counting in a hemacytometer. The remaining cells were sonicated, centrifuged at 12,000 × g for 2 min, and used for protein and enzyme assays. Catalase activity was quantitated with the method described by Abei (23). Glutathione peroxidase was assayed through the continuous monitoring of GSSG formation as described by Flohé and Gümzler (24). Briefly, the following solutions are pipetted into a semimicro cuvette: 500 µl 0.1 M phosphate buffer (pH 7), 100 µl-sample, 100 µl glutathione reductase (0.24 U), and 100 µl 10 mM GSH. The mixture is preincubated for 10 min at 37°C. Subsequently, 100 µl reduced nicotinamide adenine dinucleotide phosphate (NADPH) (1.5 mM) is added and the hydroperoxide-independent consumption of NADPH is monitored for 3 min. The overall reaction is started by adding 100 µl of prewarmed 12 mM t-butyl hydroperoxide solution, and the decrease in absorption is monitored at 340 nm for 5 min. The nonenzymatic rate is correspondingly assessed by replacing the sample by buffer. Glutathione reductase was measured by the method of Horn (25).

Effect of TGF-₁ on Glutathione Synthetic Enzymes

Glutathione synthesis is mediated by two enzymes, -GCS and glutathione synthetase (GS). The activity of each of these enzymes was measured in the cells treated as described earlier, using the method described by Nardi and coworkers (26). The method involves high-performance liquid chromatography (HPLC) with fluorometric detection to directly quantify as fluorescent derivatives the -glutamylcysteine and GSH produced by the enzymatic reactions. The incubation mixture for the assay of -GCS contained 0.1 M Tris-HCl (pH 8.2), 6 mM adenosine triphosphate (ATP), 50 mM KCl, 6 mM dithiothreitol (DTT), 20 mM MgCl₂, 3 mM L-cysteine, and 15 mM L-glutamic acid, and was incubated at 37°C for 15 min to ensure the complete reduction of thiols. The reaction was initiated by the addition of 100 µl of homogenized cell supernatant to 100 µl of reaction mixture. The GS assay was similar, with the exception that 3 mM -glutamylcysteine and 30 mM glycine were substituted for the cysteine and the L-glutamic acid, respectively. At various time intervals, 20-µl aliquots were removed and combined with 50 µl of 50 mM N-ethylmorpholine and 20 µl 1 mM monobromobimane. The mixture was incubated in the dark at room temperature for 15 min, and the reaction was stopped by the addition of 80 µl of 10% sulfosalicylic acid followed by a dilution to a final volume of 500 µl with sodium acetate (0.15 M, pH 7) before HPLC analysis. Isocratic HPLC was performed on a Beckman System Gold chromatograph (Beckman Instruments Canada, Inc., Mississauga, ON, Canada) equipped with a Shimadzu RF-551 fluorescence detector (Shimadzu Corp., Tokyo, Japan). Excitation and emission wavelengths were 370 nm and 485 nm, respectively. The column was an ultrasphere ODS 5 µm (4.6 × 25 cm; Beckman Instruments Canada Inc.). The mobile phase was 10% methanol containing 0.25% acetic acid adjusted to pH 3.8, at a flow-rate of 1.5 ml/min. The amounts of -glutamylcysteine and GSH were quantitated by comparison with -glutamylcysteine and GSH standards derivatized and analyzed as described earlier. One unit of activity was defined as the activity necessary to form 1 µmol of product per minute at 37°C (26).

Western Blot Analysis of -GCS protein

-GCS was purified from 400 ml of heparinized human blood according to the procedure described by Seelig and Meister, with the following two modifications (27): (1) The red-blood-cell supernatant was treated with 31.3 g/100 ml (rather than 3.3 g/100 ml [27]) of solid ammonium sulfate, as described by Sekura and Meister (28). (2) After elution from the DE-52 cellulose column, the fractions containing -GCS activity were pooled, concentrated in an Amicon ultrafiltration cell equipped with a YM-10 membrane (Amicon Co., Lexington, MA), and introduced into the top of a column of Sephacryl S-200 (2.6 × 110 cm; Pharmacia Fine Chemicals, Piscataway, NJ) that had been previously equilibrated with a solution of 50 mM Tris/HCl, 5 mM sodium L-glutamate, and 5 mM MgCl₂ (pH 7.5). Fractions containing -GCS activity were pooled, and all other purification procedures were done as described previously (27). During the purification procedure, -GCS activity was followed in the manner described by Seelig and Meister (29). The purified protein (5 µg) was analyzed by electrophoresis on a sodium dodecyl sulfate (SDS)-7.5% polyacrylamide gel according to Laemmli (30), and was revealed with Coomassie blue stain. Antibodies to the purified enzyme were raised in rabbits by intradermal injection of 150 µg antigen emulsified in an equal volume of Freund's complete adjuvant, followed by fortnightly intradermal injections of 50 µg antigen emulsified in Freund's incomplete adjuvant for a period of 6 wk. A549-cell extracts for Western blot analysis were prepared after 72 h incubation with and without 5 ng/ml TGF-₁. The cells were removed from plastic dishes with 0.1% trypsin (Sigma), washed three times in PBS, lysed by sonication on ice, and centrifuged, and the supernatants were assayed for total protein (31). Samples containing 900 µg protein were electrophoresed on a 7.5% SDS-polyacrylamide gel. The separated proteins were electrotransferred to a nitrocellulose membrane (Bio-Rad Laboratries Ltd., Mississauga, ON, Canada). The blot was incubated for 1 h at room temperature in Tris-buffered saline (TBS) containing 5% nonfat dry milk (NFDM) and Tween 0.1%, and then in TBS containing 5% NFDM and rabbit antihuman--GCS antiserum at a final dilution of 1:1,000 for 1 h at 37°C. The membrane was washed three times for 15 min each in TBS and Tween 0.1%, and a goat antirabbit-IgG antibody-peroxidase conjugate in TBS, 5% NFDM was added and allowed to remain for 1 h at 37°C. The membrane was washed three times for 15 min each in TBS and Tween 0.1%. Protein bands were revealed with the ECL Western blotting chemiluminescence procedure (Amersham International, Buckinghamshire, UK). The specificity of the antibody for human--GCS was verified by repeating the Western blot assay on the crude extract of human red-blood-cell lysate.

RNA Extraction and Northern Blot Analysis

A549 cells were seeded at a density of 1.5 × 10⁶ cells per 100-mm cell culture dish in RPMI with 10% FBS, and were incubated at 37°C in 5% CO₂. All cells were harvested at 72 h. The effect of TGF-₁ on -GCShs messenger RNA (mRNA) expression was determined by adding 2 ng/ml TGF-₁ at 1 h, 3 h, 8 h, or 24 h before harvesting the cells for mRNA extraction. Total cell RNA was isolated with a one-step guanidinium-phenol-chloroform extraction procedure (32). RNA was separated by electrophoresis on 1% agarose and transferred onto a hybond-N+ membrane (Amersham, Oakville, ON, Canada) for analysis. Membranes were prehybridized for 4 h in a mixture containing 120 mM Tris, 600 mM NaCl, 0.1% Na₄P₂O₇, 8 mM EDTA, 0.2% SDS, 625 µg/ml heparin, and 10% dextran sulfate at pH 7.4. Hybridization was performed overnight at 68°C in the same buffer. The human -GCShs probe was obtained from the American Type Culture Collection (ATCC; GenBank/EMBL: M90656) (33) and labeled with the multiprime DNA labeling system (Amersham) using (-³²P)dCTP (specific activity > 3,000 Ci/mmol/L; Amersham). The membrane was then washed once at room temperature for 20 min in 2× SSC and for 1 h at 68°C in 0.1% SDS, 0.1× SSC, and was rinsed at room temperature in 0.1× SSC. The membrane was exposed to X-OMAT film (Kodak, Rochester, NY) with intensifying screens at 80°C. As a control for RNA integrity, the blot was hybridized with a 1-kb Pstl cDNA probe (ATCC) of the housekeeping gene glyceraldehyde phosphate dehydrogenase (GAPDH). Signal intensity was quantitated densitometrically with a Pharmacia LKB Ultroscan XL (Pharmacia Biotech, Uppsala, Sweden). Densitometric values are expressed as the ratio of -GCShs/GAPDH densitometric quantifications.

Stability of -GCShs mRNA in the Presence of TGF-₁

To determine whether TGF-₁ accelerated the degradation of mRNA expressed by the -GCShs gene in the A549 cells, cells were seeded at a density of 1.5 × 10⁶ cells/ 100 mm and cultured in RPMI with 10% FBS under 5% CO₂ at 37°C, for 48 h, followed by the addition of 5 ng/ml TGF-₁ to half of the culture dishes for 24 h. The cells were then washed three times in PBS and incubated in the presence or absence of 8 µg/ml actinomycin D for 0 to 24 h. Cells were then harvested and total RNA extracted for Northern blot analysis as described earlier. The RNA from cells exposed to actinomycin D was hybridized to the -GCShs and GAPDH probes. After signal intensity was quantitated densitometrically, the ratio of -GCShs/GAPDH mRNA was calculated.

Nuclear Run-on Assays

A549 cells were seeded at a density of 1.5 × 10⁶ cells/100 mm and cultured in RPMI with 10% FBS under 5% CO₂ at 37°C for 48 h, followed by the addition of 5 ng/ml TGF-₁ for 0 h, 1 h, or 3 h. Cells were then washed in PBS, centrifuged, and lysed in 1 ml of lysis solution (10 mM Tris-HCl [pH 7.4], 5 mM MgCl₂, 10 mM KCl, and 0.5% NP-40) (34). After centrifugation, the nuclei were resuspended in nuclei storage buffer (50 mM Tris-HCl [pH 7.4], 5 mM MgCl₂, 40% glycerol, and 0.5 mM DTT) containing 100 U/ ml RNAguard (Pharmacia). Nascent transcripts were elongated in vitro for 30 min at 28°C, and run-on analysis was done as described previously (35). Nuclear (-³²P)-UTP-labeled RNAs were hybridized to the following linearized DNAs: human liver -GCShs complementary DNA (cDNA) cloned in Bluescript SK⁺ (ATCC, GenBank/EMBL: M90656) (33), rat GAPDH cDNA cloned in pBR322 (36), and Bluescript SK⁺ vector (Promega Corp., Madison, WI) as a nonspecific hybridization control. Signal intensity was quantitated densitometrically with a Pharmacia LKB Ultrascan XL densitometer (Pharmacia Biotech Inc.). Results are representative of three independent assays.

Statistics

Results are expressed as mean ± SE. Data were analyzed with the Student's t-test, and for data including multiple groups, with analysis of variance (ANOVA) with Fisher's PLSD post hoc test. A value of P < 0.05 was considered significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Susceptibility to H₂O₂-mediated Injury

Lung epithelial cells were injured with increasing concentrations of H₂O₂, with a cytotoxicity index of 35 ± 5% at 2 mM H₂O₂ (n = 8; Figure 1A). In the presence of 200 µM BSO, a potent inhibitor of GSH synthesis, cellular GSH + GSSG concentrations consistently decreased to less than 20% of the control value (data not shown), and the susceptibility of the A549 cells was markedly increased (cytotoxicity index of BSO-treated cells at 2 mM H₂O₂ = 63 ± 5.8%; P < 0.001 compared with control, n = 4 experiments, each in triplicate). Similarly, the addition of 2 ng/ml TGF-₁ also increased the susceptibility of the alveolar epithelial cells to H₂O₂-mediated injury at all concentrations of H₂O₂ (n = 8, P < 0.05 at all concentrations of H₂O₂ compared with control). The enhanced susceptibility of A549 cells to H₂O₂-mediated cytotoxicity was maximal at 2 ng/ml TGF-₁ (P < 0.05 compared with no TGF-₁ for all H₂O₂ concentrations; Figure 1B).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Effects of GSH depletion and TGF-1 exposure on the susceptibility of A549 cells to an H2O2 stress. (A) A549 cells were seeded at 2.5 × 104/cm2 and cultured in Ham's F12 medium alone (open circles), or in the presence of 2 ng/ml TGF-1 (closed circles) or 200 µM BSO (open squares) under 5% CO2 for 72 h at 37°C. Cytotoxicity experiments were then performed as described in the text. (B) Cells were incubated with 0 to 5 ng/ml TGF-1 for 72 h, followed by a 7-h cytotoxicity assay in the presence of 2 mM H2O2. Each experiment was performed in triplicate and repeated four times.

Antioxidant Enzyme Activities

The A549 cells were found to have readily detectable levels of activity of the major antioxidant enzymes normally involved in the detoxification of cellular H₂O₂ (i.e., glutathione reductase, glutathione peroxidase, and catalase). In the presence of 2 ng/ml TGF-₁, the level of glutathione peroxidase activity remained unchanged (95 ± 4.5% of control, P > 0.5; Table 1), whereas a modest decrease was observed in the activities of both glutathione reductase and catalase (69 ± 9.6% of control; P < 0.05, n = 3; and 69 ± 3.9% of control; P < 0.01, n = 4, each respectively, compared with no TGF-₁).

Effect of TGF-₁ on A549 GSH + GSSG Content

Incubation of the lung epithelial cells in the presence of 0 to 5 ng/ml TGF-₁ for 72 h induced a concentration-dependent depletion of cellular GSH + GSSG levels with a maximal effect observed at 5 ng/ml TGF-₁ (GSH + GSSG without TGF-₁ = 3.90 ± 0.95 nmol/mg protein, graphed as 100% control in Figure 2A; 5 ng/ml TGF-₁, GSH + GSSG = 1.03 ± 0.54 nmol/mg protein, 21.7 ± 8.7% control, n = 4 experiments in triplicate; P < 0.01). The depletion of lung epithelial-cell GSH + GSSG in the presence of TGF-₁ was completely blocked by coincubation of TGF-₁ with 4 µg/ml of a neutralizing monoclonal antibody specific to TGF-₁. The ratio of GSH to total glutathione (GSH + GSSG) was not affected by TGF-₁, and was consistently greater than 95% (data not shown). The depletion of lung epithelial-cell GSH + GSSG was observed as early as 24 h, and the effect was maximal at 72 h (24 h: GSH + GSSG = 79.3 ± 3.2% of control; 48 h: GSH + GSSG = 62.3 ± 1.3% of control; 72 h, GSH + GSSG = 35.6 ± 1.8% of control, P < 0.01 for each compared with control, n = 3 experiments performed in triplicate, Figure 2B).

View larger version (11K): [in this window] [in a new window]   Figure 2.   Modulation of A549 glutathione (GSH + GSSG) content by TGF-1. (A) A549 cells were seeded at 2.5 × 104/cm2 and cultured in Ham's F12 medium with 0 to 5 ng/ml TGF-1 in the presence (closed circles) and absence (open squares) of 4 µg/ml anti-TGF-1 neutralizing monoclonal antibody for 72 h under 5% CO2 at 37°C. The cells were then harvested, lysed, and assayed for GSH + GSSG and total protein content as described in MATERIALS AND METHODS. (B) Time course of TGF-1 (2 ng/ml) modulation of GSH + GSSG content. Results represent the mean ± SEM of four separate experiments, each performed in triplicate. *P < 0.05, **P < 0.01.

-GCS Purification and Antiserum

To determine whether TGF-₁ induces a decrease in -GCShs protein synthesis, -GCS was purified from human red blood cells. Coomassie blue staining of SDS-polyacrylamide gels loaded with purified -GCS subjected to electrophoresis under reducing conditions revealed two bands corresponding to the expected molecular weights of the heavy and light subunits of -GCS (Figure 3A). Rabbit antiserum raised against this antigen was used for Western blot analysis (Figure 3B). The bands at 74 kD in the crude lysate of human red blood cells (Lane 1) and in the purified -GCS antigen (Lane 2) correspond to the expected molecular weight of -GCShs. A second band, corresponding to the expected molecular weight of the light subunit of -GCS (-GCSls), was copurified with -GCShs, as evidenced on SDS-polyacrylamide gels subjected to electrophoresis and on Western blots of the purified antigen. For reasons we have not explored, this band was not apparent in the crude lysates of human red blood cells. Immunoabsorption of A549 lysates with rabbit antiserum linked to cyanogen bromide-activated Sepharose 4B beads resulted in a loss of -GCS activity, thus indicating that the 74-kD protein detected with the antiserum was -GCShs (data not shown).

View larger version (43K): [in this window] [in a new window]   Figure 3.   Characterization of the purified -GCS enzyme and the rabbit anti--GCS antiserum. (A) Electrophoresis on 7.5% polyacrylamide gel with DTT of -GCS purified from human red blood cells and stained with Coomasie blue. The localizations of the bands correspond to the apparent molecular weights (Mr values) for the heavy (~ 74 kD) and light (~ 27 kD) subunits of -GCS. (B) Western blot of crude human red blood cell extract (Lane 1) and -GCS purified from human red blood cells (Lane 2). The anti--GCS antiserum used for the Western blot was developed by immunizing rabbits with the purified antigen illustrated in panel A.

TGF-₁ and Glutathione Synthetic Enzymes

The A549 cells were found to have clearly detectable levels of activity of both enzymes involved in GSH synthesis (-GCS = 0.443 ± 0.083 mU/mg protein, GS = 0.584 ± 0.187 mU/mg protein). Whereas 2 ng/ml TGF-₁ had no detectable effect on the activity of GS, it markedly decreased cellular activity of -GCS, the rate-limiting step in GSH synthesis (GS = 102 ± 4.3% of control, P > 0.5; -GCS = 33.5 ± 1.5% of control, P < 0.01; n = 4, Figure 4A). Western blot analysis of untreated A549 cells revealed a band corresponding to the apparent molecular weight of -GCShs. In contrast, cells treated with TGF-₁ for 72 h showed a marked decrease in -GCShs, which is known to provide the catalytic activity of the enzyme (Figure 4B).

View larger version (17K): [in this window] [in a new window]   Figure 4.   TGF-1 modulation of GSH synthetic enzyme activities and -GCShs protein. A549 cells were grown to confluence in the absence and presence of 2 ng/ml TGF-1 for 72 h under 5% CO2 at 37°C. The cells were harvested, counted, lysed, and used for (A) assays of total protein and -GCS and glutathione synthetase activity. Results are expressed as a percentage of the enzyme activities observed in cells not treated with TGF-1, and represent the mean ± SEM of four separate experiments. (B) Western blot analysis, under reducing conditions, of -GCShs protein in the crude extracts of A549 cells treated without () and with (+) TGF-1 for 24 h, 48 h, and 72 h. One representative result of three separate experiments is shown. *P < 0.01.

Effect of TGF-₁ on -GCS Gene Expression

To determine whether the TGF-₁-mediated decrease in lung epithelial-cell -GCS activity was mediated by downregulation of -GCShs gene expression, mRNA levels for -GCShs and the housekeeping gene GAPDH were compared in the presence and absence of 2 ng/ml TGF-₁ at different times. In the presence of TGF-₁, a time-dependent decrease was observed in the level of expression of -GCShs gene expression over a 24-h period (Figure 5A). This effect was not observed with the housekeeping gene GAPDH, thus suggesting that it was not a general effect of TGF-₁ on cellular gene expression. The ratio of -GCShs to GAPDH mRNA expression significantly decreased at both 8 h and 24 h (8 h: 41 ± 14% of control, P < 0.05; 24 h: 18 ± 8% of control, P < 0.01; Figure 5B).

View larger version (48K): [in this window] [in a new window]   Figure 5.   Expression of mRNA for the genes encoding either -GCShs or GAPDH. The cells were seeded at 2.5 × 104/cm2 with or without 2 ng/ml TGF-1 for the various times shown. (A) Northern blotting of -GCShs and GAPDH mRNA from one representative experiment of five. (B) Mean ratio (± SE) of the densitometric signals of -GCShs/GAPDH from five separate experiments with Northern blot analysis. *P < 0.05, **P < 0.01 compared with time = 0 h.

Effect of Actinomycin D on -GCShs mRNA in the Presence of TGF-₁

The rate of -GCShs mRNA degradation in the A549 cells was not affected by the addition of 5 ng/ml TGF-₁. The half-life of -GCShs mRNA after actinomycin D-induced transcriptional arrest was 13.8 h in the presence of TGF-₁ and 13.6 h in control cells (Figure 6). These results suggest that the marked decrease in -GCShs mRNA observed in the presence of TGF-₁ is not the result of enhanced mRNA degradation.

View larger version (21K): [in this window] [in a new window]   Figure 6.   Post-transcriptional effect of TGF-1 on -GCShs mRNA levels. A549 cells were seeded at a density of 1.5 × 106 cells/100 mm, cultured for 48 h, and then incubated in the presence (closed circles, heavy lines) or absence (open circles, light lines) of 5 ng/ml TGF-1 for 24 h. The cells were then incubated in the presence or absence of 8 µg/ml actinomycin D for 0 to 24 h. At the specified times, the cells were harvested and total RNA was extracted for Northern blot analysis. The ratios of -GCS/ GAPDH mRNA are expressed as percents of the ratios at time = 0 h. Shown are the regression lines and the 95% confidence intervals for the means. Equations for linear regressions were: without TGF-: y = 109.0  4.35x (r = 0.98); with TGF-: y = 106.7  4.1x, (r = 0.99). Data represent one of three separate experiments.

Nuclear Run-on Assays

To determine whether TGF-₁ affected -GCShs expression at the level of transcriptional initiation, cells were treated for 1 h and 3 h with 5 ng/ml TGF-₁, and the cell nuclei were prepared for nuclear run-on assays. Equal counts of radioactively labeled transcripts were hybridized to DNA fragments immobilized on nitrocellulose filters. Signals were compared with stable GAPDH transcriptional levels. The observed downregulation was sustained at 44% after 3 h (P < 0.05) (Figure 7). Signals were specific to DNA fragments, since no hybridization was obtained with a nonspecific plasmid hybridization control (data not shown).

View larger version (48K): [in this window] [in a new window]   Figure 7.   Transcriptional effect of TGF-1 on -GCShs gene expression. A549 cells were seeded at a density of 1.5 × 106 cells/ 100 mm and cultured in RPMI with 10% FBS under 5% CO2 at 37°C for 48 h, followed by the addition of 5 ng/ml TGF-1 for 0 h, 1 h, or 3 h. Cells were then washed and lysed, and the nuclei were isolated. Nascent transcripts were elongated in vitro for 30 min at 28°C for run-on analysis. Nuclear-labeled RNAs were hybridized to the following linearized DNAs: human liver -GCShs cDNA and rat GAPDH cDNA. Signal intensity was quantitated densitometrically with a Pharmacia LKB Ultroscan XL. Results are representative of three independent assays.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Exogenous TGF-₁ enhanced the susceptibility of the human alveolar epithelial cell line A549 to H₂O₂-mediated cytotoxicity. The increase in susceptibility to oxidant injury was associated with a mild decrease in two of the major antioxidant enzymes involved in the detoxification of cellular H₂O₂, catalase and glutathione reductase, an effect similar to that described in rat hepatocytes (37). Although the decrease in activity of these enzymes was modest, TGF-₁ induced a marked depletion of lung epithelial-cell GSH at 72 h, which could be completely blocked by an anti-TGF-₁-specific antibody. The decrease in GSH + GSSG was associated with a 3-fold decrease in the activity of the rate-limiting enzyme for GSH synthesis, -GCS, whereas the second enzyme in GSH synthesis, GSH synthetase, was unaffected by treatment with TGF-₁. The decrease in activity paralleled a decrease in both -GCShs protein and mRNA expression. The half-life of -GCShs mRNA expression in A549 cells was not affected by TGF-₁. In contrast, -GCShs gene transcription was significantly lower in the presence of TGF-₁.

Glutathione synthesis requires the availability of precursor amino acids and a two-step ATP-dependent enzymatic reaction (38). The first step, catalyzed by -GCS, results in the addition of L-glutamate to L-cysteine. The second step is catalyzed by GSH synthetase and involves the addition of glycine to L--glutamyl-L-cysteine to form GSH. The -GCS-catalyzed reaction is the rate-limiting step in GSH synthesis (5). -GCS is a heterodimer (74 kD heavy subunit, 27.7 kD light subunit) in which each subunit is encoded by separate genes (39). The light subunit is linked to the heavy subunit through disulfide bond formation, probably during isolation (40). The relative expression ratio of the light to heavy subunits is highly variable in different tissues (41).

The heavy subunit of -GCS provides all the catalytic activity of the enzyme; however, its activity is significantly regulated by the light subunit (41, 42). The catalytic activity of the heavy subunit of -GCS is subject to feedback inhibition by GSH. Several observations suggest that the effect of TGF-₁ on -GCS was not simply due to inhibition of the enzyme's catalytic activity. First, the cellular levels of GSH + GSSG were never increased during incubation with TGF-₁, thus excluding the possibility of GSH-mediated feedback inhibition on -GCS. Second, Western blot analysis clearly showed that TGF-₁ induced a decrease in -GCShs protein. Moreover, our studies indicate that TGF-₁ decreased the steady-state expression of -GCShs mRNA to less than 20% of the level in untreated cells at 24 h. These observations strongly support the concept that TGF-₁-mediated depletion of lung epithelial-cell glutathione is, at least in part, related to the marked decrease in -GCShs expression.

The effect of TGF-₁ on -GCShs mRNA may be transcriptional or post-transcriptional. The half-life of -GCShs mRNA in A549 cells treated with TGF-₁ and actinomycin D was 13.8 h, a value similar to that of cells not treated with TGF-₁ at 13.6 h. These results indicate that post-transcriptional modification of -GCShs mRNA cannot account for the marked suppression of -GCShs mRNA in the presence of TGF-₁. However, the nuclear run-on assays consistently demonstrated a TGF-₁-dependent decrease in the rate of transcription of the -GCShs gene. These data indicate that TGF-₁ downregulates transcription of the gene encoding -GCShs.

The recent cloning of the 5'-flanking region of the human liver -GCShs gene has led to the identification of nucleotide sequences in the 5'-flanking region consistent with a putative antioxidant response element (ARE), as well as several promoter-selective transcription factor-1 (Sp-1) binding sites, activator protein-1 (AP-1)-like and putative AP-2 binding sites, a consensus AP-1 site, and a consensus metal-responsive element (43). The cloning data suggest that -GCShs gene transcription is highly regulated. TGF- can downregulate the expression of several mammalian genes through binding of a fos-containing protein complex to a sequence of the promoter region termed TGF- inhibitory element or TIE (44, 45). Comparison of the promoter region of the -GCShs gene with the TIE consensus sequence revealed several sequences that are similar but not identical to the TIE consensus sequence. Although the present study did not allow us to define the mechanism by which TGF-₁ regulates -GCS gene transcription, it is possible that a fos-containing protein complex binds to a sequence of the -GCShs gene promoter similar to the TIE binding site. We are currently investigating this hypothesis.

GSH levels fluctuate during the cell cycle, and GSH synthesis is normally stimulated prior to cell division (46). In addition, TGF- induces a reversible growth arrest of normal epithelial cells in the late G₁ phase of the cell cycle by regulating cyclin and cyclin-dependent kinase (CDK) activity at various levels (47). It is therefore possible that TGF--mediated suppression of -GCShs transcription is induced through changes in cyclin-CDK pathways associated with TGF- exposure. It is also possible that in addition to suppressing -GCShs transcription, TGF- accelerates -GCS degradation. However, protein degradation was not adressed in the present study.

TGF- plays an essential role in wound repair, and its overexpression in lung tissues has been associated with various fibrotic lung diseases (8, 48, 49). Alveolar epithelial cells from the lungs of patients with IPF, a severe fibrotic lung disease, express high levels of TGF-₁, particularly in areas adjacent to fibrosis (9, 10). We had previously demonstrated that the alveolar ELF of patients with IPF is markedly deficient in GSH (4). On the basis of the present study, we suggest that TGF-₁ may contribute to ELF GSH depletion in IPF by downregulating -GCShs gene transcription. It is important to note that the cell line used in the present study was a neoplastic lung alveolar epithelial-cell line, and one must be cautious in extrapolating the observations made with this cell line to a non-neoplastic lung disease such as IPF. However, it has been shown that non-neoplastic lung alveolar Type II epithelial cells secrete TGF-₁, -₂, and -₃, and that their proliferative response is regulated by TGF-_1-3 (50). These observations indicate that non-neoplastic lung Type II cells have the necessary receptors and signaling pathways to fully respond to TGF-₁. Therefore, the effects of TGF-₁ on GSH metabolism reported in this study are likely to have direct relevance to non-neoplastic alveolar Type II cells such as those present in patients with IPF.

In summary, the present study demonstrates that in vitro exposure of the lung epithelial cell line A549 to TGF-₁ induces an increase in susceptibility of these cells to H₂O₂-mediated cytotoxicity, and a marked decrease in cellular GSH synthesis. The TGF-₁-mediated reduction in GSH synthesis is associated with a decrease in both the -GCS protein and the levels of mRNA expression for the gene encoding -GCShs. TGF-₁ decreases -GCShs mRNA expression by downregulating transcription of the -GCShs gene. In this context, it is likely that TGF-₁ interactions with epithelial cells play a significant role in regulating GSH synthesis in fibrotic lung tissues. The overexpression of TFG- observed in fibrotic lung disorders may decrease lung alveolar GSH, a condition known to accelerate lung fibrosis and increase epithelial-cell susceptibility to oxidant-mediated injury (20).