In vitro effects of hyperoxia on alveolar type II pneumocytes: inhibition of glutathione synthesis increases hyperoxic cell injury.

Exp Lung Res Downloaded from informahealthcare.com by Cornell University on 05/24/12 For personal use only.

In Vitro Effects of Hyperoxia on Alveolar Type I1 Pneumocytes: Inhibition of Glutathione Synthesis Increases Hyperoxic Cell Injury Colette Aerts, Benoit Wallaert, and Cyr Voisin

ABSTRACT: An in vitro model of alveolar epithelial oxidant injury was developed based on exposure to hyperoxia of cultured guinea pig type II pneumocytes using a biphasic cell culture system in aerobiosis. The present study investigates the roles of intracellular antioxidant enzymes and of glutathione in providing protection against hyperoxia. A 2day type II cell culture in normoxia was associated with a significant decrease in protein, catalase, and Cu-Zn SOD cell content, whereas ATP cell content, Mn-SOD, and glutathione peroxidase (GPx) activities did not change and glutathione cell content significantly increased. Exposure of type II cells to hyperoxia did not induce significant changes in cell content in protein, SOD, catalase, GPx, or glutathione cell content when cornpared to control cells (exposed to normoxia). With ATP cell content expressed as a cell injury index (CIO, type II cell injury was found to increase with increasing 0, concentrations. Indeed, a 2-day 50% 0, and 95% 0, exposure resulted in a CII of - 7.5 k 6.2% and 17.9 f 5.9%, respectively, LDH release by type II cells was not significantly increased after hyperoxic exposure. Cell i n j q efects of hyperoxia did not correlate with the endogenous antioxidant enzyme activities (SOD,Mn-SOD, catalase). In marked contrast, there was a signif2cant correlation between the CII and total glutathione content of type II cells (p < .OI). This correlation was largely due to the close relationship between CII and reduced glutathione. Hyperoxic induced cell injury (as demonstrated by CII > 0) was clearly associated with significantly lower intracellular glutathione level when compared to experiments without hyperoxia induced cell injury (CII < 0). In addition, in the presence of buthionine sulfoximine (BSO), the dbility of type 11 cells to

From the Laboratoire de Pathologie Respiratoire Expirimentale et de Pollution Atmosphbique, and IN59800 Lille, France. Address correspondence to B. Wallaert, MD, Dtipartement de Pneumologie, H6pital A . Culmette, Bld J. Leclercq, 59037 Lille Cidex France. Received 18 April 1991. Accepted 30 March 1992.

SERM C]F 90-06, Institut Pasteur de Lille,

Experimental Lung Research 18:845-861 (1992) Copyright 0 1992 by Hemisphere Publishing Corporation

845

846

C. Aerts et al.

synthetize new glutathione was severely impaired, whereas ATP cell content and cell antioxidant enzyme activities did not change. As a consequence, the reduction of intracellular glutathione significantly increased the srmeptibility of cells to hyperoxia injury (p < .05). The results strongly support the hypothesis that the regulation of glutathione levels is an important mechanism in protecting byperoxia-induced type II cell injury.


INTRODUCTION The alveolar epithelium is characterized by the presence of two major cell types: the membranous pneumocytes (type I cells) and the granular pneumocytes (type I1 cells). Type I1 cells are thought to play a minor role in gas exchange as compared to type I cells. The main functions of type I1 cells are synthesis and secretion of pulmonary surface active material. However, there is evidence that a second major function of type I1 cells, as progenitors of type I cells, is to regenerate a continuous epithelium after alveolar injury, in particular after injury by oxidant gases. Type I1 cells are more resistant to oxidative injury than are type I cells, but the reasons for this apparent resistance are as yet unclear [11. The effect of oxygen on pneumocytes I1 has been extensively studied by the exposure of animals to 0,-enriched atmosphere. The type I1 cells were examined in situ or in vitro after cell isolation and showed important differences from one animal species to another [l-41, Little is known about biological effects of hyperoxia on type I1 cells in vitro [ M I . In a recent study we demonstrated that guinea pig alveolar epithelial type I1 cells were able to survive using an in vitro cell biphasic culture 191. Therefore, we initiated this study to determine the cell injury effects of variable hyperoxia (50 or 95%) on alveolar epithelial type I1 cells. Since oxygen metabolites are normally scavenged by the antioxidant enzymes [lo], we also investigated the roles of intracellular antioxidant enzymes in providing protection against the cell injury effects of hyperoxia. In addition, since treatments associated with diminished levels of glutathione increased pulmonary oxygen toxicity [11-13], we tested the hypothesis that glutathione might have a protective effect in preventing hyperoxia injury. Given the apparent specificity of buthionine sulfoximine (BSO) in selectively inhibiting the synthesis of glutathione [14], this compound was used in a series of experiments to determine the effects of depletion in intracellular glutathione content on type I1 cell hyperoxia-induced injury.

MATERIAL AND METHODS

Cell Isolation Type I1 alveolar epithelial cells were obtained from male and female albinos Dunkin Hartley guinea pigs weighing 400-500 g. Cell isolation was performed according to the method previously described [15].


Type I1 Cells and Hyperoxic Injury

847

Guinea pigs were anesthetized by intraperitoneal injection of 100 mg of sodium pentobarbital. The chest was opened and the carotids were sectioned. Physiologic saline solution with ethylene diamine tetra-acetic acid (EDTA) 3 mM was injected at maximal manual pressure via the inferior vena cava to wash the vasculature until the liquid flowing away via the carotid was clear. Manual ventilation of the lungs through a tracheal cannula was carried out simultaneously. The lungs were removed from the thorax and transferred to a petri dish. The first lung lavage was performed with Hank’s solution buffered with hydroxyethylpiperazine ethanesulfonic acid (Hepes) 18 mM without calcium and magnesium, and with gentamycin (50 mg/rnL) and amphorericin B, to eliminate part of the free lung cells. An emulsion of fluorocarbon and albumin (0.5 mL of fluorocarbon diluted with 4.5 mL of 1% albumin Hank’s Hepes, p H 7.4) was instilled through the tracheal tube to facilitate the removal of macrophages by increasing their density. The lungs were placed, at 37”C, in a bottle containing a physiological solution with antibiotics and fungicides to allow remaining macrophages to ingest the heavy emulsion. After 15 min, the lungs were lavaged 10 times with 50 mL Hank’s Hepes to remove additional macrophages as well as fluorocarbon. The enzymatic digestion of the lungs was carried out by instillation of 5 mL Hank’s solution buffered with Hepes (pH 7.4) containing crude trypsin 1.5 mg/mL (Sigma Chemical Co., St. Louis, MO), elastase 1.5 U/mL (Sigma), and deoxyribonuclease (DNAase) 30 pg/mL (Sigma). The lungs were immersed in saline solution at 37°C. After a 10 min contact, a 5 mL solution of trypsin and elastase was again instilled; the lungs were immersed in saline solution at 37°C for 10 min, transferred to a petri dish, and filled with a solution of soya bean trypsin inhibitor (Sigma) in Hank’s Hepes (pH 7.4) solution containing DNAase 30 pg/mL and fetal calf serum l o % , to stop the action of trypsin and elastase. Glutathione 6.2 pg/mL was added to the solution. The peripheral lung tissue was removed from the bronchial tree with sharp scissors and the tissue fragments were sectioned into small pieces. The suspension of tissue was shaken vigorously for 2 min and filtered through gauze and nylon (NY 25 HC) to eliminate the type I pneumocytes. The separation of type I1 cells was obtained by velocity sedimentation in a unit gravity cell separator using Percoll. Fractions 15-18 were observed by phase contrast microscopy to eliminate fractions containing polymorphonuclear cell. After adherence on glass of fractions 15-18, the purity of type I1 pneumocytes was 91 f 0.2%. The type I1 cell fraction showed 94 f 4% trypan blue dye exclusion. Positive identification of type I1 cells was made by the following criteria: the presence of dark cytoplasmic granularity using a modified Papanicolaou stain and light microscopy [16], the lack of cellular uptake of myeloperoxidase stain using light microscopy, and the presence of numerous characteristic lamellar bodies using electron microscopy [ 151.

848

C. Aerrs

et

al.


Culture of Type I1 Pneumocytes in Aerobiosis Cell numbers were determined manually with a standard hemocytometer. Cell viability was judged by the trypan blue dye exclusion test. The cells were suspended in nutrient medium to obtain lo7 cellslml. The cell cultures were performed immediately after isolation according to the method previously described for alveolar macrophages [17]. The type I1 pneumocytes were layered on a triacetate cellulose membrane (Gelman Sciences, Inc., Ann Arbor, MI) of 0.2-pm porosity. This membrane was applied to the surface of a reservoir filled with nutrient medium in order to be saturated by capillary action. The cells were in direct contact with air without any interposition of liquid medium. The nutrient medium was Eagle's basal culture medium (BME, Gibco Cergy Pontoise, France), glutamine 2 mM (Gibco) enriched with 10% fetal calf serum, antibiotics, and 2.5 pg/mL amphotericin B. To determine the role of glutathione metabolism in the resistance of type I1 cells to hyperoxia, we also incubated monolayers of type I1 cells with BSO 1 pM/mL, an inhibitor of the enzyme gamrnaglytamylcysteine synthetase [ 141.

Oxygen Exposure Type I1 pneumocytes were exposed for 2 days to hyperoxia or to normoxia, each in triplicate, as previously described [9, 171. In all studies, cell cultures were maintained at 37 "C in humidified airtight incubation chambers (Lequeux, Paris, France), The chambers were flushed with gas mixtures (containing 50 or 95% 0,,5% COz,and balance purified N,) for 2 min at a high flow rate to purge them of air and were then closed. Control cells were exposed to purified air under the same conditions. The chambers were then sealed and incubated at 37°C for 2 days. An exhaust line from the chamber was connected to an oxygen analyzer (Oxynos 100, Leybold AG, Germany), which was used to confirm that the 0, concentration in the chamber was maintained throughout each experiment. After a 2-day exposure, dishes were harvested for ATP cell content assay and biochemical analyses.

Cell Injury Assessment The cell injury was evaluated by measurement of the ATP cell content according to the method of McElroy and Seliger [18] modified by Jakubzak and Leclerc [19]. A cell injury index (CII) was used corresponding to the percent decrease in ATP content of 0, exposed cells compared with the ATP content of control cells measured at the same time of culture in gas phase:

CII

=

ATP content of control cells - ATP content of exposed cells ATP content of control cells

x 100

Lactate dehydrogenase (LDH), a cytoplasmic enzyme, was assayed to determine if the hyperoxia-induced ATP decrease from cells was also associated


849

with the loss of this marker of cellular integrity. Culture medium was collected from cell monolayers, and LDH activity was measured by spectrophotometric analysis of N A D H oxidation according to Bergmeyer and Garrehn [20]. Correction was made for LDH activity of fetal bovine serum. Results are expressed as IU/mL.


Biochemical Analysis Superoxide dismutase (SOD) assay. The membranes in contact with nutrient medium were washed with 0.85% saline solution and the membranes with type I1 cells were put in phosphate buffer, p H 7.8 (6 x lo6 cells in 2 mL). The cellular suspension was sonicated on ice (MSE sonifier at 100 W power to 20 kHz for 30 s). After centrifugation at SOOg for 10 min at 4"C, the supernatants were removed and treated with Triton X-100 0.2% v/v. After 30 min incubation, the solutions were centrifuged at 45,OOOg for 15 min at 4°C and the supernatants were assayed for SOD using the method of McCord and Fridovich [21] modified by Crapo and McCord [22] with respect to pH (pH 10). This method is based on the capacity of SOD to inhibit the cytochrome c generated during the oxidation of xanthine catareduction mediated by 02lysed by xanthine oxidase. The manganese-dependent SOD (Mn SOD) activity was measured by adding KCN 10-.'M. The unit of enzymatic activity was defined by the 50% inhibition of reduction of cytochrome c in a final volume of 3 mL. Catalase assay. Cellular extraction was performed using the same method used for SOD assay but the cells were put in phosphate buffer pH 7 (2 x lo6 cellslml). Catalase was measured using the method of Beers and Sizer [23] involving reduction of H,O, per min at 25 "C: 0.1 mL of supernatant was added to 2.9 mL of the substrate solution (H,O,); for reproductible results the should be between 0.55 and 0.52. The time required for A,,, to decrease from 0.45 and 0.40 was noted and corresponded to the decomposition of 3.45 pM H,O, in the 3-mL solution. Glutathione peroxidase (GPx) assay. The GPx activity was assayed according to the method of Paglia and Valentine [24] modified by Holmes et al. [25]. Cells were extracted by the same method as used for the catalase assay, except the protein concentration was greater (10 x lo6 in 2 mL). The reaction was initiated by the addition of 0.1 mL of 2.2 mM H,O, to the reaction mixture. The reaction mixture contained 2.46 mL of 0.05 M phosphate buffer, p H 7, containing 5 mM EDTA, 0.1 mL of 8.4 mM reduced nicotinamide-adenine dinucleotide phosphate (NADPH) 0.1 mL of 0.15 M reduced glutathione, 20 pL of 0.562 M azide, 20 p L of (4.6 U) glutathione reductase (Sigma Chemical Co.), and 200 pL of cellular extract. The change in the optical density was read at 340 nM between 2 ' and 4'. In the absence of enzyme protein the change was always less than 5 % of the total reaction. The data were expressed

850

C. Aerts et al.

as nM NADPH oxidized to NADP by using the extinction coefficient of 6.2 x lo3 M-' cm-' at 340 nM.


Zntracellular glutathione as5uy The membranes of 1.4 x 106type I1 pneumocytes were placed in 1 mL of 0.05% w/v Triton X-100 in buffer (125 mM Na phosphate, 6.3 mM Na-EDTA, adjusted to p H 7.5). Samples of the lysates (760 pL) were acidified with 40 pL of 0.1 N HCL. The protein was precipitated by the addition of 40 pL of 50% w/v sulfosalicylic acid and removed by centrifugation, The supernatants were assayed for total glutathione (GSH and GSSG) by an enzymatic recycling procedure. Glutathione was sequentially oxidized by 5-5 '-dithiobis-2-nitrobenzoic acid (DTNB) and was reduced by NADPH in the presence of glutathione reductase at 30°C [26]. 2-Nitro-5thiobenzoic acid formation was monitored per 6 min at 412 nm by comparison of the result with the standard curve. GSH standards contained Triton X100, sulfosalicylic acid and HCL in quantities identical to the samples. For the glutathione disulfide assay, 10 p L triethanolamine and 10 pL 2-vinylpyridine were added to 200 pL supernatant of the same cellular sample. After 1-h contact, the assay was the same as the assay for total glutathione.

Protein m a y . Protein content of cells was measured by the method of Lowry et al. Activities were expressed as specific activity by milligrams of protein.

Statistical Analysis Results are expressed as means f standard error of the mean (SEM). The significance of differences between the means were determined by the MannWhitney U test for unpaired data or by a Wilcoxon paired t test. Correlations were calculated by means of the Spearman rank correlation coefficient. Analysis of variance was used when there were three groups of observations [27]. A value of p < .05 was considered significant.

RESULTS Effects of Normoxia When compared to values at the time of isolation, a 2-day culture in normoxia was associated with a significant decrease in protein, catalase, and Cu-Zn SOD cell content, whereas ATP cell content, Mn-SOD, and GPx activities were not significantly different (Table 1). Glutathione cell content significantly increased, whereas the GSH/GSH + GSSE ratio did not change.

Effects of Hyperoxia Exposure to 0, did not result in significant variation in cell protein and ATP content of 0,-exposed cells, compared with the ATP content of air-exposed cells. LDH release by alveolar type I1 cells was not significantly increased,

85 1


Table 1 Biochemical Changes Induced by a 2-day Normoxic or Hyperoxic Exposure of Type I1 Cell Cultures 2-day exposure Isolation


Protein (mg/106 cells) n = 10 LDH (IUimL culture medium) n = 10

ATP (ng/mg protein) n = 16

SOD (U/mg protein)

239

Normoxia

f 16

200* 23

-

e0.02

5175 f681

k 865

Hyperoxia

Hyperoxia

50% 0,

95% 0,

*

0.057

654 1

f20.5

193*

185* f21.5

0.063 f0.03

f0.03

7232 f 1148

& 1016

0.070

5973

11.22

7.10"

f0.56

f 0.74

6.68* f0.82

8.31** f 1.29

1.86

2.46 f0.27

2.69

f0.66

2.55 f0.38

4.63* 3~0.24

4.0* k0.19

f 1.17

n = 6

Mn-SOD (U/mg protein)

f0.35

n = 6

Cu-ZnSOD (U/mg protein)

9.36

f0.82

5.75*

n = 6

Catalase (U/mg protein) n - 6 GSH-PX (ng/mg protein)

201

81*

72 *

62 * f20

f 14

+ 30

f 19

13.1

17.4"" f4.1

ND

14.7** f 1.3

3935** f 721

5338"" f 1185

* 1272

f 338

3516** -t 708

4112** f909

4475** 1263

81.6 f3.38

85 f 3.88

75 f 6.22

74 f 10.5

f4.2

n = 4

+

GSH GSSG (ng/mg protein) n

=

GSH/GSH n

=

4920*

10

GSH (ng/mg/protein) n = 10

(W

1828 4 358

+ GSSG

1577

10

*p < .01 significantly different from values at the rime of isolarion. **p < .05 significantly different from values at the time of isolation.

*

852

C. Aerts er al.

demonstrating no significant cytotoxic effect of oxygen exposure. A close relationship was found between ATP content, decrease in 0,-exposed cells, and LDH activity release (Y = 31, p < .OOl). Sixteen experiments performed with the same cellular lot showed that effects of hyperoxia depended on 0, concentration (50 versus 95%): CII significantly increased from - 7.6 6.2% to 17.9 k 5.9% (n = 16, p < .Ol). Although there was a trend toward an increased SOD (particularly Cu-Zn SOD) content in 95% 0,-exposed cells, SOD content was not significantly increased. Decline in catalase activities was not significantly different between air- and 0,-exposed cells. Similarly, GPx activity was not significantly different between air- and 95% 0,-exposed cells. Changes in total glutathione (GSSG + GSH) cell content and GSH cell content during hyperoxic exposure were not significantly different when compared to normoxic exposure. However, as indicated in Fig. 1, glutathione changes were closely related to hyperoxic cell injury. Hyperoxic exposures that exhibited a type I1 cell injury (as demonstrated by a positive CII) were associated with significantly lower values of glutathione than those observed in experiments that did not demonstrate hyperoxia-induced cell injury (negative CII).


*

Relationship Between Hyperoxic Cell Injury and Cellular Antioxidant Defenses

To better determine the mechanisms of hyperoxia cell injury, we evaluated the relationship between the cell content antioxidant defenses and the CII of 95% 0, exposure. We did not find significant correlation between SOD con-

l""""

1 f

c

I

8000-

A

exprmenls with hyperoxic induced type II cell injury rxprirnenls wirhoul hyperoxic induced lype II cell inpry

i

l

a

c

2

+

a 6000-

F .

4000-

Y

1 LI)

c3

2000-

#

ci*

§"

4

0'

Isolalion

2 day air

2 day 50% 0 2

2 day 95% 0 2

Figure 1 Type I1 cell reduced glutathione in freshly isolated cells and after a 48-h culture in normoxia and hyperoxia (50% O2o r 95% OJ. Experiments were separated into 2 groups: @, experiments in which no cytotoxic effects of hyperoxia were observed (as demonstrated by a negative cell injury index); A, experiments in which cytotoxic effects of hyperoxia were observed (as demonstrated by a positive cell injury index). Each point represents the mean and SEM of eight experiments. Results are expressed as ng/ mg protein, §Significantly different values at the time of isolation; *significantly different experiments with hyperoxia-induced cell injury.

853


-


tent of control cells and 95% 0, CII (Y -.21, n = 18). Similarly, no correlation was found between 95% 0, CII and type I1 cell catalase activity of control cells (T = - .45, n = 18). In marked contrast, we found a strong negative correlation between total glutathione content of control cells and 95% 0, CII. This correlation was largely due to the correlation between CII and reduced glutathione cell content (Table 2). There was a significant negative correlation between the glutathione content of 50% 0,-exposed cells and both the 50% 0, CII and the 95% 0, CII (n = 10). Interestingly, we also found a close relationship between cell glutathione and cellular ATP of 0,-exposed cells (Fig. 2).

Effects of Buthionine Sulfoximine on Hyperoxic Cell Injury In the presence of BSO, the ability of type I1 cells to synthetize new glutathione to replace that which is lost during normal cellular metabolism was severely impaired (Table 3). In contrast, type I1 cell antioxidant enzyme activities and cell ATP content were not influenced by BSO. Reduction of intracellular glutathione by incubating type I1 cells with BSO significantly increased the susceptibility of cells to hyperoxic injury: CII significantly increased from 11.4 f 15 to 67.4 f 10 (Fig. 3 , p < .05).

DISCUSSION Since pulmonary oxygen toxicity is a good model of acute oxidant injury, considerable attention has been accorded to histopathological and biological abnormalities after hyperoxic exposure. In a previous study, we showed that guinea pig type I1 pneumocytes could be cultured in aerobiosis in direct contact with the atmosphere. This in vitro model of biphasic cell culture in gas phase permitted a precise evaluation of the biological effects of gases on various lung cells, using experimental conditions that mimic those present in the alveolar spaces. In the present investigation, we demonstrated that hyperoxia had moderate effects on type 11 cells, in agreement with the fact that type I1 cells are considered to be an oxidant-resistant lung cell type [2]. Although Table 2 Correlation Between Glutathione Content of Cells and 0, CII ~

~

GSH

Control cells (n = 18) 50% exposed cells (n = 13) 95% exposed cells (n 10)

-

+-

GSH

GSSG

r

P

r

P

- .71

< .001

- .69

< .01

- .76 - .70

< .05

- .68

< .001 < .01 < .05

- .64

Note. Statistical analysis was performed using the Spearman rank correlation coefficient.

854

C. Aerts et al.

-

10000 c

.-Ca E P

0

w

6000

-

4000

-

--. 07 C

r

Y

-

0.85 ; p < 0.001

Q

t 0

1000

2000

3000

4000

5000

6000


GSH + GSSG ( ng I mg protein ) Figure 2 Positive correlation between glutathione content of 0, exposed cells and cellular ATE 0 , 9 5 % 0, exposure; 0, 50% 0,exposure. Results are expressed as ng/mg protein.

oxygen-induced cell injury varied widely from one experiment to another, it is clear that cell injury was always more severe with higher oxygen concentrations (95% versus 50%). The most important finding of our study was that the regulation of cell GSH levels is an important mechanism in protecting against hyperoxia-induced type I1 cell injury.

Table 3 Effects of Prior Buthionine Sulfoximine (BSO) Treatment on Antioxidant Activities of Alveolar Epithelial Type I1 Cells ~

~~

Buffer

+

Buffer

BSO 1 pM/mL

3075 f 730

299 f 3ga

GSSG (ng/mg protein)

888 f 101

295 f 39"

SOD (u/mg protein)

8.6 k 0.7

9.78 f 0.73

Mn SOD (u/mg protein)

4.22 f 0.96

4.46 f 0.83

Calatase ( d m g protein)

74 f 0.94

85

ATP

6063 k 717

6046 k 856

GSSG

+ GSH

(ng/mg protein)

* 5.1

(ng/mg protein) Note. All results are expressed as means 1 SEM of six experiments. GSSG + GSH, total glurathione; GSSG, oxidized glutathione; SOD, total superoxide disrnutase; Mn SOD, manganese-dependent SOD. aAntioxidant activity of BSO treated cells is significantly less (p < .05) than that of control cells (buffer) (Wilcoxon test).

855

Type I1 Cells and Hyperoxic Injury 120 1

% 80

-

X W P

z

A.

40

0-

-I

w

u Exp Lung Res Downloaded from informahealthcare.com by Cornell University on 05/24/12 For personal use only.

-40-

"80

'

4 Buffer

Buffer

+

BSO

Figure 3 Effects of pretreatment with buthionine sulfoximine (BSO) on the cytotoxic effects of hyperoxia. Type 11 cells were cultured either in buffer or in buffer with BSO 1 pMimL before exposure to hyperoxia (&day exposure). Cells were exposed to 95% 0,. Each data point represents the mean CII from triplicate dishes.

The mechanisms responsible for cell resistance to hyperoxia are not fully understood. Since exposure to hyperoxia enhances the intracellular production of reactive 0, species and causes cellular injury, it was reasonable to hypothesize that 0, sensitivity of specific cell types could be related in part to endogenous cellular antioxidant activities. Baseline levels of pulmonary antioxidant enzymes and oxidant-induced adaptative responses of these enzymes vary according to species [28-321. It is clear that there was an inherent variability in type I1 cells from different animals studied under the same conditions and that cells obtained from guinea pigs may respond to oxidant injury differently than the rat or rabbit cells [29, 301. For example, using rats made tolerant to oxygen, Crapo and Tierney demonstrated that increase in SOD activity changed in parallel with the development of oxygen tolerance [28]. However, other species that have been shown to have the same histological response did not have changes in their levels of pulmonary antioxidant enzyme activities [29]. Rabbits may become tolerant to the toxic effects of hyperoxia without a concommittant increase of CuZn and Mn SOD catalase or glutathione peroxidase activities in lung homogenates or type II cells [32]. Our results are in agreement with recent study from Panus et al., who demonstrated that in vitro normoxic and hyperoxic exposure of rabbit type I1 cells resulted not in cell proliferation but in depressed cell protein [HI. Similarly, antioxidant enzyme activities in vitro fall during exposure to normoxia or hyperoxia. In parallel with these declines there was an increase of GSH and GSSG. Previous studies also demonstrated that levels of some antioxidant enzymes fall in cultured alveolar epithelial type I1 cells [33, 341. One cannot exclude that the loss of catalase activity in vitro caused the cells to increase


856

C. Aerts et al.

synthesis of GPx since GPx activity significantly increases during in vitro cell culture. Recent studies suggest that GPx is of importance in the protection of cells against oxidant-induced injury under high oxygen concentrations [35]. Thus, increased GPx activities may act, at least in part, as an H,O, peroxidase in place of catalase. However, the fact that specific inhibition of glutathione synthesis markedly increased oxygen-induced injury strongly supports the concept that glutathione plays a crucial role in modulating type I1 cell response to hyperoxia [36]. Glutathione (GSH),the main intracellular nonprotein sulfhydryl, is among the most important endogenous antioxidants [37]. The tripeptide GSH is part of the glutathione redox cycle and is an essential compound that protects lung cells from 0, radicals or lipid peroxides. GSH responds to oxidant toxic changes both in maintaining proteins in their functional redox state and in detoxifying reactive lipid peroxides [38, 391. In the normal lung, more than 90% of total glutathione is in the reduced form. In our study, GSH to total glutathione ratio was lower than 90% in both air-exposed and 0,-exposed type I1 cells, suggesting that in vitro cell culture may be responsible, at least in part, for a decrease in the reduced form. However, this may also be caused by the different stages of isolation, since GSH to total glutathione ratio was not significantly different at the time of isolation and after a 2-day exposure. Exposure to hyperoxia stimulates lung GSH synthesis and results in increased GSH in lung and lung cells by 24-h exposure to hyperoxia [40, 411, but the mechanisms responsible for GSH levels increment are not yet known. Indeed, little is known about GSH transport mechanisms and turnover of GSH in lung tissue and lung cells. Under our experimental conditions we were able to demonstrate that increase in glutathione content of type I1 cells by exposure to hyperoxia was clearly associated with resistance of type I1 cells to hyperoxia (Fig. 1). Previous studies reported an elevation of GSH by hyperoxia in various cell types, such as endothelial cells, alveolar macrophages, neutrophils, and smooth muscle cells, but erythrocytes fail to increase GSH in response to hyperoxia [42441. Since oxygen toxicity varies according to the target cell type, one cannot exclude that metabolic response of type I1 cells to hyperoxia may differ from other cell types. The close relationship between GSH and ATP cell content is of particular interest since ATP is required for the GSH synthesis from amino acids. It is likely that alterations of cell membranes during hyperoxic exposure may be responsible for parallel changes of these two components. However, pretreatment of type I1 cells with BSO was not responsible for significant changes in ATP cell content, whereas GSH content dramatically decreased. As a consequence, oxygen cell injury was markedly increased supporting the hypothesis that GSH plays a crucial role in protection against hyperoxic injury. The role of glutathione and precursors in decreasing lung oxidative injury has been well documented [36, 45-49]. Our results are in agreement with



857

several recent studies suggesting that glutathione levels correlate with susceptibility to hyperoxic lung damage [36, 50-521. In the present study we found that total glutathione and reduced glutathione levels clearly correlated with an increased susceptibility of type I1 cells to hyperoxia. Smith et al. [51] demonstrated that fasted mice exposed to 100% oxygen have more lung damage and die sooner than do fed mice. The susceptibility of fasted mice to oxygen correlated with a reduced lung glutathione content and not with lung levels of SOD and catalase. In addition, there is convincing evidence that when GSH is experimentally lowered the toxic effects of oxidant stress are markedly increased [12, 13, 40, 52, 531. Recently, using depletion of tissue GSH with diethylmaleate, Weber and coworkers also demonstrated an increased toxicity of rats exposed to hyperbaric hyperoxia, whereas exogenous administration of GSH reversed these effects [53]. Our results strongly support the hypothesis that GSH plays an important role in the protection of guinea pigs type I1 cells against hyperoxia since depletion of normal cell GSH potentiates the cytotoxic effects of hyperoxia. Recently, there has been a great interest in exploring strategies for protection against oxidant damage [28, 36, 49, 54-56]. Preliminary investigations suggested that sulfhydryl reagents, especially GSH, may represent a safe approach to protecting the lung against oxidant injury [46-511. Cultured rat type I1 alveolar cells are more resistant to paraquat-induced injury in the presence of extracellular GSH [57]. Although cell membranes are thought to be impermeable to GSH [48, 491, previous studies reported a direct uptake of plasma GSH by various cell types, such as type I1 cells, intestinal epithelial cells, and kidney ceIls [57-591. Moreover, type I1 cells could use GSH to protect against oxidant-induced damage [57]. Plasma is a constant source of GSH and is supplied from the liver in a process under hormonal control [60]. Exogenous GSH can be transported into type I1 cells by a Na+-dependent system and can supplement endogenous synthesis of GSH. In addition, isolated perfused lung from rat has the ability to utilize extracellular GSH for glutathione conjugation [61]. Thus, the lung may contribute to the utilization of plasma GSH. O n the basis of these studies, precursors of GSH may be an effective way to increasing lower respiratory tract GSH. In this context, our results provide an additional basis to suggest that sulfhydryl reagents may be useful as a direct therapeutic agent to protect lung tissue against hyperoxic injury. Supported by INSERM (CJF 90-06) and by UniversitC de Lille 11. The authors thank Christine Merdy and Catherine Fourneau for technical assistance.

REFERENCES 1. Crapo JD, Barry BE, Foscue HA, Shelburne J: Structural and biochemical changes in rat lungs occurring during exposures to lethal and adaptive doses of oxygen. Am Rev Respir Dis 122:123-143, 1980.


858

C. Aerts

et

al.

2. Freeman BA, Mason RJ, Williams MC, Crapo JD: Antioxidant enzyme activity in alveolar type I1 cells after exposure of rats to hyperoxia. Exp Lung Res 10:203222, 1986. 3. Holm BA, Matalon S, Finkelstein JN, Notter RH: Type I1 pneumocyte change during hyperoxic lung injury and recovery. J Appl Physiol 65:2672-2678, 1988. 4. Gross NJ, Smith DM: Impaired surfactant phospholipid metabolism in hyperoxic mouse lungs. J Appl Physiol 51:1198-1203, 1981. 5. Simon LM, Raffin TA, Douglas WHJ. Effects of high oxygen exposure on bioenergetics in isolated type I1 pneumocytes. J Appl Physiol Respir Environ Physiol 47~98-103, 1979. 6. Haagsman HP, Schuurmans EA, Batenburg JJ, Van Golde LM: Phospholipid synthesis in isolated alveolar type I1 cells exposed in vitro to paraquat and hyperoxia. Biochem J 245:119-126, 1987. 7. Forman HJ, Fisher AB: Antioxidant enzymes of rat granular pneumocytes: constitutive levels and effect of hyperoxia. Lab Invest 45:l-6, 1981, 8. Panus PC, Matalon S, Freeman BA: Responses of type I1 pneumocyte antioxidant enzymes to normoxic and hyperoxic culture. In vitro Cell Dev Biol 25821-829, 1989. 9. Aerts C, Voisin C, Wallaert B: Type I1 alveolar epithelial cell in vitro culture in aerobiosis. Eur Respir J 1:738-747, 1988. 10. White CW, Repine JE: Pulmonary antioxidant defense mechanisms. Exp Lung Res 8:81-96, 1985. 11. Sun JD, Pickrell JA, Harkema JR, McLaughlin SI, Hahn FF, Henderson RF: Effects of buthionine sulfoximine on the development of ozone-induced pulmonary fibrosis. Exp Mol Pathol 49:254-266, 1988. 12. Deneke SM, Lynch BA, Fanburg BL: Effects of low protein diets or feed restriction on rat lung glutathione and oxygen toxicity. J Nutr 115:726-732, 1985. 13. Deneke SM, Lynch BA, Fanburg BL: Transient depletion of lung glutathione by diethylmaleate enhances oxygen toxicity. J Appl Physiol 58:571-574, 1985. 14. Griffith OW, Meister A: Potent and specific inhibitors of glutathione synthesis by buthionine sulfoximine (S-N-butyl homocystein sulfiximine). J Biol Chem 254~7558-7560,1979. 15. Kikkawa V, Yoneda K: The type I1 epithelial cell of the lung, I: method of isolation. Lab Invest 30:76-84, 1974. 16. Koss LG: The Papanicolaou stain. In: Koss LG, Diagnostic Cytology and Its Histopathologic Basis. Philadelphia, Lippincott, pp 1211-1222, 1979. 17. Voisin C, Aerts C, Jakubczak E, Tonne1 AB: La culture cellulaire en phase gazeuse: un nouveau modkle experimental d ’ h d e in vitro des activitis des macrophages. Bull Eur Physiopathol Respir 13:69-82, 1977. 18. McElroy WD, Seliger HH: The chemistry of light emission. Adv Enzymol 25~119-123, 1969. 19. Jakubczak E, Leclerc H: Etude critique des mkthodes d’extraction de I’ATP bacttrien en vue de son dosage par la mithode de bioluminescence. Regard Biochim 2~14-19, 1978. 20. Bergmeyer HU, Garvehn K: Methods of Enzymatic Analysis, Vol 2,3d ed., New York, Academic, p 233, 1978.



859

21. McCord J, Fridovich I: Superoxide dismutase: an enzymatic function for erythrocuprein (hemocuprein). J Biol Chern 244:6049-6052, 1969. 22. Crapo JD, McCord JM: Oxygen-induced changes in pulmonary superoxide dismutase assayed by antibody titrations. Am J Physiol 231:1196-1203, 1976. 23. Beers RF,Sizer IW:Spectrophotometric method for measuring the breakdown of hydrogen peroxide by catalase. J Biol Chem 196:133-140, 1952. 24. Paglia DE, Valentine WN: Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med 70:158-169, 1967. 25. HolmesB, Park BH, Malawista SE, Quie PG, Nelson DL, Good RA: Chronic granulomatous disease in females: a deficiency of leukocyte in glutathione peroxidase. N Engl J Med 283:217-221, 1970. 26. Griffith OW: Determination of glutathione and glutathione disulfide using glutathione reductase and 2 vinylpyridine. Anal Biochem 106:207-212, 1980, 27. Gremy F, Salmon D: Bases statistiques pour la recherche mkdical et biologique. Dunod, Paris, 1969. 28. Crapo JD, Tierney DF: Superoxide dismutase and pulmonary oxygen toxicity. Am J Physiol 226:1401-1407, 1974. 29. Frank L, Bucher JR, Roberts RJ: Oxygen toxicity in neonatal and adult animals of various species. J Appl Physiol 45699-704, 1978. 30. SladeR, Stead AG, Graham JA, Hatch GE: Comparison of lung antioxidant levels in human and laboratory animals. Am Rev Respir Dis 131:742-746, 1985. 31. Bryant CL, Jenkinson SG: Species variations in lung antioxidant enzyme activities. J Appl Physiol 63597-602, 1987. 32. Baker RR, Holm BA, Panus PC, Matalons S: Development of 0, tolerance in rabbits with no increase in antioxidant enzymes. J Appl Physiol 66:1679-1684, 1989. 33. Tanswell AK, Freeman BA: Differentiation - arrested rat fetal lung in primary monolayer cell culture, 111: antioxidant enzyme activity, Exp Lung Res 6: 149-158, 1984. 34. Kinnula VL, Chang L, Everitt JI, Crapo JD: Oxidants and antioxidants in alveolar epithelial type I1 cells: in situ, freshly isolated and culture cells. Am J Physiol (Lung Cell Mol Physiol 6) 262:L69-L77, 1982. 35. Michiels C, Rernacle J: Use of the inhibition of enzymatic antioxidant systems in order to evaluate their physiological importance. Eur J Biochem 177:435-444, 1988. 36. Shan X, Aw TY, Jones DP: Glutathione-dependent protection against oxidative injury. Pharmacol Ther 47:61-71, 1990. 37. Griffith OW, Meister A: Glutathione: interorgan translocation, turnover and metabolism. Proc Natl Acad Sci USA 765606-5616, 1979. 38. Meister A, Anderson ME: Glutathione. Am Rev Biochem 52:711-760, 1983. 39. Deneke SM, Fanburg BL: Regulation of cellular glutathione. Am J Physiol (Lung Cell Mol Physiol) l:L163-L173, 1989. 40. Kimball RE, Reddy K, Peirce TH, Schwartz W, Mustafa MG, Cross CE: Oxygen toxicity: augmentation of antioxidant defense mechanism on rat lung. Am J Physiol 230:1425-1431, 1976. 41. Jenkinson SG, Black RD, Lawrence RA: Glutathione concentration in rat lung

860

C. Aerts

42. 43.

44.


45.

46. 47.

48. 49.

50.

51.

52. 52.

53.

54.

55.

56.

57.

et

al.

bronchoalveolar lavage fluid: effects of hyperoxia. J Lab Clin Med 112:345-351, 1988. Deneke SM, Steiger V, Fanburg BL: Effects of hyperoxia on glutathione levels and glutamic acid uptake in endothelial cells. J Appl Physiol 63:1966-1971, 1987. Rister M, Wustrow C: Effect of hyperoxia on reduced glutathione in alveolar macrophages and polymorphonuclear leukocytes. Res Exp Med 185445-450, 1985. Phelps DT, Deneke SM, Baxter DF, Fanburg BL: Erythrocytes fail to induce glutathione in response to diethyl maleate or hyperoxia. Am J Physiol (Lung Cell Mol Physiol) 257:L272-L276, 1989. Gershman R, Gilber DL, Cacamise D: Effects of various substances on survival times of mice exposed to different high oxygen tensions. Am J Physiol 192:563571, 1958. Bulter JA, Patterson CE, Rhodes RA: Protective effect of cysteinamine on normobaric hyperoxia in rats. Am Rev Respir Dis 119:2961, 1979. Dunbar JR, Delucia AJ, Bryant LR: Glutathione status of isolated rat lungs: effects of nitrofurantion and paraquat perfusion with normoxic and hyperoxic ventilation. Biochem Pharmacol 33:1343-1348, 1984. Patterson CE, Bulter JA, Byrne FD, Rhodes ML: Oxidant lung injury: intervention with sulfhydryl reagents. Lung 163:23-32, 1985. Patterson CE, Rhodes RA: Protective role of sulfhydryl reagents in oxidant lung injury. Exp Lung Res 14:1005-1019, 1988. Van asbeck BS, Hoidal J, Vercelloti GM, Schwartz BA, Moldow CF, Jacob HS: Protective against lethal hyperoxia by tracheal insufflation of erythrocytes role of red cell glutathione. Science 227:756-759, 1985. Smith J, Anderson J, Shamsuddin M, Hsueh W: Effect of fasting on hyperoxic lung injury in mice: the role of glutathione. Am Rev Respir Dis 141:141-149, 1990. Reed DJ, Fariss MW: Glutathione depletion and susceptibility. Pharmacol Rev 36:25S-33S, 1984. Deneke SM, Gersshoff SN, Fanfurg BL: Potentiation of oxygen toxicity in rats by dietary protein or amino acid deficiency. J Appl Physiol 54:147-151, 1983. Weber CA, Duncan CA, Lyons MJ, Jenkinson SG: Depletion of tissue glutathione with diethyl maleate enhances hyperbaric oxygen toxicity. Am J Physiol (Lung Cell Mol Physiol) 2:L308-L312, 1990. Crapo JD, Douglas M, Sjostrom DK, Hasler GR, Drewr RT: The failure of aerosolized dismutase to modify pulmonary oxygen toxicity. Am Rev Respir Dis 115:1027-1033, 1977. Padmanabhan RV, Guidapaty R, Llieneri E, Schwartz BA, Hoidal JR: Protection against pulmonary oxygen toxicity in rats by the intratracheal administration of liposome-encapsulated superoxide dismutase or catalase. Am Rev Respir Dis 132:164-167, 1985. Buckley BJ, Tanswell AK, Freeman BA: Liposome-mediated augmentation of catalase in alveolar type I1 cells protects against H,O, injury. J Appl Physiol 63~359-367,1987. Hagent TM, Brown LA, Jones DP: Protection against paraquat-induced injury by

Type I1 Cells and Hyperosic Injury

58. 59. 60.


61.

86 1

exogenous GSH in pulmonary alveolar type I1 cells. Biochem Pharmacol 35:4537-4542, 1986. Lash LH, Hagent TM, Jones DP: Exogenous glutathione protect epithelial cells from oxidative injury. Proc Natl Acad Sci USA 83:4641-4645, 1986. Hagent TM, Awt Y, Jones DP: Glutathione uptake and protection against oxidative injury in isolated kidney cells. Kidney Int 34:74-81, 1988. Sies H, Graf P: Hepatic thiol and glutathione efflux under the influence of vasopressin phenylephrine and adrenaline. Biochem J 226:545-549, 1985. Berggren M, Dawson J, Molpeus P: Glutathione biosynthesis in the isolated perfused rat lung: utilization of extracellular glutathione. FEBS Lett 176:189-192, 1984.

Amiloride-inhibitable Na+ conductive pathways in alveolar type II pneumocytes.

Dichloroacetate Decreases Cell Health and Activates Oxidative Stress Defense Pathways in Rat Alveolar Type II Pneumocytes.

Hyperoxic exposure in humans. Effects of 50 percent oxygen on alveolar macrophage leukotriene B4 synthesis.

Micronodular hyperplasia of type II pneumocytes.

Complex effects of in vitro hyperoxia on alveolar macrophage arachidonic acid metabolism.

Rat lung alveolar type I epithelial cell injury and response to hyperoxia.

Effects of inhibition and induction of cytochrome P-450 isozymes on hyperoxic lung injury in rats.

Taurine protects against oxidant injury to rat alveolar pneumocytes.

The effect of hyperoxic exposure on antioxidant enzyme activities of alveolar type II cells in neonatal and adult rats.

Autophagy regulates hyperoxia-induced intracellular accumulation of surfactant protein C in alveolar type II cells.

Alveolar Pneumocytes Entrapping Pathologists.

Expression and function of aquaporin-1 in hyperoxia-exposed alveolar epithelial type II cells.

Effect of fasting on hyperoxic lung injury in mice. The role of glutathione.

Hypertrophic alveolar type II cells from silica-treated rats are committed to DNA synthesis in vitro.

rIL-10 enhances IL-10 signalling proteins in foetal alveolar type II cells exposed to hyperoxia.

Effects and mechanisms of alveolar type II epithelial cell apoptosis in severe pancreatitis-induced acute lung injury.

Substrate utilization for phosphatidylcholine synthesis by type II pneumocytes of neonatal rats.

Hydrogen peroxide release from alveolar macrophages and alveolar type II cells during adaptation to hyperoxia in vivo.

Hyperoxia inhibits nitric oxide treatment effects in alveolar epithelial cells via effects on L-type amino acid transporter-1.

The effect of hyperoxia on migration of alveolar macrophages in vitro.

Synergetic effect of α-lipoic acid with keratinocyte growth factor on protecting alveolar epithelial type II cells of rat fetus from hyperoxia -induced injury.

Protein phosphorylation and dephosphorylation in type II pneumocytes.

Breaking the in vitro alveolar type II cell proliferation barrier while retaining ion transport properties.

Influence of streptozotocin-induced diabetes on adenylate cyclase activity in cultured type II pneumocytes.