INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SEVERE HYPOXIA INDUCES SIGNIFICANT lung damage. The mechanisms of hypoxic lung damage, which are similar to those observed in acute respiratory distress syndrome, are complicated and poorly understood. Hypoxia induces destructive-exudative changes within lung parenchyma, including the surfactant system-forming structures (delamellation of type II pneumocytes; edematous changes, damage to the alveolar lining layer, accumulation of alveolar macrophages) (36). Disturbances in the lung surfactant system produce changes in the lung lipid composition, which are similar in hypoxia of different genesis (15-18, 34). The main changes include the following: 1) in most cases, the overall phospholipid content was reduced; 2) a significant change in the relative distribution of the phospholipid classes was noted, including a marked decrease in phosphatidylglycerol levels and a compensatory increase in the relative amounts of the minor components (phosphatidylinositol, phosphatidylethanolamine, phosphatidylserine, sphingomyelin); 3) although the concentration of phosphatidylcholine, the most abundant lung phospholipid, is moderately reduced, two very important ratios, lysophosphatidylcholine to phosphatidylcholine and sphingomyelin to phosphatidylcholine, are dramatically increased; 4) the relative amount of palmitic acid, the major fatty acid of phosphatidylcholine, is significantly decreased under the acute hypoxic respiratory damage, whereas the relative amount of unsaturated fatty acids in this phospholipid is increased; and 5) the relative amount of dipalmitoyl phosphatidylcholine, the most abundant single surfactant component, is dramatically reduced. This leads to two main consequences: 1) disturbances in lung biomechanics and 2) acute lung edema (26, 33). An increase in the lung elasticity decreases the tidal volume and increases the respiratory frequency. These changes, in combination with lung edema, decrease the efficiency of the lung gas exchange (26).

Activation of lipid peroxidation and limitation in cellular antioxidant defense during hypoxia also contribute to lung injury (2, 20-22, 25, 32, 35, 40). It is very important to note that hypoxic activation of lipid peroxidation takes place on the background of the decrease in the antioxidant defense. The latter include the decrease in the activity of superoxide dismutase (21) and in the ratio of reduced to oxidized glutathione (8, 35). It has been suggested that free radicals might be implicated in the pathophysiology of many types of hypoxia (2). In addition to cellular membrane damage and alterations in lung lipid composition, lipid peroxidation generates biologically active substances of lipid origin (prostaglandins, leukotrienes, free fatty acids, lung inflammatory mediators, endothelium-derived constricting and relaxing factors, etc.). These in turn can produce significant damage in lungs and other organs, inducing downstream pathological changes, such as hypoxia-induced pulmonary hypertension (9, 33, 39).

Recent evidence points to the involvement of apoptosis in cellular damage induced by hypoxia (1, 6, 14, 19, 37, 41). More specifically, it appears that hypoxia-inducible factor-p53-caspase 3 apoptosis signaling pathways may be involved (14, 19, 37).

Because hypoxic lung damage clearly involves a decrease in the normal antioxidant defense, we proposed to restore the antioxidant defense during hypoxia through the use of liposomes containing -tocopherol (27). The present paper examines the influence of liposomal -tocopherol (LAT) on a chain of events that are affected by severe normobaric hypoxia. These include the breathing pattern, lung lipid composition, lipid peroxidation, eicosanoid production, antioxidant and antiapoptotic defense, and apoptosis signaling and induction.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LAT. To prepare large, multilamellar vesicles, the solution of egg lecithin and -tocopherol (3:1, wt/wt) was dried in a rotary vacuum evaporator. Hydration of the dry lipid film was accomplished by adding 140 mM NaCl and 40 mM Tris-HCl buffer (pH = 7.4) to the film and agitating. The small, unilamellar liposomes were produced by 22-kHz sonication of large, multilamellar vesicles at 20°C within 3 min. Liposomes were purified by gel filtration on a Sephadex-G50 column (Pharmacia). The mean size of liposomes was 100 nm (range 20-200 nm), and the working concentration was 10 mg/ml. Empty liposomes were prepared by using the same procedure without adding -tocopherol.

Animal model. Healthy male Wistar rats (180-200 g) were anesthetized by intraperitoneal injection of -chloralose (50 mg/kg) and urethane (500 mg/kg). Anesthetics were introduced intraperitoneally as a solution of 2 mg/ml of -chloralose and 20 mg/ml of urethane. The total amount of the introduced solution was 25 ml/kg. Previous studies (26, 32) showed that this type of anesthetization produces minimal changes in investigated parameters and stable values for 120-180 min. Acute hypoxia was produced by 120-min inhalation of the gas mixture with 6% oxygen. After tracheotomy, a three-way tube was inserted into the trachea. The inlet of the tube was connected to a rubber bag containing room air or the gas mixture with 6% oxygen. To prevent rebreathing, the gas mixture was inhaled from the bag through the tube by use of a compressor. The volumetric gas flow of the compressor was stabilized at 500 ml/min. The maximum dead space of a tube was <0.6 cm³. This method of hypoxia production was routinely used in our laboratories and showed reproducible results in different types of experiments (26, 27).

Breathing and gas exchange. Parameters of breathing and gas exchange were measured by using a pneumoresistor, a fast pressure transducer, a mass spectrometer, an analog-to-digital converter, and a computer, as previously described (26). Briefly, the pneumoresistor consisted of a three-way tube commonly connected to the airway by means of a cannula after tracheotomy. The pressure drop across the resistor was proportional to the volumetric gas flow. The pressure-flow characteristics of the installation were linear because the gas flow rate was <25 ml/s. The inlet of the capillary of the mass spectrometer was placed into the intratracheal tube. The amplitude of the gas flow and the oxygen and carbon dioxide concentrations were subsequently sampled at 10 ms, and breath-by-breath data were averaged for each rat every 15-30 s. The computer analysis included the calculation of tidal volume, respiratory frequency, oxygen consumption, and partial pressures of oxygen and carbon dioxide in the inspired and alveolar air. Arterial and mixed venous blood were sampled from the right common carotid artery and right atrium by using polyethylene catheters. Blood gases and pH were analyzed by a blood-gas analyzer. On the basis of these measurements, the lung diffusing capacity was calculated according to the method of Piiper as previously described (26). Lung diffusing capacities, parameters of breathing pattern, and gas exchange over time were measured on the same animals of each group. In some of the animals, at the end of the experiments the breathing was stopped by (D-)tubocurarine (Neopharma), and lung pressure-volume curves were registered.

Tissue hypoxia and lipid peroxidation. The concentrations of lactic and pyruvic acids in lung tissue homogenates were measured with the standard enzymatic methods (Sigma Chemical). The concentration of lactic acid and lactate-to- pyruvate ratio were used as indexes of tissue hypoxia. The level of thiobarbituric acid reactive substances was determined by the modified method of Ohkawa et al., as previously described (29, 32), and the values of lipid peroxides were expressed as nanomoles of malondialdehyde per gram of wet lung tissue weight, with tetramethoxypropane (Sigma Chemical) used as the external standard. Total lipids were extracted from the lung homogenates, and the diene conjugation absorption was evaluated by using the ultraviolet spectrum of absorption of the lipid solution in methanol-hexane (5:1) at 233 nm (10). Lipid peroxides and conjugated dienes were used as indexes of lipid peroxidation in lung tissue. To analyze the state of lung antioxidant systems, the activity of ascorbate- and NADPH-dependent lipid peroxidation were studied during the incubation of lung homogenates at 37°C for 20 min. For the induction of ascorbate-dependent lipid peroxidation 1 mM ascorbate, 1 mM ADP, and 20 µM Fe²⁺ were added to the incubation media, whereas 1 mM NADH, 1 mM ADP, and 20 µM Fe²⁺ were used to initiate NADPH-dependent lipid peroxidation. After the incubation, the concentration of thiobarbituric reactive substances (lipid peroxides) in lung homogenates was measured as described. The degree of activation, which is inversely proportional to the power of antioxidant systems, was expressed for each group (II-VI) as the ratio of the nontreated air-breathing control (group I). In addition, the expression of genes encoding two main forms of superoxide dismutase (Mn- and Cu-Zn-SOD) was measured by RT-PCR.

Lung lipid composition and eicosanoids. Prostaglandin F₂ (PGF₂), leukotrienes B₄ and C₄ (LTB₄ and LTC₄), and lipid composition were analyzed in lung homogenates. Phospholipids were separated by two-dimensional thin-layer chromatography on Siluphol plates (Kavalier) using solutions of chloroform-methanol-benzene-ammonia 28% in the ratio of 65:30:10:6 (by volume) for the first dimension and chloroform-methanol-benzene-acetone-acetic acid-water in the ratio of 70:30:10:5:4:1 (by volume) for the second dimension. The following identified phospholipids were analyzed: phosphatidylcholine, lysophosphatidylcholine, phosphatidylserine, phosphatidylinositol, phosphatidylethanolamine, phosphatidylglycerol, and sphingomyelin. The quantitative analysis of phospholipids was made on the basis of phospholipid standards from Sigma Chemical. PGF₂, LTB₄, and LTC₄ levels were determined by using radioimmunoassay kits (Amersham). In addition, the expression of genes encoding two major lung surfactant-associated proteins, B and C, in lung homogenates were measured by RT-PCR.

Gene expression. Gene expression was analyzed by RT-PCR. Total cellular RNA was isolated by using RNeasy kit (Qiagen) and QIAshredder micro spin homogenizer (Qiagen). First-strand cDNA was synthesized by Ready-To-Go You-Prime First-Strand Beads (Amersham Pharmacia Biotech) according to manufacturer instructions with 1 µg of total cellular RNA and 100 ng of random hexadeoxynucleotide primer (Amersham Pharmacia Biotech). After synthesis, the reaction mixture was immediately subjected to PCR, which was carried out by using GenAmp PCR System 2400 (Perkin-Elmer). The pairs of primers used to amplify each type of cDNA are detailed in Table 1. The PCR regimen was: 94°C/30 s, 55°C/1 min, 72°C/1 min for 30 cycles. The -actin mRNA was used as an internal standard. PCR products were separated in 4% NuSieve 3:1 Reliant agarose gels (BMA) in 1×TBE buffer (0.089 M Tris borate, 0.002 M EDTA, pH 8.3; Research Organics) by submarine electrophoresis. The gels were stained with ethidium bromide, digitally photographed, and analyzed by use of Gel Documentation System 920 (NucleoTech). Gene expression was calculated as the ratio of mean band density of analyzed RT-PCR product to that of the internal standard (-actin) coamplified with the gene of interest.

Apoptosis. The direct measurement of apoptosis intensity in lung homogenates was made by using a cell death-detection ELISA Plus kit (Roche) as previously described (28, 29, 31). The method is based on measuring the enrichment of the tissue by histone-associated DNA fragments (mono- and oligonucleosomes), which accumulate in cytoplasm during apoptosis, by use of antihistone and anti-DNA antibodies. We have previously demonstrated a very good correlation between this method and the other conventional method of apoptosis detection based on the labeling of apoptotic single- and double-stranded DNA breaks (nicks) by TUNEL (28, 31). In addition, the expression of genes encoding proteins involved in apoptosis signaling, execution, and antiapoptotic defense (caspases 9 and 3 and Bcl-2, respectively) was measured by using quantitative RT-PCR. To further characterize apoptosis, the activity of the main initiator (caspase 9) and the executor (caspase 3) of apoptosis was measured by using colorimetric protease assay kits (Pan-Vera). The assay is based on the spectrophotometric detection of the chromophore p-nitroanilide (pNA) after cleavage from the substrates X-p-NA, where X stands for amino acid sequence recognized by the specific caspase (DEVD and LEHD for caspases 3 and 9, respectively).

Morphological studies. Lung tissue for morphological studies was fixed in 10% neutral formalin and embedded into celloidin-paraffin wax. Ten-micrometer tissue sections were stained with hematoxylin-eosin. Microscopic examination of slides prepared from the lung tissue samples was conducted on all 18 samples (6 animals from each group and 3 tissue sections from each animal) taken at random from each described group. Images shown in the paper are typical for air-breathing controls, untreated hypoxic controls, and hypoxic rats treated with LAT.

Statistics. The difference between variants was considered significant if P < 0.05, determined by single-factor ANOVA. Data are expressed as means ± SD.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Breathing pattern and gas exchange. LAT as well as empty liposomes did not significantly change breathing pattern, gas exchange, and lung diffusing capacity of air-breathing rats (Fig. 1). Data from breathing-pattern analysis indicate that severe hypoxic hypoxia shifts the breathing pattern toward rapid and shallow breathing. We found a significant increase in the breathing frequency and marked decrease in the tidal volume. Injection of LAT after 30 min from the beginning of hypoxia significantly decreased breathing frequency and increased tidal volume. As a result, the same level of ventilation was achieved in a more effective way, by relatively low breathing frequency and high tidal volume. At the same time, the injection of empty liposomes did not induce significant changes in the breathing pattern. Analysis of gas exchange showed that severe hypoxia decreased oxygen consumption more than two times. During hypoxia, lung diffusion capacity increased only at the beginning of inhalation, followed by a progressive drop to the value approximately two times lower compared with control. Intratracheal application of liposomes with -tocopherol significantly improved gas exchange. It led to the marked increase in oxygen consumption and lung diffusion capacity. Empty liposomes did not significantly change lung gas exchange and oxygen diffusion.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   Breathing frequency (A), tidal volume (B), oxygen consumption (C), and diffusion capacity of lung (D) in untreated (A), empty-liposome-treated (A+L), and liposomal -tocopherol (LAT)-treated (A+LAT) air-breathing rats and untreated (H), empty liposome-treated (H+L), and LAT-treated (H+LAT) hypoxia-breathing rats. Means ± SD are shown. Number (n) of measurements in each group was A, n = 20; A+L, n = 6; A+LAT, n = 7; H, n = 22; H+L, n = 11; H+LAT, n = 21 (A, B, C) and A, H, H+LAT, n = 6; A+L, A+LAT, H+L, n = 4 (D). ×P < 0.05 compared with control; +P < 0.05 compared with hypoxia without LAT.

Lung pressure-volume curves. Empty liposomes as well as LAT did not significantly influence the lung pressure-volume curves in rats breathing room air (Fig. 2). Analysis of lung pressure-volume curves demonstrated that hypoxia significantly decreased lung elasticity. In fact, to achieve the same lung volume of 2 ml, diaphragm and lung muscles should produce a pressure drop of ~6 cmH₂O in control and ~15 cmH₂O in hypoxic lungs. Application of LAT almost completely normalized lung pressure-volume curves despite the continued hypoxic exposure. Sixty minutes after the liposomes were injected, the pressure required to increase lung volume to 2 ml decreased to ~7 cmH₂O. At the same time, treatment with empty liposomes does not significantly change the pressure-volume curve of rats breathing hypoxic gas mixture.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Typical lung pressure-volume curves in A, A+L, A+LAT, H, H+L, and H+LAT rats. Measurements were made after 120 min of breathing room air or hypoxic gas mixture.

Lung hypoxia, lipid peroxidation, and eicosanoids. Empty liposomes and liposomes containing -tocopherol did not induce statistically significant changes in lung concentration of lactate, pyruvate, lipid peroxides, conjugated dienes, free fatty acids, or eicosanoids in air-breathing rats (Table 2). They also did not significantly change lung antioxidant defense against induced NADPH- and ascorbate-dependent lipid peroxidation (Fig. 3). As expected, inhalation of gas mixture with 6% oxygen induced hypoxia in lung tissue, which led to the increase in lactate concentration and lactate-to-pyruvate ratio (for the raw data, see Table 2). Accumulated lung lactate might also have systemic, not pulmonary, origin. In addition, hypoxia activated lipid peroxidation. As a result, the concentration of lipid peroxidation products [thiobarbituric acid active substances (lipid peroxides) and conjugated dienes] was significantly elevated during hypoxic exposure. Hypoxia significantly altered lung antioxidant defense. In fact, the increase in the production of lipid peroxides after the induction of NADPH- and ascorbate-dependent lipid peroxidation was significantly higher in nontreated hypoxia-breathing rats and those treated with empty liposomes compared with all groups of air-breathing animals (Fig. 3). In contrast, treatment of hypoxia-breathing rats with LAT significantly decreased (up to 50%) the accumulation of lipid peroxides after the induction of both types of lipid peroxidation, which reflected an increase in lung antioxidant defense. We also found that hypoxia increased the concentration of free fatty acids and studied biologically active substances of lipid nature, namely PGF₂, LTB₄, and LTC₄. The application of LAT significantly countered lung tissue hypoxia and lipid peroxidation. While preventing accumulation of free fatty acids and LTC₄, LAT produced a marked increase in PGF₂. In contrast, empty liposomes did not produce significant changes in lung lactate, pyruvate, and eicosanoid lung concentrations in hypoxia-breathing rats (Table 2). They also did not significantly influence the cellular antioxidant defense in normoxia and hypoxia (Fig. 3).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   NADPH- (A) and ascorbate-dependent (B) induced lipid peroxidation in A, A+L, A+LAT, H, H+L, and H+LAT rats. Lung tissue samples were taken after 120 min of breathing room air or hypoxic gas mixture. Means ± SD from 6 independent measurements are shown. ×P < 0.05 compared with control; +P < 0.05 compared with hypoxia without LAT.

Lung phospholipid pattern. Severe hypoxia significantly changed lung phospholipid composition (Table 2). A noticeable decrease in the lung phosphatidylcholine concentration was accompanied by the marked increase in the lysophosphatidylcholine. This produced a significant decrease in the phosphatidylcholine-to-lysophosphatidylcholine ratio. Other changes in the lung phospholipid pattern included an increase in the phosphatidylserine and a decrease in the phosphatidylinositol, phosphatidylethanolamine, and phosphatidylglycerol levels. Injection of LAT significantly improved lung phospholipid pattern toward normal. It increased phosphatidylcholine and decreased lysophosphatidylcholine and sphingomyelin levels. This in turn significantly improved the phosphatidylcholine-to-lysophosphatidylcholine and phosphatidylcholine-to-sphingomyelin ratios. In addition, injection of LAT normalized phosphatidylserine levels and increased phosphatidylinositol, phosphatidylethanolamine, and phosphatidylglycerol levels. In contrast, empty liposomes did not induce statistically significant changes in lung phospholipid pattern in hypoxia-breathing rats. Empty liposomes as well as LAT did not change lung phospholipid pattern in air-breathing rats.

Gene expression. Severe hypoxic hypoxia led to the overexpression of genes encoding hypoxia-inducible factor 1 (HIF-1), both Ca²⁺-dependent and independent phospholipase-A₂ (cPLA2 and ciPLA2), heat shock protein 70 (HSP70), and caspases 3 and 9 (see Fig. 4 for the representative data and Table 3 for the mean data). At the same time, hypoxia downregulated genes encoding Mn- and Cu-Zn-cofactored superoxide dismutases (Mn-SOD, Cu-Zn-SOD), surfactant-associated proteins B and C (SFTPB, SFTPC), and Bcl-2 protein. The application of LAT limited the expression of genes encoding hypoxia-inducible factor, phospholipases, and caspases. On the other hand, it upregulated HSP70, Mn-SOD, Cu-Zn-SOD, SFTPB, SFTPC, and BCL2 genes (Fig. 4, Table 3).

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Typical gel electrophoresis images of RT-PCR products of studied genes in A, A+L, A+LAT, H, H+L, and H+LAT rats. Lung tissue samples were taken after 120 min of breathing room air or hypoxic gas mixture. For abbreviations see Table 1.

View this table: [in this window] [in a new window]   Table 3.   Gene expression (% of internal standard, -actin) in nontreated and treated with empty liposomes and LAT, air- and hypoxia-breathing rats

Apoptosis induction. Analysis of the enrichment of lung cells with mono- and oligonucleosomes showed that severe hypoxia progressively induced apoptosis in lung tissue (Fig. 5). In fact, the amount of cells undergoing apoptosis increased more than two times after 60-90 min of hypoxic exposure. The fact of hypoxic apoptosis activation is also supported by the overexpression of genes encoding the apoptosis initiator caspase 9 and apoptosis executor caspase 3 and by downregulation of the main antiapoptotic protein, Bcl-2 (Fig. 4, Table 3). This finding is also supported by the observed significant increase in the activity of both studied caspases (Fig. 6). The use of LAT significantly limited apoptosis in lung tissues. Apoptosis was 1.6 times and almost two times less pronounced on 60 and 90 min of hypoxia (30 and 60 min after the application of LAT), respectively. Injection of LAT also led to the downregulation of the expression of both caspases studied and a decrease in their activities, almost completely normalized the expression of the BCL2 gene, and inhibited apoptosis. Empty liposomes did not change any of the studied indexes of apoptosis during hypoxia. Neither empty liposomes nor LAT induced apoptosis in air-breathing rats (compare Figs. 4-6).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Apoptosis induction in lung tissues of A, A+L, A+LAT, H, H+L, and H+LAT rats. Means ± SD from 4 independent measurements in each time point are shown. Each point on the curve within a group represents different animals. ×P < 0.05 compared with control; +P < 0.05 compared with hypoxia without LAT.

View larger version (19K): [in this window] [in a new window]   Fig. 6.   Activity of caspase 9 (A) and caspase 3 (B) in A, A+L, A+LAT, H, H+L, and H+LAT rats. Lung tissue samples were taken after 120 min of breathing room air or hypoxic gas mixture. Means ± SD from 6 independent measurements are shown. ×P < 0.05 compared with control; +P < 0.05 compared with hypoxia without LAT.

Lung morphology. Morphological analysis of lungs showed that severe hypoxic hypoxia led to the development of serious lung injury (Fig. 7). Among others, the most common morphological features of hypoxic lungs included perivascular edema, lymphoid infiltration of peribronchiolar connective tissue, infiltration and edema of intra-alveolar septa, desquamation of alveolar epithelium, microatelectases, hyaline thrombi, and partial spasms of lung vessels. After the application of LAT, mentioned disturbances were almost completely eliminated.

View larger version (58K): [in this window] [in a new window]   Fig. 7.   Typical images of rat lung slices (×530) stained with hematoxylin-eosin. Top: untreated air-breathing control. Middle: untreated hypoxic lungs. Bottom: hypoxic lungs treated with LAT. Samples were taken after 120 min of breathing room air or hypoxic gas mixture. A, alveolus; B, bronchus; C, capillary; E, perivascular edema; I, lymphoid infiltration of peribronchial connective tissue; H, alveolar septum hemorrhage.

Mortality. In the group of animals breathing hypoxic gas mixture, 61% of rats (35 of 57) died during 120 min of exposure. Empty liposomes did not significantly change this mortality during hypoxia (17 of 28 rats died). LAT significantly reduced the mortality to 30% (9 of 30 rats died).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The data analysis showed that severe hypoxic hypoxia led to the lung injury that is responsible for the disturbances of lung breathing pattern and gas exchange. The hypoxic lung is characterized by the following features.

The first feature is lung tissue hypoxia and an activation of lipid peroxidation, phospholipases, and cyclo- and lipoxygenases. Although an increase in activity of phospholipase A₂ under hypoxic conditions of different genesis as well as after elevated production of eicosanoids might be considered as well established fact (33), the activation of lipid peroxidation directly under very low oxygen partial pressure could be somewhat unexpected. Evidences of oxidative stress (an increase in plasma glutathione disulfide concentration) was reported previously in hypoxic rats (8). We also previously found that severe hypoxia activated lipid peroxidation in rat brain (27, 32). later this finding was confirmed in erythrocytes, spleen, brain, liver, heart and other organs (25, 40). Recently it has been shown that generation of oxygen radicals contributes to lung damage and development of hypoxic pulmonary hypertension (20). It has been also suggested that free radicals may be implicated in the pathophysiology of acute mountain sickness because of their ability to initiate and propagate cell membrane damage (2). An activation of lipid peroxidation under conditions of decreased oxygen supply might be explained by the production of reactive oxygen species in mitochondrion because of the limitation of electron and proton transfer in the respiratory chain (32) and by the disturbances of antioxidant cellular defense induced by hypoxia. Present data showed that hypoxia downregulated the expression of genes encoding mitochondrial manganese-containing superoxide dismutase and the cytosolic copper and zinc-containing superoxide dismutase. The similar results were also obtained in alveolar type II epithelial cells and lung fibroblasts under another type of hypoxia (21). Significant decrease in reduced and increase in oxidized glutathione contents have been recently observed in the muscles and blood of hypoxia-exposed rats compared with unexposed rats (35). Glutathione reductase and glutathione S-transferase activities were also decreased in the livers and erythrocytes of hypoxia-exposed rats (35). The decrease in the antioxidant defense might also explain in part our observation of downregulation of gene encoding Bcl-2 protein, which also protects cells against lipid peroxidation (23, 41). In the present work, the direct measurement of accumulation of lipid peroxidation products after the induction of ascorbate- and NADPH-dependent lipid peroxidation in lung homogenates demonstrated the disturbances in lung antioxidant defense during hypoxia. Taken together, all these pathological changes led to the accumulation of underoxidized products of anaerobic glycolysis, lipid peroxides, free fatty acids, prostaglandins, and leukotrienes in lungs. Most of these biologically active products in turn are capable of damaging lung tissue (13, 24, 33).

In addition to generation of biologically active substances of lipid origin, the activation of phospholipase A₂, lung tissue hypoxia, and lipid peroxidation led to the changes in lung lipid composition. Hypoxia also significantly downregulated expression of genes encoding surfactant proteins B and C. It is known that the hydrophobic surfactant proteins B and C appear to be especially important in the surface-spreading characteristics of pulmonary surfactant (11, 12). These changes lead to disturbances in surfactant function and increases in lung elasticity, which in turn shift breathing pattern toward rapid and shallow breathing. The latter decreases the efficacy of ventilation and gas exchange. Therefore, the combined disturbance in surfactant function, lung biomechanics, breathing pattern, and gas exchange is the second important feature of hypoxic lungs.

The third important feature of the hypoxic lung is an activation of apoptosis in lung cells registered in the present study. Although the fact of apoptosis induction during hypoxia was confirmed in several independent studies (1, 3-5, 14, 37, 38), the exact mechanisms of this effect still remain unknown. Induction of the overexpression of the HIF-1 gene, which we found in the lung tissue, might play a trigger role in this process. It was found that, under hypoxia, HIF-1 protein binds to aryl hydrocarbon receptor nuclear translocator to activate expression of genes important for cell survival, including glycolytic enzymes, growth factors, and vasoactive peptides (6, 7). Alternatively, HIF-1 can bind to the tumor suppressor p53 and promote p53-dependent apoptosis (3, 6, 37). It was also reported that hypoxia causes an increase in p53 protein levels (14). It was shown that Fas antigen also plays a role in hypoxic apoptosis induction in cardiomyocytes (38). Our observation of downregulation of the gene encoding a key player in antiapoptotic cellular defense, Bcl-2 protein, could also promote apoptosis in hypoxic lung. A similar mechanism was reported for hypoxia in the brain (1, 4). In addition to the downregulation of the BCL2 gene, we also observed the overexpression of genes encoding caspases 3 and 9 and increases in the activity of these caspases. Comparison of data obtained in the present study with those reported for brain cells under hypoxia (19) and for cancer cells under the action of different anticancer drugs (28-30) allows us to suggest a similar mechanism of apoptosis induction. This mechanism includes an alteration of mitochondrial metabolism leading, in combination with the downregulation of Bcl-2 protein, to the cytochrome c release into the cytosol. Binding of cytochrome c to the apoptosis-activating factor Apaf-1, dATP, and procaspase 9 forms the so-called "apoptosome." Formation of apoptosome leads to the release of active caspase 9, which triggers the caspase 3 and other caspases and initiates apoptosis.

Shift in surfactant and lung tissue lipid composition limited the antihydration effect of these surface active agents. This, in combination with lung damage, led to the development of limited acute hypoxic interstitial lung edema. This acute hypoxic lung edema is the fourth important feature of hypoxic lungs. Lung edema and a decrease in major lung phospholipids disturb oxygen diffusion, which in turn increased the severity of hypoxia. This forms a vicious cycle. In 60% of animals, an activation of this vicious cycle of hypoxic lung damage led to the development of "classical" lung edema, the cessation of spontaneous breathing, and death.

As expected, intratracheal application of LAT normalized lung phospholipid composition and inhibited lipid peroxidation in lung tissues, which in turn limited lung damage and improved breathing pattern, oxygen diffusion, and lung gas exchange. However, the antihypoxic and antioxidant action of LAT cannot be explained only by the direct substitution of the deficit of phosphatidylcholine and antioxidant action of -tocopherol itself. The action of liposomes turned out to be far more complex. First, the changes in lung phospholipid pattern involved phospholipids that might be found in the liposomes only in trace amounts (phosphatidylserine, phosphatidylinositol, phosphatidylethanolamine, phosphatidylglycerol, sphingomyelin, surfactant-associated proteins B and C). Second, empty liposomes of identical lipid composition did not significantly change the lung phospholipid pattern 1 h after injection and did not improve antioxidant defense against ascorbate- and NADPH-induced lipid peroxidation. Third, liposomes with -tocopherol limited the overexpression of hypoxia-inducible factor-1 and both studied forms of phospholipase A₂, whereas empty liposomes did not produce such an effect. Fourth, LAT increased the power of cellular antioxidant and antiapoptotic defense by overexpressing genes encoding Mn- and Cu-Zn-cofactored superoxide dismutases, Bcl-2, and HSP70 proteins. This in turn limited the overexpression of the studied caspases and their activity and significantly limited apoptosis in lung cells. Empty liposomes did not demonstrate antioxidant and antiapoptotic activity during hypoxia. Finally, all these positive changes decreased the mortality from ~60% in hypoxic control to ~30% in the cohort of rats treated with LAT. At the same time, liposomes without -tocopherol did not increase the survival of animals during severe hypoxia induced by the inhalation of a gas mixture containing 6% oxygen.

It is remarkable that liposomes containing -tocopherol (as well as empty liposomes) did not induce significant changes in studied parameters in air-breathing rats. In addition, LAT did not produce marked changes in rats breathing for 2 h hypoxic gas mixtures with concentration of oxygen higher than 10% in which we did not observe extensive lung damage and significant activation of lipid peroxidation (data not shown). Taken together, the data suggest that the observed positive action of LAT might be attributed to the -tocopherol itself or to the combination of phosphatidylcholine with -tocopherol. One can also speculate that, for the effective antihypoxic action of LAT, significant lung damage is required. When we applied LAT before hypoxic exposure, the positive effect of LAT started to appear only after 30 min of hypoxic exposure, after lung damage and activation of lipid peroxidation were first observed (data not shown).

Although the mechanisms of action of liposomes containing -tocopherol require further extensive investigations, the results obtained in the present study clearly indicate that such liposomes are able to produce marked antihypoxic, antioxidant, and antiapoptotic effects, suggesting the potential use of LAT for the correction of hypoxic lung injury.