Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Oxidant-mediated tissue injury and impaired vascular and bronchial function play a prominent role in pulmonary dysfunction associated with syndromes such as acute respiratory distress syndrome (ARDS) (Haddad et al., 1994; Kooy et al., 1995; Lamb et al., 1999) and ischemia-reperfusion injury (Eckenhoff et al., 1992). Under these conditions, the rates of superoxide (O₂) and nitric oxide (NO^·) released from activated inflammatory cells such as macrophages and neutrophils as well as from enzymes such as inducible NO synthase are increased, and thus, the potent oxidant peroxynitrite may be formed. There is widespread evidence that reactive oxygen species cause direct cellular damage in different models of acute lung injury. However, sparse information is available on the endogenous metabolites of such radical-mediated tissue injury. The toxicity of peroxynitrite is attributed to the rapid reaction of NO^· with O₂ (k = 2 × 10¹⁰ M¹ s¹) (Kissner et al., 1997). This kinetics implies that NO^· is capable of outcompeting the O₂ removal via superoxide dismutase, which has a second order rate constant of k = 2 × 10⁹ M¹ s¹ (Royall et al., 1995). On the other hand, this kinetics implies that every 10-fold increase in O₂ and NO^· formation will cause a hundredfold increase in peroxynitrite generation. Peroxynitrite has strong oxidizing properties toward a large variety of biological molecules. These reactions include on the one hand killing of pathogenic microorganisms as a part of the innate host defense (Darrah et al., 2000). On the other hand peroxynitrite can nitrate free or protein-associated tyrosine residues and other phenolics (Ducrocq et al., 1999). The detection of tissue nitrotyrosine residues by specific antibodies is used as a marker for peroxynitrite action (Crow and Ischiropoulos, 1996). Peroxynitrite eventually also leads to sulfhydryl oxidation (Radi et al., 1991a), lipid peroxidation (Radi et al., 1991b), as well as structural and functional alterations of surfactant proteins (Haddad et al., 1993), i.e., to adverse effects on lung function. Vasoconstriction is either indirectly initiated via a diminished release of prostacyclin (PGI₂) or NO^·, or directly via receptor-dependent vasoconstrictors such as, among others, thromboxane and endothelins (ETs). In particular, the extremely potent vasoconstrictor ET-1 and its receptors, which are abundant in mammalian lung, cause vaso- and bronchoconstriction and increase vascular permeability (Hay, 1997). Since ET-1 acts via the two different receptors ET_A and ET_B, we used the selective ET_A receptor antagonist BQ123 (Ihara et al., 1992) and ET_B receptor antagonist BQ788 (Ishikawa et al., 1994) to discern which receptor mediates the ET-1 signal in this model. The roles of thromboxane A₂/prostaglandin H₂ (TXA₂/PGH₂) and PGI₂ release were also investigated. With peroxynitrite generated intravasally from chemically released NO^· and enzymatically formed O₂ from hypoxanthine/xanthine oxidase, we investigated the consequences of exposure of the isolated rat lung to peroxynitrite on its functional integrity. Our aims were 1) to define the mode and time course of the action of peroxynitrite on lung physiology, and 2) to identify the mediator(s) that cause pulmonary dysfunction and corroborate these findings by pharmacological inhibitors.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Pentobarbital sodium (Narcoren) was from Merial GmbH, Hallbermoos, Germany; glucose was from Riedel-de Haën AG, Seelze, Germany; and HEPES was from ICN Biomedicals Inc., Cleveland, OH. DL-Lysine-mono-acetylsalicylic acid (Aspisol) was purchased from Bayer, Leverkusen, Germany. Hypoxanthine (6-hydroxypurine) and xanthine oxidase (XO) [in a suspension of 2.4 M (NH₄)₂SO₄, 1 mM sodium phosphate buffer, 1 mM EDTA, and 1 mM sodium salicylic acid, pH 7.8], 2',7'-dichlorodihydrofluorescein diacetate (DCHF), aprotinin, leupeptin, pepstatin, cycloheximide, and actinomycin D were from Sigma, Deisenhofen, Germany. Sulotroban (BM13.177) was a gift from Roche Diagnostics Inc., Mannheim, Germany. The NO donor sodium nitroprusside (SNP) was purchased from Research Biochemicals International, Natick, MA. The ET_A and ET_B receptor antagonists BQ123 and BQ788 were from Alexis Inc., Grünberg, Germany. ET-1 and PGI₂ EIA detection kit were purchased from Biotrend, Köln, Germany. The thromboxane B₂ EIA detection kit, which measures the stable secondary product of thromboxane A₂, was obtained from Cayman distributor in Massy Cedex, France.

Isolated Perfused Rat Lung. The lungs of female Wistar rats (weight 200-250 g; Harlan-Winkelmann, Borchen, Germany) were prepared after terminal i.p. anesthesia by 160 mg/kg pentobarbital sodium (Merial GmbH) and perfused as formerly described (Uhlig and Wollin, 1994a). All equipment was obtained from Hugo Sachs Electronics (March-Hugstetten, Germany). Lungs were perfused at constant hydrostatic pressure (12 cm of H₂O) through the pulmonary artery, which resulted in a flow rate of approximately 35 ml/min. As perfusion medium a Krebs-Henseleit buffer (38°C) containing 2% bovine serum albumin (fraction V; Serva, Heidelberg, Germany), 0.1% glucose (Riedel-de Haën Inc., Seelze, Germany), and 0.3% HEPES (ICN Biomedicals Inc.) was used. The total amount of recirculating buffer was 100 ml. The lungs were suspended by the trachea and ventilated by negative pressure ventilation (inspiratory pressure, 7 cm of H₂O; expiratory pressure, 2 cm of H₂O) with 80 breaths/min, resulting in a tidal volume of approximately 2 ml. Every 5 min a deep inspiratory breath (20 cm of H₂O) was performed. Artificial thorax chamber pressure was measured with a differential pressure transducer (Validyne DP 45-14), and air flow velocity with a pneumotachograph tube (Fleisch type 0000) connected to a differential pressure transducer (Validyne DP 45-15). The perfusate flow (Narcomatic RT-500) and the arterial and venous pressure (Statham P23BB) were continuously monitored. The pH of the perfusate before entering the lung was kept at 7.25 to 7.35 by automatic bubbling of the buffer with CO₂ as soon as the pH exceeded this range. A weight transducer was integrated into the chamber lid and allowed continuous assessment of lung weight. Data were recorded on a Pentium II computer using the Matlab Software package (MathWorks, Inc., Natick, MA). For lung mechanics, the data were analyzed by applying the following formula:
where P is chamber pressure, C pulmonary compliance, V_T tidal volume, and R_L airway resistance. All lung physiology parameters were normalized to time point 0, i.e., after the end of the preconditioning perfusion of 40 min.

Measurement of Perfusate Samples. Samples taken from the perfusate were stored at 20°C. Rat ET-1 was assessed by human endothelin-1 EIA detection kit detecting an amino acid sequence common to bovine, rat, canine, murine, and porcine ET-1. The relative cross reactivity of the detecting antibody to ET-1 was indicated as 100%, to ET-2 3%, to ET-3 < 0.1%, and to big ET-1(-2, -3) < 0.1%. Recovery was 99.5% (manufacturer's information). TXB₂ and PGI₂ were measured in perfusate samples according to manufacturer's instructions.

Experimental Design. To obtain a stable baseline, all lungs were perfused for 10 min before any substance was added to the recirculating buffer. To obtain formation of peroxynitrite, 1 mU/ml xanthine oxidase and 0.34 mM SNP were infused together at 0 min, whereas hypoxanthine, the substrate of the xanthine oxidase reaction, was already given at 30 min. The inhibitors were usually injected before peroxynitrite formation: 190 µM TXA₂/PGH₂ receptor antagonist BM13.177 was given at 36 min, 8 µM ET_A/ET_B receptor antagonists BQ123/BQ788, 10 µM 2',7'-dichlorofluorescein diacetate, and 640 nM actinomycin D were given at 10 min. Cycloheximide in a final concentration of 177 µM was applied either at 10 or 80 min. None of these substances alone led to any measurable changes of the parameters investigated in the concentrations used. Perfusion and ventilation were continued for another 160 min. Hypoxanthine, sodium nitroprusside, and BQ123/BQ788 were each dissolved in PBS/MilliQ. Xanthine oxidase was purchased in liquid form. The TXA₂/PGH₂ receptor antagonist BM13.177 was dissolved in 30 µl dimethyl sulfoxide and then in 5 ml of perfusion buffer. 2',7'-Dichlorodihydrofluorescein diacetate was dissolved in ethanol and stored as 10-mg/ml stock solution at 20°C.

Preparation of Lung Tissue Homogenates. At the time point 120 min the buffer was replaced by PBS/MilliQ for another 2-min perfusion to remove the albumin-containing buffer from the lung. Lung tissue was shock frozen in nitrogen. Corresponding to the lung tissue weight, 10 volumes of PBS containing antiproteases (5 µg/ml each of aprotinin, leupeptin, and pepstatin) were added. After homogenization on ice the tissue samples were centrifuged (15 min, 17,900g at 4°C). The supernatant was used for ET-1 measurement. The ET-1 amount present in untreated lungs was subtracted from the treated lungs. The results were calculated as ET-1 (pg/ml).

Detection of Peroxynitrite by 2',7'-Dichlorodihydrofluorescein Diacetate. Ten minutes before peroxynitrite generation by xanthine oxidase and SNP, the fluorescent dye DCHF (10 µM) was added to the hypoxanthine-containing buffer. At the end of perfusion, the lung tissue was homogenized as described above. The fluorescence intensity in the supernatant was measured with a fluorescence spectrometer (excitation at 485 nm and emission at 535 nm). The results were calculated as fluorescence counts per milligram of total protein.

Statistics. Data are given as mean ± S.E.M. Due to normal distribution of the variables, ANOVA was performed. P  0.05 is considered to be significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Peroxynitrite-Induced Bronchoconstriction and Vasoconstriction. About 100 min after infusion of the peroxynitrite-generating agents into rat lungs, airway (p = 0.93) and vascular (p < 0.001) resistance started to increase (Fig. 1, A and B), whereas pulmonary compliance and tidal volume decreased from 0.36 ± 0.03 to 0.16 ± 0.03 ml/cm H₂O and 1.85 ± 0.09 to 1.00 ± 0.21 ml, respectively, after 160 min. Infusion of either reactant alone did not lead to any measurable changes (Fig. 1, A and B) in comparison to control perfusions. This demonstrates that none of the agents alone is likely to be responsible for the observed changes in lung physiology initiated by peroxynitrite. Within 160 min of perfusion of the peroxynitrite-generating agents, the lung weight was increased compared with controls (HX/XO: 0.45 ± 0.20-0.77 ± 0.14 g; SNP: 0.37 ± 0.35-0.48 ± 0.43 g; HX/XO plus SNP: 0.68 ± 0.17-1.14 ± 0.24 g) without statistical significance. Therefore, this parameter was not investigated further.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   A and B, peroxynitrite generation induced broncho- and vasoconstriction. Time course of peroxynitrite (HX/XO plus SNP)-induced changes in airway (A) and vascular (B) resistance. XO (1 mU/ml) and 0.34 mM SNP were added to the perfusate at 0 min (arrow). The substrate HX (1 mM) was infused at 30 min. Airway RL/RL 0 min and vascular RV/RV 0 min resistance are normalized to 0 min. Data are means ± S.E.M, n = 3 (p = 0.93 and ***p < 0.001, respectively).

Chemical Evidence for Peroxynitrite Formation. To verify that peroxynitrite was actually generated in the perfusate by hypoxanthine/xanthine oxidase plus SNP, the following controls were carried out. 1) Oxidation of DCHF to the fluorescent dye 2'7'-dichlorofluorescein was increased after 120 min of infusion in the simultaneous presence of SNP and hypoxanthine/xanthine oxidase from 42,000 in untreated control lungs, or 46,000 and 39,000 in HX/XO and SNP-treated lungs, respectively, to 76,000 counts of 2'7'-dichlorofluorescein/mg of protein in the homogenate of peroxynitrite-treated lungs. 2) After perfusion with DCHF in the presence of peroxynitrite-generating agents, no significant change compared with controls in the lung parameters investigated was observed (data not shown). This indicates that peroxynitrite was chemically reduced by DCHF to a form that is biologically inactive.

Role of Cyclooxygenase Products. Since there is in vitro evidence that peroxynitrite induces cyclooxygenase 2 and thus stimulates prostanoid production (Goodwin et al., 1999), we examined whether prostanoids were involved in the pulmonary dysfunction by pharmacological means. Neither nonspecific inhibition of cyclooxygenases (COX 1/2) by 500 µM acetylsalicylic acid, which inhibits prostanoid formation (Uhlig et al., 1994b), nor TXA₂/PGH₂ receptor antagonism with BM13.177 had any significant effect on the peroxynitrite-induced increase in airway and vascular resistance in peroxynitrite-treated lungs (data not shown). Moreover, no significant differences in the rates of TXB₂ or PGI₂ synthesis were found in lungs exposed to peroxynitrite (1.2 ± 0.3 pg/ml min TXB₂ and 12 ± 4 pg/ml min PGI₂) compared with lungs treated only with hypoxanthine/xanthine oxidase (1.0 ± 0.3 pg/ml min TXB₂ and 14.2 ± 4 pg/ml min PGI₂). This finding makes it unlikely that prostanoids are causal mediators of the peroxynitrite-induced lung injury.

Formation and Action of Endothelins. ET-1 concentrations were determined in the supernatant of lung homogenates 120 min after infusion of hypoxanthine/xanthine oxidase and/or SNP. The data in Fig. 2 show that the ET-1 concentration in lung homogenate supernatants from hypoxanthine/xanthine oxidase and SNP-treated lungs was significantly increased compared with treatment with either agent alone (p < 0.01). This finding demonstrates increased ET-1 production in the lung after peroxynitrite exposure.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Endothelin-1 release after peroxynitrite generation. ET-1 concentrations in tissue homogenate from controls and lungs exposed for 120 min to peroxynitrite. ET-1 was measured in the supernatant of lung homogenates by EIA. XO (1 mU/ml) was infused together with 0.34 mM SNP at 0 min. The substrate HX (1 mM) was infused at time 30 min. The increase in ET-1 is expressed as ET-1 (pg/ml) over untreated control lungs. Data are means ± S.E.M, n = 3 (**p < 0.01).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   A and B, endothelin antagonism abolished effects of peroxynitrite generation. Time course of peroxynitrite (HX/XO plus SNP)-induced changes in airway (A) and vascular (B) resistance in lungs pretreated with either 8 µM BQ123 alone, 8 µM BQ788 alone, or 8 µM BQ123 in combination with 8 µM BQ788 10 min before generation of peroxynitrite by 1 mU/ml XO, plus 0.34 mM SNP at 0 min (arrow). HX (1 mM) was infused at 30 min. Airway RL/RL 0 min and vascular RV/RV 0 min resistance are normalized to 0 min. Data are means ± S.E.M, n = 3 (p = 0.38 and p = 0.06, respectively).

Dependence of Peroxynitrite-Induced Endothelin Action on Protein Synthesis. Since endothelins are not stored in cells, we checked the influence of transcriptional or translational arrest on peroxynitrite-induced alterations of lung physiology. Inhibition of transcription by actinomycin D and of translation by cycloheximide before peroxynitrite generation completely prevented peroxynitrite-induced bronchoconstriction (p = 0.02) as well as vasoconstriction (p < 0.001; data not shown). These findings indicate that de novo peptide/protein synthesis is required for the manifestation of pulmonary peroxynitrite-induced alterations. Remarkably, infusion of cycloheximide protected against pulmonary dysfunction as late as 80 min after peroxynitrite generation, i.e., about 20 min before injury could be detected. This indicates that the peptide that is causal to lung injury is released and acts very soon after translation, as is known for endothelins.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Oxidant-mediated tissue injury as well as vascular and bronchial dysfunction play a prominent role in various lung diseases. Our study provides a new link between biologically relevant reactive oxygen reaction products and endothelin-induced lung dysfunction. Our experiments confirmed in an organ system that simultaneous generation of O₂ and NO^· yield the reactive oxidant peroxynitrite (Beckman et al., 1990). It is shown that increased amounts of ET-1 are found in lung tissue exposed to the peroxynitrite-generating agents. Finally, the study demonstrates that peroxynitrite caused broncho- and vasoconstriction by a process with a lag phase of about 100 min requiring intact protein synthesis and engagement of both the ET_A and ET_B receptor.

Evidence for Exogenous Peroxynitrite Generation. As a methodological prerequisite, any detectable influence of the separate components of the exogenous peroxynitrite-generating agents on lung physiology needed to be excluded. SNP potentiates myocardial ischemia-reperfusion injury independent of peroxynitrite generation (Cope et al., 1997), and is reductively metabolized to cyanide and nitric oxide (Ivankovich et al., 1978). This vasodilator also interferes with hypoxic pulmonary vasoconstriction and therefore promotes mismatching of ventilation with perfusion (Oates, 1996). It has to be recalled that in our isolated perfused rat lungs, neither infusion of SNP alone nor of the nonspecific competitive NO synthase inhibitor L-N^G-monomethyl-L-arginine monoacetate affect bronchial tone compared with untreated lungs (K. Eichert, unpublished results). This suggests that in the not innervated isolated organ, spontaneous, endogenous NO formation plays a minor role. The negative controls of either component of the peroxynitrite-generating agents alone indicate that the pulmonary dysfunction we observed was in fact initiated only by the reaction product of NO^· and O₂, i.e., peroxynitrite. Any alternative combination product of these agents is considered unlikely.

Peroxynitrite-Induced Broncho- and Vasoconstriction. Because COX 1/2 inhibition and TXA₂/PGH₂ receptor antagonism had no effect on peroxynitrite-induced vasoconstriction, COX-derived prostanoids are unlikely to be chief mediators in this acute lung injury model. Because PGI₂ synthesis was not impaired under our conditions, an inhibition of prostacyclin synthase by nitration via peroxynitrite detected in bovine artery strips described recently does not seem to be the mechanism leading to the increased resistance that we observed (Zou and Ullrich, 1996). An alternative explanation, i.e., that peroxynitrite-induced activation of COX 1/2 with ensuing enhanced prostaglandin synthesis is involved (Goodwin et al., 1999), also seems unlikely in perfused rat lung, since peroxynitrite treatment did not lead to a further increase in the rates of TXA₂/PGH₂ and PGI₂ release compared with lungs exposed to HX/XO, i.e., superoxide generation alone, and because the inhibition of COX 1/2 by acetylsalicylic acid did not prevent lung injury.

Evidence for ET-Mediated Broncho- and Vasoconstriction. Our differential pharmacological inhibition experiments identify endothelins as candidate pivotal mediators of broncho- and vasoconstriction in this model. The ET_A-selective antagonist BQ123 predominantly blocked peroxynitrite-induced vasoconstriction. The observation that ET_B receptor antagonism was associated with a trend to further increase the vascular resistance might be explained by an enhanced activation of the ET_A receptor by ET-1. Since the combined inhibition of both receptors completely prevented the vasoconstriction, we conclude that both receptors play a role in mediating the effects of ET-1 in this model, however, to a different extent.