Effect of ONO1714, a Specific Inducible Nitric Oxide Synthase Inhibitor, on Lung Lymph Filtration and Gas Exchange during Endotoxemia in Unanesthetized Sheep

This study protocol was approved by the institutional review board for the care of animals in Shinshu University (Asahi Matsumoto, Japan). Care and handling of animals were in accordance with the guidelines of the National Institutes of Health.

Thirteen sheep weighing 30–39 kg were used. We prepared chronic lung lymph fistulae and placed catheters for hemodynamic monitoring. Sheep were anesthetized with 12.5 mg/kg intravenous pentobarbital sodium and then ventilated with 0.5–1.0% halothane using positive-pressure ventilation. Through two right thoracotomies, the efferent lymphatic channel from the caudal mediastinal node was cannulated with a thin silicon tube. The caudal mediastinal node tail was then ligated at the free margin of the inferior pulmonary ligament to block contamination by nonpulmonary lymph. Through a left thoracotomy, we directly implanted silicon catheters into the main pulmonary artery and left atrium. A silicon tube was inserted into the thoracic aorta via  the carotid artery. An 8-French catheter sheath introducer (Cordis Laboratories, Miami, FL) was placed in the superior vena cava via  the right jugular vein. On the day before the experiment, a 7-French thermodilution Swan-Ganz catheter (Becton Dickinson, Tokyo, Japan) was passed into the pulmonary artery through the Cordis introducer. The animals were allowed to recover for at least 7 days with free intake of food and water after the surgical procedures.

All measurements were made under an awake and standing condition. Systemic arterial, pulmonary arterial, and left atrial pressures were continuously measured and recorded using calibrated pressure transducers (RMP-6008; Nihon Koden, Tokyo, Japan) and a recorder (WT-685G; Nihon Koden), respectively. Cardiac output was determined by the thermodilution method using a cardiac output computer (model 9520; Edwards, Santa Ana, CA). We collected lung lymph in heparinized tubes and measured Qlym every 30 min. The protein concentration in plasma and pooled lung lymph was determined by biuret method every 30 min, and the lymph–to–plasma protein concentration ratio (L/P) was calculated. Lymph protein clearance (Clym) was calculated by multiplying L/P by Qlym. Blood samples for blood gas analysis were drawn before endotoxin and 0.5, 1, 2, 3, 4, and 5 h after endotoxin from systemic and pulmonary arterial lines, respectively. Blood gas analysis (partial pressure of oxygen [Po²], partial pressure of carbon dioxide [Pco²], pH, oxygen saturation, and hemoglobin) were performed with use of a blood gas analyzer (ABL-2; Radiometer, Copenhagen, Denmark). Alveolar–arterial oxygen pressure difference (A-aDo²) and venous admixture (Q^S/Q^T) were calculated using standard equations. The plasma and lung lymph concentrations of nitrate–nitrite (NOx) were also measured using the Cayman Chemical Nitrate/Nitrite Colorimetric Assay Kit (Cayman Chemical Company, Ann arbor, MI). The first step is the conversion of nitrate to nitrite using nitrate reductase. The second step is the addition of Griess reagent,25which converts nitrite into a deep-purple azo compound. The plasma sample was taken from the systemic artery catheter. NOx concentrations in plasma and lung lymph were measured before endotoxin and 1, 3, and 5 h after endotoxin administration in sheep treated with and without ONO1714.

Sheep were allocated to two experimental groups. In experiment 1 (endotoxin alone; n = 6), after stable baseline measurements over 2 h, Escherichia coli  endotoxin (1 μg/kg, diluted in 30 ml sterile normal saline, E. coli  0127; B8; Difco Laboratories, Franklin Lakes, NJ) was infused. In experiment 2 (endotoxin combined with ONO1714 treatment; n = 6), ONO1714 ((1S,5S,6R,7R)-7-chloro-30imino-5-methyl-2-azabicyclo[4.1.0] heptane hydrochloride) was supplied by Ono Pharmaceutical Company Ltd. (Osaka, Japan). ONO1714 (0.1 mg/kg) was dissolved with 20 ml normal saline and intravenously administered for 5 min. The treatment of ONO1714 was performed 30 min before endotoxin infusion. Observation of systemic and pulmonary hemodynamics, Qlym, and blood samples was continued for 5 h after endotoxin in both groups.

The data are expressed as mean ± SD. Changes measuring variables over time and between groups were analyzed by two-way analysis of variance, and the differences were tested by Fisher exact test. P < 0.05 was accepted as a significant difference.

Pulmonary hemodynamic response to endotoxin in both with and without ONO1714 are shown in figure 1. In both groups, pulmonary artery pressure increased significantly within 1 h after the start of endotoxin administration (early phase) and then decreased gradually and returned to the baseline by 4–5 h (late phase). iNOS inhibition with ONO1714 resulted in slight increase in pulmonary artery pressure and decrease in cardiac output at baseline, but these changes were statistically not significant. In addition, ONO1714 did not cause any significant changes in early transient and late sustained pulmonary hypertension during endotoxemia. The time courses of cardiac output and left atrial pressure during endotoxemia were also not affected by ONO1714 treatment.

The time courses of systemic aortic pressure both with and without ONO1714 treatment were shown in figure 2. ONO1714 did not cause any change in systemic aortic pressure at baseline. Systemic arterial pressure decreased significantly and transiently at the early phase after endotoxin administration. There were no significant differences in the decrease of systemic artery pressure between groups with and without ONO1714 treatment.

The time courses in Qlym, L/P, and Clym are summarized in figure 3. There were no significant differences at baseline lung lymph balance in either group. During the early phase of endotoxemia, concomitant with the peak in pulmonary hypertension, Qlym increased approximately threefold, and L/P decreased slightly in the endotoxin-alone group. Subsequently, Qlym decreased slightly but remained significantly increased during the late phase with an increased L/P. Increased Qlym in ONO1714 treatment was low, compared with that in the endotoxin-alone group. Increased Qlym during the late phase in sheep treated with ONO1714 was significantly lower than that in endotoxin-alone sheep at one point. However, increased levels of Clym during the late phase in the ONO1714 group were significantly lower than those in the endotoxin-alone group.

The time courses of Po², A-aDo², and shunt ratio (Q^S/Q^T) are summarized in figure 4. Po²after endotoxin administration was significantly decreased within 1 h and gradually recovered to baseline at 5 h after endotoxin in both groups. However, the values of Po²in the ONO1714 group during 2–4 h were significantly higher than those in the endotoxin-alone group. In the endotoxin-alone group, A-aDo²immediately increased from 0.53 ± 0.11 at baseline to 0.87 ± 0.24 at 0.5 h. The values of A-aDo²then sustained the increased levels during the late phase and returned to the baseline level approximately 5 h after the start of endotoxin infusion. In sheep treated with ONO1714, the values of A-aDo²were significantly lower than those in the endotoxin-alone group. Likewise, the time courses of Q^S/Q^Tin both groups were almost similar to those in A-aDo². The increased values of Q^S/Q^Tduring 2–4 h after endotoxin administration in animals treated with ONO1714 were significantly lower than those in the endotoxin-alone group.

There were no significant differences at baseline values of NOx in plasma (2.78 ± 0.4 pg/ml in endotoxin alone, 3.38 ± 0.4 pg/ml in endotoxin plus ONO1714; P = not significant) and lung lymph (4.66 ± 0.6 pg/ml in endotoxin alone, 6.31 ± 0.6 pg/ml in endotoxin plus ONO1714; P = not significant) between groups with and without ONO1714 treatment. We estimated NOx values as the percent changes to baseline and graphed this in figure 5. The plasma NOx in endotoxin group gradually increased after endotoxin administration and reached approximately 100% increase at 5 h. Likewise, NOx in lung lymph increased significantly after endotoxin administration. In contrast, NOx in both plasma and lung lymph remained unchanged after endotoxin administration in animals treated with ONO1714. Therefore, ONO1714 treatment could prevent the increased NOx in plasma and lung lymph during endotoxemia.

In the current study, we have shown in unanesthetized sheep that a selective iNOS inhibitor, ONO1714, reduces lung lymph filtration and improves oxygenation during endotoxemia in awake sheep. These data suggest that NOS activation through inducible isoform contributes to the development of endotoxin-induced lung injury in sheep.

Nonselective NOS inhibitors such as N -nitro-l-arginine methyl ester (or N -monomethyl-l-arginine) aggravated endotoxin-induced pulmonary hypertension and increased extravascular water in sheep.19,20These physiologic findings by nonselective NOS inhibitors were due to the pharmacologic effects of constitutive NOS and seemed to be unsuitable in the situation of increased pulmonary permeability. Although Evgenov et al.  18demonstrated that aminoguanidine administration improved oxygenation during the late phase of endotoxemia in sheep, aminoguanidine caused significant increases in pulmonary arterial pressure and Qlym during the simultaneous phase in the same sheep model. These findings suggested that pharmacologic effects of aminoguanidine on constitutive NOS during endotoxemia were present even in aminoguanidine. In the current study, ONO1714 did not cause any pulmonary or systemic hemodynamic changes induced by endotoxin. The finding suggested that ONO1714 was superior in selectivity against iNOS as compared with aminoguanidine. Our results in vivo  were therefore similar to those in vitro .21Furthermore, ONO1714 significantly reduced Qlym and improved abnormal oxygenation during endotoxemia, suggesting that ONO1714 could attenuate endotoxin-induced lung injury in sheep. Based on the current results as well as published data,15–17it seems that selective inhibition of iNOS leads to improvement in lung edema and gas exchange during endotoxemia.

We used a newly synthesized iNOS inhibitor, ONO1714, which is more potent and selective both in vitro  and in vivo  than the previously evaluated iNOS inhibitors. ONO1714 is 10-fold more selective for iNOS than constitutive NOS in the human cell line, which is still higher than aminoguanidine. The iNOS selectivity of ONO1714 was approximately 2-fold higher than aminoguanidine in human cell lines.22In addition, inhibitory activity of ONO1714 is more than 20,000-fold more potent than aminoguanidine. In an in vivo  study, which was performed in endotoxin-stimulated mice, ONO1714 inhibited the increases in NOx accumulation in plasma 2,600-fold more than N  ^G-monomethyl-l-arginine monoacetate. These data indicated that ONO1714 was one of the most potent iNOS inhibitors in vitro  and in vivo  to date. We confirmed that ONO1714 could completely prevent NOx concentrations in plasma and lung lymph in response to endotoxin, whereas plasma NOx was not prevented by the treatment of aminoguanidine in the same endotoxemic sheep model18and other studies.15,16These observations suggest that the dose of 0.1 mg/kg ONO1714 was adequate to inhibit the iNOS activity during endotoxemia and that the inhibitory activity of ONO1714 was more potent than that of aminoguanidine in sheep.

Hypoxemia during endotoxemia is thought to be primarily due to ventilation/perfusion mismatch with physiologic shunting through collapsed alveoli.26It seems that the degree of pulmonary edema is a major determinant of Q^S/Q^T. In the current study, iNOS inhibition with ONO1714 improved abnormal oxygenation during endotoxemia. The beneficial findings were most likely responsible for the reduction of lung fluid filtration, as assessed by Qlym and Clym. On the other hand, it is well known that endotoxemia impairs hypoxic pulmonary vasoconstriction.27,28The attenuated hypoxic pulmonary vasoconstriction during endotoxemia might cause increased ventilation/perfusion mismatching, augmented right-to-left shunting of venous blood. Nitric oxide contributes to the impairment of hypoxic pulmonary vasoconstriction during endotoxemia.8,29,30We previously reported in the same sheep model that ONO1714 could restore the decreased hypoxic pulmonary vasoconstriction during endotoxemia.8We speculate that restoration of decreased hypoxic pulmonary vasoconstriction by ONO1714 could contribute to improve abnormal gas exchanges in the current study.

Nitric oxide inhibition could restore the endotoxin-induced hypotension in animal models6,17and humans.31These findings suggested that overproduction of nitric oxide was responsible for the hypotension. Likewise, administration of nonselective NOS inhibitor could restore the endotoxin-induced hypotension in sheep.32However, it has been reported that a significant increase in systemic arterial pressure is also obtained before endotoxin administration in sheep21and that prominent iNOS expression is induced 6 h after endotoxin treatment.12,33Furthermore, ONO1714 did not restore the early hypotension induced by endotoxin, although the unanesthetized sheep model used in the current study was not typical for endotoxic shock. Therefore, taken together, we believe that early hypotension after endotoxin administration in sheep was at least independent for the iNOS pathway.

Although iNOS inhibition by ONO1714 prevented the increases in nitric oxide products after endotoxin, improvement of lung injury was not complete in sheep treated with ONO1714. Similarly, several studies showed that endotoxin-induced lung injury was not completely reversed by the other NOS inhibitors and even in iNOS-deficient animals.12,13,15–17These results indicate that a self-limited contribution of the iNOS pathway and some other mediators are involved in the development of endotoxin-induced lung injury. Especially reactive oxidant species, such as superoxide anion, interact with nitric oxide in the development of acute lung injury.34Peroxynitrite, a product of nitric oxide and superoxide, has been implicated in sepsis-induced lung injury.35Because there are no data about the direct effects of ONO1714 on reactive oxygen intermediates, further studies are needed to evaluate the pharmacologic interaction of ONO1714 with oxygen radical production during endotoxemia.

Pharmacologic effects of aminoguanidine on the inflammatory disorders may vary by the dose of the agents, sepsis, or both.15–17The relatively high dose of aminoguanidine caused an increase in pulmonary artery pressure in sheep.18We previously showed that 0.2 mg/kg ONO1714 caused pulmonary vasoconstriction in normal sheep, which is thought to be an inhibitory influence on constitutive NOS.8We have not investigated the dose-related effects of ONO1714 on the degree of lung injury and gas exchanges in the current study. To further evaluate the usefulness of this agent in clinics, we need additional studies of differential dose settings, other types of inflammatory models, or both.

In summary, we have shown that a new selective iNOS inhibitor, ONO1714, attenuates increased pulmonary fluid flux and improves abnormal oxygenation during endotoxemia in unanesthetized sheep. The current study reveals that nitric oxide release by iNOS pathway partially contributes to the development of lung injury during endotoxemia and that iNOS inhibition by ONO1714, a newly developed selective iNOS inhibitor, may be a therapeutic approach in patients with sepsis, acute lung injury, or both.