INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A successful transition from gas exchange by the placenta to gas exchange by the lung depends upon a dramatic decrease in pulmonary vascular resistance (Rpv) at birth. If this normal decrease in Rpv does not occur, persistent pulmonary hypertension of the newborn (PPHN) may develop. Surgical ligation or constriction of the ductus arteriosus in the fetal lamb several days before delivery at term results in PPHN following delivery (1, 2). This lamb model closely resembles the human condition, with suprasystemic pulmonary artery pressure (Ppa) and anatomic changes including muscularization of the normally nonmuscular intraacinar arteries (3).

Nitric oxide (NO) is produced from L-arginine by an endothelial constitutive nitric oxide synthase (NOS), and is an important factor in the decrease in Rpv that occurs at birth (4, 5). NO is believed to relax vascular smooth muscle by stimulation of soluble guanylate cyclase and production of the second messenger, guanosine-3',5'-cyclic monophosphate (cGMP), from guanosine-5'-triphosphate. Inhaled NO (iNO) at doses of 5 to 80 ppm decreases Ppa and improves survival in lambs with PPHN after ductal ligation (6). Multicenter, randomized trials have shown that iNO improves oxygenation and decreases the need for extracorporeal membrane oxygenation (ECMO) in infants with PPHN (7). However, these studies also demonstrated that although iNO effectively improved gas exchange in 60% of the treated neonates, 40% did not respond adequately and required treatment with ECMO. Additional research, directed at better understanding of the mechanisms involved in the pathogenesis and treatment of PPHN, is likely to improve the efficacy of treatment with iNO.

Potential toxicities of NO and its higher oxides indicate the need for designing therapeutic strategies that will allow delivery of the lowest effective iNO concentration. iNO is typically given in combination with high concentrations of inspired O₂. These conditions may result in increased production of oxygen free radicals, such as superoxide. NO is also a free radical, and can rapidly combine with superoxide to form the potent oxidant peroxynitrite (ONOO-). We have previously shown that prophylactic treatment with recombinant human superoxide dismutase (rhSOD) reduces inflammatory changes and lung injury caused by prolonged exposure to high concentrations of oxygen and iNO (10). By scavenging superoxide, SOD may increase the bioavailability of iNO while simultaneously reducing peroxynitrite formation. In the current study, we examined the short-term effects of a single intratracheal dose of rhSOD in newborn lambs with PPHN, both alone and in combination with 5 ppm and 80 ppm of iNO.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The Laboratory Animal Care Committee at the State University of New York at Buffalo approved all protocols used in the study.

Materials

The following pharmacologic agents were used for isolated vessel protocols: indomethacin, L-norepinephrine, D,L-propranolol hydrochloride, polyethylene glycol-SOD (SOD), polyethylene glycol-catalase (CAT), and S-nitrosyl-acetylpenicillamine (SNAP). All drugs were purchased from Sigma Chemical Company (St. Louis, MO). For intact lambs, rhSOD was obtained from Bio-Technology General Corporation (Iselin, NJ). NO was obtained as 1,000 ppm in nitrogen (Matheson Gas Products Inc., Twinsburg, OH), and blended with oxygen to obtain either 0.5 or 5 ppm.

Prenatal Ligation of the Ductus Arteriosus

The technique for creating pulmonary hypertension by prenatal ligation of the ductus arteriosus has been described previously (1, 2, 11). Time-dated pregnant ewes were operated on at 127 d of gestation (term = 146 d). A left lateral thoracotomy was performed in the fourth intercostal space of the fetus, and the ductus arteriosus was ligated. The chest was closed and the fetus was returned to the uterus. The ewe was allowed to recover for 9 d.

Isolated Vessel Protocol

Following operative delivery, and before their first breath, lambs were killed by rapid exsanguination via direct cardiac puncture. We have previously described the techniques for study of isolated vessels in detail (11, 12). In the present study we used fifth-generation pulmonary arteries (13) with inside diameters of  500 µm. All pulmonary arteries were pretreated for 20 min with 10⁵ M indomethacin to prevent the formation of vasoactive prostaglandins, and with 10⁶ M propran- olol to block activation of -adrenergic receptors. Vessels were then preconstricted with a submaximal concentration of norepinephrine. Cumulative concentration-response curves for SNAP (10⁹ to 3 × 10⁶ M) were developed. Some vessels were pretreated with polyethylene glycol-bound SOD (37.5 U/ml) alone or in combination with catalase (1,200 U/ml) 20 min before relaxation with SNAP.

Newborn Lamb Procedures

Lambs were delivered and studied 9 d after ductal ligation. We have reported the techniques used for this in detail (12, 14, 15). Before delivery, the fetal trachea was intubated and catheters were placed for hemodynamic monitoring. After delivery, ventilation was initiated at an FI_O2 of 0.95; the ventilatory rate and peak inspiratory pressure (PImax) were adjusted to keep the arterial carbon dioxide tension (Pa_CO2) between 35 and 50 mm Hg. Metabolic acidosis was corrected with tris-hydroxymethyl aminomethane (THAM), and systemic hypotension was corrected by administration of maternal blood. A 60-min stabilization period occurred before initial hemodynamic recordings were made. No blood or THAM was given during any of these protocols, and ventilator settings remained constant.

Measurements. Phasic pulmonary arterial, left atrial, aortic, and airway pressures were measured with Gould Statham physiologic pressure transducers (P-23 XL; Gould Electronics, Cleveland, OH). Pulmonary blood flow was measured with an ultrasonic transit-time flow transducer placed around the main pulmonary artery, and the data were processed by a digital flowmeter. Aortic blood was collected for measurement of pH, arterial oxygen tension (Pa_O2), (Pa_CO2), hemoglobin, and base status.

Experimental protocol. Animals (n = 18 lambs) received rhSOD alone (5 mg/kg), iNO alone (5 ppm and 80 ppm), or a combination of the two in a randomized block fashion. SOD was delivered as a single bolus through the endotracheal tube immediately after intubation. Hemodynamic and blood gas values were measured at 1 h of life (before iNO administration) and following 30 min of iNO at 5 ppm (1.5 h). iNO was discontinued, and hemodynamics were allowed to return to baseline for a 30-min period (2 h). iNO was then resumed at 80 ppm, and hemodynamics and blood gas values were measured after 30 min of iNO at 80 ppm (2.5 h).

Analysis for rhSOD Concentration

SOD concentration analyses were done by radioimmunoassay (RIA) with a monoclonal antibody specific for rhSOD as previously described (16). The concentration of rhSOD has been shown to directly correlate with activity of the enzyme.

Data Analysis

All data are given as mean ± SE. Statistical comparisons were made with analysis of variance (ANOVA) for repeated measures, or with one-way ANOVA as needed. A value of p < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

For in vitro studies, pretreatment of pulmonary arteries isolated from fetal lambs with PPHN with SOD alone (37.5 U/ml) significantly enhanced relaxations induced by SNAP (Figure 1). Pretreatment with SOD in addition to catalase (1,200 U/ml) enhanced relaxations induced by SNAP to a similar degree as did SOD alone (Figure 1).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Cumulative concentration-response curves for the NO donor SNAP. Pretreatment of fifth-generation pulmonary artery rings with SOD alone, and with SOD and catalase, significantly enhanced relaxations in response to SNAP in lambs with PPHN. All rings were pretreated with indomethacin and propranolol. Relaxations are expressed as percent of a plateau constriction in response to norepinephrine (100%). Data are mean ± SEM, n = 4 lambs for each data point. *p < 0.05 versus SNAP alone.

For in vivo studies, we compared initial Ppa measured at 1 h of age in lambs that received rhSOD at the time of intubation (n = 12) with that in lambs that did not receive rhSOD (n = 6). Ppa was significantly lower in the lambs that received SOD (Figure 2).

View larger version (14K): [in this window] [in a new window]   Figure 2.   Intratracheal administration of rhSOD at the time of tracheal intubation reduced pulmonary artery pressure at 1 h of life. Data are mean ± SEM, n = 12 lambs treated with rhSOD, n = 6 lambs not treated. * p < 0.05 versus untreated lambs.

Table 1 shows the hemodynamic results for each study group. Values shown were measured at 1 h (before iNO), after 30 min of inhalation of iNO at 5 ppm (1.5 h), and after 30 min of inhalation of iNO at 80 ppm (2.5 h). No significant changes were seen in any measured parameter after either study period in the group given rhSOD alone. iNO alone, inhaled at 5 ppm and 80 ppm, significantly decreased Ppa, increased pulmonary blood flow (Qp), and decreased Rpv relative to baseline. Qp and Rpv were significantly better than in the rhSOD group at 1.5 h, and all pulmonary hemodynamics were significantly better than in the rhSOD group at 2.5 h. The combination of rhSOD + iNO also significantly decreased Ppa, increased Qp, and decreased Rpv relative to baseline after both study periods. The combination also produced hemodynamic changes that were significantly better than in the group given rhSOD alone at both 1.5 h and 2.5 h. Ppa decreased to a significantly greater extent with rhSOD + iNO than with iNO alone at 1.5 h, and Rpv decreased to a significantly greater extent at both 1.5 and 2.5 h (Figure 3).

View larger version (11K): [in this window] [in a new window]   Figure 3.   Pulmonary vascular resistance in lambs treated with rhSOD alone, iNO alone, and a combination of the two. Values shown were obtained after stabilization (1 h), after 30 min of inhalation of 5 ppm iNO (or 1.5 h), and after 30 min inhalation of 80 ppm iNO (or 2.5 h). Data are mean ± SEM, n = 6 lambs for each data point. *p < 0.05 versus rhSOD alone; p < 0.05 versus iNO alone.

PAo did not change in any group of animals after 5 ppm of iNO. However, in both the iNO and the rhSOD + iNO groups, PAo fell after 80 ppm NO. The ratio of Ppa to PAo was calculated for all groups before and after iNO. The Ppa/PAo ratio in the rhSOD + iNO group remained significantly lower throughout both study periods than it did in the groups that received either rhSOD or iNO alone (Figure 4).

View larger version (11K): [in this window] [in a new window]   Figure 4.   The ratio of pulmonary to systemic arterial pressure responses in lambs treated with rhSOD alone, iNO alone, and a combination of the two. Values shown were obtained after stabilization (1 h), 30 min inhalation of 5 ppm iNO (or 1.5 h), and 30 min inhalation of 80 ppm iNO (or 2.5 h). Data are mean ± SEM, n = 6 lambs for each data point. * p < 0.05 versus rhSOD-alone and iNO-alone groups at 1.5 and 2.5 h.

Table 2 shows the arterial blood gas results for each study group. All values were measured at 1 h (before iNO), after 30 min of 5 ppm iNO (1.5 h), and after 30 min of 80 ppm iNO (2.5 h). The mean ventilator rate (67 ± 4 [mean ± SE] breaths/min) and PImax (35 ± 1 cm H₂O) did not change in any group after either study period. No significant changes were seen in any measured parameter after either study period in the group given rhSOD alone. iNO alone at 5 ppm and 80 ppm significantly increased Pa_O2 relative to baseline. The combination of rhSOD + iNO significantly increased Pa_O2 and pH, and decreased Pa_CO2 at 5 ppm and 80 ppm. The combination also produced blood gas changes that were significantly better than in the group given rhSOD alone at both 1.5 h and 2.5 h.

Table 3 shows the concentration of rhSOD in serum obtained at 1 h and at 4 h of life, as well as in lung parenchyma obtained after killing. The rhSOD concentration was significantly greater in both serum and lung parenchyma in animals that received rhSOD than in animals given iNO and no rhSOD.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

iNO has been shown to decrease the need for ECMO in term and near-term infants with PPHN (7). However, there are potentially serious toxic effects of iNO, including methemoglobinemia, platelet dysfunction, and acute lung injury resulting from the formation of oxidants. The reported toxicities depend on dose and duration of administration.

In addition to its own potential for causing toxicity, iNO is typically delivered in conjunction with high concentrations of inspired oxygen, a condition that may increase the production of oxygen free radicals such as superoxide. Not only is superoxide toxic, but it can react with iron through the Fenton reaction to form hydroxyl radicals, which are even more potent oxidants. The relative activities of antioxidant enzyme systems in the term or near-term infant with PPHN are important, but not well understood. Previous investigators have shown that healthy neonatal rats are able to increase SOD expression after exposure to high oxygen tensions (17); but preterm primates with respiratory distress syndrome are not (18).

In addition to the foregoing concerns about high inspired concentrations of oxygen, there are complex interactions between iNO and superoxide that have the potential to generate extremely potent oxidants such as peroxynitrite (OONO-). Exposure of endothelial cells in culture to hyperoxia and NO results in accelerated cell death as compared with treatment with either agent alone (19). Furthermore, we have previously shown that the combination of hyperoxia and 100 ppm iNO increases inflammatory changes, surfactant dysfunction, and acute lung injury in healthy newborn piglets (20). Although gross histologic markers of lung damage are not changed in newborn lambs with PPHN after 24 h of hyperoxia and iNO (6), more sensitive biochemical markers have not been analyzed in this model (19).

SOD catalyzes reactions that reduce the toxic superoxide radical to hydrogen peroxide. Further action, by catalase or glutathione peroxidase, reduces hydrogen peroxide to oxygen and water. By scavenging superoxide and decreasing the formation of peroxynitrite, SOD may enhance the bioavailability of iNO, resulting in a prolongation of its effect. By reducing the formation of toxic free radicals, SOD may also improve the safety of combined oxygen and iNO therapy. Previous studies with normal piglets have shown that pulmonary administration of rhSOD reduces pulmonary inflammation from prolonged exposure to hyperoxia and high concentrations of NO (100 ppm) (10).

We considered the possibility that SOD enhanced vascular relaxation in response to SNAP by increasing hydrogen peroxide rather than by simple elimination of superoxide. This is important because hydrogen peroxide has been shown to elicit relaxation in pulmonary arteries (21). However, we found that pretreatment with catalase did not alter the vascular effects of SOD, and we conclude that SOD enhanced the vascular relaxation induced by NO in the vessels of our PPHN model via the scavenging of superoxide anion.

The ductal ligation model of PPHN represents a model of abnormally remodeled pulmonary vessels (3). We have previously reported that pulmonary arteries from lambs with PPHN have decreased relaxation and show decreased accumulation of cGMP in response to exogenous NO relative to controls (11, 15, 22). Our current studies of isolated pulmonary arteries indicate that the blunted responses to exogenous NO in the vessels of our lamb model of PPHN can be enhanced to a significant degree by pretreatment with SOD. Villamor and colleagues recently reported that pretreatment with SOD did not enhance relaxation in response to NO in pulmonary arteries isolated from normal 12- to 24-h-old piglets (23). Their findings lead us to speculate that the effects of SOD are more profound in abnormally remodeled pulmonary vessels than in normal vessels, perhaps because excess superoxide production is a part of the pathophysiology of chronic antenatal pulmonary hypertension.

In intact, instrumented newborn lambs with PPHN, a single intratracheal dose of rhSOD decreased Ppa when it was measured 1 h later. We chose rhSOD for these studies because it is a compound available for testing in human infants (24). Because PAo was similar in all three groups, the dilation was selective for the pulmonary circulation. The Ppa/PAo ratios were significantly improved in the group that received rhSOD and iNO as compared with the two other study groups. We used 5 ppm and 80 ppm of iNO with and without rhSOD. These doses were selected because they represent the threshold and maximal concentrations, respectively, that affect pulmonary vascular hemodynamics in the ductal ligation lamb model of PPHN (25). At both concentrations of iNO, rhSOD enhanced the pulmonary hemodynamic effects of iNO.

We did not determine the mechanism for the apparent pulmonary vascular dilation in response to SOD alone in newborn lambs with PPHN. However, we speculate that the relaxations were due to inactivation of excess superoxide production. Carpenter and coworkers recently reported that SOD and catalase did not affect pulmonary hypertension induced by group B streptococcus in young lambs (26). Their model of acute pulmonary hypertension is not associated with vascular remodeling, which probably explains the difference in response from that in the present study. Production of reactive oxygen species has been shown to contribute to the pathogenesis of systemic hypertension. Several possible sources of superoxide have been proposed in the diseased vessel wall: superoxide may be produced by the endothelium, smooth-muscle cells, adventitial fibroblasts, or inflammatory cells that have migrated into the vascular wall (27). In systemic resistance vessels, superoxide production increases when vessels are subjected to high perfusion pressures, and appears to interfere with NO activity (28). We further speculate that elimination of superoxide increased the bioavailability of endogenous NO for vasorelaxation. Even though the PPHN model has been shown to have reduced expression and activity of NOS, relaxations in response to SOD in in vitro systems have been reported to occur after exposure to extremely low concentrations of NO (29). Other investigators have reported that NOS may not catalyze the formation of free NO unless high concentrations of SOD are present to clear superoxide (30), further emphasizing the potential importance of SOD activity.

SOD was adequately taken up in intact lambs with PPHN, based on SOD activity in lung tissue (measured after killing and washing of SOD from lung tissue) and on serum concentrations of the enzyme (measured at 6 h of life). These findings are in agreement with those in previous studies using fluorescence labeled rhSOD, which demonstrated specific cellular uptake of this compound (31). In the present study, rhSOD was administered immediately after delivery into the liquid filled lung, which may have assisted in its homogenous distribution. However, previous studies indicate that nebulization may be preferable in the mechanically ventilated lung, and that repeated dosing may be more effective than the single dose that we delivered (32). The correct dosage, interval of dosage, and method of drug delivery in the clinical setting need further evaluation.

The improvement in oxygenation, CO₂ exchange, and pH in response to the combination of rhSOD and iNO was more striking than the improvement in pulmonary hemodynamics. This may have been due to the significantly lower Ppa/PAo ratio in the lambs that received the combination therapy. It is also possible that the location of vasodilation favored the arterioles and postcapillary venules in this group of animals. This would be expected to increase net lung liquid clearance in the postnatal lung, which could also explain the improvement in oxygenation that we observed.

The decrease in PAo after inhalation of 80 ppm NO (Table 1) indicates that at high concentrations, iNO may exert a systemic effect. This effect was not different for lambs that received iNO alone and lambs that initially received rhSOD. In the clinical setting, 80 ppm iNO was found to improve oxygenation in infants with PPHN (7). Currently, however, this is generally considered to be a high dose, and no more effective in improving oxygenation than 5 to 20 ppm (33). Dose-dependent effects of iNO on pulmonary hemodynamics have been observed in the PPHN lamb model: In previous studies, 5 ppm iNO represented the threshold dose, and maximal pulmonary vascular effects occurred after a dose of 100 ppm iNO (25). However, in the present study, there were no significant differences in pulmonary hemodynamics following treatment with 5 ppm iNO as compared with 80 ppm iNO.

PPHN is a condition resulting from a diverse set of circumstances, and caution must be exercised when applying the results of experimental models to humans. However, the ductal ligation model of pulmonary hypertension is a chronic preparation resulting in antenatal vascular remodeling (3), and has striking physiologic and anatomic similarities to PPHN in human infants (34). Conducting studies with newborn animals is important, since recent studies propose distinct developmental time lines for the enzymatic steps in the NO/cGMP vasodilatory pathway (35). Our previous studies further indicate that prenatal ligation of the ductus arteriosus alters the expression of critical enzymes in the NO pathway (38, 11), a finding that would not be expected in acute models of pulmonary hypertension following infusion of vasoconstrictive agents. The current study provides a strong rationale to further explore the role of superoxide production in the pathogenesis of pulmonary hypertension, and for therapeutic intervention strategies for this disesae that use antioxidant enzymes.