Main

ET are powerful vasoactive mediators synthesized and released by the vascular endothelium (1). Plasma levels of circulating ET-1, the predominant isoform of ET, vary with age: in the fetus and newborn infants they are higher than in neonates and infants up to 3 months of age; they then become nearly constant and similar to adult levels (2–4). ET-1 plasma levels are also elevated in newborns and children suffering from diseases associated with pulmonary hypertension (PHT) and are correlated with the severity of the disease (5–7).

ET have opposite vascular effects mediated through different receptors. ETA receptors, located on vascular smooth muscle cells, mediate ET-1 vasoconstriction. ETB receptors are found on vascular smooth muscle cells, where they mediate vasoconstriction, and also on endothelial cells, where their stimulation produces vasodilation, presumably through the endothelial synthesis of NO and prostacyclin (8,9). Activation of the ETB receptor mediates pulmonary clearance of circulating ET-1 and reuptake of ET-1 by endothelial cells (8). In addition, although it is generally proposed that endothelial NO synthesis is related to ET-B receptor stimulation (8–11), another study shows that activation of ET-A receptors also can enhance NO production in the vasculature (12). In experimental studies on the regulation of the pulmonary vascular tone in perinatal lambs, it has also been demonstrated that ET-1 produces various hemodynamic effects (vasoconstriction as well vasodilation), depending on the presence of different types of ET-receptors with opposite effects (10,13–18). Thus, the role played by endogenous ET in the regulation of basal and increased vascular tone in perinatal experimental models remains uncertain.

ET have also been hypothesized to play a role in the mediation of HPV (19). Increased levels of ET-1 in the plasma and higher levels of ET-1 mRNA and ETA receptors in the lung, were measured in rats exposed to short- and long-term hypoxia (20,21). Pretreatment with the ETA receptor antagonist BQ-123 curtails the increase in pulmonary vascular tone secondary to acute or chronic hypoxia in rats and lambs (17,22,23). Using intact newborn lambs, Wong et al. (14) found ET-1 not to be involved in the regulation of HPV. These discrepancies in the possible role of ET regulating HPV might be explained, at least in part, by the variable release of NO, which has been shown to limit hypoxic pulmonary vascular tone in intact animals (24–28).

In the context of previous experiments from our laboratory studying the possible role of endothelial mediators on pulmonary vascular tone in an anesthetized (adult) dog model, these experiments were set up to investigate the effects of bosentan on pretreated lambs with L-NA at controlled flow (28).

We started from the hypothesis demonstrated in the dog experiments, that the vasoconstrictive effects of ET are self-limited by an ET-induced synthesis of endogenous NO (28). For that purpose, we examined the effects of endogenous ET blockade on hyperoxic and hypoxic vascular tone, with and without NO synthase inhibition by L-NA in intact perinatal lambs. As ET have opposed vascular effects mediated by different receptors and ETA as well as ETB receptor stimulation enhances NO production, a complete blockade of endogenous ET requires both ETA and ETB receptor blockade (12).

The effects of hypoxia and drugs on pulmonary vascular tone were evaluated by constructing multipoint pressure/flow plots, which has been previously demonstrated in experiments from our laboratory to be the method of preference to evaluate the functional state of the pulmonary circulation (24,28,29). This approach allows discrimination between active from passive changes of Ppa and PVR. To our knowledge, this method has never been reported in a model of intact perinatal lamb.

METHODS

Experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and were approved by the Committee on the Care and Use of Animals in Research of the School of Medicine of the Université Libre de Bruxelles.

Animal preparation.

Twenty-one mixed-breed western lambs (age, 15–30 d; weight, 5.7 ± 2.4 kg) were anesthetized with ketamine (10 mg/kg i.v.) through an over-the-needle catheter inserted percutaneously in a hind leg vein. Profound anesthesia was obtained by administering pentobarbital sodium (20 mg/kg i.v.). The animals were paralyzed with pancuronium bromide (0.2 mg/kg i.v.), intubated with a cuffed endotracheal tube (4.5-mm internal diameter), and ventilated (Elema 900 C, Siemens Elema, Sölna, Sweden) with a tidal volume of 10–12 mL/kg adjusted to maintain carbon dioxide tension in arterial blood (Paco 2) between 35 and 45 mm Hg, a positive end-expiratory pressure of 3 cm H2O, a sigh, a respiratory rate of 30 breaths/min, and an Fio2 of 0.4 during the preparation. This higher than normal Fio2 was selected to maintain the lungs above the threshold for HPV (28).

Continuous infusion of a sodium bicarbonate solution (7%) was started at a rate of 1 mEq/kg/h and was adjusted to maintain arterial pH above 7.30. Antibiotic prophylaxis (cefazoline 100 mg/kg) was given at the beginning of the preparation and after 6 h.

Pentobarbital (3 mg/kg i.v.) and pancuronium bromide (0.1 mg/kg) were given on a repeated hourly basis, to maintain anesthesia and paralysis. Central body temperature was kept in the normal range (39–40°C) with an infrared rewarmer and a heating blanket throughout the preparation and the entire experiment.

A femoral artery (Ledercath 20 G) and a balloon-tipped, flow-directed pulmonary catheter (model 132-F5, Baxter Edwards, Irvine, CA) were respectively inserted after denudation of the right femoral artery and right external jugular vein. They allowed measurements of systemic and pulmonary hemodynamics and sampling of arterial and mixed venous blood and were positioned by means of pressure monitoring. A balloon-tipped catheter (model 132-F7, Baxter Edwards) was advanced into the inferior vena cava through a right femoral venotomy. Filling of the balloon with normal saline produced a titratable decrease in Q by reducing venous return. The distal port of the inferior vena caval catheter was used for fluid and drug administration. Thrombus formation along the catheters was prevented by heparin (100 IU/kg i.v.).

Glucose homeostasis was maintained by a continuous infusion of a glucose 10% in water solution (infusion rate, 2.5 mL/kg/h). After preparation, the animal was allowed to stabilize for 1 h.

Standard measurements.

Heart rate (HR) was determined from a continuously monitored electrocardiographic lead.

Pulmonary and systemic arterial pressures (Psa) were measured using Gould Statham P50 transducers (Gould, Cleveland, OH). HR and vascular pressures were displayed on a monitor (Sirecust 404, Siemens, Erlangen, Germany) and recorded on a six-channel recorder (model 2600S, Gould). The pressure transducers were zero referenced at mid-chest and vascular pressures were recorded at end expiration. Q was measured by thermodilution using injections of 5 mL of 0.9% sodium chloride at 0°C and a computer (9520-A CO-set, Edwards Laboratories, Santa Ana, CA); it was calculated as the mean of at least three successive measurements.

Arterial and mixed venous blood gases were determined immediately after drawing the samples by a tonometered automated analyzer (ABL2, Radiometer, Copenhagen, Denmark) and corrected for temperature.

The pulmonary vascular pressure signals were sampled at 200 Hz using an analog/digital converter (DAS 8-PGA, Keithley-Metrabyte, Taunton, MA) and stored and analyzed on a personal computer.

Experimental protocol.

In all lambs, a first four-point Ppa – Ppao/Q relationship was constructed at baseline (Fio2 0.4). One point was generated with the vena caval balloon being empty (first point = high Q) and three points with stepwise filling of the balloon (last point = low Q). One such relationship was generated in about 30 min. Arterial and mixed venous blood gases were measured at each point of the plot. The same procedure was repeated with an Fio2 0.1. Each Fio2 was applied for at least 10 min to allow stabilization before new measurements were done. In a series of pilot experiments (data not shown) with solvent alone, we were able to reproduce identical hypoxic responses at least up to three times.

Two further hypoxic challenges were performed 30 min after the start of either an infusion of solvent (normal saline) followed by an infusion of bosentan (Actelion, Allschwill, Switzerland) given as an i.v. bolus of 5 mg/kg followed by a constant infusion of 5 mg/kg (Group 1: n = 7); or an infusion of L-NA (Sigma-Aldrich, Bornem, Belgium) alone given as an i.v. bolus of 5 mg/kg followed by a constant infusion of 5 mg/kg (Group 2: n = 7); or an infusion of L-NA followed by an infusion of bosentan (Group 3: n = 7). The infusion rates were adapted during the experiments so that at the end of all pressure/flow measurements all lambs received the same total amount of bosentan (i.e. 10 mg/kg) and L-NA (i.e. 10 mg/kg). The dosage of bosentan has been shown to provide optimal ET blockade in anesthetized dogs and sheep (28,30). The dose of L-NA has been shown to block NO synthase in lambs (31) as well as in dogs (24). Lambs were randomly allocated to one of the three study groups. L-NA was dissolved in 25 mL of normal saline and bosentan was diluted in heated, distilled, sterile water just before the experiments.

Data analysis.

All results are expressed as means ± SEM. Q is expressed in mL/kg/min. The individual (Ppa – Ppao)/Q plots were essentially linear. The correlation coefficients of all individual plots were >0.8 except in one plot that was deleted from analysis. A least squares regression analysis was used to obtain the slopes and the zero-flow pressure intercepts of the individual (Ppa – Ppao)/Q plots. Values of Ppa – Ppao interpolated from the regression analysis for individual lambs were averaged at 50 mL/kg/min intervals of Q from 100 to 200 mL/kg/min to obtain the composite (Ppa – Ppao)/Q plots shown in the left panel of Figures 1–3. HPV was defined as the hypoxia-induced increase in the Ppa – Ppao gradient at identical Q, i.e. low and high Q (Figs. 1–3, right panels). A two-factor ANOVA for multiple measurements was used to assess the effects of hypoxia and drugs on hemodynamics and blood gases. When the F-ratio of the ANOVA reached a value of p < 0.05, modified t tests were used to determine which means differed.

Figure 1
figure 1

(A) Multipoint mean Ppa minus Ppao at standardized (100–200 mL/kg/min) Q plots at baseline (circles), after the administration of solvent (squares) and after the administration of bosentan (triangles) in seven lambs during hyperoxia (solid symbols and lines) and hypoxia (open symbols and dashed lines). Vertical bars indicate the SEM; n = 7. *p < 0.005 between hypoxia and hyperoxia for the same drug condition. (B) Hypoxic response defined as the increase in the gradient between Ppa – Ppao measured at constant Q (LQ = lowest Q of the pressure/flow plots; HQ = highest Q of the pressure/flow plots) in response to a reduction in the fraction of inspired oxygen from 0.4 to 0.1 at baseline (open bars), after administration of solvent (closed bars), and after administration of bosentan (shaded bars). Vertical bars indicate the SEM.

Full size image
Figure 3
figure 3

(A) Multipoint mean pulmonary artery pressure (Ppa) minus occluded Ppa (Ppao) at standardized (100–200 mL/kg/min ) Q plots at baseline, after the administration of L-NA followed by the administration of bosentan in seven lambs during hyperoxia (solid symbols and lines) and hypoxia (open symbols and dashed lines). Vertical bars indicate the SEM; n = 7. *p < 0.005 between hypoxia and hyperoxia for the same drug condition. (B) Hypoxic response defined as the increase in the gradient between Ppa-Ppao measured at constant Q (LQ = lowest Q of the pressure/flow plots; HQ = highest Q of the pressure/flow plots) in response to a reduction in the Fio2 from 0.4 to 0.1 at baseline (open bars), after administration of L-NA (closed bars) followed by the administration of bosentan (shaded bars). Vertical bars indicate the SEM. *p < 0.005.

Full size image

RESULTS

Results from the three study groups are presented in Tables 1–3 and Figures 1–3. At baseline, hemodynamic and blood gas values did not differ between study groups. Reduction of Q significantly decreased mean Psa, Ppa, Ppao, right atrial pressure, and mixed venous oxygen tension (Pvo2). Hypoxia significantly reduced the oxygen tension in arterial blood (Pao2) and Pvo2 and increased Ppa and Ppa-Ppao without changes in Q, indicating pulmonary vasoconstriction.

Table 1 Blood gas and hemodynamic data, solvent followed by bosentan (BOS) (group 1, n = 7)
Full size table
Table 3 Blood gas and hemodynamic data, L-NA followed by L-NA + bosentan (BOS) (group 3, n = 7)
Full size table

Effects of solvent.

Solvent had no significant effect on hemodynamics and blood gases and did not affect the (Ppa – Ppao)/Q relationships during hyperoxia or hypoxia (Table 1, Fig. 1).

Effects of bosentan alone.

Bosentan significantly decreased Ppa and Ppa – Ppao in hyperoxia as well as in hypoxia (Table 1, Fig. 1), so that HPV was not significantly affected (Fig. 1). Bosentan also decreased mean Psa at Fio2 0.4 and at Fio2 0.1 (Table 1).

Effects of L-NA alone.

L-NA increased Ppa and Ppa – Ppao in hyperoxia and, with a larger magnitude, in hypoxia (Table 2, Fig. 2), so that HPV was enhanced after L-NA (Fig. 2). L-NA also increased mean Psa (Table 2).

Table 2 Blood gas and hemodynamic data, L-NA alone (group 2, n = 7)
Full size table
Figure 2
figure 2

(A) Multipoint mean Ppa minus Ppao at standardized (100–200 mL/kg/min) Q plots at baseline (circles), after the subsequent administration of two doses of L-NA [L-NA1 (squares) and L-NA2 (triangles)] in seven lambs during hyperoxia (solid symbols and lines) and hypoxia (open symbols and dashed lines). Vertical bars indicate the SEM; n = 7. *p < 0.005 between hypoxia and hyperoxia for the same drug condition. (B) Hypoxic response defined as the increase in the gradient between Ppa – Ppao measured at constant Q (LQ = lowest Q of the pressure/flow plots; HQ = highest Q of the pressure/flow plots) in response to a reduction in the Fio2 from 0.4 to 0.1 at baseline (open bars) and after subsequent administration of two doses of L-NA [LNA1 (closed bars) and L-NA2 (shaded bars)]. Vertical bars indicate the SEM. *p < 0.005.

Full size image

Effects of bosentan after L-NA.

After L-NA, bosentan significantly decreased Ppa and Ppa – Ppao in hypoxia, but these pressure values remained higher than the baseline values, and HPV was not significantly affected (Table 3, Fig. 3). During hypoxia, L-NA shifted the (Ppa – Ppao)/Q relationship to higher pressures, and bosentan given after L-NA reduced Ppa – Ppao at all levels of Q. As observed in the first study group, bosentan decreased mean Psa (Table 3).

DISCUSSION

In our model of anesthetized intact lamb, ET receptor antagonism by bosentan decreased pulmonary and systemic vascular tone both in hyperoxia and hypoxia. NO synthase inhibition increased pulmonary vascular tone more in hypoxia than in hyperoxia so that HPV was enhanced. After L-NA, bosentan still decreased pulmonary vascular tone in hypoxia but did not affect the magnitude of HPV. This suggests that the vasodilator effect of bosentan may be due to a non-NO-dependent mechanism. Because a combined ETA and ETB receptor antagonist was used, the results from these experiments cannot identify the respective effects of these receptors on HPV.

The investigation of pulmonary hemodynamics in intact animals and humans usually rests on the determination of PVR (32). PVR is calculated as the difference between mean Ppa and Ppao, assumed to be equal to left atrial pressure, divided by Q (32). This method is based on the assumptions that the Ppa – Ppao/Q relationship is linear and passes through the origin. The latter is in fact incorrect when the lungs are diseased and/or hypoxic (24,29). When the extrapolated pressure intercept of the pressure-flow plots, i.e. the closing pressure of the pulmonary vessels or their effective downstream pressure, exceeds Ppao, the calculation of PVR cannot discriminate between passive (flow-dependent) and active changes in pulmonary artery pressure.

For that reason, it is preferable to evaluate the functional state of pulmonary circulation by multipoint pressure/flow plots, or to measure pulmonary vascular pressures at a constant flow. To our knowledge, this methodology has never been used to examine the pulmonary vascular response in the intact lamb model.

Bosentan is a nonpeptide mixed antagonist of ET-receptors, obtained by structural optimization of the less potent ETA/ETB antagonist Ro 46-2005. Bosentan competitively antagonizes the specific binding of ET-1 on human smooth muscle cells (i.e. ETA receptors) and it has been demonstrated, in porcine trachea, that it also inhibits the binding of selective ETB ligands. Contractions induced by ET-1 in isolated rat aorta (via ETA receptor stimulation) and by the selective ETB agonist sarafotoxin S6C in rat trachea, were competitively inhibited by bosentan. In vivo, bosentan inhibits the pressor response to big ET-1 both after intravenous and oral administration, and this with a long duration of action and no intrinsic agonist activity (33).

The simultaneous inhibition of ETA and ETB receptors shows that the net effect of ET in the intact anesthetized perinatal lamb results in pulmonary vasoconstriction. Coe et al. (34,35) reported that specific ETA antagonists reversed the pulmonary hypertension when induced by hypoxia in conscious perinatal lambs, whereas, contrary to our findings, these compounds had no effect on pulmonary vascular tone in normoxia. Whether additional ETB blockade by bosentan is responsible for the pulmonary vasodilation in hyperoxia is speculative, but could suggest a pulmonary vasoconstrictor effect of ETB receptors in resting conditions. It was shown that ET-induced vascular contraction is mediated by ETA and ETB receptors located on vascular smooth muscle cells (36).

When endogenous NO production was inhibited by L-NA, the pulmonary vascular tone increased in hyperoxia and even more in hypoxia, which confirms the pulmonary vasodilating action of NO. In adult animals, we previously observed that L-NA increased HPV but did not affect hyperoxic Ppa (24). This study suggests that NO mediates resting pulmonary vascular tone in intact anesthetized newborn lambs and that NO is further released during hypoxia. Thus, it appears that pulmonary vascular tone in perinatal lambs is the result of the balanced influences of NO and ET. However, other vasoconstrictors such as serotonin, angiotensin II, prostaglandin D2, and vasodilators such as prostacyclin and endothelium-derived hyperpolarizing factor may also play a role in regulating pulmonary vascular tone (37).

Bosentan did not reverse the L-NA-induced pulmonary hypertension, although it still reduced Ppa during hypoxia when given after L-NA. Perhaps a higher dosage of bosentan would be necessary.

Coe et al. (34) reported that selective ETA antagonists reversed pulmonary hypertension when induced by hypoxia, but not when elicited by an infusion of a thromboxane A2 analog. These observations suggest that ET antagonists could selectively reduce pulmonary hypertension if due to hypoxia, but not if due to other stimuli.

HPV is associated with the release of ET-1, suggesting a role for ET-1 in HPV (19,20). However, in several studies, the role of ET in HPV could not be demonstrated after administration of selective or non selective ET antagonists (14,16,28,29,38–40).

Our results are in agreement with the idea that ET do not play a major role in HPV in the perinatal lamb, even when it is enhanced by NO synthase inhibition.

The present study suggests that ET are involved in the control of pulmonary vascular tone in hyperoxia and hypoxia, and that their effects are not limited to a NO-dependent mechanism. This conclusion is limited to the acute setting of anesthetized lambs. However, we would like to stress on the use of an adequate methodology, i.e. multipoint pressure-flow relationships, in evaluating the functional status of the pulmonary circulation, which, without any doubt, adds to the scientific weight of our findings.