INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

RECENT FINDINGS from studies done in this laboratory employing isotope dilution methodology in chronically instrumented fetal lambs before and after delivery have provided new insights into the kinetics of the catecholamines norepinephrine and epinephrine in the immediate newborn period. Thus whole body studies indicated that about one-half of the birth-related surge in circulating catecholamines is related to increased norepinephrine and epinephrine release into the circulation associated with enhanced sympathoadrenal activity and the remainder to reduced catecholamine clearance from the circulation accompanying loss of the placenta (36). Subsequent pulmonary studies suggested that the lungs make a substantial contribution to whole body catecholamine kinetics after birth, accounting for approximately one-third of total body norepinephrine and one-tenth of total body epinephrine clearance from the circulation, as well as up to two-fifths of total body norepinephrine and up to one-fifth of total body epinephrine release into the circulation (37).

In the adult, considerable information is available about factors that modulate norepinephrine and epinephrine clearance and release (10, 12-21, 28, 29, 40, 41) and this has facilitated increased understanding of disturbances in sympathoadrenal function accompanying disease processes (17). By contrast, relatively little is known about these factors in the setting of the incompletely developed sympathetic innervation characteristic of the immediate newborn period (27). In particular, it is unknown to what extent pulmonary and total body norepinephrine and epinephrine removal after birth is affected by the tricyclic antidrepressant desipramine, which is a potent inhibitor of catecholamine uptake into sympathetic neurons (10, 15, 24) as well as the neuronal-like catecholamine uptake process located within the microvasculature of the lungs (4, 14). In addition, it is unknown if desipramine affects pulmonary or total body norepinephrine and epinephrine release into the circulation after birth, a question that is of particular relevance not only because desipramine has central sympathoinhibitory effects (15, 41) but also because birth is accompanied by increased central sympathoadrenal outflow (33). Furthermore, it is unclear to what extent pulmonary changes contribute to the whole body effects of desipramine on norepinephrine and epinephrine kinetics in the newborn.

Accordingly, the aim of this study was to examine the actions of desipramine on pulmonary and total body norepinephrine and epinephrine kinetics in the immediate newborn period. Experiments were performed in conscious, chronically instrumented near-term lambs 4 h after cesarean section delivery, a time point characterized by the presence of substantial pulmonary uptake and release of both norepinephrine and epinephrine (37). Pulmonary and total body norepinephrine and epinephrine kinetics were determined using a combined tracer infusion of ³H-labeled norepinephrine and epinephrine, while blood flows were measured with radioactive microspheres.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

All experiments were approved by the Monash University Animal Experimentation Committee and were in accord with guidelines established by the National Health and Medical Research Council of Australia.

Animal preparation. Six fetal lambs with known breeding dates were chronically instrumented under aseptic conditions at 133-134 days gestation (term 147 days) as previously described (35-37). Briefly, fasted Border-Leicester cross ewes were anesthetized with propofol (5 mg/kg iv), intubated, and then mechanically ventilated with 1-3% halothane and a 2:1 mixture of nitrous oxide and oxygen. The uterus was exposed through a midline laparotomy and incised over the fetal hindlimbs. Polyvinyl catheters (ID 1 mm, OD 1.5 mm) were inserted into a posterior tibial artery and lateral saphenous vein and advanced into the abdominal aorta and inferior vena cava, respectively. After delivery of the fetal head, left forelimb, and upper thorax through a second hysterotomy, a thoracotomy was performed in the third left interspace and the pericardium was opened over the pulmonary trunk and left atrium. A Teflon cannula connected to a polyvinyl catheter was inserted into the distal part of the pulmonary trunk, and a polyvinyl catheter was introduced into the left atrial cavity through a purse-string suture. The pericardium was then loosely closed, the ribs reapposed, and overlying muscle layers repaired. After ventral incision of the neck in the midline, a Teflon cannula attached to a polyvinyl catheter was inserted nonocclusively into the left carotid artery and a polyvinyl catheter was passed into the superior vena cava via the left external jugular vein. Both catheters were tunneled subcutaneously to the chest incision. A Silastic catheter (ID 0.8 mm, OD 1.7 mm) was inserted into the trachea and exteriorized through the cephalic end of the neck incision in all fetuses for later withdrawal of lung liquid. Finally, a wide-bore catheter was sutured to the anterior chest wall for measurement of amniotic fluid pressure. The fetus was then returned to the uterus, and all incisions were closed. The vascular catheters were filled with sodium heparin solution (1,000 IU/ml) and exteriorized on the right flank of the ewe. After surgery, vascular catheters were flushed every second day and refilled with sodium heparin. Antibiotics (500 mg streptomycin and 5 × 10⁶ units penicillin) were instilled into the amniotic cavity perioperatively and then administered daily, either as an intramuscular injection to the ewe or directly into the amniotic cavity when catheters were flushed.

Experimental protocol. Seven days after surgery (i.e., at a gestation of 140-141 days), low spinal anesthesia was induced in the ewe with an intrathecal injection of 3-5 ml of 0.5% bupivacaine. After withdrawal of approximately 40 ml of lung liquid via the tracheal catheter to facilitate the rapid establishment of pulmonary gas exchange after birth (3), the fetus was quickly delivered by cesarean section, the tracheal catheter was removed, and the umbilical cord was clamped and cut. The ewe was killed immediately after cesarean section delivery with an intravenous overdose of pentobarbital sodium. All lambs breathed spontaneously and rapidly established a rhythmic breathing pattern.

The baseline catecholamine kinetics study was performed in newborn lambs 4 h after cord clamping, after the commencement of a continuous and constant-rate infusion of [³H]norepinephrine and [³H]epinephrine via the hindlimb catheter 30 min beforehand. At baseline, hemodynamics were recorded; blood samples were taken for hematocrit, blood gas, and catecholamine analysis; and left ventricular (LV) output was measured with radioactive microspheres (22). Blood withdrawn during blood sampling and injection of radioactive microspheres was replaced with newborn lamb blood mixed 1:1 with a plasma substitute (Haemaccel, Behring, Marburg, Germany). Catecholamine neuronal uptake was then blocked with 2 mg/kg iv desipramine hydrochloride (Sigma, St. Louis, MO), a dose that results in >90% inhibition of whole body norepinephrine neuronal reuptake (15). With the infusion of ³H-labeled tracer continuing, measurements were repeated 30 min later.

Physiological measurements. Abdominal aortic and pulmonary arterial blood pressures were monitored with strain-gauge pressure transducers (model 1280B, Hewlett-Packard, Waltham, MA), which were calibrated against a water manometer before each experiment. Vascular pressures were referenced to atmospheric pressure at the midchest position. Mean vascular pressures were obtained electronically while heart rate was measured with a tachometer triggered by an arterial pulse. All signals were displayed on an eight-channel paper recorder (model 800Z, Neomedix Systems, Sydney, NSW, Australia).

Blood pH, PO₂, and PCO₂ were obtained at the measured rectal temperature with a blood analyzer (model 168, Corning Medical, Halstead, Essex, UK). Blood hemoglobin concentration and hemoglobin oxygen saturation were measured in duplicate with a hemoximeter (model OSM2, Radiometer, Copenhagen, Denmark).

Radiotracer technique. Stock solutions of radiolabeled norepinephrine (levo-[³H-2,5,6]norepinephrine) and epinephrine (levo-N-methyl-[³H]epinephrine; New England Nuclear, Boston, MA) were dissolved in 0.2 M acetic acid containing 1 mg/ml ascorbate and stored at 80°C to minimize degradation (11). The stock solutions were thawed immediately before the study, and a 1-ml aliquot of each radiotracer was added to 40 ml of 0.9% sodium chloride. To simultaneously measure norepinephrine and epinephrine kinetics (30), the combined radiotracers were then infused intravenously into the lambs using a syringe pump set at a rate of 0.18 ml/min, which corresponded to an infusion rate of 47.3 ± 3.4 nCi · kg¹ · min¹ for norepinephrine and 57.3 ± 5.8 nCi · kg¹ · min¹ for epinephrine. A sample of infusate was stored at 80°C for subsequent assay of norepinephrine and epinephrine content. Withdrawn blood samples were immediately transferred to tubes containing EDTA and, after centrifugation, the plasma fraction was stored in a 80°C freezer until assay.

Endogenous and tritiated catecholamines were extracted from 1-ml plasma samples with the use of alumina adsorption and separated with HPLC as previously described (10-14, 18, 30, 35-37). Concentrations of total catecholamines in plasma and 10-µl infusate samples were quantified by electrochemical detection, whereas ³H-labeled catecholamines were measured by liquid scintillation spectroscopy of the eluted fractions leaving the electrochemical cell. The recovery of dihydrobenzylamine, the internal standard used for the HPLC assay, was 90 ± 1%, compared with 88 ± 1% for norepinephrine and 87 ± 1% for epinephrine. The within-assay coefficient of variation was 1.7 ± 0.5% for norepinephrine and 3.1 ± 0.9% for epinephrine. A representative chromatogram is shown in Fig. 1. Endogenous levels of norepinephrine and epinephrine were not corrected for the contribution of exogenous ³H-labeled catecholamines, because the latter comprised <1% of circulating norepinephrine and epinephrine levels both before and after desipramine.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Illustrative chromatogram obtained by liquid chromatographic separation and electrochemical detection of arterial plasma in 4-h-old newborn lamb. Horizontal line depicts time scale (min), and vertical axis depicts electrochemical response. Peaks and plasma concentrations (pg/ml) corresponding to dihydroxyphenylglycol (DHPG), norepinephrine (NE), dihydroxyphenylalanine (DOPA), epinephrine (Epi), and dihydroxybenzylamine (DHBA) are identified.

Radioactive microsphere technique. Radioactive microspheres, 15 µm in diameter and labeled with one of five gamma-emitting isotopes (¹⁴¹Ce, ¹¹³Sn, ⁸⁵Sr, ⁹⁵Nb, or ⁴⁶Sc, New England Nuclear) were ultrasonicated for 10-15 min before injection and then injected over 30-45 s with 10 ml isotonic saline. About 0.5 × 10⁶ microspheres were injected into the left atrium while reference samples were obtained from the carotid artery and the abdominal aorta. All reference samples were drawn at a rate of 4.1 ml/min with a mechanical pump (model 901A, Harvard Apparatus). Reference sample collection was commenced 5-10 s before injection and continued for an additional 75 s after the completion of the injection.

At the end of the experiment, lambs were killed with an intravenous overdose of pentobarbital sodium and, at postmortem examination, the position of all catheters was carefully checked and the ductus arteriosus was confirmed to be markedly constricted. The lungs were placed in Formalin fixative for 7-10 days and then carbonized at a temperature of 280°C in a vented box furnace. The carbonized tissue was ground into a coarse powder and packed into plastic counting vials to a height of 2 cm. The radioactivity of the blood reference samples and the tissue vials was counted in a gamma counter (model 1282 CompuGamma, LKB-Wallac, Turku, Finland) at the appropriate window settings, and the photopeaks of individual isotopes were separated by an online computer program.

Calculation of blood flows. Radioactive microsphere measurements of tissue blood flow were calculated using the general relation Q_Tissue = [Q_Reference · R_Tissue]/R_Reference, where Q is flow (ml/min) and R is radioactivity (counts/min). With the use of the same relation, newborn LV output (Q_LV), which was assumed to be equivalent to total pulmonary blood flow, was calculated as [Q_Reference · R_LA]/R_LAA, where R_LA is the radioactivity of the label injected into the left atrial cavity and R_LAA is the average radioactivity of the carotid arterial and abdominal aortic reference samples (22, 37). The systemic contribution to pulmonary blood flow, which was derived from a combination of bronchial arterial inflow and any shunting from the aorta into the pulmonary circuit across an incompletely constricted ductus arteriosus, as well as shunting across peripheral arteriovenous anastomoses, was calculated as [Q_Reference · R_L]/R_LAA, where R_L is lung radioactivity (37).

Calculation of catecholamine clearance and spillover. Total body plasma catecholamine clearance and spillover rates were obtained with previously described formulas (10-18). Thus total body plasma clearance of catecholamines (TBCl) was calculated as IR/([³H]Cat_A · body wt), where IR is the infusion rate of ³H-labeled catecholamine (dpm/min), [³H]Cat_A is the steady-state mean systemic arterial plasma concentration of ³H-labeled catecholamine (dpm/ml), and body weight is in kilograms. To obtain an accurate estimate of total body clearance, the infusion rate of ³H-labeled catecholamine was corrected for pulmonary catecholamine extraction (19, 21). The total body fractional extraction (TBFx) of either catecholamine, i.e., the proportion extracted on first passage through the circulation, was computed as TBCl/[Q_LV · (1  Hct)], where Hct is the hematocrit. The total body spillover rate of catecholamines into plasma was equivalent to TBCl · Cat_A, where Cat_A is the arterial plasma concentration of norepinephrine or epinephrine (pg/ml).

The pulmonary fractional extraction of tritiated catecholamines (PulFx) was calculated as ([³H]Cat_PA  [³H]Cat_A)/[³H]Cat_PA, the contribution of pulmonary to total body catecholamine clearance as PulFx/TBFx, and the spillover of norepinephrine or epinephrine from the lungs into the systemic circulation as [(PulFx · Cat_PA) + (Cat_A  Cat_PA)] · [Q_LV · (1  Hct)]/body wt (14, 37).

Statistics. Physiological and catecholamine variables before and after desipramine were compared with repeated measures one-way analysis of variance, whereas norepinephrine and epinephrine kinetics were compared with the Student's t-test (39). Results are reported as means ± SE, and P < 0.05 was considered significant.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Hemodynamics, blood gas variables, and blood flows. Administration of desipramine was not associated with any significant changes in hemodynamics, systemic arterial blood gas variables, LV output, or the level of pulmonary blood flow derived from systemic sources (Table 1).

Endogenous and tritiated catecholamine plasma concentrations. Baseline endogenous systemic and pulmonary concentrations of either norepinephrine or epinephrine were not statistically different from one another, and no concentration changed significantly after desipramine (Table 2). By contrast, under baseline conditions, the [³H]norepinephrine concentration in the pulmonary trunk exceeded that of the aorta by 163 ± 37 dpm/ml (P < 0.01), whereas the corresponding difference for [³H]epinephrine was 86 ± 9 dpm/ml (P < 0.001). Desipramine increased the pulmonary arterial and aortic concentrations of [³H]norepinephrine (both P < 0.001) and [³H]epinephrine (both P < 0.005) but abolished transpulmonary [³H]norepinephrine and [³H]epinephrine differences (Table 2).

Total body catecholamine clearance, fractional extraction, and spillover. Norepinephrine total body clearance decreased from 93 ± 8 to 46 ± 4 ml · min¹ · kg¹ after desipramine (P < 0.001), and this was accompanied by a fall in the [³H]norepinephrine total body fractional extraction from 0.458 ± 0.057 to 0.215 ± 0.022 (P < 0.005; Fig. 2). Similarly, desipramine reduced epinephrine total body clearance from 82 ± 4 to 58 ± 6 ml · min¹ · kg¹ (P < 0.01) in association with a reduction in [³H]epinephrine total body fractional extraction from 0.407 ± 0.044 to 0.284 ± 0.040 (P < 0.005; Fig. 2).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Total body norepinephrine and epinephrine clearance (A) and total body [3H]norepinephrine and [3H]epinephrine fractional extraction (B) before (filled bars) and after (open bars) desipramine in newborn lambs.

Comparison of desipramine-sensitive and desipramine-resistant fractional extractions indicated that neuronal uptake contributed 52 ± 3% of total body [³H]norepinephrine but only 31 ± 5% of total body [³H]epinephrine removal (P < 0.01). Furthermore, comparison of the reductions in the fraction extraction of [³H]norepinephrine (0.243 ± 0.042) and [³H]epinephrine (0.123 ± 0.022) indicated that total body desipramine-sensitive uptake of norepinephrine was twice as efficient as for epinephrine (P < 0.01).

In addition to the reduction in catecholamine clearance, desipramine decreased norepinephrine total body spillover from 188 ± 29 to 82 ± 16 ng · min¹ · kg¹ (P < 0.005) and epinephrine total body spillover from 22.7 ± 2.7 to 13.8 ± 1.6 ng · min¹ · kg¹ (P < 0.05; Fig 3).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Total body norepinephrine and epinephrine spillover before (filled bars) and after (open bars) desipramine in newborn lambs.

Pulmonary catecholamine clearance and fractional extraction. The baseline pulmonary fractional extraction of [³H]norepinephrine was 0.133 ± 0.026, which constituted 35 ± 10% of total body [³H]norepinephrine extraction, whereas the pulmonary fractional extraction of [³H]epinephrine was 0.057 ± 0.007, which accounted for 15 ± 2% of total body epinephrine clearance. Desipramine reduced the pulmonary fractional extraction of both [³H]norepinephrine (P = 0.005) and [³H]epinephrine (P < 0.02) to undetectable levels (Fig. 4). Moreover, comparison of pre- and postblockade fractional extractions indicated that [³H]norepinephrine was extracted 2.3-fold more efficiently than [³H]epinephrine by desipramine-sensitive removal processes in the lungs.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Pulmonary [3H]norepinephrine and [3H]epinephrine fractional extraction before (filled bars) and after (open bars) desipramine in newborn lambs.

Pulmonary catecholamine spillover. Under baseline conditions, the specific activity of norepinephrine fell by 19.8 ± 5.7% (P < 0.02) and that of epinephrine by 7.2 ± 2.7% (P < 0.05) between the pulmonary trunk and aorta, indicative of a dilution of infused ³H-labeled catecholamines by endogenous catecholamines released from the lungs. However, desipramine reduced the transpulmonary difference in specific activity for norepinephrine (P < 0.02) and epinephrine (P < 0.03) to levels that were not significantly different from zero (Fig. 5).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Transpulmonary change in norepinephrine (A) and epinephrine (B) specific activity before (Pre DMI) and after (Post DMI) desipramine in newborn lambs.

Under baseline conditions, calculated pulmonary norepinephrine spillover was 79 ± 26 ng · min¹ · kg ¹, which constituted 47 ± 13% of total body norepinephrine spillover, whereas pulmonary epinephrine spillover was 3.8 ± 1.0 ng · min¹ · kg¹, which accounted for 19 ± 7% of total body epinephrine spillover. However, desipramine administration was associated with a fall in both pulmonary norepinephrine (P < 0.03) and epinephrine spillovers (P < 0.05) to levels that were not statistically different from zero (Fig. 6).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Pulmonary norepinephrine and epinephrine spillover before (filled bars) and after (open bars) desipramine in newborn lambs.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

This study has employed isotope dilution methodology to examine the effects of desipramine on pulmonary and total body clearance and spillover of norepinephrine and epinephrine in the immediate newborn period. To characterize catecholamine kinetics after birth in the absence of anesthesia and during spontaneous respiration, the study was performed in lambs that had been delivered by cesarean section close to term after adequate recovery from implantation of vascular catheters at prior in utero surgery. Previous work from this laboratory (35-38) has shown that such lambs undergo physiological changes typically associated with birth, including increases in arterial oxygenation, heart rate, systemic blood pressure, pulmonary blood flow, and circulating catecholamines, as well as reductions in pulmonary arterial blood pressure (2, 7, 8, 34). Furthermore, the use of conventional formulas to calculate pulmonary catecholamine clearances and spillovers at the 4 h time point employed in the present study was possible because such lambs have no significant right-to-left or left-to-right transductal shunting detectable on blood gas analysis (38) and left-to-right transductal flow that constitutes only 10% of systemic blood flow (37).

In accord with the higher affinity of norepinephrine as a substrate for neuronal uptake observed in the adult (10, 14, 19), desipramine also produced a greater reduction in norepinephrine total body removal relative to that of epinephrine in newborn lambs (Fig. 2). Moreover, the observation that whole body clearance of circulating [³H]norepinephrine in the present study was reduced by 51%, compared with 20-42% in the adult (10-14, 18, 19, 40), and circulating [³H]epinephrine by 29%, compared with no change (10) or a reduction of 13-19% (10, 14) in the adult, is consistent with a greater dependency of neonatal total body catecholamine clearance on desipramine-sensitive mechanisms. A likely basis for this phenomenon is apparent from examination of catecholamine total body fractional extraction data. Thus, although the total body fractional extraction of [³H]norepinephrine (0.46) and [³H]epinephrine (0.41) observed in newborn lambs was substantially lower than the corresponding ranges of 0.66-0.71 (14, 19, 21) and 0.65-0.72 (14, 21) reported in adult humans and experimental animals, the desipramine-induced fall in total body norepinephrine fractional extraction in newborn lambs (0.245) was similar to the adult range of 0.219-0.254 (14, 19), whereas the reduction in total body epinephrine fractional extraction (0.123) was even greater than the adult range of 0.061-0.086 (14, 19). These results imply that, despite the presence of incompletely developed sympathetic innervation (27), the maturity of catecholamine neuronal uptake processes in the newborn was on a par with that of the adult and that the lower total body catecholamine fractional extraction in the newborn was instead related to an immaturity of extraneuronal transporter mechanisms, a notion that will require formal testing with specific extraneuronal uptake blocking agents such as disprocynium24 (13, 20).

In adult lungs, norepinephrine and epinephrine removal occurs via a transporter situated within the endothelial cells of the pulmonary microvasculature (6, 31). However, despite its location at an extraneuronal site, this transporter is not only inhibited by specific inhibitors of neuronal catecholamine uptake such as desipramine (6, 14) but also has an amino acid sequence identical to the neuronal catecholamine transporter (5). Two findings of the present study point to a close functional similarity between the catecholamine transporter in the newborn and adult lungs. First, in accord with the two- to fourfold difference seen in adult lungs (14, 19), pulmonary norepinephrine clearance in newborn lambs was 2.3-fold more efficient than for epinephrine. Second, as in adult dogs (14), desipramine reduced pulmonary [³H]norepinephrine and [³H]epinephrine extractions in newborn lambs to undetectable levels (Fig. 4). Collectively, these results suggest that, in contrast to the major roles of both neuronal and extraneuronal uptake processes in the systemic circulation (10, 12, 13, 15, 19-21), removal of circulating catecholamines by the lungs occurs primarily via the neuronal-like uptake mechanism located within the pulmonary microvasculature (6, 31). Although the basis for this difference is not fully understood at present, a likely contributory factor is the differing spatial arrangement of uptake mechanisms at these two sites, in that the pulmonary catecholamine transporter resides on the diffusional endothelial surface (31), whereas catecholamines in systemic tissues are exposed to extraneuronal uptake mechanisms en route to the neuronal transporter situated on perivascular and parenchymal sympathetic elements (17).

In addition to its inhibitory effect on neuronal uptake processes, desipramine also acts centrally to reduce sympathetic outflow (15), an action accompanied by a marked reduction in sympathetic nerve activity to organs such as skeletal muscle (18) and the kidney (15, 41). As most of the norepinephrine released by sympathetic nerves is recaptured, an effect of desipramine on neuronal uptake alone would increase the norepinephrine concentration at the neuroeffector junction and thereby result in a greater spillover of norepinephrine into plasma (17). However, norepinephrine spillover in normal adult humans and animals is either unchanged (14, 15, 20, 41) or reduced by 13-36% after desipramine (18, 28, 40), indicating that the two actions of this compound either counteract one another or slightly favor central sympathoinhibition. In contrast, desipramine caused a 56% reduction in the total body spillover of norepinephrine in newborn lambs (Fig. 3), an observation that not only suggests that this agent exerted a predominant central sympathoinhibitory effect in the newborn, but also supports an important role for central mechanisms in the increased sympathetic activation occurring at birth (33).

Inasmuch as the developing lungs contain appreciable amounts of norepinephrine (1) and have functional evidence of sympathetic innervation (9) and given the recognized central sympathoinhibitory effects of desipramine (15, 18, 41), the reduction in norepinephrine spillover from the lungs to undetectable levels by the latter compound implies that pulmonary sympathetic activity in the immediate newborn period was principally related to the presence of a marked central sympathetic drive to the lungs. On present evidence, it is unlikely that norepinephrine arising from the newborn lungs was derived from a source other than sympathetic nerves. Specifically, recent studies in isolated rat lungs indicate that norepinephrine taken up into pulmonary endothelial cells is primarily metabolized to O-methylated compounds by catechol O-methyltransferase (COMT), an enzyme that becomes half saturated at a norepinephrine concentration of 9.8 nmol/l (4). Furthermore, saturation of pulmonary COMT results in the accumulation (4) and subsequent efflux of unchanged norepinephrine from pulmonary endothelial cells (42). However, although our baseline pulmonary arterial norepinephrine concentration of 1,888 pg/ml was equivalent to 11.2 nmol/l, it is unlikely that the foregoing mechanism made a significant contribution to pulmonary norepinephrine release in newborn lambs, because blockade of the pulmonary catecholamine transporter would then have increased norepinephrine efflux (42) and therefore spillover, whereas desipramine reduced norepinephrine spillover to near-zero levels in our study (Fig. 6).

Although desipramine does not alter total body epinephrine release in the adult (14, 20), this compound reduced epinephrine total body release by 39% in newborn lambs of the present study (Fig. 6). As with our norepinephrine data, this suggests that any increased epinephrine release into the circulation related to blockade of the pulmonary catecholamine transporter was outweighed by the central sympathoinhibitory effects of desipramine (15, 41). Furthermore, the finding that about two-fifths of the reduction in total body epinephrine was related to a marked fall in pulmonary epinephrine efflux extends our previous finding that the newborn lungs are a major extra-adrenal site of epinephrine release (37) by implying that such release can be modulated by sympathetic outflow from the central nervous system.

An important issue that is still to be fully resolved relates to the precise origin of the epinephrine released from newborn lungs. The most straightforward explanation, which is consistent with the concurrent reduction in norepinephrine and epinephrine spillovers to undetectable levels after desipramine (Fig. 6), was that epinephrine was coreleased with norepinephrine from pulmonary sympathetic nerve endings. One possible source of this epinephrine was a loading of pulmonary neuronal epinephrine stores via uptake from the circulation (29) occurring in association with the perinatal surge in plasma epinephrine concentration (33). However, the demonstration of the epinephrine-synthesizing enzyme phenylethanolamine N-methyltransferase in the developing lung and the increased activity of this enzyme in the initial days after birth (32) as well as our previous kinetic analysis (37) suggests that epinephrine is also synthesized within newborn lungs. Evidence from adult rats of lung epinephrine synthesis within an extraneuronal site (25) points to the neuroendocrine cell, which occurs in abundance in newborn lungs (23), as a likely candidate for such synthesis. Importantly, extraneuronal synthesis is not necessarily incompatible with neuronal release because data from the heart and skeletal muscle (26) suggest that epinephrine synthesized in an extraneuronal location may be sequestered into sympathetic neurons.

Perspectives