INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MAINTENANCE OF A BALANCE BETWEEN contractile and relaxant responses is critical for adequate airway conductivity. There is considerable interest in the role of nonadrenergic noncholinergic neural inhibitory (iNANC) mechanisms that mediate airway relaxant responses and the role of endogenously released nitric oxide (NO) in modulating airway smooth muscle relaxation (7, 9, 18, 19, 27, 34). This may be especially important during early development because available data obtained under in vitro conditions indicate that functional NO-mediated airway relaxant responses are enhanced at that age (17). It has been proposed that different environmental factors, such as hyperoxic stress, viral infection, and allergen sensitization, may enhance airway reactivity by impairing this NO/cGMP inhibitory pathway (2, 3, 16, 30).

Lung resistance (RL) can be partitioned into airway (Raw) and tissue (Rti) components, and the latter is thought to reflect the responses of most distal airway and parenchymal contractile elements (10). The contribution of each of these components to RL varies with age. Dreshaj and colleagues (4, 5) showed an increase in the contribution of Rti to RL with advancing maturation in the piglet. However, early in life (1-3 wk of age), the contractile responses to exogenous tachykinin peptides and histamine were relatively greater in the tissue than airway component of RL. Potter et al. (29) showed that blockade of endogenous NO production in the piglet [by using N-nitro-L-arginine methyl ester (L-NAME)] induced an increase in baseline tissue resistance, an effect that was reversed by addition of inhaled NO.

Autonomic mechanisms appear to be involved in control of peripheral airway function. By employing the retrograde catheter technique and alveolar capsule method (24, 28), it has been shown that the most peripheral airways and contractile elements within the lung parenchyma, as well as central airways, respond to changes in parasympathetic output. However, little is known about the contribution of endogenous NO in modulating cholinergic outflow to the most distal parts of the lung. We, therefore, aimed to study the role of endogenously released NO in modulating the response of airways and lung tissue to cholinergically mediated constrictor responses induced by vagal stimulation during early postnatal life. We hypothesized that the increases in both Rti and Raw, in response to vagal stimulation, are opposed by release of NO and activation of the NO/cGMP system. To test this hypothesis, we measured the response of RL and its components Rti and Raw, as well as release of cGMP, to vagal stimulation before and after NO synthase (NOS) blockade in the piglet.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All procedures were approved by the Case Western Reserve University Institutional Review Board for Animal Studies. Experiments were performed on 3- to 10-day-old piglets of either sex with a body weight of 2.4 ± 0.7 kg. Piglets were initially sedated by use of a mixture of intramuscular ketamine hydrochloride (7 mg/kg) and xylazine (1.4 mg/kg) and then anesthetized with intravenous thiopental (25-30 mg/kg). After a high cervical tracheostomy, animals were mechanically ventilated through a tightly fitted endotracheal tube connected to a volume ventilator (Harvard Apparatus, model 55-0798) that delivered a cycled volume of 10 ml/kg of 100% oxygen. The rate of the ventilator was adjusted to maintain blood-gas parameters with a pH of 7.35-7.45 and arterial PCO₂ of 35-45 Torr. During the entire experiment, we maintained a positive end-expiratory pressure (PEEP) of 3 cmH₂O. In addition, the lungs were inflated every 10 min by occluding the expiratory line of the ventilator for three consecutive volume cycles to prevent atelectasis.

A femoral artery was cannulated for continuous blood pressure monitoring and blood-gas sampling. Blood-gas tensions and pH were intermittently determined with an automated blood-gas analyzer (ABL3, Radiometer, Copenhagen, Denmark). A venous line was placed in a jugular vein for administration of fluid or drugs. Body temperature was maintained between 37.5 and 38.5°C by a heating pad. To decrease the confounding effects of anesthesia on the measured physiological parameters, animals were decerebrated. Our laboratory has previously shown (14) that decerebration does not influence cholinergic outflow to airway smooth muscle. As in our group's previous studies (6), a midline skin incision was made over the interparietal bone, and bone over that area was removed. After the dura was cut, a thin spatula was inserted at an angle of ~70° until it touched the skull base and then swung gently left and right completing decerebration at the midcollicular level.

A midline sternotomy was then performed, and the animal's chest was widely retracted. Two lightweight round base capsules (10-mm diameter) were glued with cyanoacrylate to the pleural diaphragmatic surface of both lungs as previously described by our group (4, 29). The pleura under each capsule were punctured four to five times by using a 25-gauge needle to bring the underlying alveoli into communication with the capsule chamber. The tracheal flow signal was obtained with a pediatric Fleisch pneumotachograph connected to a differential pressure transducer (Validyne, Northridge, CA) and was electrically integrated to derive volume. Both vagus nerves were exposed in the upper neck region. After separation of the nerves from the rest of the carotid sheath, they were severed, and the distal ends of both sides were placed on bipolar stimulating electrodes.

The pressure changes in the capsules and in the tracheal side tap were measured with miniature piezoresistive pressure transducers (Endevco, San Juan Capistrano, CA). Resistances calculated from the two capsules exhibited minimal differences and were averaged for each challenge. To detect possible capsule malfunction rapidly, alveolar pressures were displayed against tidal volume on a Tektronix storage oscilloscope. Commercially available software (ANADAT) was used to calculate resistances on the basis of adjustment of the equation of motion (21) and was previously employed by our laboratory (16). RL was calculated from tracheal pressure (Ptr), flow (V), and volume (V)
where K is a constant term reflecting PEEP and the error linked to the residuals of the least square adjustment method. EL represents elastance. Rti was calculated by adjustment of the equation of motion to the alveolar pressure (Palv)
Raw was calculated by subtraction
where Rti is the average obtained from both capsules.

While maintaining ventilation, we measured lung mechanics at baseline and during vagal stimulation, which was applied for 10 s at 4, 8, 16, 32, and 64 Hz. Stimulation was performed with a constant current (10 mA) and variable frequency by using a stimulator (Grass Instruments, model S11) and isolation unit (Grass Instruments, model PSSIU6). In nine animals, the NOS inhibitor L-NAME (10 mg/kg) was infused over 10 min, and vagal stimulation and lung mechanics measurements were repeated, beginning 20 min after L-NAME infusion. This protocol was consistent with our laboratory's prior in vivo studies for L-NAME infusion in rat pups (16), although we employed a smaller dose in this study to minimize nonspecific effects. It is, therefore, possible that NOS blockade was not entirely complete, although this should not have affected our overall conclusions. No experiments were performed with more selective NOS isoform blockers. In an additional six animals, we measured the responses to vagal stimulation of infusing D-NAME (10 mg/kg), the inactive stereoisomer of L-NAME.

Because the physiological consequences of NO release are associated with production of cGMP, we also sought to determine whether vagal stimulation performed before and after L-NAME administration affected cGMP levels in lung tissue. Six additional piglets in which physiological studies were not done were ventilated, their chests were opened, and they were decerebrated as in the earlier experiments. Lung tissue samples (2-4 g) were obtained from the distal part of a lower lobe at baseline and 30 s after the previously described vagal stimulation at 32 Hz. L-NAME (10 mg/kg) was then infused over 10 min as above, and lung tissue was again sampled before and after vagal stimulation. Lung tissue samples were utilized for cGMP analysis as previously described (30). One volume of tissue was mixed with 10 volumes of 1.07 N perchloric acid, homogenized, and centrifuged at 14,000 g for 10 min, and the supernatant was collected for cGMP assay. cGMP assays were performed in duplicate with a commercially available kit (Immunotech, Westbrook, ME) by use of a rabbit polyclonal cGMP antibody. The values of cGMP were expressed as femtomoles cGMP per milligram of protein.

To determine the role of cholinergic vs. noncholinergic mechanisms for NO release, seven additional piglets were treated with intravenous atropine (1 mg/kg), after which histamine was infused (10³ M solution) at an infusion rate adjusted to increase RL to 75-100% of baseline. Preconstriction enhanced the likelihood of inducing a relaxant response, and we sought to approximate the magnitude of the maximal response to vagal stimulation. Vagal stimulation was then performed, and changes in RL, Rti, and Raw were again measured. This protocol allowed us to determine whether vagal stimulation of airways preconstricted via a noncholinergic mechanism would be associated with release of NO from NOS-containing vagal preganglionic neurons in the presence of muscarinic receptor blockade. For continuous observation of changes, and for "real time" calculation of resistance in these animals, different software was used that calculates the resistance on the basis of the isovolumic method. The results obtained by this method were comparable to the results obtained by simultaneous use of ANADAT software (adjustment of the equation of motion).

Statistics. One- and two-way ANOVA with repeated measures was used to compare changes in resistance in response to vagal stimulation before and after administration of L-NAME and D-NAME. Paired t-test was used to compare changes in baseline values of resistance in response to L-NAME and responses of cGMP to vagal stimulation. A value of P < 0.05 was considered statistically significant. Data are presented as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Under baseline conditions, before vagal stimulation, Rti and Raw represented 66 ± 3 and 34 ± 3% of RL, respectively, as seen in the table. Vagal stimulation caused a significant increase in RL, reflecting an increase in both Rti and Raw (Table 1). At peak contractile response corresponding to stimulation with 32 Hz, Rti and Raw represented 59 ± 2 and 41 ± 2% of RL, respectively. These relative contributions of Rti and Raw to RL were significantly different between baseline and stimulated conditions, P < 0.05. 

Blockade of NOS induced a significant increase in baseline Rti without significant change in baseline RL or Raw, although quantitatively the responses of Rti and Raw were comparable. In these nine animals, L-NAME significantly augmented the responses of RL, Rti, and Raw to vagal stimulation (Fig. 1). Representative data for pressure changes measured in the trachea and in both capsules, in response to vagal stimulation, are presented in Fig. 2. In six additional animals, D-NAME infusion had no measurable effect on baseline Rti, Raw, and RL, or changes in Rti, Raw, and RL in response to vagal stimulation (Fig. 3).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Changes in tissue (Rti), airway (Raw), and lung (RL) resistance in response to vagal stimulation at constant current and increasing frequencies (n = 9). Administration of N-nitro-L-arginine methyl ester (L-NAME) significantly increased the responses of all 3 resistance parameters to vagal stimulation (P < 0.05, based on 2-way ANOVA).

View larger version (51K): [in this window] [in a new window]   Fig. 2.   An example of the responses of tracheal (Ptr) and alveolar (Palv) pressures from 2 capsules to vagal stimulation (10 mA, 32 Hz, 10 s) before and after L-NAME. The horizontal lines indicate duration of vagal stimulation. Administration of L-NAME caused elevation in both tracheal and alveolar baseline pressure. Mean arterial blood pressure (ABP) also consistently increased with administration of L-NAME.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Changes in Rti, Raw, and RL in response to vagal stimulation (n = 6). There was no measurable effect of D-NAME administration on these responses.

Vagal stimulation resulted in an increase in levels of cGMP from a baseline of 35.0 ± 7.5 to 45.8 ± 9.6 fmol/mg protein (P < 0.005). After L-NAME infusion, there was no significant change in baseline cGMP and no significant effect of vagal stimulation on cGMP levels (Fig. 4).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Mean responses of cGMP to vagal stimulation in the absence and presence of infused L-NAME in 6 piglets. Before NOS blockade, cGMP levels increased in each animal with a significant mean increase (P < 0.005). After L-NAME infusion, individual responses to vagal stimulation were variable and the mean effect did not approach significance (P = 0.33).

We employed histamine in the presence of atropine to test the role of cholinergic and noncholinergic mechanisms on iNANC responses of lung and tissue resistance in the preconstricted state. Histamine infusion (n = 7) increased RL from 13.3 ± 1.1 to 22.4 ± 1.3 cmH₂O · l¹ · s, Rti from 7.8 ± 0.2 to 12.8 ± 0.4 cmH₂O · l¹ · s, and Raw from 5.5 ± 0.8 to 10.5 ± 1.2 cmH₂O · l¹ · s. After histamine infusion, and in the presence of cholinergic blockade with atropine, vagal stimulation failed to elicit any further changes in RL, Rti, and Raw in these animals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our laboratory has previously shown (29) that piglets exhibit an increase in baseline RL and Rti after NOS blockade, suggesting a role for endogenous NO in modulating baseline lung mechanics at the tissue level. In the present study, we employed vagal stimulation to elicit a release of neurotransmitters that affect Raw and Rti and evaluated the role of endogenously released NO in opposing the contractile effect of endogenous acetylcholine. We have now shown under in vivo conditions that the endogenous NO/cGMP pathway plays a significant role in modulating excitatory cholinergic outflow to both airways and lung parenchyma in piglets.

The three isoforms of NOS (nNOS, eNOS, iNOS) are developmentally regulated, are abundant in late fetal life, and are thought to play an important role in the transition from fetal to neonatal life despite considerable interspecies variability (1, 11-13, 32, 35). Jakupaj et al. (17) demonstrated the importance of airway epithelium as a source of endogenous NO modulating airway caliber under in vitro conditions. After removal of tracheal epithelium, NOS blockade no longer enhanced tracheal contractile response induced by cholinergic stimulation. This is consistent with findings by Sherman et al. (32), who showed that all three isoforms of NOS are expressed in respiratory epithelium of developing sheep. They noted a differential expression of iNOS and nNOS (rather than eNOS) in distal bronchiolar and alveolar epithelium, although our present study does not allow us to speculate which NOS isoforms mediate the airway and tissue effects we have observed, in part, because of the nonspecific effects of L-NAME. Recent data employing a mature mouse model of airway responsiveness suggest that eNOS primarily modulates the increase in resistance induced by a cholinergic agonist (8). Nonetheless, we speculate that airway and alveolar epithelium are the sites for endogenous NO/cGMP elicited by vagal stimulation.

NO activates soluble guanylate cyclase in smooth muscle cells leading to the synthesis of cGMP, and resultant downstream events result in airway (or vascular) smooth muscle relaxation (11). Analysis of cGMP levels in lung tissue provided direct support for our physiological studies (employing NOS blockade) of a role for endogenously released NO in modulating airway smooth muscle relaxant responses in the lung. This protocol clearly did not allow us to identify the precise location of the cGMP released in lung tissue in response to vagal stimulation. We speculate, however, that cholinergic stimulation and activation of muscarinic receptors resulted in activation of NOS in airway epithelial cells and diffusion of NO into airway smooth muscle cells.

Innervation of the tracheobronchial tree is provided by extrinsic sources such as the vagus nerve, nerves from the sympathetic chain, and fibers from sensory ganglia, as well as intrinsic sources such as local intramural ganglia. The presynaptic component of most ganglia consists of vesicle-laden nerve terminals derived from vagal preganglionic neurons within the medulla oblongata (15). Postganglionic nerve fibers, some of which contain NOS (36), are found in the vicinity of epithelial cells, airway smooth muscle, mucosal glands, and blood vessels. Nerve terminal varicosities are at variable distances from effector cells (20 nm to 2 µm); hence, neurotransmitter release and transmission could occur via the intercellular space to receptor sites of effector cells (volume transmission), causing parallel changes in the airway circulation, airway secretion, and smooth muscle tone, as recently described (14). Hence, we believe that release of acetylcholine causes stimulation of muscarinic receptors and activation of NOS in epithelial cells with resultant release of NO, blockade of which leads to increase in airway smooth muscle tone and parenchymal constriction. This is consistent with the concept that NO works as a functional antagonist of acetylcholine (34).

In our study, vagal stimulation in the preconstricted state induced by histamine in atropine-treated piglets failed to elicit any changes in Raw or Rti. These findings demonstrate that stimulation of vagal fibers does not induce relaxation of nonspecifically constricted airway smooth muscle in the presence of cholinergic blockade. Therefore, NO release is linked to release of acetylcholine and muscarinic receptor stimulation. Our results indicate that the release of acetylcholine that causes smooth muscle contraction also activates NOS in the airway effector system, inducing release of NO that modulates the constrictor effect of acetylcholine. Although we cannot determine the site of acetylcholine-induced NO release, our data do not support the contention that stimulation of the vagus nerve elicits release of NO from NOS-containing vagal preganglionic nerve elements, given that no measurable airway smooth muscle relaxation was observed. This is in agreement with earlier findings that there is no evidence of any existing iNANC pathway in the mature pig (22).

Impairment in the inhibitory function of NO, an iNANC system mediator, may correlate with hyperreactive airway disease. Mehta et al. (25) showed that the level of exhaled NO increased transiently with histamine-induced bronchoconstriction in normal and antigen-challenged animals. However, when NOS was blocked with L-NAME, bronchial responsiveness to histamine was enhanced only in normal animals, suggesting that the NO inhibitory pathway becomes weaker after antigen sensitization. Impairment in the iNANC-NO pathway has been also observed in the airway hyperresponsiveness of allergen-sensitized rabbits and rats exposed to hyperoxia or ferrets infected with respiratory syncytial virus (2, 3, 16, 26), although the relevance of these observations to the developing human respiratory system remain uncertain.

Rti is an important component of RL during early postnatal life. In sheep, Perez-Fontán and Kinloch (28) showed that resting cholinergic outflow is preferentially distributed to the peripheral airway even at early age. The present data demonstrate that Rti contributes more to RL than does Raw during early life in the pig at both baseline and during cholinergic stimulation. Vagal stimulation significantly increased both Rti and Raw, with a greater effect on Raw. We believe that the use of alveolar capsules is an accurate method to partition RL into tissue and airway components. In this study, the contribution of Rti and Raw under baseline conditions was similar to that previously described for the newborn piglet (5, 29). In addition, resistances measured at both capsules were equal, reflecting homogenous distribution of ventilation and parallel changes throughout each experiment. The use of a constant PEEP of 3 cmH₂O decreases the confounding effects of lung volume history on Rti (23, 31). Romero et al. (31) suggested that changes in peripheral resistance, in response to cholinergic stimulation, were not affected by changes in the larger airway. This is consistent with our study, which showed that L-NAME at baseline significantly increased Rti but not Raw, although the responses were comparable in magnitude. We speculate that distortion in larger airways is probably not the mechanism for the changes seen in Rti, which may result from changes in the property of lung tissue contractile elements such as preductal smooth muscle cells and alveolar myofibroblasts. Although these structures have been reported in human and rat lungs, we acknowledge that it is unclear whether piglet lungs possess these elements (20).

Although L-NAME has been shown to increase both pulmonary and systemic pressure and, therefore, decrease pulmonary blood flow (29), it is not known how this would affect Raw or Rti. In piglets, Uhlig et al. (33) showed that vascular congestion increased rather than decreased both baseline airway and tissue resistance. Furthermore, congestion enhanced only the airway response to low-dose methacholine with minimal effect on peripheral lung tissue responsiveness to the same drug. This suggests that the response seen with L-NAME is not related to changes in pulmonary blood flow or volume but reflects a direct effect on contractile elements of lung parenchyma.

This study demonstrates the importance of endogenous NO in modulating cholinergic regulation of lung mechanics at both the tissue and the airway level during early life. Our data support the concept that endogenously released NO may dampen constrictor responses of peripheral, as well as central, lung units in early life. Although the importance of this mechanism in inhibiting airway narrowing under pathophysiological conditions in human infants remains to be defined, greater understanding of the mechanisms that downregulate or impair this system should contribute to our knowledge of the pathophysiology of lung disease at that age.