INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: airway smooth muscle; L-N^G-nitro-arginine; nitric oxide synthase; nonadrenergic, noncholinergic relaxant nerves; 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one

Nitric oxide (NO), synthesized from arginine by a family of enzymes collectively termed the NO synthases (NOS), plays an integral role in airway and pulmonary homeostasis (1). Inhibition of NO synthesis increases airways reactivity in animals and humans (2). Although many of the cells producing this gaseous autacoid and transmitter have been identified in the lung (6), the source of the NO that modulates airways reactivity has not been clearly defined. The lack of interventions that can selectively inhibit the activity of specific cell types in the airway wall, along with the lack of highly selective inhibitors for the various isozymes of NOS, contributes to the poor understanding of this aspect of airway physiology.

One source of NO in the lung is the airway parasympathetic nerves (7). Nonadrenergic, noncholinergic (NANC) parasympathetic relaxant nerves are the primary relaxant nerves innervating airway smooth muscle (ASM) in most mammalian species including humans (11). When activated, these nerves can reverse a nearly maximal contraction of isolated tracheal or bronchial smooth-muscle preparations in vitro or can completely dilate a markedly constricted airway in vivo. The relaxations produced by the parasympathetic nerves are thought to be mediated by NO and/or vasoactive intestinal peptide (VIP and related peptides) (9, 12, 13). Their wide distribution and capacity to prevent and/or reverse bronchospasm evoked by many spasmogens makes it reasonable to speculate that the function of these nerves is important in regulating airway caliber (9). Their dysfunction might also contribute to obstructive airways diseases (18, 19).

Previous studies indicate that distinct pre- and postganglionic parasympathetic nerves mediate contractions and relaxations of ASM (20). These cholinergic and noncholinergic parasympathetic nerve pathways are subject to differential reflex activation (24, 25). In a recent study of the parasympathetic innervation of the guinea pig airways, we observed that cholinergic nerves are tonically active, producing marked contractions of the airways that are necessarily dependent on ongoing afferent nerve activity arising from the lungs (26). In the present study, we addressed the hypothesis that airway noncholinergic parasympathetic nerves are similarly active at baseline, thereby blunting the contractile effects of the cholinergic nerves as well as contractions evoked by exogenously administered agonists.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Preparation

All of the study experiments described here were approved by the Johns Hopkins Medical Institutions Animal Care and Use Committee. Male Hartley guinea pigs (weighing 300 to 400 g; Hilltop, Scottdale, PA) were anesthetized with urethane (1 g/kg, intraperitoneally) and positioned ventral side up on a heated pad. The caudalmost portion of the extrathoracic trachea was cannulated and the animals were mechanically ventilated (at 60 breaths/min, 6 ml/kg, at 2 to 3 cm H₂O of positive end-expiratory pressure to preserve airway patency) following induction of paralysis (with 2 mg/kg succinylcholine chloride, subcutaneously).

We measured tracheal smooth-muscle tension in situ as described previously (26). Stainless steel hooks were passed between two cartilage rings (rings 6 and 7 caudal to the larynx) on either side of the trachea, rostral to the tracheal cannula. One hook was sutured to a fixed bar and the other hook was sutured to an isometric force transducer (Model FT03C; Grass Instruments, Quincy, MA). Optimal baseline tension was set (1.5 to 2 g) and maintained throughout the equilibration period. The lumen of the tracheal segment studied was perfused with warmed (37° C), oxygenated Krebs buffer delivered through a small slit made in the ventral trachea, caudal to the hooks. The buffer was removed from the rostralmost end of the trachea with suction.

In all experiments described in this report, the Krebs bicarbonate buffer perfusing the tracheal lumen (composition in mM: NaCl, 118; KCl, 5.4; NaHPO₄, 1; MgSO₄, 1.2; CaCl₂, 1.9; NaHCO₃, 25; dextrose, 11.1) contained 3 µM indomethacin, 2 µM propranolol, and 1 µM phentolamine. These drugs were used to block the local effects of prostaglandins and to block any effects of circulating and neurally released catecholamines on the tracheal segment studied.

The abdominal aorta was cannulated to monitor blood pressure and for drug delivery. Arterial blood pressure (ABP), tracheal insufflation pressure (PT) and tracheal smooth-muscle tension (TT) were recorded on a Grass polygraph (Model 79E).

At the conclusion of each experiment, animals were killed by inhalation of 100% CO₂, delivered through the inspiratory port of the ventilator.

Vagally Mediated Contractions and Relaxations of the Guinea Pig Trachea In Situ

The relaxant responses of the trachealis elicited by electrical stimulation of the vagus nerves were studied in situ. To prevent marked alterations in heart rate and blood pressure evoked by vagus nerve stimulation, the M₂ muscarinic receptor antagonist methoctramine (2 mg/kg) was administered intraarterially at the beginning of all experiments in which the vagus nerves were stimulated. Bilateral vagotomies were performed, and insulated platinum hook electrodes, attached to a Grass stimulator (Model S88), were placed on the peripheral ends of the cut vagus nerves. Optimal stimulus intensities (typically 6 to 15V, 1 ms pulse duration) were determined in each experiment by constructing voltage-response curves as previously described (20). Vagally mediated relaxations were evoked after administration of atropine via the tracheal perfusate and contraction of the trachealis with 10 µM histamine. When the contraction in response to histamine stabilized, the vagus nerves were again stimulated (2 to 24 Hz) until relaxant responses reached equilibrium. Intervals from 3 to 10 min were maintained between each stimulus, depending on the duration of the relaxant response evoked. Alternatively, frequency-response curves were generated cumulatively. The magnitude of the vagally mediated relaxant response was expressed as a percentage reversal of the corresponding histamine-induced contraction.

Effects of NOS and Soluble Guanylate Cyclase Inhibitors on Vagally Mediated Relaxations of the Trachealis and Baseline Cholinergic Tone In Situ

The effects of the NO synthase inhibitor L-N^G-nitro-arginine (L-NNA) (30 µM) on vagally mediated relaxations of the trachealis were studied with a nonpaired experimental design. For relaxant responses, the vagus nerves were stimulated continuously at 16 Hz to equilibrium (as described earlier) in the presence of L-NNA or its vehicle.

The cholinergic nerves innervating the guinea pig trachea are spontaneously active at baseline, producing contractions of the trachealis in situ that are reversible by administering atropine via the tracheal perfusate (26). We studied the effects of L-NNA and 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) on this baseline cholinergic tone with a nonpaired experimental design. After a 20-min equilibration period, L-NNA, ODQ, or vehicle were administered via the tracheal perfusate while baseline cholinergic tone in the trachealis was monitored. After 20 min, or when the effects of the treatments reached equilibrium, atropine was administered to confirm the cholinergic nature of the contractions evoked, as well as the cholinergic nature of the baseline tone. The effects of L-arginine on the responses evoked by L-NNA were assessed in parallel experiments.

Baseline cholinergic tone in the trachealis was expressed as a percentage of the maximum attainable contraction (26). Cholinergic tone was determined by quantifying the magnitude (in grams) of isometric tension reversed by adding the muscarinic receptor antagonist atropine (1 µM) to the buffer perfusing the tracheal lumen, and the maximum attainable contraction was elicited by adding 300 mM barium chloride (BaCl₂) to the tracheal perfusate.

The effects of L-NNA and ODQ on baseline cholinergic tone were expressed as percentage increases in baseline tone. In some experiments, the effects of bilateral vagotomy on the responses to L-NNA and ODQ were assessed.

Modulation of Electrical Field Stimulation-induced Contractions and Relaxations of Guinea Pig Tracheal Strips

The effects of L-NNA and ODQ on electrical field stimulation (EFS)- induced contractions and relaxations of guinea pig tracheal strips were also assessed (21). Tracheal strips were suspended between bipolar platinum ring electrodes in 10-ml organ baths. Contractions and relaxations evoked by EFS were measured isometrically (Grass FT03 and Grass Model 7D polygraph). Responses were evoked by EFS at optimal stimulus intensities (8V, 1 ms pulse duration, 200 to 300 mA) generated by a Grass Model S44 stimulator connected in series to a Stimusplitter (MedLab Instruments, Fort Collins, CO) and the electrodes. Relaxations were evoked after addition of atropine and contraction of the trachealis with 10 µM histamine. Frequency-response curves were constructed as described previously. Contractions and relaxations were evoked in the absence or presence of L-NNA, ODQ, or vehicle. To limit the influence of stimulating tachykinin-containing afferent nerve endings innervating the trachealis, a neurokinin₂ receptor-selective antagonist (SR 48968; 0.3 µM) was added to the tracheal perfusate (27). Contractions were expressed as a percentage of the maximum contraction evoked by barium chloride. Relaxations were expressed as a percentage reversal of the histamine-induced contractions.

Modulation of Histamine-mediated and Bradykinin-induced, Reflex-mediated Contractions of Guinea Pig Trachealis In Situ

Potential modulatory effects of the noncholinergic, parasympathetic relaxant nerves on contractions evoked by a direct-acting agonist (histamine) and an agonist that evokes tracheal contractions by initiating a cholinergic reflex (bradykinin) were studied with the in situ tracheal preparation (28). For histamine, the effects of blocking all nerve activity (tetrodotoxin [TTX]) or blocking only the NO-dependent component of the nerve-mediated effect were studied. (Histamine applied to the tracheal mucosa does not directly activate guinea pig tracheal afferent nerves and is therefore unlikely to initiate reflex alterations in ongoing nerve activity [29].) Following atropine, administration and stable contraction of the trachealis with 10 µM histamine, TTX, or vehicle was added to the tracheal perfusate. The effects of these treatments were quantified as a percentage increase in the magnitude of the histamine-induced contractions of the trachealis. In some experiments, bilateral vagotomies were performed before drug administration. In other experiments, the effects of TTX were assessed after the addition of L-NNA or L-NNA and L-arginine (1 mM) to the tracheal perfusate.

Bradykinin-mediated reflexes were evoked by administering this peptide via aerosol (1 mg/ml), intravenously (1 nmol/kg, administered via the abdominal vena cava), or by direct application to the laryngeal mucosa (3 µM). Reflex-mediated cholinergic contractions were evoked after the addition of L-NNA or its vehicle to the tracheal perfusate. Bradykinin-mediated noncholinergic relaxations were evoked after administration of atropine and contraction of the trachealis with 10 µM histamine. Reflex-mediated cholinergic contractions were expressed as a percentage increase in cholinergic tone. Reflex-mediated relaxations of the trachealis were expressed as a percentage reversal of the histamine-induced contractions. In some experiments, the effects of vagotomy on the response to bradykinin were assessed. In other experiments, the reflex effects produced by laryngeal (10 µM) or aerosolized (0.1 mM) capsaicin were studied.

Because the tracheal vasculature is preserved entirely in the preparation used in our study, it is necessary to block the local effects of intravenously administered autacoids on the tracheal segment studied in order to observe only the reflex effects evoked. Accordingly, all experiments with inhaled or intravenously administered bradykinin were conducted with the bradykinin B₂ receptor antagonist FR 173657 (3 µM) added to the tracheal perfusate (28). Furthermore, propranolol was administered systemically (1 mg/kg, intravenously) in all experiments conducted with intravenously administered bradykinin.

Statistics

All data are presented as mean ± SEM. Differences between groups were assessed through analysis of variance (ANOVA) on Statview for Macintosh (Jandel, Berkeley, CA). Values of p < 0.05 were considered significant. When significant variation between groups was detected, treatment group means were compared through Scheffe's F test for unplanned comparisons.

Drugs

Atropine sulfate, barium chloride, heparin sulfate, indomethacin, histamine, urethane (ethyl carbamate), succinylcholine chloride, L-NNA, L-arginine, phentolamine hydrochloride, and dl-propranolol hydrochloride were purchased from Sigma (St. Louis, MO). Capsaicin was obtained from ICN (Aurora, OH). Trimethaphan camsylate was acquired from Roche Laboratories (Nutley, NJ). Methoctramine hydrochloride was purchased from RBI (Natick, MA). TTX and ODQ were obtained from Tocris (Ballwin, MO). Bradykinin and SR 48968 (saredutant) were provided by Zeneca (Wilmington, DE). FR 173657 was provided by Fujisawa (Osaka, Japan). All drugs were dissolved in distilled water except for trimethaphan, methoctramine, bradykinin, and heparin, which were dissolved in 0.9% sodium chloride solution; capsaicin and indomethacin, which were dissolved in 100% ethanol; SR 48968, FR 173657, and ODQ, which were dissolved in dimethyl sulfoxide (DMSO); and L-NNA, which was dissolved in 0.1 N HCl. Capsaicin was further diluted in saline for aerosol/laryngeal challenges.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Modulation of Cholinergic Contractions and NANC Relaxations of the Trachealis Evoked by Vagus Nerve Stimulation In Vivo or EFS In Vitro

Bilateral stimulation of the vagus nerves (10 to 20 V, 1 ms pulse duration, 0.5 to 24 Hz, 1- to 3-min stimulation trains) evoked frequency-dependent cholinergic contractions of the guinea pig trachealis in situ. After addition of atropine to the tracheal perfusate and contraction of the trachealis with 10 µM histamine, bilateral stimulation of the vagus nerves also evoked frequency-dependent NANC relaxations of the trachealis. The contractions evoked by vagus nerve stimulation, which were abolished by atropine, developed faster and at much lower frequencies than the vagally mediated relaxations. In accord with the findings in previous studies (11, 15, 10, 23, 30), these contractions and relaxations were completely abolished after systemic administration of the ganglionic blocker trimethaphan (2 mg/kg) (Figure 1).

View larger version (14K): [in this window] [in a new window]   Figure 1.   Representative traces of frequency-dependent, vagally mediated contractions and relaxations of guinea pig trachea in situ. To evoke relaxations, the trachea was pretreated with atropine and contracted with 10 µM histamine. The vagus nerves were stimulated bilaterally (10 to 20 V, 1 ms pulse duration) at cumulatively increasing frequencies. Vagus nerve stimulation at 8 Hz evoked relaxations that reversed the histamine-induced contractions by 25 ± 5% and 0 ± 0% before and after ganglionic blockade with trimethaphan, respectively (n = 3; p < 0.05). Trimethaphan also abolished vagally mediated contractions (n = 3). ABP = arterial blood pressure; PT = pulmonary insufflation pressure; TT = tracheal tension.

In vitro, EFS-induced NANC relaxations were partly inhibited by the NO synthase inhibitor L-NNA (30 µM) or the soluble guanylate cyclase inhibitor ODQ (1 µM) (Figure 2). Also consistent with previously published reports (12, 13, 21, 31) was the finding that the effects of L-NNA and ODQ on the NANC relaxations were more prominent at lower stimulation frequencies (50% to 100% inhibition at frequencies < 5 Hz), only partly inhibiting the relaxations evoked by higher stimulation frequencies (10% to 50% inhibition). L-NNA also inhibited vagally mediated relaxations evoked in situ. By contrast, L-NNA and ODQ potentiated EFS-induced cholinergic contractions evoked in vitro (Figure 3). The effects of L-NNA were prevented by 1 mM L-arginine.

View larger version (6K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 2.   Effects of L-NNA and ODQ on EFS-induced relaxations in vitro and on vagally mediated relaxations in situ. Both experiments were done after atropine administration and contraction of the trachealis with 10 µM histamine. (A) Representative trace of EFS-induced (8 V, 1 ms pulse duration, 200 to 300 mA, 10 s train) NANC relaxations evoked in the absence and presence of L-NNA. In control tissues, relaxations were biphasic: A rapid and short-lived initial phase was followed by a long-lasting second phase. In accord with findings in previous studies, L-NNA inhibited the initial phase of EFS-induced relaxations (note long delay in onset of relaxations in tissues pretreated with L-NNA) and partly inhibited the longer-lasting second phase. Also, in accord with previous studies, was the finding that L-arginine (1 mM) prevented the effects of L-NNA (100 ± 9% of the control response; n = 3). Vertical and horizontal calibration bars denote 0.5 g and 3 min, respectively. (B) Mean data for frequency-response curves constructed in the absence and presence of L-NNA and ODQ. Relaxations at each frequency were evoked to equilibrium by continuous stimulation. , Control; , 10 µm L-NNA; , 1 µm ODQ. (C ) Relaxations of the guinea pig trachea in situ evoked with bilateral vagus nerve stimulation (16 Hz, 1 ms pulse duration, 10 to 20 V, 1 to 2 min train duration) in the presence of 30 µM L-NNA or vehicle. Each point is the mean ± SEM of five or six experiments. * denotes a statistically significant reduction in relaxant responses evoked by nerve stimulation relative to vehicle control (p < 0.05).

View larger version (14K): [in this window] [in a new window]   Figure 3.   Representative traces of the effects of L-NNA and ODQ on EFS-induced contractions of guinea pig trachealis. L-NNA and ODQ potentiated contractions evoked by EFS (16 Hz, 10 s, 1 ms pulse duration, 200 to 300 mA, 10 s train). Peak percentage increases in contractions evoked by EFS averaged 3 ± 3%, 52 ± 6%, and 27 ± 6% in the presence of vehicle, L-NNA, and ODQ, respectively (n = 6; p < 0.05). L-Arginine (1 mM) partly reversed the effects of L-NNA (42 ± 15% reversal; n = 3), but had no effect on the potentiation produced by ODQ (15 ± 34% enhancement; n = 3). Vertical and horizontal calibration bars denote 0.5 g and 5 min, respectively.

NO-dependent Modulation of Baseline Cholinergic Tone in the Trachealis In Situ

The results summarized in the preceding section and the results from previous studies indicate that NANC relaxations depend in part on the actions of NO (based on their sensitivity to NOS inhibitors and inhibitors of guanylate cyclase). The results also indicate that when activated, the NO-dependent component of the NANC relaxant responses can attenuate cholinergic contractions of ASM. In a previous study, we documented the presence of baseline cholinergic tone in the guinea pig trachea in situ (26). We reasoned that if the noncholinergic parasympathetic nerves were similarly active at baseline, the inhibition of their influence over ASM tone would potentiate baseline cholinergic contractions of the ASM. In accord with this hypothesis, the administration of L-NNA (30 µM) or ODQ (1 µM) to the tracheal perfusate produced gradual but marked increases in baseline cholinergic tone, which increased after administration of these drugs by 49 ± 10 [mean ± SEM]% (Figure 4). Subsequent administration of atropine completely reversed the effects of L-NNA and ODQ, as well as the preexisting baseline cholinergic tone. Vagotomy done before administration of L-NNA or ODQ eliminated baseline cholinergic tone and completely prevented the increases in contractile tone evoked by L-NNA and ODQ. L-arginine prevented the effects of L-NNA on baseline cholinergic tone.

View larger version (13K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 4.   Effects of L-NNA and ODQ on baseline cholinergic tone in guinea pig trachea in situ. Vehicle, L-NNA, or ODQ were selectively administered via the tracheal perfusate. The effects of these treatments were monitored until they reached equilibrium. Atropine was then administered to confirm the cholinergic nature of the responses elicited and to confirm the cholinergic nature of the baseline tracheal tone. (A) Representative trace of the effects of L-NNA on baseline cholinergic tone. Tone rose (~ 100 mg) gradually during continuous tracheal administration of the NOS inhibitor, reaching peak effects at 10 to 30 min after initial administration (see Figure 7). Subsequent treatment with atropine reversed the effects of L-NNA as well as the baseline cholinergic tone. ODQ produced similar effects (see Figure 4B) and, as reported previously (27), so did intravenously administered L-NNA. (B) Mean data for the effects of L-NNA, ODQ, or vehicle on baseline cholinergic tone in animals with intact (n = 5 to 18 per group) or severed (n = 3 per group) vagus nerves. The effects of L-arginine on the L-NNA- induced responses are also depicted (n = 5). The effects of the vehicles for L-NNA (0.1 N HCl; n = 15) and ODQ (DMSO; n = 3) on baseline tone did not differ and were therefore pooled. With the vagus nerves intact, baseline cholinergic tone averaged 26 ± 2% of the maximum contraction evoked by BaCl2 and did not differ among the five treatment groups (n = 45; p > 0.1). Vagotomy effectively abolished baseline tone (1 ± 1% of maximum contraction; n = 6; p < 0.05). Each bar represents the mean ± SEM from three to 18 experiments. * denotes a statistically significant increase in baseline cholinergic tone relative to vehicle control (p < 0.05). , Control; , 1 mM L-arginine; , vagotomy.

View larger version (15K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   Figure 7.   Effects of bradykinin inhalation on cholinergic tone after pretreatment with vehicle (n = 8), L-NNA (n = 7), or vagotomy (n = 3). (A) Aerosol administration of bradykinin evoked rapid and marked increases in tracheal cholinergic nerve activity. The percentage increase in cholinergic tone evoked by bradykinin did not correlate with the magnitude of baseline cholinergic tone before challenge (r2 = 0.1). (B) Mean data showing that although L-NNA markedly increased baseline cholinergic tone, the NOS inhibitor was without effect on the kinetics or the magnitude of bradykinin-mediated, reflex-induced increases in cholinergic tone. By contrast, vagotomy entirely reversed baseline cholinergic tone and prevented the effects of bradykinin on the trachealis. Comparable contractile and relaxant reflexes were initiated by capsaicin delivered by inhalation (10 µM) or delivered selectively to the laryngeal mucosa (10 µM). L-NNA also failed to modulate reflex tracheal contractions evoked by laryngeally applied bradykinin or capsaicin, which averaged 14 ± 5% (n = 5) and 8 ± 3% (n = 4) of maximum contraction in the absence and presence of 30 µM L-NNA, respectively (p > 0.1). , Vehicle; , 30 µM L-NNA; , vagotomy.

Nerve-dependent Modulation of Histamine- and Bradykinin-mediated Responses in the Trachea In Situ

As a more direct assessment of baseline noncholinergic nerve effects on ASM tone, we used TTX to selectively inhibit the influence of nerves on ASM contractions. Tracheal segments studied in situ were treated with atropine and contracted to ~ 50% of their maximum contraction by adding 10 µM histamine to the tracheal perfusate. Subsequent administration of TTX induced a slowly developing (15 to 30 min) increase in the magnitude of the histamine-induced contractions, producing an average increase in tone of 18 ± 4% of the initial contraction (range: 5% to 33% increase; n = 7). L-NNA produced similar effects to TTX on the histamine-induced contractions (22 ± 6% increase; n = 5). Pretreatment with L-NNA or prior vagotomy prevented the effects of TTX on the histamine- induced contractions. Both of these latter manipulations potentiated the contractions subsequently evoked by histamine (Figure 5). The effects of L-NNA on the contraction evoked by histamine were prevented by 1 mM L-arginine.

View larger version (13K): [in this window] [in a new window]   Figure 5.   Effect of TTX on histamine-induced contractions of guinea pig trachea in situ. Baseline cholinergic tone was reversed with atropine and the trachealis was precontracted (to 50% to 60% of maximum) with 10 µM histamine (see Figure 6B). When the contraction induced by histamine reached equilibrium, TTX or vehicle was added to the tracheal perfusate. Tone was monitored until the effects of TTX reached equilibrium (15 to 30 min). L-NNA was added before contraction of the trachealis with histamine. Histamine-induced contractions averaged 55 ± 5%, 75 ± 5%, 57 ± 4%, and 62 ± 7% of the maximum contraction when the trachealis was pretreated with vehicle (n = 17), L-NNA (n = 10), or L-NNA and 1 mM L-arginine (n = 6), or following vagotomy (n = 5), respectively. * denotes a statistically significant increase in the magnitude of the histamine-induced contraction (p < 0.05). Each bar is the mean ± SEM from three to nine experiments.

View larger version (12K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   Figure 6.   Bradykinin-mediated reflexes in the guinea pig trachea in situ. (A) With vagus nerves intact, bradykinin dose-dependently increased baseline cholinergic tone in the trachea, an effect reversed entirely by pretreatment with 1 µM atropine or vagotomy (the bradykinin B2-receptor antagonist FR 173657 [0.3 µM] was present in the tracheal perfusate at all times, thereby preventing any direct effects of the inflammatory peptide on the tracheal segment studied [27]). (B) After 1 µM atropine administration and contraction of the trachealis with 10 µM histamine, bradykinin evoked dose-dependent, reflex-mediated relaxations of the trachealis (see C and D). (C ) Mean data for the contractile and relaxant reflexes initiated by bradykinin administered intravenously (n = 6 to 8) or selectively to the laryngeal mucosa (n = 4 to 7). Reflex-mediated contractions and relaxations were prevented by interventions that included atropine (versus contractions only), vagotomy, or ganglionic blockade with 5 mg/kg trimethaphan (n  3; p < 0.01). Contractile responses are expressed as percentages of the maximum attainable contraction evoked by BaCl2. Relaxant responses are expressed as percentage reversals of the histamine-induced contractions. Comparable contractile and relaxant reflexes were evoked by aerosolized bradykinin (see Figure 7). , Control; , intervention. (D) Lack of effect of L-NNA on reflex-mediated relaxations evoked by bradykinin. Each bar represents the mean ± SEM from three to eight experiments. * denotes a statistically significant difference from the control value (p < 0.05). , Control; , 50 µM L-NNA.

Severing the vagi caudal to the recurrent laryngeal nerves, thereby disrupting all afferent nerve activity arising from the intrapulmonary airways and lungs, but preserving the parasympathetic innervation of the trachea, rendered TTX ineffective at increasing the magnitude of the histamine-induced contractions of ASM (n = 4). Thus, as in the case of the cholinergic nerves (26), the baseline activity of the noncholinergic parasympathetic nerves seems to depend on the ongoing activity of pulmonary afferent nerves. Despite this shared property, however, there was no correlation between the magnitude of baseline cholinergic tone and the effects of L-NNA on this tone (r² = 0.3) or between the magnitude of baseline cholinergic tone and the effects of TTX on the contraction evoked by histamine (r² = 0.1).

Bradykinin delivered as an aerosol to the lower airways (1 mg/ml), applied selectively to the laryngeal mucosa (3 µM), or administered intravenously (1 nmol/kg) evoked reflex tracheal contractions. We also observed that after atropine pretreatment and contraction of the trachealis with histamine, bradykinin (aerosolized, intravenous, and laryngeal) evoked reflex tracheal relaxations (Figure 6; [28]). Comparable contractile and relaxant reflexes were initiated by capsaicin delivered by inhalation (10 µM) or delivered selectively to the laryngeal mucosa (10 µM). The kinetics of the contractile and relaxant reflexes initiated by bradykinin differed substantially from one another. Whereas reflex tracheal contractions evoked by intravenous bradykinin (1 nmol/kg) reached peak magnitudes in an average of 80 ± 12 s (range: 45 to 130 s; n = 8), peak relaxant responses were attained an average of 371 ± 50 s after initiating the challenge (range: 240 to 540 s; n = 6; p < 0.05). By contrast, relaxant reflexes initiated from the larynx by bradykinin were maximal at 1 to 3 min after challenge, whereas contractile reflexes evoked from the larynx developed slowly, reaching peak magnitudes at 10 to 25 min after challenge (n = 4 to 7). Vagotomy or ganglionic blockade before bradykinin challenge prevented all of these reflex-mediated responses. L-NNA was, however, without effect on the reflex-mediated relaxant responses (Figure 6). This suggests that although the L-NNA-sensitive component of the NANC relaxant response can modulate baseline cholinergic tone, the heightened nerve activity associated with reflex activation may not enhance NO synthesis and/or activity. In accord with this hypothesis, L-NNA was ineffective at modulating the reflex tracheal contractions evoked by either bradykinin or capsaicin (Figure 7).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NANC relaxant nerves are the predominant bronchodilating influence on the airways in humans and in many mammalian species (11). Despite their likely importance in maintaining airway caliber, these nerves remain poorly characterized. The difficulty of studying this neuronal pathway is the primary reason for the poor understanding of the role of the NANC relaxant nerves in the lungs (19). Unlike contractile nerves innervating the airways, dilating nerves must be studied when the airways are preconstricted, since without such tone, dilatation becomes unmeasurable. This complicates the experimental design in such studies. Another hindrance has been the lack of useful and selective pharmacologic tools with which to study the actions of these nerves. Only recently have agents that selectively target the nonadrenergic component of airway relaxant responses been developed (12, 31).

Nerve activity in the trachea provides an accurate glimpse of the activity of the NANC relaxant nerves in the entire lung and airway tree (26, 32). Measurements of tracheal nerve activity can therefore be used to predict the role of airway autonomic nerves in regulating airways caliber. The experimental design used in the present study provides an ideal approach to studying ongoing parasympathetic nerve activity and reflex-mediated alterations in this activity. Unlike whole-lung assessments of nerve activity and reflexes, nerve activity in the trachea can be monitored without the complicating influences of the mechanical forces produced by breathing. Tracheal tone is also measured at a site distal to the sites at which airway afferent nerves are stimulated to evoke changes in the activity of the parasympathetic nerves. The approach described here also permits the use, to assess the role of nerves in regulating ASM tone, of powerful pharmacologic tools (e.g., TTX, L-NNA) that cannot be administered systemically or via inhalation without profoundly affecting extrapulmonary and/or nonneural targets. These nonneural and systemic effects might in turn initiate alterations in airway nerve activity or might profoundly influence the pulmonary response to an aerosolized agent.

NO-dependent Effects of NANC Nerves on ASM Tone

NANC relaxations of the airways are mediated by postganglionic parasympathetic nerves (11, 15, 20, 23, 30). The relaxation evoked by activation of these postganglionic nerves appears to involve NO and peptides such as VIP. Physiologic, immunohistochemical, and morphologic studies of these nerves indicate that they represent a distinct parasympathetic pathway from the well-characterized cholinergic-parasympathetic pathways innervating the airways (8, 10, 20). Consequently, it seems likely that interactions between these nerve pathways occur postjunctionally and are manifested through their opposing actions on ASM.

Drugs that modulate the NO-dependent component of the NANC relaxant response (e.g., L-NNA, ODQ) are without effect on cholinergic nerve-mediated contractions when the noncholinergic nerves are not simultaneously activated (2, 33). In the present study, we observed that L-NNA and ODQ increased baseline cholinergic tone in the trachealis and increased the magnitude of histamine-induced contractions of the trachealis in situ. The effects of these interventions on both baseline cholinergic tone and the histamine-induced contractions were pronounced. Cholinergic tone increased in these studies by nearly 50% (range: 0% to 113%), and the histamine-induced contractions increased in magnitude by nearly 40%. These data provide compelling evidence that NO synthesis by nerves can have a considerable influence on airways obstruction and airways responsiveness to exogenously administered agonists. Similar results have been reported elsewhere (2, 34).

Although there are myriad sources of NO in the tracheal wall, the observation that the effects of L-NNA and ODQ were prevented by vagotomy and mimicked by TTX provides evidence that the effects of these agents in this study were due to their ability to partly inhibit NANC relaxations. A previous study found a similar role for NO formed by airway nerves (37). The observation that L-NNA and ODQ only partly inhibit the noncholinergic-parasympathetic nerve-mediated effects on ASM tone, made in the present study and elsewhere (9, 12, 13, 21, 31), also suggests that the influence of these nerves may be underestimated. The non-NO-dependent component of nonadrenergic relaxant responses might also profoundly influence airways reactivity, particularly during periods of high-frequency parasympathetic nerve activation. Further assessment of the role of airway noncholinergic relaxant nerves in regulating airway caliber awaits the development of potent and specific inhibitors of the peptide-mediated component of this relaxant response.

L-NNA potentiated baseline cholinergic tone but had no effect on reflex-mediated tracheal contractions evoked by bradykinin or capsaicin. These results seem contradictory, and the latter observation seems to contradict the results of previous studies (5). Again, however, it is important to emphasize the key features of our experimental design. Our specific intentions with these studies were to assess the modulatory effects of noncholinergic relaxant nerves on ASM tone. NO is just one component of the noncholinergic relaxant response (Figures 2 and 6). Neither component of this relaxant response can be adequately studied pharmacologically with whole-lung measurements of ASM activity. This is particularly true of the NO-dependent component. Lung challenge with histamine or bradykinin may activate all NO-producing cells in the airways and not only nerve cells alone. It seems likely that NO from any source could modulate airways reactivity. Our results would suggest, however, that the NO derived from noncholinergic parasympathetic relaxant nerves does not modulate responsiveness to bradykinin, and further, that the NO formed in previous whole-lung challenge studies was not derived exclusively from airway nerves (5). It is also worth noting that relaxant nerves are more effective at reversing a sustained contraction of ASM than at preventing such contractions (17). The results of the present study suggest that bradykinin, which evokes profound increases in parasympathetic nerve activity (Figures 6 and 7), surmounted the low-frequency inhibitory effects of L-NNA on noncholinergic relaxations. The inability of L-NNA to inhibit reflex-mediated relaxations evoked by bradykinin, as reported here, are consistent with this notion.

Spontaneous Activity of Cholinergic and Noncholinergic Parasympathetic Nerves

The effects of L-NNA and ODQ on baseline cholinergic tone in the trachealis, and the effects of L-NNA and TTX on histamine-induced contractions of the trachealis, suggest that the noncholinergic nerves are spontaneously active during tidal breathing. This spontaneous activity in the noncholinergic parasympathetic pathway was abolished when ongoing afferent nerve activity arising from the intrapulmonary airways and lungs was disrupted by intrathoracic vagotomy in which the nerves were cut just caudal to the recurrent laryngeal nerves. We reported a similar finding in our studies of baseline cholinergic tone in the trachealis (26). These data are interpreted as evidence that baseline parasympathetic tone, both cholinergic and noncholinergic, is absolutely dependent on the ongoing activity of intrapulmonary afferent nerves. Perhaps the activity in the noncholinergic parasympathetic nerves, like cholinergic nerve activity, is driven by the rapidly adapting receptors of the peripheral airways, activated during respiration as a consequence of the mechanical forces of breathing. Previous studies document the ability of these nerves to initiate increases in noncholinergic parasympathetic nerve activity (28, 38).

Measuring the extent to which the parasympathetic nerves are active during tidal breathing would be difficult. Such an assessment would necessarily involve recordings from identified pre- and/or postganglionic nerves innervating the trachealis. This is further complicated by the observation that the noncholinergic ganglia seem to filter a large proportion of preganglionic input (20, 39), making measurements of preganglionic activity less predictive of the ultimate influence of these nerves on ASM tone. It may, however, still be possible to predict a range of parasympathetic nerve activity based on the effects of L-NNA and TTX. Thus, L-NNA virtually abolished EFS-induced relaxations evoked at stimulation frequencies less than 2 Hz and also abolished vagally mediated relaxations evoked at stimulation frequencies of less than 12 Hz (21, 39). In the present study, L-NNA was as effective as TTX in potentiating histamine-induced contractions of the trachealis. TTX completely abolishes EFS- or vagally mediated NANC relaxations of ASM. We therefore suggest that preganglionic parasympathetic nerves mediating relaxations of the trachealis are spontaneously active at a mean action potential frequency of less than 12 Hz. Extending this line of reasoning, the data also suggest that when they are reflexly activated (such as through the use of bradykinin in the present study), postganglionic noncholinergic parasympathetic nerves produce action potentials at frequencies > 12 Hz (as based on the lack of measurable effects of L-NNA on bradykinin-mediated reflexes).

Potential Physiologic Significance of Noncholinergic Parasympathetic Nerve Activity

Noncholinergic parasympathetic nerves can profoundly influence the contractility of ASM. Alterations in NO formation are similarly influential on ASM tone and on airways reactivity in humans and many other mammalian species. The results of the present study provide evidence that the effects of inhibition of NOS may be partly due to the effects of these interventions on airway parasympathetic nerves. It is therefore reasonable to speculate that dysfunction and/or dysregulation of the noncholinergic parasympathetic nerves might contribute to the pathogenesis of airways disease (5, 18, 19, 37, 40). NO is, however, only one of several potential transmitters that mediate noncholinergic parasympathetic responses. Accordingly, studies published to date, including the present study, might underestimate the influence of noncholinergic parasympathetic nerves on airway caliber.