INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Vagal afferent-released tachykinins are thought to contribute to airway reflexes including bronchoconstriction, coughing, and mucus secretion. Airway sensory innervation originates from both the superior (jugular) and inferior (nodose) vagal ganglia, as well as some cervical dorsal root ganglia (1). Normally, tachykinin immunoreactivity is observed only in jugular neurons projecting to the guinea pig airway (2). However, nodose neurons innervating the airway are pleiotropic and can become tachykininergic following airway inflammation. Preprotachykinin A (PPT-A) messenger RNA (mRNA), which codes for both substance P (SP) and neurokinin A (NKA), is significantly upregulated in the somata of nodose neurons within 12 h of airway inflammation initiated by exposing actively immunized guinea pigs to airway antigen (3). Twenty-four hours after antigenic challenge, there is a measurable increase in the expression of the tachykinin SP (3). These data indicate that the tachykininergic system of nodose neurons may be modified during allergic airway diseases.

It has been well established that neurons of the peripheral nervous system are often responsive to the neurotransmitters and neuropeptides they secrete. Therefore, it seems likely that, in addition to inducing the synthesis of SP, allergic inflammation in the airway may also evoke the expression of functional tachykinin receptors in nodose neurons. Tachykinin receptor agonists produce no measurable effects on electrophysiological properties of control nodose ganglion neurons. However, after antigenic activation of nodose ganglion mast cells in vitro, many nodose neurons reveal depolarizing SP responses that are reversibly abolished by a neurokinin (NK)-2 receptor antagonist (4). NK-2 receptor expression can occur within 5 min of mast cell activation and does not require new protein synthesis (4). Thus, regulation of NK-2 receptors relies on posttranslational mechanisms; we have designated this phenomenon "unmasking."

The current work examines whether airway inflammation brought about by antigen inhalation in vivo also unmasks functional NK-2 tachykinin responses in vagal afferent somata. As observed previously (4), vagal sensory neurons from control nodose ganglia were electrophysiologically unresponsive to tachykinins. However, within 24 h of allergen inhalation, the majority of nodose neurons were depolarized by exogenously applied tachykinins acting on NK-2 receptors. Vagotomy significantly attenuated antigen-induced unmasking of tachykinin responses, indicating that the nerve fibers connecting the airways and vagal somata are involved in the transduction of a signal essential for unmasking of tachykinin responses.

It has been well established that sensory neuron-released tachykinins can evoke a number of proinflammatory responses (e.g., vasodilation and plasma extravasation) through activation of neurokinin receptors localized to blood vessels and epithelia (5). Though the present study was carried out on the vagal afferent somata, our data suggest that at sites of airway inflammation, a new target emerges for released tachykinins, namely the primary afferent neurons themselves. A change in the tachykinin responsiveness of peripheral sensory nerve endings may provide a pathway for the processing of noxious stimuli, thereby contributing to the hyperalgesia and pain often associated with inflammation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Immunization and Antigenic Challenge

Adult male Hartley guinea pigs (100 to 200 g; Charles River, Wilmington, MA) were actively or passively immunized to ovalbumin (chicken egg albumin, OVA, grade V), as described previously (6). Active immunization was accomplished by injecting OVA (150 µl, 100 mg/ml) intraperitoneally every other day for a total of three injections. It is possible that activation of the immune system during the active immunization process contributes to the upregulation of NK-2 responses. To address this hypothesis, we bypassed the active immunization process by passively immunizing animals using serum previously collected from immunized animals (1 ml/kg, intraperitoneally). Similar results were observed following either method of immunization.

Three weeks after active immunization, or 24 h after passive immunization, the animal was placed in an acrylic chamber measuring 20 × 20 × 60 cm. An aerosol of OVA (0.3 mg/ml) was introduced into the chamber via a DeVilbiss 646 nebulizer (Somerset, PA). The animal was observed for symptoms (cough or labored breathing), and was removed from the chamber when symptoms appeared, typically within 2 to 5 min. Those animals that failed to respond within 5 min were given a larger dose of antigen (up to 1 mg/ml), which caused symptoms in all sensitized animals. This procedure was repeated on three consecutive days. Animals were killed 24 h after the last antigen challenge. Nonimmunized animals exposed to nebulized OVA and OVA-immunized animals exposed to aerosolized bovine serum albumin (BSA) served as controls.

Histological Evaluation of Airway Inflammation

To determine whether aerosolized antigen provoked airway inflammation in immunized animals, tracheas from OVA-challenged and control animals were immersion fixed in 10% formaldehyde in phosphate-buffered saline (PBS). After dehydration in ethanol and clearing with xylenes, tracheas were embedded in paraffin and 5-µm sections were prepared. Sections were rinsed in xylenes, rehydrated, stained with May-Grunwald-Giemsa, dehydrated and coverslipped. Stained sections of the airway were viewed with an Olympus BX60 microscope and photographed (Olympus PM-E35X camera). Photos were scanned using a Polaroid SprintScan 35 then viewed using Adobe PhotoDeluxe 1.0, and printed using an Epson Photo 700.

Unilateral Vagotomy

A subset of guinea pigs were subjected to unilateral vagotomy 7 to 10 d before airway antigen exposure. Surgery was performed under ketamine-xylazine (30 and 5 mg/kg intraperitoneally, respectively) anesthesia. The vagus nerve was exposed through an incision in the neck and was transected approximately 2 cm caudal to the nodose ganglion. Approximately 5 mm of the vagus was removed and the cut ends were tied off with silk suture. This procedure was performed on the left vagus in seven animals and on the right vagus in two animals. Vagotomy affected each side identically because we did not observe any left versus right differences in the percentage of SP-responsive neurons after antigen inhalation. Vagotomized animals that were not exposed to airway antigen served as controls. Nodose neurons from ganglia associated with a cut vagus nerve were not SP-responsive; only one of 39 neurons was depolarized (10 mV) by bath application of 100 nM SP.

An appropriate positive control to assess the influence of unilateral vagotomy on inflammation-induced unmasking of SP responses would be nodose neurons from the contralateral ganglion of the same animal. However, in preliminary experiments with control guinea pigs (n = 4), unilateral vagotomy alone induced SP responses in nodose neurons from contralateral ganglia (approximately 25%; 13 ± 1.0 mV; n = 8 of 32 neurons). Vagotomy-induced unmasking of SP responses in normal animals was indeed unexpected. It did not occur in sham-operated animals (n = 2), where vagus nerves were exposed and silk sutures were transiently placed around the nerves. The mechanisms underlying this phenomenon were not investigated further. To minimize interpretational difficulties, we have not used nodose neurons from contralateral ganglia as controls. Rather, we have utilized neurons from animals with both vagi intact as positive controls.

Tissue Preparation

Twenty-four hours after the last treatment with control or antigenic proteins, animals were killed by asphyxiation with CO₂, as approved by the Institutional Animal Care and Use Committee of the University of Maryland, Baltimore. Nodose ganglia were dissected bilaterally and placed in ice-cold (4° C) Locke solution (composition in mM: 136 NaCl; 5.6 KCl; 1.2 MgCl₂; 2.2 CaCl₂; 14.3 NaHCO₃; 1.2 NaH₂PO₄; and 10 dextrose) equilibrated with 95% O₂/5% CO₂, pH 7.2 to 7.4. Acutely dissociated neurons were prepared enzymatically using a modification of the procedure described by Christian and coworkers (7). Briefly, ganglia were trimmed of excess connective tissue and placed in Ca²⁺-free, Mg²⁺-free Hanks' balanced salt solution containing collagenase, type 1A (1 mg/ml, 455 U/mg; Sigma) and dispase, grade II (1 mg/ml, 0.5 U/mg; Boehringer Mannheim, Indianapolis, IN) for 6 to 8 h at 4° C, followed by 15 min at room temperature (22° to 24° C) and 15 min at 37° C. The tissue was triturated with a fire-polished Pasteur pipette, washed twice in Lebovitz L-15 medium (GIBCO BRL, Rockville, MD) containing 10% (vol/vol) fetal bovine serum (FBS; JRH Biosciences, Lexena, KS) and then dissociated neurons were resuspended in Lebovitz L-15 medium containing 10% FBS. Cell suspensions (0.15 ml) were transferred onto circular 15-mm polylysine (0.1 mg/ml poly-D-lysine; Sigma)-coated glass coverslips (Bellco, Vineland, NJ) in a 24-well culture plate. Neurons were maintained at 37° C for at least 8 h before intracellular recording.

Electrophysiological Recording and Drug Delivery

Standard intracellular recording techniques with glass microelectrodes were used for current- and voltage-clamp recordings, as described in detail by Weinreich and coworkers (4). Data acquisition and analysis of electrophysiological data were performed using pClamp 6.2 software in conjunction with a Digidata 1200 interface (Axon Instruments, Foster City, CA). Neurons were accepted for study only if they showed a stable resting membrane potential (E_m  50 mV) and had action potentials overshooting 0 mV.

Agonists were superfused over isolated neurons for 20 to 60 s. When receptor antagonists were used, neurons were superfused with these reagents for at least 2 min before the addition of agonists. The temperature of the Locke solution flowing through the recording chamber (3 to 4 ml/min) was maintained at 33° to 35° C.

Drug Solutions

Drug solutions were prepared daily from concentrated ( 10 mM) stock solutions stored at 20° C. CP99,994 was provided by Dr. Jim Heyn (Pfizer, Inc., Groton, CT) and SR48968 was a gift from Zeneca (Wilmington, DE). Chick OVA was acquired from Sigma Chemical Co., St. Louis, MO. The same lot number of OVA used to immunize an animal was used for antigen challenge of the airway.

Data Analysis

Data are expressed as mean ± SEM. Student's paired and unpaired t tests were used to assess significant differences between calculated means; p  0.05 was considered significant. The z test and chi-square tests were used to compare differences in the percentage of neurons responding to SP after various treatments. Statistical analysis was performed using Sigmastat software (Jandel Scientific, San Rafael, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histology of Antigen-challenged Airway

Using scanning electron microscopy, Fischer and coworkers (3) have reported epithelial shedding and accumulation of inflammatory cells, expected signs of allergic airway inflammation, following antigen inhalation by immunized guinea pigs. Using a similar sensitization and inhalation protocol, we too observed signs typical of airway inflammation. Histological examination of airway tissue from airway antigen-challenged guinea pigs revealed extensive eosinophilic infiltration into the airway epithelium and submucosa of the thoracic trachea and mainstem bronchi (Figure 1, right panel ). By contrast, eosinophilia was not evident in the airways of control guinea pigs (Figure 1, left panel ).

View larger version (65K): [in this window] [in a new window]   Figure 1.   Effect of antigen challenge on the cellular architecture of the airway. (Left panel ) Cross section of trachea taken from a guinea pig 24 h after a sham antigen challenge. (Right panel ) Representative cross section of guinea pig trachea taken from an animal 24 h after antigen inhalation as described in METHODS. Note the elevation in the number of eosinophils in the epithelium and submucosa of the trachea from the antigen-challenged animal. Scale bar is 30 µm.

SP Responses Recorded from Nodose Neurons after Inhalation of Aerosolized Antigen

Eighty percent of nodose neurons isolated from ganglia 24 h after antigen inhalation by immunized guinea pigs were depolarized by exogenously applied SP (100 nM). The percentage of SP-responsive neurons did not vary between immunized animals exposed to airway OVA (p = 0.447; ² = 2.7). The peak depolarization averaged 9 ± 0.8 mV (n = 32; Table 1 and Figure 2) and was accompanied by a 23 ± 3.0% (n = 20) decrease in membrane input resistance (R_in) in 74% of neurons in which R_in was monitored (n = 27). In the remaining 26%, there was either no change (n = 1) or a 20 ± 7.0% increase in R_in (n = 6). When neurons from immunized antigen-challenged guinea pigs were voltage-clamped to their E_m (approximately 60 mV), SP application produced an inward current averaging 808 ± 198 pA (n = 12) that was accompanied by an 88 ± 26.5% increase in membrane conductance (n = 11 of 12 neurons; Figure 2).

View larger version (24K): [in this window] [in a new window]   Figure 2.   Typical SP responses after inhalation of nebulized antigen (OVA) by a passively immunized guinea pig. (Upper trace) Depolarizing response produced by SP (100 nM) recorded in a nodose neuron. Downward deflections are electronic voltage transients produced by hyperpolarizing current pulses (100 pA, 300 ms, 0.6 Hz); the magnitude of these transients is a measure of membrane input resistance (Rin). The SP-induced membrane depolarization was accompanied by a decreased Rin. The resting membrane potential (Em) was 58 mV; resting Rin was 45 M. (Lower trace) Inward current in response to SP recorded in the same nodose neuron voltage clamped to 51 mV. Downward current deflections were elicited by hyperpolarizing voltage step commands (5 mV, 300 ms, 0.6 Hz) to monitor membrane conductance (Gm). The SP-induced inward current was accompanied by an increased Gm. The horizontal bar represents the time of SP application.

By contrast, acutely isolated nodose neurons from control or immunized but nonallergen-challenged guinea pigs lacked measurable electrophysiological responses to SP (0.1 to 10 µM; Table 1; see also References 4 and 8). Similarly, inhalation of a control protein (BSA) by OVA-immunized guinea pigs did not unmask depolarizing SP responses (n = 22; Table 1). In one neuron, SP evoked a small membrane hyperpolarization (3 mV). Exposure of nonimmunized guinea pigs to aerosolized antigen (OVA) also did not elicit the unmasking of SP responses (n = 23; Table 1). Resting membrane properties in nodose neurons isolated from guinea pigs after allergic airway inflammation were not significantly different from those recorded in control nodose neurons (Table 1). Together, these results suggest that the SP responses elicited after allergen inhalation by immunized guinea pigs are the consequence of a specific antigen-antibody reaction.

The pharmacology of tachykinin responses unmasked by antigen challenge in vitro has previously been characterized extensively (4). Here, we have used selective, nonpeptide neurokinin receptor antagonists to confirm that tachykinin responses unmasked by in vivo airway antigen exposure are also mediated by NK-2 tachykinin receptors. SR48968 (100 nM), a selective NK-2 receptor antagonist (9), reversibly reduced the SP-induced depolarization from 15 ± 4.8 to 3 ± 2.6 mV (n = 4; p = 0.024). CP99,994 (100 nM), a selective NK-1 receptor antagonist (10), did not significantly (p = 0.344) inhibit the SP-mediated depolarization (9 ± 2.0 versus 7 ± 2.0 mV; n = 2). These pharmacologic results suggest that allergic airway inflammation in vivo also unmasks NK-2 tachykinin responses.

Effect of Unilateral Vagotomy on the Induction of NK-2 Receptors after Inhalation of Aerosolized Antigen

To determine whether the induction of NK-2 receptors in nodose ganglion neurons requires an intact vagus nerve to relay a signal from the airway to the ganglion (e.g., through anterograde transport of trophic factors or nerve impulse activity), we performed a series of experiments examining the impact of unilateral vagotomy. Vagotomy significantly reduced the percentage of neurons with unmasked SP responses after antigen inhalation (p  0.001). SP depolarized 28% of nodose neurons isolated from ganglia with severed vagi 24 h after aerosolized antigen, compared with 80% of nodose neurons from animals with both vagi intact. In neurons with severed vagi, the magnitude of the SP depolarization (9 ± 1.3 mV; n = 28; Table 1) was not significantly different from that recorded in nodose neurons associated with intact vagi (9 ± 0.9 mV; n = 32; Table 1). These data suggest that the vagus nerve may be required, at least in part, for the unmasking of functional tachykinin responses.

The reduced number of SP-responsive neurons cannot be attributed to changes in membrane properties of axotomized nodose neurons. The E_m of these neurons did not differ significantly from those of neurons with intact vagi. Furthermore, the R_in of axotomized neurons was significantly increased (approximately 25%; p = 0.001; Table 1). Rather than diminishing the amplitude of the SP depolarization, an increased R_in would be expected to increase the voltage change, thereby favoring the detection of SP-responsive neurons.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Normally, guinea pig vagal afferent somata (nodose ganglion neurons) are electrophysiologically insensitive to exogenously applied SP over the concentration range of 0.1 to 10 µM. Twenty-four hours after inhalation of aerosolized antigen, 80% of nodose ganglion neurons were depolarized by bath applied SP (100 nM) in the absence of any change in resting membrane properties (E_m and R_in). Pharmacological analysis revealed that these SP responses were mediated by NK-2 tachykinin receptors. These changes are quantitatively similar to those observed after antigenic activation of nodose ganglion mast cells in vitro (4).

Airway inflammation may alter vagally mediated impulse traffic and/or anterograde transport of trophic factors (e.g., nerve growth factor or brain-derived neurotrophic factor), both of which may contribute to the expression of functional NK-2 receptors. We have compared the percentage of SP- responsive neurons from nodose ganglia associated with severed vagi to that of nodose neurons from animals with both vagi intact. After airway inflammation, SP depolarized 80% of control nodose neurons with intact vagi, whereas only 28% of the nodose neurons from ganglia associated with cut vagi were depolarized. Because vagotomy significantly reduced the percentage of SP-responsive nodose neurons observed after allergic inflammation of the airway, it appears that the vagus nerve is critical for the unmasking of tachykinin responses.

Why NK-2 receptor unmasking was not completely eliminated by vagotomy is unknown. However, it is likely that the remaining fraction of tachykinin-responsive neurons project their peripheral fibers to the upper airway via the superior laryngeal nerve, and are thus not affected by the vagotomy procedure. Nonetheless, in the absence of direct evidence supporting the involvement of the superior laryngeal nerve in transduction of an "unmasking signal" from the airway, bloodborne mechanisms remain a viable alternative.

In a preliminary series of experiments, we tested whether nerve impulse activity alone could unmask SP responses. Isolated vagi with attached nodose ganglia were electrically stimulated with depolarizing current pulses (0.1 to 0.5 ms, 10 to 30 Hz) at intensities sufficient to elicit A- and C-fiber compound action potentials. In three preparations, electrical stimulation ranging from 30 min to 2 h did not induce unmasking of SP responses (author's unpublished observations). Either longer time periods of impulse activity are required for unmasking of SP responses, or trophic factors may play a dominant role. Clearly additional experiments are necessary to determine the nature of the signaling pathway between the inflamed airway and the nodose ganglion.

It is unlikely that all of the nodose neurons expressing unmasked NK-2 responses innervate the airway. A number of hypotheses can be put forth to explain the large percentage of tachykinin-responsive neurons. Because vagotomy abolishes the bulk of NK-2 receptor unmasking, a likely possibility is that exaggerated impulse activity in airway-projecting nodose neurons releases paracrine signaling molecules in the ganglia (11, 12). One paracrine mediator of particular interest in this regard is nitric oxide (NO); in acutely isolated nodose ganglion neurons, a NO signaling cascade has been shown to unmask functional NK-2 receptors in approximately 65% of neurons (8). Alternatively, aerosolized antigen molecules may reach other viscera innervated by the nodose ganglia (e.g., the upper gastrointestinal tract). As with all aerosolized substances, at least a small fraction of antigen molecules will enter the gastrointestinal tract via swallowing. Local inflammation in multiple tissues could provoke the unmasking of NK-2 responses in a large percentage of nodose neurons. Finally, inflammatory mediators generated in the airway could travel to the ganglia via the bloodstream and unmask NK-2 responses in nonairway projecting nodose neurons.

In conclusion, allergic airway inflammation unmasks functional tachykinin receptors in the bulk (80%) of nodose neurons. In light of these findings, it is noteworthy that the percentage of nodose neurons that are tachykininergic increases to 50% after allergen inhalation (3). Thus, it seems likely that at least a subpopulation of nodose neurons are expressing both tachykinins and functional tachykinin receptors after allergic inflammation of the airway.

The expression of both tachykinins (SP and NKA) and their receptors (NK-2 receptors) in vagal afferent neurons may have important functional consequences. If the unmasking of NK-2 responses also occurs at vagal afferent nerve terminals, it may contribute to the heightened reflex activity (e.g., cough) that is commonly associated with allergic airway disease. Furthermore, somal release of SP (12) may provide a mechanism for excitatory signaling between primary afferent somata, perhaps adding to the synchrony of impulse traffic.