Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

It is well documented that airway mucosal inflammation and bronchial hyperreactivity occur commonly in cigarette smokers (1). Further, chronic smoking-induced bronchial hyperreactivity has been repeatedly demonstrated in various animal models (2, 3), but the underlying mechanism is not fully understood.

The entire respiratory tract is innervated by vagal afferent fibers of which ~ 75% are nonmyelinated (C) fibers (4). These bronchopulmonary C-fiber afferents are extremely sensitive to various inhaled irritants, including cigarette smoke (CS) (5). Recent reports have further demonstrated extensive innervation of airway mucosa by the tachykinin-related and calcitonin gene-related peptide (CGRP)-immunoreactive afferent endings (6, 7). Tachykinins and CGRP are synthesized in the cell bodies of small-size (predominantly C) sensory neurons in the nodose, jugular, and dorsal root ganglia, then transported to and stored in the peripheral nerve terminals. When bronchopulmonary C-fiber endings are activated, the impulses trigger the release of these neuropeptides, which can induce neurogenic inflammation in the airways (8). A recent study by Wu and Lee (11) reported that chronic exposure to CS induced airway mucosal inflammation accompanied by bronchial hyperreactivity in guinea pigs, and that the tachykininergic mechanism is involved. In view of increasing evidence of sensitization of bronchopulmonary C-fiber afferents caused by airway mucosal inflammation (12), we hypothesize that mucosa injury induced by chronic exposure of the airways to CS causes a sustained stimulation of these afferents and thereby increases the synthesis of these peptides. If this hypothesis is correct, we expect to find an increased tachykinin content in the sensory terminals innervating the airways and an increased expression of -preprotachykinin (PPT) messenger RNA (mRNA), the precursor of the tachykinins, in the nodose, jugular, and dorsal root C-fiber neuron soma of CS-exposed animals. The present experiment was carried out in guinea pigs to test this hypothesis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

The procedures described herein were performed in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health, and were also approved by the University of Kentucky Institutional Animal Care & Use Committee.

Chronic Exposure to CS

Young male Hartley guinea pigs of similar age and weight (initial weight 250 to 300 g) were randomly divided into two groups to be exposed to either mainstream CS (the CS group) or air (the control group). The chronic smoke exposure was carried out by the University of Kentucky Tobacco and Health Research Institute staff according to a standard protocol (13). In brief, each awake guinea pig was held in place by a wire mesh restrainer with only the nose and mouth exposed to an exposure chamber. A puff of smoke (35 ml, 50% concentration) generated from a University of Kentucky Reference Cigarette 2R1 was drawn into the chamber each minute. Guinea pigs inhaled smoke during the first 20 s; the residual smoke was then vacuumed away and replaced by fresh air during the remainder of the 1-min exposure. Each 2R1 cigarette (85 mm overall length and 30 mm butt length) delivers 10 puffs of cigarette smoke containing a total of 2.19 mg nicotine, 32.9 mg tar, and 38.8 mg total particulate matter. After a 1-wk acclimatization period, each guinea pig was exposed to 10 puffs of CS per cigarette, four cigarettes/d for 14 to 17 consecutive days. Daily measurement of the level of total particulate matter delivered to the animals was performed during each exposure period, and during each week the arterial carboxyhemoglobin saturation level of one randomly chosen animal was measured immediately after the smoke exposure. Control animals underwent the identical procedure and duration of smoke exposure as did the treated group, except that CS was not generated during the exposure.

Measurement of Lung Mechanics

Within 2 h after the last CS or sham (air) exposure period, guinea pigs were anesthetized with chloralose (100 mg/kg, intraperitoneally) and urethane (500 mg/kg, intraperitoneally), and supplemental doses of the same anesthetics were administered whenever necessary to maintain abolition of the withdrawal reflexes in response to pain induced by toe-pinch. A heating pad was placed under the animal to maintain the body temperature at ~ 36°C. The trachea was cannulated just below the larynx through a tracheotomy, and the lungs were ventilated with a respirator at a constant rate of 44 breaths/min and a stroke volume of 8 ml/kg of body weight. The right jugular vein and carotid artery were cannulated for intravenous injections and for arterial blood pressure (ABP) measurement, respectively. A catheter for measuring intrapleural pressure (P_ip) was inserted into the right intrapleural cavity through an incision between the fifth and sixth ribs; this incision was subsequently sutured and further sealed air-tight with silicone jelly. The pneumothorax was then corrected by briefly opening the intrapleural catheter to ambient air during a held hyperinflation of the lungs. Transpulmonary pressure (P_tp) was measured as the difference between the tracheal pressure and P_ip. Respiratory flow was measured with a heated pneumotachograph and a differential pressure transducer (Validyne MP 45-14). All signals were analyzed by an on-line computer for total pulmonary resistance (RL) and dynamic lung compliance (C_dyn) on a breath-by-breath basis (TS-100 series; Biocybernetics, Taipei, Taiwan).

Radioimmunoassay of Neuropeptides in Airway Tissues

An intrathoracic airway specimen, including the lower trachea and extrapulmonary bronchi, was removed from each animal immediately after a midline thoracotomy under anesthesia. The specimen was weighed, boiled (95°C) for 10 min in either 1 M acetic acid (1:10, wt/vol) for radioimmunoassay (RIA) of CGRP or distilled water for RIA of substance P (SP), and homogenized. CGRP was also measured in this experiment because it is known to be colocalized and coreleased with tachykinins from C-fiber endings, and is more stable than tachykinins (8). Homogenates were transferred to polypropylene tubes and centrifuged (40,000 × g, 4°C, 20 min). Supernatant fractions were removed and partially purified with C₁₈ Sep-Pak columns. After washing columns with 0.1% trifluoroacetic acid (TFA) (four washes with 5 ml each time), CGRP and SP were eluted from columns with 60% acetonitrile in 0.1% TFA. The samples were lyophilized and reconstituted in RIA buffer, containing 0.1 M phosphate buffer (pH 7.4), 0.1% NaCl, and 0.1% Triton X-100. For the measurement of CGRP, 200 µl of sample were incubated at 4°C for 24 h with 100 µl of anti-CGRP antibody (antihuman CGRP II antibody; Peninsula Lab, Belmont, CA), which cross-reacts with both CGRP I and CGRP II from both humans and rats. Standard curves were established with synthetic CGRP (rat sequence), ranging from 2.5 to 1,000 pg/assay tube. Then 100 µl of ¹²⁵I-CGRP (human sequence; Amersham, Piscataway, NJ) were added to each tube and allowed to incubate for an additional 24 h at 4°C. Finally, 100 µl of goat antirabbit immunoglobulin G (Peninsula Lab) and 100 µl of normal rabbit serum (Peninsula Lab) were added and incubated for 2 h at room temperature, after which 0.5 ml of RIA buffer was added and centrifuged (1,700 × g, 4°C) for 20 min. After supernatant fractions had been decanted, the -radioactivity in the remaining pellet was counted. The level of SP-like immunoreactivity (LI) was measured in a manner similar to that for CGRP-LI described earlier.

In Situ Hybridization

Methods used to quantify PPT mRNA levels were modified from those described previously (14). Briefly, dorsal root, jugular, and nodose ganglia were harvested, frozen, and sliced into 12-µm sections. The sections of tissue were thaw-mounted onto glass slides and frozen (80°C) for in situ hybridization processing. Sections containing the midportion of the ganglia were rapidly dried, fixed in ribonuclease (RNase)-free phosphate-buffered 4% paraformaldehyde, then sequentially washed in phosphate buffer, diethylpyrocarbonate-treated water, acetic anhydride, triethanolamine buffer, and 2× saline sodium citrate (SSC) (1× SSC = 15 mM Na citrate and 150 mM NaCl). The riboprobe directed against -PPT mRNA, a splice variant from the ppt A gene encoding tachykinins (15), was generated from a -PPT complementary DNA insert excised from the plasmid pG1-PPT (generously provided by Dr. James E. Krause, Washington University, St. Louis, MO). The riboprobe was transcribed using 50 µM total uridine triphosphate (UTP) (12.5 µM [³⁵S]UTP and 37.5 µM unlabeled UTP and T7 polymerase). Preliminary experiments determined that 75 ng/ml was the optimal concentration of probe. The quantity of 50 µl of this concentration of probe was applied to each slide. Slides were then coverslipped and incubated at 55°C for 14 to 18 h in a humidified incubator. After hybridization, slides were immersed in 4× SSC to remove coverslips, rinsed in 4× SSC, treated with RNase-A, and stringently washed. After dehydration, slides were dipped in Kodak NTB-2 emulsion. Preliminary experiments determined that 6 d was the optimal time length of exposure. The level of mRNA was quantified using the Bioquant OS/2 Image Analysis System (R&M Biometrics, Nashville, TN). Cells were imaged under brightfield microscopy at a total magnification of ×160. At this magnification, the perimeter of each labeled cell was outlined so that the area of the cell covered by grains could be measured. All cells in each ganglia that were covered by grains were analyzed. Lighting and contrast levels were standardized before taking measurements to assure that all slides were assessed under the same conditions. Background was assessed by taking measurements over unlabeled cells outside the area of interest. Cells with a value of 5 times higher than background were considered labeled.

Experimental Protocols

Three series of experiments were carried out. (1) The first series was carried out to verify whether chronic smoking-induced airway hyperresponsiveness (AHR) is mediated through the release of tachykinins. Responses of R_L and C_dyn to a bolus injection (0.2 ml volume) of capsaicin (1.68 µg/kg, intravenously) were determined in each animal before and ~ 20 min after pretreatment with CP-99994 (0.3 mg/kg, intravenously), the selective antagonist of neurokinin (NK)-1 receptor, and SR-48968 (0.3 mg/kg, intravenously), the selective antagonist of NK-2 receptor. The lungs were hyperinflated (3 times tidal volume) periodically and also at 2 min before each injection to avoid pulmonary atelectasis. These responses were then compared between control (n = 6) and CS-exposed (n = 5) animals. (2) The purpose of the second series of experiments was to determine whether the SP-LI and CGRP-LI in bronchial tissue were increased in smoke-exposed animals. Bronchial tissue was harvested within 2 to 6 h after the last CS or sham (air) exposure; RIAs of SP-LI and CGRP-LI were performed and data compared between control (n = 8) and CS- exposed (n = 8) animals. (3) The third series of experiments was performed to determine whether the increase in SP-LI in the bronchial tissue of CS-exposed guinea pigs was accompanied by an increase in synthesis of -PPT mRNA in the sensory ganglia. Guinea pigs were exposed to either CS (n = 10) or sham (n = 10) treatment. Dorsal root, jugular, and nodose ganglia were harvested within 2 to 6 h after the last chronic CS exposure and immediately frozen on dry ice for in situ hybridization. The tissue was analyzed for amount of mRNA labeling per cell. To determine whether chronic CS exposure could increase the number of cells expressing tachykinins, we analyzed the number of labeled cells within a 0.36-mm² area of tissue.

Statistical Analysis

A one-way or three-way analysis of variance (ANOVA) was used for statistical analysis wherever appropriate. When the ANOVA showed a significant interaction, pairwise comparisons were made with a post hoc analysis (unequal n honestly significant difference). Data are reported as means ± standard error of the mean (SEM). P < 0.05 was considered significant.

Chemicals

All chemicals were obtained from Sigma Chemical Co. (St. Louis, MO) unless otherwise noted. Pancuronium bromide (Gensia Laboratories, Irvine, CA) and CP-99994 (Pfizer, Groton, CT) were each diluted in saline. Stock solution of capsaicin (400 µg/ml; Sigma) was prepared in a vehicle of 10% Tween 80, 10% ethanol, and 80% isotonic saline. SR-48968 (Sanofi Recherche, Montpelliex Cedex, France) was first dissolved in polyethylene glycol (average mol. wt. 200 g/mol; Sigma) and then diluted 1:1 (vol/vol) in saline to a final concentration of 0.67 mg/ml.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The average dose of total particulate matter delivered during the chronic smoke exposure was 17.88 ± 0.67 mg/kg/d, and the average blood arterial carboxyhemoglobin level measured in the CS-exposed animals at the end of the smoke exposure was 10.8 ± 1.2%. After exposure to CS for 14 to 17 consecutive days, the average body weight of CS-exposed animals was 406 ± 7 g, which was ~ 5% less than that of the matching control group (427 ± 6 g; P < 0.05).

Study 1. There was no difference between the control and CS-exposed animals in the baseline RL (control: 0.130 ± 0.006 cm H₂O/ml/s; CS: 0.139 ± 0.015 cm H₂O/ml/s; P > 0.05) and baseline C_dyn (control: 0.88 ± 0.06 ml/cm H₂O; CS: 1.05 ± 0.13 ml/cm H₂O; P > 0.05). Capsaicin (1.68 µg/ kg, intravenously) evoked an increase in P_tp immediately after the injection in both control and CS-exposed animals, but the response was markedly greater and lasted longer in the CS animals (e.g., Figure 1). To pool the data from all the animals for statistical analysis, we chose the six breaths immediately before the capsaicin injection as the baseline and the six consecutive breaths with peak increases in RL within 30 breaths after the capsaicin injection as the peak response. The same dose of capsaicin injection evoked significantly greater changes of RL and C_dyn from the baselines in CS-exposed animals (RL = 0.635 ± 0.135 cm H₂O/ml/s; C_dyn = 0.46 ± 0.13 ml/cm H₂O), but not in the control animals (RL = 0.201 ± 0.067 cm H₂O/ml/s; C_dyn = 0.04 ± 0.03 ml/cm H₂O) (Figure 2). These augmented responses to the same dose of capsaicin were completely abolished in the same group of CS- exposed animals after the pretreatment with a combination of CP-99994 (0.3 mg/kg) and SR-48968 (0.3 mg/kg) (Figures 1 and 2).

View larger version (48K): [in this window] [in a new window]   Figure 1.   Experimental record illustrating the responses of Ptp, respiratory flow (; inspiratory flow:positive deflection), and ABP to capsaicin before and 30 min after administration of CP-99994 and SR-48968 in a control (425 g) and a CS-exposed (415 g) guinea pig. Arrows indicate times of bolus injection of capsaicin (Cap; 1.68 µg/kg) into the venous catheter.

View larger version (21K): [in this window] [in a new window]   Figure 2.   Comparison of increased airway resistance elicited by bolus injection of capsaicin (1.68 µg/kg) between control and CS-exposed animals, and the effects of pretreatment with CP-99994 and SR-48968. Open bars, baseline; filled bars, peak response to capsaicin. In each animal, each data point was averaged over six consecutive breaths. *Significant difference from baseline; significant difference from control animals. Data represent means ± SEM of six guinea pigs in the control group and five in the CS group.

Study 2. In the CS-exposed animals (n = 5), the level of SP-LI in the bronchial tissue was 49.8 ± 10.7 fmol/g of tissue and that of CGRP-LI was 37.8 ± 6.8 fmol/g; both of these levels were significantly higher than those obtained from control animals (n = 5) (SP-LI: 20.0 ± 6.5 fmol/g, P < 0.05; CGRP-LI: 15.6 ± 4.6 fmol/g, P < 0.05; Figure 3).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Effect of CS inhalation on SP-LI and CGRP-LI in bronchial tissue from control (open bars) and CS-exposed (filled bars) animals. SP-LI and CGRP-LI were measured by RIA and expressed in femtomoles per gram of tissue. *Significant difference between control and CS-exposed animals. Data represent means ± SEM of eight guinea pigs in each group.

Study 3. Representative photomicrographs depicting silver grains that represent radiolabeled -PPT antisense mRNA probes hybridized to -PPT sense mRNA in nodose ganglia are shown in Figure 4 (4A and 4C, control animal; 4B and 4D, CS-exposed animal). Quantification of message was performed under darkfield optics (Figures 4A and 4B). Brightfield micrographs were used to confirm cell boundaries (Figures 4C and 4D).

View larger version (135K): [in this window] [in a new window]   Figure 4.   Emulsion autoradiograms showing the distribution of -PPT mRNA in the jugular ganglia. Darkfield photomicrographs of (A) control and (B) CS-exposed guinea pigs. Note the increase in the density of silver grains per cell, suggesting an increase in -PPT mRNA transcription and hence SP and NKA protein synthesis. Brightfield photomicrographs of (C ) control and (D) CS-exposed guinea pigs showing cell staining. Scale bar indicates 0.2 mm.

We assessed changes in -PPT mRNA concentrations by examining gene expression per cell (area covered by silver grains per cell) and by quantifying changes in the number of neurons that expressed detectable levels of -PPT mRNA. In jugular neurons, -PPT mRNA levels per cell in CS-exposed animals were significantly greater than levels in the control animals (Figure 5) (CS: 162.0 ± 10.5 µm²/ cell; control: 121.0 ± 13.9 µm²/cell; P < 0.05). We did not detect a difference between the CS and control groups in dorsal root ganglia (CS: 193.3 ± 13.5 µm²/cell; control: 169.0 ± 17.5 µm²/cell; P > 0.05) or in nodose ganglia (CS: 144.5 ± 23.8 µm²/cell; control: 104.8 ± 17.4 µm²/cell; P > 0.05). However, the data indicate that in nodose and dorsal root ganglia, levels of -PPT A mRNA tended to increase as well. Because sensory neuron ganglion slices have irregular areas that prohibit enumeration of cells per tissue as an accurate measure of labeled-cell density, we counted the number of labeled cells in a standardized area of 0.36 mm². We did not detect a difference in the number of labeled cells between the CS-exposed and control animals in all three ganglion types (data not shown).

View larger version (15K): [in this window] [in a new window]   Figure 5.   Enhanced expression of -PPT mRNA per cell in dorsal root, jugular, and nodose ganglia. In situ hybridization histochemistry was quantified by measuring the area covered by silver grains per cell. Open bars, tissue harvested from control animals; filled bars, tissue from CS-exposed animals. *Significant difference from control animals. Data are expressed as means ± SEM with 10 guinea pigs in each group except for the nodose ganglia of the CS group (n = 9).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study demonstrates marked increases in the SP-LI and CGRP-LI in the bronchial tissue of guinea pigs chronically exposed to CS; these increases were accompanied by an increase in the expression of -PPT mRNA, a precursor to the tachykinins, in jugular ganglion neurons. Further, our results have verified that chronic exposure to CS induces AHR to capsaicin injection in guinea pigs and that the augmented responses were completely abolished by pretreatment with a combination of NK-1- and NK-2-receptor antagonists, indicating a critical role of tachykinins. Together, these data strongly suggest that the synthesis of tachykinins in the vagal sensory neurons innervating the airways is elevated resulting from chronic exposure to CS. On the basis of these results, we suggest that the increased content of tachykinins in the C-fiber terminals leads to a greater quantity to be released in response to a given level of stimulation of these afferent endings, which in turn renders a more severe bronchoconstriction. Thus, this may explain, at least in part, the AHR observed in this study. Although capsaicin was the only bronchoactive substance used to test airway responsiveness in this study, the chronic smoking-induced hyperresponsiveness to other bronchoactive agents (e.g., acetylcholine, histamine, etc.) has been documented in various species, including humans (1, 11). We chose capsaicin in the present study because of its selective and potent stimulatory effect on C-fiber afferents and its ability to evoke the release of tachykinins from these nerve endings.

Tachykinins in the lungs and airways are synthesized in the cell body of small-size sensory neurons, primarily C neurons, located in the nodose and jugular ganglia (vagal afferents) and in the dorsal root ganglia between the levels of T₁ and T₆ (spinal sympathetic afferents) (8, 9). The majority of these peptides are then transported along the axon toward the peripheral terminals, where they are stored in large-granular vesicles and released upon a surge of calcium influx resulting from membrane depolarization. Once released, these neuropeptides can act on a number of effector cells (e.g., airway and vascular smooth muscles, cholinergic ganglia, various inflammatory cells, mucous glands, etc.) and produce potent local effects such as bronchoconstriction, extravasation of macromolecules, and edema of airway mucosa. Prolonged stimulation leads to neurogenic inflammation in the airways (8, 10). Further, increasing evidence suggests that endogenous tachykinins are involved in the AHR associated with airway mucosal inflammation (9, 10).

There are three distinct types of receptors that mediate the action of tachykinins and are located in the plasma membrane of these effector cellsNK-1, NK-2, and NK-3 receptorsbut only the first two are found in the respiratory tract (16, 17). In guinea-pig lungs, activation of the NK-1 receptor causes an increase in vascular permeability to plasma macromolecules and adhesion of leukocytes to the vascular endothelium. In contrast, NK-2 receptor activation primarily mediates airway smooth-muscle contraction (8, 9). Hence, although we did not test the blocking effects of the NK-1 and NK-2 receptor antagonists separately in this study, we suggest that chronic smoking-induced AHR is mediated mainly through the NK-2 receptor in the airway smooth muscle. The NK-1 and NK-2 receptors show preferential affinities for the two major types of tachykinins in the lungs, SP and NKA, respectively. Both SP and NKA are derived from the same gene, ppt A, and are frequently colocalized and coreleased from the same neurons (8, 10).

Our data revealed increased -PPT mRNA expression in jugular ganglia. To our knowledge, this is the first study to show an increase in tachykinin gene expression in the sensory neurons of animals chronically exposed to CS. These results are similar to previous observations showing that prolonged exposure of the lungs to other inhaled irritants and the consequent airway inflammation induces a substantial increase in the synthesis of tachykinins. For example, the level of PPT A mRNA expression was significantly elevated in the nodose ganglia after airway inflammation was induced by chronic exposure to ovalbumin (OVA) (18).

Whereas the increased tachykinin expression was as we expected, we were surprised that induction was less dramatic in nodose and dorsal root ganglia. One interpretation of this result is that the effects of chronic CS exposure are more severe at the more superficial structures (e.g., mucosa and submucosa) of the airways. As such, neurons emanating from the jugular would presumably be stimulated at a greater intensity. This interpretation seems to correlate well with a study by Hunter and Undem (19), who reported that guinea-pig airway epithelium is innervated almost exclusively by jugular ganglion neurons. Alternatively, tachykinin expression in the nodose and dorsal root ganglia may be modulated to a greater extent through post-translational mechanisms. Indeed, Killingsworth and colleagues made a similar proposal to explain why tachykinin mRNA levels did not increase when their animals were exposed to OVA-induced inflammation (20).

Sensory fibers originating from the nodose and dorsal root ganglia are known to innervate lungs and airways, and at least a portion of these fibers are SP-immunoreactive (8). Despite the finding that -PPT A mRNA expression failed to reach statistical significance in nodose and dorsal root ganglia, the trend of increased expression in these ganglia was also apparent. Further, we believe that the change in mRNA levels in the pulmonary neurons may have been underestimated in our study because the vagal afferents innervating the lungs and airways made up about only 20% of the total vagal afferent fibers (4). Presumably, the remaining 80% of the neurons would not be stimulated by CS and their steady-state tachykinin mRNA levels would remain relatively constant. Therefore, any mild increase in -PPT A mRNA expression in the subset of nodose and dorsal root ganglion neurons innervating the lungs and airways could be rendered insignificant when comparisons are made on the entire set of ganglion neurons that innervates other parts of the viscera (e.g., heart and gastrointestinal tract) in addition to the lungs and airways.

The increase in the amount of -PPT A mRNA in the experimental animals suggests increased synthesis of tachykinins in response to chronic CS exposure. We cannot rule out, however, that the apparent increase in the amount of mRNA could have resulted from decreased rates of mRNA translation or degradation. Regardless, the net increase in mRNA levels (Figure 5) coupled with the increase in tachykinin-LI in bronchial tissue (Figure 3) suggest that chronic CS exposure has increased the net rate of -PPT expression (i.e., tachykinin synthesis).

Another mechanism possibly involved in the increased amount of -PPT is that chronic CS could upregulate previously quiescent neurons to begin transcribing -PPT. Because the total area of each tissue varies from slice to slice, we selected 0.36 mm² as the standard area from which we enumerated labeled cells. We have ruled out this possibility because we did not detect a difference in the number of labeled cells between CS and control groups in any of the three ganglion tissues. However, several researchers have noted that some vagal neurons can undergo "phenotypic switching," wherein previously quiescent neurons begin to transcribe -PPT. Fischer and colleagues (18) reported this phenomenon in animals exposed to OVA-induced pulmonary inflammation. Further, nerve growth factor, which is elevated in allergically inflamed tissues (21), could cause neurofilament-immunoreactive nodose neurons (e.g., A neurons) to become immunoreactive to SP (22), suggesting the involvement of neurotrophins as a possible link between allergic airway inflammation and neuronal plasticity.

Although our data indicate an increase in tachykinin synthesis in the CS-exposed animals, other factors that may contribute to AHR should also be considered. For example, it has been suggested that airway mucosal injury and inflammation induced by chronic exposure to CS may enhance the sensitivity of the bronchopulmonary C-fiber endings in the lung; a given level of stimulus such as capsaicin may, therefore, trigger a greater intensity of discharge of these afferents and release a larger amount of tachykinins. This could happen even without an increase in tachykinin synthesis. Further, various components of CS are also known to depress the activity of neutral endopeptidase (NEP) that is present on the membranes of epithelium and nerve fibers in the airways and can cleave tachykinins immediately after their release (23). Hence, increased level of tachykinins in the bronchoalveolar lavage fluid could also be associated with the inhibited enzyme activity of NEP. In addition, the chronic smoking-induced AHR may be related to an increase in the densities of tachykinin receptors in the airway smooth muscle because it has been reported that the tachykinin-receptor mRNA expression is higher in the airways of human smokers than in those of nonsmokers (24). On the other hand, the possibility of an increase in the responsiveness of airway smooth muscle per se should also be considered because an increased bronchoconstrictive response to exogenous acetylcholine has been reported in guinea pigs chronically exposed to CS (11).

In conclusion, this study demonstrates that chronic exposure to CS induces an increased expression of -PPT mRNA in jugular ganglion neurons accompanied by an increase in the SP content in the bronchial tissue of guinea pigs, indicating an increased synthesis of tachykinins in the sensory neurons innervating the airways. Thus, a greater amount of tachykinins will be released when vagal C-fiber sensory terminals in the airways are activated, which may therefore contribute, at least in part, to chronic smoking-induced AHR.