Bronchoprotector Properties of Calcitonin Gene ... - ATS Journals

0 downloads 0 Views 737KB Size Report
to 114 g/kg) in anesthesized guinea pig had no effect per se on airway resistance but caused a dose ..... (13.5 g/kg) caused PIP to increase by 14.8 1.0 mm Hg (n.
Bronchoprotector Properties of Calcitonin Gene–related Peptide in Guinea Pig and Human Airways Effect of Pulmonary Inflammation ALAIN CADIEUX, NATHALIE P. MONAST, FRANÇOIS POMERLEAU, ALAIN FOURNIER, and CHANTAL LANOUE Department of Pharmacology, Faculty of Medicine, University of Sherbrooke, Sherbrooke; and INRS-Santé, University of Quebec, Pointe Claire, Quebec, Canada

Calcitonin gene–related peptide (CGRP), a neuropeptide released from sensory nerves during axonal reflexes, has strong bronchoprotector properties in rat isolated airways. In this study, we examined this ability of CGRP to prevent agonist-induced contraction in guinea pig and human airways and determined whether inflammatory reaction affects its function. CGRP administered intravenously (0.38 to 114 mg/kg) in anesthesized guinea pig had no effect per se on airway resistance but caused a doserelated inhibition of substance P (SP; 13.5 mg/kg)-induced bronchoconstriction (60% at 114 mg/kg). Similarly, CGRP (1029 to 1026 M) prevented in a concentration-dependent manner the contraction elicited by SP (5 3 1028 M) in guinea pig isolated main bronchi and parenchymal strips, the inhibition caused by CGRP being more pronounced in distal than in proximal airways (47 and 20%, respectively, at 1026 M). The breaking effect of CGRP on SP-induced constriction was however significantly reduced (p , 0.05) in guinea pig actively sensitized to ovalbumin (OA) and the loss in its potency was of similar magnitude (. 40%) whether it was administered in vivo or in vitro. A same phenomenon was observed in human isolated peripheral bronchi. CGRP (1026 M) reduced by more than 75% the extent of the contraction evoked by 1026 M of carbamylcholine and its protector effect was totally abolished in bronchi showing clear morphological manifestation of inflammatory reaction. It is concluded that CGRP acts as a potent bronchoprotector agent on both guinea pig and human airways but its ability to limit the extent of airway responsiveness is strongly impaired in inflammatory conditions. Cadieux A, Monast NP, Pomerleau F, Fournier A, Lanoue C. Bronchoprotector properties of calcitonin gene–related peptide in guinea pig and human airways: effect of pulmonary inflammation. AM J RESPIR CRIT CARE MED 1999;159:235–243.

Evidence has been provided that activation of axon reflexes, which are triggered in response to various airway irritants, may have relevance to the pathogenesis of asthma (1). This neural mechanism would involve impulse spread through antidromic transmission to collateral sensory nerve endings and release of sensory neuropeptides such as substance P (SP), neurokinin A (NKA), and calcitonin gene–related peptide (CGRP) (2, 3). In the skin, the action of these peptides is thought to be an important component of the plasma protein extravasation and the vasodilation that result from such neuronal stimulation (1, 3).

(Received in original form November 7, 1997 and in revised form August 11, 1998) This study was supported by the Medical Research Council of Canada (MRCC No. MT-13272) and by the “Association Pulmonaire du Québec” (APQ). A.F. is a scholar of the Fonds de la Recherche en Santé du Québec (FRSQ). N.P.M. has a studentship from the FRSQ-FCAR Santé. Correspondence and requests for reprints should be addressed to Alain Cadieux, Ph.D., Associate Professor, Department of Pharmacology, Faculty of Medicine, University of Sherbrooke, 3001, 12th Avenue North, Sherbrooke, PQ, J1H 5N4 Canada. Am J Respir Crit Care Med Vol 159. pp 235–243, 1999 Internet address: www.atsjournals.org

Once it was established that SP and NKA were predominantly associated with this neurogenic mechanism in the lung, they rapidly raised enthusiastic interest and caused an explosion of research that made these neuropeptides among the most investigated chemical neuromessengers of the mammalian respiratory system yet identified (3–5). Because of their ability to induce bronchospasms both in vitro and in vivo, to cause vasodilation, to increase vascular permeability, and to stimulate tracheobronchial mucous glands, they have been regarded as the favored candidates for the bronchial vascular permeability and airway smooth muscle contraction that occur in allergic inflammatory reactions of the lung (3–5). Argument in favor of this hypothesis is that deletion of these peptides from the lung attenuates or prevents inflammatory changes and bronchoconstriction induced by nerve stimulation and chemical stimuli (3–5). Even though SP and NKA appear to fulfill several of the criteria to which presumed mediators of asthma have to respond, their real contribution to the nonspecific airway hyperresponsiveness, which is the hallmark of bronchial asthma, still remains uncertain (1–5). In contrast to this manifested and very well justified interest toward these two tachykinins, very few questions have been raised with regard to the putative implications of CGRP

236

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

in the regulation of lung functions. Composed of 37 amino acid residues, with a disulfide bridge between Cys2 and Cys7 (6), CGRP is another well-known peptide that has been shown to be widely distributed in the lung innervation of several mammalian species including humans. Peripheral branches of CGRP-containing nerve fibers are located beneath and within the airway epithelium, around blood vessels and seromucous glands, and within the smooth muscle layers of the tracheobronchial tree (6–8). Costored with SP in primary afferent sensory neurons, CGRP is also released from nerve endings upon capsaicin stimulation and after exposure to chemical irritants (9). Although the possible roles of this neurotransmitter in lung physiological and pathological processes are unknown, its production in the vicinity of bronchial smooth muscle cells suggests that CGRP could also affect adjacent tracheobronchial tone. To that effect, CGRP is often reported as a potent bronchoconstrictor (2, 3) but evidence has been provided that this peptide is without any effect on airway smooth muscle tone (7, 9). We have previously shown that CGRP exerts a potent and concentration-related inhibition on bronchospasms induced by carbamylcholine and 5-HT in rat isolated airways. Interestingly, this bronchoprotective effect of CGRP was found to be more powerfully expressed in distal than in proximal airways, suggesting that CGRP might be of functional importance mainly in the periphery of the tracheobronchial tree (10). In view of the importance given to the sensory neuropeptides in the pathogenesis of asthma and because CGRP is also associated with neurogenic reflexes in the lung, we found it worthwhile to further investigate this interesting property of CGRP to downregulate the airway responsiveness. The present experiments were then carried out to determine whether CGRP had any bronchoprotective effect in guinea pig when given intravenously, and to evaluate whether immunization of guinea pigs could influence the pharmacological activity of CGRP both in in vivo and in in vitro conditions. Investigations were also carried out in human isolated airways for comparison purposes. A preliminary account of some of this work has been presented to the 1997 International Conference of the American Thoracic Society (11).

METHODS Animals and Sensitization Procedure Adult Hartley albino guinea pigs of either sex, weighing 200 to 250 g, were actively sensitized to ovalbumin (OA) according to the method of Schultz-Dale (12). They were injected with 100 mg intraperitoneally and 100 mg subcutaneously of OA on Day 1 and a further 10 mg intraperitoneally was administered as a booster on Day 8. They were used experimentally 14 or 15 d later. Control animals received the vehicle solution only. All research protocols conform to the guiding principles for animal experimentation and were approved by the Ethical Committee on Animal Research of the Medical School of the University of Sherbrooke.

In Vivo Measurements of Airway Responses Guinea pigs were anesthetized with xylazine (20 mg/kg, intramuscularly) and ketamine hydrochloride (130 mg/kg, intramuscularly). The jugular vein and trachea were cannulated and animals were mechanically ventilated with a constant-volume rodent respirator (model 683; Harvard, Southnatick, MA) delivering a tidal volume of 5.5 ml/kg at a frequency of 60 cycles/min via the tracheal cannula. Airway resistance was measured by the overflow technique according to the method of Konzett and Rössler (13). Briefly, ventilation overflow, that is, the pulmonary insufflation pressure (PIP), was recorded at the side arm of the tracheal cannula by means of a differential pressure transducer (Gould P23 ID; Grass Instrument Co., Quincy, MA) connected to a Grass polygraph (model 7D; Grass Instrument Co.). Airway responsiveness was assessed by measuring the changes in PIP following in-

VOL 159

1999

travenous injections of the agonists. Changes in PIP, which were taken as an index of changed airway resistance, were measured and expressed in mm of mercury (mm Hg). Each agonist was injected through the jugular vein cannula in a volume of 100 ml followed by a washout with 200 ml of heparinized saline (100 U/ml). Before starting the experiment, guinea pig airway responsiveness was first tested by a single intravenous administration of 25 mg/kg of acetylcholine. Based on preliminary experiments, this resulted in an airway constriction which averaged 28.6 6 1.5 (n 5 36) and 24.0 6 1.0 mm Hg (n 5 38) in control and OA-sensitized animals, respectively. Only guinea pigs (control and OA-sensitized) showing a reasonable strong (. 20 mm Hg) airway constriction to the cholinergic agonist were used in this study. Thereafter, SP which was used as the main spasmogen, was injected intravenously every 20 min to induce bronchoconstriction. Care was taken to choose a dose of SP (13.5 mg/kg) so that the same extent of airway constriction could be induced in both animal groups. SP (13.5 mg/kg) was then administered repeatedly (approximately 3 times) until a reproducible constriction (control response) was obtained. After two SP-induced bronchoconstrictions of similar amplitude, CGRP (0.38 to 114 mg/kg) was administered 5 min before a further challenge with SP and the SP challenge was then repeated. The resulting constriction was compared with the control response, which was defined as the mean of the two preinhibition bronchoconstrictions. In some experiments, systemic arterial blood pressure was monitored at the same time via a polyethylene catheter filled with heparinized saline inserted into a carotid artery and linked to a pressure transducer (Gould P23 ID) and a Grass polygraph.

In Vitro Measurements of Airway Responses Guinea pigs were killed by exsanguination following intraperitoneal injection of pentobarbitone sodium (50 mg/kg). The trachea and the lungs were quickly removed and placed in cold Krebs solution. Both right and left main bronchi were dissected out and cut into spirals. Parenchymal strips (approximately 15 mm 3 3 mm 3 3 mm) were cut from the periphery of either right or left lower lobes of the lungs (10). Each preparation was mounted in a 5-ml organ bath containing Krebs solution (378 C) bubbled with 5% CO2 in O2. Contractions were measured isometrically with a Grass FT03 force displacement transducer and recorded on a Grass polygraph (model 7D) as tension changes (g). All tissues were subjected to an initial loading tension of 1 g and allowed to equilibrate for 60 min (with changes of bath medium every 15 min) before experimentation began. The inhibitory effect of CGRP on SP-induced contractions was investigated essentially as described previously (10). Each tissue preparation was first primed with a concentration of carbamylcholine (1026 M) to evaluate its responsiveness. When the basal tone was reestablished (after a few washouts), contractions to SP (5 3 1028 M) were elicited at 15-min intervals with a drug contact time just sufficient to record the peak response. CGRP (1029 to 1026 M) was then added to the organ bath and 5 min later a further spasmogen challenge occurred. The contraction produced in the presence of CGRP was then compared with the mean of two control responses to SP observed prior to CGRP administration. A further concentration of CGRP was tested only when responses to SP showed no further recovery. Concentrations of CGRP were tested in a random order in each experiment. It is worthy of note that the inhibitory effect of CGRP on contractions induced by SP was also examined at a dose of SP (5 3 1028 M) that induced similar maximal responses in tissues obtained from each group of animals.

Isolated Human Airways Macroscopically normal lung specimens were obtained from six patients undergoing surgery for carcinoma and from autopsy (n 5 9) performed within 6 h of death. The study protocol was approved by the Ethical Committee of the University Hospital. The ages of the subjects (all males) ranged from 17 to 88 yr (mean 6 SEM: 57.4 6 5.1 yr) and none had a known history of either asthma or atopy. The lung specimens were transported to the laboratory in ice cold preoxygenated Krebs solution. Suitable cartilaginous bronchi (internal diameter 2 to 4 mm) were dissected free of adjacent tissue and were either used immediately (autopsy cases) or after 24-h storage in preoxygenated Krebs solution at 48 C (14, 15). Isolated bronchi were cut into helical

Cadieux, Monast, Pomerleau, et al.: Bronchoprotector Properties of CGRP

237

strips and set up for isometric recording (initial tension: 1 g) under the conditions described for the isolated guinea pig airways. They were primed by adding 1023 M of carbamylcholine followed by washouts, which ensured a stable function for the rest of the experiment. Protocol for administration of compounds was also the same for animal airways except that the agonist used to elicit contractions was carbamylcholine (1026 M) instead of SP, the latter being much less potent than the cholinergic agonist in human airways (16). As was previously reported, our preliminary data also revealed that tissues obtained from autopsy showed no difference from those obtained at surgery in their responses to drugs (14) and that a 24-h storage did not either affect agonist’s response (15).

Morphometric Studies Samples of lung tissue were collected in close vicinity of each airway preparation dissected out from both guinea pig (control and OA-sensitized) and human lungs. They were fixed in Bouin’s solution and processed in a standardized conventional manner for embedding in paraffin wax. Each block was sectioned at 100-mm intervals to yield six sections of 5 mm thickness. Sections were stained with hematoxylin–eosin and examined by light microscopy. Two to three nonoverlapping fields per section were analyzed for pathological changes. Histological examination focused mainly on inflammatory reactions, such as cellular infiltration, interstitial or alveolar edema, and airway obstruction. All estimations were performed double-blind by an observer who was not aware of the origin of the tissues and of the results of the pharmacological studies.

Chemicals and Solutions The Krebs solution had the following composition (mM): NaCl 118.1, KCl 4.7, MgSO4 .7 H2O 1.2, KH2PO4 1.2, NaHCO3 25, CaCl2 2.5, and glucose 11.1 (pH 7.4). Carbamylcholine chloride, heparin, acetylcholine chloride, and ovalbumin (grade V) were purchased from Sigma Chemical Co. (St. Louis, MO). SP and rat aCGRP (CGRP) were synthesized in our laboratories. Carbamylcholine was freshly diluted in the stock Krebs solution while heparin, acetylcholine, and OA were prepared in saline (0.9% NaCl). SP and CGRP, in lyophilized form, were dissolved in distilled water at an initial concentration of 1 mg/ml and kept frozen at 2208 C. The required working solutions were made by dilution either in saline or in the stock Krebs solution immediately before the experiments. In in vivo experiments, the doses are expressed in terms of their corresponding base. In in vitro experiments, the concentrations referred to are final bath concentrations.

Figure 1. Original tracings showing changes in airway resistance (PIP) and mean arterial blood pressure (MAP) induced by intravenous injections of SP in control (A) and OA-sensitized (B) guinea pigs. Airway resistance and blood pressure were monitored simultaneously in the same animal. Closed circles 5 injection of SP (13.5 mg/kg) every 20 min; arrows 5 injection of CGRP (3.8 mg/kg) 5 min before administration of SP. Ordinate scale: changes of PIP and MAP in mm Hg. Abscissa scale: time in min.

CGRP, did not affect the hypotension evoked by CGRP (Figure 1). The protective effects of increasing doses of CGRP on airway constriction induced by SP in both control and OA-sensitized guinea pigs are illustrated in Figure 2. When added 5 min before SP, CGRP (0.38 to 114 mg/kg) prevented bronchoconstriction in a dose-related manner in control animals, provid-

Statistical Analysis All data are expressed as means 6 SEM. Throughout means values were compared by use of Student’s t test for unpaired data. A probability level of p , 0.05 was considered statistically significant.

RESULTS Effect of CGRP on SP-induced Airway Constriction in Anesthetized Guinea Pigs

Baseline PIP in anesthetized guinea pigs averaged 6.0 6 0.2 (n 5 36) and 6.2 6 0.2 (n 5 38) mm Hg in control and OA-sensitized animals, respectively. Intravenous administration of SP (13.5 mg/kg) caused PIP to increase by 14.8 6 1.0 mm Hg (n 5 57) in control animals and by 18.1 6 2.2 mm Hg (n 5 10) in OA-sensitized guinea pigs (p , 0.05). As illustrated in Figure 1, CGRP (3.8 mg/kg) did not affect baseline airway resistance by itself, but potently inhibited the response to SP in nonsensitized animals. In contrast, CGRP failed to significantly prevent the SP-induced increase in airway constriction in OAsensitized guinea pigs. In anesthetized animals, SP injected intravenously elicited not only airway constriction but also transient systemic hypotension. As expected, CGRP, which is recognized as a potent vasodilator (3), also caused a decrease in blood pressure, its effect being more pronounced and more prolonged than that of SP. Interestingly however, SP, when administered after

Figure 2. Inhibitory effect of CGRP on SP-induced increases in airway resistance (PIP: mm Hg) in control and OA-sensitized guinea pigs. Each column and bar represents the mean 6 SEM of data obtained from 8 to 57 animals (numbers on top of columns). CGRP (0.38 to 114 mg/kg) was injected 5 min before administration of SP (13.5 mg/kg). Asterisks 5 significantly different (p , 0.05) from SP injection without a preadministration of CGRP (filled columns).

238

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

ing a 60% inhibition at the highest dose used in this study. In contrast, the protective effect of CGRP on SP-induced airway constriction was deeply attenuated in OA-sensitized animals. At 114 mg/kg, the inhibition provided by CGRP reached only 36%, which means a reduction by more than 40% in its capability to exert its bronchoprotective action. In experiments where the systemic arterial blood pressure was also monitored, there was no difference in the magnitude of vasorelaxation induced by CGRP between the two groups of animals whatever the dose of CGRP used. For example, at 38 mg/kg, CGRP reduced the blood pressure by 21.5 6 1.5 (n 5 13) and 22.5 6 3.0 mm Hg (n 5 5) in control and OA-sensitized guinea pigs, respectively. Effect of CGRP on SP-induced Contraction in Guinea Pig Isolated Airways

SP (5 3 1028 M) caused similar increases in tension in both control and OA-sensitized guinea pig isolated airways (Figure 3). The magnitude of the contractions varied between 15 and 30% of maximal responses, depending on the airway preparation used (bronchus versus parenchymal strips) and the origin of the tissues (control versus OA-sensitized animals). As illustrated in Figure 3, preincubation of parenchymal strips with CGRP (1028 M) caused a slight reduction of SP-induced contraction in control tissue but did not alter the amplitude or the shape of the SP-evoked response in parenchyma from OAsensitized animals. Pretreatment with higher concentrations of CGRP (for example, 1026 M) significantly inhibited the contractions induced by SP in both groups of tissue, the protective effect of CGRP being more marked in parenchymal strips obtained from nonsensitized guinea pigs (Figure 3). Similar results were obtained in isolated bronchi, CGRP being more prone to prevent SP-induced response in control than in OAsensitized tissues (not shown).

VOL 159

1999

Expressed in terms of percentage of contraction developed by SP alone, CGRP (1029 to 1026 M) inhibited contractions of control airways in a concentration-dependent manner (Table 1). The antibronchoconstrictor effect of CGRP was however more marked in the lower segments of the tracheobronchial tree and at a concentration of 1026 M, reached an inhibitory effect of 47.3 6 3% (n 5 4) in the parenchymal strips and 20.3 6 1.5% (n 5 3) in the main bronchus. Similar concentrationdependent inhibitory effects of CGRP were observed in parenchymal strips obtained from OA-sensitized guinea pigs except that the maximal inhibition elicited by 1026 M of CGRP reached only 18.7 6 4.8% (n 5 5). Moreover, CGRP failed to significantly prevent the SP-induced bronchoconstriction in the main bronchi from these same animals (Table 1). Effect of CGRP on Carbamylcholine-evoked Contraction in Human Isolated Airways

Two types of responses were recorded in human airways after the administration of 1026 M of carbamylcholine. Despite the fact that the magnitude of the contraction elicited by carbamylcholine was similar in all tissues (0.1 6 0.02 g; n 5 15), some responses were fast in onset whereas others showed slowly developing increase in tone, which reached a plateau after 10 to 15 min (Figure 4). These two types of responses to carbamylcholine were observed in tissues obtained either from surgical resection (3 of 6) or from autopsy (3 of 9). CGRP (1029 to 1026 M), added 5 min prior to carbamylcholine, had no effect per se on the airway tone but prevented in a concentrationrelated manner the contraction in airways showing rapid response to the cholinergic agonist, providing more than 75% inhibition at the highest concentration tested. However, preaddition of CGRP (1029 to 1026 M) in slowly responding isolated bronchi had no significant effect on the magnitude or duration of carbamylcholine-induced contractions (Figure 5).

Figure 3. Original recordings of the inhibitory effect of CGRP on contractions evoked by 5 3 1028 M SP in isolated parenchymal strips from (A) control and (B) OA-sensitized guinea pigs. Contractions were elicited by SP (triangles) at 15-min intervals in the absence or in the presence of various concentrations of CGRP. Note that the inhibitory action of CGRP is more pronounced in tissues from nonsensitized animals. In each panel, the two recordings were obtained from different tissues. Ordinate scale: changes of tension in g. Abscissa scale: time in min. W 5 washouts.

Cadieux, Monast, Pomerleau, et al.: Bronchoprotector Properties of CGRP

239

TABLE 1 CONCENTRATION-DEPENDENT INHIBITION OF RESPONSES TO SP (5 3 1028 M) BY CGRP IN ISOLATED AIRWAYS FROM CONTROL AND OA-SENSITIZED GUINEA PIGS* % Inhibition Concentration CGRP (2log M) 9 8 7 6

Control

OA-sensitized

Bronchus

Parenchyma

Bronchus

Parenchyma

0.0 (3) 8.0 6 2.7 (6) 13.2 6 2.0 (5) 20.3 6 1.5 (3)

1.4 6 0.2 (3) 17.3 6 3.0 (7) 28.2 6 6.0 (5) 47.3 6 3.0 (4)

0.0 (3) 0.0 (3) 4.2 6 0.6† (4) 7.2 6 4.5† (5)

0.0 (3) 1.8 6 1.0† (3) 7.8 6 2.6† (6) 18.7 6 4.8† (5)

* Data are shown as mean 6 SEM of values from 3 to 7 animals (number in parentheses). † p , 0.05 versus control.

Morphometric Analysis

The predominant histopathological manifestation observed in lungs from OA-sensitized guinea pigs consisted of a diffuse interstitial inflammatory process involving particularly the parenchyma and the peripheral airways. In the stroma, the infiltrate consisted largely of chronic inflammatory cells with lymphocytes, plasma cells, and monocytes, resulting in a thickening of the alveolar septae (Figures 6A and 6C). Cellular infiltration, enriched in eosinophils, was also present around and within the compartments of the bronchial walls. There was no evidence of bronchial epithelial damage but several small airways were filled with plugs consisting of macrophages, mucus, lymphocytes, and eosinophils (Figures 6B and 6D). Inflammatory cells were also observed in the adventitia of small blood vessels, which were often surrounded by plasma effusion (edema) (not shown). No such histological changes were observed in lungs from nonsensitized animals. High variability was seen among human lung specimens with respect to the level of local inflammatory reactions. All of the 15 lung samples analyzed had inflammatory cells in the al-

Figure 5. Concentration-dependent inhibition of responses to carbamylcholine (1026 M) by CGRP in human isolated peripheral airways. CGRP was added in a manner similar to that described in the legend to Figure 3. Ordinate scale: degree of inhibition produced by CGRP expressed as percentage of contraction developed by carbamylcholine alone. Abscissa scale: concentration of CGRP. Each point and bar represents the mean 6 SEM of data from 3 to 7 tissues. Solid lines 5 airways showing fast (closed squares) and slow (closed circles) onset of responses to 1026 M of carbamylcholine.

veolar septae, but most were neutrophilic granulocytes and mononuclear cells; eosinophils were rare. Histological changes varied from slight perivascular and peribronchiolar local inflammation (Figures 7A and 7B) to massive pulmonary edema and intra-alveolar hemorrhage, the inflammatory reaction being however always much more severe in tissues demonstrating slow-onset response to carbamylcholine (Figures 7C and 7D). In addition, small airways from these same lung specimens (with slow-onset response) were often severely obstructed by mucus, collagen deposits, and inflammatory cells (Figure 7D). In many of them the epithelial integrity was not maintained (not shown).

DISCUSSION

Figure 4. Original recordings showing (A) fast and (B) slow onset of responses to 10 26 M carbamylcholine (arrows) in human isolated peripheral bronchi. In each panel, tracings refer to tissues obtained from different subjects. Ordinate scale: changes of tension in g. Abscissa scale: time in min. W 5 washouts.

The results of this study demonstrate that in guinea pig airways, CGRP exerted a potent and dose-related inhibitory effect on SP-induced contraction both in vitro and in vivo. Similarly, CGRP prevented in a concentration-related manner carbamylcholine-evoked contractions in human isolated peripheral bronchi. However, the bronchoprotective effect offered by CGRP in both mammalian species was found to be strongly attenuated and even vanished in airway preparations showing clear manifestations of local inflammatory reaction. This indicates that this property of CGRP to exert a braking effect on airway smooth muscle tone may be strongly altered in inflammatory conditions. Numerous in vitro and in vivo investigations have been carried out with the aim of determining the role of CGRP on tracheobronchial smooth muscle cells. Although several of these studies have, to a large extent, focused on the putative myotropic activity of CGRP, its effect on airway smooth muscle tone still remains unclear, as considerable controversial data have been reported. For example, in human isolated airways, evidence has been provided for (3) and against (9) a bronchoconstrictor action of CGRP. Similarly, Luts and coworkers (17), Bhogal and coworkers (18), and Pinto and coworkers (19) showed that CGRP does not exert any contractile or re-

240

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

Figure 6. Photomicrographs comparing the morphology of lung architecture in control (A, B) and OA-sensitized (C, D) guinea pigs. Lung samples were collected in areas adjacent to the tissue preparations used for the bioassays. Note that in OA-exposed animals, alveolar walls are thickened from the presence of lymphocytes, plasma cells, and monocytes (C). Small bronchi are also surrounded by an inflammatory infiltrate and partially obstructed by the presence of plugs consisting of macrophages, mucus, lymphocytes, and eosinophils (D). Hematoxylin–eosin staining. Original magnifications of A, C: 3250; B, D: 3335.

laxant effect in guinea pig trachea, whereas Hamel and FordHutchinson (20) and Tschirhart and coworkers (21) reported that CGRP is one of the most potent contractile agents on guinea pig trachea yet identified. On the other hand, CGRP has been found to have no effect on hilus bronchi and parenchyma in the guinea pig in vitro (9, 19, 20) and to cause no changes in airway resistance in guinea pig in vivo (7, 9, 22–24). More recently CGRP has been reported to potentiate the cholinergic contractions elicited by electrical field stimulation (EFS) in guinea pig trachea (19), to produce no bronchoconstriction in isolated and perfused guinea pig lung (25), to induced relaxations in guinea pig trachea precontracted with KCl and prostaglandin F2a (PGF2a) (18), but to be without any relaxant effect in human and guinea pig bronchi precontracted with NKA (9) and in guinea pig trachea precontracted with histamine (19). In other animal species, CGRP has been shown to have no contractile effect on the airway smooth muscle in the rabbit (26), pig (27, 28), rat (10, 20), and mouse (29) in vitro and in the sheep in vivo (30). Finally, CGRP has been noted to cause relaxations in pig trachea (27) and bronchi (28)

VOL 159

1999

Figure 7. Photomicrographs comparing the morphology of lung architecture in human airway preparations showing fast (A, B) and slow (C, D) onset of responses to carbamylcholine. Lung samples were collected in areas adjacent to the peripheral bronchi used for the bioassays. Note that the inflammatory reaction was much more severe in C and D than in A and B. Hematoxylin–eosin staining. Original magnifications of A, B, D: 3300; C: 3250.

precontracted with carbamylcholine and histamine, respectively, and in mouse airways precontracted with carbachol (29). Given this wide spectrum of pharmacological activities that are currently reported for CGRP, these results would easily suggest that the effect of this sensory neuropeptide greatly differs between animal species and different levels of the tracheobronchial tree. When analyzing in greater detail all these data, it is interesting to note that they are not as conflicting as they first appear to be. For example, regardless of the origin of the airway preparation used (animal species and level of the tracheobronchial tree), most studies examining the effect of CGRP on the baseline tension of the airway smooth muscles resulted in identical data, as no contractile or relaxant response was observed (9, 10, 17–20, 23, 24, 26, 27, 29). In studies showing a contractile action for CGRP (for example, in guinea pig trachea), there is a consensus regarding the epithelium dependency for this effect of the peptide but the magnitude of the response evoked by CGRP was each time described as being weak and of minor importance: 30% of the maximum induced with 1.5 3 1025 M of methacholine (20), 14% of the capsaicin (1025 M) maximum (21), and 8% of 50 mM KCl-induced contraction (31). Thus, when compared with SP and NKA, which

Cadieux, Monast, Pomerleau, et al.: Bronchoprotector Properties of CGRP

are well recognized potent bronchoconstrictor agents (5), the data yet accumulated on the pharmacological effects of CGRP do not really speak in favor of an efficient and potent constrictor activity for this sensory neuropeptide in airway smooth muscle. To that effect, it is worthy of note that in all experiments measuring pulmonary resistance in vivo, no bronchoconstriction or bronchodilation was detected in guinea pigs (7, 9, 22–24) and sheep (30) intravenously administered CGRP. From the studies reporting a bronchodilatory effect with CGRP, a similar statement of fact can be made, as the relaxations evoked by CGRP were characterized as being weak (27, 28), modest (18), very small and even erratic (29). Thus from the limited data yet available, it appears a bit premature and probably too strong a statement to claim that CGRP has to be regarded as a potent and efficient bronchodilatory agent. In the light of these observations, it is then tempting to speculate that either CGRP may be related to physiological processes other than those involved in the regulation of bronchomotor tone or that the predominant effect of CGRP on airway smooth muscle cells differs from a direct and potent myotropic action. Our results would support the latter hypothesis as we found that exogenously applied CGRP produces no myotropic action per se in guinea pig and human isolated airways, neither does it cause changes in airway constriction when injected intravenously in anesthetized guinea pigs. This would be in keeping with numerous observations previously made by several other groups of investigators, as discussed previously. Interestingly however, it was noted that behind this apparent lack of pharmacological activity, CGRP presents a marked ability to prevent agonist-induced bronchoconstrictions in both in vitro and in vivo conditions. In isolated airways from control guinea pigs, the protective effect offered by CGRP on airway smooth muscle tone was found to be concentrationdependent and more pronounced on parenchymal strips than in main bronchi. Accordingly, in human peripheral bronchi showing a rapid onset of response to carbamylcholine, the magnitude of the inhibition caused by CGRP was even greater than that observed in guinea pig. At the highest concentration of CGRP used in our study, the responsiveness of the distal airways to the cholinergic agonist was reduced by more than 75%. All these results would be consistent to effects in rat airways, in which it was also observed that this ability of CGRP to exert a bronchoprotector action was inversely related to the size of the airways: the smaller the caliber, the greater the inhibition (10). Thus, in contrast to the conflicting data currently reported on the myotropic effect of CGRP, our results would suggest that there are no interspecies variations in this capability of CGRP to act as a modulatory agent in mammalian isolated airways. In fact, our findings are not altogether unprecedented: recently CGRP has been shown to produce a concentration-dependent inhibition on both the nonadrenergic, noncholinergic contraction induced by EFS and the SPevoked contractile response in guinea pig bronchi in vitro (19) and to prevent leukotriene D4–elicited increase in airway resistance in guinea pig in vivo (24). Previously, we showed that CGRP also prevents in a concentration-related manner contractions elicited by carbamylcholine and 5-hydroxytryptamine in rat isolated airways (10). Thus, although very little is known about this newly identified bronchoprotector property of CGRP (10, 19, 24), the data yet accumulated indicate that CGRP may act as a nonspecific modulatory agent and that it may exert its braking effect toward a wide spectrum of chemically different contractile agonists. This makes this peptide probably the first natural and endogenous bronchoprotector agent yet discovered.

241

The ability of CGRP to prevent SP-induced airway constriction in anesthetized guinea pigs has been previously reported by Gatto and coworkers (22). These investigators showed that co-infusion of CGRP with SP or bombesin caused a marked attenuation in both SP- and bombesin-induced increases in airway resistance. Because CGRP produced no bronchoconstriction when injected alone, they concluded that CGRP antagonized the effect of SP or bombesin without affecting bronchomotor tone. Our data confirm and extend this observation by demonstrating that the bronchoprotective effect of CGRP is dose-dependent and that it can also be observed when CGRP is injected prior to the administration of SP. On the one hand, this suggests that CGRP may first interact with specific receptor sites in order to exert its breaking effect, which would be in line with the demonstration of receptors with high affinity for CGRP in guinea pig lungs (32). To that effect, it is worth noting that the protection offered by CGRP on SP-induced increase in airway resistance in vivo was significantly detected when a relatively low dose (3.8 mg/kg) of CGRP was used. On the other hand, these data indicate that CGRP can gain access to its specific receptors located in the lung via the general circulation in a relatively quite active state, which attests to a putative role for CGRP in the regulation of airway smooth muscle tone in vivo. The observation that CGRP was more efficient in attenuating the agonist-induced contractions in distal than in proximal airways is also of particular interest as it corresponds to the reported pattern of distribution of CGRP receptor sites in mammalian lungs. CGRP binding sites are very abundant in peripheral airways, including the alveolar walls, but are sparsely distributed in smooth muscle of large airways (33). CGRP may therefore exert its braking effect mainly in the periphery of the tracheobronchial tree, which means in airways that are usually responsible for the increases in lung resistance following chemical or allergic stimuli. Because it has been demonstrated that CGRP is released during anaphylaxis (1, 3) and that its plasma level is increased during acute exacerbations of asthma (34), the bronchoprotective effect reported in this study could invest CGRP with a new and putatively very important role in the regulation of bronchomotor tone in different kinds of airway diseases including bronchial hyperresponsiveness. To examine this question, experiments were repeated in guinea pigs actively sensitized to OA. It was found that the braking effect of CGRP on SP-induced increase in airway resistance was markedly impaired in the OA-exposed guinea pigs when compared with control animals. Moreover, this impairment was of similar magnitude (reduction by more than 40%) whether CGRP had been administered in vivo or in vitro. The bronchoprotective effect of CGRP on carbamylcholine-evoked contractions was also abolished in human peripheral bronchi showing a slow-onset response to the cholinergic agonist. Because morphometric analysis revealed that these latter tissues were specifically characterized by the presence of a clear and marked inflammatory reaction, as was globally the case in airways from OA-sensitized guinea pigs, our results suggest that airway inflammation alters the bronchoprotector property of CGRP in both mammalian species, whether its nature be allergic (guinea pig) or nonallergic (human subjects). From our study we cannot ascertain the exact reasons for the change in the contractile velocity of the human inflamed bronchi as well as for this impairment in the property of CGRP to fully exert its braking effect in such tissues. Even though it has often been reported that inflamed airways may present functional abnormalities with regard to their contractile or relaxant properties, there is no consensus in the litera-

242

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

ture on the mechanisms underlying such alterations (35). The fact that the cholinergic agonist generated isometric forces of a same magnitude in both groups of human bronchi could suggest that the change observed in the onset of response was ascribed to a difficulty of diffusion for carbamylcholine in the inflamed tissues to reach its receptors in the airway smooth muscle cells. A similar hypothesis could be put forward for the loss in the potency of CGRP in both mammalian species whether it was administered in vitro or in vivo. The reduction in the extent of bronchoprotection offered by CGRP could also be explained by enzymatic degradation. It is possible that because CGRP is metabolized more quickly in inflamed tissues, as is the case for several other neuropeptides (3–5), there is less of it available to protect the airways against the action of contractile agonists. Clearly, characterization of the events leading to this impairment in the bronchoprotector properties of CGRP would be important to pursue, and further investigations on the mechanism underlying this new and unique effect of this sensory neuropeptide would be necessary to reveal its exact role in the lung. Our findings on the bronchoprotector properties of CGRP may have relevance to the pathogenesis of some airway disorders. For example, bronchial asthma is defined as an inflammatory disease characterized by plasma extravasation, edema, and bronchial hyperresponsiveness. Although the exact cause of this increase in airway responsiveness is still unclear, a large variety of inflammatory mediators and neuropeptides have been implicated (1–5). As mentioned earlier, SP and NKA have received particular attention because of their ability to mimic many features of asthma, including bronchial hyperresponsiveness. Evidence for the involvement of CGRP however has been lacking. Because several inflammatory mediators typically found in tissues after allergen challenge can evoke activity in sensory nerve fibers (1–5) and consequently the release of CGRP, we hypothesize that CGRP, by its braking effect, could serve to limit the extent of airway responsiveness. An impairment in its capability to exert its bronchoprotective effect in inflamed airways might significantly contribute to the bronchial hyperresponsiveness associated with asthma. In conclusion, we have shown that CGRP possesses a marked ability to make guinea pig and human peripheral airways less sensitive to the action of spasmogenic agents and that this bronchoprotector property of CGRP is markedly attenuated in inflammatory conditions.

8.

9.

10.

11.

12. 13. 14. 15.

16.

17.

18.

19.

20.

21.

22.

23.

Acknowledgment : The authors thank Ms. Pierrette Carrier for excellent secretarial assistance.

References 1. McDonald, D. M. 1994. The concept of neurogenic inflammation in the respiratory tract. In M. A. Kaliner, P. J. Barnes, G. H. H. Kunkel, and J. N. Baraniuk, editors. Neuropeptides in Respiratory Medicine. Marcel Dekker, New York. 321–350. 2. Barnes, P. J. 1986. Asthma as an axon reflex. Lancet 1:242–245. 3. Barnes, P. J., J. N. Baraniuk, and M. G. Belvisi. 1991. Neuropeptides in the respiratory tract: part II. Am. Rev. Respir. Dis. 144:1391–1399. 4. Barnes, P. J., J. N. Baraniuk, and M. G. Belvisi. 1991. Neuropeptides in the respiratory tract: part I. Am. Rev. Respir. Dis. 144:1187–1198. 5. Joos, G. F., P. R. Germonpre, J. C. Kips, R. A. Peleman, and R. A. Pauwels. 1994. Sensory neuropeptides and the human lower airways: present state and future directions. Eur. Respir. J. 7:1161–1171. 6. Rosenfeld, M. G., J. J. Mermod, S. G. Amara, L. W. Swanson, P. E. Sawchenko, J. Rivier, W. W. Vale, and R. M. Evans. 1983. Production of a novel neuropeptide encoded by the calcitonin gene via tissuespecific RNA processing. Nature 304:129–135. 7. Lundberg, J. M., A. Franco-Cereceda, X. Hua, T. Hökfelt, and J. A. Fischer. 1985. Co-existence of substance P and calcitonin gene–related peptide–like immunoreactivities in sensory nerves in relation to car-

24.

25.

26.

27.

28.

29. 30.

VOL 159

1999

diovascular and bronchoconstrictor effects of capsaicin. Eur. J. Pharmacol. 108:315–319. Cadieux, A., D. R. Springall, P. K. Mulderry, J. Rodrigo, M. A. Ghatei, G. Terenghi, S. R. Bloom, and J. M. Polak. 1986. Occurrence, distribution and ontogeny of CGRP immunoreactivity in the rat lower respiratory tract: effect of capsaicin treatment and surgical denervations. Neuroscience 19:605–627. Martling, C. R., A. Saria, J. A. Fischer, T. Hökfelt, and J. M. Lundberg. 1988. Calcitonin gene–related peptide and the lung: neuronal coexistence with substance P, release by capsaicin and vasodilatory effect. Regul. Pept. 20:125–139. Cadieux, A., C. Lanoue, P. Sirois, and J. Barabé. 1990. Carbamylcholineand 5-hydroxytryptamine-induced contraction in rat isolated airways: inhibition by calcitonin gene–related peptide. Br. J. Pharmacol. 101: 193–199. Monast, N. P., C. Lanoue, A. Fournier, and A. Cadieux. 1997. Calcitonin gene–related peptide: an inhibitory agent in guinea pig airways (abstract). Am. J. Respir. Crit. Care Med. 155:A486. Chand, N., and P. Eyre. 1978. The Schultz-Dale reaction: a review. Agents Actions 8:171–184. Konzett, H., and R. Rössler. 1940. Versuchsanordnung zu untersuchungen an der bronchialmuskulatur. Arch. Exp. Pathol. Pharmakol. 195:71–74. Richardson, J., and J. Béland. 1976. Nonadrenergic inhibitory nervous system in human airways. J. Appl. Physiol. 41:764–771. Ellis, J. L., and B. J. Undem. 1992. Inhibition by L-N G-nitro-L-arginine of nonadrenergic–noncholinergic-mediated relaxations of human isolated central and peripheral airways. Am. Rev. Respir. Dis. 146:1543– 1547. Lundberg, J. M., C. R. Martling, and A. Saria. 1983. Substance P and capsaicin-induced contraction of human bronchi. Acta Physiol. Scand. 119:49–53. Luts, A., E. Widmark, R. Ekman, B. Waldeck, and F. Sundler. 1990. Neuropeptides in guinea pig trachea: distribution and evidence for the release of CGRP into tracheal lumen. Peptides 11:1211–1216. Bhogal, R., R. L. G. Sheldrick, R. A. Coleman, D. M. Smith, and S. R. Bloom. 1994. The effects of IAPP and CGRP on guinea pig tracheal smooth muscle in vitro. Peptides 15:1243–1247. Pinto, A., K. Sekizawa, M. Yamaya, T. Ohrui, Y. X. Jia, and H. Sasaki. 1996. Effects of adrenomedullin and calcitonin gene–related peptide on airway and pulmonary vascular smooth muscle in guinea pigs. Br. J. Pharmacol. 119:1477–1483. Hamel, R., and A. W. Ford-Hutchinson. 1988. Contractile activity of calcitonin gene–related peptide on pulmonary tissues. J. Pharm. Pharmacol. 40:210–211. Tschirhart, E., C. Bertrand, E. Theodorsson, and Y. Landry. 1990. Evidence for the involvement of calcitonin gene–related peptide in the eptihelium-dependent contraction of guinea pig trachea in response to capsaicin. Naunyn-Schmied. Arch. Pharmacol. 342:177–181. Gatto, C., R. C. Lussky, L. W. Ericksson, K. J. Berg, J. D. Wobken, and D. E. Johnson. 1989. Calcitonin and CGRP block bombesin- and substance P-induced increases in airway tone. J. Appl. Physiol. 66:573–577. Kanazawa, H., T. Kawaguchi, H. Kamoi, T. Fuji, S. Kudoh, K. Hirata, N. Kurihara, and J. Yoshikawa. 1996. Calcitonin gene–related peptide antagonizes the protective effect of adrenomedullin on histamineinduced bronchoconstriction. Clin. Exp. Pharmacol. Physiol. 23:472– 475. Nagase, T., E. Ohga, H. Katayama, E. Sudo, T. Aoki, T. Matsuse, Y. Ouchi, and Y. Fukuchi. 1996. Roles of calcitonin gene–related peptide (CGRP) in hyperpnea-induced constriction in guinea pigs. Am. J. Respir. Crit. Care Med. 154:1551–1556. Kroll, F., J. A. Karlsson, J. M. Lundberg, and C. G. A. Persson. 1990. Capsaicin-induced bronchoconstriction and neuropeptide release in guinea pig perfused lungs. J. Appl. Physiol. 68:1679–1687. Kanemura, T., J. Tamaoki, S. Horii, N. Sakai, K. Kobayashi, K. Isono, S. Takeuchi, and T. Takizawa. 1990. Calcitonin gene–related peptide augments parasympathetic contraction of rabbit tracheal smooth muscle in vitro. Agents Actions 31:219–224. Kannan, M. S., and D. E. Johnson. 1991. Functional innervation of pig tracheal smooth muscle: neural and non-neural mechanisms of relaxation. J. Pharmacol. Exp. Ther. 260:1180–1184. Martling, C. R., R. Matran, K. Alving, T. Hökfelt, and J. M. Lundberg. 1990. Innervation of lower airways and neuropeptide effects on bronchial and vascular tone in the pig. Cell. Tiss. Res. 260:223–233. Manzini, S. 1992. Bronchodilation by tachykinins and capsaicin in the mouse main bronchus. Br. J. Pharmacol. 105:968–972. Parsons, G. H., G. M. Nichol, P. J. Barnes, and K. F. Chung. 1992. Peptide mediator effects on bronchial blood velocity and lung resistance

Cadieux, Monast, Pomerleau, et al.: Bronchoprotector Properties of CGRP in conscious sheep. J. Appl. Physiol. 72:1118–1122. 31. Ninomiya, H., Y. Uchida, T. Endo, M. Ohtsuka, A. Nomura, M. Saotome, and S. Hasegawa. 1996. The effects of calcitonin gene–related peptide on tracheal smooth muscle of guinea pigs in vitro. Br. J. Pharmacol. 119:1341–1346. 32. Umeda, Y., and M. Arisawa. 1989. Characterization of calcitonin gene– related peptide (CGRP) receptors in guinea pig lung. Jap. J. Pharmacol. 51:377–384. 33. Mak, J. C. W., and P. J. Barnes. 1988. Autoradiographic localization of

243

calcitonin gene–related peptide (CGRP) binding sites in human and guinea pig lung. Peptides 9:957–963. 34. Cardell, L. O., R. Uddman, and L. Edvinsson. 1994. Low plasma concentrations of VIP and elevated levels of other neuropeptides during exacerbations of asthma. Eur. Respir. J. 7:2169–2173. 35. Solway, J., and J. J. Fredberg. 1997. Perhaps airway smooth muscle dysfunction contributes to asthmatic bronchial hyperresponsiveness after all. Am. J. Respir. Cell Mol. Biol. 17:144–146.