A comparison of sensory nerve function in human, guinea-pig, rabbit and marmoset airways

Life Sciences.Vol. 63, No. IX. pp. 1629-1642, I’+98 Copyright0 1998 Elsevier ScienceInc. Printed in the USA. All rights resewed 00243205/98 519.00 + 00 PI1 SOO24-3205(98)00432-9

ELSEVIER

A COMPARISON

OF SENSORY NERVE FUNCTION IN HUMAN, GUINEA-PIG, AND MARMOSET AIRWAYS

RABBIT

Domenico Spina, Gabriella M. Matera’, Maggi M. Riccio’ and Clive P. Page.

The Sackler Institute of Pulmonary Pharmacology, Department of Respiratory Medicine, King’s College School of Medicine and Dentistry, Bessemer Rd, London SE5 9PJ. ‘Institute of Pharmacology, II University of Naples, Via Constantinople 16, Naploli, Italy.

(Reccivcd

in final form August XI, 1998)

Summary We have investigated the role of sensory nerves in regulating airway smooth muscle function in the guinea-pig, marmoset, rabbit and man. Tissue levels of the sensory neuropeptides CGRP and substance P in the airways of the guinea-pig were significantly greater compared with the rabbit and marmoset. The relative order of tissue content was guinea-pig >>> rabbit = marmoset. Marmoset responded weakly to exogenously bronchial and tracheal preparations administered substance P and neurokinin A but contracted to methacholine and demonstrated atropine-sensitive cholinergic responses. In marmoset, rabbit and human airway preparations, capsaicin mediated weak contractile responses to exogenously administered capsaicin. However, high concentrations of capsaicin elicited a relaxation response that was epithelium-independent, cyclo-oxygenaseinsensitive, not involving nitric oxide and not dependent on the activation of capsaicin-sensitive afferents. These results suggest that rabbit and marmoset airways respond functionally in a similar way to human airway preparations and maybe more relevant than guinea-pig airways with regard to understanding the role of sensory neuropeptides in airways.

Key Words:

human, marmoset,

rabbit, guinea-pig,

trachea, bronchi, capsaicin, alferent

neurons, smooth muscle

In the guinea-pig, airway smooth muscle contraction can be elicited by electrical field stimulation that is atropine-resistant, (l-3), by capsaicin (1,4,5) and exogenous administration of the sensory neuropeptides neurokinin A and substance P (4-7). In contrast, activation of neurokinin-1 receptors on guinea-pig airway epithelium mediates a prostanoid sensitive relaxant effect (8,9). These results are consistent with studies demonstrating the release of sensory neuropeptides by D. Spina PhD. The Sackler Institute of Pulmonary Pharmacology, Department of Respiratory Medicine, King’s College School of Medicine and Dentistry Bessemer Rd, London SE5 9PJ, U.K., Phone +41-171-3463589, FAX +41-171-3463589 2Present address: Novartis Institute for Medical Sciences, 5 Gower Place London, WC 1E 6BN

1630

Sensory Nerve Function in Human Airways

Vol. 63, No. 18, 1998

capsaicin (7,10,1 I), vagal stimulation (7) and electrical field stimulation (12) of guinea-pig airways. This has led to guinea-pig airway preparations being extensively used as a model for investigating the role of sensory neuropeptides in human airway disease.

However, considerable species variation is evident, with regard to the effect of exogenously administered and endogenously released neuropeptides on airway smooth muscle function. The exogenous administration of substance P, neurokinin A and CGRP; as well as capsaicin- and electrical field stimulation-induced release of sensory neuropeptides, causes relaxation of tracheal preparations from Sprague-Dawley and Wistar rats (13-15). Interestingly, the relaxation induced by sensory neuropeptides is mediated by N&-receptors and is dependent on the presence of an intact epithelium and is inhibited by cycle-oxygenase inhibitors, implicating prostanoids in this response (13- 15). The ability of endogenously released sensory neuropeptides to mediate relaxation is consistent with the ability of various stimuli to mediate the release of substance P and CGRP from rat perfused tracheal preparations (16-18) and with the autoradiographic localisation of neuropeptide binding sites over airway epithelium and not smooth muscle in Sprague Dawley rats (19). Similarly, sensory neuropeptides mediate relaxation of mouse trachea that is also dependent on prostanoid formation within the epithelium (20). In contrast, tracheal preparations from Fischer 344 rats contract in response to substance P and neurokinin A (1521).

Substance P, and neurokinin A mediate contraction of rabbit isolated bronchus, while the excitatory response to capsaicin is modest and no evidence of an atropine-resistant response to electrical field stimulation was observed (22). The contractile response to substance P may be secondary to the stimulation of parasympathetic nerves by an NK,-receptor mechanism (23-27). Exogenous administration of substance P and neurokinin A contracts human bronchial preparations (4,6). In contrast, capsaicin elicits a modest contractile response that is at least 2-3 orders of magnitude less potent than in guinea-pig airways (4,28-3 1). Interestingly, capsaicin also appears to mediate an inhibitory response that is not dependent on the release of sensory neuropeptides (29,30). However, to date no study has convincingly demonstrated contractile responses secondary to the release of sensory neuropeptides from excitatory non-adrenergic noncholinergic (eNANC) nerves (4,32).

The aim of this study was to further investigate the role of sensory neuropeptides in the regulation of airway smooth muscle tone in a number of species including man to determine the most appropriate species for the study of neuropeptides in the airways when human tissue is not readily available. Materials and Methods

Tissue Preparation White NZ rabbits (2 - 3 kg) and marmosets (350 - 400 g) were killed with an intravenous overdose of pentobarbitone (60 mg/kg; the marmosets were being killed as part of another study and airway tissue was made available for this study); guinea-pigs (300 - 450 g) were killed by cervical dislocation and the lungs placed in ice cold (4°C) Krebs-Henseleit solution. Human lung was obtained from patients (60 + 3 y, n = 6). who were undergoing thoracotomy and human bronchi was obtained from areas of macroscopically normal tissue. Connective tissue, adherent blood vessels and parenchymal tissue were removed from the airways. Both distal tracheal and main bronchial rings (2 mm) were suspended between two L shaped steel

Vol. 63, No. 18, 1998

Sensory Nerve Function in Human Airways

1631

hooks connected to an FT03 isometric transducer under 500 mg, 1 g and 1.5 g tension for marmoset (trachea and main bronchus) and guinea-pig (trachea), rabbit (main bronchus) and human (bronchial), respectively. The tissues were placed in Krebs-Henseleit solution maintained at 37°C and aerated with 95% 0, and 5% CO,. Tissues were allowed to equilibrate for 40 min and changes with fresh Krebs-Henseleit solution made every 10 min. In some preparations, the epithelium was removed using a wooden probe as described previously (22). In some experiments, tissues were maintained in Krebs-Henseleit solution containing the cycle-oxygenase inhibitor, indomethacin (5 uM) to abolish endogenous production of prostanoids. Following a 40 min equilibration period, marmoset, guinea-pig, rabbit and human airway preparations were contracted with consecutive concentrations of methacholine (0.3 & 100 PM). Tissues were washed 5 times over a 15 min period and allowed to equilibrate for a further 30 min.

Contractile Studies Tissues from marmoset airway were exposed to a range of ‘neurogenic’ agonists since the response to these agents has not been previously characterised. Concentration response curves to increasing concentrations (0.5 or 1 log increments) of methacholine (10 nM - 100 p,M), substance P (0.1 nM - 1 PM) and neurokinin A (0.1 nM - 1 PM) were performed in marmoset bronchial and tracheal preparations. In some experiments, the neutral endopeptidase inhibitor, thiorphan (10 PM) was added to the organ bath, 30 min prior to the addition of substance P and neurokinin A.

In other experiments, tissues were placed between 2 platinum electrodes and electrically stimulated (0.1 - 30 Hz, 10 set, 0.5 ms, max voltage) in the absence or presence of the non-selective muscarinic antagonist, atropine (0.1 PM). Some tissues were electrically stimulated following 30 min incubation with thiorphan (10 PM). In additional experiments, cumulative concentrations of capsaicin (1 - 100 pM) were added to the organ bath in the absence or following 30 min incubation with thiorphan (10 PM).

Rabbit bronchial, tracheal and guinea-pig bronchial preparations were contracted with increasing concentrations of capsaicin (1 nM - 100 PM). In other experiments rabbit airway preparations were then washed 5 times over a 15 min period and allowed to equilibrate for a further 30 min prior to a consecutive administration of capsaicin (10 & 100 PM). This was followed by a further wash period and 30 min equilibration period and in some tissues, tone was raised with methacholine (1 PM) and the effect of capsaicin (10 & 100 pM) on this response was determined.

The effect of capsaicin (1 - 100 FM) on spontaneously contracted tissue was determined in the absence or presence of the cycle-oxygenase inhibitor, indomethacin (5 PM). In control tissues, capsaicin responses were elicited a second time to determine whether the response was subject to desensitisation.

Relaxation Studies In some studies, marmoset, rabbit and human airway preparations were contracted with methacholine. Once the contractile response had reached a plateau, capsaicin (1 - 100 pM) or vehicle (ethanol; 0.1 - 1 %) was added to the bath. This was also repeated in some bronchial preparations in the absence or presence of the nitric oxide synthase inhibitor, L-NGnitroarginine methyl ester (L-NAME, 100 pM, 30 min incubation period) or ruthenium red (10 PM, 20 min incubation period).

1632

Sensory Nerve Function in Human Airways

Vol. 63, No. 18, 1998

Measurement of neuropeptide levels in airway preparations Airway preparations (trachea and main bronchi) were dissected clear of connective tissue and visible blood vessels cut into small pieces and placed into 2 ml of 2N acetic acid in a boiling water bath (95°C) for 30 min. Samples were homogenised and centrifuged (2000 g) at 4°C for 30 min. The supernatant was decanted, freeze dried and stored at -18°C for subsequent determination of calcitonin gene-related peptide(CGRP) and substance P-like immunoreactivity.

Freeze dried .samples reconstituted in phosphate buffer (100 pl) and known concentrations of human cl-CGRP-II (40 - 5000 pg ml-‘) was incubated with rabbit anti-human CGRP-II antiserum (100 ~1) for 48 h at 4°C. [‘251]-iodohistidyl human CGRP (100 pl) was added to each sample and standards and incubated for a further 48 h at 4°C. Separation of bound and free antigen was performed by incubating samples and standards with 1 ml of phosphate buffer containing 7.5 % polyethylene glycol, l/200 goat anti-rabbit antiserum and l/2000 normal rabbit serum for 2 h, then centrifuged at 2000 g for 30 min at 4°C. The lower detection limit for this assay was 2 fmol ml-‘. The antiserum cross reacted 100 % with rat P-CGRP and human CGRP-I and CGRP-II and less than 0.1 % with salmon and human calcitonin (33,34).

In other experiments, freeze dried samples reconstituted in phosphate buffer (100 pl) and known concentrations of substance P (5 - 1250 pg ml-‘) was incubated with anti-substance P antiserum (100 ~1) for 24 h at 4°C. [‘*‘I]-Bolton Hunter substance P (100 ~1) was added to each sample and standards and incubated for 24 h at 4°C. Separation of bound and free antigen was performed by incubating samples and standards with 1 ml of phosphate buffer containing 7.5 % polyethylene glycol, l/200 goat anti-rabbit antiserum and 112000 normal rabbit serum for 2 h, then centrifuged at 2000 g for 30 min at 4°C. The lower detection limit for this assay was 10 fmol ml-‘. The antiserum cross reacted 1 % with neurokinin A, 0.5 % neurokinin B and less than 0.1 % physalaemin and eledoisin (33).

Drugs methylester Composition

Au-opine, capsaicin, indomethacin, methacholine, substance P, L-NC-nitroarginine (L-NAME), ruthenium red, neurokinin A, thiorphan, theophylline (Sigma). of Krebs-Henseleit solution (mM): NaCl 117.6, KC1 5.4, MgS0,.7H20 0.57, KH,.PO,

1.03, NaHCO, 25.0, Glucose 11 .l and CaC1,.2H,O 2.5. All drugs were prepared in Krebs-Henseleit solution. The stock concentration (0.01 M) of indomethacin was prepared in 0.5 % NqCO,. The stock concentration of capsaicin (0.01 M) was prepared in 100 % ethanol. Stock concentrations of substance P and neurokinin A were prepared in 10 % acetic acid and stored at -20°C. The appropriate dilution was then made in Krebs-Henseleit solution.

Contractile responses are calculated as a % of the maximum Statistical Analysis response (Emax) to methacholine. Relaxant responses are calculated as % reversal of methacholine-induced tone. In spontaneously-contracted preparations, relaxant responses are expressed as a % of the maximum relaxation to theophylline (3 mM). Both contractile and relaxant responses were expressed as the arithmetic mean + s.e.mean. Contractile potency (EC,,) represents the concentration of agonist that is required to contract airway preparations to 50% of the Emax to methacholine and is expressed as the geometric mean together with the 95% confidence limits. Analysis of covariance was used to determine the effect of treatment on frequency response curves (Minitab, release 10). Differences between means were analysed using Student’s paired and non-paired t-test and considered statistically significant if P < 0.05.

Sensory

Vol. 63, No. 18, 1998

Nerve Function

in Human

1633

Airways

Results

Marmoset: Cholinergic responses Methacholine produced a concentration dependent contraction of marmoset isolated trachea and bronchus. Epithelium removal failed to alter the contractile potency (EC,,; 95% Confidence limits) to methacholine in marmoset isolated trachea (epithelium-intact; 0.36 l.tM; 0.18 - 0.71 vs epithelium-denuded, 0.50 PM; 0.18 - 1.35; P > 0.0s) and isolated bronchus (epithelium-intact; 0.83 PM; 0.20 - 2.3 vs epithelium-denuded; 0.68 PM; 0.49 - 1.4; P > 0.05). In the presence of the cycle-oxygenase inhibitor, indomethacin (5 PM), the contractile potency to methacholine in epithelium-intact trachea (0.22 uM; 0.06 - 0.76) and bronchus (0.63 PM; 0.13 - 3.0) was not significantly different to the contractile potency observed in epithelium-intact trachea and bronchus, in the absence of indomethacin (P > 0.05).

Electrical field stimulation of marmoset isolated trachea and bronchus produced a frequency dependent contractile response (Table I) that was atropine sensitive. Epithelium removal failed to significantly alter the contractile response to electrical field stimulation (Table I; P > 0.05). In contrast, treatment of epithelium-intact bronchus with indomethacin significantly increased the contractile response to electrical field stimulation (P < 0.05).

Marmoset: Neuropeptides Substance P elicited weak and variable contractile responses (6 out of 16 preparations from 9 animals) in marmoset isolated trachea that did not appear to be dose dependent (Table II). No significant increase in the contractile response was observed following epithelium removal, although a trend was observed (P > 0.05, Table II). Treatment with thiorphan failed to increase the contractile response to substance P. Indeed, 4 out of 4 preparations tested failed to elicit a response to substance P in the presence of thiorphan. Similarly, substance P elicited weak and variable contractile responses in marmoset isolated bronchus (4 out of 8 preparations from 5 animals). No increase in the contractile response to substance P was observed TABLE I Contractile responses (% methacholine Emax) to electrical marmoset airways in the absence or presence of the epithelium. Frequency (Hz) Trachea 0.1 1 10

Bronchus 0.1 1 IO

Epithelium-intact

field

stimulation

Epithelium-denuded

0.8 + 0.8 4.6 t 2.9 25.4 + 6.1

6.9 rfr3.2 16.4 + 4.7 37.5 + 5.3

(4)

(5)

10.6 + 2.6 16.3 + 3.7 29.3 + 4.7

6.5 + 3.0 16.4 i 5.1 21.4 rf:2.8

(6)

(5)

Results expressed as mean + s.e.mean and values in parentheses of animals.

indicate the number

in

Sensory Nerve Function

1634

in Human Airways

Vol. 63, No. 18, 1998

in epithelium-denuded tissue (P > 0.05, Table II). Thiorphan did not increase the contractile response to substance P as only 1 out of 5 preparations contracted to substance P.

Neurokinin A contracted marmoset trachea albeit in a weak and variable fashion (Table II; epithelium-intact, 7 out of 10 preparations from 7 animals). Hence, no significant increase in the contractile response to neurokinin A was observed in epithelium-denuded preparations (P > 0.05) or following incubation of epithelium-intact tissues with thiorphan (P > 0.05). Similarly, neurokinin A contracted marmoset epithelium-denuded bronchus in a weak fashion (3 out of 8 preparations from 5 animals). Epithelium removal failed to augment the contractile response to neurokinin A (P > 0.05) and the response was not augmented by thiorphan (P > 0.05).

Marmoset: Capsaicin Capsaicin (100 PM) elicited weak contractile responses in all tracheal (2.3 + 0.8 % methacholine Emax, n = 12) and bronchial (8.0 + 2.2 % methacholine Emax, n = 9 preparations) preparations from 7 animals. The absence of an epithelium or the presence of thiorphan did not augment the contractile response to capsaicin. In contrast, capsaicin (100 PM) elicited relaxation (% reversal of methacholine EC&-induced tone) in epithelium-intact (51.7 2 17.0 %, n = 3), epithelium-denuded (66.8 + 14.5 %, n = 4) and in epithelium-intact marmoset trachea in the presence of indomethacin and thiorphan (56.5 + 14.4 %, n = 3). Similarly, capsaicin (100 PM) reversed methacholine contracted marmoset epithelium-intact bronchus in the absence (38.1 f. 19.5 %, n = 3) or presence (64.6 & 6.3%, n = 7) of indomethacin. Lower concentrations of capsaicin failed to elicit a significant response in trachea and bronchus. Ethanol (0.1 & 1%) reversed methacholine contracted marmoset airway preparations (5.3 + 3.5 and 32 + 12 %, respectively; n = 7). TABLE II Substance P and Neurokinin A-induced contraction marmoset bronchial and tracheal preparations

SP -10 -9 -8 -7 -6

(% methacholine

Emax) in

Trachea +EPI 2.5 2 2.5 28.1 + 11 28.8 + 11 28.8 & 11 28.8 + 11

-EPI 0 1128 16 + 12 17 + 12 17 + 12

Bronchus + EPI 7.7 + 5.6 16.2 212.4 18.3 212 26.4 212.6 28.6 k13.1

-EPI s+s 2oi20 21221 21221 21221

(5)

(7)

(6)

(3)

5.7 + 2.6 16.8 + 8.3 25.7 + 9.3 36.9 + 13 38.5 + 13.1

6.2 + 4.5 9.2 + 6.2 20.8 + 13.4 20.8 2 13.4

0 3.2 +_3.2 5.3 + 4.1 8.4 + 5.2 13.7 + 8.7

0 0 2.9 + 2.9 6.9 + 6.9 9.8 f. 9.8

(5)

(5)

(5)

(3)

NIL4 -10 -9 -8 -7 -6

Results expressed as mean & s.e.mean. animals used. +EPI, epithelium-intact; NRA, neurokinin A.

Values in parentheses indicate the number of -EPI, epithelium-denuded; SP, substance P;

Vol. 63, No. 18, 1998

Sensory Nerve Function in Human Airways

1635

The capsaicin-induced relaxation of methacholine contracted marmoset bronchus was not subject to desensitisation (Fig. la, P > 0.05 compared with the first concentration). Similarly, the capsaicin-induced relaxation was not inhibited by the nitric oxide synthase inhibitor L-NAME (P > 0.05, Fig. 2a).

Rabbit High concentrations of capsaicin (10 & 100 pM) induced modest contractile responses (Oh methacholine Emax) in rabbit epithelium-intact bronchus that was subject to desensitisation (capsaicin 100 pM; control 14.4 + 2.5 % vs consecutive dose, 1.5 + 0.6 %: n = 12 bronchial preparations from 6 animals, P < 0.05).

In contrast, capsaicin induced a relaxation response in rabbit epithelium-intact trachea contracted with methacholine, that was not subject to desensitisation (capsaicin 100 PM; control 43.2 + 10.3 % vs consecutive dose, 53.8 t 13.0 %, n = 5 preparations from 5 animals, P > 0.05). Similarly, capsaicin mediated a concentration-dependent relaxation (“/ reversal of 1 pM methacholineinduced tone) of rabbit epithelium-intact bronchus that was not subject to desensitisation (capsaicin 100 pM; control 54.9 + 6.2 % vs consecutive dose, 60.6 2 6.0 %, n = 12 preparations from 6 animals, P > 0.05, Fig. lb). Vehicle induced a small degree of relaxation in methacholinecontracted bronchus (ethanol 0.1%; -0.6 + 1.1 %; ethanol 1% 9.4 + 2.4 %, n = 10 preparations from 5 animals) and trachea preparations from 5 animals).

(ethanol

0.1%;

1.6 f. 1.O %; ethanol

1% 2.4 + 1.6 %, n = 5

The relaxation response to capsaicin was also unaltered in tissues previously desensitised to the capsaicin-induced contractile response (capsaicin 100 PM; control; 60.6 + 6.0 %, n=12 preparations from 6 animals vs desensitised bronchi, 66.7 + 5.8 %, n = 10 preparations from 5 animals, P > 0.05). Similarly, L-NAME (100 pM; Fig. 2b) and ruthenium red (10 nM; Fig. 3b) failed to significantly inhibit the capsaicin-induced inhibitory response in rabbit bronchial preparations (P > 0.05).

Human In the presence of thiorphan, capsaicin elicited weak contractile responses (% methacholine Emax) in human bronchial preparations (1 pM, 2 + 1; 10 pM, 14 + 4; 100 pM, 44 + 12; n = 11 bronchial preparations from 4 lungs). In human bronchial preparations contracted with methacholine (100 PM), capsaicin elicited a concentration-dependent relaxation that was not subject to desensitisation (Fig. lc), not inhibited by L-NAME (Fig. 2c) or ruthenium red (Fig. 3~).

Neuropeptide content We also measured tissue neuropeptide content in rabbit, marmoset and guinea-pig airways. Both CGRP and substance P were detected in the airways of all species studied but marked species differences was observed. Guinea-pig airways contained significantly greater amounts of substance P and CGRP than rabbit or marmoset airways (P < 0.05, Table III).

Discussion We have demonstrated marked species differences with regard to airway tissue content of sensory neuropeptides and excitatory responses of airway tissue to capsaicin. The neuropeptide content of guinea-pig airways is at least 10 times greater than that observed in marmoset and rabbit airways

1636

Sensory Nerve Function in Human Airways

Vol. 63, No. 18, 1998

a 3 d

100

1

-5

-4

Log, OCapsaicin (M) b

-5

-4 Loglo Capsaicin (M)

Log,* Capsaicin (M)

FIG 1 Bar graph represent capsaicin-induced relaxation in (a) marmoset (b) rabbit and (c) human bronchial preparations. First (stippled columns) and a consecutive (shaded columns) responses shown. Vertical lines represent s.e.mean.

Sensory Nerve Function in Human Airways

Val. 63, No. 18,199s

a 100 -

75 -

50 -

25 -

-6 Log 1. Capsaicin (M) b 100 7

-6

-5 LogioCapsaicin

-4 (M)

50 -

25 -

O--6

-5 Log, &apsaicin

-4 (M)

FIG 2 Bar graph represent caps&in-induced relaxation in (a) marmoset (b) rabbit and (c) human bronchial preparations in the absence (stippled columns) and presence (shaded columns) of L-NAME. Vertical lines represent s.e.mean

1637

1638

Sensory Nerve Function in Human Airways

E = e?

Vol. 63, No. 18, 1998

70 -

GS s .s r”E

45 -

88 E .iii

20 -

4 I?

-5-

-5

-4

LogioCapsaicin

(M)

b B 5

100

1

-6

-5 Log,,,Capsaicin

-4 (M)

FIG 3 Bar graph represent capsaicin-induced relaxation in (a) rabbit and (b) human bronchial preparations in the absence (stippled columns) and presence (shaded columns) of ruthenium red. Vertical lines represent s.e.mean

TABLE III Airway content (fmol mg protein-‘) of CGRP and substance P in a number of animal species.

guinea-pig rabbit marmoset human

CGRP

N

Substance P

N

13 220 t 3250* 0.38 pmol g tissue-’ (10) 4400 + 640 3220 + 1170 0.56 pmol g tissue-’ (10)

6

4410 + 380* 12 pm01 g tissue-’ (35) 127230 122240 13 pmol g tissue-’ (36)

10

15 17

Results expressed as mean + s.e.mean of N observations indicated parenthesis. * Significantly greater cf rabbit and marmoset ( P < 0.05).

15 17

in

Vol. 63, No. 18, 1998

Sensory Nerve Function in Human Airways

1639

presented in this current study. In contrast, the levels of CGRP and substance P in guinea-pig are similar to that in human airways (10,35,36). Despite this finding, contractile responses to capsaicin in human are modest and similar to that observed in rabbit and marmoset airway. Furthermore, capsaicin appeared to mediate relaxation that was epithelium-independent, cyclooxygenase independent, was not subject to desensitisation or blocked by ruthenium red and L,NAME in marmoset, rabbit and human bronchial preparations. Thus, capsaicin has been shown to exert airway smooth muscle relaxation unrelated to the activation of airway sensory nerves.

It is now well established that guinea-pig airways contain a dense plexus of neuropeptide containing nerves (2,4,5) including neurokinin A, substance P and CGRP. Bronchomotor responses to endogenously released neuropeptides increase from proximal to more distal regions of the airways (35) and correlate with the greater neuropeptide content found in the lower airways of the guinea-pig (35). We have previously documented that capsaicin is a potent spasmogen of guinea-pig bronchial preparations (37), being some 2 - 4 orders of magnitude lower than that observed in rabbit, marmoset and human bronchial preparations. Nonetheless, while the excitatory response to capsaicin in these species is modest, it is subject to desensitisation similar to that reported in guinea-pig airways. Immunohistochemical studies have shown that rabbit airways are innervated sparsely with neuropeptide containing nerves (38) and this is confirmed by the relatively low tissue levels of the neuropeptides CGRP and substance P. Autoradiographic studies have revealed the presence of neuropeptide receptors on airway smooth muscle (39) and exogenously administered substance P and neurokinin A contracts rabbit airway smooth muscle directly (22,39) and via the release of acetylcholine from parasympathetic nerves (24). The endogenous release of neuropeptides by electrical field stimulation in the rabbit has not been demonstrated (22). However, in contrast to the modest contractile response to capsaicin in rabbit, human and marmoset airway preparations, we have demonstrated that capsaicin mediates relaxation of methacholine contracted tissue that is not subject to desensitisation. This suggests that the relaxant response to capsaicin is not mediated by activation of sensory nerves. Furthermore, the relaxant response to capsaicin in rabbit and marmoset airways was not inhibited by ruthenium red, ruling out a role of the capsaicin-sensitive non selective ion channel present on a population of afferent nerves (40). Furthermore, the nitric oxide synthase inhibitor, L-NAME was without effect on the capsaicin-induced relaxant response, thus ruling out the possible involvement of nitric oxide.

There appeared to be a correlation between neuropeptide levels and the functional response to capsaicin in marmost, rabbit and guinea-pig airways. However, this correlation may be more complex when considering human airways. Immunohistochemical studies has shown that sensory neuropeptide containing nerves sparsely populate the lung in non-asthmatics (28,38) and may explain why human airway tissue is at least 2 - 4 orders of magnitude less sensitive to capsaicin, even in the presence of neutral endopeptidase inhibitors compared with guinea-pig airways (4,28,3 1) and in the present study. Interestingly, the levels of substance P-like immunoreactivity in guinea-pig (12 pmol g-’ tissue; (35)) and human airways (13 pmol g-’ tissue; (36)) are similar. Furthermore, the density of vanilloid receptors was similar in homogenates obtained from guineapig airways and from a human lung obtained from a patient with bronchial carcinoma (41). It is clear therefore, that the insensitivity of human airways to capsaicin compared with guinea-pigs is not simply a reflection of the total neuropeptide content of the airways, but possibly to the pool of neuropeptide that is available for release (42).

1640

Sensory Nerve Functionin Human Airways

Vol. 63, No. 18, 1998

We have shown that capsaicin induces an inhibitory effect in rabbit, marmoset and human bronchial preparations similar to that observed previously in man (29,30). The excitatory response in human airways are relatively weak compared to that observed in the guinea-pig but similar to that observed in rabbit and marmoset airways. The inhibitory response to capsaicin was achieved at high concentrations, was not dependent on the presence of the epithelium, products of the cyclooxygenase pathway and is insensitive to ruthenium red and L-NAME, thereby ruling out a role for sensory neuropeptides, prostaglandins and nitric oxide in this response. Capsaicin has previously been reported to have effects additional to activation of sensory nerves including stimulating prostanoid synthesis (43,44) and behaving as a non-competitive nicotinic antagonist (45). Thus, the relaxation induced by capsaicin in human airway may be secondary to the release of mediators from sensory nerves that is resistant to desensitisation (29) or alternatively, via the opening of calcium activated potassium channels (30). Recent in vivo data has also demonstrated that capsaicin can mediate bronchodilation in healthy subjects or mild asthmatics following adrenergic and/or cholinergic blockade and in whom baseline airway resistance was increased with inhaled spasmogen (46,47). Capsaicin-induced bronchoconstriction, however, was only observed in 7 of 17 asthmatics in the absence of adrenergic and/or cholinergic blockade or raised baseline tone (48), while bronchodilation was observed in 8 of 15 patients who received heart lung transplants (48). The capsaicin induced bronchodilator response might be due to the release of inhibitory neurotmnsmitters from efferent nerves and/or by a direct relaxant effect on bronchial smooth muscle (48).

Our results demonstrate that human airways, like marmoset and rabbit are considerably less responsive to capsaicin than guinea-pig and suggest that species other than the guinea-pig might be considered when investigating the role of endogenously released neuropeptides in modulating airway smooth muscle function, Furthermore, the inhibitory action of capsaicin on human airway smooth muscle should be considered when studying the effect of inhaled capsaicin in man.

Acknowledgements The research was supported by grants from the Joint Research Committee of King’s College School of Medicine & Dentistry, Purdue Frederick Inc., Norwalk, CT, USA and by the Central Research Fund, University of London.

Reference 1. N. GRUNDSTROM, R.G.G. ANDERSSON Toxicol. 49 150-157 (1981). 2. J.M. LUNDBERG

and A. SARIA,

and J.E.S. WIKBERG,

Acta Physiol.Scand.

3. B.J. UNDEM, A.C. MYERS, H. BARTHLOW 1336-1346 (1990). 4. J.M. LUNDBERG, (1983).

C.-R. MARTLING

116 473-476 (1982).

and D. WEINREICH,

and A. SARIA,

Acta Pharmacol.et

J.Appl.Physiol.

Acta Physiol.Scand.

5. J.M. LUNDBERG, A. SARIA, E. BRODIN, S. ROSELL and K. FOLKERS, Proc.Natl.Acad.Sci.USA 80 1120-l 124 (1983).

119 49-53

69

1641

Sensory Nerve Function in Human Airways

Vol. 63, No. 18, 1998

6. C. ADVENIER, 137 (1987).

E. NALINE, G. DRAPEAU

and D. REGOLI,

Eur.J.Pharmacol.

7. A. SARIA, C.-R. MARTLING, Z. YAN, E. THEODORSSON-NORHEIM, J.M. LUNDBERG, Am.Rev.Respir.Dis. 137 1330-1335 (1988). 8. E. TSCHIRHART 9. N. FROSSARD, (1989).

and Y. LANDRY,

Eur.J.Pharmacol.

K.J. RHODEN and P.J. BARNES,

139 133-

R. GAMSE and

132 103-104 (1986).

J.Pharmacol.Exp.Ther.

248 292-298

10. C.R. MARTLING, A. SARIA, J.A. FISCHER, T. HOKFELT and J.M. LUNDBERG, Regul.Pept. 20 125-139 (1988). 11. Y.P. LOU, A. FRANC0 119-127 (1992).

CERECEDA

12. S. SHAH, D. SPINA and C.P. PAGE, 13. N. FROSSARD

and F. MULLER,

and J.M. LUNDBERG, Am.J.Respir.Crit.Care

J.Appl.Physiol.

Acta Physiol.Scand.

146

Med. 153 A848.(1996).

61 1449-1456 (1986).

14. P. DEVILLIER, G.M. ACKER, C. ADVENIER, J. MARSAC, D. REGOLI and N. FROSSARD, J.Pharmacol.Exp.Ther. 263 767-772 (1992). 15. J.L. SZAREK, N.L. STEWART, 1679-1687 (1995).

B. SPURLOCK

and C. SCHNEIDER,

16. N.J. RAY, A.J. JONES and P. KEEN, Br.J.Pharmacol. 17. X.Y. HUA and T.L. YAKSH,

J.Appl.Physiol.

78

102 797-800 (1991).

J.Neurosci. 13 1947-1953 (1993).

18. X.Y. HUA, S. JINNO, S.M. BACK, E.K. TAM and T.L. YAKSH, 1147-l 154 (1994).

Neuropharmacology

33

19. K. SERTL, M.L. KOWALSKI, 1295-1299 (1988).

J. SLATER and M.A. KALINER,

Am.Rev.Respir.Dis.

138

20.

S. MANZINI,

105 968-972 (1992).

21.

G.F. JOOS, R.A. LEFEBVRE, (1994).

22.

D. SPINA, M.G. MCKENNIFF, A.J. COYLE, E.A. SEEDS, M. TRAMONTANA, F. PERRETTI, S. MANZINI and C.P. PAGE, Br.J.Pharmacol. 103 1268-1274 (1991).

23.

D.T. TANAKA and M.M. GRUNSTEIN,

J.Appl.Physiol.

60 458-463 (1986).

24.

D.T. TANAKA and M.M. GRUNSTEIN,

J.Appl.Physiol.

57 155 l-l 557 (1984).

25.

C. JOHN, S. BRUNNER and D.T. TANAKA,

26.

K. INOUE, Y. SAKAI and I. HOMMA,

27.

M.G. BELVISI, R. PATACCHINI, 103-l 10 (1994).

28.

A.A. HISLOP, J. WHARTON, K.M. ALLEN, J.M. POLAK and S.G. HAWORTH, Am.J.Respir.Cell Mol.Biol. 3 191-198 (1990).

29.

P. CHITANO, P. DI BLASI, R.E. LUCCHINI, F. CALABRO, M. SAETTA, P. MAESTRELLI, L.M. FABBRI and C.E. MAPP, Eur.J.Pharmacol. 270 167-173 (1994).

30.

J.L. ELLIS, J.S. SHAM and B.J. UNDEM, (1997).

Br.J.Pharmacol.

J.C. KIPS and R.A. PAUWELS,

Am.J.Physiol.

Br.J.Pharmacol.

Eur.J.Pharmacol.

27147-54

265 L228-L233(1993).

105 393-399 (1992).

P.J. BARNES and C.A. MAGGI,

Am.J.Respir.Crit.Care

Br.J.Pharmacol.

Med. 155 751-755

111

1642

31.

Sensory

Nerve Function

in Human Airways

Vol. 63, No. 18, 1998

M. MOLIMARD, C.A. MARTIN, E. NALINE, A. HIRSCH and C. ADVENIER, Am.J.Respir.Crit.Care Med. 149 123-127 (1994).

32. J.C. DE JONGSTE, H. MONS, I.L. BONTA and K.F. KERREBIJN, 1558-1566 (1987).

J.Appl.Physiol.

63

33. P. GEPPETTI, S. FRILLI, D. RENZI, P. SANTICIOLI, C.A. MAGGI, E. THEODORSSON and M. FANCIULLACCI, Regul.Pept. 23 289-298 (1988). 34.

S. MEINI, S. EVANGELISTA, P. GEPPETTI, A. SZALLASI, MANZINI, Am.Rev.Respir.Dis. 146 930-934 (1992).

35.

S. MANZINI, S. CONTI, C.A. MAGGI, L. ABELLI, V. SOMMA, E. DEL BIANCO and P. GEPPETTI, Am.Rev.Respir.Dis. 140 936-941 (1989).

36.

C.M. LILLY, T.R. BAI, S.A. SHORE, A.E. HALL and J.M. DRAZEN, Am.J.Respir.Crit.Care Med. 151 548-553 (1995).

37.

D. SPINA, S. HARRISON

38.

L.A. LAITINEN, A. LAITINEN, P.A. PANULA, M. PARTANEN, TERVO, Thorax 38 53 l-536 (1983).

and C.P. PAGE,

Br.J.Pharmacol.

P.M. BLUMBERG

and S.

116 2334-2340 (1995). K. TERVO and T.

39. J. BLACK, L. DIMENT, C. ARMOUR, L. ALOUAN and P. JOHNSON, J.Pharmacol.Exp.Ther. 253 38 l-386 (1990). 40.

A. DRAY, C.A. FORBES and G.M. BURGESS,

41.

A. SZALLASI,

42.

R. HAKANSON, B. BEDING, R. EKMAN, M. HEILIG, C. WAHLESTEDT SUNDLER, Neurosci. 4 943-950 (1987).

43.

H. JUAN, F. LEMBECK, S. SEEWANN and U. HACK, Naunyn Schmiedebergs Arch.Pharmacol. 312 139-143 (1980).

44.

D.L. FLYNN, M.F. RAFFERTY 198 (1986).

45.

K. NAKAZAWA, (1994).

46.

J.W. LAMMERS, P. MINETTE, M.T. MCCUSKER, J.Appl.Physiol. 64 1817-1822 (1988).

K.F. CHUNG and P.J. BARNES,

47.

J.W. LAMMERS, P. MINETTE, M.T. MCCUSKER, J.Appl.Physiol. 67 856-861 (1989).

K.F. CHUNG and P.J. BARNES,

48.

T.J. HATHAWAY, T.W. HIGENBOTTAM, J.F. MORRISON, WALLWORK, Am.Rev.Respir.Dis. 148 1233-1237 (1993).

C. GOSO and S. MANZINI,

Neurosci.Lett.

110 52-59 (1990).

Am.J.Respir.Crit.Care

and A.M. BOCTOR,

Prostaglandins

K. INOUE, S. KOIZUMI and M. IKEDA,

Med. 152 59-63 (1995). and F.

Leukot.Med. 24 195-

Br.J.Pharmacol.

113 296-302

C.A. CLELLAND

and J.