OPIOID INHIBITION OF NEURALLY MEDIATED MUCUS SECRETION IN HUMAN BRONCHI

930

for the inability to show NANC secretion of mucus in human bronchi might be the inhibitory effect of opioids. We have tested this hypothesis in human bronchi in vitro; we used capsaicin, the pungent extract of red peppers,l’ to release neuropeptides selectively from sensory nerves and the opioid receptor antagonist naloxone to reverse the effect of residual opioids.

possible explanation

Preliminary Communication OPIOID INHIBITION OF NEURALLY MEDIATED MUCUS SECRETION IN HUMAN BRONCHI DUNCAN F. ROGERS

PETER J. BARNES

METHODS

Department of Thoracic Medicine, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY induces release of substance P from stimulated mucus secretion in surgically nerves, sensory resected human bronchi in vitro. Pretreatment of the tissue with the opioid antagonist naloxone significantly enhanced secretion, possibly by blocking the inhibitory effect of opiate premedication before surgery. Capsaicin-induced mucus secretion was completely blocked by morphine, and this effect was reversed by naloxone. Thus, sensory nerve stimulation increases mucus secretion in human airways, which might contribute to the mucus hypersecretion seen after inhalation of irritants such as cigarette smoke. Secretion can be completely inhibited by opioid drugs, so

Summary

Capsaicin, which neuropeptides such

as

they may represent a new therapeutic approach to airway hypersecretion in chronic bronchitis and asthma, in which axon reflex mechanisms have been implicated. INTRODUCTION

MUCUS

hypersecretion

is

a

feature of several bronchial

diseases, including chronic bronchitis, cystic fibrosis, and some types of asthma, and contributes substantially to

Main bronchi, macroscopically normal, were obtained at lung resection for bronchial carcinoma from nine male and five female patients (mean age 63 years, range 55-70). All patients had received anticholinergic drugs and the opiate papaveretum (’Omnopon’). Bronchi were kept overnight in Krebs-Henseleit physiological saline (4°C, oxygenated with 5% carbon dioxide in oxygen) to increase viability.’’’ They were opened longitudinally and cut into sections (approximately 1-5 x2’0 cm) which were mounted flat between the two halves of plastic Ussing chambers so that the tissue separated the chambers into "luminal" (ie, mucus-producing) and "submucosal" sides.19 There was enough tissue from each patient for 2-4 chambers. Each half of the tissue was bathed in oxygenated Krebs-Henseleit solution at 37°C, containing the alphaadrenoceptor antagonist phentolamine, the beta-adrenenoceptor antagonist propranolol, and the cholinergic antagonist atropine (all at 10 umol/1) to eliminate adrenergic and cholinergic effects. Fluid from both sides of the chambers was drained at the end of each incubation period and replaced by fresh Krebs-Henseleit solution. Fluid from the luminal side was stored at - 70°C for later analysis. Basal rates of secretion were reached after 1 ’5 h incubation (two 15 min collections followed by two 30 min collections); ten further 30 min collections were made. Drugs were added in randomised order by replacing solution alone with Krebs-Henseleit containing morphine sulphate (1 µmol/l); Evans Medical Ltd, Horsham), naloxone hydrochloride (10 umol/1, DuPont Pharmaceuticals,

morbidity. The mechanisms underlying hypersecretion in disease, and even those controlling mucus secretion in health, are poorly understood, although both humoral and neural mechanisms probably play a part.1 This uncertainty may explain why there is no satisfactory treatment for mucus hypersecretion. In animals, there is evidence for neural control of airway mucus secretion by adrenergic and cholinergic mechanisms and also by non-adrenergic, noncholinergic (NANC) nerves, whose transmitters may be neuropeptides.2,3 Several neuropeptides, including vasoactive intestinal peptide and substance P, 4,5 as well as stimulation of NANC nerves, induce airway mucus secretion in various animal species.6,7Electrical stimulation of the vagus nerve causes release of neuropeptides, including substance P, from sensory nerve endings.2 Thus, sensory neuropeptides released by way of an axon reflex could be involved in airway disease.8 However, there is as yet no evidence for NANC neuronal control of mucus secretion in human airways in vitro, despite a clear effect of cholinergic nerve stimulation.9 SubstanceP-immunoreactive nerves lie close to,10 and substance P receptors have been localised on,11.12 submucosal glands in human airways. Furthermore, substance P is a potent stimulus to mucus secretion in human bronchi in vitro, 13 which raises the possibility that an axon reflex mechanism might lead to increased mucus secretion. We have shown previously that several airway responses, including bronchoconstriction and microvascular leakage, mediated by way of NANC neuronal mechanisms and the release of neuropeptides, are effectively inhibited by opioid drugs. 14-16 Since bronchi obtained after surgical resection have been

exposed

to

opioids

in the

premedication,

a

Effect of naloxone and morphine on capsaicin-induced secretion of fucose in human bronchi in vitro.

Values

=

% change in

pmol/1). Horizontal bar

rate of fucose secretion in response to capsaicin (10 = median. Significant differences indicated in table.

931 EFFECT OF NALOXONE AND MORPHINE ON CAPSAICIN-INDUCED SECRETION IN BRONCHI IN VITRO

0.05;

I

Significance of change: *p < tp < 001. tp < 0 05 compared with value for no pretreatment.

Stevenage), or the two drugs together. Capsaicin (Sigma Chemical Ltd, Poole) diluted from the stock solution (10 mmol/1 in ethanol) was added to the chambers at the beginning of incubation periods two, six, and ten after baseline measurements to give final concentrations of 0-1umol/1 or 10 lunol/1. Ethanol, at equivalent concentrations, was added as a control at the beginning of the preceding incubation period. The collected fluid was prepared and analysed as previously described20 for the secretory markers fucose, hexose, and protein. Percentage changes in rate of secretion were calculated for the Co

difference in concentration of each marker between the response to ethanol and the subsequent response to capsaicin. The significance of group changes in secretion was assessed by means of the Wilcoxon sign-rank sum test and differences between groups by the Mann-Whitney U test (both two-tailed). The null hypothesis was rejected at p 0-05. Values are expressed as median and range. RESULTS

The change in secretion in response to capsaicin was somewhat variable (see figure and table). Capsaicin (10 (Jmol/1) caused significant rises in secretion of fucose and protein, but not hexose. In the presence of naloxone the changes in secretion of all markers were significant. Capsaicin-induced secretion was inhibited to basal values by morphine, and this effect was reversed by naloxone. At the lower concentration of 0-11 lunol/1, capsaicin raised secretion of fucose by 29-8% (-5-8% to 124%) in the presence of naloxone (p < 005, n 9); it had no significant effect on secretion in any other preparation. =

DISCUSSION

which induces release of neuropeptides from significantly raised secretion of fucose and protein by human bronchi in vitro. We have previously used fucose as an endogenous marker specific for mucus glycoprotein.2° Therefore, stimulation of NANC sensory nerves induces mucus secretion. This result contrasts with the findings of a previous study9 that neural stimulation of human bronchi elicited mucus secretion only by way of cholinergic nerves. However, the electrical features of NANC nerve stimulation are poorly defined and may differ from those of cholinergic nerves. More important, however, is the finding that the response is enhanced in the presence of the opioid antagonist naloxone; this finding lends support to the argument that opioid premedication may account for the lack of response in the earlier study.9 The effect is consistent with the finding of opioid-binding sites on unmyelinated, capsaicin-sensitive sensory nerve fibres,21 the inhibition by opioids of the postsynaptic release of substance P from

Capsaicin,

sensory nerves,

sensory nerves,22 and the opioid inhibition of airway smooth muscle contraction14,15 and vascular leakage16 in response to NANC stimulation. We have now shown a similar effect in human airways. Morphine inhibited secretion and naloxone reversed the inhibition, showing that the response is mediated by way of specific opioid receptors, presumably localised to capsaicin-sensitive sensory nerve fibres related to submucosal glands. Several opioid-receptor subtypes have been differentiated. u-opioid receptors appear to mediate the inhibition of bronchoconstriction,14,15 and morphine is a selective agonist for this receptor subtype. Our fmdings may have important clinical implications. Cigarette smoke selectively triggers capsaicin-sensitive nerves in rat airways to induce microvascular leakage 23 Inhalation of cigarette smoke (and other irritants) may trigger the same nerves in human airways resulting in release of substance P, which stimulates mucus secretion.13 This hypothesis provides an explanation for hypersecretion of mucus in chronic bronchitis. Since the response can be effectively blocked by morphine acting on prejunctional receptors on sensory nerves, opioids may be useful in blocking the airway hypersecretion of chronic bronchitis. The mucus hypersecretion of asthma may also be susceptible to this treatment, since stimulation of sensory nerves by inflammatory mediators may be important.8 The development of -selective opioid agonists which could be given by inhalation or which did not cross the blood-brain barrier would be an important new therapeutic approach to mucus hypersecretion. Furthermore, opioids would inhibit abnormally stimulated mucus secretion, rather than secretion under normal cholinergic control, and might not have the disadvantage of drying up resting secretions in the way that anticholinergic drugs, such as atropine, might be expected to do. We thank Mr P. Goldstraw and the pathologists of Brompton Hospital for making tissue available to us. This study was supported by the Cystic Fibrosis Research Trust.

Correspondence should be addressed to D.

F. R.

REFERENCES Alton

EWFW, Barnes PJ. Airway secretion. In: Cohen RD, Alberti Rogers DF, KGMM, Lewis B, Denman AM, eds. The metabolic and molecular basis of acquired disease: IV Metabolic and molecular aspects of non-metabolic disease. London. Ballière Tindall, 1989: chapter 93 (in press). 2. Richardson PS, Webber SE. The control of mucous secretion in the airways by peptidergic mechanisms. Am Rev Respir Dis 1987; 136: S72-76. 3. Barnes PJ. Neuropeptides in the lung: localization, function and pathophysiologic implications. J Allergy Clin Immunol 1987; 79: 285-95. 4. Peatfield AC, Barnes PJ, Bratcher C, Nadel JA, Davis B. Vasoactive intestinal peptide stimulates tracheal submucosal gland secretion in ferret. Am Rev Respir Dis 1983; 128: 89-93. 5. Coles SJ, Neill KH, Reid LM. Potent stimulation of glycoprotein sercretion in canine trachea by substance P. J Appl Physiol 1984; 57; 1323-27. 6. Borson DB, Charlin M, Gold BD, Nadel JA. Neural regulation of 35SO4macromolecule secretion from tracheal glands of ferrets. J Appl Physiol 1984; 57: 457-66. 7. Peatfield AC, Richardson PS. Evidence for non-cholinergic, non-adrenergic nervous control of mucus secretion into the cat trachea. J Physiol 1983; 342: 335-45. 8. Barnes PJ. Asthma as an axon reflex. Lancet 1986;i: 242-45. 9. Baker B, Peatfield AC, Richardson PS. Nervous control of mucin secretion into human bronchi. J Physiol 1985, 365: 297-305. 10. Lundberg JM, Hokfelt T, Martling C-R, Saria A, Cuello C. Substance Pimmunoreactive sensory nerves in the lower respiratory tract of various mammals including man. Cell Tissue Res 1984; 235: 251-61. 11. Carstairs JR, Barnes PJ. Autoradiographic mapping of substance P receptors in lung. Eur J Pharmacol 1986; 127: 295-96. 12. Barnes PJ. Neuropeptides in human airways: function and clinical implications. Am Rev Respir Dis 1987; 136: S77-83. 13. Rogers DF, Carstairs JR, Alton EWFW, Dewar A, Barnes PJ. Tachykinins and mucus secretion in human bronchi in vitro. Am Rev Respir Dis 1988; 137 (suppl). 12. 14. Frossard N, Barnes PJ. µ-opioid receptors modulate non-cholinergic constrictor nerves in guinea pig airways. Eur J Pharmacol 1987; 141: 519-22. 15. Belvisi MG, Chung KF, Jackson DM, Barnes PJ. Opioid modulation of noncholinergic neural bronchoconstriction in guinea-pig in vivo. Br J Pharmacol 1988; 95: 413-18. 1.

932

Hypothesis LIGANDS SPECIFIC TO PERIPHERAL BENZODIAZEPINE RECEPTORS FOR TREATMENT OF PORPHYRIAS YESHAYAHU KATZ ABRAHAM WEIZMAN MOSHE GAVISH

Rappaport Family Institute for Research in the Medical Sciences and Department of Pharmacology, Faculty of Medicine, TechnionIsrael Institute of Technology, Haifa, Israel; and Geha Psychiatric Hospital, Beilinson Medical Center, Petah-Tiqva, and Sackler School of Medicine, Tel-Aviv University, Tel-Aviv data indicate that Accumulating are porphyrins physiologically endogenous ligands to mitochondrial peripheral benzodiazepine receptors. An isoquinoline carboxamide derivative that likewise binds to peripheral benzodiazepine receptors could prove therapeutically useful in porphyrias by displacing porphyrins from these receptors in mitochondria.

Summary

INTRODUCTION

HAEM

prosthetic group for mitochondrial cytochromes, haemoglobin, and other haem-proteins. Porphyrins are tetrapyrrolic pigments formed in the pathway of haem biosynthesis. Inherited or acquired enzymatic defects in this pathway cause the porphyrias, a group of disorders characterised by excessive production of porphyrins or porphyrin precursors. Porphyrias can be classified into two main categories, hepatic and erythropoietic, according to the site of expression of the metabolic error. Clinically the main effects are seen in the nervous system and the skin. The neurological disorders result in abdominal pain, peripheral neuropathy, and psychiatric disturbances, while the predominant dermatological symptom is photosensitivity. An acute attack ofporphyria can be triggered by factors including infections, emotional stress, diet, and drugs such as oestrogens and barbiturates. The most important aspect of treatment is preventive. Symptomatic treatment consists of major tranquillisers, propranolol, and haematin infusion. In dermal photosensitivity, exposure to sunlight should be serves as

the

avoided.’1 Until lately porphyrins were regarded as a byproduct, the exact function of which was not fully understood. However,

Rogers DF, Barnes PJ. Neurogenic plasma extravasation: inhibition by morphine m guinea pig airways in vivo. J Appl Physiol 1989; 66: 268-72. 17. Buck SH, Burks TF. The neuropharmacology of capsaicin: review of some recent 16. Belvisi MG,

obsevations. Pharmacol Rev 1986; 38: 179-226. 18.

Ferguson CC, Richardson JB. A simple technique for the utilization of postmortem tracheal and bronchial tissue for ultrastructural studies. Human Pathol 1978; 9:

the recent finding by Verma et al2 that porphyrins bind with nanomolar affinity to mitochondrial (peripheral-type) benzodiazepine receptors and can be considered physiologically endogenous ligands for these sites may provide an important clue to the pathophysiology of porphyrias and the means of pharmacotherapy. BENZODIAZEPINE RECEPTORS AND LIGANDS

Benzodiazepines bind to two distinct types of receptor. Central benzodiazepine receptors, located on the cell surface within the central nervous system, mediate the anxiolytic, hypnotic, and anticonvulsant effects of benzodiazepines.3,4 These receptors are coupled to the gamma-aminobutyric acid (GABA) receptors and chloride ion channels.5 Activation of the GABA receptor increases membrane permeability to chloride ions in postsynaptic neurons.6 The neuropsychiatric symptoms of porphyrias have been related to the activity of delta-aminolaevulinic acid (ALA) as a competitor of GABA at the GABA receptor in brain/,8 or to the inhibitory effect of ALA on neuronal GABA uptake.9 However, the peripheral symptoms of porphyrias cannot be explained by interference with GABA transmission. ’ In addition, benzodiazepines bind to peripheral benzodiazepine receptors, which are distributed throughout most organs including kidney, heart, lung, endocrine glands, and also brain.10,11 These receptors are located on the outer mitochondrial membrane12 and are probably coupled to the voltage-operated calcium channels.13

High-affinity

peripheral

benzodiazepine-receptor-mediated

effects possibly involve regulation of intracellular metabolic events rather than intercellular neurochemical events. Among these effects are arachidonic-acid-dependent oxidative metabolism in a murine cell line, 14 chemotaxis of human monocytes,15 and mitogenesis in lymphoma cells. 16 Hirsch et a117 have linked peripheral-type benzodiazepines and porphyrins with regulation of mitochondrial functions. Mitochondrial peripheral benzodiazepine-receptor-specific ligands, as well as porphyrins, decreased the rates of respiratory states III and IV in isolated rat kidney mitochondria. The affinity of porphyrins to peripheral benzodiazepine receptors strongly correlated with their potency as respiratory-rate modifers. In contrast, central benzodiazepine receptor ligands were inactive in this preparation. The benzodiazepine ’Ro 5-4864’ (4’-chlorodiazepam), which at low doses lacks anxiolytic and anticonvulsant properties, displays nanomolar affmity to peripheral receptors. Its affinity to central benzodiazepine receptors is four orders of magnitude lower than to peripheral benzodiazepine receptors. In contrast, clonazepam, a potent ligand for central receptors, binds weakly to peripheral benzodiazepine receptors.18 Diazepam exhibits high affinity to central receptors and moderate affinity to peripheral benzodiazepine receptors." ’PK 11195’, a non-benzodiazepine isoquinoline carboxamide derivative, labels peripheral benzodiazepine receptors selectively and with high-affinity.20 Ro 5-4864 is considered an agonist and PK 11195 an antagonist at the

peripheral benzodiazepine receptors.21 The naturally occurring porphyrins protoporphyrin IX, mesoporphyrin IX, deuteroporphyrin IX, and haemin possess high affinities for peripheral receptors. The precursors, coproporphyrin and uroporphyrin, are less potent, while biliverdin and bilirubin, the breakdown products of porphyrins, bind only weakly to these receptors,22

463-70. 19. Pack RJ, Williams

IP, Phipps RJ, Richardson PS, Rich B. A preparation for the study of secretory function of the human bronchus m vitro. Eur J Respir Dhs 1984; 65: 239-50. 20. Rogers DF, Turner NC, Marriott C, Jeffery PK. Cigarette smoke-induced ’chronic bronchitis’: a study in situ of laryngo-tracheal hypersecretion in the rat. Clin Sci 1987; 72: 629-37. SF, Murrin LC, Kuhar MJ. Presynaptic localization of opiate receptors m the vagal and accessory optic systems: an autoradiographic study. Neuropharmacology 1981; 17: 101-04. 22. Jessel TM, Iversen LL. Opiate analgesics inhibit substance P release from rat trigeminal nucleus. Nature 1977; 268: 549-51. 23. Lundberg JM, Saria A. Capsaicin-induced desensitization of airway mucosa to cigarette smoke, mechanical and chemical irritants. Nature 1983; 302: 251-53. 21. Atweh

HYPOTHESIS

We propose a new mode of treatment for porphyrias, based on the notion that activity of porphyrins at the mitochondrial peripheral benzodiazepine receptors is relevant to symptoms. Ligands specific to peripheral benzodiazepine receptors could be used to compete with and block those effects of porphyrins. Ro 5-4864 has been administered in the past to man,-" however, the use of this agent in porphyrias is problematic since Ro 5-4864 is