Epithelial factors: modulation of the airway smooth muscle tone

Airway epithelium-derived relaxing factor: myth, reality, or naivety?

The presence of a healthy epithelium can moderate the contraction of the underlying airway smooth muscle. This is, in part, because epithelial cells generate inhibitory messages, whether diffusible substances, electrophysiological signals, or both. The epithelium-dependent inhibitory effect can be tonic (basal), synergistic, or evoked. Rather than a unique epithelium-derived relaxing factor (EpDRF), several known endogenous bronchoactive mediators, including nitric oxide and prostaglandin E2, contribute. The early concept that EpDRF diffuses all the way through the subepithelial layers to directly relax the airway smooth muscle appears unlikely. It is more plausible that the epithelial cells release true messenger molecules, which alter the production of endogenous substances (nitric oxide and/or metabolites of arachidonic acid) by the subepithelial layers. These substances then diffuse to the airway smooth muscle cells, conveying epithelium dependency.

Altered airway responsiveness in adult sheep born prematurely: effects of allergen exposure

The aim of this study was to determine the effects of preterm birth per se on airway function in adult sheep. Preterm birth was induced at approximately 0.89 of term. At approximately 1 year of age the authors measured pulmonary resistance (RL) and airway responsiveness before and after house dust mite (HDM) challenge. Mature preterm sheep tended to have greater baseline RL than controls (P = .12): the smaller preterm sheep showed significantly greater RL than controls following bronchoconstrictor challenge. Preterm animals tended to have greater baseline total blood leukocyte count (P = .06). It was concluded that preterm sheep, especially with low postnatal growth, have greater airway responsiveness to bronchoconstrictor and higher baseline RL.

EETs relax airway smooth muscle via an EpDHF effect: BK(Ca) channel activation and hyperpolarization

Epoxyeicosatrienoic acids (EETs) are produced from arachidonic acid via the cytochrome P-450 epoxygenase pathway. EETs are able to modulate smooth muscle tone by increasing K(+) conductance, hence generating hyperpolarization of the tissues. However, the molecular mechanisms by which EETs induce smooth muscle relaxation are not fully understood. In the present study, the effects of EETs on airway smooth muscle (ASM) were investigated using three electrophysiological techniques. 8,9-EET and 14,15-EET induced concentration-dependent relaxations of the ASM precontracted with a muscarinc agonist (carbamylcholine chloride), and these relaxations were partly inhibited by 10 nM iberiotoxin (IbTX), a specific large-conductance Ca(2+)-activated K(+) (BK(Ca)) channel blocker. Moreover, 3 microM 8,9- or 14,15-EET induced hyperpolarizations of -12 +/- 3.5 and -16 +/- 3 mV, with EC(50) values of 0.13 and 0.14 microM, respectively, which were either reversed or blocked on addition of 10 nM IbTX. These results indicate that BK(Ca) channels are involved in hyperpolarization and participate in the relaxation of ASM. In addition, complementary experiments demonstrated that 8,9- and 14,15-EET activate reconstituted BK(Ca) channels at low free Ca(2+) concentrations without affecting their unitary conductance. These increases in channel activity were IbTX sensitive and correlated well with the IbTX-sensitive hyperpolarization and relaxation of ASM. Together these results support the view that, in ASM, the EETs act through an epithelium-derived hyperpolarizing factorlike effect.

Impairment of epithelium-dependent relaxation in coaxial bioassay by reactive oxygen species

The purpose of the present study was to investigate the effects of reactive oxygen species on the activity of epithelium-derived relaxant factor (EpDRF) released by guinea-pig tracheal epithelium. Reactive oxygen species were generated by the electrolysis of the physiological buffer in which the tissues were bathed. Epithelium-dependent relaxation induced by acetylcholine in precontracted rat anococcygeus muscle that was placed in epithelium-intact guinea-pig trachea (coaxial bioassay system) was significantly attenuated when the tissues were exposed to electrolysis. Impairment of the acetylcholine response was prevented by incubation with free radical scavengers prior to electrolysis. In isolated rings of guinea-pig trachea, the contractile responses elicited by acetylcholine, histamine and 5-hydroxytryptamine were not altered after electrolysis of the bathing solution. The results of the present study suggested that exposure to reactive oxygen species impaired EpDRF release from guinea-pig trachea epithelium but did not alter the contractility of tracheal smooth muscle.

Modulation of ET-1-induced contraction of human bronchi by airway epithelium-dependent nitric oxide release via ET(A) receptor activation

1. The purpose of this work was to investigate whether endothelin-1 (ET-1) was able to induce the release of an inhibitory factor from the airway epithelium in isolated human bronchi and to identify this mediator as well as the endothelin receptor involved in this phenomenon. 2. In intact bronchi, ET-1 induced a concentration-dependent contraction (-logEC50 = 7.92+/-0.09, n = 18) which was potentiated by epithelium removal (-logEC50 = 8.65+/-0.11, n = 17). BQ-123 , an ET(A) receptor antagonist, induced a significant leftward shift of the ET-1 concentration-response curve (CRC). This leftward shift was abolished after epithelium removal. 3. L-NAME (3 x 10(-3) M), an inhibitor of nitric oxide (NO) synthase, induced a significant leftward shift of the ET-1 CRC, and abolished the potentiation by BQ-123 (10(-8) M) of ET-1-induced contraction. 4. In intact preparations, the ET(B) receptor antagonist BQ-788 induced only at 10(-5) M a slight rightward shift of the ET-1 CRC. In contrast, in epithelium-denuded bronchi or in intact preparations in the presence of L-NAME, BQ-788 displayed a non-competitive antagonism toward ET-1-induced contraction. 5. IRL 1620, a selective ET(B) receptor agonist, induced a contraction of the isolated bronchus (-logEC50=7.94+/-0.11, n= 19). This effect was not modified by epithelium removal or by BQ-123. BQ-788 exerted a competitive antagonism against IRL 1620 which was similar in the presence or absence of epithelium. 6. These results show that ET-1 exerts two opposite effects on the human airway smooth muscle. One is contractile via ETB-receptor activation, the other is inhibitory and responsible of NO release which counteracts via ETA-receptor activation the contraction.

Epoxyeicosatrienoic acids relax airway smooth muscles and directly activate reconstituted KCa channels

Epoxyeicosatrienoic acids (EETs) relax various smooth muscles by increasing outward K+ movement, but the molecular mode of action of EET regioisomers remains to be clarified. The effects of EETs were investigated on bovine airway smooth muscle tone and on reconstituted Ca2+-activated K+ (KCa) channels. 5,6-EET and 11, 12-EET induced dose-dependent relaxations of precontracted bronchial spirals. These effects were partly abolished by 10 nM iberiotoxin. Bilayer experiments have shown that 0.1-10 microM 11,12-EET produced up to fourfold increases in the open probability of KCa channels from the cis (extracellular) side by enhancing the mean open time constant and reducing the long closed time constant, without affecting the unitary conductance. EET-induced activations were blocked by 10 nM iberiotoxin. Addition of vehicles or other lipids as well as of GTP and guanosine 5'-O-(3-thiotriphosphate) in the absence of EET had no effect on channel activity. Thus EETs directly activate KCa channels from airway smooth muscle through an interaction with the extracellular face of the channel. We propose that EETs could represent candidate molecules as epithelium-derived hyperpolarizing factors.

Airway epithelium: more than just a barrier!

Airway hyper-responsiveness and epithelial cell damage are associated commonly with asthma. The airway epithelium is a physical barrier that protects sensory nerves and smooth muscle from stimulation by inhaled irritants. In addition, epithelial cells release mediators that can inhibit bronchoconstriction by relaxing the underlying smooth muscle: so-called 'epithelium-derived relaxing factors' (EpiDRFs). Clear functional evidence for EpiDRFs is provided by experiments where different endogenous mediators induce the relaxation of tracheas containing epithelium, but cause a contraction in preparations lacking this layer. Here, Gert Folkerts and Frans Nijkamp describe the pharmacological relevance of the putative EpiDRFs, prostaglandin E2 and NO, in the modulation of airway tone under basal conditions in vitro and in vivo. Special attention is paid to the role of both EpiDRFs in the development of airway hyper-responsiveness in animal models and in patients with asthma.

Role of tachykinins in bronchial hyper-responsiveness

1. Sensory afferent fibres mediate important protective reflexes in the lung. Small, unmyelinated C-fibre nerves have both sensory afferent and effector functions. C-fibres contain a number of neuropeptides, including the tachykinins, which have pro-inflammatory effects in the airways. Following stimulation with capsaicin and other stimuli, neuropeptides are released from the nerve endings, either directly or by axonal reflexes. 2. Important tachykinin effects include smooth muscle contraction, vasodilatation and oedema, mucus secretion and inflammatory cell activation. There are also trophic effects, including proliferation of fibroblasts, smooth muscle and epithelial cells. 3. Tachykinins mediate their effects by binding to G-proteinlinked receptors. Receptor-specific agonists and antagonists are available, which have helped clarify the effects of tachykinins. These agents may have therapeutic potential. 4. Tachykinins are degraded by the enzyme neutral endo-peptidase. 5. Studies in humans in vivo show an increase in airways resistance following challenge with tachykinins. There is some evidence for an increase in tachykinins and their receptors in airway inflammation, but this has not been found in all studies. A reduction in neutral endopeptidase has been seen in some animal models of airway inflammation, but this has not been shown in human disease. 6. Trials of tachykinin receptor antagonists in human asthma have begun, but it is too early to say what their therapeutic impact will be.

Deficiency of nitric oxide in allergen-induced airway hyperreactivity to contractile agonists after the early asthmatic reaction: an ex vivo study

1. Using a guinea-pig model of allergic asthma, we investigated the role of nitric oxide (NO) in allergen-induced airway hyperreactivity after the early asthmatic reaction, by examining the effects of the NO-synthase inhibitor N omega-nitro-L-arginine methyl ester (L-NAME) on the responsiveness to methacholine and histamine of isolated perfused tracheae from unchallenged (control) animals and from animals 6 h after ovalbumin challenge. 2. All animals developed airway hyperreactivity to inhaled histamine at 6 h after ovalbumin challenge, with a mean 3.11 +/- 0.45 fold increase in sensitivity to the agonist (P < 0.001). 3. In perfused tracheal preparations from the ovalbumin-challenged guinea-pigs, the maximal responses (Emax) to methacholine and histamine were significantly enhanced compared to controls, both after intraluminal (IL) and extraluminal (EL) administration of the contractile agonists. In addition, a small but significant increase in the pD2 (-log10 EC50) for IL and EL methacholine and for IL histamine was observed. As a consequence, the delta pD2 (EL-IL) for histamine was slightly decreased from 1.67 +/- 0.13 to 1.23 +/- 0.14 (P < 0.05). However, the delta pD2 for methacholine was unchanged (1.85 +/- 0.11 and 1.77 +/- 0.12, respectively; NS). 4. Incubation of control tracheae with 100 microM L-NAME (IL) significantly enhanced the Emax for both IL and EL methacholine and histamine to approximately the same degree as observed after ovalbumin challenge, with no effect on the pD2 and delta pD2 for both agonists. On the contrary, L-NAME had no effect on Emax and pD2 values of tracheal preparations from ovalbumin-challenged guinea-pigs. 5. L-NAME (10 microM-1 mM) had no effect on methacholine-induced contraction of isolated tracheal strip preparations obtained from control animals, indicating that L-NAME has no antimuscarinic effect on tracheal smooth muscle. 6. Histological examination of the intact tracheal preparations indicated epithelial and subepithelial infiltration of eosinophils after ovalbumin challenge. However, no apparent damage of the airway epithelium was observed in these preparations. 7. The results indicate that a deficiency of NO contributes to allergen-induced airway hyperreactivity after the early asthmatic reaction and that this deficiency appears not to be due to epithelial shedding.

Epithelial modulation of cholinergic responses in rabbit trachea is partly due to neutral endopeptidase activity

By the simultaneous measurement of acetylcholine release and smooth muscle contraction in rabbit tracheal segments with and without epithelium, pre- as well as postsynaptic effects of this cell layer were studied on cholinergic neurotransmission. The epithelial cell layer exerted a presynaptic inhibitory influence on acetylcholine release, induced by KCl and electrical stimulation, with a concomitant decrease in the smooth muscle contractions. The responses elicited by exogenous acetylcholine, acting postsynaptically, were also inhibited in the presence of the epithelium. The epithelial effect was not accounted for by the production of inhibitory prostaglandins or a nitric oxide-synthase product. Furthermore, the epithelium did not function as a metabolic site for the degradation of acetylcholine. Phosphoramidon, an inhibitor of neutral endopeptidase, mimicked the effects of epithelium removal on the cholinergic responses to high frequency stimulation and on the acetylcholine-induced effects. Neutral endopeptidase inhibition did not further enhance the responses in epithelium-denuded segments. We therefore suggest that the inhibitory function of the epithelium can be partly explained by the activity of neutral endopeptidase, limiting the excitatory effects of tachykinins on cholinergic responses. An alteration in the neutral endopeptidase activity as a result of inflammatory responses and epithelial damage can contribute to the mechanism of airway hyperreactivity in asthma.

Evidence for reduction of bradykinin-induced bronchoconstriction in guinea-pigs by release of nitric oxide

1. In this study the influence of nitric oxide (NO) on the bronchoconstriction induced by bradykinin in anaesthetized and artifically ventilated guinea-pigs pretreated with atropine was investigated. 2. Aerosol administration of bradykinin (0.1-1 mM, 40 breaths) caused a dose-dependent increase in lung resistance (RL): maximum increase in RL was 2.5 fold the baseline value. Pretreatment with aerosolized NG-nitro-L-arginine methyl ester (L-NAME) or NG-monomethyl-L-arginine (L-NMMA) (1 mM, 10 breaths every 5 min for 30 min), NO synthase inhibitors, markedly increased the bronchoconstrictor response to bradykinin. L-Arginine, but not D-arginine, (3 mM, 10 breaths every 5 min for 30 min) reversed the hyperresponsiveness to aerosolized bradykinin caused by L-NAME and L-NMMA. 3. L-NAME (1 mM, 10 breaths every 5 min for 30 min) increased the bronchoconstriction induced by intravenous bradykinin (1-10 nmol kg-1). L-Arginine, but not D-arginine, (10 breaths every 5 min for 30 min) reversed the hyperresponsiveness to intravenous bradykinin caused by L-NAME. 4. The increase in RL induced by capsaicin, either aerosol (10 microM, 10 breaths) or i.v. (20 nmol kg-1) was not affected by L-NAME (1 mM, 10 breaths every 5 min for 30 min). Acute resection of the vagi did not affect the bronchoconstriction evoked by bradykinin in guinea-pigs, either in the absence or presence of L-NAME (1 mM, 10 breaths every 5 min for 30 min). 4. These results suggest that, irrespective of the route of administration, bradykinin releases NO or a related molecule which exerts a bronchodilator action that opposes the bronchoconstrictor mechanisms activated by bradykinin itself.

Epithelium modulates the kinetics of the response to substance P and its intrinsic activity in the guinea-pig trachea

The contractile response of guinea-pig tracheal preparations with or without epithelium to substance P has been studied in the presence or absence of thiorphan, an endopeptidase 24.11 inhibitor, paying special attention to the kinetics of the response. Without thiorphan, the response to substance P was greater in tracheal preparations without epithelium than in tracheal preparations with epithelium. The concentration-response curve was shifted to the left in the absence of the epithelium. In the presence of 10 microM thiorphan, the maximal contractile response induced by single doses of substance P (0.1 to 10 microM) was lower in tracheal preparations without epithelium. The maximal responses required 10 min in tracheal preparations with epithelium and 2 min in tracheal preparations without epithelium. These epithelium-dependent differences of reactivity remained in the presence of lipoxygenase or cyclooxygenase inhibitors and of selective antagonists of muscarinic, serotoninergic and histaminergic receptors, after the pre-treatment of tissues with capsaicin or compound 48/80 and in the presence of tetrodotoxin. The profile of the cumulative concentration-response curves for substance P was largely dependent on the time between two successive doses. When this time was short (2-4 min), curves established with or without the epithelium were parallel and both reached similar maximal values (2696 +/- 214 mg and 2780 +/- 62 mg, respectively). The curve in tracheal preparations without epithelium was slightly shifted to the left (EC50s: 24 +/- 10 nM and 78 +/- 19 nM). When this time was longer (10 min, ie corresponding to the time required for a full response to a single dose in intact trachea) the potency of substance P was not modified (EC50s: 13 +/- 3 nM and 52 +/- 11 nM), but a lower maximal response was observed with tracheal preparations without epithelium (1440 +/- 182 mg and 2832 +/- 209 mg). Similar results were observed with neurokinin A and neurokinin B. Thus, the removal of the epithelium led to a more rapid contraction and to a decrease of the maximal response to neurokinins, ie a decreased intrinsic activity, a property known to be drug- and tissue-dependent. These data suggest that the intrinsic activity of drugs depends on the cellular environment of the target cells in a tissue and is partly related to the diffusion and metabolism of drugs and to drug-induced hyporeactivity of the target cells.