Mechanisms of leukotriene D₄-induced constriction in human small bronchioles

Abstract

1. We examined the mechanisms underlying leukotriene D₄- (LTD₄) induced constriction of human small (300 – 500 μm i.d.) bronchioles, and the effect of LTD₄ on ion currents and Ca²⁺ transients in smooth muscle cells (SMC) isolated from these bronchioles.

2. LTD₄ caused a concentration-dependent bronchoconstriction with an EC₅₀=0.58±0.05 nM (n=7) which was not easily reversible upon washout. This bronchoconstriction was entirely dependent on extracellular Ca²⁺.

3. Blockade of L-type Ca²⁺ channels with nifedipine (10 μM) reduced LTD₄ response by 39±2% (n=8), whilst La³⁺, Gd³⁺ and SK&F 96,365 abolished LTD₄-induced bronchoconstriction completely and reversibly, suggesting the majority of Ca²⁺ entry was via non-selective cation channels.

4. Antagonists of PI-PLC (U73,122 and ET-18-OCH₃), PLD (propranolol) and PKC (cheleretrine and Ro31-8220) were without any effect on LTD₄-induced bronchoconstriction, whilst the PC-PLC inhibitor D609 caused complete relaxation. Inhibition of protein tyrosine kinase with tyrphostin A23 (100 μM) caused about 50% relaxation, although the inactive analogue tyrphostin A1 was without effect.

5. In freshly isolated SMC from human small bronchioles LTD₄ caused a slow increase of intracellular Ca²⁺ concentration, with a consequent rise of the activity of large conductance Ca²⁺-dependent K⁺ channels and the amplitude of depolarization-induced outward whole-cell current. Again, no effect of LTD₄ could be observed in the absence of extracellular Ca²⁺.

6. We conclude that LTD₄ causes constriction of these small bronchioles primarily by activating Ca²⁺ entry via non-voltage gated channels, possibly by a PC-PLC mediated pathway.

Introduction

Cysteinyl leukotrienes are potent pro-inflammatory lipid mediators that are produced by many inflammatory cells, and cause smooth muscle constriction, hyper-secretion of mucus, and increased vascular permeability. There is clear evidence that leukotriene D₄ (LTD₄) plays a major role in the pathophysiology of asthma and other inflammatory diseases (Arm & Lee, 1993). However, the specific receptors and transduction pathways remain poorly understood, particularly in human airways. Moreover, they may differ considerably between different tissues and species (Crooke et al., 1988). It has been suggested that LTD₄ induces bronchoconstriction through phosphatidylinositol phospholipase C (PI-PLC) activation, as LTD₄ increases inositol polyphosphates in guinea-pig ASM (Dumitriu et al., 1997; Howard et al., 1992), and LTD₄-induced contraction of guinea-pig ASM is reduced by PLC inhibition (Salari et al., 1993). However other signal transduction pathways are also possible, involving G_i proteins, phosphatidylcholine phospholipase C (PC-PLC)- and phospholipase D (PLD)-mediated generation of diacylglycerols (DAG), activation of protein kinase C (PKC) and sustained contraction and Ca²⁺ sensitization (Kim et al., 1998). Another alternative to the IP₃-mediated mechanism is one that includes the membrane-delimited conversion of DAG generated by PC-PLC or PLD into phosphatidic acid by DAG kinase, with subsequent Ca²⁺ release from IP₃-insensitive intracellular stores (Camiña et al., 1999).

It has been predicted that much of the resistance to airflow seen during agonist challenge is due to contraction of small (<∼2 mm i.d.) airways (Moreno et al., 1986). However there have been very few in vitro studies on these functionally significant small bronchioles, particularly in man, and the large majority of reports have been on tissue derived from trachea or main bronchus of other species. This may have important consequences, as significant pharmacological and mechanical differences have been reported between airways from different parts of the bronchial tree (Gauthier et al., 1992; Mustafa et al., 1994), including our own studies on small bronchioles of the rat (Chopra et al., 1994; 1997). Moreover we have demonstrated that the electrophysiology of human ASM differs from other species commonly used in airways research (Snetkov et al., 1995; 1998), and more recently that ASM from human small bronchioles (<1 mm i.d.) have a different distribution of ionic channels and currents than ASM from main bronchus (Snetkov & Ward, 1999). This parallels the well known variation in function that occurs between arteries of different calibre within the vascular tree (e.g. Archer, 1996).

The contraction of smooth muscle from different tissues depends to a variable extent on Ca²⁺ influx across the plasma membrane. Though a number of ion channels have been implicated in such influx, the role of L-type voltage gated Ca²⁺ channels in human ASM seems to be a relatively minor one. Attention has been recently drawn to store-operated Ca²⁺ channels (SOC) which are activated upon emptying of IP₃-dependent Ca²⁺ stores (Putney & Mckay, 1999), and to a variety of channels directly activated by second messengers such as Ca²⁺, IP₃, IP₄, arachidonic acid and its metabolites (for review see Kiselyov & Muallem, 1999).

The aim of the present study was to establish the role for different sources of Ca²⁺ in the constrictor response of functionally important human small bronchioles to LTD₄, and to elucidate the signalling mechanisms involved.

Methods

Human intralobular bronchioles were dissected from tissue obtained after lobectomies within 90 min of removal from the patient, as approved by the Guy's and St Thomas' Hospital Committee on Ethics. In total 46 lung specimens from patients aged 49 – 88 years have been used.

Contraction measurements

1 – 2 mm segments of epithelium-intact bronchioles (300 – 500 μm i.d.) were cleaned of parenchyma, mounted on a small vessel myograph (Myodata 1200, Denmark), and stretched to 80% of the maximum of the active length – tension curve as previously described (Chopra et al., 1997). The chamber contained physiological saline (PSS; for composition see below), equilibrated with 95% O₂/5% CO₂ at 37°C. After 45 min preparations were challenged with at least three 4 min exposures to PSS containing 80 mM KCl (KPSS; equimolar substitution for NaCl). Experiments were performed once the response to KPSS was repeatable and stable.

Cells isolation

Epithelium-denuded bronchiole segments were incubated for 60 min at 36°C in 2 ml oxygenated M-199 medium containing 2 mg ml⁻¹ collagenase (Sigma type XI), 1 mg ml⁻¹ papain, 2 mg ml⁻¹ soyabean trypsin inhibitor and 20 μM dithiothreitol. Then tissue was then washed with fresh M-199 medium and kept at 36°C for further 20 min, following which the tissue was triturated gently with a flame-polished Pasteur pipette in 0.5 ml M-199 in order to release single cells. Five to ten min before use a drop of the cell suspension was placed on a alcohol-cleaned glass coverslip. This procedure provided sufficient number of easy identifiable, relaxed smooth muscle cells as well as a number of active ciliated epithelial cells.

Intracellular Ca²⁺

Isolated cells were incubated with 0.5 μM Fura-PE3/AM for 30 min at room temperature. A drop of cell suspension was placed at the bottom of a recording chamber (RC-26, Warner Instrument Corp.) consisting of a 13 mm glass coverslip (thickness 0), and mounted on the stage of inverted microscope (Nikon Diaphot). Attached cells were visualized with a ×100 oil immersion UV objective, and the ratio of emission intensities at excitation wavelengths 340 and 380 nm were recorded using a microspectrofluorimeter (Cairn Research Ltd., U.K.).

Electrophysiology

Single cells were studied with standard patch-clamp techniques using an Axopatch-1C amplifier (Axon Instruments, Inc., Foster City, CA, U.S.A.), as previously described (Snetkov & Ward, 1999). The experiments were carried out at room temperature. The bath (≈0.5 ml) was continuously superfused (1-2 ml min⁻¹) with external physiological salt solution (PSS, see below). Agonists and drugs were added through the perfusion system. Whole-cell recording was performed in permeabilized patch mode with 150 μg ml⁻¹ nystatin in the pipette. Whole-cells and single channel currents were filtered at 5 or 1 kHz respectively (−3 dB, low pass 80 dB/dec, Bessel type built-in filter) and digitised with a 20 or 5 kHz sampling frequency using a CED 1401 interface and PATCH v.6 software (Cambridge Electronic Design, Cambridge, U.K.). Whole-cell current-voltage relationships were obtained using voltage ramp protocol when holding potential was kept at −60 mV and 0.5 s ramp from −100 mV to +100 mV was applied each 5 s.

Solutions and reagents

PSS contained (in mM): NaCl 118, NaHCO₃ 24, MgSO₄ 1, NaH₂PO₄ 0.435, glucose 5.56, Na-pyruvate 5, CaCl₂ 1.8, KCl 4, pH 7.4 when equilibrated with 95% O₂/5% CO₂. Ca²⁺-free solution contained no added Ca²⁺ and 1 mM EGTA. The pipette solution used both in cell-attached and whole-cell recording contained (in mM): KCl 140, MgCl₂ 2, EGTA 0.1, HEPES 10, ATP 2, GTP 0.2, pH 7.2 with KOH. The normal bath solution for patch-clamp experiments and intracellular Ca²⁺ recording contained (in mM): NaCl 135, KCl 5, MgCl₂ 2, CaCl₂ 2, glucose 11, HEPES 10, pH 7.4 with NaOH. The same solution was used for contraction experiments involving La³⁺, Gd³⁺ and Ni²⁺ to prevent precipitation. D609, U73,122, SK&F 96,365 and MDL-12,330A were obtained from Calbiochem, Notts., U.K., and 2-aminoethoxydiphenyl borate (2-APB) from Tocris Cookson Ltd., Bristol, U.K. All other drugs and reagents were purchased from Sigma Chemical Company (Poole, U.K.).

Data are presented as mean±s.e.mean, and n is the number of separate preparations. Fitting was performed using the Marquardt-Levenberg algorithm (SigmaPlot v.6, SPSS Inc.)

Results

Agonist-induced bronchoconstriction

In order to characterise the contractile response of these small bronchioles, concentration-response curves were constructed for LTD₄, LTE₄, carbachol and histamine. LTD₄ induced contraction with a threshold concentration of ∼0.03 nM and an EC₅₀=0.58±0.05 nM (n=7), which is smaller than previously reported for larger human bronchi (11 nM: Bourdillat et al., 1987; 3.5 nM: Hay et al., 1993). The maximally effective concentration was 10 nM, giving a contractile response of ∼150% of the response to KPSS. LTE₄ was significantly less potent (EC₅₀=4.6±0.3 nM (n=6; P<0.01). The EC₅₀ for carbachol and histamine were 213±17 nM (n=8) and 446±26 nM (n=6) respectively (Figure 1a). Preincubation with 10 μM indomethacin for 30 min had no effect on LTD₄-induced contraction (n=4). 3 mM L-cysteine, an inhibitor of the degradation of LTD₄ to LTE₄ by aminopeptidases (Bourdillat et al., 1987) did not alter the dose-response of LTD₄ (n=3). The CysLT₁ antagonist ICI 198,615 (0.1 μM) caused complete relaxation of preparations constricted with 10 nM LTD₄ (n=6). Prior addition of ICI 198,615 also completely blocked the response to 10 nM LTD₄, without having any effect on the response to KPSS or 10 μM carbachol (n=3).

Figure 1.

(a) Dose – response for agonist-induced contractions in human small bronchioles (n=6 – 8). (b) Contraction of human small bronchiole induced by increasing concentrations of LTD 4. W indicates repetitive washes with normal PSS.

Effect of Ca²⁺ removal

In marked contrast to the response to such agonists as carbachol (10 μM) and histamine (50 μM), LTD₄-induced contraction was not readily reversible and was sustained after several washes. However, replacement of PSS with a Ca²⁺-free solution containing 1 mM EGTA led to immediate relaxation (n=11) (Figure 1b). This was reversible upon restoration of extracellular Ca²⁺. In Ca²⁺-free solution even high concentrations of LTD₄ (10 nM) were without effect (n=7). However, reintroduction of Ca²⁺ following challenge with LTD₄ induced a maximal response, even after prolonged washing with Ca²⁺-free solution (Figure 2). As this effect was observed even when the LTD₄ antagonist ICI 198,615 was present during washout, it was not due to continued binding of LTD₄ to its receptor.

Figure 2.

LTD₄ does not contract bronchiole in Ca-free solution; reintroduction of normal Ca²⁺ induced a contractive response.

Mechanisms of Ca²⁺ entry

To investigate possible pathways of Ca²⁺ entry we studied the effects of various Ca²⁺ channels blockers on the sustained contraction induced with LTD₄. Nifedipine (10 μM), a blocker of L-type voltage-gated Ca²⁺ channels, produced only partial relaxation of bronchioles preconstricted with 10 nM LTD₄ (39±2% initial tension, n=8) (Figure 3a). Consistent with this, preincubation with nifedipine reduced the subsequent response to LTD₄ to 62±3% of control (n=7). Blockade of Ca²⁺ influx with La³⁺ (1 mM, n=7) or SK&F 96,365 (100 μM, n=6) caused complete relaxation of bronchioles constricted with 10 nM LTD₄ (Figure 3b,c respectively). Lower concentrations of both La³⁺ and SK&F 96,365 were less effective and more variable in effect between preparations; 100 μM La³⁺ caused only 25 – 30% relaxation (n=7), whereas 10 μM SK&F 96,365 caused only ∼15% relaxation. Gd³⁺ (100 μM and 1 mM) had the same effect as La³⁺, while Ni²⁺ (1 mM) was without any effect (n=4). MDL-12,330A, otherwise used as an adenylyl cyclase inhibitor, has been reported to block Ca²⁺ entry in a variety of cell types in a similar but more reproducible fashion to SK&F 96,365, independently of its effect on adenylate cyclase (Van rossum et al., 2000). In these human bronchioles it blocked 35% of the LTD₄ (10 nM) response at 10 μM and 90% at 100 μM (n=3). 2-APB, a membrane-permeable IP₃ receptor antagonist, has been shown to prevent activation of both endogenous SOC and transfected TRP3 in HEK293 cells (Ma et al., 2000). 2-APB at 75 μM had no effect on LTD₄ (10 nM) induced bronchoconstriction (n=4), although this concentration was sufficient to abolish SOC in pulmonary arteries and human cultured ASM cells (Snetkov, unpublished observations).

Figure 3.

Nifedipine blocks LTD₄-induced contraction only partially (a), while La³⁺ (b) and SK&F96,365 (c) relax bronchioles completely.

Signal transduction

Neither inhibition of PI-PLC with U73,122 (100 μM, n=5) or ET-18-OCH₃ (100 μM, n=3), or PLD with propranolol (100 μM, n=4) had any effect on LTD₄-induced bronchoconstriction. In contrast, the PC-PLC antagonist D609 (1 – 100 μM) relaxed the LTD₄-induced constriction in a concentration-dependent manner with a IC₅₀ ≈ 5 μM. At 100 μM it often produced depression below the baseline, with return to baseline tone after washout of D609 (n=8) (Figure 4). However the response to 10 μM carbachol was only reduced by D609 to 57±5% (n=6) even at 100 μM, whilst the response to KPSS was not affected (n=14). Preincubation with the PKC inhibitors cheleretrine (10 μM, n=7) or Ro31-8220 (10 μM, n=3) had no effect on LTD₄-induced constriction. In contrast the protein tyrosine kinase inhibitor tyrphostin A23 (100 μM for 10 min) reduced the LTD₄ response to 49±11% (n=4); the inactive analogue tyrphostin A1 was however without effect. The effects of inhibition of the various phospholipases and protein kinases are summarized in Figure 5.

Figure 4.

PI-PLC inhibitor U73,122 has no effect on LTD₄-induced contraction, while PC-PLC inhibitor D609 causes complete relaxation.

Figure 5.

Effects of inhibitors of PI-PLC (U73,122 and ET-18-OCH3), PC-PLC (D609), PLD (propranolol), PKC (cheleretrine and Ro31 – 8220) and PTK (tyrphostin A23) on LTD4-induced contraction. Inactive tyrphostin A1 was used as control. n=4 – 8.

Basal tone

Human small bronchioles mounted on the myograph often possessed a variable basal tone which could be reversible abolished by removal of extracellular Ca²⁺ (n=14), or addition of 1 mM La³⁺ (n=4) or 100 μM SK&F 96,365 (n=4) (Figure 6a). 10 μM nifedipine had no significant effect on basal tone (n=12). D609 (100 μM) also caused complete relaxation of basal tone (n=4) (Figure 6b), while U73,122 and ICI 198,615 were without effect (n=5). It should be noticed that the preparations with the highest Ca-dependent basal tone demonstrated the smallest response to LTD₄, and vice versa.

Figure 6.

(a) Basal tone in bronchiole could be abolished by Ca²⁺ removal or La³⁺, but not with nifedipine. (b) Variable amount of basal tone present in bronchioles dissected from the same specimen is blocked by PC-PLC inhibitor D609.

Intracellular stores

To study a role for intracellular Ca²⁺ stores, bronchioles were challenged with 10 μM cyclopiazonic acid (CPA) which induces Ca²⁺ release by inhibiting sarcoplasmic Ca-ATPase. CPA induced a maximal response in normal PSS, but did not cause any constriction in the absence of extracellular Ca²⁺ (n=3). However, restoration of normal Ca²⁺ after CPA had been washed out with a Ca²⁺-free solution caused a substantial constriction, similar to that seen for LTD₄. No constriction was ever observed with caffeine (5 mM, n=7) or ryanodine (20 μM, n=2); in fact caffeine commonly caused a small fall in baseline tone.

Intracellular Ca²⁺ and electrophysiology

10 nM LTD₄ caused visible shortening of the freshly isolated human bronchiole SMC loaded with Fura PE-3, accompanied with a slower rise in intracellular Ca²⁺ concentration, though this was much smaller than that induced with either caffeine, KPSS or CPA (Figure 7). Consistent with this rise in intracellular Ca²⁺, 10 nM LTD₄ considerably increased the open state probability of large conductance, Ca²⁺-dependent K⁺-channels recorded in cell-attached configuration (n=6). LTD₄ also greatly enhanced the activity of small conductance (∼25 pS) channels present in some patches (Figure 8a), which persisted long after the activity of large conductance K⁺ channels returned to control levels. In perforated patch whole-cell recording LTD₄ enhanced an outward whole-cell current induced with depolarisation beyond +50 mV (n=5). Such enhancement persisted during prolonged washout of agonist, and could be abolished quickly and reversibly with the removal of extracellular Ca²⁺ or addition of 100 μM La³⁺ (Figure 8b), or 10-100 μM SK&F 96,365 (n=5, data not shown). In current-clamp mode 10 nM LTD₄ failed to induce any change of the resting membrane potential (−40.−50 mV). At higher concentrations (>50 nM) LTD₄ evoked Ca²⁺ transients in Fura PE-3 loaded cells, as well as a transient inward Cl⁻ current, similar to the effects observed with caffeine (5 mM) when applied to the same cells (Figure 9).

Figure 7.

Changes in intracellular Ca²⁺ concentration measured in single smooth muscle cell isolated from human bronchiole. Data presented as Fura PE-3 emission ratio at excitation wavelengths of 340 and 380 nm.

Figure 8.

(a) LTD₄ induces rise of activity of Ca²⁺-dependent large conductance K⁺ channels and smaller non-selective cation channels in cell-attached patch from single smooth muscle cell isolated from human bronchiole. Bath and pipette solution contained 140 mM KCl; membrane potential +60 mV. (b) Whole-cell responses to voltage ramp from −100 mV to +100 mV (holding potential between ramps −60 mV). Bath solution 5 mM KCl, pipette solution 140 mM KCl.

Figure 9.

High concentration of LTD₄ evokes Ca²⁺ transients in single Fura PE-3-loaded myocytes similar to that induced with caffein (a), as well as transient Cl⁻ current in whole-cell voltage clamp condition (b).

Discussion

The results of the present study suggest that the contractile response to LTD₄ in human small bronchioles is mediated by the CysLT₁ receptor, which has recently been cloned (Lynch et al., 1999), and is dependent on a sustained influx of extracellular Ca²⁺. It has been proposed that LTD₄ primarily induces bronchoconstriction in airways via IP₃-mediated release from Ca²⁺ stores. However, removal of external Ca²⁺ has been shown to substantially reduce or abolish ASM contraction (Black et al., 1986; Matran et al., 1988). We also found that removal of external Ca²⁺ abolished LTD₄-induced bronchoconstriction (Figures 1b and 2), as well as the responses to histamine and carbachol. The LTD₄-induced increase of Ca²⁺-dependent K⁺ channels activity, which can be used to assess subplasmalemmal concentration of free Ca²⁺, was also abolished in Ca²⁺-free solution.

The above results do not necessarily rule out a role for IP₃ generation and Ca²⁺ release, as PI-PLC is Ca²⁺-sensitive, and IP₃ formation has been reported to be dependent on external Ca²⁺ (Dumitriu et al., 1997). Indeed, we have found that at very high concentrations (>50 nM) LTD₄ could induce Ca²⁺ release (Figure 9). However, neither inhibition of PI-PLC with U73,122 or ET-18-OCH₃ nor application of the IP₃ receptor antagonist 2-APB had any effect on bronchoconstriction induced by more physiologically relevant concentrations of LTD₄ (10 nM), which was in any case sufficient to induce maximum bronchoconstriction (Figure 1) in these small bronchioles. This is consistent with one other report on guinea-pig trachea, that suggested that Ca²⁺ release from stores makes little direct contribution to LTD₄-induced contraction of ASM (Cuthbert et al., 1994). Indeed, even though caffeine (5 mM) induced a massive release of Ca²⁺ from intracellular stores followed by a transient inward Cl⁻ current (Figures 7 and 9) and a longer lasting increase in Ca²⁺-dependent outward current, it did not cause any contraction in human small bronchioles. Moreover, it has been reported that in rat neonatal cardiac myocytes IP₃ generation results rather from the rise of intracellular Ca²⁺ than from the cleavage of PIP₂ (Matkovich & Woodcock, 2000). Wang & Kotlikoff (2000) have also reported that although U73,122 abolished histamine- or methacholine-induced Cl⁻ and cationic currents in equine tracheal myocytes, simultaneous application of caffeine with these agonists restored the currents in the presence of U73,122. Thus it seems unlikely that Ca²⁺ release from IP₃-sensitive stores is directly involved in activation of the contractile machinery.

Our results provide good evidence that the primary action of LTD₄ is to induce Ca²⁺ influx across the cell membrane. In many types of smooth muscle voltage-operated Ca²⁺ channels (VOC) are clearly involved in Ca²⁺ entry, but although influx of Ca²⁺ via VOCs has been implicated in guinea-pig ASM, their role in human ASM remains controversial, especially in relation to the LTD₄ response. The VOC agonist Bay K 8644 did not significantly modify the EC₅₀ of acetylcholine or histamine on human main bronchus, but potentiated the effects of KCl on that preparation (Advenier et al., 1986). In human bronchial muscle preparations preincubated for 30 min with 10 μM diltiazem, the LTD₄ cumulative concentration – effect curves following these drug treatments were similar to controls (Bourdillat et al., 1987). In human bronchial rings (2 – 5 mm) nifedipine had no effect on the LTD₄ dose – response curve or on the maximal LTD₄ contraction (Gorenne et al., 1998). On the other hand, Kohrogi et al. (1985) have reported that nifedipine decreased or reversed LTD₄-induced contraction of human bronchial strips. We found that in human small bronchioles nifedipine could only partially block or reverse LTD₄-induced contraction (Figure 3a). The possible involvement of T-type channels, which are highly sensitive to Ni²⁺, seems unlikely, as 1 mM Ni²⁺ did not affect contraction.

An alternative path for sustained extracellular Ca²⁺ entry is through Ca²⁺-permeable non-selective cation channels (NSCC) as proposed for muscarinic stimulation of equine ASM (Fleischmann et al., 1997), histamine in cultured human ASM (Murray & Kotlikoff, 1991), and specifically LTD₄ in guinea-pig and human ASM (Gorenne et al., 1998; Cuthbert et al., 1994). These channels could be blocked with some selectivity with La³⁺ and SK&F 96,365. Indeed we have found that La³⁺, Gd³⁺ and SK&F 96,365 abolished LTD₄-induced constriction in human bronchioles (Figure 3b,c), and MDL-12,330A, which has been shown to block NSCC and SOC in variety of cell types (Van rossum et al., 2000), was also an efficient inhibitor. These findings are consistent with the observation that SK&F 96,365 significantly reduced the nifedipine-resistant response to LTD₄ in human bronchial rings (Gorenne et al., 1998), and together lend support to the hypothesis that LTD₄-mediated Ca²⁺ entry is via NSCCs. We have previously characterized NSCCs in ASM from human foetal trachea and adult bronchioles (Snetkov et al., 1998; Snetkov & Ward, 1999), which have very similar properties to the channels activated by LTD₄ and shown in Figure 8a.

The only molecular entities presently characterized functionally as channels responsible for receptor-mediated Ca²⁺ influx in vertebrate and invertebrate cells are transient receptor potential (TRP) protein homologues. Of the seven mammalian TRP homologues currently cloned, TRP3, TRP6 and TRP7 have been reported to be activated by DAG (Kiselyov & Muallem, 1999). The fact that D609 abolished LTD₄-induced constriction in these small bronchioles could imply that LTD₄ stimulates PC-PLC mediated production of DAG, and thus activation of plasmalemmal TRP channels, inducing influx of extracellular Ca²⁺. Although non-specific effects of D609 cannot be entirely ruled out, it should be noted that we found its EC₅₀ for inhibition of LTD₄-induced bronchoconstriction to be ≈5 μM, which is equivalent to the K_i for PC-PLC inhibition in vitro (5 – 10 μM). As PKC antagonists were without effect (Figure 5), it seems reasonable to assume that any stimulation of PKC by DAG plays little part in LTD₄-induced bronchoconstriction.

Spontaneous activity of TRP3 and TRP7 channels in transfected cells has been reported to give rise to a background non-selective current and Ca²⁺ permeability (Zhu et al., 1998; Okada et al., 1998). Such activity could underlie the Ca²⁺-dependent but nifedipine resistant basal tone we have observed in human small bronchioles (Figure 6). It is notable that in contrast to the study on human bronchial strips of Fujiwara et al. (1993), the basal tone in these small bronchioles was apparently not related to basal release of leukotrienes, as it was not affected by ICI 198,615.

The fact that emptying of intracellular stores with CPA does not produce contraction per se, but apparently activates subsequent Ca²⁺ influx, may be interpreted as evidence for store-activated rather than second messenger-activated channels. Specifically, TRP4 channels have been suggested as a part of a native Ca²⁺ release activated channels in adrenal cells (Philipp et al., 2000). However, Schaefer et al. (2000) have reported that the properties of murine TRP4 suit this role considerably less well than reported for its bovine counterpart, and in vascular smooth muscle TRP1 has been suggested to fulfil a similar role (Xu & Beech, 2001). While the effect of CPA shows that SOC may well be present in human small bronchioles, the lack of any significant effect of low micromolar concentrations of trivalent cations or 2-APB on the LTD₄-induced bronchoconstriction, both of which are reported to block SOC, tends to suggest that SOC is not the major Ca²⁺ entry pathway during LTD₄ stimulation of human small bronchioles. Moreover, the ability of the PC-PLC antagonist D609 to block the LTD₄ response implies involvement of second-messenger operated channels such as TRP3 or TRP6.

Recent data suggest that there might be a direct molecular interaction between components of the intracellular stores (such as the IP₃ receptor itself) and plasmalemmal TRP channels (see Putney & Mckay, 1999). However, there is the possibility that Ca²⁺ released by CPA could activate phospholipid hydrolysis and produce second messenger(s) (Matkovich & Woodcock, 2000). It is very probable that more than one type of TRP channel is expressed in ASM. Indeed, both TRP6 and TRP7 have been found in lung (Boulay et al., 1997; Moore et al., 1998; Okada et al., 1999), while TRP1 has been shown to regulate endothelial permeability in the pulmonary vasculature (Moore et al., 1998).

In our experiments the PTK inhibitor tyrphostin A23 significantly suppressed LTD₄-induced contraction. This is in agreement with previous findings that PTK inhibition reduces carbachol and serotonin-induced constriction of rat airways, particularly in small bronchioles (Chopra et al., 1997), and suppresses LTD₄-induced bronchoconstriction in guinea-pig (Wong et al., 1997). Gronroos et al. (1995) have also demonstrated that tyrosine phosphorylation of PLC as well as other downstream targets modulates LTD₄-mediated Ca²⁺ influx in human epithelial cells, and in human fibroblasts PTK inhibition blocked Ca²⁺ influx induced by bradykinin and thapsigargin (Lee et al., 1993). Taken together, these data allow us to consider a possible role for tyrosine phosphorylation in mediating Ca²⁺ influx in human bronchioles.

In conclusion, we have found that the constrictor effect of LTD₄ in human small bronchioles is mediated by CysLT₁ receptors, requires Ca²⁺ entry across the sarcolemma, and seems to be independent of PI-PLC and PKC. Ca²⁺ released from intracellular stores is unable to induce contraction directly, though store emptying could be inducing Ca²⁺ influx through TRP channels. It would appear that PC-PLC plays an important role in LTD₄-induced bronchoconstriction in this tissue. This apparently does not involve activation of PKC by DAG, but we can speculate that DAG generated by this pathway is activating TRP channels and therefore Ca²⁺ influx.

Abbreviations

2-APB

2-aminoethoxydiphenyl borate

ASM

airway smooth muscle

DAG

diacylglycerols

IP₃

inositol 1,4,5-trisphosphate

KPSS

physiological salt solution containing 80 mM KCl (equimolar substitution for NaCl)

LTD₄

leukotriene D₄

NSCC

non-selective cation channels

PC-PLC

phosphatidylcholine phospholipase C

PI-PLC

phosphatidylinositol phospholipase C

PKC

protein kinase C

PLD

phospholipase D

PSS

physiological salt solution

SOC

store-operated channels

VOC

voltage-operated Ca²⁺ channels