Published ahead of print on September 18, 2003, doi:10.1165/rcmb.2003-0212OC
© 2004 American Thoracic Society DOI: 10.1165/rcmb.2003-0212OC Enhanced Myosin Phosphatase and Ca2+-Uptake Mediate Adrenergic Relaxation of Airway Smooth MuscleAsthma Research Group, Firestone Institute for Respiratory Health, St. Joseph's Hospital, and the Department of Medicine, McMaster University, Hamilton, Ontario, Canada Address correspondence to: Dr. Luke J. Janssen, L-314, St. Joseph's Hospital, 40 Charlton Ave. East, Hamilton, ON, L8N 4A6 Canada. E-mail: janssenl{at}mcmaster.ca
We examined the mechanisms underlying relaxations evoked by isoproterenol (Iso) in isolated porcine, bovine, or human tracheal and bronchial tissues (TSM and BSM, respectively). Iso had little effect against contractions evoked by high KCl, indicating that it does not directly suppress voltage-dependent Ca2+-influx nor directly inhibit myosin light chain kinase. Furthermore, Iso was equally potent against carbachol (CCh) contractions in the presence versus absence of nifedipine (10-6 M), establishing that the primary action of Iso is not through membrane hyperpolarization. However, Iso relaxations in porcine/bovine BSM were significantly suppressed by inhibitors of the internal Ca2+ pump (cyclopiazonic acid; 10-5 M) or of myosin light chain phosphatase (calyculin; 10-6 M). Myosin light chain phosphatase activity was assayed directly (using 32P-labeled myosin) and found to be enhanced in a time- and concentration-dependent fashion by Iso. Iso relaxations in human airway tissues, on the other hand, were not significantly affected by either calyculin or cyclopiazonic acid. Thus, we conclude that Iso acts largely in a voltage-independent fashion: in nonhuman airways, this involves enhanced Ca2+ pump activity (to decrease [Ca2+]i) and myosin light chain phosphatase activation (to decrease Ca2+-sensitivity of the contractile apparatus), whereas in human airways the underlying mechanisms are still unclear.
Abbreviations: 11-2[[2-(diethylamino)methyl1]-1-piperidinyl] acetyl-5,11-dihydro-6H-pyrido-[2,3-b]-benzodiazepine-6-one, AF-DX 116 • airway smooth muscle, ASM • bronchial smooth muscle, BSM • cyclopiazonic acid, CPA • N-
In the airways of all species, the primary excitatory innervation is cholinergic in nature, acting through muscarinic receptors of the M2 subtype (coupled through inhibition of adenylate cyclase) and the M3 subtype (coupled through the phosphoinositide pathway). In many mammalian species, adrenergic fibers provide the primary inhibitory neural input, signaling through ß-adrenoceptors on the airway smooth muscle (ASM). In the human airways, however, the adrenergic innervation is quite sparse, although the ß-adrenoceptors are present nonetheless and are activated by circulating epinephrine. Depending on species, there may also be an inhibitory nitrergic innervation. There is still much that we do not know regarding the signaling pathways underlying the responses to adrenergic and nitrergic stimulation in ASM. It is often promoted that they act through opening of K+ channels and subsequent cessation of voltage-dependent Ca2+-influx, perhaps because of the importance of this pathway in other muscle preparations. However, there is now extensive evidence that voltage-dependent Ca2+-influx does not play a central role in mediating bronchoconstrictor responses, and that K+ channel activation is not a central mechanism in the responses to ß-agonists (14). Needless to say, then, bronchodilators must be acting primarily in some other fashion. In general, contraction in smooth muscle is determined by the net level of phosphorylation of myosin, which in turn depends on the relative activities of myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP). Thus, many constrictors act through two signaling pathways: (i) release of internally sequestered Ca2+, which in turn enhances MLCK activity; and (ii) suppression of MLCP activity via activation of the monomeric G-protein Rho and its down-stream effector Rho-activated kinase (ROCK). Relaxants may therefore act by reversing one or both of these pathways; however, the interactions between bronchodilators and the activities of either MLCP or of Rho/ROCK are largely unstudied. A better understanding of these questions may lead to the development of entirely novel strategies for controlling asthma. We therefore set out in this study to examine the contributions of various signaling pathways, particularly that of MLCP, to relaxations in ASM. We used both tracheal and bronchial preparations, as there is evidence that these do not necessarily respond in like fashion to a variety of stimuli (see DISCUSSION). Moreover, we used preparations from bovine, porcine, and human airways to broaden the relevance (both general and clinical) of these findings, as there are frequently species-related differences in the pharmacologic and physiologic responses of ASM. We used standard muscle bath techniques to monitor mechanical activity, and biochemical techniques (Western blot and myosin dephosphorylation assay) to assess enzymatic activities.
Preparation of Isolated Tissues and Single Cells Lobes of lung and tracheae were obtained from pigs (20–90 kg) and cows (136–454 kg) killed at a local abattoir, and immediately put in ice-cold physiologic solution for transport to the laboratory. TSM was isolated by removing connective tissue, vasculature, and epithelium, then cut into strips parallel to the muscle fibers (  1 mm wide). Lobes of lung were pinned out, the overlying parenchyma and pulmonary vasculature were removed, and ring segments (4–5 mm long) of bronchi (outer diameter 2–4 mm) were excised. Portions of human lung that had been resected at St. Joseph's Healthcare (Hamilton, ON, Canada) and which had been judged by the pathologist to be macroscopically normal were also obtained (n = 20). From these, small order airways (outer diameter 0.5–1 mm) were carefully removed and cut into ring segments 4–5 mm long.
Muscle Bath Technique
Preparation of Phosphorylated Myosin Porcine stomachs were trimmed of fat and connective tissue, then minced and homogenized. From these homogenates, we extracted myosin and myosin light chain kinase using protocols published elsewhere (12). These were stored at -70°C until needed for MLCP assay, at which point myosin light chain (1 mg/ml) was incubated together with MLCK (20 µg/ml) and 32P-ATP (1 mM; 0.5 mCi) for 30 min at 25°C in phosphorylation buffer (50 mM Tris-HCl, 20 mM NaCl, 1 mM MgCl2, 0.2 mM CaCl2, 0.3 µM calmodulin, 1 µg/ml leupeptin, pH 7.4). Reactants were dialyzed eight times overnight against 30 mM Tris-HCl (pH 7.5), 50 mM KCl, 0.5 mM DTT to remove ATP.
Phosphatase Assay
Solutions and Chemicals All chemicals were obtained from Sigma-Aldrich Canada, Ltd. (Oakville, ON, Canada) with the exception of calyculin and chlorpheniramine (Tocris, Ellisville, MO). These were prepared as 10-mM stock solutions, either as aqueous solutions or in absolute EtOH (nifedipine, [+]-[R]-trans-4-[1-aminoethyl]-N-[pyridyl] cyclohexanecarboxamide dihydrochloride [Y27632], chlorpheniramine) or DMSO (cyclopiazonic acid, 11–2[{2-(diethylamino)methyl1}-1-piperidinyl] acetyl-5,11-dihydro-6H-pyrido-[2,3-b]-benzodiazepine-6-one [AFDX-116], or calyculin). Aliquots were then added to the muscle baths; the final bath concentration of DMSO and EtOH did not exceed 0.1%, which we have found elsewhere to have little or no effect on mechanical activity.
Data Analysis
Effect of Iso and CPA on KCl-Evoked Contractions We first investigated the effects of Iso in tissues preconstricted with KCl, which is presumed to evoke contractions largely via opening of voltage-dependent Ca2+ channels and subsequent activation of MLCK. Porcine TSM and BSM tissues were pretreated with atropine (10-6 M; 20 min) to prevent constriction evoked by KCl-induced release of acetylcholine, then preconstricted with either 20 or 65 mM KCl, and challenged with cumulatively increasing concentrations of Iso (Figure 1). Iso completely reversed 20 mM KCl-tone with a log IC50 of –8.1 ± 0.1 and –8.0 ± 0.1 in porcine TSM and BSM, respectively. Tone evoked by 65 mM was much less sensitive to Iso, being reduced less than 25% (hence, we did not calculate IC50 values under these conditions). In a separate set of experiments, we challenged porcine TSM and BSM tissues with increasing concentrations of KCl in the presence or absence of Iso (10-6 M). As bronchodilators may stimulate SERCA activity, we also performed these experiments in the presence or absence of cyclopiazonic acid (CPA): 10-5 M CPA has been shown elsewhere to be maximally effective in this respect (4, 5, 18). CPA alone markedly and significantly augmented the responses to submaximal concentrations of KCl (20 and 35 mM), particularly in the TSM tissues, but had no significant effect on the maximal KCl response (Figure 2), an effect which we have shown previously to be due to abolition of the superficial buffer barrier function of the sarcoplasmic reticulum (15). Iso tended to reduce the magnitudes of KCl-evoked contractions in both the TSM and BSM, although this effect did not reach statistical significance (Figure 2, right panel). High KCl causes contraction simply by raising [Ca2+]i (through voltage-dependent Ca2+-influx) and thereby stimulating MLCK. As such, our observation that Iso could completely reverse the small contractions evoked by 20 mM KCl can be interpreted either that Iso: (i) inhibits voltage-dependent Ca2+-influx (tested below using nifedipine); (ii) reverses the rise in [Ca2+]i (tested below using cyclopiazonic acid); (iii) directly inhibits MLCK; or (iv) increases MLCP activity (tested below using a radiometric phosphatase assay).
Role of Electromechanical Coupling in Iso Relaxations Figure 3 shows the Iso concentration-response relationship is negatively affected by increasing levels of cholinergic stimulation in both TSM and BSM, being shifted upward and to the right with increasing concentration of CCh. This functional antagonism is in part owing to the action of CCh through M2-receptors (which couple via Gi) (16): consistent with this, pretreatment of the tissues with the M2-receptor blocker AFDX-116 (10-6 M) caused a marked and significant leftward shift of the Iso concentration–response relationships. More importantly, however, we found that nifedipine had no significant effect whatsoever on the Iso concentration–response relationship (neither efficacy nor potency) in either TSM or BSM (Figure 3).
Role of SERCA in Iso Relaxations during Cholinergic Stimulation In TSM tissues exposed to CPA (10-5 M), Iso still caused a substantial and concentration-dependent suppression of tonic cholinergic activity (Figure 4E). The mean magnitudes of these relaxations were not significantly different from those seen in tissues not pretreated with CPA, except at the very highest concentration of Iso tested (Figure 4E). Moreover, these relaxations were significantly greater than the small and gradual decay in tone seen in time-control tissues (those constricted for 60–90 min with CCh after pretreating with CPA, but not challenged with Iso; bars in Figure 4E). The phasic activity, on the other hand, continued with no apparent sensitivity to Iso; the dotted line in Figure 4E shows the maximal magnitude of the phasic relaxations (labeled "CPA (troughs)") at the different concentrations of Iso applied. As already noted above, CPA did not evoke phasic activity when the tissues were preconstricted with both CCh and KCl. More importantly, though, the mean magnitudes of Iso-evoked tonic relaxations under this experimental condition were not significantly different from those seen in the presence of CPA and CCh alone (Figure 4E). In BSM tissues, however, Iso-evoked tonic relaxations were markedly and significantly reduced by pretreatment with CPA (10-5 M), although oscillatory phasic activity continued (Figure 4E); the small reversal of tone which was measured in Iso-challenged tissues was not significantly different in magnitude to the time-dependent decay in tone seen in the time-control tissues (bars in Figure 4E), indicating that the former was not due to adrenergic stimulation. Concurrent exposure to 65 mM KCl had no additional effect on the tonic component of relaxations, but abolished the phasic relaxations (Figure 4E).
Role of MLCP in Iso Relaxations during Cholinergic Stimulation We next examined the role of MLCP directly using a radiometric phosphatase assay (see MATERIALS AND METHODS). In bovine tissues exposed to Iso (10-6 M) alone for various periods of time before being flash-frozen and homogenized, Iso caused a dramatic time- (Figure 6A) and concentration (Figure 6C and 6D)-dependent increase in 32P-release. Although CCh (10-6 M) alone for 20 min markedly reduced the basal level of 32P-release (Figure 6B), this too was markedly increased in a concentration-dependent fashion by Iso (10-6 M; Figure 6D). The Iso-induced elevation in MLCP activity in CCh-stimulated tissues was further enhanced by pretreating the tissues for 20 min with the M2-selective muscarinic inhibitor AF-DX 116 (10-5 M) or the phosphodiesterase inhibitor isobutylmethylxanthine (10-4 M), but not when tissues were pretreated for 20 min with calyculin (10-6 M) or the nonspecific ß-receptor blocker propranolol (10-6 M) (Figure 6E).
Human Data We repeated several of the key experiments outlined above in human main stem bronchi ("TSM") and small order bronchi ("BSM"; outer diameter < 2 mm). Unfortunately, these experiments were complicated by the development of spontaneous tone, which sometimes hindered further contraction by cholinergic agonist. In one set of human TSM or BSM tissues bathed in Krebs containing L-NNA (10-4 M) and indomethacin (10-5 M), application of nifedipine (10-6 M) significantly reduced total tone (both cholinergic plus spontaneous) by 44 ± 28% in human TSM (n = 5) and by 13 ± 10% in human BSM (n = 3), but had no significant effect on the Iso concentration–response relationship compared with vehicle control (Figure 7). Another set of human BSM tissues were bathed in Krebs containing L-NNA (10-4 M), indomethacin (10-5 M), the antihistamine chlorpheniramine (10-6 M), plus the leukotriene receptor blocker MK-571 (10-6 M). Basal tone under these conditions was not significantly altered by subsequent addition of calyculin (10-7; mean increase of 4 ± 6%; n = 8), but was increased further by CPA (10-5 M; mean increase of 43 ± 20%; n = 10) or CPA plus calyculin (mean increase of 69 ± 16%; n = 6); vehicle for these agents (0.1% DMSO) had no significant effect on basal tone (mean decrease of 6 ± 4%; n = 11). More importantly, however, the Iso concentration–relaxation relationship was unaltered by any of these three experimental conditions (Figure 7).
A great deal of study has been directed at elucidating the mechanisms underlying contractions in ASM; however, the mechanisms responsible for relaxations have received relatively less attention. The majority of those studies which have been done have used tracheal preparations, usually nonhuman, despite the increasing awareness that there can be substantial regional and species-related differences with respect to airway function, and that the 4th to 6th order bronchi are primarily responsible for determining peripheral resistance to airflow. Also, many previous studies of relaxant mechanisms in ASM have focused in one way or another on the role of potassium channels. In this study, we performed a systematic study of several different relaxant signaling pathways in porcine and bovine trachea and bronchi, following this up with specific studies using human airway preparations. Many continue to promote the concept that opening of K+ channels plays some key role in ASM relaxation, by causing membrane hyperpolarization and subsequent closure of voltage-dependent Ca2+-channels. Data used to support this claim include the observations that: (i) relaxations are reduced in magnitude by high millimolar KCl (a condition which compromises agonist-mediated membrane hyperpolarization by elevating the K+ equilibrium potential), as we found to be the case in the present study; (ii) relaxations can be suppressed by K+ channel blockers (24); (iii) K+ currents can be directly activated by ß-agonists (22, 23); and (iv) agents which enhance K+ channel activity can elicit a relaxant response (3, 24). However, we found the Iso concentration–relaxation relationships in porcine and human TSM and BSM to be unaffected by pretreatment with the voltage-dependent Ca2+-channel blocker nifedipine, despite the substantial ability of nifedipine to block Ca2+-currents and inhibit depolarization-evoked contractions. It is difficult to reconcile this finding with any interpretation involving a central role for K+ channel opening in relaxation: consistent with this, many clinical studies have not found a beneficial effect of either Ca2+-channel blockers (2, 7, 8, 11, 26, 28) or of K+ channel openers (6, 21) in asthma. It may be that KCl-evoked contractions are reversed only partially by Iso simply because these two agents regulate the contractile apparatus through nonoverlapping pathways: KCl acts almost exclusively in an electromechanical fashion to increase MLCK activity, whereas Iso acts through some other signaling pathway(s) which is not under regulation by high millimolar KCl (see below). We next examined the role of SERCA in mediating airway relaxation. This was complicated by the fact that disruption of SERCA activity (using CPA, thapsigargin, or ryanodine) markedly alters the nature of cholinergic contraction by removing a major excitatory component of excitation-contraction coupling—that is, the release of internally sequestered Ca2+ (19, 20)—leaving tone entirely sensitive to the relatively smaller contributions of voltage-dependent Ca2+-influx (1, 15, 27) and of the Rho/ROCK signaling pathway (20). This could explain why tone fluctuates erratically in the presence of CPA: the phasic activity reflects the rapid changes in [Ca2+]i due to opening and closing of K+ and Ca2+ channels, whereas sustained tone reflects the slower changes in Ca2+-sensitivity. More importantly, though, we found Iso-evoked relaxations were similar in magnitude between CPA-treated and -untreated porcine TSM as well as human ASM (both large and small airways), which suggests that the major effect of Iso is not to stimulate the internal Ca2+-pump. Interestingly, however, there are important tissue- and species-related differences, because Iso appeared to be unable to evoke statistically significant relaxation in porcine BSM pretreated with CPA. The relaxations evoked by Iso in the presence of both CPA and high KCl clearly involve mechanisms other than stimulation of SERCA or of K+ channels. Contraction is triggered by phosphorylation of myosin, and this is reversed by MLCP. Thus, we explored the regulation of MLCP activity by Iso in these tissues, finding: (i) Iso directly stimulated MLCP activity in resting and carbachol-stimulated tissues in a time- and concentration-dependent fashion; and (ii) the phosphatase inhibitor calyculin markedly and significantly suppressed (but did not abolish) Iso-evoked relaxations and Iso-stimulated MLCP activity. Again, there were important tissue- and species-related differences in the sensitivity of Iso-relaxations to calyculin, with marked sensitivity being noted in porcine and bovine BSM but not in the TSM of these animals nor in the human airway preparations (large and small airways). The regulation of MLCP, particularly its activation by relaxants, is very poorly understood in ASM. We found the adrenergic effect on MLCP activity was enhanced by IBMX (which would prevent degradation of cAMP) or by the M2-selective muscarinic inhibitor AF-DX 116 (which would remove Gi-mediated functional antagonism of adenylate cyclase activity), suggesting that cAMP is involved. It is unclear from these data whether cAMP interacts directly with MCLP or acts indirectly through cAMP-dependent protein kinase. Given the importance of the monomeric G-protein Rho and its downstream effector ROCK in cholinergic tone, it is possible that Iso acts in some way by suppressing Rho and/or ROCK activities: in fact, cAMP/PKA have indeed been shown in vascular smooth muscle to inhibit ROCK activity (10). In conclusion, our data argue against any major role for K+ channels in adrenergic relaxation of ASM (except under the nonphysiologic condition of complete sarcoplasmic reticulum depletion). Instead, activation of MLCP and SERCA are much more important in this respect in porcine and bovine ASM tissues. The mechanisms coupling adrenergic stimulation to relaxation in human ASM, on the other hand, are still unclear. Finally, our data highlight important tissue-related and species-related differences in function, indicating the limitations of studies done using nonhuman TSM tissues.
The authors thank Dr. A. Yoshimura (Welfide Corporation, Osaka, Japan) for the kind gift of Y-27632. These studies were supported by an Investigator Career Award from the Canadian Institutes of Health Research, as well as operating support kindly provided by the Canadian Institutes of Health Research, the Ontario Thoracic Society of Canada, and AstraZeneca. Received in original form June 7, 2003 Received in final form September 6, 2003
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
-nitro-L-arginine, L-NNA • myosin light chain kinase, MLCK • myosin light chain phosphatase, MLCP • Rho-activated kinase, ROCK • tracheal smooth muscle, TSM • (+)-(R)-trans-4-(1-aminoethyl)-N-(pyridyl) cyclohexanecarboxamide dihydrochloride, Y27632
