INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although -adrenergic receptor agonists (-agonists) are widely used clinically as bronchodilator agents of first choice, considerable concern has been raised regarding the regular use of these agents in patients with bronchial asthma. Regular administration of inhaled -agonists may cause not only a deterioration of asthma control and an exacerbation of airway hyperreactivity (1), but also may accelerate a decline in lung function in patients with asthma (2). Ohta and coworkers have reported that many cases initially treated with repeated inhalation of -agonists every 15 min were unable to recover from their acute exacerbation of bronchial asthma and required additional treatment (3). Because a reduced bronchodilator response to a -agonist is observed in patients with asthma (4) and animal models of this disease (5, 6), dysfunction of -adrenergic receptors is a characteristic feature of asthma. The undesirable dysfunction is generally considered to be due to reduced responsiveness to inhaled -agonists induced by regular use over a long time or repeated administration of these agonists over a short time, a phenomenon referred to as desensitization. Recent reports have indicated that various proinflammatory cytokines, which participate in inflammatory responses in bronchial asthma, attenuate the response to -agonists in airway smooth muscle (7), suggesting that reduced responsiveness also occurs with -adrenergic receptors in patients with asthma untreated with -agonists. -Adrenergic desensitization may be a very important problem in both the pathogenesis and therapy of bronchial asthma.

Because a decrease in the density and affinity of -adrenergic receptors does not cause a reduction in response to -agonists in airway smooth muscle (11), signal transduction processes coupled with -adrenergic receptors may play an important role in the regulation of the mechanical response to -agonists. A recent report has shown that preincubation with cholera toxin (CTX) suppressed the subsequent reduction in -adrenergic relaxation after continuous and repeated exposure to a -agonist in guinea pig tracheal smooth muscle (12). These results reveal that since CTX irreversibly activates the -adrenergic receptor-coupled guanosine triphosphate (GTP)-binding protein (G_s), prevention of the -adrenergic desensitization may be mediated by G_s activation. The -subunit of G_s activates an intracellular cAMP-dependent protein kinase (PKA), which augments the probability that large conductance Ca²⁺-activated K⁺ (K_Ca) channels (Maxi-K⁺ channels) distributed densely on the surface of the cell membrane of airway smooth muscle are open (13, 14), whereas it activates these channels directly at the membrane boundary (15). As is well known, many -agonists cause relaxation of airway smooth muscle accompanied by activation of these channels (16). These observations suggest that K_Ca channel activity regulated by G_s is involved in subsequent reduction in -adrenergic relaxation.

This study was designed to examine whether reduced -adrenergic relaxation occurs in human tracheal smooth muscle after continuous and repeated exposure to -agonists. Additionally, we investigated the role of postreceptor-coupled signal transduction processes (G_s-K_Ca channel stimulatory linkage) as the mechanisms underlying this -adrenergic desensitization.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tissue Preparation and Tension Records

The methods were essentially similar to those described previously (17). Human tracheal tissue was obtained from 22 patients (56 to 79 yr old) at autopsy. None of the patients had a respiratory disease and received therapy for bronchodilation using -adrenergic receptor agonists. The tracheas were dissected within 2 h postmortem, and one cartilaginous length was taken from the middle portion of the trachea by the hospital pathologists. At the hospital the tissue was immediately placed in ice-cooled Krebs solution and then transported to the laboratory. The tracheal ring was opened by cutting longitudinally through the cartilaginous region. The epithelium was dissected away, and a muscle strip of 2-mm width and 5-mm length was removed. The strips were placed vertically in a 1-ml organ bath, and were perfused with solution at a constant flow rate of 2.0 ml/min throughout the experiments. The normal bathing solution had the following composition (mM): 137 NaCl, 5.9 KHCO₃, 2.4 CaCl₂, 1.2 MgCl₂, and 11.8 glucose, bubbled with a gas mixture of 99% O₂-1% CO₂. For the Ca²⁺-free solution, 2.4 mM CaCl₂ was replaced with 2.2 mM NaCl and 0.2 mM ethyleneglycol-bis-(-aminoethyl ether)-N,N',-tetraacetic acid (EGTA). One side of the tissue was connected to a hook at the bottom of the organ bath, and the other was connected to a strain gauge to measure tension isometrically. The passive tension was adjusted to 0.5 g and after equilibrating the preparation in a normal bathing solution for 1 h, 1 µM methacholine (MCh) was applied to the strips for 10 min at intervals of 20 min until the control response to 1 µM MCh was established, and then the experiments were started. The relaxation induced by exposure to the Ca²⁺-free solution was defined as complete relaxation (0% contraction). The Ca²⁺-free solution was applied at the end of each experiment to determine the level of 0% contraction. All experiments were carried out at 37° C.

Experimental Protocols

To examine the effects of activated G_s via adenosine diphosphate (ADP) ribosylation of the -subunit of G_s protein, the tissues were incubated with 2 µg/ml cholera toxin (CTX) for 6 h and then the CTX was washed out. To examine -adrenergic desensitization after continuous exposure to a -agonist, the tissues were exposed to isoproterenol (ISO) for 45 min and then the ISO was washed out. MCh- induced contraction and inhibitory effects of ISO on MCh-induced contraction were recorded under each experimental condition before and after the tissues were treated with ISO and CTX. The normal bathing solution was applied to the strips for 15 min before and after incubation with ISO and CTX for washout. Because MCh-induced contraction was reduced by exposure to ISO for 45 min, relaxant effects of ISO after MCh-induced contraction returned to the control response were expressed as the subsequent relaxant effects of ISO. On the other hand, since after exposure to CTX MCh-induced contraction did not return to the control response and ISO-induced relaxation did not alter after repeated application, the first ISO response after exposure to CTX was expressed as the subsequent ISO response. To examine the effects of -agonists after repeated exposure to -agonists, ISO was applied to the tissues for 10 min every 30 min in the presence of 1 µM MCh. To examine the effects of -agonists after continuous and repeated exposure to -agonists, concentration-inhibition curves for ISO were plotted twice with an interval of 15 min for each experimental condition. The involvement of K_Ca channels was examined by application of iberiotoxin (IbTX), a potent selective K_Ca channel inhibitor. The effect of relaxant agents on MCh-induced contraction was expressed as percent contraction by taking the control response to 1 µM MCh as 100%. Time-matched control tissues were treated similarly to the test tissues, but exposed continuously to the normal bathing solution (sham incubation) instead of ISO, CTX, forskolin, prostaglandin E₂ (PGE₂), and N⁶-dibutylyl cyclic AMP (db-cAMP). The subsequent relaxant response to ISO after exposure to these agents is compared with time-matched control tissues.

Materials

MCh, ISO, CTX, EGTA, theophylline, forskolin, histamine, leukotriene D₄ (LTD₄), PGE₂, and db-cAMP were obtained from Sigma Chemical Co. (St. Louis, MO). IbTX was obtained from Peptide Institute Inc. (Osaka, Japan). Rp diasteromer adenosine-3',5' cyclic monophosphothiate (Rp-cAMPS) was obtained from BIOLOG Life Science Institute (Bremen, Germany).

Analysis of Results

All data are expressed as mean ± S.D. The responses to an agent under each condition were described as a percentage of the maximal response. Values of concentrations of relaxant agents that produce 50% inhibition (EC₅₀) of contraction induced by 1 µM MCh were determined using linear regression analysis applied to the linear portion of each concentration-response curve. Parameters were compared using the paired and the unpaired Student's t test, and p values of < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Reduced Responsiveness to a -Agonist after Continuous Exposure to a -Agonist

MCh (1 µM) was applied to the strips, and after a 15-min washout, MCh was again applied in the presence of 0.3 µM ISO. Preexposure to 0.3 µM ISO caused a marked inhibition (16.8 ± 6.3% contraction; n = 10) of the contraction induced by 1 µM MCh (Figure 1A, upper trace). After the tissues were exposed to 3 µM for 45 min and then the ISO was washed out, contraction induced by 1 µM MCh was markedly attenuated and the reduction in MCh-induced contraction returned to the control response roughly 1 h after washout of the ISO (Figure 1B, middle trace). On the other hand, after the continuous exposure to 3 µM ISO, the inhibitory action of 0.3 µM ISO on 1 µM MCh-induced contraction was markedly attenuated (Figure 1A, upper trace) and the reduction in ISO-induced relaxation continued for roughly 4 h (unpublished observation). After the tissues were exposed to the normal bathing solution without ISO for 45 min (sham incubation), the inhibitory action of ISO on MCh-induced contraction was not reduced (Figure 1A, lower trace). The percent contraction values for MCh with ISO inhibition after continuous exposure to the normal bathing solution and 3 µM ISO were 20.9 ± 6.3% (n = 10) and 98.7 ± 8.1% (n = 10), respectively (p < 0.001; Figure 1B). When tissues were exposed to a 10,000-fold lower concentration of ISO than 3 µM for 45 min, the inhibitory action of ISO on MCh-induced contraction was also significantly attenuated (Figure 1B) and the percent contraction values for MCh with ISO inhibition increased to 90.7 ± 8.9% (n = 8; p < 0.001). To examine the role of PKA in the reduced -adrenergic relaxation, the tissues were exposed to 3 and 0.0003 µM ISO for 45 min in the presence of 300 µM Rp-cAMPS, a membrane-permeable PKA inhibitor (Figure 1B). Addition of Rp-cAMPS did not change the effects derived from continuous exposure to ISO, and the percent contraction values for MCh with ISO inhibition after exposure to 3 and 0.0003 µM ISO in the presence of Rp-cAMPS were 95.5 ± 8.9% (n = 6; not significant) and 91.6 ± 10.5% (n = 6; not significant), respectively. After exposure to 300 µM Rp-cAMPS without ISO for 45 min, the inhibitory effect of ISO was not reduced, similar to the subsequent relaxation by ISO after exposure to the normal bathing solution. This result indicates that effects of Rp-cAMPS disappeared completely before addition of subsequent ISO. When concentrations of ISO higher than 0.3 µM were cumulatively applied after exposure to 3 µM ISO for 45 min, ISO caused an inhibition of MCh-induced contraction in a concentration-dependent fashion (Figure 1C). The percent contraction values for MCh induced by 0.3, 1.0, 3.0, and 10 µM ISO were 98.7 ± 8.1% (n = 10), 39.1 ± 10.2% (n = 6), 16.4 ± 8.9% (n = 6), and 6.9 ± 4.1% (n = 6), respectively.

View larger version (20K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 1.   Reduced relaxation in response to a -agonist after continuous exposure to a -agonist. (A) A typical continuous record of contraction induced by 1 µM MCh and relaxant effects of 0.3 µM ISO on this contraction in isolated human tracheal smooth muscle before and after exposure of the strips to 3 µM ISO for 45 min (upper trace). A typical continuous record of contraction by 1 µM MCh before and after exposure to 3 µM ISO for 45 min (middle trace). A typical continuous record of contraction by 1 µM MCh and relaxant effects of 0.3 µM ISO on this contraction before and after exposure to the normal bathing solution (sham incubation) for 45 min (lower trace). (B) Relaxant effects of 0.3 µM ISO on contraction induced by 1 µM MCh after exposure to 3 µM and 0.3 nM ISO in the absence and presence of 300 µM Rp-cAMPS, and after exposure to the normal bathing solution (sham). (C ) Concentration-inhibition curves for ISO on 1 µM MCh-induced contraction after exposure of the tissues to 3 µM ISO for 45 min. The abscissa expresses molar concentration on a log scale. Relaxant effects of ISO on MCh-induced contraction (percent contraction) were expressed by taking the previous MCh-induced contraction for each experimental condition as 100%.

Role of G_s and K_Ca Channels in the Reduced Responsiveness to a -Agonist after Continuous Exposure to a -Agonist

To examine the inhibitory effects of G_s on the reduced responsiveness to a -agonist after continuous exposure to the agonist, the strips were incubated with 2 µg/ml CTX or the normal bathing solution (sham incubation) for 6 h, after which they were exposed to 3 µM ISO for 45 min. MCh (1 µM) and ISO (0.3 µM) were applied in the same way before and after exposure to ISO subsequent to CTX treatment (Figure 2A, upper trace) or sham incubation. Reduced responsiveness to ISO after continuous exposure to ISO occurred in the sham incubation, but this phenomenon did not occur in the CTX incubation. However, the restored inhibitory effect of ISO by CTX was again diminished in the presence of 30 nM IbTX (Figure 2A, upper trace). After exposure to the normal bathing solution without ISO for 45 min subsequent to CTX treatment, the reduction in the ISO-induced relaxation did not occur (Figure 2A, lower trace). The percent contraction values with ISO inhibition after preexposure to CTX and the normal bathing solution (sham incubation) for 6 h were 3.9 ± 2.1% (n = 12) and 95.6 ± 9.2% (n = 10), respectively (p < 0.001; Figure 2B). The percent contraction values increased from 3.9 ± 2.1% to 90.1 ± 9.2% (n = 12; p < 0.001) by addition of IbTX (Figure 2B). To examine the involvement of PKA with CTX incubation, the tissues were incubated with 2 µg/ml CTX in the presence of 300 µM Rp-cAMPS for 6 h. Under these conditions, the application of 0.3 µM ISO also caused a marked inhibition of contraction induced by 1 µM MCh after exposure to 3 µM ISO for 45 min, similar to incubation with CTX without Rp-cAMPS (Figure 2B). The percent contraction values with ISO inhibition were 5.1 ± 3.2% (n = 6).

View larger version (18K): [in this window] [in a new window]   View larger version (24K): [in this window] [in a new window]   Figure 2.   Inhibitory effects of Gs and KCa channels on the -adrenergic desensitization after continuous exposure to -agonists. (A) A typical continuous record of contraction induced by 1 µM MCh and the relaxant effects of 0.3 µM ISO on this contraction before and after exposure to 3 µM ISO (upper trace) and the normal bathing solution (lower trace) for 45 min subsequent to incubation with 2 µg/ml CTX for 6 h, and the relaxant effects of ISO on MCh- induced contraction in the presence of 30 nM IbTX after exposure to CTX and ISO (upper trace). (B) Relaxant effects of 0.3 µM ISO on this contraction after exposure to 3 µM ISO for 45 min subsequent to incubation with 2 µg/ml CTX for 6 h in the absence (CTX) and presence (CTX + Rp-cAMPS) of 300 µM Rp-cAMPS, and with the normal bathing solution for 6 h (sham). IbTX: Relaxant effects of 0.3 µM ISO on this contraction in the presence of 30 nM IbTX after exposure to ISO and CTX. The percent contraction values for each experimental condition are expressed as in Figure 1.

Inhibitory Effects of -Agonists on Contraction by Spasmogens Other than MCh after Continuous Exposure to Agents Other than ISO

Relaxant actions of -agonists on contraction induced by histamine and LTD₄ were examined after incubation with and without -agonists as in Figure 1. Application of 0.3 µM ISO resulted in a marked inhibition (24.5 ± 8.4% contraction; n = 6) of contraction induced by 10 µM histamine. However, the inhibitory action of ISO on histamine-induced contraction was markedly suppressed after exposure to 3 µM ISO for 45 min, as in the case of MCh-induced contraction (Figure 3A). The inhibitory action of ISO was not affected after exposure to the normal bathing solution for 45 min. The percent contraction values for histamine with ISO inhibition after the sham and ISO incubations were 24.5 ± 8.4% (n = 6) and 74.6 ± 6.6% (n = 6), respectively (p < 0.01; Figure 3A). The inhibitory action of 0.3 µM ISO on contraction induced by 3 nM LTD₄ was investigated before and after a 45-min exposure to either the normal bathing solution or 3 µM ISO. The inhibitory action of ISO on LTD₄-induced contraction was markedly attenuated after exposure to ISO, and the percent contraction values for LTD₄ with ISO inhibition after the sham and ISO incubations were 14.8 ± 7.4% (n = 6) and 96.5 ± 8.9% (n = 6), respectively (p < 0.001; Figure 3A). To examine the role of intracellular cAMP in -adrenergic desensitization, the tissues were exposed to agents that elevate the concentration of intracellular cAMP bypassing -adrenergic receptors. After a continuous 45-min exposure to 10 µM forskolin, the direct activator of adenylyl cyclase, 100 nM PGE₂, a G_s activator mediated by PG receptors, and 300 µM db-cAMP, the stable analog of cAMP, the inhibitory action of ISO on MCh-induced contraction was not reduced in comparison with a sham incubation. The percent contraction values for MCh with ISO inhibition after exposure to forskolin, PGE₂, and db-cAMP were 22.6 ± 7.4% (n = 6), 24.5 ± 8.8% (n = 6), and 26.7 ± 9.9% (n = 6), respectively (Figure 3B).

View larger version (14K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   Figure 3.   (A) Relaxant effects of 0.3 µM ISO on contraction by spasmogens other than MCh such as 3 nM LTD4 and 10 µM histamine (Hist) after exposure of the tissues to 3 µM ISO for 45 min. (B) Relaxant effects of 0.3 µM ISO on contraction induced by 1 µM MCh after exposure of the tissues to agents other than ISO such as 10 µM forskolin (Forsk), 100 nM PGE2, and 300 µM db-cAMP for 45 min. The percent contraction values for each experimental condition are expressed as in Figure 1.

Role of G_s and K_Ca Channels in the Reduced Responsiveness to a -Agonist after Repeated Exposure to a -Agonist

ISO (0.3 µM) was applied repeatedly to strips of human tracheal smooth muscle precontracted by 1 µM MCh for 10 min at intervals of 20 min (Figure 4A, upper trace). The first addition of 0.3 µM ISO resulted in a marked inhibition of contraction induced by 1 µM MCh. The percent contraction value for MCh with ISO inhibition was 9.1 ± 4.9% (n = 10; Figure 4B). The inhibitory action of ISO on MCh-induced contraction gradually diminished after repeated application of ISO, and the percent contraction value increased to 72.9 ± 9.9% upon the ninth application (n = 10; Figure 4B). After the tissues were incubated with 2 µg/ml CTX for 6 h, 0.3 µM ISO was applied repeatedly to the pretreated tissues in the presence of 1 µM MCh in the same way (Figure 4A, middle trace). However, by pretreatment with CTX, the inhibitory effect of ISO was not reduced with repeated application of ISO. The percent contraction values with ISO inhibition for the first and the ninth application of ISO were 3.6 ± 2.9% and 3.3 ± 2.2% (n = 10), respectively (Figure 4B). Furthermore, when 0.3 µM ISO was applied repeatedly to strips precontracted by MCh in the presence of 30 nM IbTX after incubation with CTX, 0.3 µM ISO-induced relaxation was decreased gradually again following repeated application (Figure 4A, lower trace). The percent contraction value with ISO inhibition was 41.2 ± 12.9% for the first application, but this value was increased to roughly 100% at the sixth application (Figure 4B). Forskolin (1 µM), theophylline (100 µM), and db-cAMP (300 µM), which elevate the concentration of intracellular cAMP bypassing -adrenergic receptors, were applied to the strips contracted by 1 µM MCh in the same way, but the inhibitory action of these agents did not change with repeated application (Figure 4C).

View larger version (21K): [in this window] [in a new window]   View larger version (20K): [in this window] [in a new window]   View larger version (19K): [in this window] [in a new window]   Figure 4.   Inhibitory effects of Gs and KCa channels on the -adrenergic desensitization after repeated exposure to a -agonist. (A) A typical example of repeated application of 0.3 µM ISO to tissues precontracted by 1 µM MCh at intervals of 20 min under the following experimental conditions: control (upper trace), preincubation with 2 µg/ml CTX for 6 h (middle trace), and preincubation with CTX and in the presence of 30 nM IbTX throughout the experiment (lower trace). (B) Relationship between percent contraction values for MCh with ISO inhibition and the number of ISO applications under the three experimental conditions described (control, open circles; preincubation with CTX, closed circles; preincubation with CTX and application of MCh and ISO in the presence of IbTX, open squares). (C ) Relationship between percent contraction for 1 µM MCh with inhibition by 1 µM forskolin (open circles), 300 µM db-cAMP (closed circles), and 100 µM theophylline (open squares) versus the number of applications of these agents. The percent contraction values for each experimental condition are expressed as in Figure 1.

Role of G_s and K_Ca Channels in the Reduced Responsiveness to a -Agonist after Repeated and Continuous Applications of a -Agonist

ISO (0.003 to 10 µM) was applied cumulatively to tissues precontracted by 1 µM MCh, after which ISO was again applied cumulatively to the identical tissues contracted by MCh after a 15-min washout. Concentration-inhibition curves for ISO were shifted to the right for the second cumulative application (Figure 5A). The EC₅₀ values for the curves for the first and the second applications were 0.11 ± 0.01 and 1.06 ± 0.08 µM (n = 8), respectively (p < 0.01). On the other hand, after the tissues were incubated with 2 µg/ml CTX for 6 h, when ISO was applied cumulatively to the tissues contracted by MCh twice as in Figure 5A, the concentration-inhibition curves for ISO did not differ for the first and the second cumulative applications (Figure 5B). The EC₅₀ values for the curves with the first and second cumulative applications were 0.010 ± 0.004 and 0.012 ± 0.006 µM (n = 8), respectively (not significant). Moreover, after pretreatment with CTX, ISO was cumulatively applied to the tissues contracted by MCh in the absence (first cumulative application) and the presence (second cumulative application) of 30 nM IbTX. The concentration-inhibition curves for ISO were markedly shifted to the right in the presence of IbTX (Figure 5C). The EC₅₀ values in these two cases were 0.011 ± 0.005 and 0.79 ± 0.06 µM (n = 8), respectively (p < 0.001). When in the presence of 30 nM IbTX throughout the experiment ISO was cumulatively applied to the tissues precontracted by 1 µM MCh as in Figure 5A, the concentration-inhibition curves for ISO did not differ for the first and the second cumulative application (Figure 5D). The EC₅₀ values for the curves in these two cases were 1.94 ± 0.79 and 3.09 ± 0.98 µM (n = 6), respectively.

View larger version (17K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   Figure 5.   Involvement of Gs and KCa channels in the prevention of -adrenergic desensitization after continuous and repeated exposure to a -agonist. (A) Concentration-inhibition curves for ISO on precontraction by 1 µM MCh before (open circles) and after (closed circles) sham incubation with the normal bathing solution for 15 min. (B) Concentration-inhibition curves for ISO on precontraction by 1 µM MCh before (open circles) and after (closed circles) incubation with the normal bathing solution for 15 min when the tissues were preincubated with CTX. (C ) Concentration-inhibition curves for ISO on precontraction by 1 µM MCh in the absence (open circles) and the presence (closed circles) of 30 nM IbTX after the tissues were preincubated with CTX. (D) Concentration-inhibition curves for ISO on precontraction by 1 µM MCh before (open circles) and after (closed circles) incubation with the normal bathing solution under the condition of presence of 30 nM IbTX throughout the experiment. The concentration-inhibition curves were generated by taking contraction induced by 1 µM MCh alone as 100% under each experimental condition. The abscissa expresses molar concentration on a log scale.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that a G_s-K_Ca channel stimulatory linkage, a postreceptor-coupled signal-transduction process, is involved in the reduced relaxation in response to a -agonist (-adrenergic desensitization) on isolated human tracheal smooth muscle after exposure to a -agonist. The subsequent reduction in -adrenergic relaxation occurred similarly in human tracheal smooth muscle with the epithelium. To make clear the effects of exposure to -agonists on the human tracheal smooth muscle, we showed data derived from the tissues without the epithelium in this study. Our findings confirm and extend the previous studies showing that relaxant effects of -agonists on muscarinic contraction diminish in isolated human bronchial smooth muscle after continuous exposure to a high concentration (1 to 10 µM) of a -agonist (18, 19). As shown in Figure 1B, ISO has a reduced relaxant effect after continuous exposure of human tracheal smooth muscle to a high concentration (3 µM) of ISO for 45 min, similar to previous observations. Moreover, we investigated whether a reduction in the mechanical response to -agonists occurs when the tissues are continuously exposed to a 10,000-fold lower concentration (0.3 nM) of a -agonist for an equivalent time period. Our results demonstrate that the relaxant effect of a -agonist also diminishes after continuous exposure to a concentration of -agonists much lower than 1 µM (Figure 1B), suggesting that -adrenergic desensitization in airway smooth muscle may occur even though plasma concentrations of -agonist decrease after the administration of an agent. Because the subsequent reduction in the inhibitory effect of ISO on contraction by spasmogens other than muscarinic agonists has never been shown yet, we sought to investigate the relaxant effects of -agonists on contraction induced by 10 µM histamine and 3 nM LTD₄ after continuous exposure to a -agonist in the same way (Figure 3A). Our results demonstrate that the reduction in the relaxant effect of a -agonist occurs after exposure to a -agonist similar to MCh if the tissues are contracted by histamine and LTD₄, which are generally considered to participate in airway contraction during asthma attacks.

The phosphorylation of -adrenergic receptors, which leads to desensitization via uncoupling G_s from the receptors, is mediated by two types of protein kinases, i.e., cAMP-dependent PKA and cAMP-independent protein kinase such as -adrenergic receptor kinase (ARK) (20). PKA-induced phosphorylation, which is produced by exposure to a low concentration of a -agonist, leads to heterologous desensitization (a nonspecific reduced response to other agonists involving cAMP) (21). On the other hand, ARK-induced phosphorylation, which is produced by exposure to a high concentration of a -agonist, leads to homologous desensitization (a specific reduced response to -agonist) (22). Hall and colleagues have shown that intracellular cAMP formation in response to ISO diminishes after incubation not only with ISO but also with forskolin and PGE₂ in cultured human airway smooth muscle, indicating that this phenomenon is mediated by heterologous desensitization (23). Although in this study we did not measure concentrations of intracellular cAMP, it is considered that the intracellular cAMP formation by ISO probably decreases after incubation with ISO, forskolin, PGE₂, and db-cAMP. However, relaxant action of ISO was not affected after continuous exposure to forskolin, PGE₂, and db-cAMP, indicating that this phenomenon is mediated by homologous desensitization (Figure 3B). These results show that with regard to -agonist response, cAMP formation by -agonists is not consistent with relaxation by these agents in human airway smooth muscle, suggesting that -agonists can cause relaxation via cAMP-independent pathways (15, 24). Application of 300 µM Rp-cAMPS, a PKA inhibitor, caused a 12.3 ± 5.9% inhibition of 0.3 µM ISO-induced relaxation (n = 6; unpublished observation). This modest effect of a PKA inhibitor on -adrenergic relaxation is consistent with previous observations that -agonists have cAMP-independent pathways (15, 24, 25), and that the degree of stimulation over basal cAMP is significantly greater with forskolin than -agonists at equivalent levels of relaxation (25, 26). As shown in Figure 1B, subsequent reduction in -adrenergic relaxation also occurred after continuous exposure to both lower and higher concentrations of -agonists even when the tissues were incubated with ISO in the presence of Rp-cAMPS. Hence, the reduced mechanical response to -agonists in human airway smooth muscle is not involved with intracellular cAMP-dependent phosphorylation and is mediated by homologous desensitization. However, this homologous desensitization, which is induced by exposure to a low concentration of a -agonist, can not be explained simply by ARK activity because a high concentration of a -agonist is needed to activate this kinase. Mechanisms underlying the reduced relaxation in response to -agonists are not in agreement with molecular mechanisms previously derived (20).

Preincubation with CTX results in an inhibition of the reduced relaxant effect of -agonists after continuous and repeated exposure to -agonists (Figures 2, 4, and 5), indicating that augmentation of G_s activity may play an important role in preventing the desensitization of -adrenergic relaxation in airway smooth muscle. Because signals can not be transduced into the postreceptor processes even though agonists bind to receptors under the condition that -adrenergic receptors are uncoupled completely from G_s (dissociation), preactivation of G_s by CTX has no effect on -adrenergic relaxation. The subsequent reduction in -adrenergic relaxation shown in this study may be mediated by mechanisms other than uncoupling of G_s derived from molecular biological methods, possibly by a decrease in G_s activity. Moreover, increase in concentrations of -agonists after continuous exposure antagonized the subsequent reduction in -adrenergic relaxation and caused roughly complete relaxation as shown in Figure 1C. Concentration-response curves for ISO shifted in parallel to the right in the second plot (Figure 5A), also indicating that the maximal mechanical response to -agonists does not reduce after exposure to -agonists. On the other hand, previous observations derived from molecular biological methods have shown that exposure to higher concentration of -agonist results in a decrease in both sensitivity and the maximal response of adenylyl cyclase to -agonists (22). These observations indicate that subsequent reduction in -adrenergic relaxation is not consistent with the molecular mechanisms. Because -adrenergic relaxation is markedly augmented after incubation with CTX (12), this enhancement seems to antagonize the reduced relaxation in response to a -agonist after exposure to a -agonist (see Figure 1C). However, when the tissues were preincubated with a lower concentration of CTX (0.02 µg/ml), which does not cause a marked increase in relaxation induced by -agonists, an inhibition of subsequent reduction in -adrenergic relaxation is also mimicked by this experimental condition (unpublished observation). The effects of CTX in this study cannot be simply explained by enhancement of -adrenergic relaxation. A previous report has shown that incubation with CTX causes an attenuation of the contraction induced by MCh (27). Inhibiting MCh-induced contraction by CTX may lead to alteration of -adrenergic sensitivity to muscarinic contraction. However, subsequent reduction in -adrenergic relaxation occurred even though MCh-induced contraction was suppressed within 1 h after a 45-min exposure to 3 µM ISO (Figure 1A, upper trace). Moreover, when MCh concentration was lowered to 0.1 µM (10-fold less concentration than that used in this study), reduced responsiveness to ISO occured similarly after exposure to continuous and repeated exposure to ISO (unpublished observation). Concentration-inhibition curves for ISO on 0.1 µM MCh-induced contraction also shifted to the right at the second cumulative application, different from Figure 5B (unpublished observation). These results indicate that inhibiting MCh-induced contraction by CTX is not involved in prevention of subsequent reduction in -adrenergic relaxation by this agent.

-Agonists augment the activity of large conductance K_Ca channels mediated independently by two pathways as follows: (1) cAMP-dependent channel phosphorylation via PKA and (2) cAMP-independent channel stimulation via G_s (25). We investigated the involvement of K_Ca channel activity regulated by G_s activity in the reduced relaxation in response to a -agonist after continuous and repeated exposure to a -agonist. A single channel record has demonstrated that extracellular application of IbTX results in an inhibition of K_Ca channel activity in a concentration-dependent fashion, and that more than 2 nM IbTX causes roughly complete closure of these channels in smooth muscle cells (28). In the presence of 30 nM IbTX, the basal mechanical tone was not entirely affected, but CTX failed to cause a suppression of the reduced relaxant effect of a -agonist (Figures 2A and 4A). These observations demonstrate that an inhibition of -adrenergic desensitization by CTX may be caused by an augmentation of K_Ca channel activity by G_s, suggesting that a suppression of the channel activity by a reduction in G_s activity leads to the reduction in the subsequent -adrenergic relaxation after exposure to a -agonist. As shown in Figure 5, we also investigated the subsequent relaxant action of a -agonist after continuous and repeated exposure to a -agonist by constructing two concentration- inhibition curves for ISO on precontraction by 1 µM MCh. Concentration-inhibition curves for ISO in isolated human tracheas were shifted to the right in the second plot, similar to previous observations (18). A rightward shift of concentration- inhibition curves for ISO at the second plot means -adrenergic desensitization. Preincubation with CTX does not cause a rightward shift of the curves at second plot (Figure 5B), indicating again that preactivation of G_s prevents subsequent reduction in -adrenergic relaxation. In the presence of 30 nM IbTX throughout the experiments, concentration-inhibition curves for ISO do not shift to the right at the second application (Figure 5D). This observation indicates that K_Ca channels may be involved in the subsequent reduction in -adrenergic relaxation because the curves must shift to the right even when these channels are fully shut down if this -adrenergic desensitization is mediated by mechanisms other than these channels. These results also indicate that G_s-K_Ca channel stimulatory linkage may be involved in the prevention of -adrenergic desensitization in human tracheal smooth muscle. In contrast, the reduced responsiveness to -agonists produced by exposure to proinflammatory cytokines is considered to be the result of an augmentation of a pertussis (PTX) toxin-sensitive G protein (G_i), which inhibits the activity of adenylyl cyclase (29, 30). Because G_i proteins suppress the activity of K_Ca channels of smooth muscle in trachea and other tissues (31, 32), G_i-K_Ca channel inhibitory linkage may be involved in the reduction of the mechanical response to a -agonist after exposure to a -agonist and cytokines. Moreover, -adrenergic relaxation is augmented by incubation with PTX (33) and the enhanced effect of -agonists is also markedly inhibited in the presence of 100 nM charybdotoxin, a selective K_Ca channel antagonist, similar to incubation with CTX (27). These observations also support the idea that the G_s-K_Ca channel stimulatory linkage may be a key process in the inhibition of the subsequent reduced relaxation in response to -agonists after exposure to a -agonist.

In conclusion, the reduced relaxant effect of -agonists on contraction induced by spasmogens participating in asthma attacks occurs after continuous and repeated exposure to a -agonist in human tracheal smooth muscle. This reduced responsiveness is mediated by homologous desensitization via cAMP-independent pathways, and is prevented by prolonged augmentation of K_Ca channel activity via irreversible G_s stimulation. Our results provide evidence that the repeated use of -agonists within short time intervals without any additional medication as therapy for acute asthma attacks may be harmful, because it leads to a desensitization of -adrenergic receptors in airway smooth muscle.