Introduction

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

An increase in smooth muscle (SM) tension and/or phosphorylation of 20-kD regulatory light chain of myosin (MLC₂₀) at a constant Ca²⁺ concentration is referred to as Ca²⁺ sensitization (1, 2). This phenomenon is thought to be mediated mainly by a G protein-coupled inhibition of SM phosphatase 1 associated with myosin (SMPP-1M) (3). RhoA p21 (6), a small guanosine triphosphatase (GTPase) of ras superfamily, and protein kinase C (PKC) - dependent pathways (7, 8) are the proposed mechanisms of the SMPP-1M inhibition. Ca²⁺ sensitization is observed not only in vascular but also in other visceral SM tissues, including airway SM. Indeed, receptor-dependent, G protein-mediated Ca²⁺ sensitization occurs in canine tracheal SM (9). We have also demonstrated that the extent of Ca²⁺ sensitization is highly dependent upon receptor type, and that guanosine 5'-O-(3'-thiotriphosphate) (GTPS)- sensitive and PKC-sensitive pathways may diverge in permeabilized canine tracheal SM (10).

Recently, it has been reported that two target proteins of rhoA p21, rho-associated coiled coil-forming protein kinase (ROCK I [11], also called p160ROCK [12]) and its isoform, ROCK II (13) (also known as ROK [14] or rho kinase [15]), play a key role in the rhoA p21-mediated Ca²⁺ sensitization. (+)-(R)-Trans-4-(1-aminoethyl)-N-(4-pyridyl)cyclohexane carboxamide dihydrochloride, monohydrate (Y-27632) effectively inhibited ROCK I as well as ROCK II both in vitro and in vivo; Y-27632 reversed GTPS-induced Ca²⁺ sensitization in -escin-permeabilized mesenteric arteries, and decreased blood pressure in the hypertensive animal models (16). Similarly, intact guinea-pig trachea contracted by histamine also responded to Y-27632 (16), suggesting that inhibition of the rho/ROCK pathway may become a new treatment not only of hypertension but also of bronchial asthma. However, the efficacy of excitatory agonists for Ca²⁺ sensitization varied in canine tracheal SM (10), and the precise mechanism of action of the different agonists also varied in rabbit vascular and ileum SM; inactivation of rhoA p21 by epidermal cell differentiation inhibitor (EDIN)-induced adenosine diphosphate (ADP)-ribosylation effectively blocked Ca²⁺ sensitization evoked by guanosine triphosphate (GTP) alone, GTP plus carbachol (CCh), GTP plus endothelin-1 (ET-1), and fluoroaluminates, but not that evoked by GTP plus phenylephrine or GTPS (17). Thus, it remains to be elucidated whether rho/ROCK signaling might contribute to Ca²⁺ sensitization in airway SM.

The aim of this study was to determine whether inhibition of rho/ROCK signaling by Y-27632 might reverse G protein-mediated and/or PKC-mediated Ca²⁺ sensitization of airway SM in vitro. We permeabilized rabbit tracheal SM with Staphylococcus aureus -toxin (10, 18), and measured isometric tension at a constant free Ca²⁺ in response to CCh plus GTP, GTPS, or a phorbol ester, 4-phorbol 12,13-dibutyrate (PDBu), followed by cumulative application of Y-27632. The inhibitory effect of Y-27632 on ET-1 plus GTP-induced Ca²⁺ sensitization was also tested in human bronchial SM permeabilized with -toxin (18). Further, we determined expression of rhoA p21, ROCK I, and ROCK II in rabbit tracheal and human bronchial SM by immunoblot analysis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

The trachea was removed from Japanese albino rabbits (weighing 2.5 to 3.0 kg) under halothane anesthesia. The anesthesia was administered by placing the animals in an anesthetic chamber until they became unresponsive to corneal reflex. When the tracheal tissue had been removed, the animals were killed by rapid exsanguination through the carotid artery, in accordance with the recommendations of the Council of Animal Care of Gunma University (Gunma, Japan), and dissected as previously described (18).

Human bronchial SM was prepared from a macroscopically normal part of the lung tissue, which was obtained at surgery for lung cancer. The surgically resected tissue was put in ice-cold Dulbecco's modified Eagle's medium, and small bronchi with an outer diameter of 2 to 4 mm were carefully dissected as previously described (18). Consent was obtained from each patient before surgery.

Isometric Force Measurement in Intact Rabbit Tracheal SM

The rabbit tracheal strips (approximately 3 mm wide and 10 mm long with cartilage and epithelium) were set between a hook and an isometric force transducer (strain gauge TB-612T; Nihon Kohden Ltd., Tokyo, Japan) connecting an amplifier (TB-611-T; Nihon Kohden) and a multipen recorder (R66; Rika Denki Ltd., Tokyo, Japan), and vertically mounted in a 10-ml Magnus tube filled with Tyrode's solution aerated continuously with 5% CO₂ in O₂. The composition of the solution was NaCl, 136.8 mM; KCl, 2.7 mM; CaCl₂, 1.8 mM; MgCl₂, 1 mM; NaH₂PO₄, 0.4 mM; NaHPO₃, 11.9 mM; and glucose, 5.6 mM; and temperature was kept at 37°C. At the onset of each experiment, tissues were subjected to an imposed tension of 1.0 g and allowed to equilibrate for 60 min. The solution was changed at 20-min intervals. To confirm stability of the preparations, application of high potassium (154 mEq/liter) and its washing were repeated twice, followed by CCh-induced contraction. Ibuprofen (2 µM), a cyclooxygenase inhibitor, was present throughout both the intact and the permeabilized experiments (see following sections) to prevent spontaneous tone development due to release of cyclooxygenase products (10, 18).

Permeabilization with -Toxin or Triton X-100

Small strips of rabbit tracheal SM (200 to 300 µm wide; 40 to 50 µm thick; 3 mm long) and human bronchial SM (150 to 200 µm wide; 20 to 30 µm thick; 3 mm long) were mounted on a bubble plate (400 µl per bubble), and isometric force development was measured using a force transducer (AE801; SensoNor, Horten, Norway). The muscle strips were stretched to 1.3 times rest length. The composition of the solutions has been reported elsewhere (10, 18). In brief, the normal relaxing solution (G1) contained 74.1 mM potassium methanesulfonate, 2 mM Mg²⁺, 4.5 mM adenosine triphosphate (Mg²⁺ salt), 1 mM [ethylene-bis(oxyethylenenitrilo)]-tetraacetic acid (EGTA), 10 mM creatine phosphate, and 30 mM 1,4-piperazinebis (ethane sulfonic acid)-KOH (pH 7.1 at 24°C, ionic strength 0.2). The same solution containing 10 mM, rather than 1 mM, EGTA and various amounts of calcium methanesulfonate was used to achieve the desired concentration of free Ca²⁺.

The tracheal strips were exposed to -toxin (16.4 µg/ ml) with calcium ionophore A23187 (10 µM) in the pCa 6.5 solution for 30 to 45 min at 30°C, and then washed in G1 (negative logarithm of free Ca²⁺ concentration) for 5 min (10). The calcium ionophore was used to deplete Ca²⁺ stores of sarcoplasmic reticulum (SR). The temperature was kept at 24°C after permeabilization to obtain a reproducible contractile response. For permeabilization with Triton X-100, tracheal SM was incubated with 0.1% Triton X-100 in G1 at 4°C for 30 min, after which the temperature was increased rapidly to 30°C by exchange of the cold bubble plate for one that had been prewarmed, and permeabilization was continued for an additional 15 min. Cold preincubation with Triton X-100 was useful in obtaining homogeneously permeabilized preparations as reported by Lee and colleagues (19). The experiments using Triton X-100-permeabilized tracheal SM were performed at 24°C in the presence of calmodulin (CaM; 1 µM) (5).

Intact Rabbit Tracheal SM

After reproducible responses to high potassium (154 mEq/ liter) were obtained, tracheal SM was contracted with CCh that was cumulatively applied to the bath (0.01 to 100 µM). In another series of experiments, when CCh (1 µM)- induced contraction became stable (approximately 20 min after the CCh application), we added Y-27632 cumulatively to the bath. Finally, complete relaxation was confirmed by a potent phosphodiesterase (PDE) inhibitor, 3-isobutyl-1- methylxanthine (IBMX; 100 µM).

-Toxin-Permeabilized Preparations

The maximum contraction induced by pCa 5.0 and relaxation in G1 was repeated twice. The second pCa 5.0 response was employed as the maximum contraction to normalize the following tension developments in each strip. When submaximum contraction induced by pCa 6.5 plus GTP (3 µM) was stable, CCh (100 µM) was added to the -toxin-permeabilized SM. At the peak of additional contractions (Ca²⁺ sensitization) Y-27632 was cumulatively added to the strip (except as shown in Figure 8b, where a single dose of Y-27632 was applied to the strip). Control experiments were run in parallel in which the same amount of the vehicle (water) for Y-27632 was added to the strips at the same intervals as the Y-27632 application. Similar protocol was carried out for ET-1 plus GTP- induced, GTPS-induced, and PDBu-induced Ca²⁺ sensitization. We used CCh to evaluate agonist-induced Ca²⁺ sensitization in rabbit trachea because CCh was the most potent Ca²⁺-sensitizing agonist in canine trachea (10), and muscarinic acetylcholine receptor-mediated signaling was preserved in -toxin-permeabilized rabbit trachea (18). On the other hand, in human bronchial SM studies we preferred ET-1 to CCh because ET-1 response was more constantly observed than was CCh response. This is presumably due to sparse distribution of muscarinic acetylcholine receptor at the distal portion of bronchial tree (20), although human bronchial tree throughout expresses endothelin B-receptor subtype that mediates ET-1-induced contraction (21).

View larger version (30K): [in this window] [in a new window]   Figure 8.   GDPS-induced or Y-27632-induced rapid relaxation of -toxin-permeabilized and Ca2+-sensitized rabbit trachea, and lack of effect of IBMX on these events. After the initial pCa 5.0 response was obtained, 100 µM IBMX (b) or its vehicle, 0.1% dimethyl sulfoxide (DMSO) (a), was present 20 min before and during the following Ca2+ sensitization experiments as indicated. When the 100 µM CCh-induced additional contraction reached a peak, we added 2 mM GDPS (a) or 100 µM Y-27632 (b) to the strips. The traces represent four sets of separate experiments. Data are summarized in c. **P < 0.01 versus the submaximum contractions induced by pCa 6.5 plus 3 µM GTP (Bonferroni method). The CCh response with or without IBMX was not significant (Mann-Whitney U test). N.S. indicates not significant.

Triton X-100-Permeabilized Preparations

In the preliminary experiments with -toxin-permeabilized SM, we observed that Ca²⁺-induced contraction was partially reversed by Y-27632. To determine whether Y-27632 might affect the activity of myosin light chain kinase (MLCK) of airway SM, we applied Y-27632 to Triton X-100-permeabilized rabbit trachea. Although MLCK and SMPP-1M system was preserved, introduction of val14p21^rhoA (a constitutive active rho mutant) to the Triton X-100-permeabilized SM failed to evoke a contraction, indicating that the rho/ROCK-mediated Ca²⁺-sensitizing system was removed in such heavily permeabilized strips (17). After the maximum contraction at pCa 5.0 with CaM (1 µM) was obtained, the strip was relaxed in G1 and then pCa 6.0 plus CaM (1 µM)-induced contraction was evoked. When the produced force was stable, GTPS was added to the strip to ascertain functional removal of the rho/ROCK pathway from the Triton X-100-permeabilized strip, followed by cumulative application of Y-27632. Finally, calyculin A (300 nM), a potent phosphatase-1 and -2 inhibitor (19), was added to the strip.

Determination of rhoA p21, ROCK I, and ROCK II in Airway SM

Four to five dissected airway SM strips were placed in 60 µl of glycerol sample buffer (containing 1% sodium dodecyl sulfate [SDS], 10% glycerol, 20 mM dithiothreitol, and 100 µg/ml bovine serum albumin [BSA]) and manually ground with a small glass-glass homogenizer on ice. The amount of 10 µl of the homogenates was used for protein assay (Bio-Rad Protein Assay; Bio-Rad, South Richmond, CA) using BSA as a standard. The resultant 50-µl homogenates were transferred to a 1-ml centrifuge tube. To precipitate the protein, 694 µl of cold distilled water, 250 µl of 24% trichloroacetic acid, and 6 µl of 2% deoxycholate were added to the tube, followed by centrifuging for 10 min at 5,000 × g. Supernatant was discarded, the resultant pellet was neutralized with 1 M Tris base, and 20 µl of sample buffer (Laemmli) was added to the pellet. Approximately 4 µg protein was subjected to each well of SDS polyacrylamide gel electrophoresis (PAGE) (acrylamide, 12.5% for rhoA p21 and SM -actin; 7.5% for ROCK I and ROCK II). The proteins were transferred to the nitrocellulose membrane, and immunostaining was carried out using the following antibodies and visualized with enhanced chemiluminescence (Amersham, Buckinghamshire, UK): monoclonal anti-rhoA p21 (1:500), monoclonal anti-ROCK II (1:1,000), polyclonal anti-ROCK I (1:1,000), and monoclonal anti-SM -actin (1:5,000) were applied to the membrane for 4 h at room temperature. After rinsing the unbound primary antibodies and blocking the nonspecific binding with nonfat milk, we used treatment with sheep antimouse or donkey antirabbit immunoglobulin (Ig)G antibodies linked to horseradish peroxides (1:5,000) for 2 h at room temperature to detect the primary antibodies.

Reagents

S. aureus -toxin was obtained from Research Biochemicals International (Natick, MA); CCh, IBMX, and CaM were from Sigma (St. Louis, MO); and ET-1, PDBu, and calyculin A were from Calbiochem (La Jolla, CA). Y-27632 was a gift from Yoshitomi Pharmaceutical Industries Ltd. (Osaka, Japan). Y-27632 was dissolved in distilled water as stock solution (10 mM), and stored at 20°C until use. Guanosine 5'-O-(2'-thiodiphosphate) (GDPS), GTP, and GTPS were from Boehringer Mannheim (Indianapolis, IN). Monoclonal anti-SM -actin antibody was purchased from Sigma. A mouse monoclonal antibody against rhoA p21 was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), and a mouse monoclonal antibody against ROK (ROCK II) was purchased from Transduction Laboratories (Lexington, KY). Rabbit anti-p160ROCK (ROCK I) polyclonal antibody was a gift from T. Ishizaki (Kyoto University, Kyoto, Japan) (22). The second antibodies (sheep antimouse or donkey antirabbit IgG) were obtained from Amersham.

Statistical Analysis

Unless noted otherwise, the developed force was normalized to the initial pCa 5.0 response in the same strip, and given as mean ± standard error. Data were compared by the Mann-Whitney U test or Student's t test with the Bonferroni correction for multiple comparisons. A P value of < 0.05 was considered to be statistically significant in the Mann-Whitney U test, and a significant level of P < 0.05/m (where m is the number of comparisons) was used in the Bonferroni method.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

Intact Rabbit Tracheal SM

As shown in Figure 1a, CCh dose-dependently contracted the intact rabbit tracheal SM with a concentration of drug producing half-maximum contraction (EC₅₀) of 0.37 ± 0.1 µM (n = 5), and reached a peak (12.0 ± 1.2 mN, n = 5) at 100 µM CCh. The CCh (1 µM)-induced contractions were reversed completely by Y-27632 with a concentration of drug producing half-maximum inhibition of effect (IC₅₀) of 1.29 ± 0.2 µM (n = 5) (Figure 1b). IBMX did not reduce the tone further (data not shown).

View larger version (19K): [in this window] [in a new window]   Figure 1.   Dose-response curves for CCh and Y-27632 in intact rabbit trachea. CCh was applied to intact rabbit trachea (n = 5) cumulatively (a). After the CCh (1 µM) response became stable, Y-27632 was cumulatively added to the strip (n = 5) (b).

Expression of rhoA p21, ROCK I, and ROCK II in Airway SM

Figure 2 shows that both rabbit tracheal and human bronchial SM expressed rhoA p21, ROCK I, and ROCK II. Although comparable densities of SM -actin (Figure 2d) suggest that similar amounts of SM proteins of rabbit and human tissues were run in the gels, ROCK II (Figure 2b) and rhoA p21 (Figure 2c) showed a tendency of lower expression in human tissue than in rabbit tissue.

View larger version (22K): [in this window] [in a new window]   Figure 2.   Immunoblot analysis of rhoA p21, ROCK I, and ROCK II in airway SM. Arrows indicate standard molecular weight. Crude homogenates of rabbit tracheal (left lane) and human bronchial SM (right lane) were run in SDS-PAGE. The data represent four sets of separate experiments. Panels a, b, c, and d indicate ROCK I, ROCK II, rhoA p21, and SM -actin, respectively.

Effect of Post-Treatment with Y-27632 on Agonist-Induced Ca²⁺ Sensitization in Rabbit Tracheal and Human Bronchial SM Permeabilized with -Toxin

The pCa 5.0-induced contractions before Ca²⁺ sensitization were 2.97 ± 0.1 mN in rabbit trachea (n = 45) and 1.92 ± 0.4 mN in human bronchial SM (n = 10). The stable contractions induced by pCa 6.5 plus GTP (3 µM) in trachea (Figure 3b) and by pCa 6.7 plus GTP (3 µM) in bronchus (Figure 4b) were 12.2 ± 5.3% (n = 6) and 21.9 ± 2.1% (n = 4), respectively. In the presence of GTP (3 µM), CCh (100 µM) induced an additional contraction at a fixed free Ca²⁺ concentration of pCa 6.5 in -toxin-permeabilized trachea (Figure 3a). The peak force achieved 84.4 ± 3.0% (n = 6) of the initial contraction at pCa 5.0, followed by spontaneous decline of force in the control strips. The resultant force approximately 60 min after the CCh application was 34.1 ± 9.9% (n = 6). The resultant contractions were reversed to the initial submaximum level of contraction at pCa 6.5 by a nonpermeable G-protein inhibitor, GDPS (1 mM), indicating that the CCh-induced additional contraction was due to G-protein-mediated Ca²⁺ sensitization in completely permeabilized preparations, and that the sensitization was at least partially retained at this point (Figure 3a). As shown in Figure 3b, Y-27632 also reversed the CCh (100 µM)-induced Ca²⁺ sensitization in a dose-dependent manner. The peak contraction of the CCh-induced Ca²⁺ sensitization was 86.9 ± 11.4% (n = 6). Note that the force returned to the basal tone at 100 µM of Y-27632 (2.69 ± 3.0%, n = 6). The dose-response relationship is shown in Figure 3c. The IC₅₀ value was 1.44 ± 0.3 µM (n = 6).

View larger version (25K): [in this window] [in a new window]   Figure 3.   The inhibitory effect of Y-27632 on CCh-induced Ca2+ sensitization in rabbit trachea permeabilized with -toxin. When the pCa 6.5 plus GTP (3 µM)-induced submaximum contraction became stable, CCh (100 µM) was added to the strip. After the additional contraction reached a peak, Y-27632 (b) or the vehicle (a) was cumulatively applied to the strip. A G-protein inhibitor, GDPS, was added as indicated in control experiments (a). Data from six experiments are summarized in c. Y-27632 (filled circle) or the vehicle (open circle) was applied to the strip. **P < 0.01 versus control.

View larger version (25K): [in this window] [in a new window]   Figure 4.   The inhibitory effect of Y-27632 on ET-1-induced Ca2+ sensitization in human bronchial SM permeabilized with -toxin. When the pCa 6.7 plus GTP (3 µM)-induced submaximum contraction became stable, ET-1 (1 µM) was added to the strip. After the additional contraction reached a peak, Y-27632 (b) or the vehicle (a) was cumulatively applied to the strip. A G-protein inhibitor, GDPS, was added as indicated in control experiments (a). Data from four to six experiments are summarized in c. Y-27632 (filled circle) or the vehicle (open circle) was applied to the strip. **P < 0.01 versus control.

Similar findings were observed in ET-1-induced Ca²⁺ sensitization of the -toxin-permeabilized human bronchial SM (Figure 4). The peak force developments induced by ET-1 in Y-27632-treated and nontreated (control) strips were 64.7 ± 4.1% (n = 6) and 59.0 ± 5.1% (n = 4), respectively. The ET-1-induced contractions were reversed to basal level (0.21 ± 1.0%, n = 6) by Y-27632 at 300 µM, whereas the resultant force before GDPS application in control strips was 36.5 ± 6.8% (n = 4). Note that GDPS relaxed the ET-1-induced contraction to the prior submaximum contraction level at pCa 6.7. Therefore, these results indicate again that the extent of ET-1-induced Ca²⁺ sensitization was reduced but still substantially retained at this point. The IC₅₀ value for Y-27632 in ET-1-induced contraction of bronchial SM was 4.10 ± 1.1 µM (n = 6).

Selectivity of Y-27632 toward ROCK but Not MLCK

In -toxin-permeabilized strips, the mechanism of the Ca²⁺ sensitizing system was retained almost completely. In contrast, Triton X-100 permeabilization entirely disrupted the SMPP-1M regulatory system for Ca²⁺ sensitization while preserving MLCK and the function of calyculin A (a potent phosphatase inhibitor). As shown in Figure 5a, Y-27632 partially but significantly relaxed the pCa 6.0- induced contraction in the -toxin-permeabilized strips; Y-27632 at 300 µM decreased the contraction to 32.1 ± 2.1% (n = 3), normalized to the initial contraction at pCa 6.0. The time-matched control strips did not show any decline of force (trace not shown). In Triton X-100-permeabilized strips, the maximum contraction at pCa 5.0 with CaM (1 µM) was 1.07 ± 0.2 mN (n = 7). There was no response to GTPS, and no difference in force reduction between Y-27632-treated strips (5.6 ± 10.3%, normalized to the peak force at pCa 6.0, n = 4) and control strips (4.8 ± 3.7%, n = 3) before calyculin A application (Figures 5b and 5c). Calyculin A contracted both -toxin- and Triton X-100-permeabilized strips either with or without Y-27632.

View larger version (23K): [in this window] [in a new window]   Figure 5.   Comparison of Y-27632 effect on Ca2+-induced contraction between -toxin- and Triton X-100-permeabilized strips. When the pCa 6.0 response became stable, Y-27632 was applied to -toxin- (a) or Triton X-100- (c) permeabilized rabbit tracheal SM. The same amount of vehicle was added to the time-matched control strip permeabilized with Triton X-100 (b). CaM (1 µM) was present in Triton X-100- (b and c) but not in -toxin- (a) treated preparations. The final calyculin A (300 nM)-induced contraction was achieved to the initial maximum contraction level at pCa 5.0 in both -toxin- (a) and Triton X-100- (b and c) permeabilized strips. The traces represent at least three sets of separate experiments.

GTPS-Induced Ca²⁺ Sensitization

GTPS, a stable GTP analogue, activates both trimeric and monomeric GTPase including rhoA p21. Indeed, GTPS was a potent Ca²⁺ sensitizing agent in -toxin-permeabilized canine trachea (10). When the pCa 6.5 response was stable, we cumulatively added GTPS to the strip. GTPS dose-dependently increased the Ca²⁺ sensitivity, and the force increased from 3.32 ± 0.5% before GTPS application to 85.4 ± 3.6% (n = 6) at 30 µM of GTPS (Figure 6a). The response to GTPS reached a peak at 3 µM with EC₅₀ of 0.16 ± 0.03 µM (n = 6) (Figure 6b). Single (noncumulative) application of GTPS at 10 µM (a supramaximum concentration) produced the largest Ca²⁺ sensitization. The peak contractions were at 104.6 ± 6.6% in the control group and 105.3 ± 6.6% in the Y-27632-treated group (before Y-27632 application, n = 6). Although the GTPS-induced force development declined spontaneously, as did CCh plus GTP (Figure 7a), Y-27632 apparently accelerated relaxation with the IC₅₀ value of 1.15 ± 0.3 µM (n = 6) (Figure 7b). Data are summarized in Figure 7c.

View larger version (20K): [in this window] [in a new window]   Figure 6.   GTPS-induced Ca2+ sensitization in rabbit trachea permeabilized with -toxin. To determine the dose-response relationship in GTPS-induced Ca2+ sensitization we cumulatively added GTPS to the strip that was contracted by pCa 6.5 (a). Dose-response curves for GTPS are shown in (b) (n = 6).

View larger version (24K): [in this window] [in a new window]   Figure 7.   Inhibition of GTPS-induced Ca2+ sensitization by Y-27632 in rabbit trachea permeabilized with -toxin. When the 10 µM GTPS-induced additional contraction reached a peak, we added the vehicle (a) or Y-27632 (b) to the strips. The traces represent six sets of separate experiments. Data from Y-27632- treated (filled circle, n = 6) and control strips (open circle, n = 6) are summarized in c. *P < 0.05 versus control; **P < 0.01 versus control.

Rapid Relaxation of Ca²⁺-Sensitized Airway SM by Noncumulative Application of Y-27632, and Lack of Effect of IBMX

To minimize the time-dependent factors, we added a supramaximum concentration of Y-27632 (100 µM) to the strips contracted by CCh (100 µM) at the peak in the presence of pCa 6.5 plus GTP (3 µM), resulting in a rapid relaxation (Figure 7b). The half-time (t_1/2) of relaxation was 70.6 ± 11.8 s (n = 4, P < 0.01) in the Y-27632-applied group and 80.9 ± 3.7 s (n = 4, P < 0.01) in the GDPS- applied group. Note that t_1/2 of spontaneous force reduction was approximately 30 min (1,637.5 ± 44.5 s, n = 6, Figure 3a). The amplitude of CCh response reached a comparable level with the initial pCa 5.0-induced contraction, indicating that CCh sensitized the contractile machinery apparatus to near maximum level in the presence of GTP. The possibility of incomplete permeabilization was ruled out by the finding that GDPS (2 mM) completely reversed the CCh response to the prior level (9.05 ± 2.1%, n = 4, Figure 8a). To examine whether the Y-27632 effect was due to inhibition of PDE activity, we treated the permeabilized strip with IBMX, a potent PDE inhibitor. As shown in Figure 8b, IBMX (100 µM) was present 20 min before and during the responses to CCh and Y-27632. The submaximum contraction induced by pCa 6.5 with GTP (3 µM) showed a tendency of reduction in the presence of IBMX (4.25 ± 0.3% in the IBMX-treated group and 5.96 ± 0.5% in the control group, n = 4). Even in the presence of IBMX, however, CCh-induced full Ca²⁺ sensitization was still observed (95.2 ± 5.9% in the IBMX-treated group, 95.0 ± 5.9% in the control group, n = 4), and Y-27632 (100 µM) relaxed the contraction (4.24 ± 2.5%, n = 4). Thus, IBMX failed to affect both the CCh-induced Ca²⁺ sensitization and Y-27632-induced relaxation (Figure 8c).

PDBu-Induced Ca²⁺ Sensitization in Rabbit Tracheal SM Permeabilized with -Toxin

The prior submaximum contractions at pCa 6.5 without GTP were 2.29 ± 0.4% in the control group (n = 7), and 2.00 ± 0.6% in the Y-27632-treated group (n = 5). A PKC activator, PDBu (1 µM), gradually caused Ca²⁺ sensitization without any spontaneous force reduction in control experiments of rabbit tracheal SM permeabilized with -toxin (Figure 9a). The developed force became stable 35 to 45 min after the PDBu application. The peak amplitudes achieved were 68.9 ± 5.7% in the Y-27632-treated group (n = 5) and 75.0 ± 7.5% in the control group (n = 7), followed by cumulative addition of Y-27632 or vehicle to the strips. Although inhibition of PDBu-induced contraction by Y-27632 was smaller than that of GTPS- induced contraction, Y-27632 dose-dependently relaxed the strip (Figure 9b). The resultant force in the presence of Y-27632 (100 µM) was 48.4 ± 4.2% (n = 7) (Figure 9c).

View larger version (24K): [in this window] [in a new window]   Figure 9.   Partial inhibition of PDBu-induced Ca2+ sensitization by Y-27632. Y-27632 (b) or the vehicle (a) was added to the strips as indicated. The traces represent at least five sets of separate experiments. Data from Y-27632-treated (filled circle, n = 7) and control strips (open circle, n = 5) are summarized in c. *P < 0.05 versus control.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

Importance of Ca²⁺ Sensitization on Sustained Phase of Airway SM Contraction

Ca²⁺-CaM complex activates MLCK, leading to phosphorylation of MLC₂₀ (23). This is necessary and sufficient to initiate SM contraction. In airway SM, like other types of SM (1), inositol 1, 4, 5-triphosphate (InsP₃)-mediated Ca²⁺ release is responsible for this initial step of CCh-triggered contraction (18). In contrast, to maintain force developments, contribution of Ca²⁺ sensitization has been expected from the experiments of simultaneous measurement of force and intracellular Ca²⁺ concentration in intact (nonpermeabilized) vascular (24) and airway SM (25). Recently, this was supported by the results from inhibition of rhoA p21 activity in intact SM by glucosylation of rhoA p21 with Clostridium difficile toxin B (26) or by in vivo ADP-ribosylation of rhoA p21 with chimeric toxin DC3B (27). Until Y-27632 was reported, however, to what extent rho/ ROCK-mediated Ca²⁺ sensitization contributed to maintenance of force in intact airway SM contraction had not been known. In the present study we confirmed that post-treatment with Y-27632 caused complete relaxation in intact and -toxin-permeabilized rabbit tracheal SM. The values of IC₅₀ in intact and in -toxin-permeabilized tracheal SM were very close. When Y-27632 relaxed the intact strips contracted by CCh, mainly through inhibition of rho/ROCK-mediated Ca²⁺ concentration, the IC₅₀ values should be comparable between intact and permeabilized strips. Thus, the similar IC₅₀ values of intact versus permeabilized strips support the idea that relaxation of intact trachea was due to selective inhibition of the rho/ROCK pathway by Y-27632. Similar experiments were performed with human bronchial SM to estimate the clinical applicability of Y-27632 as a bronchodilator. ET-1-induced Ca²⁺ sensitization was also reversed by Y-27632 in human bronchial SM permeabilized with -toxin. Thus, it is likely that Ca²⁺ sensitization through rho/ROCK signaling plays a central role in the sustained phase of airway SM contraction, including in human airway SM. The IC₅₀ values in human bronchus were larger than those in rabbit trachea (IC₅₀ for Y-27632: 4.10 µM versus 1.44 µM, respectively). A relatively small amount of expressed proteins, rhoA p21 and ROCK II (Figure 2), may account for the difference in the IC₅₀ values. Further studies are required for qualitative and quantitative determination of these proteins in human airway SM.

Inhibitory Action of Y-27632 toward ROCK but Not MLCK

Although Ca²⁺-induced contraction of -escin-permeabilized rabbit mesenteric artery was not affected significantly by Y-27632 in the original report (16), Y-27632 apparently relaxed the Ca²⁺-induced contraction of -toxin- permeabilized rabbit tracheal SM in the present study (Figure 5a). However, it is not likely that this was due to inhibition of MLCK by Y-27632 because Y-27632 had no apparent effect on Ca²⁺-induced contraction of Triton X-100-treated strips (Figure 5c). Both Y-27632-treated and control strips showed approximately 5 to 6% decline of force, which is a "run down" phenomenon usually observed in preparations permeabilized by saponin, -escin, and Triton X-100, but not -toxin (5, 28). Complete loss of GTPS response in the Triton X-100-treated strips suggests that rho/ROCK signaling was disrupted entirely by treatment with Triton X-100 (17). In contrast, the initial pCa 5.0 plus CaM-induced contraction and the final calyculin A-induced contraction indicate that the regulatory mechanism of MLC₂₀ phosphorylation level by MLCK and SMPP-1M was preserved even in such heavily permeabilized strips (5).

Rho kinase (ROCK II) is known not only to inhibit phosphatase activity of SMPP-1M but also to directly phosphorylate Ser 19 of MLC₂₀, as does MLCK (29). In the -escin-permeabilized strip, influx of 150-kD protein such as IgG and efflux of lactate dehydrogenase (approximately 140 kD) were observed (28), although proteins greater than 1 kD were not able to pass through the pores produced by -toxin (10, 28). Thus, partial leakage of soluble compartment(s) of the rho/ROCK pathway may be the reason why Ca²⁺-induced contraction of the -escin-permeabilized strip was relatively insensitive to Y-27632, although this speculation should be examined further. The EC₅₀ value for GTPS in rabbit trachea permeabilized with -toxin (the present study) was approximately five to six times smaller than that in bovine tracheal SM permeabilized with saponin (30) (EC₅₀ for GTPS: 0.16 versus 0.9 µM); whereas the 61-kD protein, Ca²⁺-independent MLCK, was permeable as in -escin-treated portal vein SM (28), supporting this possibility. Taken together, in -toxin-permeabilized SM it is more likely that Y-27632 inhibited basal activity of ROCK I (and/or ROCK II) but did not affect MLCK activity, as reported (16).

Difference in the Effects of GDPS and Y-27632

Why did Y-27632 more effectively reverse the agonist- induced Ca²⁺ sensitization than did GDPS (Figures 3 and 4)? We propose a possibility that basal ROCK activity was present and GDPS did not affect the activity. This idea is supported by the following reasons: (1) EDIN, which inactivates rhoA p21 via ADP-ribosylation, inhibited submaximal Ca²⁺-induced contractions, whereas GDPS did not affect the pCa-tension curves in the permeabilized rabbit mesenteric artery (17). This is very similar to the results of this study using Y-27632. (2) Amano and colleagues demonstrated that ROCK II phosphorylated MLC₂₀ in the absence of rhoA p21, and that the GTPS-bound form of rhoA p21 accelerated this reaction by approximately twice in vitro (29), suggesting that GTP may not be essential for the basal activity of ROCK. (3) Although GDPs effectively reversed agonist- (this study) and fluoroaluminates-induced Ca²⁺ sensitization (10), GTPS (10 µM)-induced Ca²⁺ sensitization was totally resistant to the GDPS (1 mM) treatment (data not shown). Because Y-27632 effectively reversed the GTPS response (Figure 7), efficacy, potency, and action sites of the reagents seemed to be different. Hence, it is not surprising that Y-27632 was a more effective relaxant than GDPS. However, it remains to be elucidated whether direct ROCK-induced phosphorylation of MLC₂₀ contributes to the basal ROCK activity in situ because inhibition of SMPP-1M has been proposed as the main mechanism of Ca²⁺ sensitization (1).

Spontaneous Decline of CCh-Induced, GTPS-Induced, and ET-1-Induced Ca²⁺ Sensitization

CCh-induced and ET-1-induced Ca²⁺ sensitization reduced with time in SM permeabilized with -toxin. Although it is not known whether the diminished agonist response occurred at the receptor site, the postreceptor site, or both, the time-dependent reduction in GTPS-induced force suggests that the postreceptor site seems to be more responsible for the desensitization to the excitatory agonists. Gong and associates reported similar desensitization to GTPS accompanied by translocation of rhoA p21 from cytoplasm to membrane fraction in -toxin-permeabilized vascular SM (31). This is also the reason why the maximum response to GTPS was smaller in the cumulative application protocol (85%; Figure 6) than that in the noncumulative application protocol (105%; Figure 7).

The time-dependent force reduction of G protein- mediated Ca²⁺ sensitization was a physiologic phenomenon but not a technical artifact. Although an approximately 30% decline of force appeared in the phenylephrine-induced contractions of intact vascular SM (16), the IC₅₀ values are quite comparable between the original and our studies. Unfortunately, a time-matched control trace of the GTPS response was not shown in the original paper (16). Our traces of post-treatment with Y-27632 were very similar to that in the original paper. Therefore, the time-dependent force reduction of the GTPS response should be observed in -escin-permeabilized vascular SM. However, the extent of force reduction might be only to a lesser extent because relative and/or absolute Ca²⁺ sensitization would be small in -escin-permeabilized SM (see below). Furthermore, pCa 6.0 response in the original paper also showed an approximately 30% decline of force (16). By contrast, the responses to pCa 6.0 (data not shown) and to PDBu (Figure 9) in this study were well maintained, and the absolute force of contraction was much larger in the present study than in the original study. These differences were presumably caused not only by the preparations (differences in strip size and in SM types, vascular versus airway SM) but also by the permeabilization method (-escin versus -toxin). -escin- but not -toxin- permeabilized strips show "run down," as up to 150-kD- size proteins leak from the pores produced by -escin (28). So far, the most successful permeabilization method for the long-term maintenance of both receptor coupling and contractility is by -toxin (18, 28). Indeed, the amplitude of contractions at pCa 5.0 and also at pCa 6.5 with Ca²⁺ sensitizing agents in permeabilized preparations was larger than CCh (100 µM)-induced maximum contractions in intact preparations before permeabilization of the same strips (10). The extents of Ca²⁺ sensitization before Y-27632 application in the -toxin-permeabilized airway SM were 86.9 ± 11.4% (CCh, n = 6), 59.0 ± 5.1% (ET-1, n = 4), and 105.3 ± 6.6% (GTPS, n = 6) (normalized to the initial pCa 5.0 response). In the original paper, pCa 5.0 response was not shown (16). Personal communication with the authors, Uehata and coworkers, confirmed that the extent of GTPS-induced contraction was 35.6 ± 6.5% (n = 4) in their -escin-permeabilized vascular SM. These are reasonable values because they are consistent with data from using the -escin-permeabilized vascular SM (28); however, they are much smaller than those of -toxin-permeabilized airway SM in previous (10) and present studies. Thus, one possible explanation for the substantial force reduction in this study is presumably the initial large response to the Ca²⁺ sensitizing agents (e.g., approximately three times larger GTPS response; this study [105%] versus the original report [36%]) in the -toxin-permeabilized SM, where preservation of the signal transduction system for Ca²⁺ sensitization was nearly complete. Another possibility is that rabbit trachea is a more phasic type SM than rabbit mesenteric artery, because we succeeded in Ca²⁺ release experiments from the SR with -toxin-permeabilized rabbit trachea (18). In general, it is difficult to carry out the Ca²⁺ release protocol using tonic-type SM such as rabbit femoral artery. Hence, serious changes in tension with G protein-mediated Ca²⁺ sensitizing agents reveal the time course of the G protein-mediated Ca²⁺ sensitization in the most successfully permeabilized airway SM rather than technical error.

In this case it is important to determine the statistical significance between the time-matched control group and the Y-27632-treated group because time-dependent factors were involved in both groups. At 3 µM and more in rabbit and 10 µM and more in human airways, Y-27632 significantly decreased the force developments. Further, it should be noted that in the control strips GDPS induced force reduction, indicating that agonist-triggered Ca²⁺ sensitization still remained at this point. Therefore, the IC₅₀ data with statistical significance to time-matched strips are useful in comparing our data with the original data, and our data will provide further information on the time course of G protein-mediated Ca²⁺ sensitization in airway SM.

To minimize the time-dependent factors, especially diffusional delays, we applied a single dose of Y-27632 (100 µM) to the sensitized strip at the peak response to CCh (100 µM). As a consequence, the relaxation induced by Y-27632 became more apparent (Figure 8). The extent of CCh (100 µM) response was comparable with the pCa 5.0 response, and relaxant effects of GDPS and Y-27632 were also comparable. These results indicate that CCh-induced full Ca²⁺ sensitization was reversed completely by GDPS at the G-protein(s) level and by Y-27632 at the ROCK level.

Interestingly, PDBu-induced Ca²⁺ sensitization was well maintained in -toxin-permeabilized trachea. It is not clear whether the different time course of produced force between PDBu and other Ca²⁺ sensitizing agents we used was due to qualitative or quantitative difference in PKC activity in rabbit tracheal SM (see below). However, complete relaxation of agonist-induced Ca²⁺ sensitization in both rabbit and human tissues by Y-27632 suggests that a common mechanism through the rho/ROCK pathway is generally present in the sustained phase of agonist-evoked airway SM contraction.

PDBu-Induced Ca²⁺ Sensitization

Introduction of activated PKC caused Ca²⁺ sensitization in single SM (7). A PKC activator, PDBu, also evoked Ca²⁺ sensitization in various types of SM (7, 8), including airway SM (10), and this is mediated mainly by inhibition of SMPP-1M (8). Although we could not exclude the possibility of inhibition of PKC activity by Y-27632, especially at high concentrations (16), partial inhibition by Y-27632 of PDBu-induced Ca²⁺ sensitization suggests a minor role of rho/ROCK in signaling downstream of PKC. However, Y-27632 inhibited Ca²⁺-induced contraction of -toxin- permeabilized SM in this study, raising another possibility: that the partial inhibition of PDBu-induced force might be due to a decrease in the basal ROCK activity. Further studies are required to clarify this point. Nevertheless, because Y-27632 completely reversed agonist- and GTPSinduced Ca²⁺ sensitization, rho/ROCK-mediated signaling plays a much more important role in receptor-dependent, G protein-mediated Ca²⁺ sensitization than does PDBu-sensitive PKC in airway SM. This interpretation is supported by the findings that several PKC inhibitors failed to block acetylcholine-induced Ca²⁺ sensitization of canine trachea permeabilized with -escin (32). However, we could not rule out the possibility that atypical PKC (insensitive to phorbol ester) may contribute to the agonist-evoked Ca²⁺ sensitization in SM (33).

Putative Other Mechanisms of Y-27632 Effect

In -toxin-permeabilized and A23187-treated strips, the actions of ion channels in plasma membranes and of SR for Ca²⁺ storage were functionally removed. Direct evidence of agonist-induced Ca²⁺ sensitization has been established by the receptor-coupled permeabilization techniques with -toxin and -escin (1). Under these experimental conditions, putative other mechanisms of Y-27632 effect were limited. Myosin light chain phosphorylation theory indicates that MLCK and SMPP-1M activity ratio is the major mechanism of force development, although thin-filament proteins such as caldesmon and calponin may contribute to force generation (1). However, we have demonstrated that Y-27632 did not affect the MLCK activity in situ (in this study); we also reported that phosphorylation of caldesmon by mitogen-activated protein kinase (a proposed mechanism of increase in caldesmon activity in situ) had no effect on tension (34), and that contribution of calponin to Ca²⁺ sensitization was minor (only 25%) in canine trachea (10). We also observed that neither a tyrosine kinase inhibitor, genistein, nor phospholipase A₂ inhibitors (quinacrine and manoalide) prevented CCh-induced Ca²⁺ sensitization of rabbit airway SM (unpublished observation). Further, Y-27632 did not affect the calyculin A-induced contraction (16 and Figure 5a), indicating that this compound did not inhibit adenosine triphosphatase activity of myosin. Taken together, it is substantially reasonable that Y-27632 mainly affected the SMPP-1M activity via inhibition of the rho/ROCK pathway, although selectivity of Y-27632 toward other kinases, especially toward atypical PKC in situ, remains to be elucidated.

	Conclusion

Top Abstract Introduction Materials and Methods Results Discussion Conclusion References

In conclusion, rho/ROCK-mediated Ca²⁺ sensitization is important for the sustained phase of contraction in airway SM, and Y-27632 selectively blocks this pathway, leading to complete relaxation of airway SM. Hence, inhibition of rho/ROCK signaling may become a new strategy to resolve airflow limitation in diseases such as bronchial asthma.