INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MECHANICAL STRAIN is an important modulator of many cellular processes including signal transduction, growth, and contraction (4, 8, 17). Phosphorylation of the 20,000-Da myosin light chain is the central regulatory mechanism of smooth muscle contraction. Mehta et al. (21) and Yoo et al. (27) found that the level of myosin light chain phosphorylation elicited by acetylcholine and carbachol is linearly dependent on muscle length in airway smooth muscle. This finding indicates that mechanical strain modulates muscarinic receptor-mediated airway smooth muscle activation. Yoo et al. found that muscarinic receptor-mediated intracellular Ca²⁺ concentration is also length dependent. However, Ca²⁺ sensitivity of myosin light chain phosphorylation does not change with muscle length in smooth muscle (12, 22). These results suggest that the primary target of mechanosensitive modulation may be signal transduction in the cell membrane of airway smooth muscle.

Muscarinic receptors are coupled to phospholipase C activation via heterotrimeric G proteins (24). Matsumoto et al. (20) found that mechanical stretch per se activates phospholipase C activity in smooth muscle cells. Yoo et al. (27) found that carbachol-activated phosphatidylinositol (PI) turnover was significant at optimal length (L_o) but not at 10% L_o in airway smooth muscle. However, it is not known how mechanical strain modulates muscarinic receptor-mediated PI turnover in airway smooth muscle.

Receptor-mediated enzyme activation may be described most simply by Michaelis-Menten kinetics: V = V_max × [agonist]/(K_m + [agonist]), where V is the enzymatic activity at a given agonist concentration ([agonist]), V_max is the maximal enzymatic activity, and K_m is the half-maximal [agonist]. Therefore, mechanical strain may modulate muscarinic receptor-mediated PI turnover by modulating V_max, K_m, or both. Recently, Youn et al. (28) found that mechanical strain modulates the maximal level of myosin light chain phosphorylation that can be elicited by carbachol in bovine tracheal smooth muscle. If signal transduction is the primary target of mechanosensitive modulation, we hypothesized that mechanical strain should also modulate the maximal PI turnover that can be elicited by carbachol. In a dose-response curve, the term "maximal PI turnover" refers to the highest level of PI turnover reached at the maximal carbachol concentration ([carbachol]). The first goal of this study was to test this hypothesis by measuring carbachol-activated PI turnover at different [carbachol] values ranging from 10⁸ to 10³ M and at different muscle lengths ranging from 20 to 100% L_o.

Fluoroaluminate activates heterotrimeric G proteins directly (3, 25) and thereby activates PI turnover in airway smooth muscle (15). To determine whether the length-dependent step is upstream from heterotrimeric G proteins, we measured fluoroaluminate-activated PI turnover at different muscle lengths ranging from 20 to 100% L_o. If the length-dependent step is upstream from heterotrimeric G proteins, then there would not be length sensitivity in fluoroaluminate-induced PI turnover.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tissue preparation. Bovine tracheae were collected from a slaughterhouse and transported to the laboratory in cold (4°C) physiological salt solution (PSS) containing (in mM) 140.1 NaCl, 4.7 KCl, 1.2 Na₂HPO₄, 2.0 MOPS (pH 7.4), 0.02 Na₂EDTA, 1.2 MgSO₄, 1.6 CaCl₂, and 5.6 D-glucose. Adventitial and mucosal layers were carefully dissected away, and smooth muscle strips were prepared along the direction of muscle bundles. One end of the muscle strip was clamped to a stainless steel clip connected to a force transducer (Grass FT.03). The other end of the muscle strip was clamped to a stainless steel clip connected to a length manipulator (Narishige). The muscle strips were then equilibrated for 2 h in PSS (37°C, pH 7.4) bubbled with air and adjusted to L_o for maximal active force development as previously described (14). The protocol consisted of first stretching the muscle strip so that total force was ~10 g and then releasing the muscle strip rapidly to a passive force that was ~2.5 g. Muscle length associated with this passive force was previously found to be optimal for force development induced by K⁺ depolarization (14). The muscle strips at L_o were ~15 mm long × 5 mm wide × 0.5 mm thick. The muscle strips at L_o were then stimulated by K⁺ depolarization with K-PSS, a solution similar to PSS in composition except that 104.95 mM NaCl was substituted for by an equimolar concentration of KCl. Active force developed in this contraction was recorded as F_o, which was used as a standard to normalize forces induced by carbachol or fluoroaluminate in subsequent contractions at different muscle lengths.

The following protocol was used to adjust muscle lengths ranging from 0.2 to 1.0L_o. Actual muscle length at L_o was measured in millimeters with a caliper (0.1-mm resolution). The amount of shortening necessary to change the muscle length to a desired fraction of L_o was calculated. The muscle strip was then quickly released by the calculated amount with the length manipulator (0.1-mm resolution). The force immediately after the quick release was considered to be the passive force at that particular length. The force induced by carbachol or fluoroaluminate above the passive force in subsequent contractions was considered to be the active force.

Measurement of PI turnover. The lithium method is similar to that of Berridge et al. (2) and has been previously described (27). Berridge et al. (2) found that Li⁺ caused only a slight increase in [³H]inositol phosphate accumulation in unstimulated tissues but greatly amplified [³H]inositol phosphate accumulation in carbachol-activated tissues. Grandordy et al. (11) also included LiCl in the [³H]inositol loading solution to study cholinergic receptor-mediated PI turnover in airway smooth muscle. The rationale for including LiCl in the loading solution in this study was to allow equilibration of Li⁺ in the tissue before its activation by carbachol. Accordingly, equilibrated muscle strips were incubated in 10 ml of PSS (37°C) containing myo-[2-³H(N)]inositol (5 µCi/ml; New England Nuclear) and 10 mM LiCl for 2 h. The solution was continuously stirred by a miniature magnetic stir bar. Phosphoinositides were unlikely to be labeled to equilibrium during the 2-h [³H]inositol loading in this study. Therefore, the [³H]inositol phosphate production reflected the flux of PI turnover. In each experiment, two muscle strips dissected from the same trachea were loaded with [myo-³H]inositol. After being loaded, both muscle strips were washed extensively with nonradioactive PSS (37°C). One strip was then incubated in 10 ml of nonradioactive PSS containing 10 mM LiCl for 30 min. The other muscle strip was incubated for 30 min in 10 ml of nonradioactive PSS containing 10 mM LiCl and various carbachol or fluoroaluminate concentrations depending on the experiment. Subsequently, the muscle strips were quickly frozen in a chloroform-methanol (1:1) solution previously cooled on dry ice for 1 h.

Frozen muscle strips were homogenized in 3 ml of a chloroform-methanol (1:1) solution on ice. The homogenate was mixed with 1.5 ml of 0.1 M HCl and 1.5 ml of chloroform and centrifuged. After centrifugation, the aqueous phase was removed and kept for further analysis. The organic phase was mixed with 0.75 ml of chloroform and 0.75 ml of 0.1 M HCl and centrifuged again. The aqueous phase was removed, combined with the first aqueous extract, and neutralized with 0.1 N NaOH. The extract was then analyzed by anion-exchange chromatography with AG1-X8 resin (formate form; Bio-Rad). The column was sequentially eluted with 4-ml volumes of the following solutions: solution 1, 11 rinses of distilled water; solution 2, 14 rinses of 5 mM sodium tetraborate and 60 mM sodium formate; solution 3, 9 rinses of 0.15 M ammonium formate and 0.1 M formic acid; solution 4, 14 rinses of 0.3 M ammonium formate and 0.1 M formic acid; and solution 5, 10 rinses of 0.75 M ammonium formate and 0.1 M formic acid. The column was calibrated with standards (New England Nuclear) of myo-[³H]inositol, [³H]inositol 4-monophosphate (IP₁) [³H]inositol 1,4-bisphosphate (IP₂) and [³H]inositol 1,4,5-trisphosphate (IP₃), which were found to elute in solutions 1, 3, 4, and 5, respectively. Radioactivity of the eluted fractions was measured by liquid scintillation counting with a scintillation fluid of high efficiency (Ultima Gold, Packard). Aliquots of loading solution were also sampled and counted in each experiment; these values were used to normalize data for minor differences in specific activity of the loading solution. Muscle wet weight was calculated for each experiment by subtracting the clamp weight measured at the end of an experiment from the combined weight of tissue and clamp measured at the beginning of an experiment. Total [³H]inositol phosphates [sum of [³H]IP₁, [³H]IP₂, and [³H]IP₃ are expressed in disintegrations per minute (dpm) per gram of wet weight by converting counts per minute to disintegrations per minute with experimentally measured counting efficiencies for the different solutions.

Statistics. Data are presented as means ± SE; n is the number of tracheal rings. Student's t-test was used for statistical comparison of two means; P < 0.05 was considered significant. The correlation between two variables such as total [³H]inositol phosphates and muscle length was analyzed by Pearson's correlation analysis; P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Length dependencies of basal and 1 µM carbachol-induced active force and PI turnover. Figure 1A shows the length dependencies of basal and active force. Basal force in unstimulated tissues was 0.013 ± 0.004 F_o at 0.2L_o and then increased to 0.023 ± 0.012 F_o at 1.0L_o (Fig. 1A, ). The correlation between basal force and muscle length was not significant. Active force in 1 µM carbachol-activated tissues was 0.027 ± 0.007 F_o at 0.2L_o and then increased length dependently to 1.28 ± 0.13 F_o at 1.0L_o (Fig. 1A, ). The correlation between active force and muscle length was significant (P < 0.05). Figure 1B shows the length dependencies of a 30-min accumulation of total [³H]inositol phosphates in unstimulated and carbachol-activated tissues. Total [³H]inositol phosphate accumulation in unstimulated tissues was 0.26 ± 0.02 × 10⁵ dpm/g at 0.2L_o and increased to 0.50 ± 0.22 × 10⁵ dpm/g at 1.0L_o (Fig. 1B, ). The correlation between basal [³H]inositol phosphate production and muscle length was not significant. In 1 µM carbachol-activated tissues, total [³H]inositol phosphate accumulation was 0.63 ± 0.33 × 10⁵ dpm/g at 0.2L_o and then increased length dependently to 1.68 ± 0.75 × 10⁵ dpm/g at 1.0L_o (Fig. 1B, ). The correlation between [³H]inositol phosphate production and muscle length was significant in carbachol-activated tissues (P < 0.05).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Length dependencies of basal and carbachol-induced active force (A) and accumulation of total [3H]inositol phosphates (sum of [3H]inositol 4-monophosphate, [3H]inositol 1,4-bisphosphate, and [3H]inositol 1,4,5-trisphosphate; B). Active force is expressed as a fraction of active force (Fo) induced by K+ depolarization at beginning of each experiment. Lo, optimal length. Data are means ± SE; n = 5-9 tracheal rings. P values, significant correlation with muscle length; ns, insignificant correlation with muscle length.

Because muscle lengths were held constant during loading with [³H]inositol and activation by 1 µM carbachol, the decrease in carbachol-induced [³H]inositol phosphate production at a short muscle length (0.2L_o) could be due to a decrease in [³H]inositol loading, phospholipase C activity, or both. To differentiate these possibilities, we loaded some tissues with [³H]inositol at 1.0L_o and subsequently activated them at 0.2L_o. As shown in Table 1, both basal and carbachol-activated total [³H]inositol phosphate accumulations at 0.2L_o yielded similar results regardless of whether the tissues were loaded with [³H]inositol at 0.2 or 1.0L_o. These results indicate that the decrease in carbachol-induced suprabasal [³H]inositol phosphate production at 0.2L_o reflected a decrease in phospholipase C activity.

Length dependencies of 100 µM carbachol-induced active force and PI turnover. If mechanical strain regulates maximal PI turnover coupled to muscarinic-receptor activation, then PI turnover at a maximal [carbachol] value should also be length dependent. Figure 2A shows the concentration dependence of carbachol-induced contractions. As shown in Fig. 2A, maximal active force was reached at 100 µM carbachol. Figure 2B shows the active force and [³H]inositol phosphate production induced by 100 µM carbachol at 0.2, 0.6, and 1.0L_o. Active force increased length dependently from 0.014 ± 0.006 F_o at 0.2L_o to 1.95 ± 0.15 F_o at 1.0L_o (Fig. 2B, ). The correlation between active force and muscle length was significant (P < 0.05). The 30-min accumulation of [³H]inositol phosphates also increased length dependently from 3.80 ± 0.56 × 10⁵ dpm/g at 0.2L_o to 8.33 ± 1.69 × 10⁵ dpm/g at 1.0L_o (Fig. 2B, ). The correlation between [³H]inositol phosphate production and muscle length was significant (P < 0.05). Comparison of Figs. 1B and 2B indicates that raising [carbachol] from 1 to 100 µM induced proportional increases in [³H]inositol phosphate production at all measured muscle lengths. As a result, the slopes of length dependence at 1 µM carbachol (1.32 ± 0.50 × 10⁵ dpm/L_o) and 100 µM carbachol (5.65 ± 1.98 × 10⁵ dpm/L_o) were not significantly different. Therefore, these results indicate that PI turnover remains length dependent at maximal [carbachol] values.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Concentration dependence of carbachol-induced active force at Lo (A) and length dependence of 100 µM carbachol-induced active force (B, ) and 30-min accumulation of total [3H]inositol phosphates (IP; B, ). Active force is expressed as fraction of either maximal force (A) or fraction of Fo induced by K+ depolarization at beginning of each experiment (B). Data are means ± SE; n = 5-9 tracheal rings. P values, significant correlation with muscle length.

Concentration dependence of carbachol-induced [³H]inositol phosphates production at 0.2 and 1.0L_o. To further test the hypothesis that mechanical strain may regulate maximal PI turnover coupled to muscarinic-receptor activation, we compared the concentration dependencies of carbachol-induced [³H]inositol phosphate production at 0.2 and 1.0L_o. As shown in Fig. 3A, carbachol-induced suprabasal [³H]inositol phosphate production was concentration dependent at both 0.2 and 1.0L_o, reaching maximal levels at 100 µM carbachol. The maximal [³H]inositol phosphate production (at 100 µM carbachol) was significantly higher at 1.0L_o (7.8 ± 1.7 × 10⁵ dpm/g) than at 0.2L_o (3.5 ± 0.6 × 10⁵ dpm/g). There was a significant difference between the 0.2 and 1.0L_o data points at matched [carbachol] values (Fig. 3A). At other [carbachol] values (1, 10 and 1,000 µM), [³H]inositol phosphate production was also significantly higher at 1.0L_o than at 0.2L_o. To compare the sensitivity of carbachol-induced PI turnover at 0.2 and 1.0L_o, we plotted the two sets of data as the percentage of their respective maximal values: PI turnover induced by a given [carbachol] was divided by the maximal PI turnover in concentration-response experiments. As shown in Fig. 3B, the two normalized concentration-response relationships at 0.2 and 1.0L_o were similar. Five of the six pairs of data points overlapped within 1 SE. Two-way ANOVA of the data indicated an insignificant difference between the two normalized concentration-response relationships at 0.2 and 1.0L_o. These results indicate that a length-dependent change in maximal PI turnover was necessary and sufficient to explain the 0.2 and 1.0L_o data sets.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   A: concentration dependence of carbachol-induced 30-min accumulation of total [3H]inositol phosphates at 0.2 and 1.0Lo. [Carbachol], carbachol concentration. B: normalized concentration dependence of carbachol-induced 30-min accumulation of [3H]inositol phosphates at 0.2 and 1.0Lo. Data are means ± SE; n = 5-6 tracheal rings. Data in A and B are the same except for normalization by maximal values in B. * Significant difference between the 2 groups.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   A: concentration dependence of fluoroaluminate-induced active force at Lo. Fluoroaluminate concentration was controlled by changing NaF concentration ([NaF]) at 10 µM AlCl3. B: length dependence of fluoroaluminate (5 mM NaF plus 10 µM AlCl3)-induced and carbachol-induced active force. C: length dependence of fluoroaluminate (5 mM NaF plus 10 µM AlCl3)-induced and carbachol-induced 30-min accumulation of total [3H]inositol phosphates. Data are means ± SE; n = 5-7 tracheal rings. P values, significant correlation between dependent and independent variables.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Airway smooth muscle cells in vivo constantly undergo cycles of shortening and lengthening during lung ventilation. Mechanical strain is known to modulate airway smooth muscle activation, but relatively little is known about how mechanical strain modulates signal transduction in airway smooth muscle cells. PI turnover catalyzed by phospholipase C is the major signal transduction mechanism coupled to muscarinic receptors (7, 24). Matsumoto et al. (20) reported direct activation of phospholipase C activity by mechanical stretch in otherwise unstimulated smooth muscle. However, stretch-induced phospholipase C activity returned to the control level rapidly (after 1.7 s) despite maintenance of mechanical strain. The transient nature of stretch-activated phospholipase C has also been observed in vascular smooth muscle cells (18, 26). These observations are consistent with the finding in this study that steady-state PI turnover in unstimulated airway smooth muscle is not significantly correlated with muscle length (Fig. 1).

Yoo et al. (27) investigated how mechanical strain modulates muscarinic receptor-coupled PI turnover. They found that PI turnover induced by 1 µM carbachol was significant at L_o but not at 10% L_o. This finding indicates that the level of muscarinic receptor-coupled PI turnover is not only dependent on [agonist] but is also dependent on the mechanical strain imposed on the airway smooth muscle. Two unanswered questions remained: 1) how does receptor-coupled PI turnover change with muscle length and 2) does mechanical strain modulate the maximal response (V_max) or the affinity of interaction (K_m) of the signal transduction cascade? These two questions were addressed in this study.

In this study, we measured PI turnover in airway smooth muscle by measuring total [³H]inositol phosphate accumulation in the presence of Li⁺ because muscarinic receptor-mediated [³H]inositol phosphate accumulation has been shown to be sustained for at least 30 min (5, 7, 27). We found that muscarinic receptor-mediated PI turnover was linearly dependent on muscle length at both 1 and 100 µM carbachol (Figs. 1 and 2). The observed linearity indicates that mechanical strain modulates receptor-mediated PI turnover in a graded manner. It is noteworthy that 100 µM carbachol is the maximal concentration for eliciting contraction in bovine tracheal smooth muscle (Fig. 2A). Therefore, the observed linear length dependence of PI turnover at 100 µM carbachol suggests that affinity of interaction (K_m) is unlikely to be the target of modulation of mechanical strain. This is because two concentration-response relationships different only in affinity (1/K_m) should converge to the same maximal value (V_max) at a maximal [agonist]. A more definitive test of this suggestion was done by measuring [carbachol]-PI turnover relationships at 0.2 and 1.0L_o (Fig. 3A). The half-maximal [carbachol] of ~3 µM found in this study is similar to that reported by Chilvers and Nahorski (7). A change in maximal PI turnover alone was necessary and sufficient to explain the two concentration-response relationships (Fig. 3B). These results indicate that maximal PI turnover coupled to muscarinic receptors is the target of modulation by mechanical strain in airway smooth muscle. This finding suggests that mechanical strain regulates the number of functional units in the cascade leading from muscarinic receptor to phospholipase C. Muscarinic receptors are coupled to phospholipase C via G proteins. Therefore, mechanical strain may regulate the number of receptors, G proteins, or phospholipase C enzymes on the cell membrane of an airway smooth muscle cell.

Fluoroaluminate activates heterotrimeric G proteins directly, thus bypassing the step of receptor activation (1, 25). Fluoroaluminate has been shown to activate PI turnover (15) and myosin light chain phosphorylation (13) in airway smooth muscle. In this study, we found that fluoroaluminate-induced PI turnover at a maximal [fluoroaluminate] remains linearly dependent on muscle length (Fig. 4). This finding suggests that mechanical strain regulates the number of functional G proteins and/or phospholipase C enzymes in the cell membrane. Mechanistically, this suggests that mechanical strain regulates the compartmentalization of G proteins and/or phospholipase C in airway smooth muscle cells. Possible molecular mechanisms include mechanosensitive modulation of 1) formation of caveoli where signal transduction molecules are concentrated (1) and 2) translocation of phospholipase C between the cytosol and cell membrane (9, 10, 19, 23). Receptor-mediated signal transduction in airway smooth muscle is often presented as a unidirectional cascade. The findings from this and other studies (21, 27) indicate that mechanical strain (cell length) is an important feedback modulator of signal transduction in airway smooth muscle. If this feedback mechanism is understood at the molecular level, it may offer a new approach to control airway smooth muscle contractility and therefore airway resistance.

Phosphorylation of the 20,000-Da myosin light chain is the central regulatory mechanism of contractile filaments in smooth muscle (16). It may be informative to compare the effects of mechanical strain on the upstream process of PI turnover with the downstream process of myosin light chain phosphorylation. Recently, Youn et al. (28) reported that mechanical strain regulates maximal myosin light chain phosphorylation in airway smooth muscle. Hai and Ma (13) found that maximal fluoroaluminate-induced myosin phosphorylation is also linearly dependent on muscle length. The striking similarity in the dependencies of muscarinic receptor-mediated PI turnover and myosin phosphorylation on muscle length suggests that signal transduction may be the primary target of modulation by mechanical strain. That is, length-dependent changes in myosin phosphorylation may result from length-dependent changes in signal transduction at the cell membrane. IP₃ is known to induce Ca²⁺ release from the sarcoplasmic reticulum in airway smooth muscle (24). Due to the finite amount of phosphoinositide and sarcoplasmic reticular Ca²⁺, IP₃ is important in the initiation but not in the maintenance of contraction in smooth muscle. The signal transduction that regulates force maintenance in smooth muscle is not fully understood, although protein kinase C activation by diacylglycerol has been implicated. The observed pattern of how mechanical strain modulates PI turnover in this study suggests the possibility that signal transduction mechanisms involved in force maintenance may be similarly regulated by mechanical strain. In theory, the results may also be explained by postulating that phospholipase C activity is regulated by intracellular Ca²⁺ concentration. However, Grandordy et al. (11) and Chilvers et al. (6) found that K⁺ depolarization failed to activate PI turnover in airway smooth muscle. Another possibility is that multiple steps in the activation-contraction cascade are all linearly dependent on muscle length. The observed length dependence of K⁺ depolarization-induced myosin phosphorylation (14) appears to support this alternative hypothesis. If this alternative hypothesis is correct, then mechanosensitive feedback must be critical to the function of airway smooth muscle cells so that redundant mechanosensitive feedback is built into each level of the activation-contraction cascade. One possible function of mechanosensitive feedback is to prevent airway collapse at small-airway diameter when the mechanical advantage of transmural pressure is reduced as predicted by Laplace's law.