Effects of Intravenous Anesthetics on Ca2+ Sensitivity in Canine Tracheal Smooth Muscle

After approval by the Institutional Animal Care and Use Committee, mongrel dogs (20–25 kg) of either sex were anesthetized using an intravenous injection of sodium pentobarbital (50 mg/kg) and were killed by exsanguination. A 10- to 15-cm portion of extrathoracic trachea was excised and immersed in chilled physiologic salt solution of the following composition: NaCl: 110.5 mM; NaHCO³: 25.7 mM; dextrose: 5.6 mM; KCl: 3.4 mM; CaCl²: 2.4 mM; KH²PO⁴: 1.2 mM; MgSO⁴: 0.8 mM. The adventitia and mucosa were removed after cutting the visceral side of the cartilage, then connective tissues were carefully removed during microscopic observation.

Muscle strips (width, 0.1–0.2 mm; wet weight, 0.05–0.1 mg) were mounted in 0.1-ml glass cuvettes and continuously superfused at 1.2 ml/min with physiologic salt solution (37°C) aerated with 94% oxygen and 6% carbon dioxide, providing a pH of ≈ 7.4, a partial pressure of oxygen of ≈ 400 mmHg, and a partial pressure of carbon dioxide of ≈ 39 mmHg in the physiologic salt solution. One end of the muscle strips was anchored with stainless steel microforceps to a stationary metal rod, and the other end was anchored to a calibrated force transducer (model KG4; Scientific Instruments, Hedelberg, Germany). The initial gap between microforceps (i.e.,  initial muscle length) was set at 5 mm. During a 3-h equilibration period, the length of the muscle strips was increased after subsequent isometric contractions (of 1-min duration) induced by 1 μM acetylcholine until isometric force was maximal (optimal length). Each muscle strip was maintained at this optimal length for the remainder of the experiment. These tissues produced maximal isometric forces of 1–3 mN when stimulated with 1 μM acetylcholine.

The muscle strips were permeabilized with β-escin, as previously described, 13and validated in the Mayo Clinic and Mayo Foundation laboratory. 1β-escin creates pores in the smooth muscle cell plasma membrane, thus allowing substances of small molecular weight, such as Ca²⁺, to freely diffuse across the cell membrane. Accordingly, [Ca²⁺]ⁱcan be manipulated and controlled by changing the concentration of Ca²⁺in the solution that bathes the smooth muscle strip. Larger cellular proteins necessary for contraction are preserved. Additionally, the membrane receptor–coupled mechanisms that modulate Ca²⁺sensitivity remain intact and can be activated. 1 

After optimal length was set, subsequent experimental protocols were performed at room temperature (25°C) and without aeration of the solutions. Muscle strips were superfused for 20 min, with a relaxing solution that contained 100 μM β-escin. The relaxing solution was made up in the following composition using the algorithm of Fabiato and Fabiato:147.5 mM MgATP, 4 mM ethylene glycol-bis (β-aminoethyl ether)-N,N,N′N′-tetraacetic acid (EGTA), 20 mM imidazole, 1 mM dithiothreitol, 1 mM free Mg²⁺, 1 nM free Ca²⁺, 10 mM creatine phosphate, and 0.1 mg/ml creatine phosphokinase. Ionic strength was kept constant at 200 mM by adjusting the concentration of potassium acetate. The pH was adjusted to 7.0 at 25°C with potassium hydroxide. After the permeabilization procedure, strips were superfused with the relaxing solution for 10 min to wash out the excess β-escin. The calcium ionophore A23187 (10 μM) was added to the relaxing solution and all subsequent experimental solutions to deplete the sarcoplasmic reticulum Ca²⁺stores and maintain [Ca²⁺]ⁱ. 1Solutions of varying free Ca²⁺concentrations used in the subsequent experiment also were prepared using the aforementioned algorithm. 14 

Experiments with intact strips were designed to determine what concentrations of drugs should be used in subsequent studies of permeabilized strips. This protocol was performed at 37°C in physiologic salt solution. Muscle strips were contracted for 10 min with 0.03 μM acetylcholine, which produced approximately 50% of the maximal force induced by 100 μM acetylcholine in preliminary studies (data not shown). After stable contractions were obtained, cumulative doses of ketamine (10⁻⁷− 10⁻³M), propofol (10⁻⁶− 10⁻³M), or midazolam (10⁻⁷− 1.8 × 10⁻⁴M) were applied, and concentration–response curves for these anesthetics were generated. The effect of 10% intravenous fat emulsion, similar to the vehicle for propofol, also was evaluated. The effect of halothane on intact CTSM was reported in a previous study. 15 

Experiments with permeabilized strips determined the effects of anesthetics on the Ca²⁺sensitivity in the absence and the presence of muscarinic receptor stimulation. A set of four β-escin permeabilized strips was prepared from the same dog for each experiment. All strips first were contracted maximally with 10 μM Ca²⁺; all subsequent force measurements were normalized to these contractions. Strips were superfused with relaxing solution that contained 5 mM inorganic phosphate for 10 min and then superfused with relaxing solution for 10 min to remove inorganic phosphate. Inorganic phosphate reduces the time necessary for relaxation by accelerating the rate of actomyosin cross-bridge detachment. 16Each set of four strips was studied as two pairs. All four strips were contracted with 0.3 μM Ca²⁺for 10 min. One pair was then stimulated with 10 μM acetylcholine and 10 μM guanosine 5′-triphosphate; the other pair continued to be exposed to Ca²⁺alone. After 10 min, one of four drugs (one drug for each set of muscles) was added to one strip of each pair for 15 min. For ketamine, propofol, and midazolam, an approximate EC⁸⁰concentration was used (200, 270, and 100 μM, respectively). For halothane, a concentration equivalent to approximately 3 MAC (for dogs, corrected to room temperature 17) was used. The remaining strip of each pair was not exposed to drugs and served as a time control. In experiments with propofol, an equivalent volume of the vehicle 10% intravenous fat emulsion was added at the appropriate time for the time-control strips.

Halothane was delivered to solutions via  a calibrated vaporizer. Each solution was equilibrated with halothane for at least 5 min (enough time to become equilibrated in our system) before being introduced to the superfusion system. At the end of the protocol, the concentrations of halothane in the solutions at the cuvette were determined by gas chromatography from anaerobically obtained samples using an electron capture detector (model 5880A; Hewlett-Packard, Waltham, MA) according to the method of Van Dyke and Wood. 18 

Halothane was purchased from Ayerst Laboratories (New York, NY). Adenosine 5′-triphosphate, disodium salt was purchased from Research Organics (Cleveland, OH). Propofol was purchased from Zeneca Pharmaceuticals (Wilmington, DE). Intravenous fat emulsion, 10%, was purchased from Baxter Health Care (Deerfield, IL). All other drugs and chemicals were purchased from Sigma Chemical (St. Louis, MO). The calcium ionophore A23187 was dissolved in dimethyl sulfoxide (0.05% final concentration). Propofol was diluted from an aqueous emulsion (5.6 × 10⁻²M) in 10% (wt/vol) soybean oil, 2.25% glycerol, and 1.2% purified egg lecithin. All other drugs and chemicals were prepared in distilled filtered water.

Data are expressed as the mean ± SD; n represents the number of dogs. Concentration–response curves for ketamine, propofol, and midazolam were fitted by nonlinear regression (SigmaPlot for Windows Version 4.00; SPSS Inc., Chicago, IL) and used to calculate EC⁸⁰values.

In the first protocol, relaxation produced by anesthetics was expressed as a percent change from the initial force produced by 0.03 μMacetylcholine. In the second protocol, forces were expressed as a percentage of the maximal force induced by 10 μM Ca²⁺determined in each individual strip before the experimental protocol. The decrease in force produced by anesthetics was expressed as a percent relaxation from initial force (before exposure to drug). Initial force was adjusted for the effect of time by using the change in force of the time-matched control strip in each set according to the following formula: where A¹is the force of the anesthetic-exposed strip just before exposure to anesthetic, A² is the force of the anesthetic-exposed strip at the end of anesthetic exposure, C¹and C² are the forces of the control strip at the matched times with A¹and A², respectively. All force values represent a change from baseline force (i.e.,  in relaxing solution). Statistical assessments were made using the paired t  test. A P  value < 0.05 was considered to be statistically significant.

Ketamine, propofol, and midazolam all produced concentration-dependent relaxation of intact CTSM strips contracted with acetylcholine (figs. 1A–C). EC⁸⁰values for this effect were 199, 274, and 99 μM for ketamine, propofol, and midazolam, respectively. The vehicle for propofol, 10% intravenous fat emulsion, did not affect contraction (data not shown).

In β-escin–permeabilized tissue, 0.3 μM Ca²⁺induced stable contractions of 7.4 ± 5.5% (n = 80 strips) of the maximal force induced by 10 μM Ca²⁺(fig. 2). The subsequent addition of 10 μM acetylcholine and 10 μM guanosine 5′-triphosphate further increased the force to 38.7 ± 10.4% of maximal force after 10 min, indicating an increase in Ca²⁺sensitivity produced by acetylcholine and guanosine 5′-triphosphate. None of the three intravenous anesthetics studied significantly changed force, either in the presence or the absence of receptor stimulation with acetylcholine (figs. 2 and 3). Consistent with our previous work, 1halothane (0.75 ± 0.05 mM) significantly reduced force in strips stimulated with acetylcholine (by 27.8 ± 13.4%) but did not change force in strips stimulated only with Ca²⁺(figs. 2 and 3).

The major finding of this study in β-escin–permeabilized CTSM is that ketamine, propofol, or midazolam did not affect Ca²⁺sensitivity, either in the presence or the absence of muscarinic receptor stimulation. Therefore, the effect of halothane to inhibit agonist-induced increases in Ca²⁺sensitivity, confirmed in this study, is not a general feature of drugs with anesthetic properties.

Many anesthetic agents act as bronchodilators in vivo . 19,20These drugs depress neural transmission in pathways that innervate airway smooth muscle and also directly affect the smooth muscle cell. In general, these direct effects are caused by a decrease in [Ca²⁺]ⁱ, a decrease in the force maintained for a particular [Ca²⁺]ⁱ(i.e.,  the Ca²⁺sensitivity), or a combination of both mechanisms. It is clear that during submaximal contraction of airway and other types of smooth muscle, anesthetic-induced relaxation is associated with a decrease in [Ca²⁺]ⁱ. This is true for the volatile anesthetics 3,21,22and for the three intravenous anesthetics evaluated in our study: ketamine, 7,10propofol, 10and midazolam. 11For these intravenous anesthetics, decreases in [Ca²⁺]ⁱare caused by inhibition of Ca²⁺influx through L-type voltage-operated Ca²⁺channels. 6,10Although volatile anesthetics may affect intracellular Ca²⁺stores, 3these intravenous anesthetics do not appear to do so in studies to date. 9,11The barbiturates, the other major class of intravenous induction agents used clinically, also may depress airway neural transmission in intact animals but, in most recent studies, do not affect or actually cause contraction of isolated airway smooth muscle. 23,24In preliminary studies, we found little effect of thiopental on canine airway smooth muscle (data not shown) and therefore elected not to study this agent further.

Although it is clear that anesthetics relax airway smooth muscle in part by decreasing [Ca²⁺]ⁱ, inhibition of Ca²⁺sensitivity also may be important. We evaluated Ca²⁺sensitivity by using a β-escin–permeabilized smooth muscle preparation. The creation of pores in the plasma membrane permits the manipulation of the intracellular environment by changing the composition of the fluid that bathes the smooth muscle, such that Ca²⁺sensitivity may be directly studied. 1,13Contraction of this preparation is produced by exposing it to Ca²⁺; exposure to muscarinic receptor agonists produces additional force (fig. 2). The binding of Ca²⁺to calmodulin increases myosin light chain kinase activity and phosphorylation of the 20-kd regulatory myosin light chain (rMLC). 25,26Regulatory myosin light chain phosphorylation allows the binding of myosin to actin, which increases actomyosin adenosine triphosphatase activity and causes contraction. 2Muscarinic receptor stimulation activates a cascade of G proteins that inhibit smooth muscle protein phosphatases, increasing regulatory myosin light chain phosphorylation and, thus, force. 27,28A previous study from our laboratory showed that permeabilization of CTSM with β-escin eliminates Ca²⁺gradients across the sarcolemma and maintains [Ca²⁺]ⁱduring muscarinic receptor stimulation and that the coupling of membrane receptors to mechanisms that increase Ca²⁺sensitivity remain intact and can be activated. 1 

We confirmed our previous finding that halothane significantly decreases Ca²⁺sensitivity during muscarinic receptor stimulation of the permeabilized preparation exposed to a submaximal concentration of Ca²⁺. 1However, none of the three intravenous anesthetics studied (at approximate ED⁸⁰concentrations for relaxation of intact muscle) significantly affected the force produced by constant submaximal [Ca²⁺]ⁱ, either in the presence or the absence of muscarinic receptor stimulation. Yoshimura et al.  11also found no effect of midazolam on Ca²⁺sensitivity measured in an intact CTSM preparation. Taken together with previous studies of the effects of these drugs on [Ca²⁺]ⁱ, we conclude that, at concentrations producing near-maximal effect, these drugs relax CTSM exclusively by decreasing [Ca²⁺]ⁱ, without affecting Ca²⁺sensitivity. Although comparisons should be made with caution, these ED⁸⁰concentrations of drugs considerably exceed the reported plasma concentrations that produce surgical anesthesia in humans (approximately 2– 60 μM, 2920–50 μM, 30,31and 1–2 μM 32for ketamine, propofol, and midazolam, respectively). When the fact that these drugs are significantly protein-bound (> 95% for propofol and midazolam) 30,31,33is also taken into account, it is almost certain that they would not affect Ca²⁺sensitivity as clinically used.

In conclusion, three intravenous anesthetics (ketamine, propofol, and midazolam) directly relax canine airway smooth muscle stimulated with a muscarinic agonist without affecting Ca²⁺sensitivity. Thus, inhibition of agonist-induced increases in Ca²⁺sensitivity of CTSM is not a common property of anesthetics but is unique to the volatile agents.

The authors thank Dr. Tetsuya Kai, Department of Anesthesiology and Critical Care Medicine, Faculty of Medicine, Kyushu University, Fukuoka, Japan, for helpful suggestions and Darell L. Loeffler for measuring anesthetic concentration by gas chromatography.