Five distinct muscarinic receptor subtypes have been identified, and each subtype is encoded by distinct cellular genes. Muscarinic receptor sequences, m1-m5, encode the pharmacological receptor subtypes M1-M5. Previous investigations suggest that there are at least M1, M2, and M3 muscarinic receptors in the human airway (1). M1 receptors are found in airway ganglia and alveolar walls and facilitate neurotransmission through parasympathetic ganglia and enhance cholinergic reflexes. M2 receptors are localized to cholinergic postganglionic nerves at the prejunctional level and inhibit acetylcholine release but counteract β-agonist-induced bronchodilation. M3 receptors are found on airway smooth muscles, submucosal glands, and airway vascular endothelium. Activation of M3 receptors produces airway smooth muscle contraction via phosphoinositide hydrolysis (2).
We have previously reported that propofol (3) and ketamine (4) inhibit histamine-induced bronchoconstriction in vivo, but thiopental (5) worsens this response. Propofol also reduces basal airway tone, and pretreatment with atropine abolishes these propofol effects (3). In addition, we recently reported that clinically relevant plasma concentrations of propofol antagonize methacholine-induced bronchoconstriction in dogs with and without vagotomy (6). In contrast, thiopental increases basal tone, and this is completely antagonized by atropine (5). In vitro/ ex vivo reports showed similar results. We reported that ketamine (7,8) and propofol (9) inhibited contraction of guinea pig tracheal smooth muscle induced by carbachol. Lenox et al. (10,11) showed that thiopental directly produces airway smooth muscle contraction, although they suggest that the contraction may be because of the production of thromboxane A2(11). Using isolated porcine tracheal smooth muscle, Wilson et al. (12) reported that the relaxant effects of ketamine might be caused by decreasing the excitability of postsynaptic nicotinic receptors and affecting muscarinic receptors, smooth muscle, or a combination of these. These in vivo and in vitro reports suggest that the effects of IV anesthetics on airway smooth muscles may be caused by modulation of muscarinic receptors.
In the present study, we examined if IV anesthetics propofol, thiopental, and ketamine interact with recombinant human M1-M3 muscarinic receptors expressed in Chinese hamster ovary cells.
Materials and Methods
CHO-M1, M2, and M3 cells (13) were maintained in α Minimal Essential Medium supplemented with 100 IU/mL of penicillin, 100 μg/mL of streptomycin, 2.5 μg/mL of Fungizone, and 10% fetal calf serum. Cultures were maintained at 37°C in 5% CO2/humidified air at 37°C, fed every 2–3 days, and passaged when confluent. All cells were harvested for use by the addition of 0.9% saline containing HEPES (10 mM)/EDTA (0.02%). Cells were homogenized at 4°C using an Ultra-Turrax (T25; IKA Labortechnik, Staufen, Germany) in assay buffer (20 mM of HEPES and 1 mM MgCl2 at pH 7.4). The homogenate was centrifuged at 20,000 g for 10 min and the pellet resuspended in assay buffer. This procedure was repeated two more times. Membranes of each cell line were prepared and used fresh daily.
The binding of l-[N-methyl-3H]scopolamine methyl chloride ([3H]NMS) was performed in 1-mL volumes of the assay buffer containing approximately 100 μg of membranes at 37°C for 60 min. Nonspecific binding was defined in the presence of 10−5 M atropine. After incubation, each sample was filtered (and washed) under vacuum through Whatman GF/B filters (Semat, UK) using a Brandel cell harvester. Filter retained radioactivity was extracted for at least 8 h in 4 mL of scintillation fluid. In displacement studies, the interaction of thiopental (10−6–10−3 M), propofol (10−6–10−3 M), ketamine (10−5–10−2 M), and a nonanesthetic barbiturate, barbituric acid, used as a negative control (10−6–10−3 M) with M1-M3 muscarinic receptors was determined by displacement of 0.4 nM [3H]NMS (n = 5–10). Atropine (10−11–10−5 M;n = 5) was included as a reference compound. The concentration of displacers producing 50% displacement of specific binding (IC50; half-maximal inhibitory concentration) was obtained by computer assisted curve fitting (nonlinear regression; sigmoid concentration-response curve, variable slope with bottom and top set at 0% and 100%, respectively, using GraphPad-PRISM, GraphPad Inc, San Diego, CA) and corrected for the competing mass of [3H]NMS using Kds of 0.11, 0.15, and 0.12 nM for [3H]NMS at M1, M2, and M3 receptors, respectively (13), according to Cheng and Prusoff (14) to yield Ki: Ki = IC50/[1+([3H]NMS used/Kd)].
Intracellular Ca2+ ([Ca2+]i) was measured, as described previously (15). Briefly, confluent CHO-M1-M3 cells were harvested and resuspended in Krebs/HEPES buffer, with pH 7.4. The cells were then loaded with fura2-AM (3 μM) for 30 min at 37°C, followed by 20 min postincubation at room temperature to allow complete ester hydrolysis. Finally, [Ca2+]i was measured fluorometrically in a Perkin-Elmer LS50B fluorometer with excitation wavelengths set at 340 and 380 nm and emission set at 510 nm. [Ca2+]i was calculated from the 340:380 ratio, according to Grynkiewicz et al. (15), where Rmax and Rmin were determined using Triton ×100 (0.1% vol/vol) and EGTA (4.5 mM; pH >8.0). Mean Rmax and Rmin values were 6.25 (95% confidence interval, 5.79 to 6.71) and 0.66 (95% confidence interval, 0.64 to 0.68), respectively. IV anesthetics showing significant receptor interaction in binding studies were preincubated (at 10−5–10−3 M) 15 min before challenge with the nonselective muscarinic agonist methacholine (1 mM).
Data are expressed as mean ± sem and analyzed statistically with analysis of variance. P < 0.05 was considered significant.
Atropine and ketamine concentration dependently displaced [3H]NMS binding to CHO-M1, M2, and M3 cells with pKi (mean) values of 9.36 ± 0.04 (0.43 nM) and 4.34 ± 0.14 (45 μM), 9.16 ± 0.08 (0.69 nM) and 3.53 ± 0.10 (294 μM), and 8.40 ± 0.08 (4.0 nM) and 3.61 ± 0.02 (246 μM), respectively (Fig. 1A–C). Thiopental significantly interacted with M3 (pKi [Ki] = 4.12 ± 0.06 [75 μM]) but not M1 or M2 receptors. Propofol and barbituric acid did not interact with any muscarinic receptor (Fig. 1A–C).
Because there was an interaction at clinically achievable concentrations for ketamine at M1 receptors, we measured [Ca2+]i. However, ketamine did not affect either basal or methacholine stimulated increase in [Ca2+]i in these cells. Similar experiments were performed in M3 cells with thiopental that concentration dependently inhibited methacholine stimulated increase in [Ca2+]i (Fig. 2) such that at 1 mM, >75% inhibition was observed.
Radioligand binding assays indicate that clinically achievable concentrations of thiopental and ketamine interact with M3 and M1 muscarinic receptors, respectively, when heterologously expressed in CHO cells. Propofol and barbituric acid were inactive. In addition, thiopental reduced the methacholine-induced increase in [Ca2+]i in CHO-M3 cells. Surprisingly, and despite a significant interaction at the M1-receptor, ketamine did not affect either the basal or methacholine-induced increase in [Ca2+]i in CHO-M1 cells.
Muscarinic receptor-second messenger coupling is subtype selective with M1 and M3 coupling to phospholipase C to generate inositol(1,4,5)triphosphate resulting in an increase in [Ca2+]i and with M2 coupling to adenylate cyclase to decrease intracellular cyclic adenosine monophosphate formation. Taken at face value, our data suggest that thiopental may act as a M3-antagonist, and because of the lack of interaction at M1 and M2, this may be selective. A caveat to this interpretation is that we have not used M4 and M5 (as these do not occur in the airway) and that the interaction is weak (when compared with atropine) but nevertheless clinically achievable. Ketamine interaction at the M1 receptor (in the clinical range) seems to be of no direct functional consequence, and we have no explanation for this puzzling result. This is particularly puzzling in that we used up to 1 mM ketamine in our functional experiments.
We have previously reported that thiopental inhibits carbachol-evoked [3H]norepinephrine release and increase in [Ca2+]i from SH-SY5Y human neuroblastoma cells (16). This response is driven by the native M3 muscarinic receptor, and in a series of parallel [3H]NMS binding experiments, thiopental displaced [3H]NMS with an estimated pKi of 3.56 (compared with 4.12 in this study using CHO cells), essentially confirming this result.
As activation of M3 receptors located in the airway produces contraction of airway smooth muscle, thiopental should inhibit the contraction as an M3 antagonist. Indeed, several reports (17,18) showed that thiopental inhibits acetylcholine-induced contraction of guinea pig tracheal smooth muscles. However, we previously reported that thiopental increases basal airway tone, which is abolished by atropine pretreatment in dogs. Moreover, Brown and Wagner (19) also demonstrated that local infusion of thiopental into the bronchial artery did not attenuate methacholine-induced bronchoconstriction in sheep. These data may indicate a further species effect in interpretation of our results.
In the present study, at least at the level of [Ca2+]i, ketamine (and propofol) do not have any functional activity at M1-M3 receptors, and it is therefore likely that this inactivity may also occur in the airway. However, previous studies showed that propofol and ketamine inhibited carbachol-induced contraction of tracheal smooth muscle (7–9). Brown and Wagner (19) also reported that clinical and supraclinical infusions of propofol into the bronchial artery significantly decreased neurally mediated- and methacholine-induced bronchoconstriction, respectively. Ketamine only decreased neurally mediated constriction. We cannot explain the discrepancy between the present data and the previous in vitro and in vivo data. However, we suggest that this interaction does not occur directly at the level of the receptor but may be downstream of receptor activation, although we have no experimental evidence to support this suggestion. In the absence of a simple receptor based-explanation, further studies will be required to determine the mechanism by which these IV anesthetics exert effects on airway tone.
We thank Professor NJM Buckley (University of Leeds, UK) for providing CHO-M1, M2, and M3 cells.
1. Gal TJ. Bronchial hyperresponsiveness and anesthesia: physiologic and therapeutic perspectives. Anesth Analg 1994; 78: 559–73.
2. Barnes PJ. Muscarinic receptor subtypes in airways. Life Sci 1993; 52: 512–27.
3. Hirota K, Sato T, Hashimoto Y, et al. Relaxant effect of propofol on the airway in dogs. Br J Anaesth 1999; 83: 292–5.
4. Hirota K, Hashimoto Y, Sakai T, et al. In vivo
spasmolytic effect of ketamine and adrenaline on histamine-induced airway constriction: direct visualization method with a superfine fibreoptic bronchoscope. Acta Anaesthesiol Scand 1998; 42: 184–8.
5. Hirota K, Ohtomo N, Hashimoto Y, et al. The effects of thiopental on airway calibre in dogs: direct visualization method using a superfine fibreoptic bronchoscope. Br J Anaesth 1998; 81: 203–7.
6. Kabara S, Hirota K, Hashiba E, et al. Comparison of relaxant effects of propofol on methacholine-induced bronchoconstriction in dogs with and without vagotomy. Br J Anaesth 2001; 86: 249–53.
7. Sato T, Hirota K, Matsuki A, et al. Ketamine inhibits the tonic response to carbachol and histamine in the guinea pig trachea. Eur J Anaesthesiol 1998; 15: 486–92.
8. Sato T, Hirota K, Matsuki A, et al. The relaxant effect of ketamine on guinea pig airway smooth muscle is epithelium-independent. Anesth Analg 1997; 84: 641–7.
9. Hashiba E, Sato T, Hirota K, et al. The relaxant effect of propofol on guinea pig tracheal muscle is independent of airway epithelial function and beta-adrenoceptor activity. Anesth Analg 1999; 89: 191–6.
10. Lenox WC, Mitzner W, Hirshman CA. Mechanism of thiopental-induced constriction of guinea pig trachea. Anesthesiology 1990; 72: 921–5.
11. Curry C, Lenox WC, Spannhake EW, Hirshman CA. Contractile responses of guinea pig trachea to oxybarbiturates and thiobarbiturates. Anesthesiology 1991; 75: 679–83.
12. Wilson LE, Hatch DJ, Rehder K. Mechanisms of the relaxant action of ketamine on isolated porcine trachealis muscle. Br J Anaesth 1993; 71: 544–50.
13. Cembala TM, Sherwin JD, Tidmarsh MD, et al. Interaction of neuromuscular blocking drugs with recombinant human m1–m5 muscarinic receptors expressed in Chinese hamster ovary cells. Br J Pharmacol 1998; 125: 1088–94.
14. Cheng YC, Prusoff WM. Relationship between the inhibition constant (Ki
) and the concentration of inhibitor which causes 50% inhibition (IC50
) of an enzymatic reaction. Biochem Pharmacol 1973; 22: 3099–108.
15. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+
indicators with greatly improved fluorescence properties. J Biol Chem 1985; 260: 3440–9.
16. Lambert DG, Willets JM, Atcheson R, et al. Effects of propofol and thiopentone on potassium- and carbachol-evoked [3
H]noradrenaline release and increased [Ca2+
from SH-SY5Y human neuroblastoma cells. Biochem Pharmacol 1996; 51: 1613–21.
17. Fletcher SW, Flacke W, Alper MH. The actions of general anesthetic agents on tracheal smooth muscle. Anesthesiology 1968; 29: 517–22.
18. Okumura F, Denborough MA. Effects of anesthetics on guineapig tracheal smooth muscle. Br J Anaesth 1980; 52: 199–204.
19. Brown RH, Wagner EM. Mechanisms of bronchoconstriction by anesthetic induction agents: propofol versus ketamine. Anesthesiology 1999; 90: 822–8.