Local anesthetics inhibit lipid mediator signaling (lysophosphatidate, thromboxane) by acting on intracellular domains of the receptor or on the G protein. On receptors for polar agonists, the ligand-binding pocket could form an additional site of interaction, possibly resulting in superadditive inhibition. The authors therefore investigated the effects of local anesthetics on m1 muscarinic receptor functioning.
The authors expressed receptors in isolation using Xenopus oocytes. Using a two-electrode voltage clamp, the authors measured the effects of lidocaine, QX314 (permanently charged), and benzocaine (permanently uncharged) on Ca2+-activated Cl- currents elicited by methylcholine. The authors also characterized the interaction of lidocaine with [3H] quinuclydinyl benzylate ([3H]QNB) binding to m1 receptors.
Lidocaine inhibited muscarinic signaling with a half-maximal inhibitory concentration (IC50 18 nm) 140-fold less than that of extracellularly administered QX314 (IC50 2.4 microm). Intracellularly injected QX314 (IC50 0.96 mm) and extracellularly applied benzocaine (IC50 1.2 mm) inhibited at high concentrations only. Inhibition of muscarinic signaling by extracellularly applied QX314 and lidocaine was the result of noncompetitive antagonism. Intracellularly injected QX314 and benzocaine inhibited muscarinic and lysophosphatidate signaling at similar concentrations, suggesting an action on the common G-protein pathway. Combined administration of intracellularly injected (IC50 19 microm) and extracellularly applied QX314 (IC50 49 nm) exerted superadditive inhibition. Lidocaine did not displace specific [3H]QNB binding to m1 receptors.
m1 Muscarinic signaling is inhibited by clinically relevant concentrations of lidocaine and by extracellularly administered QX314, suggesting that the major site of action is a extracellular domain of the muscarinic receptor. An additional less potent but superadditive inhibitory effect on the G-protein is suggested.
ALTHOUGH the sodium channel is usually considered the primary site of action of local anesthetics, several other sites of interaction exist that may be of relevance to local anesthetic effects and side-effects. 1Previous investigations from our laboratory have shown that local anesthetics inhibit several G-protein–coupled receptors. Lysophosphatidate (LPA) signaling is inhibited by lidocaine and bupivacaine. 2Clinically relevant concentrations of lidocaine, bupivacaine, or ropivacaine inhibit thromboxane A2receptor functioning. 3In contrast, local anesthetics do not affect the angiotensin II1Areceptor, another member of the G-protein coupled receptor superfamily, which couples, as we demonstrated, to the same intracellular signaling pathway downstream of the G-protein as does the LPA receptor. 2This suggests that the site of local anesthetic action is the G-protein or the receptor itself rather than the intracellular signaling pathway. These findings were confirmed by experiments demonstrating lack of local anesthetic effect on signaling induced by direct activation of G proteins or inositoltrisphosphate (IP3) receptors. 2,4On lipid mediator receptors, the main local anesthetic binding site appears to be intracellular because the permanently charged and therefore membrane-impermeant lidocaine analog QX314 inhibited lipid mediator receptor functioning only when injected intracellularly. 3,4Together, these findings suggest that local anesthetics act on the intracellular domains of lipid mediator receptors or on the associated G proteins. The latter seems more likely because intracellular QX314 shows a virtually identical concentration–response relation on LPA and thromboxane A2signaling 4(Hoenemann CW and Durieux ME, unpublished observation), suggesting the common G-protein, rather than the diverse receptors as site of action.
However, receptors for more polar compounds have a polar agonist binding pocket, and it is conceivable that local anesthetics interact with such receptors at this extracellular site as well. Several groups have demonstrated interactions of local anesthetics with muscarinic receptor binding, 5–10although such interactions usually necessitated high (supraclinical) local anesthetic concentrations. If such receptors contained an extracellular local anesthetic binding domain in addition to the intracellular domain described previously, superadditive interaction would be possible. If so, functional inhibition might necessitate much lower concentrations of local anesthetics than those necessitated in the binding studies. To test this hypothesis, we studied the effect of lidocaine on m1 muscarinic receptor signaling.
In addition to providing a suitable model for testing our hypothesis, muscarinic receptors are interesting subjects of anesthetic mechanisms research for several other reasons. First, as mentioned previously, local anesthetics have been shown 5–10to interact with these receptors. However, those studies usually investigated binding, not function. In addition, the specific receptor subtypes involved were rarely determined. Second, we previously demonstrated that volatile anesthetics 11,12and ketamine 13,14interact significantly with muscarinic signaling, and it would be of interest to compare the actions of local and general anesthetics in a similar model. Third, muscarinic signaling has major clinical implications for anesthesiologists. The m1 receptor is the best defined of all five muscarinic acetylcholine receptors. It is primarily found in the brain and is linked to phosphatidylinositol metabolism. m1 And m3 muscarinic receptors are largely responsible for maintenance of airway tone, and local anesthetics are known to be effective in preventing bronchospasm. m2 Muscarinic receptors are primary mediators of vagal tone in the heart. Muscarinic acetylcholine signaling also plays important roles in the central nervous system (CNS). Level of consciousness is modulated significantly by brain stem muscarinic signaling. 15Inhibition of muscarinic signaling (by reducing acetylcholine levels, inhibiting its release, or administering scopolamine) decreases minimum alveolar concentration of inhaled anesthetics. In contrast, physostigmine administration increases minimum alveolar concentration 16and reverses the action of propofol on the CNS. 17
Our study was designed to address the following questions:
Do clinically relevant concentrations of local anesthetics inhibit m1 muscarinic receptor signaling?
Is the inhibition competitive or noncompetitive?
Is the site of action located extracellular, intracellular, or both?
How much do interactions with the binding site contribute to the inhibitory effect?
Material and Methods
Oocyte Experiments
Oocyte Harvesting.
The study protocol was approved by the Animal Research Committee at the University of Virginia, Charlottesville, Virginia. Female Xenopus laevis toads were obtained from Xenopus I (Ann Arbor, MI), housed in an established frog colony, and fed regular frog brittle twice weekly. Surgery for oocyte harvesting was performed once every 2 months at most. Animals were killed after six operations. For removal of oocytes, a frog was anesthetized by immersion in 0.2% 3-amino-benzoic-methyl-ester until it was unresponsive to a painful stimulus (toe pinching). Surgery was performed while animals were positioned on ice. A 1-cm long incision was made in a lower abdominal quadrant, and a lobule of ovarian tissue containing approximately 200 oocytes was removed and placed in modified Barth solution (NaCl 88 mm, KCl 1 mm, NaHCO32.4 mm, CaCl20.41 mm, MgSO40.82 mm, Ca2NO30.3 mm, Gentamicin 0.1 mm, and HEPES 15 mm, pH adjusted to 7.6). The oocytes were defolliculated by gentle shaking for approximately 2 h in calcium-free OR2 solution (0.1% collagenase type Ia, 82.5 mm NaCl, 2 mm KCl, 1 mm MgCl2, and 5 mm HEPES, pH adjusted to 7.5). The oocytes were then returned to the modified Barth solution, and microscopic observation confirmed the absence of follicle cells.
Competitive RNA Synthesis and Injection.
The rat m1 muscarinic acetylcholine receptor complementary DNA (cDNA) was obtained from Dr. T. I. Bonner (National Institute of Mental Health, Bethesda, MD). It consists of a 2.8-kilobasepair fragment in a commercial vector (pGEM1; Promega, Madison, WI). The construct was linearized by digestion with the restriction endonuclease Hind III, and complementary RNA (cRNA) was prepared by transcription in vitro using the bacteriophage RNA polymerase T7. A capping analog (7mGpppG) was included in the reaction to generate capped transcripts because these are translated more efficiently in the oocyte. The resulting cRNA was quantified by spectrometry, and 5 ng cRNA in a 30 nl volume was injected into the oocyte using an automated microinjector (Nanoject; Drummond Scientific, Broomall, PA). The adequacy of injection was confirmed by a slight increase in cell size during injection. The cells were then cultured in modified Barth solution for 72 h at 18°C before study.
Electrophysiologic Recording.
Experiments were performed at room temperature. A single defolliculated cell was placed in a continuous-flow recording chamber (0.5 ml volume) and superfused with 2 ml/min Tyrode solution (NaCl 150 mm, KCl 5 mm, CaCl22 mm, MgSO41 mm, dextrose 10 mm, and HEPES 10 mm, pH adjusted to 7.4). The oocyte was positioned close to the inflow tubing so that complete exposure to test solutions was obtained in 5.3 ± 0.5 s (n = 20). Microelectrodes were pulled in one stage from capillary glass (BBL with fiber; World Precision Instruments, Sarasota, FL) on a micropipette puller (model 700C; David Kopf Instruments, Tujunga, CA). Tips were broken to a diameter of approximately 10 μm. They were filled with 3 m KCl, and tip resistances were usually 1–3 MΩ. The cell was voltage clamped using a two-microelectrode oocyte voltage clamp amplifier (OC725A; Warner Corporation, New Haven, CT), connected to a data acquisition and analysis system running on an IBM-compatible personal computer. The acquisition system consisted of a DAS-8A/D conversion board (Keithley-Metrabyte, Taunton, MA), and analysis was performed with a custom-written program (OoClamp software). 18All measurements were performed at a holding potential of −70 mV. Cells that did not show a stable holding current of less than 0.5 μA during a 5-min equilibration period (less than 5% of cells tested) were excluded from analysis. Membrane current was sampled at 125 Hz and recorded for 5 s before and 45 s after application of the test compounds. Responses were quantified by integrating the current trace by quadrature and are reported as microcoulomb (μC), as described previously. 19–21
Drug Administration.
Acetyl-β-methylcholine bromide (MCh), used as agonist for the m1 muscarinic receptor, was diluted in Tyrode solution to the needed concentration and superfused (2 ml/min) over the oocyte for 10 s. Agonist binding to muscarinic receptors activates one or more heterotrimeric G proteins (primarily Gq and G11) 22, which in turn activate phospholipase C. Phospholipase C releases IP3and diacylglycerol from cell membrane–bound phosphatidylinositolbisphosphate.IP3releases Ca2+from intracellular stores; diacylglycerol activates protein kinase C. Intracellular free Ca2+activates a Ca2+-dependent Cl−channel in the cell membrane of the oocyte, inducing an inward Ca2+-activated Cl−current [ICl(Ca)].
Lidocaine was also diluted in Tyrode solution to various concentrations and superfused (2 ml/min) for 10 min. Benzocaine and QX314 were diluted and administered extracellularly in the same manner as lidocaine.
For intracellular administration of QX314, a third micropipette was inserted into the voltage-clamped oocyte. The micropipette was connected to an automatic microinjector (Nanoject; Drummond Scientific, Broomall, PA). Under voltage clamp, 50 nl (approximately 10% of total oocyte volume) of various QX314 concentrations were injected.
Experimental Protocol.
Control Response.
After voltage clamping, the oocyte was allowed to stabilize for 5 min. During this time the oocyte was superfused continuously with Tyrode solution. The control response was then determined: MCh was superfused for 10 s, and membrane current was recorded for 5 s before and 55 s after the beginning of agonist application. For determination of the concentration–response relation for MCh, the experiment ended at this point. All other studies continued with measurement of the treatment response in the same oocyte.
Treatment Response.
Local anesthetic (or Tyrode solution in certain experiments) was administered to the oocyte by superfusion for 10 min. Subsequently, the treatment response was induced in the same oocyte using the same concentration of MCh as used for the control response.
Recovery Response.
To determine if any treatment effect observed could be reversed, the same oocyte was superfused with Tyrode solution for 10 min and the recovery response was measured using the same MCh concentration as used before.
Intracellular Injections.
Control and treatment experiments with intracellular QX314 injection were performed in different oocytes. In the control oocytes, 50 nl of a 150 mm KCl solution was injected; in the treatment group, we injected 50 nl of KCl solution containing various concentrations of QX314. Injection was followed by superfusion with Tyrode solution for 10 min, preventing an extracellular effect of any QX314 unintentionally leaked from the cell. ICl(Ca)was then induced by superfusion of MCh at half-maximal effect concentration (EC50), as described previously.
Binding Experiments
Membrane Preparation.
CHO cells stably transfected with the muscarinic m1 receptor were homogenized in 10 vol of ice-cold homogenization buffer (Tris 50 mm, MgCl25 mm, ethylene diamine tetraacetic acid [EDTA] 5 mm, ethylene glycol tetraacetic acid 1 mm, and aprotinin 2 μg/ml, pH adjusted to 7.5) with an Overhead stirrer (Wheaton Instruments, Millville, NJ) three times for 15 s at medium speed. The homogenate was centrifuged for 30 min at 4°C and 500 ×g . The supernatant was adjusted to 107 mm KCl and 20 mm 3-N-[morpholino]propane sulfonic acid (MOPS) (pH 7.4), mixed, incubated for 10 min on ice, and centrifuged for 60 min at 160,000 ×g at 4°C. The pellet was resuspended in 160 mm KCl/20 mm Tris (pH 7.4) with a short burst of an Overhead stirrer at medium speed and centrifuged at 160,000 ×g for 45 min at 4°C. The final pellet was resuspended in homogenization buffer and stored in aliquots at −20°C. Protein concentration was determined by the Lowry method using bovine serum albumin for standards.
Ligand Binding.
Muscarinic m1 receptor density and equilibrium dissociation constants in CHO cell membranes were determined by specific binding of a muscarinic m1 receptor antagonist, [3H] quinuclydinyl benzylate ([3H]QNB; 0.1–16 nm). One hundred–microliter aliquots of membrane (15 μg of protein) were incubated with [3H]QNB in assay buffer (Tris 20 mm, NaCl 100 mm, and EDTA 0.5 mm, pH 7.4) for 90 min at 21°C. Membranes were collected onto Whatman GF/C glass fiber filters, which were washed three times for 10 s with ice-cold buffer (Tris 10 mm and MgCl25 mm, pH 7.4). Radioactivity trapped on filters was counted using a scintillation counter. All reactions were performed in triplicate. Nonspecific binding was determined by adding 5 μm atropine to displace specific binding of [3H]QNB.
Specific binding was fit to a single-site binding model using nonlinear least square curve fitting of the untransformed data to calculate receptor density (Bmax) and dissociation constants (Kd).
To determine interaction of lidocaine with specific binding of [3H]QNB, 100-μl aliquots of membrane (15 μg of protein) were incubated with various concentrations of lidocaine (10−2–10−10m) and [3H]QNB (at Kd) in assay buffer (Tris 20 mm, NaCl 100 mm, and EDTA 0.5 mm, pH 7.4) for 90 min at 21°C, and ligand binding was determined as previously described.
Analysis
Results are reported as mean ± SEM. Measurements of at least 12 oocytes were averaged to generate each data point. Because variability among batches of oocytes is common, responses were at times normalized to control response. Statistical tests employed are indicated in the Results section. P < 0.05 was considered significant. Concentration–response curves were fit to the following logistic function, derived from the Hill equation: y = ymin+ (ymax− ymin) {1 − xn/(x50n+ xn)}, where ymaxand yminare the maximum and minimum response obtained, n is the Hill coefficient, and x50is the EC50(for agonist) or the half-maximal inhibitory concentration (IC50, for antagonist). To assess if combined administration of intracellularly injected and extracellular QX314 exert superadditive inhibition, we applied the combination in a fixed ratio based on their IC50s. The interaction was analyzed by isobolographic analysis. Ninety-five percent confidence intervals for the isobologram were calculated from the SEM.
Isobolograms
Isobolographic analysis is a nonmechanistic method of characterizing the effect resulting from the administration of two compounds by employing equieffective concentrations of individual drugs and combinations of these. Application of each drug alone is used to determine the isobolar points on the axes (a,0 and 0,b). Pure “additivity” is represented by the isobole of additivity, which is based on the equation: x/a + y/b = 1, and results in a straight line connecting the axial points. 23,24If the actual measured concentration of the combination of both drugs falls below the isobolar plot of the 95% confidence interval, superadditivity is suggested.
Materials
Molecular biology reagents were obtained from Promega (Madison, WI), and other chemicals were obtained from Sigma (St. Louis, MO). CHO cells (CRL-1982), stably transfected with the rat muscarinic m1 receptor, were purchased from ATCC (Manassas, VA). QX314 was a gift from Astra Pharmaceuticals, L.P. (Westborough, MA).
Results
Functional Expression of m1 Muscarinic Receptors in Xenopus Oocytes
Whereas uninjected oocytes were unresponsive to MCh, oocytes injected with m1 muscarinic receptor cRNA responded to application of MCh (10−4–10−9m) with a transient ICl(Ca)(fig. 1A). We previously demonstrated that this response is mediated by m1 muscarinic receptors because it is inhibited by atropine and pirenzepine. 11
We determined the concentration–response relation for the response. As shown in figure 1B, the response was concentration dependent. EC50, calculated from the Hill equation, was 5.7 ± 5.2 × 10−7m. Maximal responses of 14.6 ± 1.8 μC were obtained at a MCh concentration of 0.1 mm. Calculated maximal effect (Emax) was 13.8 ± 2.2 μC. These findings compare closely with data reported in our previous studies. 11–14
To determine if the recombinantly expressed receptors desensitized after repeated agonist administration, we induced three consecutive ICl(Ca)with MCh in the same oocyte. Two MCh concentrations were studied: 10−5m (fig. 1C) and 10−6m (fig. 1D). A recovery time of 10 min separated each agonist application (see Experimental Protocol). Average response sizes for the three applications were 2.5 ± 0.4, 2.2 ± 0.4, and 3.2 ± 0.7μC (microcoulohbs) with 10−5m MCh, and 2.9 ± 0.6, 2.8 ± 0.5 and 2.2 ± 0.4μC with 10−6m MCh. No statistically significant difference between the means of the three response sizes was found (P = 0.095 and 0.308, repeated measures analysis of variance [ANOVA]), confirming that the muscarinic receptor does not desensitize. Variability among oocyte batches is common. Figure 1E, obtained from a frog with relatively great response sizes, shows the possibility that receptor desensitization is not dependent on response size. MCh 10−5m was used as agonist. Average response sizes for the three applications were 10.2 ± 1.0, 11.3 ± 1.1, and 12.8 ± 1.8μC. No statistically significant difference between the means of the three response sizes was found (P = 0.129, repeated measures ANOVA), confirming that lack of desensitization of the muscarinic receptor is not response size dependent. Therefore, we used a control–treatment–recovery paradigm in subsequent experiments.
Lidocaine Suppresses MCh-induced ICl(Ca)Reversibly and in a Concentration-dependent Manner
To determine if local anesthetics inhibit muscarinic functioning, we chose the most commonly used local anesthetic, lidocaine. Administration for 10 min of various concentrations of lidocaine resulted in a concentration-dependent inhibition (figs. 2A and 2B) of muscarinic responses, evoked by stimulation with MCh at EC50(0.57 μm). Curve fitting revealed an IC50for lidocaine of approximately 1.8 ± 1.0 × 10−8m, a concentration much lower than that needed for blockade of sodium channels 25and also much lower than that reported for interference with muscarinic receptor binding. 26Maximal inhibition was obtained with lidocaine 1 mm; at this concentration, muscarinic responses were inhibited by 95.4%.
Next we determined the reversibility of the inhibitory effect of lidocaine on m1 muscarinic signaling. We used the most efficacious lidocaine concentration (10−3m) because we anticipated that it would have a greater likelihood of inducing nonreversible effects. After measurement of the control response, the oocyte was superfused with lidocaine 10−3m for 10 min, and the treatment response was obtained. The oocyte was then perfused with Tyrode solution for 10 min and a recovery response was obtained. As shown in figure 2C, the lidocaine effect was reversible. Percent inhibition by lidocaine was similar to that shown in figure 2A. Control,treatment, and recovery responses to MCh 0.57 μm were 8.5 ± 0.8, 0.4 ± 0.3, and 6.8 ± 0.7 μC, respectively (n = 8). There is no statistically significant difference between the control and recovery response (P = 0.149, paired t test). The holding current of the oocytes did not change significantly during these experiments.
Lidocaine Acts Primarily as a Noncompetitive Antagonist
To obtain additional information about the site of lidocaine’s interaction, we studied the type of antagonism of lidocaine on muscarinic signaling by determining if lidocaine inhibition could be overcome with greater concentrations of MCh. Responses to various concentrations of MCh (10−9–10−3m were measured under control conditions or after a 10-min superfusion with lidocaine at IC5018 nm, n = 20 at each MCh concentration;fig. 2D). The calculated control EC50for MCh was 4.2 ± 1.2 × 10−7m; Emaxwas 7.6 ± 0.3 μC. In the presence of lidocaine, the concentration–response curve shifted not significantly to the right (EC506.8 ± 3.1 × 10−7m;P = 0.445, t test), but Emaxwas significantly reduced (P < 0.001, t test) to 4.4 ± 0.4 μC, or 43% inhibition at Emax. These results suggest noncompetitive antagonism because despite high agonist concentrations, the lidocaine effect is not overcome. These findings suggest that lidocaine does not interact primarily with the ligand binding site. To eliminate oocyte variability, we confirmed the observed noncompetitive antagonism in a single oocyte (n = 20;fig. 2E). We performed four consecutive measurements. First, the oocyte was stimulated with 1 μm MCh, and the corresponding ICl(Ca) was determined. Average response size was 5.1 ± 0.5 μC. Next, the same oocyte was superfused for 10 min with lidocaine at approximately IC50(18 nm) and then stimulated with the same concentration of MCh. Lidocaine reduced the average response size to 54 ± 5.6% of the control response. A third measurement was performed after another 10-min superfusion with lidocaine at approximately IC50(18 nm), this time using a fully efficacious MCh concentration of 1 mm. Average response size with the higher agonist concentration was similar to that obtained using 1 μm MCh (44.2 ± 3.2% of the control response, P > 0.05, repeated measurements ANOVA, Turkey post hoc test). Although holding current of the oocytes did not change significantly during these experiments, we wished to confirm that the effect obtained was reversible. Thus, we superfused the oocyte with Tyrode solution for 10 min and then stimulated it with 1 μm MCh. Mean response size recovered to 89.3 ± 4.3% of the control response (P > 0.05, repeated measurements ANOVA, Turkey post hoc test).
Muscarinic Signaling Is Inhibited by Extracellularly Administered QX314
To assess lidocaine’s site of action on muscarinic signaling, we superfused the oocytes with the permanently charged and therefore membrane impermeant lidocaine analog QX314. In contrast to our results obtained from experiments with QX314 on LPA signaling, where extracellularly administered QX314 was without effect, 4extracellular QX314 blocked m1 muscarinic signaling. Curve fitting revealed an IC50of 2.4 ± 0.6 × 10−6m (fig. 3A), approximately two orders of magnitude less potent than lidocaine. Although Hill coefficient (0.39) of the inhibition curve is not typical for a ligand-gated receptor effect and might be in part the result of nonspecific action of QX314, muscarinic signaling can be inhibited by local anesthetics acting on an extracellular site.
QX314 Administered Extracellularly Acts as a Noncompetitive Antagonist
If extracellularly applied QX314, being a charged molecule, acts on the polar extracellular agonist binding site, we would expect competitive antagonism for QX314 on muscarinic signaling. Therefore, we elicited responses to various concentrations of MCh before and after administration of QX314 at IC50 (n > 12 at each MCh concentration). As shown in figure 3B, MCh EC50was 1.8 ± 0.7 × 10−5m for the control responses and 2.8 ± 0.5 × 10−5m for the treatment responses, a difference which did not reach statistical significance (P = 0.245, t test). In contrast, calculated Emaxfor the control group (9.1 ± 0.8 μC) was significantly different from that of the treatment group (7.2 ± 0.3 μC;P = 0.048, t test).
Although reduction of Emaxby QX314 is not 50% as expected for a theoretical purely noncompetitive effect, these results are in agreement with the results using lidocaine described previously and make competitive antagonism of QX314 on muscarinic functioning un-likely. We therefore suggest that the extracellular inhibitory effect of local anesthetics is mainly exerted by noncompetitive antagonism.
Benzocaine and Intracellular QX314 Inhibit Muscarinic and Lysophosphatidate Signaling to Similar Degrees
After investigating the effect of extracellularly administered QX314, we studied the effect of the permanently uncharged, and therefore membrane permeant, local anesthetic benzocaine. Benzocaine showed IC50at a concentration of 1.2 ± 0.9 × 10−3m (fig. 3C). Thus, nonpolar local anesthetics are not effective muscarinic blockers.
Next we studied the effects of intracellularly injected QX314 on muscarinic signaling. Figure 3Dshows the concentration–response relation for the QX314 effect. Curve fitting to the Hill equation revealed an IC50of 9.6 ± 2.0 × 10−4m, approximately three orders of magnitude less potent than extracellularly applied QX314. Local anesthetics acting intracellularly could in- terfere at the receptor or G-protein because we have demonstrated lack of interaction with the distal signaling pathway. 2A differentiation between these sites can be suggested by comparing effects of local anesthetic on two different Gq-coupling receptors expressed in oocytes. In this case, the intracellular receptor domains would be divergent, whereas the G-proteins coupled to the two different receptors would be identical. If the local anesthetic acts on the G-protein, similar pharmacologic parameters would be expected to be obtained in both experiments, whereas very different results would be likely when the local anesthetic acts on the receptor. Therefore, we compared the effects of QX314 and benzocaine on m1 receptors with data we obtained previously using LPAreceptors. Figures 4A and 4Bshow the concentration–response relation for benzocaine and intracellular QX314 on these two receptor systems (LPA data from Sullivan et al. 4). For QX314, not only are calculated IC50s very similar (9.6 ± 2.0 × 10−4m for MCh, 7.2 ± 1.0 × 10−4m for LPA), but maximal degree of inhibition (74.6 and 71.6%) and slope of the inhibition curve (0.48 and 0.47) were also similar. For benzocaine, calculated IC50s are close to each other (1.2 ± 0.8 × 10−3m for MCh, 1.0 ± 0.6 × 10−3m for LPA), whereas slope of the inhibition curve (0.42 and 0.68) and maximal degree of inhibition (62.1 and 72.2%) showed slightly more divergence. Nonetheless, the similarity of the curves suggest a common site of action on muscarinic and LPA signaling by benzocaine and intracellular QX314. This suggests that local anesthetics may inhibit G-protein function.
Combined Administration of Intracellularly Injected and Extracellularly Applied QX314 Inhibits in a Superadditive Manner
We tested if intracellularly injected and extracellularly administered QX314 act in a superadditive manner by measuring the effects of various concentrations of intracellular and extracellular QX314 (400:1 concentration ratio) on MCh-elicited responses. Curve fitting using the Hill equation revealed IC50s of 1.9 ± 0.5 × 10−5m intracellular QX314 / 4.9 ± 1.4 × 10−8m extracellular QX314, approximately 50-fold less than those obtained when the compounds were studied in isolation and similar to the IC50obtained with lidocaine. As shown in the isobologram in figure 4C, measured IC50for the combination is outside the 95% confidence interval (sum of fraction 0.041) for purely additive action, supporting our hypothesis of two superadditive sites of action: An extracellular site which is not the ligand-binding domain, and an intracellular site which is likely to be the G-protein (fig. 4D). Lidocaine would have access to both sites, which would explain its high potency.
Inhibition of Muscarinic m1 Receptor Signaling by Lidocaine Is Not the Result of Interaction with the Ligand Binding Pocket
We determined that lidocaine acts primarily as an noncompetitive antagonist on muscarinic m1 receptors. This strongly suggests an action outside the ligand-binding domain. If so, ligand binding should not be affected significantly by concentrations of lidocaine that inhibit signaling. Previous binding studies suggest this, 6,26but to our knowledge, it has not been tested in a system expressing only m1 receptors. Therefore, we studied the effect of lidocaine on [3H]QNB binding to m1 muscarinic receptors. We first characterized [3H]QNB binding to membranes prepared from CHO cells stably transfected with the rat muscarinic m1 receptor. Over a range of 0.1–16 nm, free drug specific binding was saturable and reached a maximum at 1.8–2.4 nm. The saturation curve and Scatchard analysis (fig. 5A) conform closely to a single-site model with a Kd of 0.23 ± 0.01 nm (n = 3) and Bmaxof 549 ± 17 nm (n = 3).
The action of lidocaine was tested in concentrations ranging between 10−10and 10−2M (fig. 5B). At concentrations that inhibit m1 muscarinic signaling, lidocaine did not affect specific binding of [3H]QNB to the muscarinic m1 receptor. Lidocaine concentrations of 10−3and 10−2m modestly reduced specific binding, but in this concentration range, nonspecific effects could not be excluded. Calculated IC50using the Hill equation was 39 ± 4 mm (n = 5). These results are in full agreement with the data reported above indicating a primarily noncompetitive antagonism of lidocaine on muscarinic m1 receptors.
Discussion
In the current study we investigated the effect of local anesthetics on the functioning of muscarinic receptors of the m1 subtype, the most common subtype in the CNS. We demonstrated that lidocaine inhibits signaling of muscarinic receptors expressed recombinantly in Xenopus oocytes. Its IC50(18 nm) is significantly less than that needed for blocking sodium channels (60–200 μm, depending on the state of the sodium channel). 25Clinically relevant blood concentrations during intravenous infusion or epidural anesthesia are in the range 1–15 μm, corresponding to a plasma level of 0.3–4.5 μg/ml. Based on these results, we postulate that m1 muscarinic receptors may be a target for local anesthetics in the CNS.
Although best known for their ability to block sodium channels, local anesthetics interact with other cellular systems as well, and some of their general anesthetic properties may result from interactions with these other targets. Several studies report that local anesthetics, whether administered intrathecally, intravenously, or intramuscularly, significantly reduce the needed concentrations for volatile anesthetics, such as nitrous oxide and halothane, 27and for intravenous anesthetics, such as thiopental, midazolam, and propofol. 28–30Inagaki 31demonstrated that epidural administration of lidocaine delays awakening from isoflurane anesthesia. The mechanisms underlying these results have not been studied in detail; our findings, interpreted in view of the effects of CNS muscarinic blockade on consciousness, 32suggest that suppression of muscarinic signaling may be involved.
Part of the current study was designed to compare local anesthetic effects on muscarinic signaling with those on LPA signaling, as reported previously, 4and methodology was therefore maintained constant between the two studies. Nonetheless, our findings were quite different. LPA 4and thromboxane A23receptors are inhibited by local anesthetics, but the primary site of action seems to be at the intracellular domains of the receptor, or at the G protein. Extracellular administration of QX314 inhibited neither LPA nor thromboxane A2signaling, but intracellular QX314 inhibits both signaling pathways effectively. Benzocaine inhibits as well and shows profound superadditive interaction with intracellular QX314 on LPA signaling. The signaling pathway downstream of the G-protein is not affected by local anesthetics (responses induced by guanosine 5′-0-(3-thiotriphosphate) [GTPγS] injection were not inhibited). Thus, local anesthetics appear to inhibit LPA and muscarinic m1 receptor signaling via an action on the intracellular domain of the receptor, the associated G-protein, or coupling between the two. However, because G-protein activation with GTPγS bypasses several critical G-protein functions (e.g. , receptor–G-protein coupling and GTPase activity), we could not exclude a direct effect of local anesthetics on the G-protein itself.
Local anesthetic effects on muscarinic signaling were quite different. The inhibitory potency of lidocaine was several orders of magnitude greater than that observed on LPA or thromboxane A2receptors. The effect was noncompetitive and was not associated with a decrease in agonist binding. Extracellular QX314 inhibited muscarinic signaling also, although with significantly less potency than lidocaine. Benzocaine and intracellularly injected QX314 inhibited with a potency similar to that observed on lysophophosphatidate and thromboxane A2receptors. Intracellular and extracellular QX314 inhibited in a superadditive manner. To explain these findings, we postulate the presence of two interacting local anesthetic binding sites in the early steps of the muscarinic signaling pathway (fig. 4D). The primary site would be a polar, hydrophilic, noncompetitive site on the extracellular receptor domains, and the secondary site would be localized on the intracellular receptor domains or on the G protein. In view of the remarkable similarity of the inhibition curves of intracellular QX314 and benzocaine on muscarinic, LPA, 4and thromboxane A2(Hoenemann CW and Durieux ME, unpublished observation) receptors, we hypothesize that this latter site is localized on the common G protein, rather than on the highly diverse receptors. Alternatively, benzocaine could act on a nonpolar extracellular site, but the similar inhibitory effects of benzocaine on different receptors make this explanation less likely. Lidocaine’s remarkable potency could be explained because it would have access to both sites, resulting in superadditive interaction.
Several studies have shown inhibition of muscarinic receptor binding by various local anesthetics. In general, the necessitated concentrations were high, and a variety of mechanisms have been invoked to explain the findings. Nonspecific alteration of surface charge and fluidity of the cell membrane 33–35resulting in a dislocation of the receptor and effector components have been proposed. Aguilar et al. 26suggested that local anesthetics bind not to the receptor site but to a nearby accessory site because they found local anesthetics did not provide protection against the deleterious effect of Triton X-100, whereas agonists and antagonists did. These findings are consistent with our results.
Fairhurst et al. 6investigated the effect of lidocaine on [3H]QNB binding to muscarinic receptors in a rat brain cortex preparation. They found that lidocaine inhibits [3H]QNB binding, with inhibitory potency correlating with pH (i.e. , with the amount of the charged form of lidocaine). In their study, competitive inhibition occurred only at higher concentrations (IC50for lidocaine 0.13 mm). The IC50for lidocaine obtained in our binding assay was two orders of magnitude higher (39 mm). This discrepancy could be explained by differences in the employed models. While we performed our binding assay as described on membranes prepared from stably transfected CHO cells, Fairhurst et al. 6obtained their results from rat brain cortex preparations, probably containing multiple receptor subtypes. QNB is known to interact with accessory sites in addition to the agonist binding region of the muscarinic receptor. 36,37Therefore, it is conceivable that high concentrations of local anesthetics displace [3H]QNB binding by interaction with an accessory receptor region while having no effect on agonist binding to the receptor. However, the results by Fairhurst et al. 6also fit partially to a theoretical model indicating that carbachol, [3H]QNB, and lidocaine (in higher concentrations) might competitively displace one another at the same agonist binding site.
Other interactions, direct and indirect, between local anesthetics and muscarinic signaling are possible and were not investigated in our study. In some tissues and brain regions, muscarinic receptors might be capable of interacting with sodium channels. Cohen-Armon et al. 9found that local anesthetics inhibit batrachotoxin-enhanced binding affinity for muscarinic receptor agonists. Recently, Horio et al. 10demonstrated that local anesthetics inhibit muscarinic receptor desensitization in guinea pig ileal longitudinal muscle. Some local anesthetics exert this effect by competitive antagonism, whereas others act on an noncompetitive site. Alteration of the conformation of the ligand binding site followed by modification of ligand receptor interaction were hypothesized to explain these findings. In our model, the m1 receptor does not significantly desensitize, so we could not investigate this issue.
One significant limitation of all studies cited is the lack of definition of the muscarinic subtypes investigated. All studies used tissues which may express a variety of receptor subtypes, making interpretation difficult. In brain, all five muscarinic receptor subtypes are expressed, 38and differences between the m1–m3–m5 and m2–m4 groups in structure and effector coupling are such that results can not be extrapolated easily. For example, significant m2 receptor block by local anesthetics would result in tachycardia, which is clearly not one of the common side effects of lidocaine. These problems are overcome by recombinant expression of a single subtype in isolation.
The current study shows that local anesthetics inhibit muscarinic m1 receptors expressed recombinantly in Xenopus oocytes. Lidocaine inhibits at concentrations significantly less than those necessitated for blocking sodium channels. We suggest that this inhibitory effect is because of superadditive interactions between noncompetitive antagonism on an extracellular polar site on the muscarinic receptor molecule and an intracellular site, probably on the coupled G-protein. Whereas the intracellular site appears to be the same on muscarinic, LPA, and thromboxane A2receptors, the lipid mediator receptors lack the extracellular polar local anesthetic binding domain.
The authors thank Prof. Dr. med. Eike Martin (Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany) for his support and G. Paul Matherne, M.D., Associate Professor, Department of Pediatrics, University of Virginia, Charlottesville, Virginia, for his assistance with the binding studies.