| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Although the effects of local anesthetics on sodium channels and various other channels and receptors have been intensively investigated, there is little information available about their effects on N-methyl-D-aspartate (NMDA) receptors. We examined the effects of four local anesthetics (procaine, tetracaine, bupivacaine, and lidocaine) on NMDA-induced currents by using a whole-cell patch-clamp technique in dissociated mouse hippocampal pyramidal neurons. Procaine and tetracaine produced a reversible and concentration-dependent inhibition of NMDA-induced currents, but lidocaine showed little inhibition at 1 mM or less. The half-maximal inhibition values (mM; mean ± SEM) for procaine, tetracaine, bupivacaine, and lidocaine at -60 mV were 0.296 ± 0.031, 0.637 ± 0.044, 2.781 ± 0.940 (extrapolated data), and 7.766 ± 14.093 (extrapolated data), respectively. Procaine 0.2 mM reduced the maximal NMDA-induced currents without affecting the 50% effective concentration values for NMDA. The inhibition by procaine exhibited voltage dependence and was more effective at negative potentials. These results indicate a noncompetitive antagonism of procaine on NMDA receptors and suggest that the inhibition is the result of a channel-blocking mechanism. IMPLICATIONS: We examined the effects of four local anesthetics (procaine, tetracaine, bupivacaine, and lidocaine) on NMDA-induced currents by using a whole-cell patch-clamp technique in dissociated mouse hippocampal pyramidal neurons. Both procaine and tetracaine produced a reversible and concentration-dependent inhibition of the NMDA-induced currents.
The primary sites of action of local anesthetics are voltage-gated sodium channels (1), but their effects on various other ion channels and receptors (26) have also been reported. Studies have shown that, although the majority of anesthetics have -aminobutyric acid type A (GABAA)-receptor agonist properties (7), a number of anesthetics have N-methyl-D-aspartate (NMDA)-receptor, a subtype of glutamate receptor, antagonist properties (810). The NMDA receptor is deeply involved in neuronal plasticity during memory acquisition (11) and in pathological states, such as nociception (12), hypoxic neuronal death (13), and drug addiction (14). Intrathecally administered NMDA antagonists reduce the minimum alveolar concentration of isoflurane (15) and prevent pain sensitization (12). Local anesthetics not only provide analgesia, but also have neuroprotective effects against ischemic neuronal damage (16), probably because of the blockade of Na+ influx. Of all local anesthetics, procaine suppressed both the influx of extracellular Ca2+ and the release of Ca2+ from the intracellular Ca2+ stores in ischemic conditions (17). Although the main mechanism for the prevention of Ca2+ influx by local anesthetics is mediated by the blockade of Na+ influx through sodium channels, we hypothesized that procaine and other local anesthetics also directly inhibit Ca2+ influx through NMDA receptors in neurons. However, there is little available information about the effects of local anesthetics on the NMDA receptors. Only one article has shown that neither etidocaine nor procaine inhibited NMDA receptor gated channels (6). The purpose of this study was to investigate the effects of four local anesthetics (procaine, tetracaine, bupivacaine, and lidocaine) on the NMDA receptors in mouse CA1 pyramidal neurons by using electrophysiological methods.
This study was approved by the Animal Research Committee of Kansai Medical University. Postnatal (714 days) ICR mice were anesthetized with diethyl ether and decapitated. The brain was rapidly dissected, and the frontal section of the brain, including the hippocampus, was cut into 500-µm slices with a microslicer (YIS-102; Astec, Fukuoka, Japan) in ice-cold Krebs solution at 2°C4°C. The slices were preincubated in the Krebs solution for 60 min at room temperature (20°C25°C). They were then enzymatically treated with 0.0125% Pronase (Calbiochem-Novabiochem Corporation, La Jolla, CA) at 31°C for 720 min, followed by 0.0125% thermolysin at 31°C for 720 min. Thereafter, they were incubated in the Krebs solution for another 60 min at room temperature. The hippocampal CA1 region was then micropunched out from the slices with a small needle and transferred to a culture dish (Falcon 3801; 35-mm diameter) filled with oxygenated Krebs solution. The pyramidal neurons in the CA1 region were mechanically dissociated under a phase-contrast microscope by using a series of fire-polished micropasteur pipettes with a variety of orifice sizes (from 0.1- to 0.5-mm diameter). The isolated pyramidal neurons adhered to the bottom of the dish within 30 min. The ionic composition of the external solution (Krebs solution) was as follows (in mM): NaCl 121.5, KCl 4.7, CaCl2 2.5, MgCl2 1.2, NaHCO3 15.5, KH2PO4 1.2, and glucose 11.5, adjusted to pH 7.45 by bubbling through 95% oxygen and 5% CO2 gas. The internal solution (pipette solution) contained (in mM) K-aspartate 103, KCl 20, NaCl 10, Mg-ATP 5, creatine-phosphate 20, EGTA 5, CaCl2 · 2H2O 1.03, and HEPES 5, adjusted to pH 7.0 with KOH. Rapid application of drugs was achieved by using the Y-tube method (18). The Y-tube was made from polyethylene (2-mm diameter), and the outlet tip of the Y tube (0.1-mm diameter) was set approximately 0.51.0 mm away from the neuron recorded. This system enabled the complete exchange of external solution surrounding the recorded neuron within 1020 ms.
Electrophysiological recordings were performed in a conventional whole-cell configuration by using a patch-clamp technique (19) at room temperature. The electrode resistance, when filled with the internal solution, was 6 to 8 M Thermolysin, NMDA, procaine, tetracaine, bupivacaine, and lidocaine were obtained from Sigma Chemical (St. Louis, MO). Pronase and glycine were obtained from Calbiochem and Nacalai Tesque (Kyoto, Japan), respectively. Pronase and thermolysin were dissolved in Krebs solution just before use. Glycine, NMDA, procaine, tetracaine, bupivacaine, and lidocaine were directly dissolved in Mg2+-free external solution (in mM: NaCl 123.3, KCl 4.7, CaCl2 2.5, NaHCO3 15.5, KH2PO4 1.2, and glucose 11.5, adjusted to pH 7.45 by bubbling through 95% oxygen and 5% CO2 gas) just before use to avoid Mg2+ block. Concentration-inhibition, concentration-response, and current-voltage curves of the averaged data were generated with Origin 6.0 software (Microcal Software, Northampton, MA). The data are represented as mean ± SEM, and at least five neurons were used for each data point. The results obtained were statistically analyzed by the unpaired Students t-test or one-way analysis of variance (ANOVA) followed by the Scheffé multiple comparison test. A P value <0.05 was considered significant.
The effects of four local anesthetics (procaine, tetracaine, bupivacaine, and lidocaine) on NMDA-induced currents were investigated. Preapplication of local anesthetics in the absence of the agonists did not produce any current response. Coapplication of 0.3 mM NMDA and 0.3 µM glycine to the neurons held at a holding potential of -60 mV induced apparent inward currents that rapidly reached the peak and then declined to the steady-state level in the Mg2+-free external solution. When one of four local anesthetics (1 mM) was coapplied with NMDA and glycine during the steady-state phase of the response, procaine, tetracaine, and bupivacaine, but not lidocaine, reversibly inhibited the steady-state current, as shown in Figure 1. In three of five neurons, lidocaine also showed a slight inhibition of steady-state NMDA currents.
The effects of four local anesthetics on the steady-state NMDA-induced currents are summarized in Figure 2. The current amplitudes in the presence of local anesthetics were normalized to the control steady-state current just before the application of local anesthetics.
Data were fitted with the following equation:
where I is the steady-state NMDA-induced current in the presence of local anesthetics, IMAX is the maximum steady-state NMDA-induced current (control; in the absence of local anesthetics), C is the concentration of local anesthetics, n is the Hill coefficient, and IC50 is the half-maximal inhibition. Procaine and tetracaine (10 µM to 1 mM) inhibited the NMDA-induced currents in a concentration-dependent manner. Bupivacaine at 1 mM significantly inhibited the NMDA-induced current (ANOVA, p < 0.05 versus the currents in the absence of bupivacaine). Lidocaine, however, failed to exert significant inhibition of the NMDA-induced currents of 1 mM or less. The obtained IC50 values (mM) from the fitting for procaine, tetracaine, bupivacaine, and lidocaine at -60 mV were 0.296 ± 0.031 (n = 5), 0.637 ± 0.044 (n = 5), 2.781 ± 0.940 (extrapolated data, n = 5), and 7.766 ± 14.093 (extrapolated data, n = 5), respectively. To characterize the method of inhibition by local anesthetics, we studied the effects of 0.2 mM procaine on the concentration-response relationships for the NMDA-induced currents (Fig. 3). The current amplitudes were normalized to the control steady-state currents just before the application of procaine. Procaine 0.2 mM effectively suppressed the maximal current responses. The half-maximal concentration values (µM) for NMDA in the absence and presence of 0.2 mM procaine were 42.8 ± 4.7 and 51.1 ± 4.9, respectively. These values were not significantly different (Students t-test). In addition, the Hill coefficient values were not affected by procaine (0.72 ± 0.05 in the absence of procaine and 0.67 ± 0.03 in the presence of procaine) (Students t-test).
To analyze the mechanism of procaine-induced noncompetitive inhibition of NMDA responses, we studied the current-voltage relationships for NMDA-induced currents in the absence or presence of 0.6 mM procaine. The current amplitudes were normalized to the control current responses (steady-state currents, see symbol * in Fig. 4), which were induced by coapplication of 0.3 mM NMDA and 0.3 µM glycine at -60 mV in the absence of procaine. The inhibition by procaine exhibited voltage dependence and was more effective at hyperpolarized potentials (Fig. 4). The extent of inhibition was significantly dependent on the membrane potential (ANOVA, p < 0.001). Procaine reduced the current responses to 25% ± 3% (n = 5) of the control responses at -80 mV, whereas it reduced the current responses to only 84% ± 2% (n = 5) at +60 mV. We analyzed the voltage-dependent inhibition by procaine by using the Woodhull model (20), on the assumption that procaine does not move from the blocking site to the cytoplasm:
equation
where P is the probability that the NMDA receptor channel is not blocked, [procaine]o is the concentration of procaine outside the cell, Kd(0) is the equilibrium dissociation constant of procaine from the binding site at 0 mV, z is the valence of procaine,
In this investigation, we showed that some local anesthetics, such as procaine and tetracaine, inhibited NMDA-induced currents in a concentration-dependent manner, but lidocaine showed little inhibition at 1 mM or less. The IC50 values (mM) for procaine, tetracaine, bupivacaine, and lidocaine at -60 mV were 0.296 ± 0.031, 0.637 ± 0.044, 2.781 ± 0.940 (extrapolated data), and 7.766 ± 14.093 (extrapolated data), respectively. The mode of inhibition by procaine seems to be noncompetitive and channel-blocking.
Several potential limitations of our model should be considered. First, in the study of concentration-inhibition relationships, reliable data could not be obtained for more than 1 mM bupivacaine and lidocaine because of the failure of complete gigaohm sealing. The acutely dissociated neurons seemed to be weak compared with Xenopus oocytes or mammalian cell lines, e.g., HEK293 cells expressing recombinant NMDA receptors, probably because they were enzymatically treated. This may be why the neurons became leaky with more than 1 mM local anesthetics. Thus, the concentration-inhibition curves of bupivacaine and lidocaine for more than 1 mM had to be extrapolated, and the IC50 values for bupivacaine and lidocaine were not reliable. Yet bupivacaine significantly inhibited the NMDA-induced currents at 1 mM compared with those without bupivacaine. However, lidocaine did not show significant inhibition even at 1 mM. Second, only procaine was used to characterize the inhibitory effect of local anesthetics on the NMDA receptor channels because the inhibition by bupivacaine and lidocaine at concentrations <1 mM was not sufficient for the characterization, and the inhibition by tetracaine seemed similar to that of procaine. Third, we used acutely dissociated hippocampal neurons from postnatal (714 days) mice to investigate native NMDA receptors because several discrepancies (21) have been reported when comparing the pharmacological and biophysical properties of native NMDA receptors with those of recombinant ones. In addition, the NMDA receptor subunits expressed in a postnatal mouse hippocampus are Some local anesthetics suppressed the NMDA receptor channel activity in a concentration-dependent manner. Århem and Frankenhaeuser (25) reported that the IC50 value for procaine on sodium permeability in a single myelinated fiber was 0.21 mM, which is very similar to the IC50 value for procaine in the NMDA-induced currents obtained in this study. A previous clinical study showed that the concentrations of procaine at 15 and 105 minutes after intrathecal injection of 150 mg procaine were approximately 18.3 and 1.1 mM, respectively (26), and the concentration of tetracaine in the cerebrospinal fluid during spinal anesthesia was in the range of approximately 50400 µM (27). Thus, the concentrations of procaine and tetracaine required to inhibit NMDA receptors are well within clinically relevant concentrations. The potency for local anesthetics to produce both tonic and phasic inhibition of sodium channels is dependent on their structure, hydrophobicity, and pKa (dissociation constants) (28,29). Local anesthetics with a more hydrophobic nature, such as tetracaine and bupivacaine, are more potent inhibitors of sodium channels than their less hydrophobic congeners. However, procaine, the least hydrophobic of the four local anesthetics we examined, produced the most potent inhibitory effect on the NMDA receptors. The effective form of local anesthetic is an important factor in understanding how these compounds interact with ion channels. Local anesthetics dissociate according to the pKa and the pH of the solution. Because the pKas of procaine, tetracaine, bupivacaine, and lidocaine are 8.9, 8.5, 8.1, and 7.9, respectively, the degree of NMDA receptor antagonism by a local anesthetic seems to be related to its pKa. The larger the pKa, the higher the level of the cationic form at pH 7.45 and the more potently a local anesthetic inhibits NMDA-induced currents. The electrically charged cationic form of local anesthetics may be more active in reducing NMDA-induced currents. Procaine reduced the maximal current response of the NMDA receptor channels without affecting the half-maximal concentration values, and this indicates a noncompetitive antagonism of NMDA receptor channels by procaine. The inhibition by procaine exhibited voltage dependence and was more effective at negative potentials, indicating that procaine can enter the channel from the outside and bind to the site to prevent Na+ and Ca2+ influx. Location of the binding site is estimated to be the halfway point of the channel.
Lu and Bieger (6) reported that neither etidocaine nor procaine inhibited glutamate receptor gated channels, such as (±)- Lu and Bieger (6) also demonstrated that the currents through a nicotinic cholinoceptor, which is another ligand-gated cation channel, were blocked by both etidocaine and procaine. Furthermore, Yost and Dodson (4) and Cuevas and Adams (5) demonstrated that the inhibition of the nicotinic acetylcholine receptor by procaine was concentration dependent, voltage dependent, and enhanced by hyperpolarization. This is similar to our characterization of the effects of procaine on NMDA receptors. The NMDA receptor antagonists inhibit pain sensitization (12) and protect against various sorts of brain damage (13). Local anesthetics themselves also block pain propagation and reduce neuronal damage (16) through voltage-gated sodium channel blockade. The concentration of local anesthetics in the cerebrospinal fluid is thought to reach hundreds of micromolar levels soon after the direct injection of local anesthetics into humans. Thus, the NMDA receptor antagonist properties of some local anesthetics may be clinically relevant and beneficial for pain treatment. Tetracaine and bupivacaine are already in prevalent use as local anesthetics for spinal or epidural anesthesia. Procaine might be the most promising local anesthetic for neuroprotection because, in addition to sodium channel and NMDA receptor blockade, it inhibits ryanodine Ca2+ release channels in the endoplasmic reticulum (17). In conclusion, we demonstrated that local anesthetics such as procaine and tetracaine inhibited NMDA-induced currents in a concentration-dependent manner and that the mode of inhibition by procaine was noncompetitive and channel-blocking. Our results imply that some local anesthetics, which have NMDA antagonist property, may be more beneficial for pain control and neuroprotection because NMDA receptor antagonists inhibit the progress of pain sensitization and neuronal damage.
The authors thank Atsushi Nagata of Kansai Medical University and Masahiro Sugimoto of Osaka University Medical School for their help in preparing the figures.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|