Local Anesthetics Inhibit Muscarinic Receptor-mediated Activation of Extracellular Signal-regulated Kinases in Rat Pheochromocytoma PC12 Cells 

Procaine hydrochloride, tetracaine hydrochloride, lidocaine hydrochloride, bupivacaine hydrochloride, atropine sulfate salt, tetradotoxin, veratridine, ouabain, calyculin A, carbachol, and PMA were purchased from Sigma Chemical Company (St. Louis, MO). Pirenzepine dihydrochloride and 4-diphenyacetooxy-1,1-dimethylpiperidinium, methiodide, and pertussis toxin were obtained from Research Biochemicals International (Natick, MA). Dulbecco modified Eagle medium and horse serum were from Life Technologies (Grand Island, NY). Fetal bovine serum was from Nippon Bio-supply Center (Tokyo, Japan). Anti-phosphotyrosine mouse monoclonal antibody was from Upstate Biotechnology Incorporated (Lake Placid, NY). Anti-ERK mouse monoclonal antibody was from Affiniti Research Products Limited (Nottingham, UK). The goat anti-mouse immunoglobulin G horse-radish peroxidase-coupled secondary antibody and the enhanced chemiluminescence system used for Western blot analysis were obtained from Amersham Life Science (Buckinghamshire, UK). Other reagents were of the highest quality available.

A PC12 cell line was supplied by Dr. Y. Sugimoto (Shirakawa Institute of Animal Genetics, Fukushima, Japan). Monolayer cultures of the cells were maintained in 100-mm-diameter tissue culture dishes in Dulbecco modified Eagle medium supplemented with 10%(vol/vol) fetal bovine serum and 5%(vol/vol) horse serum in a humidified atmosphere containing of 5% carbon dioxide at 37°C. Stock cultures were subcultured routinely at a cell density of 2–3 × 10⁶/dish at least once a week, and culture media were renewed every 2 days.

PC12 cells were subcultured in 60-mm-diameter tissue culture dishes at 1 × 10⁶cells/dish and grown for 4 days. The cells were washed twice with 2 ml buffer A (25 mM Hepes, p  H 7.4, 125 mM NaCl, 5 mM KCl, 1 mM MgCl², 1 mM CaCl², 5 mM glucose, and 1 mg/ml bovine serum albumin) and preincubated in 3 ml buffer A with or without local anesthetics or Na⁺current–modifying reagents (tetrodotoxin, veratridine, and ouabain) at 37°C for 10 min. The cells were stimulated with 1 mM carbachol or 200 nM PMA at 37°C for durations indicated in each experiment. The reaction was terminated by aspiration of the reaction buffer and washing twice with 2 ml ice-cold phosphate-buffered saline (8% NaCl, 0.2% KCl, 2.88% Na²HPO⁴· 12H²O, 0.2% KH²PO⁴, p  H 7.4). The washed cells were scraped quickly into 120 μl RIPA buffer (10 mM Tris-HCl, p  H 7.4, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 150 mM NaCl, 1 mM EDTA, 0.5 mM phenylmethylsulfronyl fluoride, 10 μg/ml leupeptin, 1 mM Na³VO⁴, 10 mM NaF, and 0.1 mM Na²MoO⁴) and transferred to a microcentrifuge tube. After incubation on ice for 30 min, the suspension was centrifuged at 13,000g  for 20 min to obtain the cell extract. Seventy micrograms of protein was subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (10%) and transferred electrophoretically to a nitrocellulose membrane. After the membranes were blocked with TBS-T (10 mM Tris-HCl, p  H 7.5, 150 mM NaCl, and 0.1% Tween 20) containing 2% bovine serum albumin, membranes were incubated with anti-phosphotyrosine antibody or anti-ERK antibody at room temperature for 90 min and then with the goat anti-mouse immunoglobulin G horseradish peroxidase–coupled secondary antibody at room temperature for 60 min. Detection was performed using an enhanced chemiluminescence system. The density of protein bands was analyzed using a densitometer (Densitograph; Atto Corporation, Tokyo, Japan).

Data are presented as the mean ± SD from three experiments. Differences between values were evaluated using analysis of variance; when P < 0.05, differences were considered significant.

Treatment of cells with 1 mM carbachol induced rapid and transient increases in tyrosine phosphorylation of several proteins with approximate molecular weights of 111, 91, 84, 74, 65–70, 44, and 42 kd. Among these, two phosphoproteins with molecular masses of 44 and 42 kd were most distinct, and phosphorylation of both proteins reached a peak 2 min after stimulation of carbachol (fig. 1A). These two proteins were identified as active forms of ERK1 and ERK2 by Western blot analysis using anti-ERK antibody. As shown in figure 1B, two bands of ERKs were presented as doublets 2–15 min after carbachol stimulation. Tyrosine-phosphorylated bands of 44- and 42-kd proteins were overlapped with upper bands of ERKs doublets, respectively (figs. 1A and 1B). As determined by densitometric analysis, active forms of ERKs reached the maximum (about 20 times more than nonstimulated levels for ERK2 and 10 times more than nonstimulated levels for ERK1) at 2 min and returned to the basal level 15 min after stimulation (fig. 1C).

Such carbachol-induced tyrosine phosphorylation of 44- and 42-kd proteins was blocked completely by preincubation with 50 μM atropine. This indicates that carbachol-induced activation of ERK1 and ERK2 was through the mAChRs in PC12 cells (figs. 2A and 2B). The mAChR-mediated ERK activation was insensitive to pertussis toxin (fig. 3A). Pretreatment of PC12 cells with pertussis toxin (10 ng/ml) for 21 h failed to suppress the carbachol-induced activation of ERK1 and ERK2. To determine which subtype of mAChRs was involved, the cells were pretreated with 10 μM pirenzepine, the M¹receptor antagonist, and with 10 μM 4-diphenyacetooxy-1,1-dimethylpiperidinium, the M³receptor antagonist. The carbachol-induced activation of ERKs was blocked completely by 4-diphenyacetooxy-1,1-dimethylpiperidinium but not by pirenzepine (fig. 3B).

In PC12 cells preincubated with 1–10 × 10⁻⁴M procaine for 10 min, carbachol-induced protein tyrosine phosphorylation decreased significantly. Procaine suppressed tyrosine phosphorylation of ERKs in a concentration-dependent manner and almost completely abolished tyrosine phosphorylation of these proteins at 5 × 10⁻⁴M (fig. 4). Pretreatment of PC12 cells with tetracaine, lidocaine, and bupivacaine (5 × 10⁻⁴M) also inhibited carbachol-induced tyrosine phosphorylation of ERKs (fig. 5).

Phosphorylation of ERK1 and ERK2 was prevented by procaine in a dose-dependent manner (fig. 6A). Some significant effect (P < 0.05) was apparent with 10⁻⁴M procaine. Activation of these ERKs was almost completely suppressed in the presence of 5 × 10⁻⁴M procaine in PC12 cells (figs. 6B and 6C). The presence of tetracaine, lidocaine, and bupivacaine at 5 × 10⁻⁴M similarly inhibited the activation of both ERKs (fig. 7). Although procaine and tetracaine appeared to be more potent than lidocaine and bupivacaine, the effects of each local anesthetic did not differ significantly (P > 0.05) at 5 × 10⁻⁴M.

A commonly used activator of PKC as a receptor-bypass stimuli, PMA caused activation of ERK1 and ERK2 and peaked 15 min after stimulation in PC12 cells (data not shown). We then considered whether procaine exerted inhibitory effects on PMA-induced ERK activation. Although 5 × 10⁻⁴M procaine almost completely suppressed carbachol-induced ERK activation (fig. 8A), it did not have any effects on PMA-induced ERK activation (fig. 8B).

Carbachol-induced ERK activation was not affected by preincubating PC12 cells with tetrodotoxin or veratridine at 1 × 10⁻⁶M (fig. 9A). In addition, at 5 ×10⁻³M ouabain also had no effects on the carbachol-induced activation of ERKs (fig. 9B).

Because we knew that the combination of NaF plus AlCl³(presumably acting as AlF⁴⁻) can directly activate G proteins, 16PC12 cells were exposed to NaF plus AlCl³to determine whether the ERK cascade downstream of the G proteins in PC12 cells is influenced by procaine. AlF⁴⁻caused a time-dependent activation of ERK1 and ERK2 (data not shown). However, 5 × 10⁻⁴M procaine did not affect the AlF⁴⁻-induced ERK activation (fig. 10A). In addition, pretreatment of PC12 cells with 10 nM calyculin A, a potent phosphatase inhibitor, did not modify the effects of procaine on carbachol-induced ERK activation (fig. 10B).

The molecular mechanisms underlying local anesthetic action for neuronal activities are not well understood. We have presented data showing that carbachol induces ERK1 and ERK2 activation that is mediated by the activation of M³mAChR in PC12 cells and that procaine inhibits carbachol-mediated tyrosine phosphorylation and ERK activation in a concentration-dependent manner. Although tetracaine, lidocaine, and bupivacaine similarly inhibited carbachol-induced tyrosine phosphorylation and ERK activation, neither Na⁺channel modifiers (such as tetrodotoxin, an Na⁺channel blocker, or veratridine, an Na⁺channel activator) nor ouabain, an Na⁺–K⁺pump inhibitor, affected ERK activation. The inhibitory effects of local anesthetic were not observed in both PMA-induced ERK activation and AlF⁴⁻-induced ERK activation. In addition, a potent phosphatase inhibitor did not modify the effects of procaine on the carbachol-induced ERK activation. Therefore, the site(s) of inhibition of local anesthetics is probably located at mAChR or an interface between the receptors and G protein (fig. 11), rather than in the downstream of G protein. The ERK signaling cascade was recently shown to play an important role in the receptor-mediated signaling processes in neuronal cells. 10,11The current results would provide an important clue for better understanding the molecular mechanism of the local action of an anesthetic.

The MAPKs are a family of serine–threonine protein kinases that are activated in response to various stimuli and play important roles in cellular signal transduction. 9Among three major MAPK cascades, ERK, Jun N-terminal kinase/stress-activated protein kinase, and p38, which was identified recently, 10the ERK studied in the current investigation has been shown to be involved in development, proliferation, plastic functions such as long-term potentiation, 17and apoptosis. 18PC12 cells are stimulated with carbachol via  heterotrimetric G protein–coupled mAChR, 19,20followed by the direct binding of the adapter protein Grb2 to Sos. 10Then the Grb2–Sos complex activates the nucleotide exchange activity on Ras, which leads to activation of the Raf–ERK pathway. 10In addition to this pathway, the Ras-independent pathway may involve PKC. 10 

Phospholipase C activated by heterotrimeric G protein acts on phosphatidylinositol 4,5-bisphosphate to produce inositol(1,4,5)-triphosphate and diacylglycerol, both of which serve as second messengers to mobilize calcium and activate PKC, respectively. 21,22Protein kinase C could be involved in the ERK activation by Raf phosphorylation on serine residues. 23Our current findings that carbachol induced tyrosine phosphorylation of p42- and p44-kd proteins and ERK activation in PC12 cells correspond with results observed in oligodendrocytes, the myelin-producing cells of the central nervous system. 24Atropine and 4-diphenyacetooxy-1,1-dimethylpiperidinium, an M³mAChR antagonist, blocked concentration-dependent MAPK activation induced by carbachol. Activation of ERKs is implicated in the transmission of the signal for mAChR. 24–26Although it is not clear whether the inhibition of M³mAChR-mediated tyrosine phosphorylation and ERK activation with procaine, lidocaine, bupivacaine, and tetracaine at clinically relevant concentrations is related to their anesthetic action, to toxic side effects, or both, many pharmacologic properties of local anesthetics can be attributed to their actions on signal transduction molecules.

Local anesthetics are known to bind to mAChRs 27and nicotinic AChRs. 1With open channels, anesthetic binding to the AChR occurs immediately after agonist stimulation. 1,27Cohen-Armon et al.  19,20have found with membrane prepared from brain that the binding of agonists to mAChRs and to the Na⁺channels in the open state is a coupled event mediated by G proteins. When an agonist-like carbachol binds to its specific receptor, mAChR, the G protein mediates the activity of transmembrane channels through which ions flow, leading to changes in transmembrane voltage that could trigger signal transmission, and thus signal transduction molecules could be altered. Inhibitors such as procaine, as a cationic inhibitor of the AChR, can bind to the inhibitory site before the channel opens and regulates the ion permeability of the AChR. 1,27Local anesthetics can allosterically interact with mAChR, 28,29and it has been reported that lidocaine interacts with primary and allosteric recognition sites on mAChR. 29After we plotted a curve fit for Hill equation, we found that the Hill coefficient for the procaine concentration–inhibition relation was 2.4 (fig. 6D), it can be assumed that the inhibition of local anesthetic on ERK activation involves multiple sites of action. The current observations may suggest that suppression of mAChRs would be involved, at least in part, in the local anesthetic inhibition of carbachol-induced ERK activation.

Modulation of Na⁺channels by PKC is likely to have important effects on signal transduction and synaptic transmission in the central nervous system. 2The action of ion channels and Na⁺–K⁺pumps, which are closely associated with PKC activity, has a central role in the regulation of neuronal excitability, so PKC has been considered as a major site of signal transduction molecules for anesthetic action. 6,30The increased activity of PKC modulates neuronal signal transduction by phosphorylation of several membrane proteins, including voltage-dependent Na⁺and other ion channels. 31The phosphorylation of Na⁺channels (α- subunits) could occur by altering PKC activity. 32Although PKC-activated Raf is thought to be Ras independent, stimulation of PKC in COS cells leads to Ras activation followed by formation of the Ras–Raf-1 complex, 33suggesting that PKC could activate ERKs by a mechanism distinct from that initiated by mAChR stimulation by carbachol. Activation of ERKs via  the M³mAChR subtype, such as M¹and M⁵, is independent of intracellular or extracellular Ca²⁺but depends in part on PKC. 26 

The effects of local anesthetics on PKC, however, are complicated. In a recent in vivo  study, Nivarthi et al.  6found that the PKC activity of the spinal cord was increased with clinically relevant concentrations of intrathecal procaine and tetracaine. In in vitro  experiments, dibucaine and tetracaine competitively inhibit the binding of PMA to its receptor. 34Unlike staurosporin, an inhibitor of PKC, neither 10⁻⁵to 10⁻³M ropivacaine, 10⁻³M bupivacaine, nor 1–3 × 10⁻³M lidocaine exerts any effects on the PMA-induced inhibition of phosphoinositide breakdown in human SK-N-MC neuroblastoma cells. 35Because Na⁺channels in PC12 cells could be blocked completely by 5 × 10⁻⁴M procaine, 36the fact that PMA-induced ERKs activation was not affected by procaine suggests that PMA-stimulated PKC activity remained intact in our experiments. Therefore, inactivation or activation of PKC, in the presence of local anesthetics, might not influence the effects on ERKs in any important way. Therefore, the effects of local anesthetics on carbachol-induced ERK activation is unlikely to be involved in the inhibition via  Na⁺channels or the PKC-dependent pathways at the concentration used in the current experiments.

We can only speculate why the Na⁺current-modifying reagents did not affect mAChR-mediated ERK activation. Several Na⁺transport pathways, such as the Na⁺channel and the Na⁺–K⁺pump, are implicated in changes in the intracellular Na⁺concentration. The membrane depolarization of the voltage-gated Na⁺channel affects G protein–coupled mAChR only if the voltage-gated Na⁺channels can be activated, 20and the blockade of Na⁺channels by tetrodotoxin decreases muscarinic agonist binding to the receptors, whereas an Na⁺channel activator such as veratridine increases the binding. 19 

Furthermore, it has been reported that the tyrosine phosphorylation process seems to directly control the Na⁺–K⁺pump activity in the proximal convoluted tubule, 37and veratridine stimulates phosphoinositide breakdown, which is inhibited by local anesthetic. 38Therefore, we assumed that tetrodotoxin, veratridine, and ouabain could have some effect on muscarinic signal transduction of ERKs. However, we observed no effect in the presence of those Na⁺-modifying reagents. It is possible that such Na⁺-modifying reagents may not affect cellular events via  pertussis toxin–insensitive G proteins, such as carbachol-induced ERK activation. In human neutrophils, tetrodotoxin and veratridine have been reported not to affect PMA-induced O²⁻generation, but eight local anesthetics, including procaine, suppress it in their concentration-dependent and lipid solubility–dependent manners. 39 

The application of carbachol reduces the peak Na⁺current without changing the voltage dependency of the channels, and the activation of mAChR strongly modulates Na⁺channel activity. 40In addition, the channel-mediated Na⁺entry and Na⁺–K⁺pump activity are functionally interdependent, 41and binding to mAChR is voltage dependent. 20Therefore, it is conceivable that the stimulation by carbachol might have already altered the channel functions and Na⁺–K⁺-pump functions of PC12 cells, so neither tetrodotoxin, veratridine, nor ouabain may cause any change in mAChR-mediated events. Although we cannot completely exclude the possibility that Na⁺-modifying reagents could affect ERK activation, the inhibition of mAChR-induced ERK activation by local anesthetics seems likely to be independent of blockade of Na⁺currents and electrochemical gradients across the PC12 cell membrane.

It is unclear whether a potential interaction with M³mAChR plays any role in local anesthetic action or state. As shown in the current results and by findings of other investigations, M³mAChRs are coupled to a pertussis toxin–insensitive G protein and seem to regulate the activation of membrane-bound guanylyl cyclase. 14,42The M³mAChRs have been reported to exist extensively in the spinal cord 43and in brain regions, including the cerebral cortex, caudate nucleus, globus pallidus, the substantia nigra, and the hypothalamus, 44and to play some role in neuronal cellular activities, such as N-methyl-D-aspartate receptor-mediated adenosine release in the cortex. 44An activator of G protein, AlF⁴⁻, caused time-dependent ERK1 and ERK2 activation, but 5 × 10⁻⁴M procaine did not affect the AlF⁴⁻-induced activation (fig. 10A). Neither did a phosphatase inhibitor modify the effects of procaine on the carbachol-induced ERK activation. Therefore, these data may provide further evidence that the inhibition of muscarinic signaling by local anesthetics that we observed could exist in some specific sites of interaction with M³mAChR signaling molecules, probably at the mAChR, an interface between mAChR and G protein, or membrane lipid microdomains around the receptors and G-protein molecules, but this is unlikely because of the effect on G proteins per se . The inhibition of M³mAChR-mediated ERK activation should be considered a convincing candidate for one target of local anesthetic action. Because little work has been done on the local anesthetic effects on ERK signaling, further studies are needed to determine the precise sites of local anesthetic action on the ERK signaling molecules.

In conclusion, the local anesthetics procaine, lidocaine, tetracaine, and bupivacaine, at their clinically relevant concentrations, inhibit carbachol-induced ERK activation, which is implicated in the transmission of signals via  M³mAChR-mediated cellular functions in PC12 cells. The target(s) that local anesthetics affect may be upstream of G proteins, including mAChR. Our findings may provide evidence that the inhibition of the ERK cascade is crucial for the action of local anesthetics. The lack of effect of Na⁺-current–modifying reagents on the muscarinic receptor-mediated ERK activation may suggest a great diversity of electrophysiologic and pharmacologic properties of local anesthetic action in signaling.