Local anesthetics (LAs) are known to inhibit voltage-dependent Na+ channels, as well as K+ and Ca2+ channels, but with lower potency. Since cellular excitability and responsiveness are largely determined by intracellular Ca2+ availability, sites along the Ca2+ signaling pathways may be targets of LAs. This study was aimed to investigate the LA effects on depolarization and receptor-mediated intracellular Ca2+ changes and to examine the role of Na+ and K+ channels in such functional responses.
Effects of bupivacaine, ropivacaine, mepivacaine, and lidocaine (0.1-2.3 mm) on evoked [Ca2+](i) transients were investigated in neuronal SH-SY5Y cell suspensions using Fura-2 as the intracellular Ca2+ indicator. Potassium chloride (KCl, 100 mm) and carbachol (1 mm) were individually or sequentially applied to evoke increases in intracellular Ca2+. Coapplication of LA and Na+/K+ channel blockers was used to evaluate the role of Na+ and K+ channels in the LA effect on the evoked [Ca2+](i) transients.
All four LAs concentration-dependently inhibited both KCl- and carbachol-evoked [Ca2+](i) transients with the potency order bupivacaine > ropivacaine > lidocaine >/= mepivacaine. The carbachol-evoked [Ca2+](i) transients were more sensitive to LAs without than with a KCl prestimulation, whereas the LA-effect on the KCl-evoked [Ca2+](i) transients was not uniformly affected by a carbachol prestimulation. Na+ channel blockade did not alter the evoked [Ca2+](i) transients with or without a LA. In the absence of LA, K+ channel blockade increased the KCl-, but decreased the carbachol-evoked [Ca2+](i) transients. A coapplication of LA and K+ channel blocker resulted in larger inhibition of both KCl- and carbachol-evoked [Ca2+](i) transients than by LA alone.
Different and overlapping sites of action of LAs are involved in inhibiting the KCl- and carbachol-evoked [Ca2+](i) transients, including voltage-dependent Ca2+ channels, a site associated with the caffeine-sensitive Ca2+ store and a possible site associated with the IP(3)-sensitive Ca2+ store, and a site in the muscarinic pathway. K+ channels, but not Na+ channels, seem to modulate the evoked [Ca2+](i) transients, as well as the LA-effects on such responses.
THE majority of work concerning the mechanisms of local anesthetics (LAs) has focused on LA actions on the Na+and K+channels. LAs are known to reduce both Na+and K+currents, 1–3which contribute to their anesthetic action. However, neuronal excitability is also determined by Ca2+availability which is controlled by regulatory mechanisms of cytosolic Ca2+([Ca2+]i). Alteration of those regulatory mechanisms by LAs can lead to changes in presynaptic transmission and postsynaptic excitability. A few studies have shown that bupivacaine, ropivacaine, and lidocaine can affect the voltage-dependent Ca2+currents in different systems. 4–6
In the neuron-like human neuroblastoma cell line, SH-SY5Y, we have previously demonstrated that volatile anesthetics can either augment or attenuate intracellular Ca2+responses to stimuli. 7The net result is determined by the action of volatile anesthetics on the entire spectrum of Ca2+regulatory mechanisms, including Ca2+-entry, -mobilization, and -removal (-extrusion and -sequestration). Due to the essential role of intracellular Ca2+in physiological functions and neuronal excitability, we wanted to determine whether LAs also affected intracellular Ca2+signaling and the regulation of Ca2+homeostasis in a similar manner as the volatile anesthetics.
In the present study, we have chosen four LAs commonly used in our clinical practice. Bupivacaine, ropivacaine, mepivacaine, and lidocaine are all amide-linked LAs. Bupivacaine, ropivacaine, and mepivacaine differ only in the length of an alkyl side chain (R) on the piperidine ring contained in their amine domain (R = C4H9for bupivacaine, R = C3H7for ropivacaine, and R = CH3for mepivacaine), while lidocaine differs from the other three in the composition of its amine group (no piperidine ring) and the length of its intermediate chain. These four LAs have different hydrophobicities (the octanol/buffer partition coefficients) correlating with their very different anesthetic potency. 8,9
To assess the LA-effects on evoked [Ca2+]itransients, single (100 mm potassium chloride [KCl] or 1 mm carbachol) or sequential (100 mm KCl followed by 1 mm carbachol and vice versa ) stimuli were used. In addition, we examined whether the LA-effects on the evoked [Ca2+]itransients were related to their actions on Na+and/or K+channels, using coapplication of LAs and Na+or K+channel blockers. Our results showed distinct LA-effects on KCl- and carbachol-evoked [Ca2+]itransients with no clear involvement of Na+channels, but involvement of K+channels. In addition, the filling status of the Ca2+stores appeared to modulate the LA effect on KCl- and carbachol-evoked [Ca2+]itransients.
Materials and Methods
All chemicals used for this study were obtained from Sigma Chemicals (St. Louis, MO), unless otherwise indicated.
Human SH-SY5Y neuroblastoma cells (originally provided by June Biedler, Ph.D., Sloan-Kettering Institute for Cancer Research, Rye, NY, at the time that the cells were provided; current position and affiliation, Distinguished Resident Scientist, Fordham University, Bronx, NY), passages 58–77, 90, 100–101, were maintained in RPMI 1640 medium containing 12% fetal bovine serum, penicillin (100 units/ml), streptomycin (100 μg/ml), and fungizone (1 μg/ml), passaged every 3 weeks. All cell culture components were GIBCO BRL products and purchased from Life Technologies (Rockville, MD).
Cells in confluent culture were loaded with 5 μm Fura-2/AM (Molecular Probes, Eugene, OR) for 30 min at 37°C in the original culture medium. After removal of the dye-containing medium, the cells (still attached) were rinsed twice with Dulbecco phosphate-buffered saline without CaCl2and MgCl2(Gibco BRL, Rockville, MD) with 10 mm glucose-addition (pH 7.4), then gently resuspended in an incubation buffer (pH 7.4) containing 140 mm NaCl, 5 mm KCl, 5 mm NaHCO3, 1 mm MgCl2, 10 mm HEPES, and 10 mm glucose. Cells were incubated in incubation buffer containing 1.5 mm CaCl2for 1 h at room temperature to allow recovery. After a brief centrifugation (1 min at 300 rpm) prior to each measurement, the cell pellet was resuspended in the same incubation buffer and thermo-equilibrated at 37°C for 5 min. After a short baseline recording, 1.5 mm CaCl2was added to the suspension 3 min before cells were challenged by either 100 mm KCl or 1 mm carbachol, or one of the two sequential stimulation modes (see section on Sequential Stimulation and fig. 1). Fluorescence ratios were monitored at 510 nm after alternated excitation at 340 and 380 nm by an SLM 8000 Aminco fluorescence spectrofluorometer (Aminco, Urban, IL). Maximal and minimal fluorescence ratios were acquired by using 0.01% sodium dodecyl sulfate and 60 mm ethylene glycol-bis(β-aminoethyl ether)N,N,N′,N′-tetraacetic acid, and [Ca2+]iwas then determined from fluorescence ratios based on the method of Grynkiewicz et al. 10and a Kdof fura-2 for Ca2+of 224 nm, as described previously. 7
In order to investigate the contribution of different calcium regulatory mechanisms on evoked [Ca2+]itransients and to monitor the dynamic interaction between different intracellular Ca2+stores, sequential stimulation was applied as follows: the cells, in the presence and absence of LA ± tetrodotoxin or tetraethylammonium (TEA), were challenged by either 100 mm KCl followed by 1 mm carbachol (a muscarinic receptor agonist and Ca2+-releaser from the IP3-sensitive store), or 1 mm carbachol followed by 100 mm KCl (fig. 1).
In the drug-treated groups, cells were preincubated with LA ± tetrodotoxin/TEA for 5 min at 37°C prior to CaCl2-addition as follows:
LA (−) tetrodotoxin/TEA: Preincubation with one of the following LAs: bupivacaine (0.75% Sensorcaine-MPF, Astra Pharmaceutical Products, Westborough, MA, USA), or ropivacaine (0.5% Naropin, Astra USA), or mepivacaine (1.5% Polocaine-MPF, Astra Pharmaceutical Products, USA), or lidocaine (2% Lidocaine HCl, Abbott Laboratories, Chicago, IL, USA). The concentrations applied varied from 0.1 to 2.3 mm for each LA.
Tetrodotoxin/TEA (−) LA: Preincubation with either 1 μm tetrodotoxin, or 10 mm TEA;
Tetrodotoxin (+) LA: Preincubation with 1 μm tetrodotoxin plus one of the LAs at 2.3 mm in both KCl- and carbachol-stimulation.
TEA (+) LA: Preincubation with 10 mm TEA plus one of the LAs.
The LA concentrations applied in this series of measurements were as follows: in KCl-stimulation, 0.25 mm for bupivacaine, 0.5 mm for ropivacaine and lidocaine, 1 mm for mepivacaine, respectively; in carbachol-stimulation, 0.25 mm for all four LAs. The selection of these LA-concentrations was based on the dose–response curve for each LA (see Results) for their inhibition of KCl- or carbachol-evoked [Ca2+]itransients. The concentrations chosen resulted in partial inhibition of the evoked [Ca2+]Itransients so that a further effect by TEA in either direction could be observed.
Comparison between different groups was performed using one-way ANOVA (Student-Neuman-Keuls test or Dunnett test) conducted by Sigmastat software (Jandel Scientific, San Rafael, CA). Comparison between two curves using F test was conducted by GraphPad Prism (Graphpad Software, San Diego, CA). A significance was considered to be P < 0.05.
Local Anesthetics Inhibited Both KCl- and Carbachol-evoked [Ca2+]iTransients
In the absence of LA, the basal [Ca2+]imeasured was 61 ± 19 nm (mean ± SD, n = 536) in SH-SY5Y cells incubated in nominally Ca2+-free buffer. Upon adding 1.5 mm CaCl2, the [Ca2+]ireached a higher steady state level of 93 ± 42 (mean ± SD, n = 157). None of the LAs showed significant effect on this Ca2+-evoked [Ca2+]iincrease prior to stimulation (fig. 2).
Figures 1A and 1Bare control traces representing the two sequential stimulation modes used in the present study. While the KCl-evoked [Ca2+]itransient is dependent on the external Ca2+, the carbachol-evoked [Ca2+]itransient can be induced in the nominally Ca2+-free medium in freshly prepared cells (data not shown). Both KCl- and carbachol-evoked [Ca2+]itransients were significantly inhibited in a concentration-dependent fashion by bupivacaine, ropivacaine, mepivacaine, and lidocaine at concentrations 0.1–2.3 mm, which are lower than those applied to nerves during spinal or nerve blocks 11,12but higher than those found in the plasma during epidural anesthesia. 13,3Nonlinear fitting of the dose–response curves (figs. 3A and 3B) showed the order of the IC50values as follows: bupivacaine < ropivacaine < lidocaine < mepivacaine and bupivacaine ≈ ropivacaine < lidocaine ≈ mepivacaine for KCl-evoked and carbachol-evoked [Ca2+]itransient without a prestimulation, respectively (table 1). The difference of IC50values between the most (bupivacaine) and the least (mepivacaine) potent LAs for inhibition of the KCl-evoked [Ca2+]itransients was elevenfold (fig. 3A&table 1). Statistical analysis also showed that the differences between four LAs for their inhibitory effects on the KCl-evoked [Ca2+]itransients at different LA concentrations were mostly significant (P < 0.05, table 2a). In contrast, the IC50range (0.13–0.28 mm) for inhibiting the carbachol-evoked [Ca2+]itransients was much narrower (fig. 3B, table 1). In fact, the potency of the four LAs was shown to be very similar (P > 0.05, table 2) in their inhibition of the carbachol-evoked [Ca2+]itransients.
Using sequential stimulation, i.e. , 100 mm KCl followed by 1 mm carbachol or vice versa , the LA effects on both KCl-evoked and carbachol-evoked [Ca2+]itransients were further examined. All four LAs inhibited both carbachol-evoked and KCl-evoked [Ca2+]itransients in a concentration-dependent manner when the intracellular Ca2+stores were partially depleted by a KCl or a carbachol prestimulation, respectively. Concentration–response curves (figs. 3C and 3D) showed an order of the IC50values bupivacaine < ropivacaine < lidocaine ≈ mepivacaine, essentially indistinguishable between lidocaine and mepivacaine (P > 0.05). The potency of bupivacaine and ropivacaine for inhibiting the KCl-evoked [Ca2+]itransients remained unchanged when the intracellular Ca2+stores were partially depleted by a carbachol prestimulation, but mepivacaine was found to be less potent (P < 0.05) and lidocaine more potent (P < 0.05) following carbachol prestimulation (fig. 3A,vs. fig. 3C, table 1, row a vs. row b). In contrast, the carbachol-evoked [Ca2+]itransients became less sensitive to the LAs and displayed a wider IC50range (table 1, row c vs. row d) when the Ca2+stores were partially emptied by a KCl prestimulation (fig. 3B,vs. fig. 3D).
Effects of Tetrodotoxin and TEA on KCl-evoked and Carbachol-evoked [Ca2+]iTransients and Effects of their Coapplication with Different LAs
The SH-SY5Y cells possess tetrodotoxin-sensitive Na+currents. 14,15Tetrodotoxin at 1 μm (a concentration shown to completely block the voltage-dependent Na+currents in the SH-SY5Y cells) 15,16,17had no significant effect on either KCl-evoked or carbachol-evoked [Ca2+]itransients (fig. 4). In the presence of tetrodotoxin, LA inhibition of evoked [Ca2+]itransients was identical to inhibition in the absence of tetrodotoxin, indicating that LA block of Na+channels does not contribute to LA inhibition of evoked [Ca2+]itransients.
TEA (10 mm) alone produced a small yet significant increase in the KCl-evoked [Ca2+]itransients. In the presence of any of the LAs, no significant difference was found between groups (±) TEA (fig. 5A). Since TEA augmented the KCl-evoked [Ca2+]itransients, the results obtained in the presence of LA would suggest a greater degree of inhibition and an involvement of K+channels in KCl-evoked [Ca2+]itransients, which was blocked by LAs.
In contrast, the carbachol-evoked [Ca2+]itransients were significantly attenuated by TEA alone, and coapplication of TEA and any LA showed further inhibition (fig. 5B).
This study demonstrates that LAs decreased KCl-evoked and carbachol-evoked [Ca2+]itransients. Their potency appeared to be modulated by the filling level of intracellular Ca2+stores (except for bupivacaine) and affected by TEA-sensitive K+channels, but not tetrodotoxin-sensitive Na+channels.
Intracellular Ca2+Stores Modulate LA Action on KCl-evoked and Carbachol-evoked [Ca2+]iTransients
Previously, in SH-SY5Y cells, we have shown that there are two major intracellular Ca2+stores, the caffeine-sensitive and the IP3-sensitive, which are functionally connected. The KCl-evoked [Ca2+]itransients are dependent on Ca2+release from the caffeine-sensitive stores, while the carbachol-evoked [Ca2+]itransients are mainly attributed to Ca2+release from the IP3-sensitive stores. 7Moreover, the carbachol-evoked [Ca2+]itransients were largely (∼90%) inhibited by a muscarinic antagonist atropine (1 μm), indicating that the carbachol-evoked [Ca2+]itransients in our SH-SY5Y cells were predominantly mediated by activation of muscarinic receptors (data not shown). Using a sequential stimulation protocol, we investigated potential LA targets associated with both intracellular Ca2+stores.
The LAs attenuated the KCl-evoked [Ca2+]itransients in a concentration-dependent manner. Prestimulation with carbachol did not alter the potency of bupivacaine and ropivacaine in inhibiting the KCl-evoked [Ca2+]itransients, suggesting that the observed inhibitory effect was unrelated to an action on the IP3-sensitive stores. However, the potency of mepivacaine and lidocaine was altered by carbachol prestimulation, suggesting an ad-ditional effect of these two drugs on the carbachol-depletable Ca2+store/IP3-sensitive pathway. Following carbachol prestimulation, the mepivacaine potency in inhibiting the KCl-evoked [Ca2+]itransients was increased, while lidocaine potency decreased (table 1), suggesting that mepivacaine and lidocaine either affected an additional target of the carbachol-depletable Ca2+store differently or they were acting at different sites on that store.
All four LAs reduced the carbachol-evoked [Ca2+]itransients in a concentration-dependent fashion. The KCl prestimulation decreased the potency of ropivacaine, mepivacaine, and lidocaine in inhibiting the carbachol-evoked [Ca2+]itransients. This suggests that part of the action of LA might involve prevention of a contribution of Ca2+from the caffeine-sensitive store to the IP3-sensitive Ca2+response, as was reported previously with halothane and isoflurane. 7When the caffeine-sensitive store is depleted, the LAs appear to be less potent. The Ca2+contribution of the caffeine-sensitive store could result from either a direct Ca2+release through the ryanodine receptors 18into the cytoplasm or a more direct Ca2+translocation within the endoplasmic reticulum from the caffeine- to the IP3-sensitive stores. 7
A comparison of the slopes of the LA concentration–response curves (the Hill coefficients) for inhibition of evoked [Ca2+]itransients prior to or following a prestimulation (table 1) reveals that predepletion of a Ca2+store may alter the cooperative interaction between LAs and/or their potential target. This appears especially true for ropivacaine in carbachol-evoked [Ca2+]itransients following KCl prestimulation, where the cooperativity seems to be significantly increased (higher Hill coefficients, table 1).
In Xenopus oocytes, LAs were also shown to inhibit muscarinic receptor-mediated signaling through an action on the receptors and the coupled G-proteins, but not on the downstream events involving the phospholipase C-IP3-Ca2+pathway. 19–23In the present study, since a direct LA-action on the sites associated with the IP3-sensitive Ca2+store was not apparent for bupivacaine and ropivacaine (see the previous paragraph), the elements upstream of the store, such as the muscarinic receptors and/or coupled Gq-protein(s), need to be considered as the possible sites for the LA action. Mepivacaine and lidocaine, however, seem to affect the IP3-sensitive Ca2+store. It is important to mention that our data only indicate an LA target associated with the IP3-sensitive Ca2+store, but not the exact nature of this site. We cannot rule out that the IP3-mediated Ca2+signaling pathway in oocytes is controlled differently than in neuronal cells and, therefore, responds to LAs differently.
Since the potency of bupivacaine in inhibiting both the KCl-evoked and the carbachol-evoked [Ca2+]itransients remained unchanged with prestimulation (table 1), bupivacaine appears to act mostly on other sites, possibly the sites associated with the plasma membrane, to inhibit the evoked [Ca2+]itransients rather than on sites associated with either caffeine- or IP3-sensitive stores, or other intracellular compartment. However, the possibility that bupivacaine equally suppresses the evoked [Ca2+]itransients mediated through these two Ca2+stores cannot be ruled out.
Roles of Voltage-dependent Na+and K+Channels on the LA Action on Carbachol- and KCl-evoked [Ca2+]iTransients
Activation of Na+currents, which occurs during KCl-induced depolarization, is not essential for the generation of the KCl-evoked and carbachol-evoked [Ca2+]i-transients. In these neuronal cells, the LA-block of Na+currents does not contribute to the LA inhibition of the evoked [Ca2+]itransients.
In SH-SY5Y cells, the TEA-sensitive K+channel has been shown to be predominantly responsible for the delayed rectifier K+currents. 15,17During depolarization, opening of these K+channels will oppose the depolarizing effect on inward Ca2+currents. Therefore, blocking of TEA-sensitive K+channels will be expected to increase the Ca2+channel activity and in turn, the KCl-evoked [Ca2+]itransients, as demonstrated in figure 5A. In the presence of either of the four LAs, the level of the [Ca2+]itransients was similar with or without TEA. This suggests that the TEA-sensitive K+channels are blocked by LAs. One possibility is that LAs inhibited TEA-sensitive K+channels and hence an additional blockade by TEA did not further affect the LA-action on the KCl-evoked [Ca2+]iresponse. However, blockade of K+channels does not explain the overall inhibition of the KCl-evoked [Ca2+]i-response, since TEA alone increased this response (fig. 5A). Most likely, LA caused reduction of the KCl-evoked [Ca2+]i-response was mediated through LA action on voltage-dependent Ca2+channels, since we have previously shown that the generation of the KCl-evoked [Ca2+]itransients are dependent on the extracellular Ca2+and functional voltage-dependent Ca2+channels, 7which have been shown to be blocked by LA. 4,5,24–26
The carbachol-evoked [Ca2+]itransients were significantly lower in the presence of TEA, suggesting that K+channels contribute to the carbachol-evoked [Ca2+]itransients. In the presence of LA, the inhibition by TEA was still visible in addition to the LA effect on the carbachol-evoked [Ca2+]itransients (fig. 5B). Moreover, the sum of the individual LA inhibition and TEA inhibition was larger than the inhibition caused by a coapplication of both drugs. These results suggest that LA and TEA were acting on different and overlapping sites, such as the same TEA-sensitive K+channels. The K+channels might modulate the carbachol-evoked [Ca2+]itransients through direct or regulatory action on the muscarinic receptor and/or coupled G-proteins.
Hydrophobicity and Local Anesthetic Action
In general, the higher hydrophobicity is attributed to the larger size of the alkyl substituents in the molecule of an LA and correlates with a higher local LA. 9Comparison of the partition coefficients (total drug/ml octanol ÷ total drug/ml buffer, pH 7.4, 8a parameter for the hydrophobicity) of the LAs shows a descendent potency order: bupivacaine (346/560) > ropivacaine (115/n.d.) > lidocaine (43/110) > mepivacaine (21/42) (values represent the partition coefficient determined at 25/36°C; n.d. = not determined), which not only resembles the potency order of these drugs in blocking conduction but also in inhibiting the KCl-evoked [Ca2+]itransients (see Results). 9The presence of a secondary potential LA target on the carbachol-sensitive Ca2+store might be correlated with the smaller hydrophobic amine domain in the molecules of mepivacaine and of lidocaine. In contrast, the LA potency for inhibition of the carbachol-evoked [Ca2+]itransients was similar for four LAs, showing little correlation to their very different partition coefficients.
In summary, this study has demonstrated inhibitory effects of several LAs on the KCl-evoked and carbachol-evoked [Ca2+]itransients mediated through different pathways. The major LA inhibition seems to occur at sites associated with the caffeine-sensitive Ca2+store (such as ryanodine receptor) and/or sites associated with plasma membrane, such as voltage-dependent Ca2+channels, muscarinic receptor, and/or coupled G-protein. A direct action of LA on the IP3-sensitive store is not apparent for bupivacaine and ropivacaine, but LA may affect the IP3-sensitive Ca2+releases indirectly through their inhibitory effect on the muscarinic receptor and/or coupled G-protein. In contrast, mepivacaine and lidocaine might act on both the caffeine- and IP3-sensitive stores. Although a blockade of Na+currents did not affect the evoked [Ca2+]itransients or the LA effects on them, the K+currents seem to modulate both the KCl-evoked and carbachol-evoked [Ca2+]itransients through different pathways, and their blockade may affect the LA action on intracellular [Ca2+]iresponses.