| ||||||||||||||
|
|
|||||||||||||
Department of Anesthesia and Critical Care, University of Chicago, Chicago, Illinois
Address correspondence and reprint requests to Dr. Khaled M. Houamed, Department of Anesthesia and Critical Care, The University of Chicago Medical Center, 5841 S. Maryland Ave., MC 4028, Chicago, IL 60637. Address e-mail to khouamed{at}midway.uchicago.edu
| Abstract |
|---|
|
|
|---|
Implications: We describe differential effects of anesthetics on cloned small conductance calcium-activated potassium channels from brain that may play a role in generating the effects or side effects of anesthetics.
| Introduction |
|---|
|
|
|---|
In addition to [Ca2+]i, neuronal SK channels are also modulated by second messenger systems activated by G protein coupled receptors (GPCRs), such as ß-adrenergic, muscarinic, metabotropic glutamate, and GABAB, as well as receptors for adenosine and dopamine. The inhibition of SK channels in the hippocampus during fight-or-flight situations is thought to be the molecular underpinning of attentiveness (1,2). Generally, in the brain, modulation of SK channels is a major effector mechanism for GPCRs affecting behavior (1,3). Underscoring the important role of SK channels in behavioral phenotype are the effects of apamin injection into the brain. This specific SK channel blocker causes long-term sleep disturbances, hyperexcitability, changes in learning ability, seizures, and neuronal degeneration (4,5). Two reports have also linked defects in these channels with predisposition to schizophrenia, bipolar disorder, and epilepsy (6,7).
Previous studies have shown that various anesthetic drugs interact with sAHP in neuronal preparations (e.g., see Ref. 8). As previously explained, the magnitude and time-course of sAHP are dependent not only on the calcium channel activity and [Ca2+]i metabolism, but also on upstream processes such as receptor, G protein, and kinase activities. The advent of SK channel cloning (9,10) has made possible the study of the direct interaction of these channels with drugs. We investigated the effects of anesthetics and related drugs on the predominant neuronal SK channel subtype, SK2, expressed in a heterologous system.
| Methods |
|---|
|
|
|---|
The thin-wall borosilicate glass patch pipettes used to measure whole-cell SK2-channel currents had tip resistance of 410 M
when filled with an intracellular solution of the composition, (in mM): K+ methyl sulfate, 137.5; CaCl2, 2; MgCl2, 1; EGTA, 3; HEPES, 10; ATP, 3; and glucose, 5; pH, 7.4. We determined analytically the calcium affinity for the EGTA batch used in this study (11,12). By using this affinity, the calculated free calcium ion concentration [Ca2+] of this solution (approximately 0.8 µM) is sufficient for near maximal activation of the SK2 channels (9,12). The cell under study was constantly perfused with a modified Ringers solution [composition, (in mM): NaCl, 117; KCl, 30; CaCl2, 2; MgCl2, 1; HEPES, 10; glucose, 5; NaHCO3, 2; pH, 7.4] by using a rapid microperfusion device. This device consisted of a 500 µm-diameter glass pipette connected, via a 10:1 manifold and valves, to reservoirs containing test solutions. The outlet of the microperfusion device was mounted on a micromanipulator and maneuvered adjacent to the cell under study. The time constant of this system was estimated, from junction potential changes on switching the perfusate from Ringers solution to 10% Ringers solution, to be approximately 1 s (11). In addition, the recording chamber was constantly perfused with a modified mammalian Ringers solution.
A different perfusion setup was used to apply the inhaled anesthetics. This setup was as previously described (13). It consists of an 80-mL perfusion chamber continuously perfused at 25 mL/min. Drugs were rapidly delivered to the cell by local perfusion by using an RSC-100 motor-driven solution exchange device (Biologic, Claix, France). The speed of exchange as measured by dye application was <1 s (13). Anesthetic losses for this system have been measured by using gas chromatography (M. D. Krasowski, unpublished data, 1999) and have been found to be 5%10% of the applied minimum alveolar anesthetic concentration (MAC). MAC values were taken to be: isoflurane, 370 µM; enflurane, 420 µM; halothane, 290 µM; sevoflurane, 360 µM; and chloroform, 910 µM and are similar to published values (14). There is currently no published concentration for 1 MAC desflurane, however, it is likely to be close to 280 µM (E. I Eger, II, oral communication, September 1999).
The cells were voltage-clamped at -100-mV holding potential; the potassium reversal potential under these conditions was approximately -30 mV. SK2 channel current was observed as a sustained inward current developing rapidly as the interior of the cell equilibrated with the large [Ca2+] pipette solution. To quantify the magnitude of the SK2 current, we applied supramaximal concentration (10 µM) of the specific SK2 channel blocker dequalinium. Brief application of this drug maximally and reversibly blocked the SK2 current (Figure 1). In control experiments (data not shown), nontransfected HEK 293 cells did not express a dequalinium-sensitive current. Furthermore, the nondesensitizing dequalinium-sensitive current was only observed in SK2-expressing cells when they were dialyzed with large [Ca2+] solutions. Additionally, experiments with changing the extracellular potassium concentration indicate that the [Ca2+]i-activated, dequalinium-sensitive current is carried by potassium ions (data not shown). We internally normalized all our block experiments to the block achieved by 10 µM dequalinium in each cell. Normalized block by drugs (mean ± SD) was plotted against drug dose and fitted, by using a least-squares minimization algorithm, with the expression:
|
|
|
| Results |
|---|
|
|
|---|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Ketamine, a dissociative anesthetic, was the most potent blocker we tested; it blocked up to 70% of recombinant SK2 channel current with an apparent KD of 470 µM. Our data are consistent with data obtained on other native potassium channels (15). Our results suggest that ketamine blocks SK2 channels with equal potency to its block of sodium channels; it is, however, three times more potent at blocking SK2 channels than voltage-gated, delayed rectifier, potassium channels (50% inhibiting concentration [IC50] approximately 900 µM). Other studies have found ketamine a more potent blocker of ATP-gated channels (KD approximately 63 µM) (16), and large conductance calcium- and voltage-gated potassium channels (KD approximately 230 µM) (17). However, in contrast to our study, in both these studies (16,17)) ketamine was applied to the intracellular aspect of the channel. Ketamine blocks N-methyl-D-aspartate receptors at approximately 10 times larger potency than we observed for SK2 channels (18). Moreover, the concentrations for blocking N-methyl-D-aspartate receptors closely overlap with concentrations for clinical anesthesia (1060 µM) (14). Taken together with previous studies, our data suggests that block of SK channels is unlikely to be the main conduit of ketamine anesthetic action, however, that at clinical concentrations of the drug, there may be a measurable effect on SK channels.
The local anesthetic, lidocaine, blocked approximately 80% of the recombinant SK2 channel current with an apparent KD of 4.44 mM. Brau et al. (19) reported approximately fourfold smaller KD (1.2 mM) for lidocaine block of voltage-gated K channels in peripheral nerve; an uncharacterized, voltage-independent, K channel was much more sensitive to block (KD approximately 220 µM). Kindler et al. (20) reported that lidocaine blocks a recombinant two pore domain potassium channel with similar potency (KD = 222 µM). The transient outward current, KA, was also much more sensitive to lidocaine (IC50 of approximately160 µM) (21). Lidocaine also interferes with intracellular calcium regulation with similar potency to its effect on SK channels (22). Such effects, in vivo, would be expected to indirectly modulate SK channel currents. In contrast to our study and the others cited previously, Sbarbaro et al. (23) showed that the antiarrhythmic effect of lidocaine occurs at plasma concentrations of a few µM, suggesting that block of SK channels is unlikely to occur in a clinical setting.
Our results indicate that large concentrations of ethanol and butanol block recombinant SK2 channels by up to approximately 50%. The apparent KDs for the alcohol block were 74 and 104 mM for ethanol and butanol, respectively. Previous work on sAHPs, which are mediated by native SK channels, yielded contradictory data. Carlen et al. (8), found that ethanol, at the concentrations we used, augmented the SK channel-mediated sAHP in hippocampal neurons. Others report that ethanol generated opposite effects; a depression of the sAHP in hippocampal and cerebellar Purkinje neurons consistent with an SK channel block (24,25). Direct comparison between our study and studies in native neurons, are complicated by the pleiotropic effects of ethanol on cellular signal transduction systems. Ethanol modulates calcium channels and various kinase systems, all of which would be expected to ultimately influence the sAHP waveform (26). In conclusion, our results show, for the first time, that ethanol at quasi-anesthetic concentrations may directly modulate SK channels.
Methohexital blocked roughly 30% of SK2 current with an apparent KD of 0.8 mM. This compares with its block of inward rectifier potassium channels in mast cells (IC50 approximately 0.15 mM) (27). Because barbiturates also block calcium channels (28), it is expected that the block of SK channels and block of calcium channels will combine to produce additive attenuation of neuronal sAHPs.
We also examined the effect of inhaled anesthetics on recombinant SK2 channels. None of the inhaled anesthetics, applied at clinical (1 MAC) or supraclinical (10 MAC) concentrations, altered the magnitude of SK2 currents significantly. In general, however, larger concentrations of inhaled anesthetics depressed the SK channels more, except chloroform, which, interestingly, caused a small potentiation. This effect of chloroform is reminiscent of its action on recombinant Shaker B voltage-gated potassium channels (29). In our study, halothane was without effect, whereas chloroform and isoflurane had opposite effects on the SK potassium channels. In general, our data showing that inhaled anesthetics depress SK channels are in agreement with previous work on other types of potassium channels, including those gated by membrane voltage (29), large conductance, voltage- and calcium-activated (30), and G proteins (31). Most significantly, Scharff and Foder (32) showed that halothane inhibits the Gardos channel in red blood cells; this channel (SK4) has recently been cloned (12) and shown to be similar in sequence and properties to the channel in our study. Moreover, several studies have shown that inhaled anesthetics inhibit voltage-gated calcium channels, which normally would mediate the calcium influx necessary for SK channel activation (e.g., see ref. 33).
In conclusion, we have reported the interactions of anesthetic molecules on recombinant SK channels. Our study, to our knowledge, is the first such study on recombinant SK channels. Our approach follows a recent tradition of investigating drug action at ion channels by using recombinant systems. The obvious strengths of this approach, simplicity and reproducibility, are countered by its limitations, i.e., it is overly reductive. Whereas we studied the effects on a homomeric recombinant channel, its native counterpart may be a heteromer. Furthermore, the membrane microenvironment for the recombinant SK channels in our expression system is likely to be different from that of a myelinated neuron, affecting the phase partitioning of the anesthetics and their apparent affinity. The two functions of SK channels, regulation of electrical excitability in neurons and hormone release in neuroendocrine cells, depend on the interplay between different cellular components, including calcium influx, other modulatory receptors, and protein phosphorylation reactions, as well as membrane voltage. It follows that the effects of anesthetics on these SK channel-mediated cellular events would be the sum of the anesthetic sensitivities of all of the cellular components involved. Our study did not address the molecular mechanisms of SK channel block by anesthetics. Moreover, several SK channel subtypes have now been identified in neurons, smooth muscle, and other tissues. Clearly, it would be desirable to investigate the mechanisms of block and the SK channel-subtype specificity of anesthetic action.
| Acknowledgments |
|---|
The authors would like to thank Neil Harrison, for support, and L. K. Kaczmarek and W. Joiner (Yale University) for providing materials.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. C. Button, K. Gardiner, T. Marqueste, and P. F. Gardiner Frequency-current relationships of rat hindlimb {alpha}-motoneurones J. Physiol., June 15, 2006; 573(3): 663 - 677. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Han, N. Kim, H. Joo, and E. Kim Ketamine blocks Ca2+-activated K+ channels in rabbit cerebral arterial smooth muscle cells Am J Physiol Heart Circ Physiol, August 7, 2003; 285(3): H1347 - H1355. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J. Kendig In vitro networks: subcortical mechanisms of anaesthetic action Br. J. Anaesth., July 1, 2002; 89(1): 91 - 101. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. W. Urban Current assessment of targets and theories of anaesthesia Br. J. Anaesth., July 1, 2002; 89(1): 167 - 183. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y.-J. Cao, J. C. Dreixler, J. D. Roizen, M. T. Roberts, and K. M. Houamed Modulation of Recombinant Small-Conductance Ca2+-Activated K+ Channels by the Muscle Relaxant Chlorzoxazone and Structurally Related Compounds J. Pharmacol. Exp. Ther., March 1, 2001; 296(3): 683 - 689. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|