An increase in intracellular calcium concentration ([Ca2+]i) in neurons has been proposed as an important effect of volatile anesthetics, because they alter signaling pathways that influence neurotransmission. However, the existing data for anesthetic-induced increases in [Ca2+]i conflict.


Changes in [Ca2+]i were measured using fura-2 fluorescence spectroscopy in rat cortical brain slices at 90, 185, 370, and 705 microM isoflurane. To define the causes of an increase in [Ca2+]i, slices were studied in Ca2+-free medium, in the presence of Ca2+-channel blockers, and in the presence of the Ca2+-release inhibitor azumolene. The authors compared the effect of the volatile anesthetic with that of the nonanesthetic compound 1,2-dichlorohexafluorocyclobutane. Single-dose experiments in CA1 neurons in hippocampal slices with halothane (360 microM) and in acutely dissociated CA1 neurons with halothane (360 microM) and isoflurane (445 microM) also were performed.


Isoflurane at 0.5, 1, and 2 minimum alveolar concentrations increased basal [Ca2+]i in cortical slices in a dose-dependent manner (P < 0.05). This increase was not altered by Ca2+-channel blockers or Ca2+-free medium but was reduced 85% by azumolene. The nonanesthetic 1,2-dichlorohexafluorocyclobutane did not increase [Ca2+]i. In dissociated CA1 neurons, isoflurane reversibly increased basal [Ca2+]i by 15 nM (P < 0.05). Halothane increased [Ca2+]i in dissociated CA1 neurons and CA1 neurons in hippocampal slices by approximately 30 nM (P < 0.05).


(1) Isoflurane and halothane reversibly increase [Ca2+]i in isolated neurons and in neurons within brain slices. (2) The increase in [Ca2+]i is caused primarily by release from intracellular stores. (3) Increases in [Ca2+]i occur with anesthetics but not with the nonanesthetic 1,2-dichlorohexafluorocyclobutane.

VOLATILE anesthetics appear to produce their effects by changing the activity of ion channels involved in neurotransmission. A direct interaction of the anesthetic molecule with an ion channel protein at a lipophilic interface is the most commonly proposed mechanism. [1,2]Alternatively, anesthetics may have an indirect effect, activating or inhibiting signaling pathways that regulate ion channel function.

One signaling molecule with potentially substantial effects on various ion channels is intracellular calcium ([Ca2+]i). Previous studies have shown that increased [Ca2+]imay activate potassium channels, [3]potentiate [Greek small letter gamma]-aminobutyric acid currents, [4]and depress both calcium channel [5,6]and N-methyl-D-aspartate receptor (NMDA) currents [7]via Ca2+-dependentmechanisms. Elevations in [Ca2+]i, therefore, could explain the augmentation of inhibitory processes and the inhibition of excitatory pathways that characterize the state of anesthesia at various foci. However, the evidence regarding the anesthetic effect on [Ca2+]iis controversial. We are aware of only two other studies that have shown an increase in [Ca2+]iin neurons produced by volatile anesthetics. Bicker et al. [8]found in cortical neurons that increases in [Ca2+](i) induced by approximately 1 minimum alveolar concentration (MAC) isoflurane were larger at higher temperatures. Recently, Franks et al. [9]reported that halothane, isoflurane, and xenon significantly increased [Ca (2+)]iin mouse cortical neurons. In contrast, in other experiments with cultured neurons, investigators concluded that anesthetics did not increase basal [Ca2+]i, [10,11]and studies in mouse brain synaptosomes showed that volatile anesthetics increased [Ca2+]ionly at very high concentrations in the millimolar range. [12]Based on these negative reports, Franks and Lieb [1]dismissed the possibility that volatile anesthetics increase [Ca2+]iin neurons.

The goal of the current study was to determine again whether volatile anesthetics increase [Ca2+]iin central nervous system neurons. Therefore, we measured the effects of isoflurane, halothane, and the nonanesthetic 1,2-dichlorohexafluorocyclobutane (2N) on [Ca2+]iin cortical brain slices, CA1 neurons in intact hippocampal brain slices, and acutely dissociated CA1 neurons with the fluorescent Ca2+indicator fura-2. Using both models allowed us to determine whether observed elevations in [Ca2+]iwere caused by increases in [Ca2+]iin neurons or simply by changes occurring in nonneuronal cells in the slices. We also determined whether increases in [Ca2+]iresulted from the influx of Ca2+from the extracellular fluid or from Ca2+release from intracellular stores. Although little is known about the role of the Ca2+-releasechannels in the central nervous system, a recent study described for the first time the distribution of the ryanodine receptor isoforms in the human brain. [13]The widespread expression of all isoforms (ryr 1, ryr 2, and ryr 3) in human hippocampus suggests a role of this gene family in Ca2+signaling, Ca2+homeostasis, and possibly also in fundamental processes such as synaptic plasticity. Finally, we determined whether the effects of a nonanesthetic compound (1,2-dichlorohexafluorocyclobutane) were similar to those of volatile anesthetics.

Preparation of Brain Slices and Dissociated Neurons

Neocortical brain slices were prepared from 13- to 23-day-old Sprague-Dawley rats according to methods that were approved by the Committee on Animal Research, University of California, San Francisco. After decapitation during 2% halothane anesthesia, brain hemispheres were dissected rapidly and slices were prepared in ice-cold artificial cerebrospinal fluid (aCSF). The aCSF used to prepare and maintain the slices consisted of 116 mM NaCl, 25 mM NaHCO3, 5.4 mM KCl, 1.8 CaCl2mM, 0.9 mM MgCl2, 0.9 mM NaH2PO4, and 10 mM glucose, pH 7.40-7.45, oxygenated with 95% oxygen and 5% carbon dioxide. Transverse cortical slices were prepared (300 [micro sign]m thick) using a vibratome (Campden Instruments, Cambridge, UK), and hippocampal slices were prepared using a tissue chopper (Stoelting Co., Wood Dale, IL). Typically, 8-12 slices were obtained from each animal, and only a few slices within one study group were obtained from the same animal. Slices were stored at room temperature for 45 min to recover from the trauma of slicing and then transferred to vials of oxygenated aCSF containing 2 [micro sign]M fura-2 acetoxymethyl ester (Molecular Probes, Eugene, OR) to permit measurement of [Ca2+]i. Fura-2 is primarily a cytosolic dye used to mark cytosolic changes in calcium. It was dissolved in 20% F-127 pluronic acid (Molecular Probes) in anhydrous dimethyl sulfoxide. After 1 h, slices were transferred to fresh oxygenated aCSF containing 2 [micro sign]M fura-2 for an additional 30 min to facilitate dye loading. [14]Fluorescence signals from fura-loaded slices were five to seven times greater than background fluorescence.

Dissociated neurons were prepared by enzymatic digestion of dissected CA1 regions of hippocampal slices of 1- to 14-day-old rats in 0.1% trypsin and 0.05% pronase in Ca2+-freebuffer for 30 min at 37 [degree sign]C. [15]After gentle tituration with a fire-polished pipette, neurons were applied to glass coverslips and loaded with 1-3 [micro sign]M fura-2 for 15 min.

Measurements of [Ca2+]i

Neocortical brain slices were studied in a Hitachi F-2000 fluorometer (Tokyo, Japan), as described previously. [8,16]Slices were fixed to a nylon mesh holder and placed in a capped quartz fluorometer cuvette containing oxygenated aCSF and a stir bar. [Ca2+]iwas determined by dual-excitation fluorescence spectroscopy at 37 [degree sign]C. Slices were excited alternately at 340 and 380 nm, and the emitted light intensity at 510 nm was recorded every 0.5 s. [Ca2+]iwas estimated using the Grynkiewicz equation [17]:Equation 1where R is the fluorescence emission ratio (340:380 nm), Kdis the temperature-corrected dissociation constant of fura-2 (Kd= 224 nM), [18]F2min/F2maxis the ratio between the 380-nm fluorescence intensities in the absence of Ca2+and in the presence of saturating [Ca (2+)]i, and Rminand Rmaxare the fluorescence ratios in the absence of Ca2+and in the presence of saturating [Ca2+]i, respectively. Tissue autofluorescence (fluorescence in the absence of fura-2) was always subtracted before computing the ratios. F2min, F2max, R (min), and Rmaxwere determined as follows: Fluorescence emission at saturating [Ca2+] was measured in slices after treatment with 10 [micro sign]M ionomycin (a calcium ionophore) and 1 mM ouabain. Fluorescence emission at 0 [Ca2+] was measured by replacing the aCSF with calcium-free aCSF containing 5 mM EGTA.

[Ca2+]iin CA1 neurons in hippocampal slices and in acutely dissociated CA1 neurons was measured using a PTI D104 fluorometer-inverted microscope system (PTI, South Brunswick, NJ) as described in detail before. [19]Hippocampal slices or coverslips with plated neurons were placed in a 200-[micro sign]l temperature-controlled (37 [degree sign]C) recording chamber on the stage of an inverted Nikon Diaphot 200 microscope (Nikon Inc. Instrument Group, Melville, NY). The CA1 region of the hippocampal slices was viewed using a microscope. The sealed recording chamber was perfused continuously with aCSF bubbled with 95% oxygen and 5% carbon dioxide at a rate of 4 ml/min. Isoflurane and halothane were added by switching the perfusate from standard aCSF to aCSF bubbled with anesthetic vapor (using a standard calibrated vaporizer) for at least 30 min. Measurements of [Ca2+]iwere acquired from regions of interest in the hippocampal slices (CA1 neurons in a 50 x 10 [micro sign]M region) or from individual dissociated neurons by adjusting slit apertures in the photomultiplier tube section of the fluorometer-microscope system. Dissociated neurons were identified morphologically by their pyramidal cell body and prominent processes before they were studied, and the slit aperture was focused on the cell body. Calibration experiments and calculation of [Ca (2+)]iwere performed as described before.

Administration and Determination of Anesthetics

After recording baseline fluorescence for 50-100 s, aCSF saturated with isoflurane or 2N was injected by syringe into the cuvette of the Hitachi F-2000 fluorometer. The actual concentration of isoflurane or 2N in the cuvette was measured by gas chromatography. Trial experiments with isoflurane and 2N were conducted to determine the actual agent concentration and the amount of agent lost from the experimental apparatus. Therefore, different volumes of agent-saturated aCSF were injected into a cuvette containing aCSF to a total volume of 2 ml. Samples of the solutions were taken from the cuvette at the beginning and at the end of the experiments. Samples were equilibrated in air in glass syringes in a constant-temperature water bath. Gas in these syringes was analyzed using a gas chromatograph (Gow Mac, Bridgewater, NJ) calibrated according to the appropriate standards. Measurements at the beginning and end of the experiment showed a loss of approximately 15% for isoflurane and 25-30% for 2N during the experiment. Based on these determinations, the volume of agent-saturated aCSF was adjusted to give target anesthetic concentrations (based on MAC values for rats given by Franks and Lieb [20]) at the middle time point of the studies. 1,2-dichlorohexafluorocyclobutane was purchased from Lancaster Synthesis (Windham, NH).

Losses from the fluorometer-inverted microscope system also were determined in trial experiments. The anesthetic concentrations in the perfusate in the recording chamber were approximately 0.8 of the concentrations expected, when aCSF was equilibrated with a given setting on the vaporizer. The actual concentrations of halothane and isoflurane in the recording chamber also were determined by gas chromatography, as described before.

Experimental Design

Isoflurane-induced changes in [Ca2+]iin cortical brain slices were studied using the Hitachi F-2000 fluorometer under the experimental conditions:(1) at 90, 185, 370, and 705 [micro sign]M isoflurane (equivalent to approximately 0.25, 0.5, 1, and 2 MAC, respectively) and standard aCSF (containing 1.8 mM Ca2+);(2) at 0.5, 1, and 2 MAC isoflurane and standard aCSF containing agents to specifically block the N-type and P/Q-type Ca2+channels and Na+channels (0.5 [micro sign]M [Greek small letter omega]-conotoxin GVIA [Research Biochemicals, Natick, MA], 0.1 [micro sign]M [Greek small letter omega]-agatoxin IVa [Pfizer, Groton, NJ], and 0.5 [micro sign]M tetrodotoxin [Sigma, St. Louis, MO], respectively). These blockers prevent more than 90% of changes in [Ca2+]iresponse to a depolarizing stimulus in this preparation [21];(3) at 0.5, 1, and 2 MAC isoflurane in Ca2+-freeaCSF (Ca2+< 1 [micro sign]M; Corning ion selective electrode #476041; Corning Incorporated Science Products Division, Corning, NY) to determine whether Ca2+influx from extracellular medium contributed to the increase in [Ca2+]iproduced by isoflurane;(4) at 2 MAC isoflurane with standard aCSF after 30 min of incubation with 100 [micro sign]M azumolene (1[[[5-(4-bromophenyl)-2-oxazolyl]methylene]amino]-2,4-imidazolidine-dione), a dantrolene analog with less fluorescence at 340 and 380 nm and no fluorescence at the emission wave-length of 510 nm (HP 8452 Diode Array Spectrophotometer, Hewlett-Packard, Palo Alto, CA)(this treatment was chosen to block possible isoflurane-mediated release of Ca2+from intracellular stores [22]); and (5) exposure to 30 and 60 [micro sign]M 2N, corresponding to approximately 1.5 and 3 times the MAC value for this compound predicted by the Meyer-Overton hypothesis based on lipid solubility. [23] 

CA1 neurons in intact hippocampal slices were studied at 360 [micro sign]M halothane (approximately 1.2 MAC), and cell bodies of dissociated neurons were studied at 360 [micro sign]M halothane (approximately 1.2 MAC) and at 445 [micro sign]M isoflurane (approximately 1.3 MAC) in the microscope fluorometer with an attached perfusion system that allowed the anesthetic to be washed out from the slices or neurons.


Results are reported as the mean +/− SD. The n values correspond to the number of brain slices studied. Within each study group, data were analyzed using the paired Student t test for comparisons between baseline [Ca (2+)]iand [Ca2+]iafter exposure to anesthetic. To test for a dose-dependent effect of the increase of [Ca2+]i, linear regression analysis was used (JMP; SAS Institute, Cary, NC). In addition, analysis of variance was used for comparison of the baseline [Ca2+]iwith all treatment groups. P < 0.05 was considered significant.

[Ca2+]iin Neocortical Brain Slices

(Figure 1, A and B) show tracings of the ratio of the two excitation wavelengths (340:380 nm) and [Ca2+]i, respectively, before and during exposure of a cortical brain slice to 705 [micro sign]M isoflurane (approximately 2 MAC). [Ca2+]iin untreated brain slices during control conditions (standard aCSF) was 146 +/− 41 nM (n = 67;Figure 2A), which is similar to previous studies. [8]An increase in [Ca2+](i) after exposure to 2 MAC isoflurane was observed in all slices (Figure 1C). Exposure to 0.5, 1, and 2 MAC isoflurane increased [Ca2+]iby 8 +/− 11 nM, 23 +/− 19 nM, and 22 +/− 15 nM (mean +/− SD), respectively (Figure 2A). These changes in [Ca2+]iwere dose dependent (P < 0.05) and represent an increase from baseline [Ca2+]i of 5-15%. No increase in [Ca2+]i was observed with 0.25 MAC isoflurane (149 +/− 41 nM vs. 149 +/− 53 nM).

Relative to untreated slices, baseline [Ca2+]iin cerebrocortical slices did not differ after pretreatment of slices with 0.5 [micro sign]M [Greek small letter omega]-conotoxin GVIA, 0.1 [micro sign]M [Greek small letter omega]-agatoxin IVa, and 0.5 [micro sign]M tetrodotoxin. Exposure to 0.5, 1, and 2 MAC isoflurane increased [Ca2+]iby 28 +/− 16, 32 +/− 20, and 41 +/− 33 nM, respectively, similar to that in untreated slices (Figure 2B).

In calcium-free extracellular solution, baseline [Ca2+]idid not differ from that in slices in 1.8 mM Ca2+(P > 0.05 by analysis of variance), and 1 and 2 MAC isoflurane increased [Ca2+]iby 9 +/− 8 nM and 31 +/− 29 mM, respectively (P < 0.05;Figure 2C).

In azumolene-pretreated slices, the increase in [Ca2+]iat 2 MAC isoflurane was significantly smaller than that in untreated slices (5 +/− 13 nM vs. 22 +/− 15 nM, respectively; P < 0.05;Figure 3).

We also exposed cortical brain slices to the nonanesthetic compound 2N. At concentrations as great as three times the predicted MAC, this compound had no effect on [Ca2+]i(151 +/− 51 vs. 156 +/− 52 nM; n = 16; P > 0.05). Because of the very low saline-gas partition coefficient at 37 [degree sign]C (0.00011), [23]we injected a relatively large volume (800 [micro sign]l) of 2N-saturated saline into the cuvette to achieve three times the predicted MAC. To control whether brain slices remained adequately oxygenated during this procedure, we measured the oxygen tension in the cuvette using an oxygen electrode (Cameron Instrument Co., Port Aransas, TX). The mean oxygen tension after addition of 800 [micro sign]l 2N-saturated solution was 236 +/− 13 mmHg (n = 4).

Hippocampal Slices and Dissociated CA1 Neurons

[Ca2+]iwas measured in the CA1 cell bodies within intact hippocampal slices using the microscope fluorometer. Baseline [Ca2+]iduring perfusion of slices with standard aCSF without additional stimulus was 108 +/− 37 nM and increased to 139 +/− 48 nM by superfusing the slices with aCSF bubbled with 1.2 MAC halothane (n = 12; P < 0.05). This increase was reversible after the anesthetic was washed out (Figure 4).

[Ca2+]ialso was measured in acutely dissociated CA1 neurons from hippocampi of 1- to 14-day-old rats. Neurons were identified by their shape, long cell processes, and phase-bright appearance. Baseline [Ca (2+)]iwas 186 +/− 48 nM (n = 14) in these dissociated neurons. In all neurons, [Ca2+]iincreased with exposure to either 1.2 MAC halothane (29 +/− 17 nM; n = 7) or 1.3 MAC isoflurane (15 +/− 9 nM; n = 7)(P < 0.05). As in the hippocampal slices, the increases of [Ca2+]iin the dissociated neurons were reversible with washout of the anesthetics (Figure 5).

Previous reports of the effect of volatile anesthetics on [Ca2+]iare contradictory. Two studies have shown an anesthetic-induced increase in [Ca2+]iin neurons, [8,9]whereas other investigators found no effect or one only at high concentrations. [10-12]Several factors might have contributed to a failure to detect an increase in [Ca2+]i. First, cultured neurons may be vulnerable to a loss of sensitivity to anesthetics, possibly by downregulation of volatile anesthetic-sensitive Ca2+-adenosine triphosphatase (ATPase) pumps. Second, studies with negative results [10,11]used relatively high concentrations of fura-2 (approximately 5 [micro sign]M), which can buffer small calcium changes. Finally, increases in [Ca2+]iare temperature dependent, [8,9]suggesting that the use of room temperature conditions in both studies may have masked the detection of any change in [Ca (2+)]i.

Several studies have proposed that increases in [Ca2+]imay be responsible in part for the actions of volatile anesthetics in neurons. Krnjevic [3]suggested that anesthetics might increase K+-conductanceby increasing [Ca2+]iin neurons. Nicoll and Madison [24]observed such hyperpolarization for various anesthetic compounds. Mody et al. [4]suggested that augmented [Greek small letter gamma]-aminobutyric acid A currents induced by volatile anesthetics might result from increased [Ca2+]ibecause BAPTA, a calcium chelator, reduced the augmentation. In other tissues, volatile anesthetics are recognized more widely to increase [Ca2+]i. [25-27] 

The mechanism by which anesthetics increase [Ca2+]iin neurons remains unclear. Our results suggest that they act predominately by inducing a release from intracellular sites rather than a Ca2+-influxfrom the extracellular medium. That is, we found that eliminating Ca2+from the extracellular medium and blocking common voltage-gated Ca2+channels of neurons did not affect the increase in [Ca2+]icaused by isoflurane. Studies of the effect of ethanol in hippocampal neurons [28]also found that neither removal of extracellular calcium nor blockade of calcium channels could prevent the agent-induced increase in [Ca2+](i), indicating intracellular rather than extracellular origins. In L6 muscle cells, approximately 50% of the halothane-induced increase in [Ca2+]iis released from intracellular stores. [26] 

In our experiments, we could, at least in part, block the isoflurane-induced increase of [Ca2+]iby pretreating the brain slices with azumolene. A water-soluble dantrolene analog, azumolene, has been shown to cause approximately 30% inhibition of Ca2+release from muscle sarcoplasmic reticulum. [22]Although the roles of Ca2+-releasechannels are well established for skeletal and cardiac muscle, less is known about their roles in the human central nervous system. A recent study showed widespread expression of all ryanodine receptor isoforms in human hippocampus, implying an important role of this gene family in Ca2+homeostasis and Ca2+-signalingpathways. [13]It is intriguing to assume that dantrolene and azumolene also block neuronal intracellular Ca2+-releasechannels, thereby diminishing the observed isoflurane-induced increase of [Ca2+]i.

A second possible mechanism is the inhibition of plasma membrane Ca2+-ATPase, resulting in calcium accumulation in the cytosol. [9]Plasma membrane Ca2+-ATPaseis highly conserved phyllogenetically and is present in various tissues, which may explain the diverse actions of volatile anesthetics. Franks et al. [9]and Janicki et al. [29]reported that anesthetics inhibit plasma membrane Ca2+-ATPase, thereby contributing to the anesthetic-induced increase in [Ca2+]i. However, inhibition of a single pump mechanism would not seem to be a sufficient explanation for the increase in [Ca2+]i, because other mechanisms regulate the resting [Ca2+]i. [30]Calcium disposal mechanisms are unlikely to be saturated by increases in [Ca2+]iin the range of 30 nM. It seems more likely that anesthetics alter the set point of [Ca2+]iregulation, such that a higher level of resting [Ca2+]ioccurs in the presence of the anesthetic. Although the observed changes of [Ca2+](i) are small, they are similar to those associated with synaptic activity. [31]It has been shown that an increase in [Ca2+]iof 10-30 nM produced a twofold synaptic enhancement. [32] 

An increase in [Ca2+]icaused by volatile anesthetics might be related to anesthesia because (1) the increases in [Ca2+]iare significant in size, at approximately one tenth of those that occur with maximal stimulation of NMDA receptors or activation of voltage-gated calcium channels with a depolarizing stimulus of 50 mM KCl [8,11,21];(2) the changes in [Ca2+]iproduced by halothane and isoflurane are reversible, matching the rapid reversal of clinical anesthesia;(3) the changes in [Ca2+]iare dose dependent and occur in the clinical range of anesthetic concentrations;(4) the increase in [Ca2+]ihas been observed in cortical and hippocampal neurons, sites possibly involved as loci contributing to the anesthetic state in intact animals;(5) an increase in [Ca2+]idid not occur with 2N, a nonanesthetic compound; and (6) an increase in [Ca2+]iprovides a plausible link to a myriad of processes already shown to be involved in some actions of anesthetics, including allosteric modulation of important ion channels via phosphorylation control pathways.

Inhibition of neurotransmission at central synapses is likely to be an important locus of anesthetic action. Anesthetics might interrupt synaptic transmission by acting presynaptically to cause a Ca2+-dependentinhibition of neurotransmitter release. It also has been suggested that halothane potentiated inhibitory postsynaptic currents via increased [Ca2+]i, based on the partial prevention of depression by the calcium buffer 1,2bis(2-aminophenoxy)ethane-N,N,N'-tetraacetic acid or the calcium release inhibitor dantrolene. [4]However, other investigators could not confirm such a role for [Ca2+]iin modulating inhibitory synapses in cultured rat hippocampal neurons. [33] 

Elevated [Ca2+]ihas important inhibitory influences in the postsynapse. The activity of the NMDA receptor is decreased by increased [Ca2+]i. Inhibition of this receptor, a proposed target for anesthetic action, [11]is controlled by calcium-bound calmodulin, which activates calcineurin. [34]Rosenmund and Westbrook [7]showed that increased [Ca2+]idecreases NMDA channel activity by increasing actin depolymerization. With respect to volatile anesthetics, elevated [Ca (2+)]icould tip the balance from an active to an inactive NMDA receptor.

Our results differ from those of previous studies that found a decrease of [Ca2+]iproduced by volatile anesthetics and barbiturates. [35,36]However, both these studies examined the effect of anesthetics on a potassium chloride-evoked increase in [Ca2+]iand not from changes of baseline [Ca2+]i. We propose that a small increase in [Ca2+]iinduced by volatile anesthetics may have complex effects on the regulation of ion channels by means of calcineurine phosphatases, and Ca2+-calmodulin-dependentprotein kinase II. [37]An elevated [Ca2+]ican cause enhanced inhibition (e.g., Ca2+-sensitivepotassium channel currents and [Greek small letter gamma]-aminobutyric acid A receptor currents) and decreased excitation (inhibition of excitatory neurotransmitter release, inhibition of postsynaptic receptors).

Study Limitations

The limitations of measuring [Ca2+]iin brain slices are well recognized. Most important is the injury layer in each slice, which may respond differently to volatile anesthetics than undamaged tissue does. In addition, the calibration of [Ca2+]imeasurements in slices is difficult because of the challenges imposed by equilibrating intracellular and extracellular pools of Ca2+. We overcame both these difficulties, in part, by also studying morphologically intact dissociated neurons in which Ca (2+) levels are more easily calibrated. The fact that volatile anesthetics produced increases in [Ca2+]iby amounts in cortical and hippocampal slices similar to those in dissociated neurons indicates that data in slices are representative of changes in [Ca2+]i. This suggests that the changes in [Ca2+]iobserved in neurons in intact slices were not due to the signal contribution from nonneuronal cells or from neurotransmitter release, because the very low density of cells on the coverslips used for the dissociated neurons and the constant perfusion system would have eliminated any released neurotransmitter.

Our results suggest that (1) volatile anesthetics reversibly increase [Ca2+]iin central nervous system neurons, (2) the increase of [Ca2+]iis, in part, the result of a release of Ca2+from intracellular stores, and (3) the increase in [Ca2+]iis not observed in a compound similar in structure to anesthetics but devoid of anesthetic potency.

The authors thank Pompili Ionescu, M.D., for his assistance with the gas chromatographic analysis; Robert Brooks, Ph.D. (Procter and Gamble Pharmaceuticals, Norwich, NY) for providing azumolene; Edmond Eger II, M.D., for providing the nonanesthetic compound 2N; Dennis Fisher, M.D., for statistical advice; and Winifred von Ehrenburg for editorial assistance.

Franks NP, Lieb WR: Molecular and cellular mechanisms of general anaesthesia. Nature 1994; 367:607-14
Franks NP, Lieb WR: Inhibitory synapses. Anaesthetics set their sites on ion channels. Nature 1997; 389:334-5
Krnjevic K: Central actions of general anaesthetics, Molecular Mechanisms in General Anaesthesia. Edited by MJ Halsey, RA Millar, JA Sutton. Edinburgh, Churchill Livingstone, 1974, pp 65-89
Mody I, Tanelian DL, MacIver MB: Halothane enhances tonic neuronal inhibition by elevating intracellular calcium. Brain Res 1991; 538:319-23
Johnson BD, Byerly L: Ca2+channel Ca2+-dependentinactivation in a mammalian central neuron involves the cytoskeleton. Pflugers Arch 1994; 429:14-21
Fomina AF, Levitan ES: Control of Ca2+channel current and exocytosis in rat lactotrophs by basally active protein kinase C and calcineurin. Neuroscience 1997; 78:523-31
Rosenmund C, Westbrook GL: Calcium-induced actin depolymerization reduces NMDA channel activity. Neuron 1993; 10:805-14
Bickler PE, Buck LT, Hansen BM: Effects of isoflurane and hypothermia on glutamate receptor-mediated calcium influx in brain slices. Anesthesiology 1994; 81:1461-9
Franks JJ, Wamil AW, Janicki PK, Horn JL, Franks WT, Janson VE, Vanaman TC, Brandt PC: Anesthetic-induced alteration of Ca2+homeostasis in neural cells: A temperature-sensitive process that is enhanced by blockade of plasma membrane Ca2+-ATPaseisoforms. Anesthesiology 1998; 89:149-64
Bleakman D, Jones MV, Harrison NL: The effects of four anesthetics on intracellular [Ca2+] on cultured rat hippocampal neurons. Neuropharmacology 1995; 34:541-51
Puil E, el-Beheiry H, Baimbridge KG: Anesthetic effects on glutamate-stimulated increase in intraneuronal calcium. J Pharmacol Exp Ther 1990; 255:955-61
Daniell LC, Harris RA: Neuronal intracellular calcium concentrations are altered by anesthetics: Relationship to membrane fluidization. J Pharmacol Exp Ther 1988; 245:1-7
Martin C, Chapman KE, Seckl JR, Ashley RH: Partial cloning and differential expression of ryanodine receptor/calcium-release channel genes in human tissues including the hippocampus and cerebellum. Neuroscience 1998; 85:205-16
Gray AT, Buck LT, Feiner JR, Bickler PE: Interactive effects of pH and temperature on N-methyl-D-aspartate receptor activity in rat cortical brain slices. J Neurosurg Anesthesiol 1997; 9:180-7
Friedman JE, Haddad GG: Major differences in Cai 2+response to anoxia between neonatal and adult rat CA1 neurons: Role of Cao 2+and Nao 2+. J Neurosci 1993; 13:63-72
Eilers H, Bickler PE: Hypothermia and isoflurane similarly inhibit glutamate release evoked by chemical anoxia in rat cortical brain slices. Anesthesiology 1996; 85:600-7
Grynkiewicz G, Poenie M, Tsien RY: A new generation of Ca2+indicators with greatly improved fluorescence properties. J Biol Chem 1985; 260:3440-50
Shuttleworth TJ, Thompson JL: Effect of temperature on receptor-activated changes in [Ca2+]iand their determination using fluorescent probes. J Biol Chem 1991; 266:1410-4
Bickler PE, Hansen BM: Hypoxia-tolerant neonatal CA1 neurons: Relationship of survival to evoked glutamate release and glutamate receptor-mediated calcium changes in hippocampal slices. Dev Brain Res 1998; 106:57-69
Franks NP, Lieb WR: Selective actions of volatile general anaesthetics at molecular and cellular levels. Br J Anaesth 1993; 71:65-76
Hampson AJ, Bornheim LM, Scanziani M, Yost CS, Gray AT, Hansen BM, Leonoudakis DJ, Bickler PE: Dual effects of anandamide on NMDA receptor-mediated responses and neurotransmission. J Neurochem 1998; 70:671-6
Tian Q, Katz AM, Kim DH: Effects of azumolene on doxorubicin-induced induced Ca2+release from skeletal and cardiac muscle sarcoplasmic reticulum. Biochim Biophys Acta 1991; 1094:27-34
Koblin DD, Chortkoff BS, Laster MJ, Eger EI II, Halsey MJ, Ionescu P: Polyhalogenated and perfluorinated compounds that disobey the Meyer-Overton hypothesis. Anesth Analg 1994; 79:1043-8
Nicoll RA, Madison DV: General anesthetics hyperpolarize neurons in the vertebrate central nervous system. Science 1982; 217:1055-7
Iaizzo PA, Olsen RA, Seewald MJ, Powis G, Stier A, Van Dyke RA: Transient increases of intracellular Ca2+induced by volatile anesthetics in rat hepatocytes. Cell Calcium 1990; 11:515-24
Klip A, Hill M, Ramlal T: Halothane increases cytosolic Ca (2+) and inhibits Na+/H +exchange in L6 muscle cells. J Pharmacol Exp Ther 1990; 254:552-9
Wheeler DM, Rice RT, Hansford RG, Lakatta EG: The effect of halothane on the free intracellular calcium concentration of isolated rat heart cells. Anesthesiology 1988; 69:578-83
Mironov SL, Hermann A: Ethanol actions on the mechanisms of Ca2+mobilization in rat hippocampal cells are mediated by protein kinase C. Brain Res 1996; 714:27-37
Janicki PK, Horn JL, Singh G, Janson VE, Franks WT, Franks JJ: Reduced anesthetic requirements, diminished brain plasma membrane Ca2+-ATPasepumping, and enhanced brain synaptic plasma membrane phospholipid methylation in diabetic rats: effects of insulin. Life Sci 1995; 56:L357-63.
Ghosh A, Greenberg ME: Calcium signaling in neurons: Molecular mechanisms and cellular consequences. Science 1995; 268:239-47
Regehr WG, Tank DW: Dendritic calcium dynamics. Curr Opin Neurobiol 1994; 4:373-82
Regehr WG, Delaney KR, Tank DW: The role of presynaptic calcium in short-term enhancement at the hippocampal mossy fiber synapse. J Neurosci 1994; 14:523-37
Jones MV, Harrison NL: Effects of volatile anesthetics on the kinetics of inhibitory postsynaptic currents in cultured rat hippocampal neurons. J Neurophysiol 1993; 70:1339-49.
Lieberman DN, Mody I: Regulation of NMDA channel function by endogenous CA2+-dependent phosphatase. Nature 1994; 369:235-9
Miao N, Frazer MJ, Lynch C: Volatile anesthetics depress Ca (2+) transients and glutamate release in isolated cerebral synaptosomes. Anesthesiology 1995; 83:593-603
Miao N, Nagao K, Lynch C: Thiopental and methohexital depress Ca2+entry into and glutamate release from cultured neurons. Anesthesiology 1998; 88:1643-53
Braun AP, Schulman H: The multifunctional calcium/calmodulin-dependent protein kinase: From form to function. Ann Rev Physiol 1995; 57:417-45