Ca2+ influx is a key component of neuronal intracellular Ca2+ ([Ca2+]i) regulation. The authors hypothesized that volatile anesthetic inhibition of neuronal activity is mediated by inhibition of Ca2+ influx via two major mechanisms: plasma membrane Na+-Ca2+ exchange (NCX) and the novel mechanism of Ca2+ influx triggered by endoplasmic reticulum Ca2+ depletion (store-operated Ca2+ channels [SOCCs]).
Differentiated rat pheochromocytoma cells loaded with the Ca2+ indicator fura-2 were Na+-loaded with 0 Ca2+, 145 mm Na+ Tyrode's and 5 microm cyclopiazonic acid plus 10 microm ryanodine (functionally isolating plasma membrane). Influx-mode NCX was rapidly reactivated by 0 Na+ and 2.5 mm Ca2+. The protocol was repeated in the presence of volatile anesthetics (0.5-1.5 minimum alveolar concentration [MAC] halothane, isoflurane, or sevoflurane) or other drugs to characterize NCX. To examine SOCCs, endoplasmic reticulum Ca2+ was depleted by cyclopiazonic acid in 0 extracellular Ca2+, and Ca2+ influx was triggered by rapid reintroduction of extracellular Ca2+. The protocol was repeated in the presence of anesthetics or other drugs to characterize SOCCs.
Influx via NCX was not inhibited by voltage-gated Ca2+ channel blockers but was sensitive to NCX inhibitors. Halothane and isoflurane (0.5-1.5 MAC) significantly inhibited NCX (P < 0.05; paired comparisons), whereas sevoflurane at less than 1.5 MAC did not inhibit NCX. SOCC-mediated Ca2+ influx was insensitive to a variety of Ca2+ channel blockers but was inhibited by Ni2+. Such influx was sensitive only to halothane at greater than 1 MAC but not isoflurane or sevoflurane.
These data indicate that volatile anesthetics, especially halothane and isoflurane, interfere with neuronal [Ca2+]i regulation by inhibiting NCX but not SOCC-mediated Ca2+ influx (except high concentrations of halothane).
VOLATILE anesthetics influence Ca2+regulation in a number of cell systems. Nowhere are these effects more important than in neuronal cells in the action of volatile anesthetics. Understanding anesthetic effects on intracellular Ca2+([Ca2+]i) regulation is important in at least two ways: (1) Synaptic transmission in itself is modulated by local Ca2+concentrations at the presynaptic terminal,1–5and (2) global Ca2+modulates expression and regulation of several genes that affect neuronal function, plasticity, and survival.4,6Although there is extensive literature on synaptic transmission and volatile anesthetic effects,7–13there is relatively little literature on anesthetic effects on specific intracellular mechanisms that modulate Ca2+during neuronal activation.
Ca2+regulation in neuronal cells is complex.6,14,15Ca2+influx is key to modulation of intracellular processes. Influx itself is known to occur through several mechanisms, including N-type and T-type Ca2+channels, nonspecific cation channels, receptor-operated specific and nonspecific channels, and finally Na+–Ca2+exchange (NCX). In this regard, the bidirectional NCX may play an important role both in Ca2+influx and removal of Ca2+during neuronal activation and in excitotoxicity.16,17Furthermore, NCX-mediated Ca2+influx and efflux can modulate synaptic transmission.18Given that Ca2+transport via NCX is much faster than via voltage-gated channels, NCX may be important in both short-term and long-term modulation of neuronal activation.
As with peripheral tissues, endoplasmic reticulum (ER) Ca2+release (and perhaps mitochondrial Ca2+as well) contributes to total Ca2+, at least at the global level. ER Ca2+release is known to occur through both inositol trisphosphate (IP3)19and ryanodine receptor (RyR)20channels. There is now considerable evidence from different cell types, including neurons, that Ca2+influx also occurs through specific store-operated Ca2+channels (SOCCs; also termed capacitative Ca2+entry ) in response to ER Ca2+depletion, thus allowing for replenishment of intracellular Ca2+stores.21–25Such influx does not seem to be mediated via voltage-gated or receptor-operated channels.21In neuronal cells, SOCC-mediated Ca2+influx seems to be present in several different cell types and is involved in modulation of neuronal activity and in synaptic plasticity.22,26–29The existence of SOCCs in neurons points to a novel mechanism of neuronal modulation that is yet to be explored fully.
Volatile anesthetics have the potential to influence each and every one of the Ca2+regulatory mechanisms mentioned above. Previous studies have used electrophysiologic techniques to determine that anesthetics inhibit neuronal L-type,30N-type,31and T-type Ca2+channels.32Whether NCX is affected in neurons is not known. However, in cardiac muscle, we have previously demonstrated that volatile anesthetics inhibit NCX.33,34This Ca2+regulatory mechanism is particularly important in neurons where fast rates of action potential generation and changes in intracellular Ca2+require rapid transfer of Ca2+in and out of the cell. At the level of the synapse, local densities of influx channels and exchangers likely influence synaptic transmission. Anesthetic inhibition of SOCC-mediated Ca2+influx may be particularly important. Previous studies have demonstrated that anesthetics actually deplete ER Ca2+stores in different cell types.35,36This should normally trigger SOCC-mediated Ca2+influx. However, inhibition of this mechanism by anesthetics would maintain the ER in a state of depletion, preventing the normal Ca2+response of the neuron to electrical or agonist stimulation. Given the relative novelty of the SOCC mechanism, there are currently no data on anesthetic effects on this mechanism in neurons. However, we have recently demonstrated that clinically relevant concentrations of halothane, isoflurane, and sevoflurane inhibit SOCC-mediated Ca2+influx in airway smooth muscle.37
In the current study, we examined the effects of clinically relevant concentrations of halothane, isoflurane, and sevoflurane on influx-mode NCX and SOCC-mediated Ca2+influx in differentiated rat pheochromocytoma (PC12) cells, which served as an immortalized model of neuronal cells. We hypothesized that volatile anesthetic inhibition of neuronal activity is mediated by inhibition of Ca2+influx via NCX and the novel mechanism of SOCCs.
Materials and Methods
PC12 Cell Culture
PC12 cells (American Type Culture Collection, Manassas, VA) were cultured in incubation flasks with RPM-1640 medium, 5% heat-inactivated fetal bovine serum, 10% heat-inactivated horse serum, 100 μg/ml streptomycin, and 100 U/ml penicillin at 37°C in a humidified atmosphere of 5% CO2–95% air. At 60–75% confluence, cells were resuspended and seeded on eight-well coverslip-bottom chambers for plating (2- to 3-day incubation). Nerve growth factor (per supplier recommendations) was then used to induce cell differentiation. Serum starvation for growth arrest was performed only for 24 h before experiments. Continued cellular response to agonists and maintained [Ca2+]iconcentrations were used as surrogate indicators for lack of apoptotic cell death. Finally, using light microscopy, differentiated PC12 cells were identified morphologically by neurite outgrowth.
Differentiated PC12 cells were incubated in 2.5 μm fura-2/AM (Molecular Probes, Eugene OR) and visualized using a MetaFluor real-time fluorescence imaging system (Universal Imaging, Downingtown, PA) on a Nikon Diaphot inverted microscope (Fryer Instruments, Edina, MN). Cells were initially perfused with normal Tyrode’s solution containing 145 mm NaCl, 4 mm KCl, 1 mm MgCl2, 1 mm CaCl2, 10 mm glucose, and 10 mm HEPES (pH 7.4; 25°C). A custom-built fluid-level controller was used to perfuse the cells, which minimized changes in fluid level but allowed for rapid exchange of perfusate (< 300 ms). Images (640 × 480 pixels) were acquired at 30 frames/s at 400× magnification (40×/1.25 oil-immersion lens). The [Ca2+]iresponses of 25 cells/chamber were obtained using individual, software-defined regions of interest. Fura-2 levels were calibrated for [Ca2+]iusing the technique of Grynkiewicz.38
As in previous studies,37a calibrated on-line vaporizer was used to add halothane (Wyeth-Ayerst Laboratories, Philadelphia, PA), isoflurane (Abbott Laboratories, Deerfield, IL), and sevoflurane (Abbott) to the aerating gas mixture. Aqueous anesthetic concentrations equivalent to 0.5, 1, and 1.5 minimum alveolar concentration (MAC) at room temperature (25°C) were measured for halothane and isoflurane by gas chromatography and an electron capture detector (Hewlett-Packard 5880A, Palo Alto, CA) and for sevoflurane using a flame ionization detector. Aqueous halothane concentrations equivalent to 1 and 2 MAC were 0.32 ± 0.08 and 0.50 ± 0.09 mm halothane, 0.32 ± 0.08 and 0.50 ± 0.09 mm isoflurane, and 0.48 ± 0.06 and 0.70 ± 0.08 mm sevoflurane, respectively.
Drugs and Chemicals
All drugs were obtained from Calbiochem/EMD Biosciences (San Diego, CA) unless stated otherwise. Cell culture media were obtained from Hyclone (Logan, UT).
Effects of Volatile Anesthetics on Influx-mode NCX
The technique for evaluating the effect of volatile anesthetics on influx mode of NCX has been previously published in cardiac muscle33and is illustrated in figure 1. PC12 cells were initially perfused with normal Tyrode’s, and resting [Ca2+]iconcentrations were recorded to ensure cell stability. Stable cells were identified, and regions of interest for [Ca2+]imeasurement were outlined. Cells were then Na+-loaded by perfusion with Tyrode’s solution containing 0 Ca2+(Ca2+-free Tyrode’s with 5 mm EGTA) and normal Na+, 5 μm cyclopiazonic acid (CPA; inhibitor of the sarcoendoplasmic reticulum Ca2+adenosine triphosphatase), 10 μm Xestospongin D (IP3receptor channel blocker), and 10 μm ryanodine (RyR channel blocker). Therefore, the ER was effectively inhibited, and the plasma membrane was functionally isolated. After Na+-loading cells for 2 min, perfusion was rapidly switched (< 300 ms) to Tyrode’s solution containing 0 Na+, normal Ca2+, CPA, Xestospongin D, and ryanodine, selectively activating the influx mode of NCX.33The rate of increase of [Ca2+]iwas measured and served as an index of NCX activity in the influx mode. After 1 min of measurement, the cells were washed for 10 min with normal Tyrode’s solution. Ryanodine is not easily washed out by this technique. However, its presence was not expected to adversely affect subsequent manipulations.
After the wash, the above protocol was repeated with preexposure to 0.5, 1, or 2 MAC volatile anesthetic (halothane, isoflurane, or sevoflurane) before rapid activation of NCX. In control experiments, the protocol was repeated without volatile anesthetics. The observed Ca2+influx was verified to be NCX mediated by repeating the above protocol in a separate set of cells using the NCX inhibitor KB-R7943 (KBR; 10 μm). In other cells, the effect of specific voltage-gated Ca2+channel inhibitors (the L-type Ca2+channel antagonist nifedipine, 1 μm; P/Q type Ca2+channel inhibitor ω-agatoxin IVA, 1 nm; N-type Ca2+channel inhibitor ω-conotoxin MVIIA, 1 nm; and T-type Ca2+channel inhibitor mibefradil, 5 μm) and receptor-operated channel inhibitor (SKF-96365, 10 μm) on the observed Ca2+influx was examined. To eliminate the contribution of voltage-gated channels per se , an additional set of experiments was performed in the simultaneous presence of nifedipine, conotoxin, agatoxin, and mibefradil.
The technique for examining SOCC-mediated Ca2+influx has also been recently described.37,39For these experiments, a Hanks balanced salt solution with or without 2.5 mm Ca2+was used. After baseline Ca2+concentrations were measured, extracellular Ca2+was removed by exposure to 0 Ca2+Hanks balanced salt solution (5 mm EGTA). In the continued absence of extracellular Ca2+, cells were also exposed to 1 μm nifedipine and 10 mm KCl to ensure that L-type Ca2+channels were not activated. Cells were then rapidly exposed to 1 μm CPA in 0 Ca2+Hanks balanced salt solution. This technique passively depletes ER Ca2+by preventing reuptake during ongoing Ca2+leak from the ER. In addition, Ca2+influx is also prevented because no extracellular Ca2+is present. When the [Ca2+]iresponse to CPA was verified, 2.5 mm extracellular Ca2+was rapidly reintroduced (in the continued presence of CPA). The observed [Ca2+]iresponse was considered to represent SOCC-mediated Ca2+influx.
The observed Ca2+influx was characterized using various pharmacologic inhibitors of other Ca2+influx mechanisms. The entire protocol was performed in the presence of one of the following agents: (1) 1 nm ω-conotoxin MVIIA to inhibit N-type Ca2+channels; (2) 1 nm ω-agatoxin IVA to inhibit P/Q-type Ca2+channels; (3) 5 μm mibefradil (Sigma, St. Louis, MO) to inhibit T-type Ca2+channels; (4) 10 μm KBR to inhibit NCX; (5) 10 μm SKF-96365; or (6) 1 μm or 1 mm Ni2+or La3+. To eliminate the contribution of voltage-gated channels per se , an additional set of experiments was performed in the simultaneous presence of nifedipine, conotoxin, agatoxin, and mibefradil.
Because Ca2+influx may occur via several mechanisms that are not dependent on store depletion, a set of control experiments was performed where, after removal of extracellular Ca2+, ER stores were not depleted (no CPA exposure). Extracellular Ca2+was then reintroduced as for the SOCC experiments.
Effect of Volatile Anesthetics on Store-operated Ca2+Influx
Store-operated Ca2+channel–mediated Ca2+influx was first established by performing a control CPA protocol as above. Cells were then washed for 15–20 min with Hanks balanced salt solution to replenish ER Ca2+stores. Extracellular Ca2+was then removed, and the cells were reexposed to CPA. When a [Ca2+]iresponse to CPA was observed, the cells were exposed for 1 min to 0.5, 1, or 1.5 MAC halothane, isoflurane, or sevoflurane in the continued presence of CPA. This technique was used to ensure that ER Ca2+release was itself not influenced by volatile anesthetics, but the anesthetics were present in sufficient concentration before SOCC activation. In the continued presence of CPA and anesthetic, extracellular Ca2+was rapidly reintroduced. In control experiments, the SOCC protocol was performed twice without anesthetic.
At least 15 cells were analyzed for each protocol. Data were compared using a paired t test because the protocols involved a control followed by anesthetic/drug effects in the same cell. Repeated-measures analysis of variance with drug/anesthetic and anesthetic concentration as variables was used for multiple comparisons, with Bonferroni corrections, and Tukey post hoc analyses. A P value less than 0.05 was considered significant (two tailed). All data are expressed as mean ± SE.
Baseline [Ca2+]iconcentrations in differentiated PC12 cells ranged between 75 and 110 nm (n = 376). Na+-loading of cells with 0 Ca2+, 140 mm Na+Tyrode’s solution resulted in a moderate decrease in [Ca2+]i(fig. 1A), presumably due to active efflux-mode NCX and Ca2+extrusion across the plasma membrane. After this initial decrease, [Ca2+]iconcentrations stabilized. Subsequent, rapid reintroduction of extracellular Ca2+and removal of Na+resulted in a rapid rate of increase in [Ca2+]i. The Ca2+influx rate (calculated from the steepest portion of the ascending curve) ranged from 3.5 to 14.6 nm/s during the first run of the protocol. Repetition of the protocol in control experiments resulted in a 5 ± 3% decrease in influx rate. Accordingly, time-related bias in the protocol was ignored.
The observed influx was significantly inhibited by the NCX inhibitor KBR (P < 0.05; fig. 1B; n = 22). In contrast to approximately 90% inhibition by KBR, inhibitors of voltage-gated Ca2+channels had minimal effects, with inhibition ranging from 5 to 9% (fig. 1B; n = 18 for each inhibitor). In the experiment where several inhibitors of voltage-gated channels were present, the total inhibition of influx was 16% (n = 30). These data verified that the observed Ca2+influx occurred via NCX.
Effects of Volatile Anesthetics on Influx-mode NCX
Compared with NCX-mediated Ca2+influx in the initial control part of the protocol, exposure to 0.5, 1, or 1.5 MAC halothane or isoflurane significantly slowed the rate of increase in [Ca2+]iconcentrations (fig. 2; P < 0.05 compared with control for each halothane or isoflurane concentration; n = 26 for each anesthetic and concentration). For halothane, there was concentration-dependent decrease in NCX-mediated Ca2+influx. At each concentration, halothane was more potent than isoflurane in inhibiting Ca2+influx. In comparison to these anesthetics, only sevoflurane at greater than 1 MAC inhibited NCX to a significant extent (fig. 2).
In the absence of extracellular Ca2+, 1 μm CPA resulted in a slow and occasionally transient increase of [Ca2+]i, which reached a peak value between 125 and 200 nm (n = 467). Subsequent rapid reintroduction of extracellular Ca2+resulted in a sustained increase of [Ca2+]i, which was approximately 100–150% of the peak CPA response (fig. 3).
In a previous study, we reported that SOCC-mediated Ca2+influx is not mediated via voltage-gated Ca2+channels and is inhibited by Ni2+and La3+.39After control activation of SOCCs with CPA and a wash, cells were exposed to nifedipine, ω-conotoxin, agatoxin, mibefradil, 1 μm or 1 mm NiCl2or LaCl3before reactivation of SOCC-mediated Ca2+influx (n = 19 for each drug). None of the inhibitors of voltage-gated Ca2+channels resulted in greater than 8% inhibition of the observed influx. In the experiment where several inhibitors of voltage-gated channels were present, the total inhibition of influx was 11% (n = 22). Both 1 μm Ni2+and La3+significantly inhibited influx (fig. 3; P < 0.05 compared with control), indicating that SOCC-mediated Ca2+influx in PC12 cells is sensitive to both ions. In other cells, neither SKF-96365 (n = 14) nor KB-R7943 (n = 18) significantly inhibited influx (fig. 3).
In control experiments where ER stores were not depleted, reintroduction of extracellular Ca2+resulted in a significantly slower rate of Ca2+influx compared to that after ER depletion (14.4 ± 4.5% rate of Ca2+increase compared with SOCCs). This influx was significantly inhibited by combined preexposure to conotoxin, agatoxin, nifedipine, and mibefradil (91.5 ± 8.5% reduction compared with control; P < 0.05; n = 24).
Effect of Volatile Anesthetics on SOCC-mediated Ca2+Influx
Repetition of the SOCC protocol resulted in a 5–8% decrease in the observed Ca2+influx (rundown control). In contrast to the significant effects of halothane, isoflurane, and sevoflurane on NCX-mediated Ca2+influx, of the three anesthetics, only halothane had any significant inhibitory effect on SOCC-mediated Ca2+influx (fig. 4; P < 0.05 compared with control for each anesthetic at each MAC; n = 29 for each anesthetic and concentration).
In the current study, we examined the effects of clinically relevant concentrations of halothane, isoflurane, and sevoflurane on influx-mode NCX and SOCC-mediated Ca2+influx in differentiated PC12 cells. We found that halothane, isoflurane, and, to a lesser extent, sevoflurane inhibit NCX-mediated Ca2+influx, thus decreasing [Ca2+]i. We also demonstrated the existence of a robust Ca2+influx mechanism that is triggered by depletion of ER Ca2+stores. Such influx is not mediated by voltage-gated channels. However, in contrast to these inhibitory effects on NCX, volatile anesthetics seem to have only minimal effects on SOCC-mediated Ca2+influx in these cells. These data suggest that anesthetic inhibition of neuronal activity may be mediated, at least in part, by inhibition of NCX, but not SOCCs, indicating differential anesthetic sensitivity of Ca2+regulatory mechanisms in neuronal cells.
The current study used differentiated rat pheochromocytoma (PC12) cells as an immortalized model of neuronal cells. These cells have been used extensively for several years, are known to maintain a differentiated neuroendocrine phenotype, and serve as a convenient model system for cell biologic studies on the action of drugs or other interventions on neuronal morphology and function. Although these cells are actually part of the sympathetic nervous system, it is important to consider that they may not necessarily represent “higher-order” neurons (as in the brain and spinal cord), which are in the postdifferentiated state. However, PC12 cells express several of the neuronal receptors, channels, and signal transduction pathways found in higher-order neurons40and also release neurotransmitters as do other neurons. Furthermore, PC12 cells respond to neurotrophic factors that modulate higher-order neuronal growth and function, an example being nerve growth factor used for differentiation. Accordingly, the findings of the current study on Ca2+regulation using PC12 cells can likely be translated at least to volatile anesthetic effects on Ca2+regulation in the sympathetic nervous system and perhaps to cortical and other neuronal cell types that are involved in the clinical manifestations of anesthetics. Whether anesthetic effects on neurotransmission and other neuronal processes in higher-order neurons can also be modeled by PC12 cells remains to be determined.
In the NCX protocol, we observed a small but consistent decrease in [Ca2+]iduring Na+-loading of PC12 cells (fig. 1A). Furthermore, in the SOCC protocol, removal of extracellular Ca2+also resulted in a small decrease in [Ca2+]i. In previous studies using similar protocols to study NCX (cardiac muscle)33,41and SOCCs (airway smooth muscle),37,39such a decrease was less obvious and was only occasionally observed. Both of these decreases are most likely due to a robust NCX in PC12 cells. In the NCX protocol, extracellular Na+in the absence of Ca2+allows for efflux mode NCX and thus Ca2+extrusion and decreased [Ca2+]i, whereas in the SOCC protocol, zero extracellular Ca2+again allows for efflux via NCX and other mechanisms. Regardless of the mechanisms involved, the decrease in [Ca2+]iis unlikely to have affected the overall results, because in the NCX protocol, the rate of increase of [Ca2+]i(and not absolute concentrations) was measured, whereas in the SOCC protocol, subsequent exposure to CPA increased [Ca2+]i.
Role of NCX in Neuronal Cells
Ca2+influx in neuronal cells is known to occur through several mechanisms, including voltage-sensitive Ca2+channels, nonspecific cation channels, receptor-operated specific and nonspecific channels, and NCX.6,14,15NCX is expressed in high concentrations in neurons,16–18,42–44especially in regions such as synapses, where large amounts of Ca2+cross the plasma membrane.45In this regard, efflux-mode NCX may contribute to the maintenance of neuronal Ca2+homeostasis, especially during depolarization and neurotransmitter release. The role of influx-mode NCX is less clear in neurons that are activated at rapid rates, because depolarization-evoked Ca2+transients are extremely brief. However, NCX may regulate resting [Ca2+]iand Ca2+store content, thus indirectly modulating transmitter release.44,46Repeated depolarizations may increase intracellular Na+, triggering influx-mode NCX, and thus modulating [Ca2+]iconcentrations.
Previous studies had suggested the existence of NCX in PC12 cells based on radioactive Na and Ca fluxes.47Although adrenal chromaffin cells are known to express NCX,48there is relatively little data on NCX in PC12 cells. Our data confirm a very robust NCX activity in these cells. Given the fact that NCX exists in other neuronal cell types, our data are significant because they relate to volatile anesthetic effects on this Ca2+regulatory mechanism.
Volatile Anesthetic Effects on NCX
Given the concurrent effects of anesthetics on multiple [Ca2+]iregulatory mechanisms, particularly Ca2+influx channels and ER, as well as the lack of specific inhibitors of the influx and efflux modes of NCX, detailed studies on anesthetic and NCX interactions have been lacking. In fact, most of the published literature, including from our group, exists in cardiac muscle. Haworth and Goknur49found that halothane, isoflurane, and enflurane all completely inhibit cardiac NCX. In recent studies on adult versus neonatal cardiac muscle, we also found that both halothane and sevoflurane inhibit NCX-mediated Ca2+influx as well as efflux.33,41Furthermore, we were the first to report that both halothane and sevoflurane blunt the relation between NCX-mediated Ca2+influx and intracellular Na+, as well as between NCX-mediated Ca2+efflux and [Ca2+]iand extracellular Na+.33
There are currently no published studies on volatile anesthetic interactions with NCX in neuronal cells. Therefore, the current study is the first to demonstrate an inhibitory effect of clinically relevant concentrations of volatile anesthetics on NCX in neuronal cells. Such an inhibitory effect is consistent with previous findings in cardiac muscle, where NCX-mediated Ca2+influx may play a part in overall modulation of [Ca2+]iconcentrations,41as in neuronal cells. In fact, we found that volatile anesthetics have a proportionately greater inhibitory effect on influx-mode NCX compared with efflux-mode NCX.33The results of the current study are interesting in that both halothane and isoflurane seem to potently inhibit NCX, whereas sevoflurane at less than 1 MAC does not have a significant effect. These data suggest that volatile anesthetic agents may differ in the mechanisms by which they inhibit neuronal activity in producing anesthesia.
Role of SOCC-mediated Influx in Neuronal Cells
In nonneuronal cells, there is now considerable evidence, including our own, for Ca2+influx via SOCCs in response to ER Ca2+depletion.21,37,39SOCC-mediated Ca2+influx (also termed capacitative Ca2+influx ) is thought to serve the purpose of ER Ca2+replenishment. However, the transfer of Ca2+from the extracellular environs to the ER actually involves an increase in cytosolic Ca2+, i.e. , [Ca2+]i. Such Ca2+influx is thought to occur via specific channels rather than receptor-operated or voltage-operated Ca2+channels, determined from electrophysiologic data and the sensitivity or insensitivity of these channels to various pharmacologic blockers.21ER-mediated Ca2+signaling is obviously important in neuronal cells. However, there are little data on SOCC-mediated Ca2+influx in this tissue. In the central nervous system, SOCC-mediated Ca2+influx has been demonstrated in astrocytes26and neuroblastoma cell lines.27There is now considerable evidence for SOCC-mediated Ca2+influx in PC12 cells.22,28,29The results of the current study are consistent with previous findings of a robust SOCC-mediated Ca2+entry upon ER depletion.
Initial studies on SOCC-mediated Ca2+influx suggested that such influx occurs largely in response to depletion of IP3-sensitive Ca2+stores.21However, in recent studies using airway smooth muscle,39we demonstrated that SOCC activation is dependent not only on IP3-sensitive Ca2+stores, but also on ryanodine-sensitive stores (i.e. , Ca2+release via RyR channels). Similar dependence of SOCCs on RyR channels has also been reported in mouse anococcygeus muscle.50In the current study, we did not specifically examine IP3- versus RyR-dependent SOCCs. However, SOCC-mediated Ca2+influx due to release from both stores has been previously demonstrated in PC12 cells.28This raises the issue of depletion of both stores using pharmacologic approaches. Several studies have used either CPA or the irreversible sarcoendoplasmic reticulum Ca2+adenosine triphosphatase inhibitor thapsigargin. In the current study, the use of the reversible drug CPA allowed for comparison of drug effects in the same cell. Unlike several Ca2+channels usually expressed by neuronal cells, SOCCs are not activated by changes in membrane potential.21In accordance, in the current study, the robust Ca2+influx after ER Ca2+depletion by CPA was found to be largely insensitive to nifedipine, ω-conotoxin, agatoxin, or mibefradil, confirming that L-type, N-type, P/Q-type, or T-type Ca2+channels do not mediate the observed influx. Consistent with other studies in nonneuronal cells, we also found that SOCCs are blocked by Ni2+and La3+.39,50,51Although both ions can nonspecifically inhibit Ca2+influx, inhibition of influx after ER depletion by micromolar concentrations of either ion strongly indicates the presence of SOCCs.
It is interesting that, given the myriad of Ca2+channels expressed in neuronal cells, SOCC-mediated influx is still so robust. However, a consistent finding (including in the current study) has been the relatively small increase in [Ca2+]iwith inhibitors of the sarcoendoplasmic reticulum Ca2+adenosine triphosphatase (e.g. , thapsigargin, CPA), suggesting only a small ER store component.52SOCC-mediated influx may then functionally compensate for the small ER response with agonist or electrical stimulation. Recent evidence of SOCC-mediated Ca2+influx in hippocampal pyramidal and dentate granule cells shows that [Ca2+]iresponse from such influx is comparable to that induced by N -methyl-d-aspartate.53Furthermore, such influx seems to be activated by N -methyl-d-aspartate. These data suggest that SOCC-mediated Ca2+influx plays a role in synaptic plasticity of neuronal cells. Again, given the recent evidence for such influx in other neuronal cells, our data in PC12 cells are of significance, especially because they pertain to volatile anesthetic effects.
Volatile Anesthetics Effects on SOCC-mediated Ca2+Influx
As with NCX, there are currently few data on anesthetic effects on SOCC-mediated Ca2+influx. In endothelial cells, Tas et al. 54found that isoflurane inhibits histamine-induced Ca2+influx via SOCCs. However, in another study by the same group, enflurane was found to enhance Ca2+influx in rat glioma cells.55We recently reported that halothane, isoflurane, and, to a lesser extent, sevoflurane inhibit SOCCs in airway smooth muscle.37
In contrast to the limited previous data, the current study found that even high concentrations of volatile anesthetic do not significantly inhibit SOCC-mediated Ca2+influx in PC12 cells. Although somewhat surprising, given the potent inhibition of this mechanism in airway smooth muscle, these data raise several issues. SOCC-mediated Ca2+influx occurs via transient receptor potential (TRP) channels.24Cell types differ in the relative expression of TRP isoforms (TRPC subfamily). Furthermore, these isoforms also differ in the extent to which they mediate Ca2+influx, as well as their sensitivity to various agonists and modulators. Currently, it is not entirely clear whether SOCCs are mediated via different isoforms in different tissues. For example, in hippocampal neuronal cells, TRPC1 and TRPC3 isoforms mediate influx via SOCCs.56TRPC1 and TRPC3 messenger RNAs were endogenously expressed in PC12 cells.22Accordingly, it is possible that neuronal cells differ from other cell types in the relative roles of TRPC isoforms in mediating SOCCs. In this regard, these isoforms may further differ in their sensitivity to inhibition by volatile anesthetics. Further studies are required to examine this issue.
Clinical Significance of Volatile Anesthetic Effects
Increase of [Ca2+]iplays an important role in neurons, both in synaptic transmission and in neuronal function in general. In this regard, both NCX-mediated and SOCC-mediated Ca2+influx likely play important roles. NCX may regulate resting [Ca2+]iand ER Ca2+stores, thus indirectly modulating transmitter release. Furthermore, NCX may be triggered by repeated depolarizations that increase intracellular Na+. Accordingly, anesthetic inhibition of NCX has the potential to prevent ER Ca2+repletion and modulate synaptic transmission, thus contributing to anesthesia.
In addition to basic [Ca2+]iregulation in neurons, NCX may play a role in neuronal excitotoxicity, e.g. , due to N -methyl-d-aspartate, by promoting Ca2+influx in reverse mode.57Accordingly, volatile anesthetic inhibition of NCX may provide significant cellular protection against excitotoxic injury from increased Ca2+. Reverse-mode NCX also contributed to increased [Ca2+]iduring ischemia–reperfusion injury, and inhibition of NCX has been shown to be neuroprotective in vitro . NCX inhibition with KBR has been shown to prevent ischemia–reperfusion injury in the cerebral cortex.58Similar NCX inhibition by volatile anesthetics may also help to minimize neuronal reperfusion injury.
In contrast to NCX inhibition, the lack of anesthetic effect on SOCCs may highlight the point that volatile anesthetic inhibition of neuronal function may not represent a generalized, nonspecific inhibition of several regulatory mechanisms. The fact that even higher concentrations of potent anesthetics such as halothane did not inhibit SOCCs suggests that at least one Ca2+regulatory mechanism may remain intact in neurons exposed to anesthetics, allowing for compensation and recovery from anesthetic effects. How such differential effects of anesthetics on NCX versus SOCCs affects synaptic transmission remains to be determined.
Finally, an interesting finding has been the relative lack of effect of sevoflurane on NCX and SOCCs. Comparisons were made between anesthetic agents at the same MAC values. Accordingly, the lack of effect of sevoflurane may be related to lower potency of the agent in modulating NCX or SOCC protein function. Further investigations are necessary to determine whether the three anesthetics differ in their mechanism of interaction with Ca2+regulatory proteins, regardless of concentration.
The authors thank Gary C. Sieck, Ph.D. (Professor and Chair), for support and Jeffrey P. Bailey, B.S. (Research Technologist, Department of Physiology and Biomedical Engineering, Mayo Clinic College of Medicine, Rochester, Minnesota), for technical assistance.