Prior studies indicated that a Ca2+-dependent release of ATP can be initiated from the soma of sympathetic neurons dissociated from guinea pig stellate ganglia. Previous studies also indicated that Ca2+-induced Ca2+ release (CICR) can modulate membrane excitability in these same neurons. As Ca2+ release from internal stores is thought to support somatodendritic transmitter release in other neurons, the present study investigated whether CICR is essential for somatic ATP release from dissociated sympathetic neurons. Caffeine increased intracellular Ca2+ and activated two inward currents: a slow inward current (SIC) in 85% of cells, and multiple faster inward currents [asynchronous transient inward currents (ASTICs)] in 40% of cells voltage-clamped to negative potentials. Caffeine evoked both currents when cells were bathed in a Ca2+-deficient solution, indicating that both were initiated by Ca2+ release from ryanodine-sensitive stores in the endoplasmic reticulum. Sodium influx contributed to generation of both SICs and ASTICs, but only ASTICs were inhibited by the presence of the P2X receptor blocker PPADs. Thus ASTICs, but not SICs, resulted from an ATP activation of P2X receptors. Ionomycin induced ASTICs in a Ca2+-containing solution, but not when it was applied in a Ca2+-deficient solution, demonstrating the key requirement for external Ca2+ in initiating ASTICs by ionomycin. Pretreatment with drugs to deplete the internal stores of Ca2+ did not block the ability of ionomycin or long depolarizing voltage steps to initiate ASTICs. Although a caffeine-induced release of Ca2+ from internal stores can elicit both SICs and ASTICs in dissociated sympathetic neurons, CICR is not required for the somatic release of ATP.
- nonclassical transmitter release
- autonomic neurons
- asynchronous transient inward currents
- slow inward currents
classically, it was proposed that calcium (Ca2+)-dependent neurotransmitter release primarily occurs at highly organized active zones in nerve terminals (11, 19, 24). However, neurotransmitter release is also now known to occur from somatodendritic areas of neurons (9, 10, 16, 17, 26, 27, 30, 31). Although nonclassical transmitter release is Ca2+ dependent, it can occur from areas lacking highly organized release sites. Even though well-documented to occur, the mechanisms regulating somatodendritic transmitter release are not yet clearly established.
Previously, our laboratory demonstrated the regulated release of ATP quanta from guinea pig sympathetic neurons dissociated from the stellate ganglion (26). The released ATP activates P2X receptors on the same cell to generate asynchronous transient inward currents (ASTICs) in cells voltage-clamped to negative potentials. The ATP release is Ca2+ dependent and blocked by inclusion of botulinum toxin E in the patch pipette, suggesting the requirement of snare proteins in the release process. A quite similar Ca2+-dependent release of ATP has been observed in dissociated dorsal root ganglion cells (31).
Midbrain neurons release dopamine from somatodendritic sites, as well as from their nerve terminals (10). In guinea pig midbrain neurons, the somatodendritic release of dopamine is reported to be facilitated by mobilization of Ca2+ from intracellular stores (16). These results from central nervous system (CNS) neurons suggest that Ca2+-induced Ca2+ release (CICR) might be involved in somatic transmitter release. In contrast, Ford et al. (6) have recently reported that, in the case of mouse midbrain neurons, somatodentritic dopamine release is primarily dependent on Ca2+ entry, with little contribution from mobilization of Ca2+ from internal stores. Thus the contribution of Ca2+ release from internal stores to somatodentritic transmitter release may be species as well as cell dependent.
Previously, our laboratory showed that caffeine can release Ca2+ from ryanodine-sensitive intracellular Ca2+ stores in the dissociated guinea pig stellate neurons and also that a rise in intracellular Ca2+ concentration ([Ca2+]i) due to CICR activates a tetraethylammonium-sensitive conductance that suppresses action potential generation by depolarizing ramps (12). Consequently, the dissociated sympathetic stellate neurons make an excellent model system to test whether CICR is required for the initiation of somatic ATP release.
The present study was done to test whether a caffeine-induced rise in [Ca2+]i without Ca2+ influx can elicit ATP quantal release and also whether depletion of [Ca2+]i stores eliminates somatic ATP quantal release evoked by Ca2+ influx through voltage-dependent Ca2+ channels (VDCCs) or by transport through the plasma membrane by ionomycin.
MATERIALS AND METHODS
Ganglion dissection and cell culture.
The studies were performed on sympathetic neurons dissociated from the guinea pig stellate ganglia. Animals were killed by isoflurane overdose, followed by exsanguination using procedures approved by the University of Vermont Institutional Animal Care and Use Committee and in accordance with guidelines established by the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize the number of animals used and their suffering. The methods for cell dissociation were slightly modified from those described previously (12, 26). In brief, the right and left stellate ganglia were exposed rostral to the first rib, removed, and placed in Krebs solution containing (in mM) 120.9 NaCl, 5.9 KCl, 25 NaHCO3, 1.2 NaH2PO4, 2.5 CaCl2, 1.2 MgCl2, and 11 glucose (bubbled with 95% O2-5% CO2). The isolated ganglia were pinned to the Sylgard-184 (Dow Corning, Midland, MI) floor of a 35-mm Petri dish, and the connective tissue capsule was removed. The ganglia were minced and enzymatically dissociated in 2 ml of a low-Ca2+ HEPES-buffered salt solution containing 121 mM NaCl, 5.9 mM KCl, 26 mM Na-HEPES, 1.2 mM MgCl2, 0.1 mM CaCl2, 8 mM glucose, 3 mg/ml trypsin type XII-S (11,500 BAEE U/mg; Sigma-Aldrich, St. Louis, MO), and 15 mg/ml collagenase A (0.25 U/mg; Roche Molecular Biochemicals, Indianapolis, IN). The tissue containing enzyme solution was rotated on a nutator (BD, Franklin Lakes, NJ) in a 37°C incubator with an atmosphere of 5% CO2-95% air for 1 h. Immediately following the incubation period, the tissue was transferred to 1 ml of “wash” solution for 30 min to decrease enzyme activity. The wash solution contained Eagle's MEM (M7278; Sigma-Aldrich) plus 10% fetal calf serum, 0.1% bovine serum albumin, 0.1 mg/ml pyruvic acid, 1 mg/ml DNase, 200 U/ml penicillin, 0.2 mg/ml streptomycin, and 0.2 mg/ml gentamicin. Tissue was then transferred to 1 ml of the low-Ca2+ HEPES-buffered Krebs and gently triturated to break up the remaining tissue fragments. As the cells were dispersed by trituration, the remaining tissue pieces were transferred to another tube with low-Ca2+ Kreb's for further trituration. Tubes were centrifuged at 100 g for 5 min. The supernatant was discarded, and the remaining pellet was resuspended and kept overnight in medium containing MEM, 2.5 mM CaCl2, 8 mM glucose, 1 mg/ml DNase, 200 U/ml penicillin, 0.2 mg/ml streptomycin, and 0.2 mg/ml gentamicin. Dissociated neurons were seeded on glass coverslips (15 mm; Fisher Scientific, Hampton, NH) and stored in the incubator overnight until used for experiments.
Measurement of intracellular Ca2+ transients.
The cells were bathed in a buffered physiological solution, which contained (in mM) 121 NaCl, 5.9 KCl, 2.5 CaCl2, 1.2 MgCl2, 26 Na-HEPES, 8 glucose, pH 7.3. All experiments were done at 37 ± 1°C.
Changes in [Ca2+]i were determined from variations in the fluorescence intensity of fluo 3, as described previously (12). Cells were loaded with 5 μM fluo 3-AM with pluronic F-127 [both from Molecular Probes, Eugene, OR, and stored as stock solutions in dimethylsulfoxide (DMSO)]. Both the loading (15 min) and AM-ester cleavage steps (at least 15 min) were performed at room temperature. Images were acquired (sample rate 0.33–1 Hz) on a DeltaVision deconvolution microscopy system. Filter sets appropriate for FITC were used. For the imaging studies, tetrodotoxin (TTX; 300 nM, Tocris Bioscience, Ellisville, MO) was included in the bath solution to eliminate any spontaneous action potential firing, which might cause a rise in intracellular Ca2+ due to Ca2+ influx. Caffeine solutions were freshly dissolved in the physiological buffer solution each day and were applied by bath perfusion for ∼60 s.
Regions of interest corresponding to the cytoplasm of the neurons were selected from the raw image files, and average brightness over time plots were generated. Files were corrected for variations in lamp intensity (recorded by a diode simultaneously with image acquisition). The data sets were then corrected for dye bleaching using a single- or double-exponential decay algorithm and normalized to this decay curve to give the fluo 3 fluorescence ratio, as shown in Fig. 1. Because caffeine often induced multiple Ca2+ transients, all of the caffeine-induced peaks in the fluo 3 fluorescence ratio were integrated over time. Integrating the Ca2+ transients was considered more appropriate than simply measuring the peak amplitude of the increase in fluo 3 signal (Fig. 1). All corrections and integrations were performed with Microcal Origin 7.0 (Northampton, MA).
Recordings were made from neurons 24 h after dissociation. All experiments were conducted at 33–34°C, with the temperature maintained by an in-line heater (Warner Instruments, Hamden, CT). Whole cell currents were recorded under voltage-clamp conditions using the perforated patch configuration of the patch-clamp recording technique (8), as described in our laboratory's prior studies (14, 26). Cells were superfused with a HEPES-buffered salt solution (extracellular solution) composed of (in mM) 121 NaCl, 26 Na-HEPES, 5.9 KCl, 1.2 MgCl2, and 2.5 CaCl2 (pH 7.36). In some experiments, cells were bathed in a Ca2+-deficient solution in which the Ca2+ was replaced by magnesium or a Na+-deficient solution in which Na+ was replaced by N-methyl-d-glucamine (NMG). TTX (300 nM) was added to block voltage-dependent Na+ channels. Patch pipettes were backfilled with a CsAsp/CsCl recording solution containing 140 mM aspartic acid, 30 mM CsCl, 10 mM HEPES, 5 mM MgCl2, and 0.2 mg/ml amphotericin B (pH 7.15–7.20 with CsOH). Cesium was used in the pipette solution to block voltage-dependent K+ currents. Pipette resistances were 3–5 MΩ when pipettes were filled with the recording solution, and the electrode shanks were coated with dental wax to reduce electrode capacitance. Voltage commands were applied, and currents were recorded using the Axopatch 1-C amplifier coupled with the pCLAMP software (version 9.2) and the Digidata 1322A acquisition board (Axon Instruments, Union City, CA). The analog signal was filtered with a low-pass Bessel filter (2 kHz) and digitized at a sampling rate of 10 kHz for caffeine- and ionomycin-induced currents and 5 kHz for currents evoked by voltage steps. Recordings were stored on the hard drive of a personal computer.
Liquid-junction potentials existed between intracellular and extracellular solutions. The reported membrane potentials have not been corrected for these potentials.
Reverse transcription-polymerase chain reaction.
Methods for determining transient receptor potential canonical (TRPC) transcript levels were similar to those described in prior studies (29). Stellate ganglia were dissected under RNase-free conditions, and total RNA was extracted using Tri reagent (Sigma). The total RNA quantity for each ganglion was determined with a Nanodrop ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE). One microgram of RNA per sample was used to synthesize complementary DNA with the M-MLV Reverse Transcriptase (Promega, Madison WI) and a mix of oligo(dT) and random hexamer primers. Amplified stellate DNA product from specific primers was ligated into pCR2.1 TOPO using TOPO TA cloning kit (Invitrogen, Carlsbad, CA) to generate plasmid standards. The nucleotide sequences of the inserts were verified by automated fluorescent dideoxy dye terminator sequencing (Vermont Cancer Center DNA Analysis Facility). Amplification of the guinea pig cDNA templates was performed using HotStart IT SYBR Green qPCR Master Mix (US Biochemical) with the following primers: TrpC-1 sense: 5′-TCTATGATAAAGGGTATACACCA-3′; TrpC-1 antisense: 5′-AAGCACGATGACAACCACAACAT-3′; TRPC-3 sense: 5-AACTCTGCTTTTACCACTGTAGAAGAAA-3′; TRPC-3 antisense: 5′-GATCATATTTAAGCACAACTGAAGTCACT; TRPC-4 sense: 5′-GGCATACGATGTGAAAAGCAGA-3′; TRPC-4 antisense: 5′-AGCCCAAATATTGACCAAAACAGT-3′; TrpC-5 sense: 5′-TTTAGACTCCTCAGATGATGT-3′; TrpC5 antisense: 5′-TTGTTCTTCCTGTCCATCA-3′; TrpC6 sense: 5-CACGTCCGCATCATCCTCAATTTC-3; TRPC6 antisense: 5-CGAGGTGCAAAGTACAACCCA-3.
All drugs were obtained from commercial sources: thapsigargin (1 μm), cyclopiazonic acid (CPA; 10 μM) and phorbol-12-myristate-13-acetate (PMA, 1 μM) from Calbiochem (La Jolla, CA); pluronic F-127 and fluo 3-AM from Molecular Probes (Eugene, OR); TTX (300 nM) from Tocris; ionomycin, pyridoxal phosphate-6-azophenyl-2, 4-disulfonic acid (PPADS) (10 μM) and caffeine (10 mM) from Sigma-Aldrich (St Louis, MO). Thapsigargin, CPA, PMA, fluo 3-AM, and pluronic F-127 were diluted each day from frozen aliquots of concentrated DMSO stock solutions. For a vehicle control, DMSO was added at the final concentration to the control solution. All drugs were added to the extracellular solution and applied by gravity flow perfusion.
Data are reported as means ± SE. Statistical significance of responses to different concentrations of caffeine was determined by ANOVA. The χ2 analysis was used to determine if the frequency of cells with ASTICs was statistically different between control and test conditions. Differences were considered significant at P < 0.05.
Caffeine elicits a concentration-dependent elevation of intracellular Ca2+.
Caffeine can elicit release of Ca2+ from ryanodine-sensitive internal stores in guinea pig sympathetic neurons dissociated from the stellate ganglia (12). We used caffeine in this study as a tool to test whether a rise in [Ca2+]i, which was due to the release of Ca2+ from ryanodine/caffeine-sensitive internal stores, could elicit ASTICs. To determine the appropriate caffeine concentration to use, experiments were done to establish the concentration dependence of the caffeine-induced change in [Ca2+]i in the dissociated guinea pig stellate neurons.
Measurements of the fluorescence ratio (F/Fo) of the Ca2+ sensitive dye fluo 3 were used to determine the concentration dependence for the caffeine-induced change in [Ca2+]i in the dissociated stellate neurons (Fig. 1A). Different cells from multiple dissociations were challenged with 5, 10, or 20 mM caffeine (a single concentration for each cell), and the fluo 3 F/Fo was determined. The averaged caffeine-induced increase in F/Fo was concentration dependent, with the caffeine-induced increase in F/Fo smaller at 5 mM than with either 10 or 20 mM (Fig. 1B). As reported previously, the pattern of the caffeine response differed between cells. In some cells, a single caffeine-induced Ca2+ transient rose quickly and declined, whereas, in other cells, multiple Ca2+ transients were produced (Fig. 1A). Because multiple peaks were noted in many cells, we integrated F/Fo over time to quantify the rise in Ca2+ (Fig. 1A). Given that the averaged results suggested 10 mM caffeine elicited a maximum rise in [Ca2+]i, this concentration was chosen for subsequent experiments.
Caffeine elicited both a slow inward current and ASTICs.
Application of 10 mM caffeine for 30–60 s could elicit two different types of inward current in cells held at −60 mV. In ∼85% of the neurons, a caffeine-induced slow inward current (SIC) was produced. This current generally peaked within 1–6 s and lasted <15 s (Fig. 2A). In 33% of cells that responded to caffeine, a second SIC was recorded either just before or shortly after the caffeine application was terminated (data not shown). In ∼40% of the cells, caffeine also elicited much faster and briefer inward currents, which were superimposed on the SIC (Fig. 2A). These transient inward currents were similar in time course to the ATP-generated ASTICs described previously (26). Caffeine could elicit both SICs and ASTICs in cells that were bathed in a Ca2+-deficient solution in which Mg2+ was substituted for Ca2+ (Fig. 2B). Thus both of these currents were initiated by a caffeine-induced rise in [Ca2+]i caused by the release of Ca2+ from internal stores and did not require Ca2+ influx.
Na+ influx contributes to the generation of both SICs and ASTICs.
Previously, Tompkins and Parsons (26) demonstrated that Na+ influx through an ATP-gated cationic channel was required for generation of ASTICs in cells voltage clamped to negative potentials. We tested whether the caffeine-activated SICs also were dependent on Na+ influx. To test the potential involvement of Na+ influx, caffeine was applied to cells maintained in a Na+-deficient solution in which the Na+ was replaced in the bath solution by NMG. No SICs or ASTICs were recorded during a 30-s exposure to caffeine in seven cells voltage clamped to −60 mV and bathed in the Na+-deficient solution.
PPADS blocks the caffeine-induced ASTICs, but not the SICs.
Tompkins and Parsons (26) reported that exposure to 10 μM PPADS essentially eliminated ASTICs, indicating that these transient inward currents were mediated by the activation of P2X receptors. In the present study, we tested whether PPADS would also block the caffeine-induced SICs, as well as the caffeine-induced ASTICs.
Cells were perfused with bath solution containing 10 μM PPADS for ∼3 min, and then the solution was switched to one containing PPADs plus 10 mM caffeine for 60 s. Caffeine stimulated SICs in 7 of 10 cells tested. However, caffeine did not elicit ASTICs in any of the cells pretreated with PPADS. The lack of ASTICs in PPAD-pretreated cells was significantly different than that found in control cells (ASTICS in 5 of 12 control cells, no ASTICS in 10 PPAD-treated cells, χ2 = 5.39, P < 0.05). These results indicate that the caffeine-induced SICs appear to be generated by a mechanism independent of P2X receptor activation and confirm that ASTICs activated during caffeine application are also mediated by P2X receptor activation.
PMA does not potentiate SIC or ASTIC initiation by caffeine.
Phorbol esters, such as PMA, are known to enhance transmitter release in a variety of preparations (13). In our laboratory's initial report, we demonstrated that, during PMA exposure, the frequency of ASTICs evoked by ionomycin increased significantly (26). As part of the present study, we tested whether PMA would enhance ASTIC generation evoked by 10 mM caffeine. Cells were pretreated with 1 μM PMA for 2 min and then perfused with a solution containing 10 mM caffeine and 1 μM PMA for 60 s. In companion control experiments, caffeine was applied to cells exposed to DMSO, the vehicle for PMA.
PMA treatment did not affect SIC or ASTIC generation by caffeine, as determined by χ2 analysis. In 10 cells treated only with DMSO, caffeine elicited a SIC in all cells tested, and ASTICs accompanied the SIC in 4 of the 10 cells. Caffeine evoked a SIC in seven of eight cells pretreated with PMA, and in four of these seven, ASTICs were evident.
The activation of ASTICs by ionomycin requires external Ca2+, but not release of Ca2+ from internal stores.
In our laboratory's previous study, we demonstrated that ASTIC frequency was dependent on extracellular Ca2+ (26). During experiments in which ASTICs were activated by ionomycin application, ASTIC frequency was enhanced when the external Ca2+ concentration was elevated. When external Ca2+ was replaced by Mg2+, ASTIC generation ceased. It has been suggested that ionomycin raises intracellular Ca2+ by first depleting Ca2 from intracellular stores, which, in turn, stimulates store-operated Ca2+ influx across the plasma membrane (15). Two different experimental protocols were used in the present study to determine whether the ionomycin-induced initiation of ASTICs required an initial phase of Ca2+ release from internal stores followed by Ca2+ influx. In the first series of experiments, we bathed the cells for 5 min in a Ca2+-deficient solution that contained 3 μM ionomycin, and in which the Mg2+ had been substituted for Ca2+. During a 5- to 6-min exposure to ionomycin in the Ca2+-deficient solution, there was no increase in ASTIC activity above that seen before exposure to the ionophore (Fig. 3A). When the bath solution was changed to the control solution containing 2.5 mM Ca2+ and 3 μM ionomycin, ASTICs were initiated within 30–60 s. When the ionomycin-containing solution bathing the cell was changed back to the Ca2+-deficient solution, ASTIC activity quickly ceased again (Fig. 3A). Thus the presence of Ca2+ in the bathing solution was required for the stimulation of ASTICs by ionomycin. Identical results were obtained in three cells.
We also noted that, when external Ca2+ was present, a −5- to −30-pA change in the holding current commonly developed during the ionomycin treatment. This shift in holding current was not observed when ionomycin was applied in the Ca2+-deficient solution. Although not studied further, we suggest that the Ca2+-dependent inward shift in holding current that occurred during ionomycin treatment might result from activation of a current similar to the SIC, which is activated by the caffeine-induced transient rise in Ca2+.
A second series of experiments tested further whether Ca2+ released from internal stores was required for the ionomycin-induced initiation of ASTICs. Previously, our laboratory reported that exposure to 10 mM caffeine, along with 1 μM thapsigargin, quickly depleted internal ryanodine/caffeine-sensitive Ca2+ stores (12), which did not refill following treatment with thapsigargin, a potent inhibitor of the SERCA pump (25). In the present study, we used caffeine/thapsigargin pretreatment to deplete the intracellular Ca2+ stores, and then tested whether ionomycin treatment could stimulate ASTIC activity. First, cells were exposed to caffeine and thapsigargin for 2–5 min, followed by a solution containing 3 μM ionomycin and 1 μM thapsigargin. The results presented in Fig. 4B are representative of results from 8 of 10 cells tested and demonstrate that ionomycin initiated ASTIC activity in cells depleted of their caffeine/ryanodine-sensitive Ca2+ stores. In the other two cells, ASTICs were recorded at a low frequency throughout the recording, even before caffeine and thapsigargin or ionomycin was applied. No obvious increase in ASTIC frequency was detected after ionomycin application in these two cells.
The generation of ASTICs by depolarizing voltage steps does not require Ca2+ release from internal stores.
Our laboratory also demonstrated in our previous study that initiation of ASTICs occurs in response to depolarizing voltage steps that activate VDCCs to allow Ca2+ influx (26). Ca2+ influx through VDCCs also can initiate Ca2+ release from internal stores by a CICR mechanism in the dissociated guinea pig stellate neurons (12). Furthermore, the rise in Ca2+ concentration near the inner surface of the plasma membrane due to CICR activates a membrane conductance, which regulates excitability (12). Depletion of ryanodine/caffeine-sensitive stores disrupted the CICR-induced modulation of excitability (12). In the present study, we hypothesized that a rise in [Ca2+]i due to CICR might be required along with Ca2+ influx for the initiation of somatic ATP release during voltage steps. Thus we tested whether depletion of endoplasmic reticulum Ca2+ suppressed the generation of ASTICs initiated by depolarizing voltage steps (26).
The following protocol was used in the experiments. A series of depolarizing steps, either 250 or 500 ms in duration, were applied from a holding potential of −80 to 0 mV to activate VDCCs before and following pharmacological-induced depletion of the internal Ca2+ stores. Two different drug treatments were used to deplete the internal Ca2+ stores: pretreatment with 10 mM caffeine and 1 μM thapsigargin (2 cells), or pretreatment with 10 mM caffeine and 10 μM CPA (3 cells). With thapsigargin treatment, the thapsigargin remained in the bath for 10–15 min and then was washed out before the voltage steps were applied. The recordings were made after removing thapsigargin to avoid any possible inhibition of Ca2+ currents by thapsigargin. When CPA was used, it remained in the bath when the second series of depolarizing steps was made. With the depolarizing steps, the inward current generated during the step was followed by a large tail current, when the voltage was returned to the holding potential (19). Both before and following the drug treatment, ASTICs were superimposed on the decay phase of the tail current and after the current had ended (Fig. 4). In all five cells, the apparent frequency of ASTICs was similar before and after depletion of the internal Ca2+ stores.
Guinea pig stellate ganglia express multiple TRPC channel transcripts.
Members of the canonical TRPC channel family are thought to form nonselective cation channels, which can be activated by multiple mechanisms, including a rise in intracellular Ca2+ (2, 18). Consequently, we established using RT-PCR what transcripts for TRPC channels are present in extracts of the guinea pig stellate ganglia. As shown in Fig. 5, transcripts for multiple TRPC channels are expressed by the guinea pig stellate neurons. Thus the combination of TRPC channels is a possible candidate for the nonselective cation channel activated to give rise to the SIC.
The key result of the present study is that CICR is not required for the generation of ASTICs, even though a caffeine-induced release of Ca2+ from internal stores can elicit both SICs and ASTICs in dissociated sympathetic neurons voltage-clamped at negative potentials. Rather, we suggest that a rise in [Ca2+]i due to Ca2+ influx, either through activated VDCCs or via transport across the plasma membrane by ionomycin, is the primary mechanism for somatic release of ATP in the dissociated stellate neurons under the different conditions of the present study. Similar to the ASTICs, the tail currents that followed long depolarizing steps remained after depletion of internal Ca2+ stores, indicating they too were not dependent on CICR.
It has been suggested that ionomycin raises [Ca2+]i by releasing Ca2+ from internal stores, which, in turn, activates store-operated Ca2+ entry (4, 15). Ionomycin also can act as a mobile Ca2+ ionophore (4). The present study was not designed to investigate the mechanism by which ionomycin increased [Ca2+]i. However, any initial ionomycin-induced release of Ca2+ from internal stores apparently did not raise [Ca2+]i sufficiently to initiate SICs or ASTICs when external Ca2+ was replaced by Mg2+. In contrast, 10 mM caffeine, which consistently stimulated Ca2+ release from internal stores, also stimulated SICs and ASTICs.
Although measured in different cells under different conditions, there were many similarities in the properties of the caffeine-induced rise in [Ca2+]i and the SICs generated by caffeine. First, caffeine elicited a rise in global Ca2+ and generated a SIC in a comparable percentage of cells. In addition, caffeine elicited multiple Ca2+ transients in 50% of cells tested, whereas 33% of the voltage-clamped cells exposed to caffeine elicited more than one SIC. Third, caffeine could initiate SICs (this study) and could elicit a rise in [Ca2+]i in a Ca2+-deficient solution, which Mg2+ was substituted for Ca2+ in the external solution (12). Thus both types of caffeine response could be generated by release of Ca2+ from internal stores.
Previously, Tompkins and Parsons (26) found that, during exposure to PMA, ionomycin-induced ASTIC generation was increased. Ionomycin was used in the prior study to stimulate ASTIC generation to avoid any potential effects of PMA on VDCCs. In the present study, we found that PMA had no noticeable effect on SIC or ASTIC generation by caffeine. The lack of potentiation of caffeine-induced ASTIC generation by PMA quite likely is due to differences in the time course of the change in intracellular Ca2+ generated by ionomycin vs. caffeine. During exposure to ionomycin, intracellular Ca2+ rises and remains elevated for the duration of exposure. ASTIC activity is initiated, gradually reaches a steady-state level, and then remains roughly constant over the duration of ionomycin exposure. Under these conditions, ASTIC activity was increased during exposure to PMA. In contrast, caffeine evokes a transient elevation of [Ca2+]i. We suggest that the duration of Ca2+ elevation was too brief for PMA to significantly enhance ASTIC generation. Phorbol ester treatment also is suggested to enhance neuronal clearance of cytosolic Ca2+ (28). This latter action could potentially shorten the duration of the caffeine-evoked Ca2+ transient and counteract any potentiating effect on ASTIC generation.
We suggest that the caffeine-induced SIC, the tail current elicited by depolarizing steps, and the small inward shift in holding current during ionomycin application (when Ca2+ was present) most likely represent a nonselective cation current activated by a rise in [Ca2+]i. Both the tail current and SIC were recorded in cells treated with PPADs, suggesting that neither results from the activation of P2X receptors by ATP. Previously, Tompkins and Parsons (26) demonstrated that the tail current was Ca2+ dependent, not being present when barium (Ba2+) replaced Ca2+ as the charge carrier traversing activated VDCCs. In addition, the tail current was either not present or greatly reduced in amplitude when NMG was substituted for external Na+, but remained in a chloride-deficient solution (J. D. Tompkins and R. L. Parsons, unpublished observations). We show here that the SIC was not seen in the Na+-deficient solution. Thus both the SIC and the tail current are activated by a rise in [Ca2+]i and require Na+ influx in cells voltage-clamped to negative potentials. The inward shift in holding current was only apparent during ionomycin application in the Ca2+-containing solution.
Subunits of the canonical TRPC channel family are suggested to comprise a number of nonselective cation channels, which can be activated by variable mechanisms (18). It was considered plausible that members of this cation channel family, some of which can be activated by rise in intracellular Ca2+, might form the ionic pathway responsible for generation of the SIC. RT-PCR analysis indicated that multiple TRPC channel transcripts are present in extracts of guinea pig stellate ganglia. The TRPC channel transcript expression indentified for the guinea pig stellate neurons is similar to that found previously for rat superior cervical ganglia (1). Future studies, which are outside of the scope of this study, need to focus on a potential role of TRPC channels in SIC generation.
Previously, Tompkins and Parsons (26) noted that depolarizing steps elicited ASTICs and a tail current when Ca2+ was the charge carrier through activated VDCCs. In contrast, only ASTICs, but not the tail current, were recorded when Ba2+ replaced Ca2+ in the bath solution. This suggests that the ionic selectivity of the “Ca2+ sensor,” which gates ASTIC generation, is different from that gating the depolarization-activated tail current. ASTICs and the tail current also were recorded when Ca2+ was replaced by strontium (Sr2+) (J. D. Tompkins and R. L. Parsons, unpublished observation), indicating that Ca2+ and Sr2+, but not Ba2+, can activate the underlying putative nonselective cation conductance suggested to be responsible for the depolarization-activated tail current.
In a number of neurons, CICR is known to contribute to the rise in Ca2+ that supports neurotransmitter and/or neuropeptide release from active zones in nerve terminals or from ectopic release sites (5, 23, 30). In some instances, there is a critical need for a CICR-induced rise in [Ca2+]i to support release. For example, the somatodendritic release of dopamine from neurons in the substantia nigra is greatly enhanced by Ca2+ released from both inositol 1,4,5-trisphosphate receptor and ryanodine receptor-sensitive endoplasmic reticulum Ca2+ stores (16). However, the need for Ca2+ release from internal stores to support somatodendritic dopamine release appears to be species dependent, as pharmacological depletion of internal Ca2+ stores did not affect somatodendritic dopamine release in slices of mouse midbrain (6). In contrast, data obtained in the present study demonstrate that CICR is not a requisite for somatic quantal ATP release from dissociated guinea pig sympathetic stellate neurons under the conditions of the present experiments. Thus, although these neurons have a demonstrated CICR mechanism, the rise in [Ca2+]i due to this mechanism is not required for quantal ATP release. The differing requirements for the participation of Ca2+ release from internal stores in different neurons indicate that mechanisms regulating somatic transmitter release very likely are specific to each neuron type.
Both Ca2+ influx and Ca2+ release from internal stores can initiate ASTIC generation. However, from comparison of the prior work of Tompkins and Parsons (26) and results of the present study, it would appear that ASTIC generation may be more effectively initiated by Ca2+ influx, suggesting Ca2+ elevation in local domains may be important in regulating the somatic release of ATP. This question cannot be answered with the imaging techniques used in the present study. Rather, elucidation of changes in local domains vs. global elevations in the activation of ATP release requires more sophisticated imaging approaches such as total internal reflection fluorescence microscopy, a technique used previously to analyze Ca2+ domains, and changes in cytosolic Ca2+ in astrocytes (21, 22).
Conclusions and physiological relevance.
ATP is a key signaling molecule in both the CNS and peripheral nervous system (3). Although known for some time to be involved in classical neurotransmission, ATP now is thought also to mediate communication between neurons and adjacent nonneural cells. This was initially suggested for neuron to glial cell signaling in the CNS, but recent studies document ATP signaling between neurons and glial cells in the enteric nervous system (7). ATP can be released from the soma of dissociated DRGs, an observation suggesting that ATP might mediate local communication between sensory neurons and satellite cells (31). Likewise, somatic release of ATP within autonomic ganglia could be a signaling mechanism between adjacent neurons or nonneural cells. The potential somatic release of ATP within intact sympathetic ganglia needs to be investigated in future studies.
Results reported in this paper were obtained using multiuser research core facilities supported by National Center for Research Resources Grant 1 P20 RR-016435.
No conflicts of interest, financial or otherwise, are declared by the author(s).
- Copyright © 2010 the American Physiological Society