HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 96/3/1093    most recent 00283.2006v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (4) Citing Articles via Google Scholar Google Scholar Articles by Jackson, J. G. Articles by Thayer, S. A. Search for Related Content PubMed PubMed Citation Articles by Jackson, J. G. Articles by Thayer, S. A.

Mitochondrial Modulation of Ca²⁺-Induced Ca²⁺-Release in Rat Sensory Neurons

Department of Pharmacology, University of Minnesota Medical School, Minneapolis, Minnesota

Submitted 15 March 2006; accepted in final form 2 June 2006

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Ca²⁺-induced Ca²⁺-release (CICR) from ryanodine-sensitive Ca²⁺ stores provides a mechanism to amplify and propagate a transient increase in intracellular calcium concentration ([Ca²⁺]_i). A subset of rat dorsal root ganglion neurons in culture exhibited regenerative CICR when sensitized by caffeine. [Ca²⁺]_i oscillated in the maintained presence of 5 mM caffeine and 25 mM K⁺. Here, CICR oscillations were used to study the complex interplay between Ca²⁺ regulatory mechanisms at the cellular level. Oscillations depended on Ca²⁺ uptake and release from the endoplasmic reticulum (ER) and Ca²⁺ influx across the plasma membrane because cyclopiazonic acid, ryanodine, and removal of extracellular Ca²⁺ terminated oscillations. Increasing caffeine concentration decreased the threshold for action potential-evoked CICR and increased oscillation frequency. Mitochondria regulated CICR by providing ATP and buffering [Ca²⁺]_i. Treatment with the ATP synthase inhibitor, oligomycin B, decreased oscillation frequency. When ATP concentration was held constant by recording in the whole cell patch-clamp configuration, oligomycin no longer affected oscillation frequency. Aerobically derived ATP modulated CICR by regulating the rate of Ca²⁺ sequestration by the ER Ca²⁺ pump. Neither CICR threshold nor Ca²⁺ clearance by the plasma membrane Ca²⁺ pump were affected by inhibition of aerobic metabolism. Uncoupling electron transport with carbonyl cyanide p-trifluoromethoxy-phenyl-hydrazone or inhibiting mitochondrial Na⁺/Ca²⁺ exchange with CGP37157 revealed that mitochondrial buffering of [Ca²⁺]_i slowed oscillation frequency, decreased spike amplitude, and increased spike width. These findings illustrate the interdependence of energy metabolism and Ca²⁺ signaling that results from the complex interaction between the mitochondrion and the ER in sensory neurons.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Ca²⁺-induced Ca²⁺-release (CICR) was originally described for excitation-contraction coupling in cardiac myocytes (Fabiato and Fabiato 1975) and is the process by which a transient elevation of the intracellular Ca²⁺ concentration ([Ca²⁺]_i) is amplified by the activation of ryanodine receptors (RyRs) on the sarcoplasmic reticulum. RyRs are also expressed in nervous tissue (Furuichi et al. 1994; Lai et al. 1992; Lokuta et al. 2002; McPherson et al. 1991), and RyR-mediated CICR from the endoplasmic reticulum (ER) has been described for neurons (Friel and Tsien 1992; Kuba 1980; Shmigol et al. 1995; Thayer et al. 1988; Usachev and Thayer 1997; Usachev et al. 1993). Indeed, RyR-mediated CICR regulates numerous neuronal processes, including synaptic plasticity (Kohda et al. 1995), neurotransmitter release and exocytosis (Peng 1996; Smith and Cunnane 1996), and differentiation and neurite outgrowth (Gomez et al. 1995; Holliday et al. 1991).

The openings of RyRs in response to Ca²⁺ are normally uncoordinated events, limited temporally and spatially (Bootman et al. 2001). These elemental events, however, have the potential to transform small, graded Ca²⁺ increases into regenerative responses by the subsequent recruitment and synchronization of Ca²⁺ release (Berridge 1998; Usachev and Thayer 1999b). In sensory neurons, regenerative CICR occurs after [Ca²⁺]_i reaches a specific threshold at which point Ca²⁺ release no longer depends on Ca²⁺ influx (Usachev and Thayer 1997). Indeed, when RyRs are sensitized, regenerative CICR will establish stable oscillations in [Ca²⁺]_i (Friel 1995; Usachev and Thayer 1999b).

Ca²⁺ regulatory processes operate interdependently to control the fidelity of information encoded in the Ca²⁺ signal (Berridge 1998; Thayer et al. 2002). CICR is particularly sensitive to Ca²⁺ regulatory processes because Ca²⁺ conveys the stimulus and mediates the response (Usachev and Thayer 1999b). Mitochondria might be expected to exert a strong influence on ER Ca²⁺ signals because of their close proximity to the ER (Rizzuto et al. 1998), their large capacity for Ca²⁺ (Werth and Thayer 1994), and their production of ATP (Hajnoczky et al. 1995; Nicholls and Budd 2000).

In this report, we studied CICR oscillations to determine how CICR is influenced by Ca²⁺ regulatory processes. We found that modulating the sensitivity of RyR to [Ca²⁺]_i or modulating cellular Ca²⁺ homeostasis altered oscillation frequency and the shape of the [Ca²⁺]_i spike. Mitochondria exerted a powerful influence on CICR by buffering Ca²⁺ and providing ATP. We also discovered that specific Ca²⁺ regulatory mechanisms were supported by ATP from different sources, with the ER Ca²⁺-ATPase relying on mitochondrially generated ATP, whereas the plasma membrane Ca²⁺ pump proved insensitive to inhibition of oxidative phosphorylation. These findings illustrate the complex interplay between mitochondria and ER Ca²⁺ regulatory processes in sensory neurons.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Cell culture

Rat dorsal root ganglion (DRG) neurons were grown in primary culture as described previously (Thayer and Miller 1990). In brief, 1- to 3-day-old Sprague-Dawley rats were killed by decapitation with sharp scissors under a protocol approved by the University of Minnesota Institutional Animal Care and Use Committee. Ganglia were dissected from the thoracic and lumbar regions, incubated at 37°C in collagenase-dispase (V. alginolyticus/ B. Polymyxa; 0.8 and 6.4 U/ml, respectively; Roche Diagnostics, Indianapolis, IN) for 45 min, dissociated by trituration through a flame-constricted pipette, and then plated onto laminin-coated (50 mg/ml) glass coverslips (25 mm diam). Cells were grown in Ham's F12 medium supplemented with 5% heat-inactivated horse serum, 50 ng/ml NGF-7S (mouse submaxillary gland; Sigma), 4.4 mM glucose, 2 mM L-glutamine, modified Eagle's medium vitamins, and penicillin–streptomycin (100 U/ml and 100 mg/ml, respectively). Cultures were maintained at 37°C in a humidified atmosphere of 5% CO₂. Cells were used on the second and third days in vitro.

Microfluorometric recording of [Ca²⁺]_i

[Ca²⁺]_i was recorded from cultured rat DRG neurons using indo-1 or mag-indo-1-based microfluorometry (Grynkiewicz et al. 1985). Cells were loaded with 2 µM indo-1 acetoxymethyl ester (AM) for 45 min at 22°C in HEPES-buffered Hanks salt solution (HHSS) containing 0.5% bovine serum albumin or with mag-indo-1 AM at a concentration of 7.5 µM in 0.02% Pluronic F-127 at 37°C for 30 min. The cells were washed for 30 min in dye-free HHSS at 37°C before initiating recording. HHSS was composed of the following (in mM): 20 HEPES, 137 NaCl, 1.3 CaCl₂, 0.4 MgSO₄, 0.5 MgCl₂, 5.0 KCl, 0.4 KH₂PO₄, 0.6 Na₂HPO₄, 3.0 NaHCO₃ and 5.6 glucose. Cells were rinsed in HHSS and the indicator allowed to de-esterify for 10 min prior to start of the experiment. Cells were placed in a flow-through chamber (Thayer and Miller 1990) (10-s solution exchange) that was mounted on the stage of an inverted epi-fluorescence microscope (Nikon) equipped with a 70x objective (NA 1.15). Experiments were performed at room temperature (22°C). The dye was excited at 350 nm (10 nm band-pass), and emission was detected at 405 (20) and 490 (20) nm. Fluorescence was monitored by a pair of photomultiplier tubes (Thorn, EMI, Fairfield, NJ) operating in photon-counting mode. The output signals were then passed through a frequency to voltage converter and digitized using a Digidata 1322a A/D Converter (Axon Instruments, Foster City, CA). Data were stored and analyzed on a personal computer.

Fluorescence changes were converted to [Ca²⁺]_i by using the formula [Ca²⁺]_i = K_d(R – R_min)/(R_max – R), where R is 405/490 nm fluorescence ratio (Grynkiewicz et al. 1985). The dissociation constant (K_d) for indo-1 was 250 nM, and was the ratio of fluorescence emitted at 490 nm and measured in the absence and presence of Ca²⁺. R_min, R_max, and were determined by bathing intact cells in 2 µM ionomycin in Ca²⁺-free buffer (1 mM EGTA) and saturating Ca²⁺ (5 mM Ca²⁺). Values for R_min, R_max, and were 0.4, 3.65, and 3.34, respectively. After completion of each experiment, cells were wiped from the microscope field using a cotton-tipped applicator. Background light levels were determined at each wavelength (5% of cell counts) and subtracted prior to calculating ratios. Potential cytoplasmic contamination with mag-indo-1 prevented us from calibrating the indicator; thus we reported [Ca²⁺]_ER as the ratio (R) of fluorescence intensity of the Ca²⁺-bound (405 nm) relative to the Ca²⁺-free form (490 nm) of the dye. To evoke action potentials in intact neurons, extracellular field stimulation was used (Piser et al. 1994). Exponential functions were fitted to the data using a nonlinear, least-squares curve fitting algorithm (Origin 4.1 software; OriginLab, Northampton, MA) (Usachev et al. 2006). Data are presented as means ± SE.

Simultaneous whole cell patch-clamp and microfluorimetric recording

Electrical measurements and [Ca²⁺]_i were recorded from cultured rat DRG neurons by using the whole cell patch-clamp technique (Hamill et al. 1981) in combination with indo-1-based microfluorimetry. Microfluorimetry was performed as described in the preceding text. Indo-1 (50 µM) was loaded into the cells via the patch pipette. Patch-clamp data were acquired using an Axopatch 200A amplifier. Both the patch clamp and the optical signals were digitized using a Digidata 1322a A/D converter, stored on a personal computer and analyzed with pClamp 9.0 (Axon Instruments). Patch pipettes were pulled from borosilicate glass (2–4 M; Narishige, Tokyo, Japan) on a Sutter Instruments (Novato, CA) P-87 micropipette puller and filled with the following solution (in mM): 125 potassium gluconate, 10 KCl, 3 Mg-ATP, 1 MgCl₂, 10 HEPES, and 0.05 indo-1, pH 7.25 with KOH, 290 mosM/kg with sucrose. Extracellular recording solution contained (in mM): 140 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 10 HEPES, and 10 glucose, pH 7.35 with NaOH, 310 mosM/kg with sucrose. Background light levels were collected in the cell-attached configuration and subtracted prior to calculating the ratios. Data are presented as means ± SE. Fluorescence intensities were converted to [Ca²⁺]_i as described in the preceding text.

Simultaneous confocal imaging of intracellular and ER [Ca²⁺]

[Ca²⁺] imaging was performed on a Fluoview 300 laser scanning confocal microscope attached to an inverted Olympus IX70 microscope equipped with an PlanApo 60x objective (NA 1.40; Olympus Optical, Tokyo, Japan). Cultures were first loaded with the low-affinity Ca²⁺ indicator mag-fluo-4-AM (10 µg/ml) at 37°C for 45 min and then loaded with X-rhod-1 AM (2 µM; room temperature; 45 min). Cells were rinsed in HHSS and the indicators allowed to de-esterify for 15 min prior to start of the experiment. Mag-fluo-4 and X-rhod-1 were excited with the 488 (Argon) and 540 nm (He-Ne) laser lines and the fluorescence imaged at 510–540 and >605 nm, respectively (Li et al. 2003).

Reagents

Indo-1, Mag-Indo-1, X-rhod-1, and Mag-Fluo-4 were obtained from Invitrogen (Carlsbad, CA). All other reagents were purchased from Sigma (St. Louis, MO).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
CICR oscillations in rat dorsal root ganglia neurons

We have previously identified a regenerative form of CICR in rat dorsal root ganglion (DRG) neurons in culture that is due to Ca²⁺ release from ryanodine-sensitive stores (Usachev and Thayer 1997). A small proportion (<5%) of these neurons display [Ca²⁺]_i oscillations in the presence of caffeine (5 mM) and Ca²⁺ influx (Usachev and Thayer 1999a). Supplementing the culture medium with NGF (50 ng/ml; n = 5), BDNF (50 ng/ml; n = 4), or GDNF (50 ng/ml; n = 4) did not increase the proportion of oscillating neurons. Here we employed CICR oscillations to examine the interplay between various Ca²⁺ regulatory processes in single DRG neurons in primary culture. [Ca²⁺]_i was recorded from medium to large DRG neurons (38–65 µm diam) using indo-1-based microfluorimetry (Werth et al. 1996). CICR oscillations depended on both Ca²⁺ influx and mobilization of Ca²⁺ stores (Fig. 1A). Depolarization of DRG neurons with 25 mM K⁺ alone caused a modest (20 ± 10 nM; n = 3) elevation of [Ca²⁺]_i that returned to basal levels following cessation of the stimulus. Application of caffeine (5 mM) produced a more robust increase (700 ± 200 nM; n = 3) in [Ca²⁺]_i that recovered to near basal levels within 5 min in the maintained presence of caffeine. Concomitant application of 25 mM K⁺ and 5 mM caffeine elicited repetitive Ca²⁺ oscillations. These oscillations were remarkably stable and could last several hours in the maintained presence of these agents. Oscillations were analyzed for changes in amplitude, width, and frequency. The mean amplitude for a [Ca²⁺]_i spike was 177 ± 6 nM, the mean frequency was 1.5 ± 0.3 spikes/min, and the mean width was 24 ± 1 s (n = 209). Cells capable of regenerative CICR were identified by stable [Ca²⁺]_i oscillations in the presence of 25 mM K⁺ and 5 mM caffeine (Fig. 1).

View larger version (26K): [in this window] [in a new window]   FIG. 1. Ca2+-induced Ca2+-release (CICR) oscillations require Ca2+ influx and Ca2+ release. [Ca2+]i was measured in single dorsal root ganglion (DRG) neurons in culture using indo-1-based microfluorimetry as described in METHODS. Drugs were applied by superfusion at the times indicated by the horizontal bars. A: neurons were subjected to depolarization (25 mM K+) and Ca2+ store mobilization (5 mM caffeine) separately and jointly as indicated by application bars. These treatments in combination elicited repetitive [Ca2+]i oscillations. Trace is representative of 3 recordings in which these treatments were applied sequentially. B, D, and E: representative recordings depict cells in the continuous presence of 25 mM K+ and 5 mM caffeine. B: ryanodine (100 nM) inhibited CICR oscillations (n = 4). C: representative trace from the same oscillating cell in 25 mM K+ and either 5 (thick bar) or 2.5 (thin bar) mM caffeine (n = 16). D: cyclopiazonic acid (CPA; 100 nM) terminated CICR oscillations in 6 of 7 cells. E: oscillation frequency slowed when extracellular Ca2+ was reduced fivefold from 1.26 mM (thick bar) to 0.25 mM (thin bar; n = 8).

The ryanodine receptor is sensitive to caffeine (Zucchi and Ronca-Testoni 1997). Consistent with this observation, caffeine concentration modulated CICR oscillations (Fig. 1C). Decreasing caffeine concentration from 5 to 2.5 mM decreased oscillation frequency by 36 ± 11% (P < 0.0001; n = 16) and increased inter-spike interval by 57 ± 12% (P < 0.001; n = 16). Decreasing caffeine concentration increased peak width by 83 ± 14% (P < 0.001; n = 16). Bath application of cyclopiazonic acid (CPA; 100 nM), a reversible antagonist of the sarcoplasmic-endoplasmic reticulum Ca²⁺-ATPase (SERCA), blocked Ca²⁺ oscillations in six of seven cells (Fig. 1D). Oscillations resumed after washout of CPA. Combined, these experiments indicate that CICR oscillations depend on the release from, and subsequent refilling of, ryanodine-sensitive Ca²⁺ stores.

An increase in [Ca²⁺] at the mouth of the ryanodine receptor triggers release of Ca²⁺ from intracellular stores. Furthermore, the stores are refilled with Ca²⁺ by uptake from the cytoplasm. Thus we hypothesized that decreasing basal [Ca²⁺]_i would decrease CICR oscillation frequency. In Fig. 1E, we show that reducing extracellular [Ca²⁺] from 1.26 to 0.25 mM (a 1:5 dilution) decreased oscillation frequency by 44 ± 14% (P < 0.01; n = 8).

Mitochondria modulate regenerative CICR

Mitochondria play an important role in the regulation of [Ca²⁺]_i. They buffer Ca²⁺ transients, accumulating Ca²⁺ when [Ca²⁺]_i is elevated above a set point and slowly releasing Ca²⁺ back to the cytosol when [Ca²⁺]_i falls below the set point (Budd and Nicholls 1998). Additionally, close apposition of the ER and mitochondria has been observed (Rizzuto et al. 1998). Thus mitochondria might be expected to alter Ca²⁺ oscillations.

Ca²⁺ accumulated by mitochondria is released via Na⁺-Ca²⁺ exchange (Baron and Thayer 1997; Werth and Thayer 1994). To test the contribution of mitochondria to changes in spike duration, we applied CGP37157 (1 µM; IC₅₀ = 0.80 µM), an inhibitor of mitochondrial Na⁺-Ca²⁺ exchange (Cox et al. 1993), to DRG neurons undergoing CICR oscillations (Fig. 2A). Inhibition of Na⁺-dependent Ca²⁺-efflux from mitochondria decreased oscillation frequency by 26 ± 6% (P < 0.001; n = 15) and decreased spike width by 45 ± 5% (P < 0.001; n = 15). The reduced frequency may be due to CGP37157-dependent lowering of inter-spike [Ca²⁺] resulting in increased time to reach the threshold [Ca²⁺]_i needed to initiate a spike. The decreased spike width in the presence of CGP 37157 suggests that Ca²⁺ release from mitochondria prolongs the duration of the increase in [Ca²⁺]_i.

View larger version (48K): [in this window] [in a new window]   FIG. 2. Mitochondria regulate CICR by buffering Ca2+ and producing ATP. [Ca2+]i was measured in single DRG neurons in culture using indo-1-based microfluorimetry as described in METHODS. Application bars indicate the duration of treatments. Recordings were in the continued presence of 5 mM caffeine and 25 mM K+. A: inhibiting Na+-dependent Ca2+ efflux from mitochondria with 1 µM CGP37157 slowed CICR oscillations (n = 15). B: uncoupling electron transport with 50 nM carbonyl cyanide p-trifluoromethoxy-phenyl-hydrazone (FCCP) slowed the frequency, increased the amplitude, and reduced the width of CICR oscillations (n = 6). Inset: individual [Ca2+]i spikes on an expanded time scale in the presence (red) and absence (black) of FCCP from the times indicated by the boxes. C: inhibiting the ATP synthase with 1 µM oligomycin B slowed CICR oscillations (n = 10). D: representative trace shows effects of caffeine (5 mM), oligomycin (1 µM), and FCCP (50 nM) on a cell held at –70 mV in the whole cell configuration with 3 mM Mg-ATP and 50 µM indo-1 in the pipette (n = 4). Oligomycin did not affect CICR in the presence of cytoplasmic ATP.

Both Ca²⁺ uptake and ATP production are coupled to the mitochondrial proton gradient. To distinguish changes in Ca²⁺ buffering from changes in ATP production, we examined the effects of the ATP-synthase inhibitor oligomycin B. As seen in Fig. 2C, oligomycin (1 µM) decreased oscillation frequency by 69 ± 10% (P < 0.001; n = 10). Oligomycin B and FCCP applied together stopped oscillations (n = 3; data not shown). The slowed oscillation frequency produced by oligomycin B suggests that mitochondrially generated ATP regulates CICR oscillations.

To determine the role of mitochondrially produced ATP directly, ATP concentration was held constant at 3 mM using the whole cell configuration of the patch clamp technique (Fig. 2D). Cells with the ability to oscillate were first identified in 25 mM K⁺ and 5 mM caffeine. Caffeine was removed and cells allowed to equilibrate before forming a seal. Cells were held between –55 and –70 mV. Re-addition of caffeine resulted in a large transient increase in [Ca²⁺], with [Ca²⁺] oscillations starting when [Ca²⁺]_i recovered near basal levels. In the maintained presence of ATP, oligomycin B (1 µM) no longer decreased spike frequency. This result indicates that mitochondrial inhibitors slowed oscillation frequency by decreasing ATP production. This slowing might result from loss of ATP-dependent sensitization of the ryanodine receptor (Laver et al. 2001) or might be due to slowed ATP-dependent SERCA activity (Landolfi et al. 1998). When ATP concentration was held at 3 mM, FCCP increased oscillation frequency by 46 ± 8% (P < 0.05) relative to oscillations in oligomycin B, suggesting that mitochondrial Ca²⁺ uptake increases the time it takes [Ca²⁺]_i to rise to the threshold for regenerative Ca²⁺ release.

Caffeine but not aerobic ATP modulates the threshold for regenerative CICR

Binding sites for ATP have been identified on each of the three RyR isoforms (Zarka and Shoshan-Barmatz 1993) and ATP potentiates [³H]-ryanodine binding to RyRs isolated from brain (Mcpherson and Campbell 1993). Millimolar concentrations of ATP stimulate Ca²⁺ release and augment CICR in skinned muscle fibers (Duke and Steele 1998; Endo et al. 1970), sarcoplasmic reticulum vesicles (Meissner et al. 1986), and through single channels (McGarry and Williams 1994; Meissner et al. 1988; Rousseau et al. 1986). In Fig. 4, we examined the effects of changing caffeine concentration and of ATP depletion on the threshold for action potential-induced regenerative CICR. Previous work demonstrated that all-or-none CICR displayed a distinct threshold for activation (Usachev and Thayer 1997). In this series of experiments, neurons previously identified as capable of oscillating were treated with 5 mM caffeine in HHSS. Ca²⁺ influx was driven by action potentials delivered using extracellular field stimulation at a frequency of 1 Hz. The electrical stimulation was maintained until the Ca²⁺ increase became regenerative and thus no longer dependent on maintained influx. We defined the inflection point where the [Ca²⁺]_i increase becomes supralinear as the threshold for regenerative CICR. In untreated neurons (Fig. 3, A and B), this threshold for CICR remained constant during repeated stimulation. The average threshold for the first three responses was 81 ± 4 nM, which was similar to the average of the last three responses (79 ± 4; n = 8). Decreasing caffeine concentration from 5 to 2.5 mM increased the [Ca²⁺]_i threshold required to evoke a regenerative response (Fig. 3, C and D) from 134 ± 13 to 151 ± 16 nM (P < 0.05; n = 9), consistent with the sensitization of the ryanodine receptors by caffeine (Usachev and Thayer 1997). The increased threshold required a longer train of action potentials to trigger CICR (23 ± 7 to 108 ± 20 s; P < 0.001; n = 9). Thus caffeine lowers the threshold [Ca²⁺] required to evoke regenerative Ca²⁺ release.

View larger version (33K): [in this window] [in a new window]   FIG. 4. Plasma membrane Ca2+-ATPase (PMCA)-mediated Ca2+ clearance is not modulated by aerobic ATP. [Ca2+]i was measured in single DRG neurons in culture using indo-1-based microfluorometry as described in METHODS. A: inhibiting PMCA function by increasing pH0 from 7.45 to 7.75 increased CICR oscillation frequency in the continued presence of 5 mM caffeine and 25 mM K+ (n = 12). B: representative trace depicts a protocol for studying PMCA-mediated recovery kinetics of evoked [Ca2+]i transients. Brief trains of action potentials (4 s, 2–6 Hz) were evoked every 4 min by extracellular field stimulation (filled triangles). The SERCA inhibitor, CPA (5 µM), was present throughout the recording (n = 8). C: representative Ca2+ transients from B (indicated by boxes) are normalized and superimposed. D: inhibiting ATP synthesis (oligomycin B; 1 µM) did not alter recovery from field stimulation evoked [Ca2+]i transients. Oligomycin B was applied at the time indicated by the horizontal bar (n = 9). E: representative [Ca2+]i responses from D (indicated by boxes) in control (black) and oligomycin-treated (1 µM; red) conditions are normalized and superimposed.

View larger version (32K): [in this window] [in a new window]   FIG. 3. Caffeine concentration but not aerobic ATP controls the threshold for Ca2+ release from the endoplasmic reticulum. A–F: [Ca2+]i was measured in single DRG neurons in culture using indo-1-based microfluorimetry as described in METHODS. Electrical stimulation (1 Hz; indicated by horizontal bars) was applied until threshold was reached, and the response became regenerative. A: CICR can be reproducibly evoked by extracellular electrical field stimulation in the presence of 5 mM caffeine (n = 8). B: representative transients (black and red boxes) from A are superimposed. C: same DRG neuron was treated with 5 (wide bar) and 2.5 (narrow bar) mM caffeine by bath application (n = 9). Increasing caffeine concentration decreased the threshold for field stimulation induced CICR. D: representative transients evoked in 5 (black) and 2.5 (red) mM caffeine are superimposed. Note that the threshold for a regenerative response (inflection point in Ca2+ trace) was higher in 2.5 mM caffeine and thus required a more prolonged period of Ca2+ influx (red horizontal bar) to evoke a regenerative response. E: inhibiting ATP synthase with oligomycin B (1 µM; n = 10) did not alter CICR threshold. F: representative [Ca2+]i transients from (E) evoked in the absence (black) and presence of oligomycin B (1 µM; red) are superimposed.

Plasma membrane Ca²⁺-ATPases do not depend on aerobically derived ATP

In rat sensory neurons, one of the principle components of Ca²⁺ clearance is the plasma membrane Ca²⁺-ATPase (PMCA). PMCAs exchange 2 H⁺ for each Ca²⁺ extruded. Elevating extracellular pH (pH₀) inhibits PMCA function by decreasing the protons available to exchange for Ca²⁺ (Benham et al. 1992; Park et al. 1996). We hypothesized that increased pH₀ would elevate basal [Ca²⁺]_i and thus accelerate spike frequency. As demonstrated in Fig. 4A, increasing pH₀ from 7.45 to 7.75 increased oscillation frequency by 73 ± 31% (P < 0.001; n = 12).

These results indicate that PMCAs regulate CICR by controlling basal [Ca²⁺]. We decided to determine whether inhibition of mitochondrial ATP synthesis altered PMCA activity. Neurons displaying CICR oscillations were identified in the presence of caffeine and 25 mM K⁺ and returned to HHSS. [Ca²⁺]_i increases were elicited by extracellular field stimulation (4s, 2–6 Hz) every 4 min. In sensory neurons, these small calcium transients recover to basal levels by the combined processes of sequestration and efflux (Benham et al. 1992; Usachev et al. 2002). Thus to study PMCA function in isolation, SERCA-mediated Ca²⁺ was blocked by CPA (5 µM). [Ca²⁺]_i transient amplitudes were kept <400 nM; under these conditions, recovery does not depend on mitochondrial uptake or Na⁺-dependent Ca²⁺ exchange across the plasma membrane (Usachev et al. 2002; Werth and Thayer 1994). A time constant for recovery was calculated by fitting the recovery phase of the [Ca²⁺]_i transients to a monoexponential decay function ( = 23 ± 2 s; n = 16). As depicted in Fig. 4B, these action potential elicited [Ca²⁺]_i responses were consistent in amplitude and rate of recovery. In Fig. 4C, normalized [Ca²⁺]_i transients from the recording in Fig. 4B were superimposed to demonstrate the reproducibility of the recovery rate. The mean recovery rate for the last three responses was 97 ± 11% of the mean of the first three (n = 8). As shown in Fig. 4, D and E, treatment with oligomycin B did not significantly decrease the rate of recovery. The mean recovery rate for the last three responses in the presence of 1 µM oligomycin B was 95 ± 10% of the mean of the first three control responses (n = 8). Thus it appears that PMCA activity in sensory neurons does not depend on mitochondrially derived ATP. Furthermore, our ability to evoke action potentials in the presence of oligomycin B suggests that other plasma membrane functions remained intact. Thus with glucose present in the bathing medium (5.6 mM), glycolysis supports plasma membrane ATPases in the presence of oligomycin.

Aerobically derived ATP regulates the rate of ER refilling

We next tested the hypothesis that sequestration of Ca²⁺ into the ER by SERCA is dependent on ATP derived from mitochondrial oxidative phosphorylation. To directly visualize [Ca²⁺]_ER, we loaded DRG neurons with the low-affinity Ca²⁺ indicator mag-indo-1 AM (K_d = 35 µM) using conditions that promote sequestration of the dye into the ER (Solovyova et al. 2002; Usachev et al. 2006). The contribution of cytosolic Ca²⁺ changes to changes in mag-indo-1 fluorescence is minimal due to the low affinity of the indicator for Ca²⁺.

Application of the SERCA inhibitor, CPA (5 µM) in Ca²⁺-free media depleted the ER of Ca²⁺ (Fig. 5A). Re-addition of Ca²⁺ to the medium refilled the ER. The time constant () for refilling was calculated by fitting the time course of the recovery to a monoexponential equation (Fig. 5B). The mean time constant for refilling was 152 ± 50 s (n = 8). As shown in Fig. 5A (and superimposed in Fig. 5B), the ER could be repeatedly depleted and refilled with the refilling rate reproducible for a given cell (₂/₁ = 0.75 ± 0.2; n = 8). Even though the cells started with varied [Ca²⁺]_ER, the standard protocol employed here returned [Ca²⁺]_ER to reproducible levels. The amplitude of the second refilling returned to 86 ± 10% of the original (Fig. 5E). Application of oligomycin B (1 µM) before the second refilling significantly decreased the rate of refilling (₂/₁ = 2.2 ± 0.5; P < 0.01; n = 7). The modest decline in the total extent of refilling in oligomycin-treated cells was not statistically significant. In Fig. 5D, traces of the first and second phases of refilling (from Fig. 5C) are superimposed on an expanded time scale to illustrate the slowed refilling in the presence of oligomycin B. These data are consistent with the hypothesis that mitochondrial derived ATP supports SERCA-mediated ER refilling and is responsible for the slowed CICR oscillations (Fig. 2C) following the inhibition of aerobic ATP synthesis.

View larger version (27K): [in this window] [in a new window]   FIG. 5. Inhibition of aerobic ATP production inhibits refilling the endoplasmic reticulum (ER) with Ca2+.[Ca2+] in the ER lumen was recorded using mag-indo-1. A: representative [Ca2+]ER recording. Blocking SERCA with CPA (5 µM) in Ca2+-free buffer (indicated by horizontal bars) produced a leak of Ca2+ from the ER, resulting in a decrease in [Ca2+]ER. Return of Ca2+ to the media allowed the ER to refill with Ca2+. This refilling process could be evoked repeatedly. B: superimposed, representative traces show reproducibility of the ER refilling between 1st (black) and 2nd (red) evoked [Ca2+]ER response from the recording in (A). ER refilling was well described by a monoexponential equation (heavy line). C: oligomycin B (1 µM) decreased the rate of ER refilling evoked using the same protocol as in (A). D: [Ca2+]ER traces from the recording in C were superimposed to show the rate of refilling in the absence (black) and presence of oligomycin (red). ER refilling was described by a monoexponential equation (heavy line). E: bar graph summarizes the change in refilling rate () from 8 control (white) and 7 oligomycin treated (red) neurons. The time constant from the 2nd refilling is normalized to the 1st and the change is expressed as a ratio (2/1). F: bar graph summarizes changes in the extent of refilling in the absence (white) and presence (red) of oligomycin B. [Ca2+]ER was defined as the difference between the stable plateau reached after refilling and the [Ca2+]ER in the depleted stores just prior to returning Ca2+ to the bath. [Ca2+]ER from the second refilling was normalized to the first refilling and expressed as a ratio ([Ca2+]ER 2/[Ca2+]ER 1).

In DRG neurons that oscillate, a small fraction display complex Ca²⁺ waveforms, reminiscent of burst firing of action potentials. Recovery of the [Ca²⁺]_i transients to basal levels in these neurons consisted of three phases (Fig. 6A) : there was an initial rapid decrease in [Ca²⁺]_i, a protracted plateau phase followed during which [Ca²⁺]_i remained elevated and secondary, smaller Ca²⁺ spikes were observed, and finally there was a rapid decrease in [Ca²⁺]_i to original resting levels. In cells spontaneously displaying these complex waveforms, the mean amplitude of the secondary [Ca²⁺]_i spikes was 31 ± 5% less than the primary [Ca²⁺]_i transient. The mean width of the primary [Ca²⁺]_i transient in cells with complex waveforms was 70 ± 8 s (n = 6), substantially larger than the mean spike width displayed in the general population of oscillating neurons. We therefore hypothesized that these secondary release events were the result of prolonged elevation of [Ca²⁺]_i. Decreasing caffeine concentration, a treatment that increases peak width, increased the duration of the plateau and the number of spikes per cluster (Fig. 6A).

View larger version (40K): [in this window] [in a new window]   FIG. 6. Complex CICR waveforms result from an extended plateau phase of the CICR [Ca2+]i spike. A–C: [Ca2+]i was measured in single DRG neurons in culture using indo-1-based microfluorometry as described in METHODS. Application bars indicate the duration of treatments. Recordings were in the continued presence of 5 mM caffeine and 25 mM K+ A: decreasing caffeine concentration increases plateau phase duration and increases the number of secondary CICR events in CICR oscillations displaying complex waveforms (n = 6). B: inhibition of Na+-dependent Ca2+-efflux (CGP37157; 1 µM) occasionally induces doublets (n = 4). C: bath application of CGP37157 (1 µM) in oscillations displaying complex waveforms normalizes the [Ca2+]i spiking pattern (n = 3). D: representative trace depicts [Ca2+]i (blue) and [Ca2+]ER (black) in a DRG neuron displaying a complex waveform. [Ca2+]i and [Ca2+]ER were simultaneously imaged in single DRG neurons in culture using the argon and HeNe laser lines of a confocal microscope to excite the Ca2+ indicators X-rhod-1 (ex. 540) and mag-fluo-4 (ex. 488). Representative trace shows the change in fluorescence (F) normalized to the initial fluorescence intensity (Fo) of mag-fluo-4 (black) and X-rhod-1 (blue) from a region encompassing the DRG cell body.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Regenerative CICR transforms [Ca²⁺]_i responses that are small, graded, and localized into large, all-or-none, global [Ca²⁺]_i signals. We show here that a subset of sensory neurons display a robust CICR response mediated by ryanodine-sensitive Ca²⁺ stores and that in the maintained presence of Ca²⁺ influx and caffeine, these cells exhibit sustained oscillations in [Ca²⁺]_i. We studied these oscillations to learn how other Ca²⁺ regulatory processes influence CICR. Modulating the sensitivity of the RyR to [Ca²⁺]_i altered oscillation frequency and the shape of the [Ca²⁺]_i spike. Mitochondrial Ca²⁺ buffering affected spike shape and mitochondrial ATP production modulated oscillation frequency. Aerobically produced ATP specifically modulated the rate of ER refilling but did not alter PMCA-mediated Ca²⁺ clearance or CICR threshold.

What is the role of regenerative CICR in sensory neurons? Depletion of ER Ca²⁺ can trigger apoptosis (Hajnoczky et al. 2003; Mengesdorf et al. 2001), but the rapid refilling of the Ca²⁺ store and the sustained nature of the oscillations suggest a more physiological role. In peripheral neurons, Ca²⁺ release from ryanodine-sensitive Ca²⁺ stores activates K⁺ channels that mediate a slow afterhyperpolarization (Moore et al. 1998; Nohmi et al. 1983), yet the massive regenerative response described here would seem excessive for this type of membrane localized signaling. During axon outgrowth, the growth cones of sensory neurons display ryanodine-sensitive [Ca²⁺]_i oscillations that appear to play a role in axon guidance (Gomez et al. 1995), yet the responses described here were recorded from the soma. We did not observe prominent neurite outgrowth during the time frame that [Ca²⁺]_i oscillations were evoked. RyR-mediated Ca²⁺ release in sensory neurons can be triggered by release of Ca²⁺ from inositol trisphosphate-sensitive Ca²⁺ stores that are also present (Hoesch et al. 2002); perhaps the principle role of RyR-mediated CICR is to amplify Ca²⁺ signals initiated by activation of the phospholipase C signaling cascade in sensory neurons. Ryanodine-sensitive Ca²⁺ stores play a role in neurosecretory processes (Emptage et al. 2001; Narita et al. 2000), and somatic release of neuropeptides from DRG neurons may influence other cells in the ganglion (Harding et al. 1999). The global Ca²⁺ wave that results from regenerative CICR in these cells propagates to the nucleus (Usachev and Thayer 1997) and [Ca²⁺]_i oscillations optimize the nuclear translocation of certain transcription factors (Tomida et al. 2003), suggesting that this regenerative type of signaling might participate in coupling excitation to transcription.

Oscillatory behavior was most prevalent in DRG neurons of medium to large diameter and in cells that were in culture <3 days. Large DRG cells relay low-threshold mechanical and proprioceptive stimuli (Hendry et al. 1999). We have not linked regenerative CICR to this particular sensory modality other than to suggest that it may be a mechanism to enhance information transfer in large neurons. Cells with low surface to volume ratios may be more dependent on intracellular Ca²⁺ stores to propagate [Ca²⁺]_i signals. The reduced prevalence of oscillatory cells with age in culture suggests that oscillations may be associated with growth and extension of processes rather than with their length. Acutely dissociated DRG neurons isolated from adult rats and mice display [Ca²⁺]_i increases in response to caffeine application (Kruglikov et al. 2004; Shmigol et al. 1996) and caffeine-induced [Ca²⁺]_i oscillations have been described for other peripheral neurons, including neurons of the sympathetic ganglia (Friel 1995). The survival of the various DRG subtypes depends on specific trophic factors (Oakley et al. 1997; Stucky et al. 1998, 2002). However, the proportion of DRG neurons that display the [Ca²⁺]_i oscillation phenotype was not affected by the presence of trophic factors in the cell culture medium.

Caffeine, at millimolar concentrations, sensitizes ryanodine receptors to [Ca²⁺]_i (Endo et al. 1970; Thayer et al. 1988), and we showed previously that caffeine concentration modulates the threshold for regenerative CICR in sensory neurons (Usachev and Thayer 1997). Here we showed that increasing the caffeine concentration increased the frequency of [Ca²⁺]_i oscillations, consistent with the predicted lowering of the [Ca²⁺]_i threshold for triggering CICR. Interestingly, increasing the caffeine concentration also decreased the duration of the individual transients. This observation is qualitatively similar to the observation that inhibition of RyRs by tetracaine yielded a lower frequency of spontaneous release events while simultaneously enhancing the amount of Ca²⁺ released per burst (Overend et al. 1997; Velez et al. 1997). The increased width of the [Ca²⁺]_i spike at lower caffeine concentrations may result from an increase in the driving force for Ca²⁺ to exit the ER. Other studies have shown a steep relationship between ER Ca²⁺ content and the magnitude of the cytosolic [Ca²⁺] increase (Trafford et al. 2000). Overall, our observations support a model in which caffeine controls oscillation frequency through its effects on RyR sensitivity and that changes in frequency modulate the duration of the individual [Ca²⁺]_i transients through secondary effects on luminal Ca²⁺ concentration.

We thought that the oscillations might serve as a useful assay to identify physiological agents that sensitize the RyR. However, we have not to date found physiological conditions that will substitute for caffeine. Other laboratories found that cyclic ADP ribose sensitizes the ER to [Ca²⁺]_i (Currie et al. 1992; Hua et al. 1994), although we have not found conditions in which this agent affects CICR in DRG neurons (Usachev and Thayer 1997). Thus we can only speculate on the physiological conditions that might elicit the regenerative responses studied here. However, factors that modulate regenerative responses likely influence elementary Ca²⁺ release events that occur in sensory neurons with prominent RyR-mediated Ca²⁺ release (Pacher et al. 2002).

Mitochondria play a complex role in neurons; they serve as a nexus where neuronal survival and death cues are integrated, provide a mechanism for local energy supply via oxidative phosphorylation, and regulate Ca²⁺ signals (Jacobson and Duchen 2004). Mitochondria take up Ca²⁺ into the mitochondrial matrix through the Ca²⁺ uniporter by a process driven by the large membrane potential established during oxidative phosphorylation (Kirichok et al. 2004). Ca²⁺ is recycled to the cytosol via a Na⁺-dependent Ca²⁺-exchanger (Baron and Thayer 1997; Zhang and Lipton 1999). These two processes enable mitochondria to blunt the amplitude of [Ca²⁺]_i increases and to translate increases in Ca²⁺ load to changes in the duration of [Ca²⁺]_i transients (Werth and Thayer 1994). Thus depolarizing the mitochondrial membrane potential with FCCP increased the amplitude and shortened the duration of individual [Ca²⁺]_i spikes. Inhibition of mitochondrial Na⁺/Ca²⁺ exchange in DRG neurons lowers the plateau phase of [Ca²⁺]_i recovery because the balance of Ca²⁺ release from mitochondria relative to Ca²⁺ clearance from the cytoplasm reaches a new steady state (Baron and Thayer 1997; Werth and Thayer 1994). Treatment with CGP37157, an inhibitor of Na⁺-dependent Ca²⁺-exchange accelerated the return to basal [Ca²⁺]_i producing, in most cells, a corresponding decrease in the spike frequency and a decrease in the spike width. However, in some cells, particularly those with prominent mitochondrial contributions to spike width, inhibition of Na⁺/Ca²⁺ exchange did not reduce the recovery to basal [Ca²⁺]_i but rather resulted in a plateau [Ca²⁺]_i near threshold for triggering CICR and thus increased the frequency of CICR oscillations. This observation raises some concern about potential adverse Ca²⁺ signaling events that might result from pharmacologic inhibition of mitochondrial Na⁺/Ca²⁺ exchange, a strategy proposed to improve ATP production in metabolic disorders (Cox and Matlib 1993). Taken together, the results from uncoupling electron transport and inhibition of Na⁺/Ca²⁺ exchange in mitochondria identify a transfer of Ca²⁺ from the ER to the matrix that is not detected by a cytoplasmic indicator. A similar shuttle of Ca²⁺ between the ER and mitochondria was shown to have a pacemaker role in regulating inositol triphosphate (IP₃)-induced Ca²⁺ oscillations in HeLa cells (Ishii et al. 2006). These observations are consistent with the close apposition of mitochondria to the ER (Rizzuto et al. 1998) and privileged mitochondrial uptake of Ca²⁺ released by IP₃-sensitive Ca²⁺ stores (Csordas et al. 1999; Rizzuto et al. 1993). Here, we showed that this type of Ca²⁺ signaling is present in neurons and can be extended to Ca²⁺ mobilization mediated by RyRs.

Mitochondria also modulated CICR through their contribution to ATP production as indicated by decreased oscillation frequency in the presence of oligomycin B or FCCP. This effect was prevented when ATP concentration was clamped to 3 mM by intracellular perfusion. Three potential targets for ATP regulation of CICR oscillations were considered: RyRs, PMCAs, and SERCAs. Although ATP has been shown to regulate Ca²⁺ release through RyRs in skinned muscle fibers (Duke and Steele 1998; Endo et al. 1970), sarcoplasmic reticulum vesicles (Meissner et al. 1986), and through single channels (McGarry and Williams 1994; Meissner et al. 1988; Rousseau et al. 1986), inhibition of mitochondrial ATP synthesis failed to alter the threshold for regenerative CICR in DRG neurons. The PMCAs and SERCAs, (which mediate Ca²⁺ clearance and sequestration, respectively) were other obvious targets, as they utilize ATP to transfer Ca²⁺ against its concentration gradient. PMCA-mediated Ca²⁺ clearance was not sensitive to mitochondrial poisons. However, we noted a significant decrease in the rate of SERCA-mediated ER refilling, suggesting that the decrease in CICR oscillation frequency was secondary to SERCA inhibition. That the two pumps should differ in their reliance on aerobic metabolism is interesting as both are P-type ATPases that share a common reaction mechanism (Kuhlbrandt 2004). Close apposition of the ER and mitochondria has been suggested to account for privileged uptake of IP₃- mediated Ca²⁺ release from the ER by the low-affinity mitochondrial Ca²⁺ uniporter (Csordas et al. 1999; Rizzuto et al. 1998). Mitochondria produce functional microdomains of ATP (Kennedy et al. 1999) consistent with the observation that local ATP levels change in regions close to the ER (Han et al. 1992). In principle, glycolytically derived ATP should be sufficient to maintain these processes. However, Kaasik and colleagues demonstrated that mitochondrially derived ATP is more effective than exogenously added ATP in sustaining Ca²⁺ uptake by the SR for contraction (Kaasik et al. 2001). Blockade of oxidative phosphorylation has also previously been shown to slow refilling of IP₃-sensitive Ca²⁺ stores in BHK-21 cells (Landolfi et al. 1998). SERCA-mediated refilling of the ER with Ca²⁺ seems to be especially sensitive to local changes in ATP concentration making ER Ca²⁺ signaling particularly sensitive to metabolic changes. Mitochondrial uptake of Ca²⁺ released from ryanodine-sensitive stores likely activates Ca²⁺-sensitive dehydrogenases to accelerate aerobic metabolism in DRG neurons as shown for hepatocytes (Hajnoczky et al. 1995). Thus the ER and mitochondria communicate via Ca²⁺ and energetic signals.

The plasma membrane Ca²⁺ pumps were not dependent on ATP derived from oxidative phosphorylation, consistent with a similar finding reported for peripheral nerve terminals (Gover et al. 2004). Glycolytic activity preferentially drives a number of plasma membrane ion transport systems, including the K_ATP channel (Weiss and Lamp 1987, 1989), the Na⁺/K⁺ pump, (Glitsch and Tappe 1993), and the Na⁺/H⁺ exchanger (Wu and Vaughan-Jones 1994). The coupling of Ca²⁺ transport systems to different metabolic pathways is likely to be broadly important to neuronal signaling and survival.

Mitochondrial regulation of CICR raises the possibility that pathological changes in cellular energy metabolism, such as those that accompany diabetic neuropathy, might have profound effects on ER Ca²⁺ signaling (Kruglikov et al. 2004). Perhaps the reliance of SERCA-mediated Ca²⁺ uptake on aerobic metabolism contributes to the sensitivity of the CNS to oxygen deprivation (Erecinska and Silver 2001). Ca²⁺ levels within the ER lumen influence protein processing and trigger certain forms of apoptosis (Hajnoczky et al. 2003; Mengesdorf et al. 2001; Paschen 2004).

	CONCLUSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
CICR oscillations illustrate the complexity and interdependence of the Ca²⁺ regulatory mechanisms operating in an intact neuron. Competition between various reuptake and clearance mechanisms allows neurons to shape Ca²⁺ signals, altering their spatial and temporal attributes. The interplay between mitochondria and ER Ca²⁺ release may influence neuronal energy metabolism and Ca²⁺-triggered changes in excitability, secretion, transcription, and apoptosis.

	GRANTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
The National Science Foundation (IOB0110409) and the National Institute of Drug Abuse (DA-07304 and DA-11806) supported this work. J. Jackson is supported by National Institute of Drug Abuse Research Service Award T32 DA-07097.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
We thank W. Lin, T. Johanns, A. Marsh, and S. Derrington for excellent technical assistance.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Address for reprint requests and other correspondence: S. A. Thayer, Dept. of Pharmacology, University of Minnesota Medical School, 6-120 Jackson Hall, 321 Church St., Minneapolis, MN (E-mail: sathayer{at}umn.edu)

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION CONCLUSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Baron KT and Thayer SA. CGP37157 modulates mitochondrial Ca²⁺ homeostasis in cultured rat dorsal root ganglion neurons. Eur J Pharmacol 340: 295–300, 1997.[CrossRef][ISI][Medline]

Benham CD, Evans ML, and McBain CJ. Ca²⁺ efflux mechanisms following depolarization evoked calcium transients in cultured rat sensory neurones. J Physiol 455: 567–583, 1992.[Abstract/Free Full Text]

Berridge MJ. Neuronal calcium signaling. Neuron 21: 13–26, 1998.[CrossRef][ISI][Medline]

Bootman MD, Lipp P, and Berridge MJ. The organisation and functions of local Ca(2+) signals. J Cell Sci 114: 2213–2222, 2001.[ISI][Medline]

Budd SL and Nicholls DG. Mitochondria in the life and death of neurons. Essays Biochem 33: 43–52, 1998.[ISI][Medline]

Cox DA, Conforti L, Sperelakis N, and Matlib MA. Selectivity of inhibition of Na⁺-Ca²⁺ exchange of heart mitochondria by benzothiazepine CGP-37157. J Cardiovasc Pharmacol 21: 595–599, 1993.[ISI][Medline]

Cox DA and Matlib MA. Modulation of intramitochondrial free Ca²⁺ concentration by antagonists of Na⁺-Ca²⁺ exchange. Trends Pharmacol Sci 14: 408–413, 1993.[CrossRef][Medline]

Csordas G, Thomas AP, and Hajnoczky G. Quasi-synaptic calcium signal transmission between endoplasmic reticulum and mitochondria. EMBO J 18: 96–108, 1999.[CrossRef][ISI][Medline]

Currie KPM, Swann K, Galione A, and Scott RH. Activation of Ca²⁺-dependent currents in cultured rat dorsal root ganglion neurons by a sperm factor and cyclic ADP-ribose. Mol Biol Cell 3: 1415–1425, 1992.[Abstract]

Duke AM and Steele DS. Effects of caffeine and adenine nucleotides on Ca²⁺ release by the sarcoplasmic reticulum in saponin-permeabilized frog skeletal muscle fibers. J Physiol 513: 43–53, 1998.[Abstract/Free Full Text]

Emptage NJ, Reid CA, and Fine A. Calcium stores in hippocampal synaptic boutons mediate short-term plasticity, store-operated Ca²⁺ entry, and spontaneous transmitter release. Neuron 29: 197–208, 2001.[CrossRef][ISI][Medline]

Endo M, Tanaka M, and Ogawa Y. Calcium induced release of calcium from the sarcoplasmic reticulum of skinned skeletal muscle fibers. Nature 228: 34–36, 1970.[CrossRef][Medline]

Erecinska M and Silver IA. Tissue oxygen tension and brain sensitivity to hypoxia. Respir Physiol 128: 263–276, 2001.[CrossRef][ISI][Medline]

Fabiato A and Fabiato F. Dependence of the contractile activation of skinned cardiac cells on the sarcomere length. Nature 256: 54–56, 1975.[CrossRef][Medline]

Friel DD. Calcium oscillations in neurons. In: Calcium Waves, Gradients and Oscillations, New York: Wiley, 1995, p. 210–234.

Friel DD and Tsien RW. A caffeine- and ryanodine-sensitive Ca²⁺ store in bullfrog sympathetic neurons modulates effects of Ca²⁺ entry on [Ca²⁺]. J Physiol 450: 217–246, 1992.[Abstract/Free Full Text]

Furuichi T, Furutama D, Hakamata Y, Nakai J, Takeshima H, and Mikoshiba K. Multiple types of ryanodine receptor/Ca²⁺ release channels are differentially expressed in rabbit brain. J Neurosci 14: 4794–4805, 1994.[Abstract]

Glitsch HG and Tappe A. The Na⁺/K⁺ pump of cardiac Purkinje cells is preferentially fuelled by glycolytic ATP production. Pfluegers 422: 380–385, 1993.

Gomez TM, Snow DM, and Letourneau PC. Characterization of spontaneous calcium transients in nerve growth cones and their effect on growth cone migration. Neuron 14: 1233–1246, 1995.[CrossRef][ISI][Medline]

Gover TD, Kao JPY, and Weinreich D. Calcium regulation in nociceptive peripheral nerve terminals. Soc Neurosci Abstr 209: 9, 2004.

Grynkiewicz G, Poenie M, and Tsien RY. A new generation of Ca²⁺ indicators with greatly improved fluorescence properties. J Biol Chem 260: 3440–3450, 1985.[Abstract/Free Full Text]

Hajnoczky G, Davies E, and Madesh M. Calcium signaling and apoptosis. Biochem Biophys Res Commun 304: 445–454, 2003.[CrossRef][ISI][Medline]

Hajnoczky G, Robb GL, Seitz MB, and Thomas AP. Decoding of cytosolic calcium oscillations in the mitochondria. Cell 82: 415–424, 1995.[CrossRef][ISI][Medline]

Hamill OP, Marty A, Neher E, Sakmann B, and Sigworth F. Improved patch-clamp techniques for high-resoluti