Mitochondrial Role in Action Potential Generation: Effects of Membrane Potential Dissipati, Papers of Theatre

Download Mitochondrial Role in Action Potential Generation: Effects of Membrane Potential Dissipati and more Papers Theatre in PDF only on Docsity! T C M N K A D l A a d s 8 g d t t ( s A r m o m t d A m g p r t c o t a t e n t d I A r * 8 E A e 5 v C d c m w Neuroscience 124 (2004) 327–339 0 d HE MODULATION OF ACTION POTENTIAL GENERATION BY ALCIUM-INDUCED CALCIUM RELEASE IS ENHANCED BY ITOCHONDRIAL INHIBITORS IN MUDPUPPY PARASYMPATHETIC EURONSK c b C m d C K 1 m i w a l s 2 d g s s C ( i T v p v r c E ( o d a a n c n i m r H a a . L. BARSTOW, S. A. LOCKNAR, L. A. MERRIAM ND R. L. PARSONS* epartment of Anatomy and Neurobiology, University of Vermont Col- ege of Medicine, 89 Beaumont Avenue, Burlington, VT 05405, USA bstract—Previously, we demonstrated that outward currents ctivated by calcium-induced calcium release (CICR) opposed epolarization-induced action potential (AP) generation in dis- ociated mudpuppy parasympathetic neurons [J Neurophysiol 8 (2002) 1119]. In the present study, we tested whether AP eneration by depolarizing current ramps could be altered by issipating the mitochondrial membrane potential and thus in- errupting mitochondrial Ca2
buffering. Exposure to the pro- onophore carbonyl cyanide m-chlorophenylhydrazone CCCP; 2 M) alone or in combination with the mitochondrial ATP ynthase inhibitor oligomycin (8 g/ml), increased the latency to P generation. Exposure to the electron transport chain inhibitor otenone (10 M) alone or in combination with oligomycin (8 g/ l) similarly increased the latency to AP generation. CCCP and ligomycin or rotenone and oligomycin treatment caused rhoda- ine 123 loss from mitochondria within a few minutes, confirming hat the mitochondrial membrane potential was dissipated during rug exposure. Oligomycin alone had no effect on the latency to P generation and did not cause loss of rhodamine 123 from itochondria. The increase in latency induced by CCCP and oli- omycin was similar when recordings were made with either the erforated patch or standard whole cell patch recording configu- ation. Exposure to the endoplasmic reticulum Ca-ATPase inhibi- or thapsigargin (1 M), decreased the latency to AP generation. In ells pretreated with thapsigargin to eliminate CICR, CCCP and ligomycin had no effect on AP latency. Pretreatment with iberio- oxin (IBX; 100 nM), an inhibitor of large conductance, calcium- nd voltage-activated potassium channels, reduced the extent of he CCCP- and oligomycin-induced increase in latency to AP gen- ration. These results indicate that treatment with CCCP or rote- one to dissipate the mitochondrial membrane potential, a condi- ion which should minimize sequestration of Ca2
by mitochon- ria, facilitated the Ca2
-induced Ca2
release activation of BX-sensitive and IBX-insensitive conductances that regulate P generation. © 2004 IBRO. Published by Elsevier Ltd. All ights reserved. Corresponding author. Tel:
1-802-656-2230; fax:
1-802-656- 704. -mail address: rodney.parsons@uvm.edu (R. L. Parsons). bbreviations: AP, action potential; BAPTA, 1,2-bis(2-aminophenoxy) thane-N,N,N,N-tetraacetic acid; BCECF, 2,7-bis(2-carboxyethyl)- -(and-6-)-carboxyfluorescein; BK, large conductance, calcium- and oltage-activated potassium; [Ca2
]i, intracellular Ca 2
concentration; CCP, carbonyl cyanide m-chlorophenylhydrazone; CICR, Ca2
-in- uced Ca2
release; DMSO, dimethylsulfoxide; EGTA, ethylene gly- ol-bis(-aminoethyl ether)N,N,N,N-tetraacetic acid; ER, endoplas- ic reticulum; IBX, iberiotoxin; SMOCs, spontaneous miniature out- 2
Eard currents; VDCC, voltage-dependent Ca channels. 306-4522/04$30.00
0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reser oi:10.1016/j.neuroscience.2003.12.010 327ey words: protonophore, rotenone, carbonyl cyanide m- hlorophenylhydrazone (CCCP), autonomic neurons, mem- rane excitability. alcium-induced Ca2
release (CICR) is a well-docu- ented mechanism by which Ca2
influx through voltage- ependent Ca2
channels (VDCC) can initiate release of a2
from internal stores (Marrion and Adams, 1992; uba, 1994; Verkhatsky and Shmigol, 1996; Berridge, 998). In mudpuppy parasympathetic neurons a CICR echanism stimulates spontaneous miniature hyperpolar- zations (in current clamp) and spontaneous miniature out- ard currents (SMOCs; in voltage clamp) which are initi- ted by the simultaneous activation of approximately 20 arge conductance, calcium- and voltage-activated potas- ium (BK) channels (Merriam et al., 1999; Scornik et al., 001). More recently, we showed that CICR-activated con- uctances regulate the latency to action potential (AP) eneration (Parsons et al., 2002). Our earlier studies howed that this modulation of AP latency was reduced ignificantly or eliminated by conditions that: (a) reduced a2
influx through VDCCs, (b) blocked BK channels or c) depleted the caffeine-sensitive intracellular Ca2
stores n the endoplasmic reticulum (ER; Parsons et al., 2002). hus, in the proposed model, activation of only a few oltage-gated calcium channels, during subthreshold de- olarizations, allows sufficient Ca2
influx to locally acti- ate ryanodine receptors and initiate CICR. The Ca2
eleased from internal stores raises the intracellular Ca2
oncentration ([Ca2
]i) in restricted domains between the R and the plasma membrane to levels high enough 40 M; Scornik et al., 2001) to activate plasma membrane utward currents that decrease the effectiveness of the epolarizing current (Parsons et al., 2002). Thus, CICR cts as an amplification mechanism for Ca2
, allowing ctivation of Ca2
-dependent outward currents that are ot activated directly by Ca2
influx alone. Cytoplasmic Ca2
levels are tightly regulated by intra- ellular buffers and active and passive transport mecha- isms that extrude Ca2
from cells or accumulate Ca2
in ntracellular stores. Mitochondria, in addition to being the ajor ATP producing organelle in neurons, also play a key ole in the homeostasis of intracellular Ca2
(Babcock and ille, 1998; Rizzuto et al., 1999; Duchen, 2000; Gunter et l., 2000; Nicholls and Budd, 2000; Camello-Almaraz et l., 2002). Mitochondria can be closely associated with the R and plasma membrane; thus mitochondria are critically ved. p d m a g i a c f h d c c d m p c m r m s t a G t o l A r A w g k U t G m T n n U s ( 1 5 b E V s ( u 1 C a r 8 K a i M 1 1 ( n t t t d t m c I p A a d m a t b p w  O M d t m w a s N s ( o t t z a s s b U a w r c c i m O fl m r a C t s e u K. L. Barstow et al. / Neuroscience 124 (2004) 327–339328ositioned to buffer transient rises in Ca2
within local omains (Landolfi et al., 1998; Rizzuto et al., 1998). Consequently, we hypothesized that mitochondria ight modulate the CICR-induced regulation of AP gener- tion. Mitochondria might be important buffering or- anelles that regulate the extent of the CICR-induced rise n Ca2
within local domains near the plasma membrane nd as a consequence, regulate the ability of depolarizing urrents to move the membrane potential to the threshold or AP generation. In the present study, we have tested this ypothesis by measuring the change in AP latency pro- uced by pharmacologically inhibiting the ability of mito- hondria to sequester Ca2
. We utilized two drugs that ause dissipation of mitochondrial membrane potential by ifferent mechanisms. Our results demonstrate that treat- ent with the protonophore carbonyl cyanide m-chloro- henylhydrazone (CCCP; 2 M) or the electron transport hain inhibitor rotenone (10 M), in combination with the itochondrial ATP synthase inhibitor oligomycin (8 g/ml), educes the mitochondrial membrane potential. Both treat- ents cause the mitochondrial membrane potential to dis- ipate and as a consequence, mitochondrial Ca2
seques- ration should be diminished (Friel and Tsien, 1994; Budd nd Nicholls, 1996; Nicholls and Budd, 2000; Medler and leason, 2002). During both drug treatments, the latency o AP generation was significantly lengthened. Elimination f CICR, by a thapsigargin-induced depletion of intracellu- ar Ca2
stores, blocked the mitochondrial modulation of P latency; thus, the effect on latency to AP generation equired CICR to be intact. EXPERIMENTAL PROCEDURES ll experiments were performed on parasympathetic neurons that ere dissociated from mudpuppy (Necturus maculosus) cardiac anglia and maintained in culture for 12–36 h. Mudpuppies were illed by rapid decapitation, following procedures approved by the niversity of Vermont Institutional Animal Care and Use Commit- ee and methods described in the National Institutes of Health uide for the Care and use of Laboratory Animals. All efforts were ade to minimize the number of animals used and their suffering. he method of cell dissociation used a combination of collage- ase, type I (Sigma Chemical Co., St. Louis, MO, USA) and eutral protease (Roche Molecular Biochemicals, Indianapolis, IN, SA) following methods described previously (Merriam and Par- ons, 1995). All experiments were done at room temperature 21–22 °C) using mudpuppy physiological solution that contained: 10 mM NaCl, 3.6 mM CaCl2, 2.5 mM KCl, 10 mM NaHEPES, mM glucose, pH 7.3. Tetrodotoxin (0.3 M) was included in the ath solution in some optical experiments. lectrophysiological methods oltage recordings were made using either the perforated patch or tandard configuration of the whole cell patch clamp technique Hamill et al., 1981; Horn and Marty, 1988) and were controlled sing the current clamp bridge mode of an Axoclamp 2A/Digidata 200/pClamp 6.0.3 acquisition system (Axon Instruments, Union ity, CA, USA). Voltage responses were digitized at 1 kHz and cquired on line. For most experiments, the perforated patch ecording mode was used with pipette solution composition: 0 mM K aspartate, 40 mM KCl, 5 mM MgCl2, 10 mM HEPES– OH, pH 7.2 and the patch pipettes were backfilled with 0.2 mg/ml mphotericin B (Sigma). For standard whole cell recordings the bnternal solution contained: 80 mM Kaspartate, 40 mM KCl, 2 mM gCl2, 0.4 mM NaGTP, 3 mM MgATP, 3 mM phosphocreatine, 0 mM HEPES–KOH, pH 7.2 and either 5 mM EGTA or 5 mM ,2-bis(2-aminophenoxy) ethane-N,N,N,N-tetraacetic acid BAPTA). The average resting membrane potential of the dissociated eurons is approximately 50 mV (Scornik et al., 2001). Although he resting membrane potential varied between cells, control and est whole cell recordings were obtained from the same cell with he membrane potential set between 55 and 60 mV. Following rug treatments, there was no consistent or sustained change in he resting membrane potential. Occasionally, there was a 5–10 V depolarization, which lasted 1 min, following a switch from ontrol solution to the solution containing CCCP and oligomycin. n these cells, experiments were started after the membrane otential returned to the initial level. Current ramps (400 ms) were applied to determine the latency to P generation. The rate of depolarization with the current ramp was djusted under control conditions to make it possible to measure a ecrease or increase in latency to the first AP. The latency, deter- ined as the time interval from onset of the current ramp to the point t which the rising phase of the AP crossed 0 mV, was compared in he same cell prior to and during drug application. Cell input resistance was monitored from the change in mem- rane potential produced by 500 ms, 5 pA hyperpolarizing current ulses. The control input resistance was 2.40.2 G in 41 cells, hen measured with the membrane potential maintained between 55 and 60 mV. ptical methods itochondrial accumulation of rhodamine 123 was used to assess issipation of the mitochondrial membrane potential following drug reatment (Johnson et al., 1981). Control cells were treated for 10 in with 5 M rhodamine 123 in control mudpuppy solution, then ashed three times with control solution without rhodamine 123 nd imaged after 5 min with the DeltaVision Restoration Micro- cope (Applied Precision, Issaquah, WA, USA) using a 60 (1.4 A) oil-immersion lens and FITC filters for excitation and emis- ion. Some cells were pretreated with a solution containing CCCP 2 M), oligomycin (8 g/ml), or a combination of CCCP and ligomycin for 10 min before the addition of rhodamine 123 solu- ion and rhodamine wash-out. Therefore, the total time of drug reatment for these cells was 25 min before imaging. A stack of -slices 0.2 m apart was acquired and the entire stack (reported s F/Fo) deconvolved using softWoRx software (Applied Preci- ion). For Fig. 2A–D, a projection of five slices near the cell urface was generated (total thickness of 1 m) and adjusted for rightness and contrast with Adobe Photoshop 4.0.1, San Jose, SA. In other cells (results shown in Fig. 2E), a time-series was cquired after rhodamine 123 pretreatment. A single optical slice as acquired at a rate of 0.1 Hz during CCCP and oligomycin or otenone and oligomycin bolus application. Mitochondrial fluores- ence was determined from pixel intensity within identified mito- hondria and was corrected for any decrease in fluorescence ntensity due to bleaching and/or extrusion of rhodamine 123 from itochondria with a single exponential decay algorithm (Microcal rigin 7.0; Microcal, Northampton, MA, USA), to give F/Fo. Changes in [Ca2
]i were assessed from variations in fluo-3 uorescence intensity. Cells were incubated for 15 min in control udpuppy solution containing 5 M fluo-3-AM and 0.02% Plu- onic F-127. They were rinsed three times with control solution nd the AM ester was allowed to cleave for at least 45 min. onfocal images were obtained with a Noran Oz system (Middle- on, WI, USA) attached to a Nikon Diaphot 200 inverted micro- cope. A 60 (1.2 NA) water immersion lens was used with an xcitation wavelength of 488 nm. The 500 nm long-pass filter was sed for data acquisition. Images were acquired at 0.1 Hz with a ath flow of approximately1 ml/min. Runs were a minimum of 20 t m r e l H o r d p m c d C u t t w p T l i T f l p o r i t d m t p w t t p a F g s ( s h r F w r t d a r ( K. L. Barstow et al. / Neuroscience 124 (2004) 327–339 331In a second series of experiments, we determined the ime course of loss of rhodamine 123 fluorescence from itochondria during exposure to CCCP and oligomycin or otenone and oligomycin. As shown in Fig. 4E, during xposure to either drug treatment, there was a progressive oss of rhodamine 123 fluorescence from mitochondria. owever, in CCCP and oligomycin, the mitochondrial flu- rescence reached a minimum within 2 min whereas in otenone and oligomycin the fluorescence took longer to issipate. These experiments indicated that with bath ap- lication of either the CCCP or rotenone, the mitochondrial embrane potential was dissipated, an observation indi- ating that ability to sequester Ca2
should be greatly iminished. Given that the change occurred more rapidly in CCP and oligomycin than in rotenone and oligomycin, we sed CCCP and oligomycin treatment in the following elec- rophysiological studies. The results also demonstrated hat a 5 min equilibration period in CCCP and oligomycin as adequate to dissipate the mitochondrial membrane 0 50 100 150 200 250 300 La te nc y (m s) * Con C+O Wash A 0 2 4 6 8 10 (m V ) Con C+O Wash * B ig. 2. The CCCP- and oligomycin-induced increase in latency to AP eneration was reversible. Results summarized in panel (A) demon- trate in four cells that the increase in latency by CCCP and oligomycin C
O) completely recovered after 10 min of wash with drug-free olution. In contrast, in these same four cells, the amplitude of the yperpolarization produced by a 500 ms, 5 pA current pulse did not ecover fully after 10 min of wash with drug-free solution.otential. wreatment with CCCP and oligomycin increases the atency to AP generation in the presence of ntracellular EGTA or BAPTA he initial latency measurements were made with the per- orated patch recording configuration in which the intracel- ular environment remains intact and is not dialyzed by ipette solution. We next determined whether CCCP and ligomycin could affect the latency to AP generation when ecordings were made with the standard whole cell record- ng configuration in which the internal environment is con- rolled by dialyzing with Ca2
buffers and intracellular me- iators including ATP. Two series of current ramp experi- ents were done with the standard whole cell recording echnique. In the first, EGTA (5 mM) was included in the ipette solution whereas in the second, BAPTA (5 mM) as present. The results summarized in Fig. 5 indicate hat, during exposure to CCCP and oligomycin, the latency o AP generation was increased as observed with the erforated patch recordings and furthermore, the CCCP- nd oligomycin-induced increase in latency was similar A 0 50 100 150 200 250 300 350 La te nc y (m s) Con Rot Con R+O * *B Control Rotenone 0 mV ig. 3. The latency to AP generation is increased following treatment ith rotenone (Rot) alone or Rot and oligomycin. Panel (A) shows a ecording from an individual cell prior to and during Rot (10 M) reatment. In this example, the latency to AP generation by a 400 ms epolarizing current ramp was determined before (Con) and following 5 min Rot treatment. Panel (B) presents bar graphs summarizing esults from multiple cells prior to and during exposure to 10 M Rot n 3 cells) or 10 M Rot and 8 g/ml oligomycin (R
O; n 6 cells).hether EGTA (509%, n 4 cells; Fig. 5A) or BAPTA ( t t F p r R fl a o K. L. Barstow et al. / Neuroscience 124 (2004) 327–3393324812%, n 5 cells; Fig. 5B) was used to buffer Ca2
in A C oligomycin E -1 0 1 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 F/ F o ti CCCP + olig ig. 4. Changes in rhodamine 123 fluorescence demonstrated drug attern of rhodamine 123 fluorescence in four different cells: (A) Con espectively, as described in Experimental Procedures. One microm estoration Microscope and processed as described in Experimental P uorescence during CCCP and oligomycin or rotenone and oligomycin s described in the Experimental Procedures. Data presented as mean ligomycin. Drugs were added at time 0.he pipette solution. In these experiments, the input resis- c ance decreased during exposure to CCCP and oligomy- CCCP + oligomycin CCCP 3 4 5 6 7 inutes) rotenone + oligomycin dissipation of the mitochondrial membrane potential. In (A–D), the ), cells pretreated with CCCP and oligomycin, oligomycin, or CCCP, k projection of 5 slices 0.2 m thick. Acquired with the Deltavision s. Calibration bar 10 m. (E) Change in mitochondrial rhodamine 123 nt. Fluorescence was normalized to a first order bleaching curve (Fo) for five cells in CCCP and oligomycin and seven cells in rotenone andB D 2 me (m omycin -induced trol; (B–D eter thic rocedure treatme S.E.M.in. However, the decrease was significantly less when B ( T a W c f u a i d 1 e i i t m e t t w p d d a t e t b t d o P e l P B g m s i l 5 t o d d p c C a s w F c l K. L. Barstow et al. / Neuroscience 124 (2004) 327–339 333APTA (335%, n 5 cells) was present than when EGTA 614%, n 4 cells) was present. he increase in latency to AP generation by CCCP nd oligomycin treatment requires CICR e next determined whether the CCCP- and oligomy- in-induced increase in latency required that CICR be unctional. In these and all subsequent experiments, we sed the perforated patch recording configuration to void dialyzing the cell interior. Thapsigargin treatment nhibits the ER Ca2
-ATPase resulting in a progressive epletion of internal Ca2
stores (Thomas and Hanley, 994). Addition of 10 mM caffeine along with thapsigargin mpties the stores more rapidly so caffeine was always ncluded in the thapsigargin pretreatment in order to elim- nate CICR (Parsons et al., 2002). Similar to the observa- ions in our previous paper (Parsons et al., 2002), treat- ent with thapsigargin decreased the latency to AP gen- ration (335%; n 4 cells; Fig. 6A). Thapsigargin reatment did not significantly change the cell input resis- ance (22%, n 4 cells). In subsequent experiments, we pretreated neurons ith 1 M thapsigargin for at least 10 min (with caffeine resent for the first 3– 4 min) to eliminate CICR and then etermined the latency in the same cells prior to and uring exposure to the CCCP and oligomycin solution lso containing thapsigargin. In six cells pretreated with hapsigargin, CCCP and oligomycin exposure had no 0 mV C C A B C C 0 mV EGTA BAPTA ig. 5. Treatment with CCCP and oligomycin increases the latenc onfiguration. With either EGTA (A) (n 4 cells) or BAPTA (B) (n 5 c atency to AP generation to the same extent.ffect on latency to AP generation (Fig. 6B). Even ghough the latency to AP generation was not increased y CCCP and oligomycin after CICR was eliminated in hese six cells, the cell input resistance was significantly ecreased by 607% during exposure to CCCP and ligomycin. retreatment with IBX reduces, but does not liminate the effect of CCCP and oligomycin on the atency to AP generation reviously, we demonstrated that exposure to the potent K channel blocker IBX, decreases the latency to AP eneration suggesting BK channels activated by CICR odulate excitability (Parsons et al., 2002). In the present tudy we tested whether pretreatment with IBX could elim- nate the CCCP- and oligomycin-induced increase in the atency to AP generation. Cells were pretreated for at least min with 100 nM IBX, a concentration previously shown o inhibit the SMOCs generated by synchronous activation f BK channels by CICR (Merriam et al., 1999). We then etermined the latency to AP generation in IBX prior to and uring exposure to CCCP and oligomycin. In six IBX- retreated cells, exposure to CCCP and oligomycin in- reased the latency to AP generation by 378%. Although CCP and oligomycin increased the latency to AP gener- tion in IBX-pretreated cells, the change in latency was ignificantly less than that noted with cells not pretreated ith IBX (564%, n 13). In contrast, the CCCP and oli- 0 50 100 150 200 250 300 La te nc y (m s) 1C + OControl * 1C + OControl 0 50 100 150 200 250 300 La te nc y (m s) * generation in recordings made with the standard whole cell patch e pipette solution, exposure to CCCP and oligomycin increased the + O ontrol + O ontrol y to AP ells) in thomycin-induced decrease in the input resistance in the D d r T w c s t p a C m m m a c r h t o d r n a r o c r t i s d t c l c t e p c t r h v f a A t t u e t w t C a w w l i c p r m 1 T fl M a ( B p r I i t e 2 ( fl 1 m C c r fl o o o o L o r m c t g s m l s m c s e p e 9 b g K. L. Barstow et al. / Neuroscience 124 (2004) 327–339336DISCUSSION issipation of the mitochondrial membrane potential ecreases the effectiveness of depolarizing current amps to generate APs he present study demonstrated that following treatment ith CCCP, CCCP and oligomycin, rotenone and oligomy- in or rotenone alone, the latency to AP generation was ignificantly increased. We attribute the change in latency o AP generation to a CCCP- or rotenone-induced dissi- ation of the mitochondrial membrane potential and result- nt inhibition of Ca2
sequestration, which enhanced ICR. A CCCP- and rotenone-induced dissipation of the itochondrial membrane potential was confirmed with easurements of rhodamine 123 fluorescence. Rhoda- ine 123 was concentrated in mitochondria in control cells s well as in cells treated only with oligomycin, but not in ells exposed to CCCP alone, CCCP and oligomycin or otenone and oligomycin. The membrane potential change produced by a small yperpolarizing current step commonly is used to estimate he cell input resistance. During exposure to CCCP and ligomycin, but not rotenone and oligomycin, there was a ecrease in the input resistance. This decrease in input esistance could have potentially decreased the effective- ess of the depolarizing current ramp. However, we do not ttribute the increase in latency to a decrease in cell input esistance. A consistent decrease in input resistance was nly observed in the presence of CCCP, whereas the hange in latency to AP generation was still observed with otenone treatment. Also, in many other experiments, here was no consistent correlation between the change in nput resistance determined from hyperpolarizing current teps and the change in latency to AP generation during epolarizing current ramps. For example, in cells pre- reated with thapsigargin, exposure to CCCP and oligomy- in decreased the input resistance without changing the atency to AP generation. The results summarized above indicate that the de- rease in input resistance was not a critical factor in de- ermining the latency. Two possibilities are suggested to xplain this observation. First, the decrease in the hyper- olarization from which input resistance was calculated ould represent a drug-induced modulation of a conduc- ance activated by hyperpolarization, such as the inward ectifier, that might have abbreviated the extent of the yperpolarization. Alternatively, the conductances acti- ated by the depolarizing current ramp near the threshold or AP generation were sufficiently great in magnitude that passive change in input resistance had minimal effect. n analysis of membrane conductances activated near the hreshold for AP generation required to answer this ques- ion is beyond the scope of the present study, but will be ndertaken in future studies. The latency to AP generation was not affected by xposure to oligomycin alone. Therefore, we concluded hat the CCCP-induced change in latency to AP generation as not due to an inhibition of mitochondrial ATP produc-ion. This conclusion is supported by the observation that tCCP and oligomycin increased the latency to AP gener- tion equally well when studied with either the standard hole cell recording configuration with the cell dialyzed ith ATP or the perforated patch configuration which al- ows the intracellular milieu to remain relatively intact. Also, n all recording modes, glucose was present in the extra- ellular solution so one might expect that the glycolytic athway should have maintained the cytosolic ATP/ADP atio for periods longer than the duration of the experi- ents (Kauppinen and Nicholls, 1986; Werth and Thayer, 994; White and Reynolds, 1995; Peng, 1998). reatment with CCCP, but not rotenone, increased uo-3 fluorescence itochondria not only produce metabolic energy, but also re key organelles contributing to the regulation of [Ca2
]i Babcock and Hille, 1998; Duchen,2000; Nicholls and udd, 2000). Dissipation of the mitochondrial membrane otential decreases the ability of mitochondria to buffer ises in [Ca2
]i (Duchen, 2000; Nicholls and Budd, 2000). n addition, Ca2
within the mitochondria can be released nto the cytoplasm during protonophore exposure and pro- onophore treatment has been reported to cause a small levation of [Ca2
]i at rest in some neurons (Nohmi et al., 000; Medler and Gleason 2002), but not in others Vanden Berghe et al., 2002). In the present study, the uo-3 fluorescence ratio F/Fo increased by approximately 0% during exposure to CCCP alone or CCCP and oligo- ycin, but not during treatment with oligomycin alone. The CCP- and oligomycin-induced increase in fluo-3 fluores- ence was maintained throughout a 10-min exposure and eversed following drug removal. As no change in fluo-3 uorescence occurred with oligomycin alone, the elevation f fluo-3 fluorescence was not thought to be due to an ligomycin-induced inhibition of ATP synthesis or a sec- ndary effect due to inhibition of other ATPases that can ccur with high concentrations of oligomycin (Fortes and ee, 1984). In addition, during exposure to rotenone and ligomycin, there was no consistent change in the fluo-3 atio. As both CCCP and rotenone should dissipate the embrane potential, both would be expected to cause a hange in the fluo-3 ratio if it reflected a rise in [Ca2
]i due he release of Ca2
from mitochondria. Based on the more radual decrease in rhodamine 123 fluorescence, we as- ume that rotenone takes longer to fully dissipate the itochondrial membrane potential than CCCP. This could ead to a slower release of Ca2
from mitochondria, but it hould still have been detectable with our Ca2
imaging ethods. Therefore, since only CCCP appeared to in- rease the fluo-3 ratio, we tentatively concluded that the ustained small increase in fluo-3 fluorescence ratio during xposure to CCCP might reflect an undefined effect of the rotonophore on dye fluorescence or binding affinity. Even if global [Ca2
]i actually was increased to a small xtent (perhaps by 10–15 nM assuming a [Ca2
]i of 60– 0 nM at rest) during CCCP exposure, it should not have een a factor determining the effect on latency to AP eneration. This conclusion is based on the observation hat the CCCP-induced increase in latency to AP genera- t p W d q i d B e p s e w p c r t e T g C i d c d a w d 1 f a C e r l C c W d c C u e I o 2 a m e r m A a 3 c o p c i m s i h a r C a t d g t a g t t m P s m R a p c a s r c t d e m d 2 t ( l c t 1 c o 1 V A i w i i o B w r K. L. Barstow et al. / Neuroscience 124 (2004) 327–339 337ion was similar when recordings were made with either the erforated patch or standard whole cell recording mode. ith the standard whole cell recording mode, a slowly eveloping small elevation in global [Ca2
]i, should be uickly buffered by the Ca2
chelator (EGTA or BAPTA) ncluded in the pipette solution. During exposure to protonophore, the cytosolic pH can ecrease (Nicholls and Budd, 2000). We found that the CECF fluorescence ratio changed during protonophore xposure, an observation suggesting that the intracellular H changed. However, the change in the F/Fo ratio was mall corresponding to very small shifts in pH units (Wong t al., 2001). In addition, the average change in F/Fo ratio as biphasic, initially an increase indicating an increase in H followed by a decrease consistent with a fall in pH. In ontrast, no consistent change in BCECF fluorescence atio was noted during exposure to rotenone. Therefore, he increase in latency for AP generation was not consid- red to be related to small shifts in intracellular pH. he mitochondrial modulation of the latency to AP eneration requires CICR ICR is an effective mechanism by which Ca2
influx can nitiate the release of Ca2
from ER stores into restricted omains near the plasma membrane, causing the local oncentration of Ca2
to rise sufficiently to activate Ca- ependent ion channels, including BK channels (Kuba et l., 1983; Kaczorowski et al., 1996; Scornik et al., 2001), hich generates the SMOCs recorded in mudpuppy car- iac neurons (Satin and Adams, 1987; Merriam et al., 999). The combined treatment of thapsigargin and caf- eine effectively depletes Ca2
stores in the ER (Thomas nd Hanley, 1994) and eliminates CICR. The effect of CCP and oligomycin on the latency to AP generation was liminated by thapsigargin, demonstrating that CICR was equired for the observed CCCP-induced increase in the atency to AP generation. ICR-activated IBX-sensitive and IBX-insensitive hannels contribute to the regulation of AP latency e previously showed that the BK channel inhibitor IBX ecreases the latency to AP generation, suggesting that urrents flowing through BK channels contribute to the ICR-activated outward current opposing the ramp current sed to initiate AP generation (Parsons et al., 2002). How- ver, in our prior study, the decrease in latency following BX treatment (165%) was half that following elimination f CICR by thapsigargin treatment (326%; Parsons et al., 002). SMOCs, which are generated by the synchronous ctivation of BK channels, are eliminated equally by treat- ent with either thapsigargin (Parsons et al., 2002) or by xposure to 100 nM IBX (Merriam et al., 1999). These prior esults suggest that activation of IBX-insensitive channels ust also contribute to the CICR modulation of latency to P generation. In the present study, we found that CCCP nd oligomycin increased the latency to AP generation by 78% in cells pretreated with 100 nM IBX to block BK hannels. This increase in AP latency was less than that btained for cells not pretreated with IBX (564%). The wresent observations provide additional support for the onclusion that both IBX-sensitive and IBX-insensitive onic conductances, activated by CICR, contribute to the odulation of the latency to AP generation. In our previous tudy, we also noted that treatment with apamin, an inhib- tor of small conductance Ca2
-activated K
channels, ad no effect on the latency to AP generation produced by depolarizing current ramp (Parsons et al., 2002). This esult suggests that apamin-sensitive small conductance a2
-activated K
channels were not involved (Parsons et l., 2002). In recent unpublished results, we found that reatment with 20 M glibencamide to inhibit ATP-depen- ent K
channels had no affect on the latency to AP eneration (Barstow and Parsons, unpublished observa- ions). Thus, it appears that activation of KATP channels lso does not contribute to the regulation of latency to AP eneration. Additional experiments are under way to iden- ify the IBX-insensitive ionic conductance(s) contributing to he CICR-mediated regulation of AP generation in these udpuppy neurons. roposed mechanism by which mitochondrial equestration of Ca2
might affect the CICR- odulation of the latency to AP generation hodamine 123 staining demonstrated that mitochondria re densely distributed throughout the cytoplasmic com- artment; thus they are positioned to potentially affect hanges in Ca2
levels by buffering Ca2
entering VDCCs nd/or Ca2
domains near ryanodine-sensitive release ites on the ER. Within local domains near VDCCs or yanodine-sensitive release sites, the Ca2
can reach mi- romolar concentrations and mitochondria could regulate he amplitude, duration and spread of Ca2
within these omains (Rizzuto et al., 1998), thereby modulating CICR ffectiveness (Babcock et al., 1997). Previously, we esti- ated that the very brief, local rise in Ca2
that occurs uring CICR must reach at least 40 M (Scornik et al., 001). This is well above the 200–500 nM thought to be he threshold for Ca2
sequestration into mitochondria Friel and Tsien, 1994; Werth and Thayer, 1994; Co- egrove et al., 2000). EGTA and BAPTA have comparable Ca2
buffering apacity although BAPTA can buffer fast transient eleva- ions of Ca2
much more efficiently than EGTA (Naraghi, 997). Therefore, results with EGTA or BAPTA often are ompared to distinguish Ca2
transients that are very fast r occur in local domains (Deisseroth et al., 1996; Neher, 998). In mudpuppy cardiac neurons Ca2
influx through DCCs directly activates BK channels that contribute to P repolarization (Parsons et al., 2002). In ongoing exper- ments we have found that the rate of AP repolarization as consistently slower when 5 mM BAPTA was included n the pipette solution than when 5 mM EGTA was included n the pipette solution (Barstow and Parsons, unpublished bservations). This observation suggests that 5 mM APTA, but not 5 mM EGTA, can buffer the rise in [Ca2
]i hich directly activates BK channels that participate in AP epolarization. In contrast, the latency to AP generation as increased by CCCP and oligomycin to the same ex- t p u p f m f b t t b w s a C w A g R s o J B B B B C C D D F F G H H J K K K K L M M M M N N N N P P P R R S S T V K. L. Barstow et al. / Neuroscience 124 (2004) 327–339338ent whether the recordings were made with the perforated atch configuration or with the standard whole cell config- ration with either 5 mM EGTA or 5 mM BAPTA in the ipette solution. Thus, the elevation of Ca2
due to release rom internal stores by CICR must occur in discrete do- ains where BAPTA at a concentration of 5 mM is inef- ective. This observation suggests that mitochondria may e positioned such that they buffer the CICR-activated ransient rise in Ca2
that occurs in local domains between he ER Ca2
release sites and opposing plasma mem- rane. Interruption of mitochondrial sequestration of Ca2
ould increase the Ca2
concentration within the re- tricted domain between the plasma membrane and ER, nd as a consequence, could increase the efficiency of ICR, which in turn increases the magnitude of the out- ard currents that oppose the depolarizing current ramps. cknowledgements—This work was supported in part by NSF rant IBN-0076741 and NIH grant HL-65481. The DeltaVision estoration microscope is provided through the Imaging Core upported by NIH grant P20 RR-16435 from the COBRE program f the National Center for Research Resources. We thank Dr. ohn Tompkins for reviewing this manuscript. REFERENCES abcock DF, Herrington J, Goodwin PC, Park YB, Hille B (1997) Mitochondrial participation in the intracellular Ca2
network. J Cell Biol 136:833–844. abcock DF, Hille B (1998) Mitochondrial oversight of cellular Ca2
signaling. Curr Opin Neurobiol 8:398–404. erridge MJ (1998) Neuronal calcium signaling. Neuron 21:13–26. udd SL, Nicholls DG (1996) A reevaluation of the role of mitochondria in neuronal Ca2
homeostasis. J Neurochem 66:403–411. amello-Almaraz C, Salido GM, Pariente JA, Camello PJ (2002) Role of mitochondria in Ca2
oscillations and shape of Ca2
signals in pancreatic acinar cells. Biochem Pharmacol 63:283–292. olegrove SL, Albrecht MA, Friel DD (2000) Dissection of mitochon- drial Ca2
uptake and release fluxes in situ after depolarization- evoked [Ca2
]i elevations in sympathetic neurons. J Gen Physiol 115:351–369. eisseroth K, Bito H, Tsien RW (1996) Signaling from synapse to nucleus: postsynaptic CREB phosphorylation during multiple forms of hippocampal synaptic plasticity. Neuron 16:89–101. uchen MR (2000) Mitochondria and calcium: from cell signalling to cell death. J Physiol 529:57–68. ortes PA, Lee JA (1984) Steady-state levels of phosphorylated inter- mediates of (Na, K)-ATPase monitored with oligomycin and an- throyloubain. J Biol Chem 259:11176–11179. riel DD, Tsien RW (1994) A FCCP-sensitive Ca2
store in bullfrog sympathetic neurons and its participation in stimulus-evoked changes in [Ca2
]i. J Neurosci 14:4007–4024. unter TE, Buntinas L, Sparagna G, Eliseev R, Gunter K (2000) Mitochondrial calcium transport: mechanisms and functions. Cell Calcium 28:285–296. amill OP, Marty A, Neher E, Sakmann B, Sigworth FJ (1981) Im- proved patch clamp techniques for high-resolution current record- ing from cells and cell-free membrane patches. Pflugers Arch 391: 85–100. orn R, Marty A (1988) Muscarinic activation of ionic currents mea- sured by a new whole-cell recording method. J Gen Physiol 94: 145–159. ohnson LV, Walsh ML, Bockus BJ, Chen LB (1981) Monitoring of relative mitochondrial membrane potential in living cells by fluores- cence microscopy. J Cell Biol 88:526–535.aczorowski GJ, Knaus HG, Leonard RJ, McManus OB, Garcia ML(1996) High conductance calcium-activated potassium channels: structure, pharmacology, and function. J Bioenerg Biomembr 28: 255–267. auppinen RA, Nicholls DG (1986) Failure to maintain glycolysis in anoxic nerve terminals. J Neurochem 47:1864–1869. uba K (1994) Ca2
-induced Ca2
release in neurones. Jpn J Physiol 44:613–650. uba K, Morita K, Nohmi M (1983) Origin of calcium ions involved in the generation of a slow afterhyperpolarization in bullfrog sympa- thetic neurones. Pflugers Arch 399:194–202. andolfi B, Curci S, Debellis L, Pozzan T, Hofer AM (1998) Ca2
homeostasis in the agonist-sensitive internal store: functional inter- actions between mitochondria and the ER measured in situ in intact cells. J Biol Chem 142:1235–1243. arrion NV, Adams PR (1992) Release of intracellular calcium and modulation of membrane currents by caffeine in bull-frog sympa- thetic neurones. J Physiol 445:515–535. edler K, Gleason EL (2002) Mitochondrial Ca2
buffering regulates synaptic transmission between retinal amacrine cells. J Neuro- physiol 87:1426–1439. erriam LA, Parsons RL (1995) Neuropeptide galanin inhibits cono- toxin GVIA-sensitive calcium channels in parasympathetic neurons. J Neurophysiol 73:1374–1382. erriam LA, Scornik FS, Parsons RL (1999) Ca2
-induced Ca2
release activates spontaneous miniature outward currents (SMOCs) in parasympathetic cardiac neurons. J Neurophysiol 82: 540–550. araghi M (1997) T-jump study of calcium binding kinetics of calcium chelators. Cell Calcium 22:255–268. eher E (1998) Vesicle pools and Ca2
microdomains: new tools for understanding their roles in neurotransmitter release. Neuron 20: 389–399. icholls DG, Budd SL (2000) Mitochondrial and neuronal survival. Physiol Rev 80:315–360. ohmi M, HuaS -Y, Liu C, Kuba K (2000) Ryanodine- and thapsigar- gin-insensitive Ca2
-induced Ca2
release is primed by lowering external Ca2
in rabbit autonomic neurons. Pflügers Arch Eur J Physiol 440:588–599. ark K-S, Jo I, Pak YK, Bae S-W, Rhim H, Suh S-H, Park SJ, Zhu MH, So I, Kim KW (2002) FCCP depolarizes plasma membrane poten- tial by activating proton and Na
currents in bovine aortic endothe- lial cells. Pflügers Arch Eur J Physiol 443:344–352. arsons RL, Barstow KL, Scornik FS (2002) Spontaneous miniature hyperpolarizations affect threshold for action potential generation in mudpuppy cardiac neurons. J Neurophysiol 88:1119–1127. eng YY (1998) Effects of mitochondrion on calcium transients at intact presynaptic terminals depend on frequency of firing. J Neu- rophysiol 80:695–708. izzuto R, Pinton P, Brini M, Chiesa A, Filippin L, Pozzan T (1999) Mitochondria as biosensors of calcium microdomains. Cell Calcium 26:193–199. izzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, Pozzan T (1998) Close contacts with the endoplasmic reticulum as determinants of mitochondrial Ca2
responses. Sci- ence 280:1763–1766. atin LS, Adams PR (1987) Spontaneous miniature outward currents in cultured bullfrog neurons. Brain Res 401:331–339. cornik FS, Merriam LA, Parsons RL (2001) Number of KCa channels underlying spontaneous miniature outward currents in mudpuppy cardiac neurons. J Neurophysiol 85:54–60. homas D, Hanley MR (1994) Pharmacological tools for perturbing intracellular calcium storage. In: A practical guide to the study of calcium in living cells (Nuccitielli R, ed), pp 65–89. New York: Academic Press. anden Berghe P, Kenyon JL, Smith TK (2002) Mitochondrial Ca2
uptake regulates the excitability of myenteric neurons. J Neurosci 22:6962–6971.