Markus HothSchwindling, Peter Lipp, Lars Kaestner and Ariel Quintana, Eva C. Schwarz, Christian  the Plasma MembraneRequires Translocation of Mitochondria toRelease-activated Calcium Channels Sustained Activity of CalciumMechanisms of Signal Transduction:

doi: 10.1074/jbc.M607896200 originally published online October 20, 20062006, 281:40302-40309.J. Biol. Chem. 

  10.1074/jbc.M607896200Access the most updated version of this article at doi:

  .JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2006/12/26/M607896200.DC1.html

  http://www.jbc.org/content/281/52/40302.full.html#ref-list-1

This article cites 36 references, 20 of which can be accessed free at

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

Sustained Activity of Calcium Release-activated CalciumChannels Requires Translocation of Mitochondria to thePlasma Membrane*□S

Received for publication, August 17, 2006, and in revised form, October 13, 2006 Published, JBC Papers in Press, October 20, 2006, DOI 10.1074/jbc.M607896200

Ariel Quintana‡1, Eva C. Schwarz‡, Christian Schwindling‡, Peter Lipp§, Lars Kaestner§, and Markus Hoth‡

From the ‡Department of Physiology, §Institute for Molecular Cell Biology, Saarland University, 66421 Homburg, Germany

A rise of the intracellular Ca2� concentration has multiple sig-naling functions. Sustained Ca2� influx across plasma membranethrough calcium release-activated calcium (CRAC) channels isrequired forT-cell development in the thymus, gene transcription,and proliferation and differentiation of naıve T-cells into armedeffectors cells. Intracellular Ca2� signals are shaped by mitochon-dria,which functionas ahighlydynamicCa2�buffer.However, theprecise role ofmitochondria for Ca2�-dependent T-cell activationis unknown. Here we have shown that mitochondria are translo-cated to theplasmamembraneasaconsequenceofCa2� influxandthat this directed movement is essential to sustain Ca2� influxthrough CRAC channels. The decreased distance between mito-chondria and the plasmamembrane enabledmitochondria to takeup large amounts of inflowing Ca2� at the plasma membrane,thereby preventing Ca2�-dependent inactivation of CRAC chan-nels and sustaining Ca2� signals. Inhibition of kinesin-dependentmitochondrial movement along microtubules abolished mito-chondrial translocation and reduced sustained Ca2� signals. Ourresults show how a directedmovement of mitochondria is used tocontrol important cellular functions such as Ca2�-dependentT-cell activation.

Ca2� entry throughcalciumrelease-activatedcalcium(CRAC)2channels, which are opened following depletion of Ca2� stores, isnecessary for T-cell activation (1–4). The sustained activity ofCRAC channels is needed for transcription of early genes (5), forT-cell development in the thymus (6), and for the control of anti-genic responsiveness and tolerance (7).Very recently, severalmol-ecules in this signal transduction pathway were identified. Theseinclude ORAI1 (also called CRACM1), which is a necessary com-

ponent forCRACchannelactivity inT-cells (8, 9), STIM1,which isprobably the long-soughtCa2� sensor that links store depletion toCRAC opening (10–12), and WAVE2, which is also involved inCRAC channel activation, probably through an interaction withthe cytoskeleton (13).Apart from the major role of mitochondria as cellular energy

sources and their role in cell death (14, 15), mitochondria havebeen identified as regulatory elements for CRAC channels (2, 16,17). Mitochondria can reduce Ca2�-dependent inactivation ofCRAC channels in submembranous Ca2� microdomains byquickly removing the inflowingCa2�, highlighting the importanceof organelle localization within a cell. Mitochondrial localizationand movements are determined by microtubules (18–21) andmicrofilaments (20, 22). Recently, Ca2� signaling and mitochon-drialmovementshavebeen linked throughanunknownCa2� sen-sor molecule that translates Ca2� signals into the microtubularmotor protein-based mitochondrial movements (20, 23). Thesedata suggest that mitochondrial movements may not always berandombut couldbedirected to certain cellular domains andhavespecific localization-dependent functions.

EXPERIMENTAL PROCEDURES

Cells—Human Jurkat T-cell lines were isolated and grown asdescribed previously (24, 25).Reagents—All chemicals and antibodies not specificallymen-

tionedwere from Sigma (highest grade). Other reagents used inour experiments included Fura-2/AM, thapsigargin (TG),MitoTracker� Green FM, BAPTA (all fromMolecular Probes),mouse anti-human CD3-RPE-conjugated mAb (DakoCytoma-tion), mouse anti-human CD45-Alexa Fluor�488 mAb (Sero-tec), di-8-aminonaphthylethenylpyridinium (Invitrogen), andBTP2 (Altana Pharma).Fluorescence Microscopy and Ca2� Imaging—Imaging experi-

mentsweredoneaspreviously described (26).A40� (Uplan/Apo,numerical aperture 1.0) or 100� (Uplan/Apo numerical aperture1.35, oil immersion) objective were used. Ca2� Ringer’s solutioncontained (inmM): 155NaCl, 4.5 KCl, 1 CaCl2, 2MgCl2, 10 D-glu-cose, and5Hepes (pH7.4withNaOH).CaCl2was substitutedby1mM EGTA to prepare Ca2�-free Ringer’s solution.

To visualizemitochondria, cellswere incubatedwith 100–200nMMitoTracker� for 30 min at 22�23 °C. Excess dye was removed bywashingtwice.Cellswere illuminatedat490nm.DCLP500(UV)wasused as a dichroicmirror andLP515 as an emission filter.Bead Stimulation—As described previously (27), we followed

the standard procedure for absorbing proteins on polystyrenemicroparticles (size�0.5�m) established by Polysciences Europe

* This project was funded by the Deutsche Forschungsgemeinschaft (SFB530, Project A3 and B6 and Graduate College “Molecular, physiological andpharmacological analysis of cellular membrane transport”), HomburgerForschungsförderungsprogramm, and the Hochschulbauförderungsge-setz Projects 136-364 and 136-418. The costs of publication of this articlewere defrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18 U.S.C.Section 1734 solely to indicate this fact.

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Fig. 1.

1 To whom correspondence should be addressed: Institut fur Physiologie,Gebaude 58, Universitat des Saarlandes, D-66421 Homburg/Saar, Ger-many. Tel.: 49-6841-1626465; Fax: 49-6841-1626060; E-mail: ptaqgo@uniklinikum-saarland.de.

2 The abbreviations used are: CRAC, Ca2� release-activated Ca2�; BTP2,bis(trifluoromethyl)pyrazole 2; TG, thapsigargin; BAPTA, 1,2-bis(2-amino-phenoxyl)ethane-N,N,N�,N�-tetraacetic acid; mAb, monoclonal antibody;NFAT, nuclear factor of activated T-cells.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 281, NO. 52, pp. 40302–40309, December 29, 2006© 2006 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

40302 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 52 • DECEMBER 29, 2006

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

GmbHCompany (www.polysciences.com).Twoalterations in theprocedure were required to optimize our results. In step 1, weemployed 100 �l of a 2.5% suspension of beads (� 2.1� 107beads), and in step7,weadded50�gof theprotein tobe absorbed.Chemical composition and pH of buffers were the same as rec-ommended in the protocol. Azid-free anti-human CD3 mAbs(Euroclone) were passively coupled to microparticles (diame-ter � 5.83 �m). They were stored at 4 °C in the specified stor-age buffer until use. Beads were washed twice with phosphate-buffered saline before resuspending them in the Ringer’ssolution used for the experiments.Confocal Microscopy—For confocal imaging, a Nipkow disc-

based scanning head (QLC-100, VisiTech International) was

attached to an upright microscope(Eclipse 600, Nikon) equipped witha 100� water lens (numerical aper-ture 1.1, Nikon). The light sourcewas a 488-nm solid-state laser(Saphire 488–30, Coherent). Adichroic mirror between the micro-lens and pinhole disc reflects theemission light that passes a 500-nmlong-pass barrier filter. For detec-tion, a charge-coupled device cam-era (OrcaER, Hamamatsu Photon-ics) in the 2 � 2 binning mode,resulting in a 672 � 512 pixel reso-lution, was used. Acquisition wascontrolled by the VoxCellScan Soft-ware (VisiTech International).Two-dimensional time lapse data

at a maximum rate of 20 images/swere analyzed using ImageJ (WayneRasband,National Institute ofMentalHealth). Stacks were deconvolvedusing AutoDeblur X (AutoQuant).2-Photon Microscopy—The same

experimental set-up as previouslydescribed was used (28). The emis-sion beam splitter contained a dich-roic mirror at 565 nm followed by achannel 1 band pass at 610/75 nmand a channel 2 long pass at 500 nm.The 2-photon excitation setting was800 nm for dyes with 1-photon exci-tation near 488 nm. Complete pic-tures of the sample were taken every5minwith a resolution of 512 � 512or 2048 � 2048 pixels.To stain mitochondria, cells were

loadedwith 200nMMitoTracker� for1 h. The plasma membrane wasstained with a mouse anti-humanCD45-Alexa Fluor�488 mAb. Cellswere incubated for 30 min withantibodies.Electrophysiology—Patch clamp

experiments were done exactly asdescribed previously (26). The standard pipette solution forwhole-cell patch clamp recordings contained (in mM): 0.05InsP3, 5 � 10�8 TG, 140 Cs-aspartate, 2MgCl2, 5Mg-ATP, 0.5Tris-GTP, 2.5 malic acid, 2.5 Na-pyruvate, 1 NaH2PO4, and 10Hepes (pH 7.2 with CsOH). This solution was either supple-mented with 0.66 CaCl2 and 1.2 EGTA ([Ca2�]free � 5 nM) toweakly buffer [Ca2�]i or with 10 mM EGTA to strongly buffer[Ca2�]i. The extracellular solution contained (in mM) 101.66NaCl, 3 KCl, 20.66 CaCl2, 1.22 MgCl2, 6.66 D-glucose, and 3.33Hepes (pH 7.4 with NaOH).Electroporation of Antibodies—Electroporation was carried

out as described previously (27) with RPMI 1640 medium con-taining 0.1 mg/ml anti-kinesin 5B mAb or 0.1 mg/ml anti-hu-

FIGURE 1. Sustained Ca2� signals induce directed translocation of mitochondria toward the plasmamembrane. A, Ca2� response of Fura-2/AM-loaded T-cells (average of 253 cells) depicted as a function oftime. B, MitoTracker� fluorescence and infrared images taken in parallel from a single MitoTracker� Green/AM-loaded T-cell at the time points indicated. Note the significant accumulation of MitoTracker� fluores-cence between both concentric circles only during the Ca2� plateau. The outer circle reflects the position ofthe plasma membrane, whereas the inner circle is positioned 0.99 �m apart from the outer one. C, statis-tical analysis of the mitochondrial localization between the concentric circles in 19 T-cells at the sametime points as in A. D, statistical analysis of mitochondrial population redistribution in T-cells(14 cells) under resting conditions after TG-induced store depletion and at the end of the Ca2� plateau(compare A). The relative fluorescence (as a fraction of the total fluorescence) between five differentconcentric circles (resulting in four concentric areas spaced 0.99 �m apart; see inset) is shown for eachcondition. E, color-coded [Ca2�]i images, infrared images, and [Ca2�]i kinetic from a typical Fura-2/AM-loaded Jurkat T-cell stimulated by anti-CD3 beads in 1 mM Ca2� Ringer’s solution. F, MitoTracker� fluores-cence and infrared images from a single MitoTracker� Green/AM-loaded Jurkat T-cell before and at theend of the Ca2� plateau �1000 s after stimulation with anti-CD3 beads in 1 mM Ca2� Ringer’s solution.Statistical analysis of the mitochondrial localization close to the plasma membrane in 19 bead-stimulatedJurkat T-cells is shown. IR, infrared image; Mito, MitoTracker� image.

Mitochondria Localization and CRAC Activity

DECEMBER 29, 2006 • VOLUME 281 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 40303

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

man IgGAb. Cells were pulsed with 300 V and 400microfaradsand then cultivated until use (30 min after transfection).DataAnalysis and Statistics—Datawere analyzed usingTILL

Vision (TILL Photonics), Igor Pro (Wavemetrics), andMicrosoft Excel. All values are given as mean � S.E.;n � number of cells. In case data points were normally distrib-

uted, a paired or unpaired two-sidedStudent’s t test was used. If normaldistribution could not be con-firmed, a non-parameterized test(Mann-Whitney) was carried out.Levels of significance are indicatedin Figs. 1–5 (* refers to p 0.05, **refers to p 0.01, and *** refers top 0.001).

RESULTS

Ca2� Influx Induces Mitochon-drial Translocation toward thePlasma Membrane—To test thehypothesis of whether or notdirected mitochondrial movementwas important for Ca2� signalingand Ca2�-dependent T-cell activa-tion, we used different microscopytechniques to analyze the intracellu-lar localization of mitochondria.Because CRAC channels are stimu-lated by Ca2� store depletion andinactivated by store refilling (2), westimulated T-cells with TG, whichhas been shown to maximally acti-vate CRAC channels by blockingthe sarcoendoplasmatic reticulumCa2� ATPase pump (29). The intra-cellular localization of mitochon-dria was analyzed before (restingcondition) and after T-cell stimula-tion by the application of TG in theabsence and in the presence ofextracellular Ca2�, a standard pro-tocol to activate Ca2� influxthrough CRAC channels (Fig. 1A).We found that sustained Ca2�

influx (Ca2� plateau) was correlatedwith mitochondrial translocationtoward the plasma membrane. Thisis evident when comparing thechanges in a subplasma membraneMitoTracker� fluorescence afterstimulation between concentric cir-cles, the outer of which reflects theposition of the plasma membrane,the inner one placed 0.99 �m apartfrom the outer one (Fig. 1, B andC). The analysis with two concen-tric circles was chosen becauseunstimulated T-cells were fairly

round and did not change their shape significantly during thap-sigargin treatment. Directed movement toward the plasmamembrane was only seen during the sustained Ca2� plateau (at1000 s after cell stimulation) but not during store depletion (at500 s after cell stimulation) or during the very first seconds ofthe Ca2� influx phase. A similar mitochondrial translocation

FIGURE 2. Ca2� influx through CRAC channels controls the translocation of mitochondria to theplasma membrane. A, statistical analysis of the mitochondrial localization close to the plasma membranebetween the two different concentric circles (as in Fig. 1B) at different times after initiation of Ca2� influxby 1 �M TG as shown in the inset. B, statistical analysis of the mitochondrial localization close to the plasmamembrane under resting conditions (open bars) and at the end of the Ca2� plateau (shaded bars) afterstimulation with (in �M) 1 TG (see inset in A) in 0.1 (9 cells), 0.25 (9 cells), 0.5 (9 cells), 1 (13 cells), and 20 (10cells) mM Ca2� Ringer’s solution. C, statistical analysis of the mitochondrial localization close to the plasmamembrane under resting conditions, at the end of the Ca2� plateau, and 15 min after removing extracel-lular Ca2� solution in 12 cells. D, statistical analysis of the mitochondrial localization close to the plasmamembrane after loading the cells for 30 min with (in �M) 20 (26 cells), 50 (24 cells), and 100 (22 cells)BATPA/AM. E, statistical analysis of the mitochondrial localization close to the plasma membrane in thepresence of 100 nM BTP2 under resting conditions, after TG-induced store depletion, and at the end of theCa2� plateau in 17 cells. F, confocal and 2-photon analysis of mitochondrial movement. Confocal fluores-cence images from single MitoTracker�/AM/di-8-aminonaphthylethenylpyridinium co-labeled JurkatT-cells before and after inducing Ca2� influx by the re-addition of 1 mM Ca2� Ringer’s solution in TGpretreated or nonpretreated cells. The statistical analysis of the fraction of mitochondria located in a ring1 �m away from the plasma membrane is depicted at the bottom. Averaged were 23 (resting) and 12 (forthe other conditions) cells. For 2-photon measurements, cells were labeled with MitoTracker�/AM andmouse anti-human CD45-Alexa Fluor�488 mAb (to stain the plasma membrane). The statistical analysis ofthe fraction of mitochondria located in a ring 300 nm away from the plasma membrane is depicted atthe bottom. Averaged were 38 cells for each condition.

Mitochondria Localization and CRAC Activity

40304 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 52 • DECEMBER 29, 2006

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

toward the plasma membrane was also observed in CD3�

peripheral blood lymphocytes during the Ca2� plateau (datanot shown). We next analyzed the complete mitochondrialpopulation by spacing five concentric circles within the cellswith a constant distance of 0.99 �m (Fig. 1D). The average dis-tance between mitochondria and the plasma membrane did notchange during store depletion compared with resting conditions;however, during the Ca2� plateau, the fluorescence between theouter circles was significantly increased. In parallel, the Mito-Tracker� fluorescence decreased between the inner circles, indi-cating that the complete mitochondrial population had movedtoward the plasmamembrane.To analyze whether mitochondrial movement toward the

plasma membrane also occurred in T-cells in the case of T-cellreceptor stimulation, we used anti-CD3-coated beads to stim-ulate the cells. Fig. 1E shows a typical bead experiment in whichthe contact between bead and cell induces a rise of the intracel-lular Ca2� concentration ([Ca2�]i), as previously reported (27).

In parallel, mitochondrial move-ment toward the plasma membranewas also observed after bead stimu-lation, as evident from Fig. 1F.[Ca2�]i signals and the amount ofmitochondrial movement werecomparable with the cells stimu-lated with TG. In the analysis, weonly included cells that did notchange their shape following beadstimulation. We observed also,however, translocation of mito-chondria to the plasma membranein cells that did change their shapesignificantly following bead contact.Because Ca2� depletion did not

induce any mitochondrial translo-cation toward the plasma mem-brane, we tested mitochondrialmovement in the following experi-ments by inducing Ca2� influxthrough TG in 1 mM Ca2� Ringer’ssolution, as shown in the inset of Fig.2A. A translocation ofmitochondriatoward the plasma membrane wasalready measured 2 min after theonset of Ca2� entry, suggesting theexistence of a fast Ca2�-dependentmitochondrial movement mecha-nism (Fig. 2A). The dependence ofmitochondrial movement on Ca2�

influx was supported by the correla-tion between the translocation ofmitochondria toward the plasmamembrane and the increase of theextracellular Ca2� concentration(Fig. 2B), which enhances the driv-ing force for Ca2� entry, as previ-ously reported (24). Even in thepresence of 0.1 or 0.25 mM external

Ca2�, which generates only very little Ca2� entry throughCRACchannels, a smallmovement ofmitochondria toward theplasma membrane was observed. The movement appears to besaturated with 1 mM external Ca2� present. Another argumentfor the dependence of the mitochondrial movement on Ca2�

influx can be gathered from the observation thatmitochondrialtranslocation toward the plasma membrane was completelyreversed after removing the extracellular Ca2� solution (Fig.2C) and was abolished in T-cells loaded with the Ca2� chelatorBAPTA (Fig. 2D). Because Ca2� influx in T-cells dependsstrictly on the activity of CRAC channels (24, 30), we analyzedthe intracellular localization of mitochondria in T-cells treatedwith the CRAC channel blocker 3,5-bis(trifluoromethyl)pyra-zole derivate BTP2 (31). 100 nM BTP2 significantly reducedmitochondrial movement toward the plasma membrane (Fig.2E). BTP2 did not completely prevent the movement, whichwas expected, because in the presence of 100 nM BTP2, Ca2�

entry through the CRAC channel is reduced to a level similar to

FIGURE 3. Mitochondria translocation toward the plasma membrane along microtubules is required tosustain long-lasting Ca2� signal in T-cells. A–D, identical experiments as shown in Fig. 1, A–C, except thatcells were pre-incubated with 2 �M nocodazole for 35 min in A and B or different times in C and D: 7 min (green,average of 233 cells), 20 min (blue, average of 83 cells), or 35 min (red, average of 44 cells). E, statistical analysisof the mitochondrial localization close to the plasma membrane in T-cells under resting conditions and 15 minafter stimulation with 1 mM Ca2� Ringer’s solution containing 1 �M TG, as shown in F. Analyzed were 40 cells(Resting), 22 cells (1Ca � TG), 20 cells (TG � nocodazole). F, Ca2� response of Fura-2/AM-loaded T-cells (pre-treated for 35 min or not pretreated with 2 �M nocodazole). IR, infrared image; Mito, MitoTracker� image.

Mitochondria Localization and CRAC Activity

DECEMBER 29, 2006 • VOLUME 281 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 40305

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

conditions in which 0.1–0.25 mM are present in the externalsolution (compare Ref. 31 and Fig. 2B). Directedmitochondrialmovement toward the plasma membrane in T-cells followingCa2� influx through CRAC channels was also confirmed byconfocal microscopy and by 2-photon microscopy (Fig. 2F). Inboth cases, the mitochondrial population was translocated toareas close to the plasma membrane after the Ca2� signalreached the plateau followingCa2� influx induced by the re-ad-dition of 1mMCa2� solution inTG-pretreatedT-cells, an effectnot observed after the re-addition of 1 mM Ca2� solution with-out TG pretreatment. In addition, we also confirmed mitochon-drial movement toward the plasma membrane in T-cells trans-fected with mitochondrial green fluorescent protein (data notshown). The results shown in Figs. 1 and 2 demonstrate that thedirected movement of mitochondria toward the plasma mem-brane depends onCa2� influx throughCRAC channels in T-cells.Mitochondria Translocation toward the Plasma Membrane

Is Required for Sustained Ca2� Signals—The mitochondrialmovement toward the plasma membrane following the activa-tion of Ca2� influx through CRAC channels was almost com-

pletely abolished by pre-incubationof the cells with nocodazole, amicrotubule inhibitor (Fig. 3, A andB) but only slightly with latrunculinB, a microfilament inhibitor (datanot shown). The inhibition of mito-chondrial movement toward theplasma membrane by nocodazolegreatly reduced the sustained Ca2�

plateau compared with control con-ditions (Fig. 3,C andD). The disrup-tion of microtubules induced bynocodazole is time-dependent,because the percentage of sub-plasma membrane MitoTracker�fluorescence between the two con-centric circles as well as the sus-tainedCa2� signal decreasewith thenocodazole incubation time (Fig. 3,C andD). TheCa2� re-addition pro-tocol usually generates very highCa2� influx signals, because theslow up-modulation of the plasmamembrane Ca2� ATPases cannotprevent the “overshoot” of the intra-cellular Ca2� concentration (32).To avoid this [Ca2�]i overshoot, wealso compared the effect of nocoda-zole to control conditions bydirectly activatingCa2� influx in thepresence of external Ca2�. Underthese conditions, mitochondrialmovement toward the plasmamembrane was completely abol-ished by nocodazole (Fig. 3E) andthe concomitant [Ca2�]i rise wasalso drastically reduced (Fig. 3F).Nocodazole did not interfere with

mitochondrial Ca2�uptake or plasmamembraneCa2�ATPaseactivity (supplemental Fig. 1). We therefore postulate that thesustained high [Ca2�]i, which depends on Ca2� influx throughCRAC channels, is maintained through stimulation-dependentmitochondrial movement toward the plasma membrane. Atestable prediction of this model is that inhibition ofmitochon-drial movement into the vicinity of the plasma membraneshould decrease CRAC current amplitudes.Mitochondrial Movement toward the Plasma Membrane

Controls CRAC Channel Activity—To analyze CRAC channelactivity, we used the whole-cell configuration of the patchclamp technique. CRAC channels were activated throughdepletion of the endoplasmic reticulum Ca2� store using acombination of InsP3 and TG. In a first series of experiments,we comparedCRACchannel inactivation using a small concen-tration of the Ca2� chelator EGTA in the pipette, which is notsufficient to prevent Ca2�-dependent inactivation of CRACchannels (16). Under these conditions, mitochondrial Ca2�

uptake close to the channels is required to maintain CRACchannel activity by reducing local accumulation of Ca2� near

FIGURE 4. Patch clamp analysis of CRAC channel activity reveals its dependence on mitochondrial move-ment toward the plasma membrane. Examples of CRAC channel recordings at �80 mV are shown with 0.66mM Ca2�/1.2 mM EGTA � nocodazole (A and B) or 10 mM EGTA � nocodazole (C and D) in the pipette solution.E, statistical analysis of the CRAC current 3 min after establishing the whole-cell configuration as a fraction ofthe maximal current for the conditions shown in A–D.

Mitochondria Localization and CRAC Activity

40306 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 52 • DECEMBER 29, 2006

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

sites that govern Ca2�-dependent CRAC channel inactiva-tion (Fig. 4A; compare also Ref. 16). Because pre-incubationof T-cells with nocodazole inhibits the fast movement of mi-

tochondria along microtubulestoward the plasma membrane afterthe Ca2� influx through CRACchannels, our model predicts thatCRAC currents should be inacti-vated faster under these conditions.We indeed found a significant inac-tivation of CRAC channels innocodazole-treated T-cells com-pared with control conditions (Fig.4, B and E). To verify that this is nota direct effect of nocodazole on theCRAC channels or an unspecificside effect, we repeated the experi-ments in the presence of 10 mMEGTA in the pipette solution, whichis sufficient to buffer all incomingCa2�, thereby preventing CRACchannel inactivation. Under theseconditions, nocodazole had noeffect on CRAC channel inactiva-tion (Fig. 4, compareC andD; statis-tics in E). Because mitochondrialCa2� uptake close to the plasmamembrane is not required to pre-vent Ca2�-dependent CRAC chan-nel inactivation in the presence of10 mM EGTA (Fig. 4C), we canexclude that potential unspecificside effects of nocodazole wereresponsible for the inactivation ofCRAC currents observed in Fig. 4B.The time course of CRAC inactiva-tion (Fig. 4B) is comparable with theonset of mitochondrial movementtoward the plasma membranewithin the first 2 min following acti-vation of Ca2� influx (Fig. 2A).From the experiments in Figs. 3 and4, we conclude that mitochondrialmovement toward the plasmamembrane is required to maintainCRAC channel activity and sus-tained high [Ca2�]i signals.Kinesin Controls Mitochondrial

Translocation toward the PlasmaMembrane and Sustained Ca2�

Signals—Kinesin motor proteinscontrol the anterograde move-ment of different organelles andcargos along microtubules (33).Because the translocation of mito-chondria toward the plasma mem-brane is an anterograde motion,we introduced 0.1 mg/ml anti-ki-

nesin 5B mAb into T-cells using electroporation and ana-lyzed in-parallel mitochondrial translocation and [Ca2�]isignals following TG stimulation. Anti-kinesin 5B mAb

FIGURE 5. Kinesin motor proteins control both mitochondrial translocation toward the plasma membraneand sustained Ca2� signals. A, MitoTracker� fluorescence and infrared images from single MitoTracker� Green/AM-loaded/anti-kinesin 5B or anti-IgG mAb-transfected T-cells before and after stimulation with 1 �M TG in 1 mM

Ca2� Ringer’s solution. MitoTracker� and infrared images were taken in-parallel before (Resting) and at the end ofthe Ca2� plateau (at 1000 s after cell stimulation). B, statistical analysis of the mitochondrial localization close to theplasma membrane in transfected resting T-cells (n � 29) at the end of the Ca2� plateau in anti-IgG mAb-transfectedT-cells (n � 14) and at the end of the plateau in anti-kinesin 5B mAb-transfected T-cells (n � 41). C, identical exper-iments as the one presented in Fig. 1A, except that cells were transfected with anti-IgG mAbs as the control oranti-kinesin 5B mAbs. The [Ca2�]i at 900–1000 s was analyzed for anti-IgG mAbs (865 � 31 nM, 457 cells) andanti-kinesin 5B mAb (414 � 17 nM, 320 cells)-treated cells. IR, infrared image; Mito, MitoTracker� image.

FIGURE 6. Model for the mitochondrial movement-dependent, efficient T-cell activation. Mitochondrial trans-location toward the plasma membrane is essential for a sustained [Ca2�]i signal and subsequent transcription factoractivity. This directed and Ca2� influx-dependent mitochondrial relocalization allows efficient T-cell activation, mostlikely by initiating a closer contact between mitochondria and CRAC channels. This close localization facilitates alarger and more sustained Ca2� influx by reducing the Ca2�-dependent CRAC channel inactivation, which subse-quently leads to higher [Ca2�]i and a concomitant change in the transcription factor profile from NFAT-AP1 to NFAT,NFAT-AP1, and NF�B (see also Refs. 3 and 7).

Mitochondria Localization and CRAC Activity

DECEMBER 29, 2006 • VOLUME 281 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 40307

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

completely blocked the anterograde mitochondrial translo-cation toward the plasma membrane (Fig. 5A; statistics in B)compared with control conditions (electroporation withanti-human IgG mAb), and this movement inhibition corre-lated with a significant reduction of the sustained Ca2� sig-nal (Fig. 5C), stressing again the importance of mitochon-drial localization beneath plasma membrane for Ca2�

signals and Ca2�-dependent signal transduction in T-cells.The introduction of the anti-kinesin 5BmAb induced a dras-tic reduction of the initial amplitude of the Ca2� signal uponCa2� re-addition, which was not observed in cells treatedwith nocodazole (Fig. 3C). We believe that this discrepancymay be due to the finding that mitochondria collapse close tothe nucleus once the function of kinesin has been disrupted(21, 34, 35), whereas nocodazole may simply freeze the cur-rent position of mitochondria. The collapse of mitochondriaclose to the nucleus has been shown to reduce store-depend-ent Ca2� signals (34).

The data with the anti-kinesin 5BmAb explain the fast trans-location of mitochondria toward the plasma membraneobserved following the Ca2� rise in T-cells, because kinesinmotor proteins move along microtubules by a hand-over-handmechanism that allows rapid transport of cargoes over longdistances (36).

DISCUSSION

We have shown that stimulation of Ca2� entry in T-cellsfacilitates translocation of the complete mitochondrial popula-tion toward the plasma membrane in a Ca2�/kinesin-depend-ent manner along microtubules (Fig. 6). Ca2� influx throughCRAC channels triggers directed mitochondrial motility. Theexact mechanism of the directed movement, however, remainsunknown. The mitochondrial movement toward the plasmamembrane in turn sustains CRAC channel activity and high[Ca2�]i by reducing Ca2�-dependent CRAC channel inactiva-tion. We refer to this as a double positive feedback loop that isinitiated by CRAC channel activation through depletion ofCa2� stores; Ca2� influx through activated CRAC channelstriggers mitochondrial translocation to the plasma membrane,which in turn keeps CRAC channels active. There are numer-ous examples of such double positive feedback loops; for exam-ple, InsP3 activates InsP3-R, which releases Ca2�, which bindsto InsP3-R, which in turn releases more Ca2�.According to a model proposed by Hajnoczky and co-work-

ers (20), the high submembranous [Ca2�]i (in our case gener-ated through CRAC activity) would decrease mitochondrialmotility thereby immobilizing them close to the plasma mem-brane. There are, however, somedifferences between themodelof Hajnoczky and co-workers (20) and our data. Hajnoczky andco-workers show that mitochondrial motility is reduced in amyoblast cell line (H9c2) after Ca2� release and recovered by50% when the Ca2� signal reaches a plateau. They concludethat mitochondria are immobilized close to microdomains ofhigh Ca2� near the endoplasmic reticulum. In T-cells, we donot observe such a high Ca2� release, and T-cell activation islargely determined by Ca2� influx through CRAC channels.Ca2� release in T-cells is usually initiated in large areasthroughout the cytosol. Therefore, there would probably be no

preferential direction for mitochondria movement. Hence, inthe “release” case, Ca2�-dependent mitochondria movementtoward the endoplasmic reticulum would be extremely hard tomeasure, because suchmovementwould almost not differ fromrandom movement. In the “influx” case, there is a preferentialdirection for a Ca2�-dependent movement, namely toward theplasmamembrane, where they would be finally immobilized bythe high [Ca2�]i microdomains close to CRAC channels.Therefore, the data by Hajnoczky and co-workers (20) and ourdatamay be reconcilable, because after an initialmovement duetoCa2� signals (whichmaynot be distinguishable from randommovement in case the endoplasmic reticulum is the Ca2�

source), the subsequent high [Ca2�]i may immobilize mito-chondria (in our case at the plasma membrane). In addition, inflat cells, such as adherent myoblasts, there may be no directedmovement of mitochondria to the plasma membrane, becausemitochondria are already close to the influx channel (in zdimension).Because Ca2� influx through CRAC channels increases the

accumulation of mitochondria close to the plasma membraneand this localization near CRAC channels sustains CRAC activ-ity, we propose that CRAC channels and mitochondria controleach other in a double positive feedback loop. The concomitantincrease in global [Ca2�]i shifts the transcription factor profileaway from mostly NFAT-AP1 to a combination of NFAT,NFAT-AP1, and NF�B (Fig. 6; see also Refs. 3 and 7). Conse-quently, this transcriptional profile change leads to a differen-tial protein expression. Thus, a small change in mitochondriallocalization is likely translated into a large effect on T-cellactivity.REFERENCES1. Lewis, R. S., and Cahalan, M. D. (1995)Annu. Rev. Immunol. 13, 623–6532. Parekh, A. B., and Putney, J. W. Jr. (2005) Physiol. Rev. 85, 757–8103. Quintana, A., Griesemer, D., Schwarz, E. C., andHoth,M. (2005) Pfluegers

Arch. 450, 1–124. Gallo, E. M., Cante-Barrett, K., and Crabtree, G. R. (2006) Nat. Immunol.

7, 25–325. Dolmetsch, R. E., Xu, K., and Lewis, R. S. (1998) Nature 392, 933–9366. Bhakta, N. R., Oh, D. Y., and Lewis, R. S. (2005)Nat. Immunol. 6, 143–1517. Macian, F., Garcia-Cozar, F., Im, S. H., Horton, H. F., Byrne, M. C., and

Rao, A. (2002) Cell 109, 719–7318. Feske, S., Gwack, Y., Prakriya, M., Srikanth, S., Puppel, S. H., Tanasa, B.,

Hogan, P. G., Lewis, R. S., Daly, M., and Rao, A. (2006) Nature 441,179–185

9. Vig,M., Peinelt, C., Beck, A., Koomoa, D. L., Rabah, D., Koblan-Huberson,M., Kraft, S., Turner, H., Fleig, A., Penner, R., andKinet, J. P. (2006) Science312, 1220–1223

10. Liou, J., Kim, M. L., Heo, W. D., Jones, J. T., Myers, J. W., Ferrell, J. E. Jr.,and Meyer, T. (2005) Curr. Biol. 15, 1235–1241

11. Zhang, S. L., Yu, Y., Roos, J., Kozak, J. A., Deerinck, T. J., Ellisman, M. H.,Stauderman, K. A., and Cahalan, M. D. (2005) Nature 437, 902–905

12. Spassova, M. A., Soboloff, J., He, L. P., Xu, W., Dziadek, M. A., and Gill,D. L. (2006) Proc. Natl. Acad. Sci. U. S. A. 103, 4040–4045

13. Nolz, J. C., Gomez, T. S., Zhu, P., Li, S., Medeiros, R. B., Shimizu, Y.,Burkhardt, J. K., Freedman, B. D., and Billadeau, D. D. (2006) Curr. Biol.16, 24–34

14. Newmeyer, D. D., and Ferguson-Miller, S. (2003) Cell 112, 481–49015. Duchen, M. R. (2000) J. Physiol. (Lond.) 529, 57–6816. Hoth, M., Button, D. C., and Lewis, R. S. (2000) Proc. Natl. Acad. Sci.

U. S. A. 97, 10607–1061217. Glitsch, M. D., Bakowski, D., and Parekh, A. B. (2002) EMBO J. 21,

6744–6754

Mitochondria Localization and CRAC Activity

40308 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 52 • DECEMBER 29, 2006

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from

18. Ball, E. H., and Singer, S. J. (1982) Proc. Natl. Acad. Sci. U. S. A. 79,123–126

19. Knowles, M. K., Guenza, M. G., Capaldi, R. A., and Marcus, A. H. (2002)Proc. Natl. Acad. Sci. U. S. A. 99, 14772–14777

20. Yi, M., Weaver, D., and Hajnoczky, G. (2004) J. Cell Biol. 167, 661–67221. Varadi, A., Johnson-Cadwell, L. I., Cirulli, V., Yoon, Y., Allan, V. J., and

Rutter, G. A. (2004) J. Cell Sci. 117, 4389–440022. Morris, R. L., and Hollenbeck, P. J. (1995) J. Cell Biol. 131, 1315–132623. Yoon, Y. (2005) Science’s STKE 2005, pe1824. Fanger, C.M., Hoth,M., Crabtree, G. R., and Lewis, R. S. (1995) J. Cell Biol.

131, 655–66725. Schwarz, A., Tutsch, E., Ludwig, B., Schwarz, E. C., Stallmach, A., and

Hoth, M. (2004) J. Biol. Chem. 279, 5641–564726. Philipp, S., Strauss, B., Hirnet, D., Wissenbach, U., Mery, L., Flockerzi, V.,

and Hoth, M. (2003) J. Biol. Chem. 278, 26629–2663827. Quintana, A., and Hoth, M. (2004) Cell Calcium 36, 99–109

28. Tutsch, E., Griesemer, D., Schwarz, A., Stallmach, A., andHoth,M. (2004)Eur. J. Immunol. 34, 3477–3484

29. Zweifach, A., and Lewis, R. S. (1993) Proc. Natl. Acad. Sci. U. S. A. 90,6295–6299

30. Feske, S., Prakriya, M., Rao, A., and Lewis, R. S. (2005) J. Exp. Med. 202,651–662

31. Zitt, C., Strauss, B., Schwarz, E. C., Spaeth, N., Rast, G., Hatzelmann, A.,and Hoth, M. (2004) J. Biol. Chem. 279, 12427–12437

32. Bautista, D. M., Hoth, M., and Lewis, R. S. (2002) J. Physiol. (Lond.) 541,877–894

33. Vale, R. D. (2003) Cell 112, 467–48034. Varadi, A., Cirulli, V., and Rutter, G. A. (2004) Cell Calcium 36,

499–50835. Hirokawa, N., and Takemura, R. (2005) Nat. Rev. Neurosci. 6, 201–21436. Yildiz, A., Tomishige, M., Vale, R. D., and Selvin, P. R. (2004) Science 303,

676–678

Mitochondria Localization and CRAC Activity

DECEMBER 29, 2006 • VOLUME 281 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 40309

at Biom

edical Library, U

CSD

on September 22, 2014

http://ww

w.jbc.org/

Dow

nloaded from