Excessive Na�/H� Exchange in Disruption of Dendritic Na�

and Ca2� Homeostasis and Mitochondrial Dysfunctionfollowing in Vitro Ischemia*□S

Received for publication, January 6, 2010, and in revised form, September 1, 2010 Published, JBC Papers in Press, September 3, 2010, DOI 10.1074/jbc.M110.101212

Douglas B. Kintner‡, Xinzhi Chen‡§, Julia Currie‡, Vishal Chanana¶, Peter Ferrazzano¶, Akemichi Baba�,Toshio Matsuda�, Mike Cohen**, John Orlowski‡‡, Shing-Yan Chiu§§, Jack Taunton**, and Dandan Sun‡¶1

From the ‡Department of Neurological Surgery, the §Neuroscience Training Program, the §§Department of Physiology, and the¶Waisman Center, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin 53705, the �Graduate Schoolof Pharmaceutical Sciences, Osaka University, Osaka 565-0871, Japan, the **Program in Chemistry and Chemical Biology andDepartment of Cellular and Molecular Pharmacology, University of California, San Francisco, California 94143, and the‡‡Department of Physiology, McGill University, Montreal H3G 1Y6, Canada

Neuronal dendrites are vulnerable to injury under diversepathological conditions. However, the underlying mechanismsfor dendritic Na� overload and the selective dendritic injuryremainpoorly understood.Our current studydemonstrates thatactivation ofNHE-1 (Na�/H� exchanger isoform1) in dendritespresents amajor pathway for Na� overload. Neuronal dendritesexhibited higher pHi regulation rates than soma as a result of alarger surface area/volume ratio. Following a 2-hoxygenglucosedeprivation and a 1-h reoxygenation, NHE-1 activity wasincreased by �70–200% in dendrites. This elevation dependedon activation of p90 ribosomal S6 kinase.Moreover, stimulationof NHE-1 caused dendritic Na�

i accumulation, swelling, and aconcurrent loss of Ca2�

i homeostasis. The Ca2�i overload in

dendrites preceded the changes in soma. Inhibition ofNHE-1 orthe reverse mode of Na�/Ca2� exchange prevented thesechanges.Mitochondrialmembrane potential in dendrites depo-larized 40 min earlier than soma following oxygen glucose dep-rivation/reoxygenation. BlockingNHE-1 activity not only atten-uated loss of dendritic mitochondrial membrane potential andmitochondrial Ca2� homeostasis but also preserved dendriticmembrane integrity. Taken together, our study demonstratesthat NHE-1-mediated Na� entry and subsequent Na�/Ca2�

exchange activation contribute to the selective dendritic vulner-ability to in vitro ischemia.

Neuronal dendrites are vulnerable to injury under diversepathological conditions, including cerebral ischemia, epilepsy,andAlzheimer disease (1, 2). The hallmark of dendritic injury isthe formation of focal swelling or beads along the length of thedendritic arbor (3). However, the underlying mechanisms forthis selective dendritic injury remain poorly understood. The

initial NMDA or kainite-mediated swelling in dendrites of cul-tured neurons depends on intracellular accumulation of Na�

and Cl� but not Ca2� (4). On the other hand, excessive Ca2�

entry plays a role in the long lasting structural damage anddelayed recovery in hippocampal slices in response to NMDA(4, 5). A correlation between dendritic bead formation andATPreduction/mitochondrial dysfunction has been demonstratedin cultured hippocampal neurons following glutamate expo-sure (6). However, the relationship between selective dendriticdamage, loss of Na� and Ca2� homeostasis, andmitochondrialdysfunction following ischemia remains to be defined.NHE-1 (Na�/H� exchanger isoform 1) is a plasma mem-

brane protein present in virtually all mammalian cells and playsa central role in intracellular pH (pHi) and cell volume regula-tion (7). NHE-1 activity is directly activated by intracellularacidification and/or by protein phosphorylation mediated byERK-p90 ribosomal S6 kinase (p90RSK)2 in ischemic neurons(8). Excessive NHE-1 activation results in intracellular Na�

accumulation, which subsequently promotes Ca2� entry viareversal of Na�/Ca2� exchange (NCXrev) and plays an impor-tant role in myocardium ischemia/reperfusion injury (9). Werecently reported that NHE-1 activity in the soma of neuronsand astrocytes is stimulated following ischemia, and inhibitionof NHE-1 activity is neuroprotective (8, 10). In addition, inhi-bition of NHE-1 either pharmacologically or by genetic knock-down reduces infarction at 24 h following in vivo focal ischemia(11). However, it remains unexplored whether concurrent acti-vation of NHE-1 and NCXrev contributes to the selective vul-nerability of postsynaptic neuronal dendrites to ischemicdamage.In the current study, we demonstrated that neurons exhib-

ited robust NHE-1-dependent pHi regulation in their dendritesas a result of their large surface area/volume ratio. Further, invitro ischemia (oxygen glucose deprivation and reoxygenation,* This work was supported, in whole or in part, by National Institutes of Health

Grants RO1NS48216 (to D. S.), R01 GM071434 (to J. T.), MS RG-4054-A-8 (toS.-Y. C.), P30 HD03352 (to the Waisman Center), and CTSA 1UL1RR02511(University of Wisconsin Institute for Clinical and Translational Research)(to P. F.). This work was also supported by American Heart AssociationGrant EIA 0540154 (to D. S.).

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Tables 1 and 2 and Figs. 1 and 2.

1 To whom correspondence should be addressed: T513 Waisman Center,1500 Highland Ave., Madison, WI 53705. Tel.: 608-263-4060; Fax: 608-263-1409; E-mail: sun@neurosurgery.wisc.edu.

2 The abbreviations used are: RSK, ribosomal S6 kinase; NCX, Na�/Ca2�

exchange; NCXrev, Na�/Ca2� exchange reversal; NHE, Na�/H� exchanger;OGD, oxygen glucose deprivation; REOX, reoxygenation; EMEM, Eagle’sminimum essential medium; FCCP, carbonyl cyanide p-trifluoromethoxy-phenylhydrazone; �m, mitochondrial membrane potential; A/V, area/vol-ume; FMK, fluoromethylketone; Ca2�

i and Ca2�m, intracellular and mito-

chondrial Ca2�, respectively; Na�i, intracellular Na�; �i, intrinsic buffer

power.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 285, NO. 45, pp. 35155–35168, November 5, 2010© 2010 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35155

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

OGD/REOX) stimulated NHE-1 activity in large dendrites (Lg-dendrites). NHE-1-mediated Na� entry and subsequent stim-ulation of NCXrev activity contributed to selective ischemicdamage of dendrites. The underlying mechanisms involved theloss of mitochondrial Ca2� homeostasis and mitochondrialmembrane dysfunction.

EXPERIMENTAL PROCEDURES

Materials—Hanks’ balanced salt solution was from Mediat-ech Cellgro (Manassas, VA). Neurobasal medium, B-27 supple-ment, fura-2/AM, SBFI/AM, BCECF/AM, rhod-2/AM, Mito-Tracker Green, TMRE, calcein/AM, JC-1, Vybrant� DiO,SYTO 60, and 4-bromo-A-23187 were from Invitrogen. Sapo-nin, tetraphenylboron, gramicidin, and monensin were pur-chased from Sigma. RU360 was from EMB Chemicals (Gibbs-town, NJ). Pluronic F-127 was from BASF Corp. (Parsippany,NJ). HOE 642 was a kind gift from Aventis Pharma (Frankfurt,Germany). SEA0400 was a kind gift from Taisho Pharmaceuti-cal Co. Ltd. (Omiya, Saitama, Japan). BI-D1870 was purchasedfrom the School of Life Science, University of Dundee (Dundee,Scotland).Pure Cortical Neuron Cultures—Pure cortical neurons from

embryonic day 14–16mouse fetuses (SV129/Black Swiss) wereprepared as described previously (8). The cortices wereremoved fromE14–16 fetuses and treatedwith 0.5mg/ml tryp-sin at 37 °C for 25min. The cells were centrifuged at 300� g for4 min. The cell pellet was diluted in B-27 supplemented neuro-basal medium (2%) containing 0.5 mM L-glutamine and penicil-lin/streptomycin (100 units/ml and 0.1 mg/ml, respectively).The cells were seeded at a density of 1 � 105 cells/cm2 on glasscoverslips in 6-well plastic plates coatedwith poly-D-lysine. Theculturesweremaintained in an incubator (model 3130, ThermoForma, Waltham, MA) with 5% CO2 and atmospheric air at37 °C. Half of the medium was replaced twice a week. 10–15-day cultures were used in the study.OGD Treatment—10–15-day neuronal cultures grown on

coverslips in 6-well plates were rinsed with an isotonic OGDsolution (pH 7.4) containing 0 mM glucose, 21 mM NaHCO3,120 mM NaCl, 5.36 mM KCl, 0.33 mM Na2HPO4, 0.44 mM

KH2PO4, 1.27 mM CaCl2, and 0.81 mM MgSO4. This solutionhas a K� concentration (�5.8 mM) that is similar to that of theneurobasal medium (5.6 mM) used for cell cultures. The cellswere incubated in 1 ml of OGD solution for 2 h in a hypoxicincubator (model 3130, Thermo Forma) containing 94%N2, 1%O2, and 5%CO2. Normoxic control cells were incubated for 2 hin 5%CO2 and atmospheric air in a buffer identical to the OGDsolution except for the addition of 5.5 mM glucose. REOX wasachieved by the addition of glucose (5.5 mM) and incubation at37 °C in 5% CO2 and atmospheric air. Alternately, REOX wasperformed on the microscope stage by superfusion withHCO3

�-EMEM at 37 °C, equilibrated with 5% CO2 and �18%O2 (monitored by an in-line oxygen electrode, model 16-730;Microelectrodes, Bedford, NH).pHi Measurement—pHi measurement and prepulse treat-

ment were performed as described previously with some mod-ifications (8). Briefly, pure neuronal cultures grown on cover-slips were incubated with 2.5–5 �M BCECF/AM for 30 minduring normoxia or during the last 30 min of REOX at 37 °C.

The coverslips were washed with HCO3�-free HEPES-EMEM

andplaced in a temperature-controlled (37 °C) open bath imag-ing chamber (model RC24, Warner Instruments, Hamden,CT). The chamber was mounted on the stage of the TE 300inverted epifluorescence microscope, and 1–3 neurons werevisualized with a �100 oil immersion objective. The cells wereexcited every 10–30 s at 440 and 490 nm, and the emissionfluorescence at 535 nm was recorded. Images were collectedusing a Princeton Instruments MicroMax CCD camera andanalyzed with MetaFluor image-processing software. Fluores-cence changes in regions of interest in soma, Lg-dendrites, andsmall dendrites (Sm-dendrites) were determined. Lg-dendriteswere defined as dendritic segments with a width of 5.3 � 1.2�m, whereas Sm-dendrites were ones with a width of 1.8 � 0.4�m. The ratio of the background-corrected fluorescence emis-sions (F490/F440) for each region was calibrated using the highK�/nigericin technique (8). pHi values were calculated forsoma, Lg-dendrites, and Sm-dendrites using the respectiveBCECF calibration values collected from each region.For the prepulse treatment, cells were subjected to an acid

load by a transient application (1.5 min) of a 30 mM NH4�/NH3

solution. NH4�/NH3 solutions were prepared by replacing 30

mM NaCl in the HEPES-buffered solution with an equimolarconcentration of NH4Cl. pHi recovery rates were determinedfrom the slope of a fitted linear regression within the firstminute after NH4

�/NH3 prepulse (8). To minimize differentialallosteric effects of H� on NHE-1 activity, pHi recovery rateswere measured at pHi �6.2 throughout the study. In the Na�-free experiments, NaCl in the HEPES-buffered solution wasreplaced with an equimolar concentration of NMDG. NMDG-substituted Na�-free solutions (�5 min) do not cause cellswelling in acutely isolated CA1 neurons (12).Determination of Intrinsic Buffer Power (�i)—�i was deter-

mined in somata, Lg- and Sm-dendrites over a range of pHi bysubjecting the cells to progressively decreasing concentrationsofNH4

� inNa�-freeHEPES-EMEMas described previously (8).The total H� net efflux rate (JH�, mMH�/min) was determinedin three neuronal regions by multiplying �i by �pHi/�t at pHi�6.2. In some experiments, JH� was also calculated in the pres-ence of HCO3

�. The buffering by CO2/HCO3� was determined

as �HCO3

� � 2.3 � [HCO3�]i, where [HCO3

�]i � S � PCO2 �10(pHi � pK), where S � 0.0314, PCO2 � 40 mm Hg, and pK �6.12. At pH 6.2, the contribution of �HCO3

� to total buffering innormoxic cells was �6%.Intracellular Na� Measurement—Intracellular Na� concen-

tration ([Na�]i) was measured with the fluorescent dyeSBFI/AM as described previously with some modifications(14). Cultured neurons grown on coverslips were loaded with30 �M SBFI/AM plus 0.02% pluronic acid during a 45-minREOX following a 2-h OGD. The coverslips were placed in theopen bath imaging chamber and superfused (1 ml/min) withHCO3

�-EMEM at 37 °C. Using the Nikon TE 300 inverted epi-fluorescence microscope and a �100 oil immersion lens, neu-rons were excited at 345 and 385 nm, and the emission fluores-cence at 510 nm was recorded. Regions of interest (1–3 cells/area) were drawn to determine SBFI fluorescence changes insoma, Lg-dendrites, and Sm-dendrites. The 345/385 ratioswere analyzed with the MetaFluor image-processing software.

Dendritic Damage

35156 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Absolute [Na�]iwas determined for each cell by performing anin situ calibration as described previously (14). Multiple timepoint data acquisition induced phototoxicity in neurons.Therefore, [Na�]iwas only determined in normoxic controls or45-min REOX-treated neurons.IntracellularCa2�Measurement—Neurons grownon cover-

slips were incubated with 5 �M fura-2 AM during a 2-h OGD.Following OGD, the cells were placed in the open bath imagingchamber and superfused (1 ml/min) with HCO3

�-EMEM at37 °C. Using the Nikon TE 300 inverted epifluorescencemicro-scope and a �100 oil immersion objective lens, neurons wereexcited every 5 min at 345 and 385 nm, and the emission fluo-rescence at 510 nm was recorded. Images were collected andanalyzedwith theMetaFluor image-processing software. At theend of each experiment, the cells were exposed to 1 mMMnCl2in Ca2�-free HCO3

�-EMEM and 5 �M 4-bromo-A-23187. TheCa2�-insensitive fluorescence was subtracted, and the MnCl2-corrected 345/385 emission ratios were converted to [Ca2�] asdescribed previously (14).Measurement of Mitochondrial Ca2�—Neurons on cover-

slips were incubated at 37 °C for 60 min with 200 nM Mito-TrackerGreen and 9�M rhod-2/AM,whichwas reducedwith aminimumof sodium borohydride inHCO3

�-EMEMcontaining3 mM sodium succinate (14). Coverslips were then incubatedfor 2 h under either OGD or normoxia conditions. For REOX,coverslips were placed in the perfusion chamber on the stage ofthe Leica DMIRE2 confocal microscope and superfused (1ml/min) with HCO3

�-EMEM at 37 °C. Cells (1–3 in the field)were visualized with a �100 oil immersion objective andscanned sequentially for MitoTracker Green (excitation 488nm (argon laser line), emission 500–545 nm) and rhod-2 (exci-tation 543 nm (HeNe laser), emission 544–677 nm). TheMito-Tracker Green signal was used to maintain focus prior to eachsequential scan. Sequential scans were analyzed using the Leicaconfocal software. Average grayscale values were collectedfrom regions of interest around mitochondrial clusters exhib-iting colocalization of MitoTracker Green and rhod-2. Ca2�

mlevels were expressed as relative change of rhod-2 signals fromthe base-line values, and summarized data represent the aver-age of the calculated values from 2–3 cells as described previ-ously (14).Measurement of Mitochondrial Membrane Potential (�m)—

The fluorescent probe JC-1 was used to monitor �m asdescribed previously (14). Neurons on coverslips were loadedwith 9�M JC-1 during 2 h ofOGDat 37 °C. FollowingOGD, thecells were placed in the temperature-controlled open bathimaging chamber and superfused (1 ml/min) with HCO3

�-EMEM at 37 °C. Cells were visualized using the Nikon TE 300inverted epifluorescence microscope and a �60 oil immersionobjective. Cells were excited at 480 nm, and emission fluores-cence images were recorded at 535 nm (the monomer) and 640nm (JC-1 aggregates). The ratio of the aggregate to monomerfluorescence was measured in regions of interest in soma, Lg-dendrites, and Sm-dendrites (2–5 cells/area). In this study, weapplied 1.0�MFCCP for 1min to determine themaximal loss ofJC-1 signals. �m was expressed as the percentage of the maxi-mal FCCP-induced change under normoxic controls (14). Webelieve that the FCCP-sensitive loss of JC-1 signals largely

reflects changes in�m and is not affected by plasmamembranepotentials. This is based on a study where 2.5 �M FCCP causedthe immediate collapse of �m and complete depolarization ofplasma membrane potential. However, a low concentration ofFCCP (0.25 �M) had no effect on plasma membrane potential(15).To further confirm that the changes in JC-1 reflect changes

of �m, we also conducted some parallel experiments using thecationicmembrane-permeant fluorescence probeTMRE.Neu-rons were loaded with 5 nM TMRE and 200 nM MitoTrackerGreen in a buffer supplemented with 1 �M tetraphenylboronfor 30 min 37 °C. The coverslips were then placed in the perfu-sion chamber on the stage of the Leica DMIRE2 confocalmicroscope and superfused (1 ml/min) with HCO3

�-EMEM at37 °C supplemented with 5 nM TMRE. Cells (1–3 in the field)were visualized with a �100 oil immersion objective andscanned sequentially for MitoTracker Green (excitation 488nm (argon laser line), emission 500–545 nm) and TMRE (exci-tation 543 nm (HeNe laser), emission 544–677 nm). Sequentialscans were analyzed using the Leica confocal software. TheMitoTracker Green signal was used to maintain focus prior toeach sequential scan and to identify mitochondrial clustersexhibiting colocalization of MitoTracker Green and TMRE.Maximal �m dissipation was induced by FCCP (1.0 �M) at theend of each experiment. At a concentration of 5 nM, TMREbehaves in the non-quench mode and decreases its fluores-cence intensity when �m is reduced. Data are expressed as rel-ative percent change in FCCP-sensitive TMRE signals.Determination of Surface Area/Volume Ratio in Soma and

Dendrites—To determine differences in the ratio of surfacearea to volume in soma and dendrites, neurons grown on cov-erslips were loaded with 0.5 �M calcein/AM (cytosol dye) and 5�M SYTO 60 (nucleus dye) for 30 min at 37 °C. The coverslipswere then placed in the perfusion chamber on the stage of aLeicaDMIRE2 confocalmicroscope and visualizedwith a�100oil immersion objective. A 110-�m-thick image stack (300slices at 512� 512 pixels) was collected sequentially (excitation488 nm (argon laser line), emission 500–545 nm; excitation 543nm (HeNe laser), emission 544–677) and imported into ImageJ(version 1.41, National Institutes of Health). A cellular region(soma, nucleus, and Lg- or Sm-dendrites) was defined, and thesurface area was calculated by summing the product of theregion perimeter with the distance between each image section(0.38 �m). The volume of the region was calculated with theregion area and section distance. The soma volume was cor-rected by subtracting the calculated volume for the nucleus. Noattempt was made to correct for the intracellular volumes ofendoplasmic reticulum or mitochondria.Detection of Dendritic Beading Formation (Varicosities)—To

monitor dendritic beading formation, neurons grown on cov-erslips were loaded with the plasma membrane dye Vybrant�DiO as per the manufacturer’s instructions. Following OGD,the coverslips were placed in the open bath imaging chamberand superfused (1 ml/min) with HCO3

�-EMEM at 37 °C on thestage of a Leica DMIRE2 confocal microscope. A single neuronwas visualized with a �100 oil immersion objective andscanned (512� 512, 200Hz)with an argon laser (excitation 488nm, emission 500–545 nm). The imageswere analyzed for den-

Dendritic Damage

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35157

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

drite beading with ImageJ analysis software. Beads with a diam-eter �4 times larger than the width of the corresponding den-drite were counted in a 90 � 90-�m area. Data represent theaverage of the calculated values from three or four experiments.Immunoblotting—Cells were washed with ice-cold PBS and

lysedwith 30 s sonication at 4 °C in anti-phosphatase buffer (pH7.4) containing 145 mM NaCl, 1.8 mM NaH2PO4, 8.6 mM

Na2HPO4, 100 mMNaF, 10 mMNa4P2O7, 2 mMNa3VO4, 2 mM

EDTA, and 0.2 �M microcystin and protease inhibitors asdescribed previously (11). Protein content was determined bythe bicinchoninic acid method. Protein samples (40 �g/lane)and prestained molecular mass markers (Bio-Rad) were dena-tured in SDS 2� sample buffer and then electrophoreticallyseparated on 8% SDS gels. The resolved proteins were electro-phoretically transferred to a PVDF membrane (14). The blotswere incubated in 7.5% nonfat dry milk in Tris-buffered saline(TBS) overnight at 4 °C and then incubated for 1 h with poly-clonal anti-NHE-1 (1:500), polyclonal anti-NHE-2 (1:500) (16),polyclonal anti-NHE-3 (1:1000; Alpha Diagnostic Interna-tional, San Antonio, TX), polyclonal anti-NHE-5 (1:1000). Theblots were rinsed with TBS and incubated with horseradishperoxidase-conjugated secondary IgG for 1 h. Bound antibodywas visualized using an enhanced chemiluminescence assay(Amersham Biosciences).Immunofluorescence Staining—Cells grown on coverslips

were fixed in 4% paraformaldehyde in PBS for 15 min. Afterrinsing, cells were incubatedwith a blocking solution for 20minfollowed by application of a primary polyclonal antibody forNHE-1 (1:50; Abcam Inc., Cambridge, MA). After rinsing inPBS, cells were incubated with Alexa FluorTM 488 goat anti-rabbit IgG (1:200; Invitrogen) for 1 h. The coverslips were thencovered with Vectashield mounting medium (Vector Labora-tories, Burlingame, CA). Fluorescence images were captured bythe Nikon TE 300 inverted epifluorescence microscope (�40)using a Princeton Instruments MicroMax CCD camera andMetaMorph image-processing software.Statistics—Statistical significance was determined by Stu-

dent’s t test or an analysis of variance (Bonferroni post hoc test)in the case ofmultiple comparisons. A p value smaller than 0.05was considered statistically significant. n values represent thenumber of cultures in each experiment.

RESULTS

Surface Area/Volume (A/V) Ratio in Soma and Dendrites—In order to accurately calculate ionic flux rates in soma anddendrites, we first estimated surface area to volume (A/V)ratios in these cellular regions. Fig. 1A shows a single slice two-dimensional image of cultured neurons from a confocal stackimage (300 slices, 110�mthick). Thearrows in Fig. 1A illustratethe areas in the somata, Lg-dendrite, and Sm-dendrite wherethe A/V ratios and ionic changes were determined. Fig. 1B is athree-dimensional reconstruction of the stack of images withMetamorph software, highlighting neuronal morphology withthe distinctly higher A/V ratios in Lg- and Sm-dendrites. TheA/V ratio in Lg-dendrites was 3.8 times larger than somata (Fig.1C). An �7 times larger A/V ratio was estimated for Sm-den-drites. Interestingly, 2 h of OGD and 1 h of REOX did not

FIGURE 1. Surface area/volume ratios in soma and dendrites. A, a singletwo-dimensional (2-D) confocal image taken from a 300-slice image stack(110 �m thick, 512 � 512 pixels) of a neuron loaded with calcein/AM.Soma, Lg-dendrites, and Sm-dendrites are indicated (arrows). B, the 300-slice image stack was rendered into a three-dimensional (3-D) image usingMetamorph software to illustrate A/V ratios determined in soma, Lg-den-drites, and Sm-dendrites (arrowheads). C, A/V ratios were calculated inthree regions of normoxic control neurons or neurons subjected to 2 h ofOGD and 1 h of REOX. Data are mean � S.E. (error bars), n � 4. *, p 0.05versus soma. D, �i was determined in three regions under normoxia andfollowing 2-h OGD/1-h REOX. �i in each region was plotted against pHiand fit with a linear regression. The fitted slopes for each region were notsignificantly different under either normoxic (p � 0.97) or OGD/REOX (p �0.474) conditions.

Dendritic Damage

35158 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

significantly change the A/V ratio either in soma or indendrites.We then determined �i in three regions as shown in Fig. 1D.

�i in each region was plotted against pHi and fit with a linearregression. The slopes of the lines in the three regions were notsignificantly different under normoxic control or OGD/REOXconditions. These findings imply that the changes of pHi regu-lation may result from altered function of H� transporters,such as NHEs.Changes of pHi in Soma andDendrites followingOGD/REOX—

Lg-or Sm-dendrites exhibited more alkaline resting pHi valuesthan soma under normoxic conditions (Fig. 2A). Inhibition ofNHE-1 with its potent inhibitor HOE 642 (1 �M) or the newlydeveloped NHE-1 kinase p90RSK inhibitor fluoromethylketone(FMK; IC50 of 15 nM, (17)) acidified pHi and significantlydecreased pHi recovery rates (Fig. 2, A and B). OGD/REOXcaused an alkalization of pHi in soma (a shift from 6.96 � 0.03to 7.19 � 0.05; p 0.05). Inhibition of NHE-1 reversed the

OGD/REOX-mediated increase in pHi (Fig. 2C). Moreover,inhibition of theNHE-1 kinase p90RSKwith FMKprevented thepost-OGD alkalization. OGD/REOX did not trigger additionalchanges in pHi in dendrites. However, either HOE 642 or FMKsignificantly acidified dendrites following OGD/REOX. Thesedata suggest that NHE-1 activation plays a role in resting pHimaintenance and contributes to the intracellular post-OGDalkalization.Increased H� Efflux in Soma and Dendrites following OGD/

REOX—We further determined NHE-1 activity in soma anddendrites by measuring the pHi recovery rate following theNH3/NH4

� prepulse-induced acidification. As shown in Fig. 3A,when neurons were exposed to 30 mM NH3/NH4

�, pHi in Lg-dendrites rose rapidly as NH3 diffused into the cell and com-bined with H� to formNH4

� (a and b) and then declined slowly(b and c). Returning cells to the standard HCO3

�-free HEPES-EMEM solution caused pHi to decrease due to the rapid diffu-sion of NH3, which was dissociated from the newly formedNH4

�, and trappingH� inside the cells (c and d). Both normoxiccontrol and OGD/REOX-treated cells were able to restore pHito their basal levels (Fig. 3A). However, the pHi recovery rateincreased by �2-fold in Lg-dendrites following OGD/REOX(1.23 � 0.24 unit/min versus 0.57 � 0.05 unit/min in normoxicneurons, p 0.05).

pHi recovery rates were significantly higher in the Lg-den-drites (90%) and in the Sm-dendrites (330%) than the somaunder normoxic conditions (Fig. 3B). The apparent higher pHirecovery rates in the dendrites could result from the larger A/Vratios in the dendrites. Thus, we corrected the pHi recovery rateforA/V ratio in the three different regions. After the correction,the rates were similar in all three regions under normoxic con-trol conditions (Fig. 3C). 2-h OGD/1-h REOX triggered a fur-ther increase in the H� efflux in the soma (264%), the Lg-den-drites (218%), and the Sm-dendrites (69%; Fig. 3B). After thecorrection for the A/V ratio, the OGD/REOX-induced eleva-tion in pHi recovery rates remained significant in soma andLg-dendrites (Fig. 3C).This finding was further validated by calculating JH� (Fig.

3D). Dendrites exhibited smaller JH� than soma under nor-moxia and OGD/REOX conditions. The OGD/REOX-medi-ated selective stimulation of JH� persisted in soma and Lg-den-drites but not in Sm-dendrites. Similar changes of JH� wereobserved in the presence ofHCO3

� (21mM; Fig. 3D). The lack ofchanges in the A/V ratios and�i followingOGD/REOX suggestthat theOGD/REOX-induced stimulation of pHi recovery ratesmainly reflect JH�.Differential NHE-1 Activity in Soma and Dendrites—We

directly evaluated NHE-1-dependent pHi regulation activity inthe soma and the dendrites using theNHE-1 inhibitorHOE642(1 �M, IC50 of 0.08 �M) at a concentration that inhibits only theNHE-1 isoform (18). As shown in Fig. 4A, the OGD/REOX-mediated elevation of the H� extrusion rate in the somata wasnearly abolished in the presence of HOE 642. However, HOE642 only partially blocked the elevated pHi recovery rate in thedendrites (�50–60%). To determine the possible role of otherisoforms of NHE in neuronal processes, we examined theeffects of removing extracellular Na�, which inhibits the func-tion of all NHE isoforms by abolishing the inward Na� driving

FIGURE 2. A, resting pHi in the soma, Lg-dendrites, and Sm-dendrites undernormoxic conditions. In the drug treatment experiments, neurons wereexposed to either HOE 642 (1 �M) or FMK (3 �M) for 30 min prior to the pHidetermination. B, pHi regulation in the soma, Lg-dendrites, and Sm-dendritesunder normoxic conditions. Data are mean � S.E. (error bars), n � 3– 4. #, p 0.05 versus corresponding normoxia; *, p 0.05 versus corresponding soma.C, changes of pHi at 60 min of REOX following 2 h of OGD. In the drug treat-ment experiments, either HOE 642 (1 �M) or FMK (3 �M) was present onlyduring the 60-min REOX. Data are mean � S.E., n � 3– 4. *, p 0.05 versuscorresponding normoxia; #, p 0.05 versus OGD/REOX.

Dendritic Damage

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35159

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

force. In the absence of extracellular Na�, H� extrusion wasabsent in soma, similar to the NHE-1 inhibition via HOE 642(Fig. 4A). In the dendrites, the pHi recovery rate was eliminatedby �76–86%. Inhibiting all NHE isoforms with a general NHEinhibitor EIPA (100 �M) had a similar effect as removing extra-cellular Na�. Moreover, the residual Na�-independent H�

extrusion in the Sm-dendrites could be mediated by vacuolar

FIGURE 3. Increased H� efflux rate in somata and dendrites followingOGD/REOX. A, representative pHi changes in Lg-dendrites subjected to NH4

�/NH3 prepulse-mediated acid-loading. pHi recovery rate was determined byfitting a slope to the pHi values within the first minute following the prepulsein either normoxic or 2-h OGD/1-h REOX-treated neurons. pHi recovery rateswere determined at �6.2 to normalize for the allosteric regulation of H� onNHE-1 activity. B, summary data of pHi recovery rates under normoxic andOGD/REOX conditions. Data are mean � S.E. (error bars), n � 3– 4. *, p 0.05versus corresponding normoxia. #, p 0.05 versus corresponding soma. C, pHirecovery rates were corrected for the relative differences in surface area tovolume ratios in three regions. Data are mean � S.E., n � 3– 4. *, p 0.05versus corresponding normoxia; #, p 0.05 versus corresponding soma.D, proton flux (JH�) was calculated at pH �6.2 during pH recovery followingNH4

�/NH3 prepulse. Data are mean � S.E. n � 3– 4. *, p 0.05 versus corre-sponding normoxia; #, p 0.05 versus corresponding soma.

FIGURE 4. Differential NHE-1 activity in soma and dendrites. A, 2-h OGD/1-h REOX-treated neurons exhibited different pHi recovery rates followingthe NH4

�/NH3 prepulse. To inhibit NHE-1 activity, 1 �M HOE 642 was presentduring the 60-min REOX. In some prepulse studies, Na�-dependent H� extru-sion was blocked by replacing NaCl in the HEPES-buffered solution with anequimolar concentration of NMDG. Na�-free HEPES-EMEM, either with orwithout 1 �M bafilomycin, was used to evaluate function of other NHE iso-forms and vacuolar H�-ATPases. Inhibition of all NHE isoforms was examinedwith the general NHE inhibitor EIPA (100 �M). Data are mean � S.E (error bars),n � 3–5. *, p 0.05 versus soma under OGD/REOX; #, p 0.05 versus corre-sponding OGD/REOX. B, HOE 642-sensitive pHi regulation. OGD/REOX dataare from the experiments in Fig. 3B. Data are mean � S.E. n � 3–5. *, p 0.05versus soma under normoxia. #, p 0.05 versus corresponding normoxia.C, NHE-1, NHE-2, NHE-3, or NHE-5 protein expression in cultured cortical neu-rons. Cerebellar tissue was used as positive control for NHE-3, and blots wereprobed for �-tubulin as a loading control. D, expression of NHE-1 protein insoma (arrow), Lg-dendrite (arrowhead), and Sm-dendrite (open arrowhead).Negative Control, primary antibody was omitted. Scale bar, 10 �m.

Dendritic Damage

35160 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

H�-ATPase. Inhibition of vacuolar H�-ATPase with a specificinhibitor, bafilomycin (1 �M), abolished the residual H� extru-sion in both the Lg- and the Sm-dendrites (Fig. 4A). These dataimply that NHE-1 is the dominant isoform in soma. However,in the dendrites, pHi regulation is governed byNHE-1 as well asother NHE isoforms and H�-ATPases.

The HOE 642-sensitive portion of the H� extrusion rate wasobtained under normoxic control and OGD/REOX conditions(Fig. 4B). Consistently, NHE-1 activity (HOE 642-sensitive por-tion) in the soma and the Lg-dendrites was significantly ele-vated following OGD/REOX. Sm-dendrites exhibited a signifi-cantly higher basal activity of NHE-1, and OGD/REOX did notcause additional activation. Fig. 4C demonstrates that NHE-1 isthe dominant form in cortical neurons, whereas the NHE-3isoform is restricted to the cerebellum (19). Localization ofNHE-1 protein in soma and Lg- and Sm-dendrites was alsoshown in Fig. 4D.NHE-1-mediatedNa�Entry in SomaandDendrites following

OGD/REOX—The robust NHE-1 activation in the soma andthe dendrites following REOX led us to speculate that NHE-1plays a role in the dendritic Na�

i dysregulation following invitro ischemia. [Na�]i in the soma and the dendrites was mon-itored under normoxic controls and at 45 min REOX. Therewere no significant differences in the base-line [Na�]i betweenthe soma and the dendrites (Fig. 5A, arrowhead). A 45-minREOX following 2 hOGD led to an increase in [Na�]i through-

out the neuron. Localized increasesin [Na�]i were detected in the den-drites (Fig. 5A, arrow). Summarydata show that [Na�]i increasedfrom a resting level of 12.5 � 0.3 to44.4� 2.6mM in the soma (p 0.05;Fig. 5B). The Sm-dendrites exhib-ited the largest Na� accumulation(56.5 � 4.1 mM, p 0.05). Inhibi-tion ofNHE-1withHOE 642 duringREOX abolished the OGD/REOX-induced Na� overload in both thesoma and the dendrites. The rise inHOE 642-sensitive changes in[Na�]i was shown in Fig. 5C. Thesedata imply that OGD/REOX-medi-ated accumulation of [Na�]i islargely mediated via NHE-1activation.We recently reported that stimu-

lation of NHE-1 depends on activa-tion of the ERK-p90RSK signaltransduction pathways and phos-phorylation of NHE-1 (8). In thecurrent study, we examinedwhether direct inhibition of theNHE-1 kinase p90RSK with itspotent inhibitor BI-D1870 (BI-D,IC50 of 10–30 nM (20)) could reduceOGD/REOX-mediated NHE-1 acti-vation. The OGD/REOX-inducedNa�

i loading in Lg- and Sm-den-drites was abolished by BI-D1870 (Fig. 5B). This effect is similarto the one mediated by the NHE-1 inhibitor HOE 642. Thus,p90RSK function is largely responsible forNHE-1 activation andNa� entry. These data also imply that the initial increased intra-cellular acidification associated with OGD (8) is not sufficientto drive NHE-1 activity but requires altered phosphorylation ofthe transporter.We further compared the effects of BI-D along with FMK on

inhibition of NHE-1 activation. Both BI-D and FMK abolishedthe OGD/REOX-mediated stimulation of pHi recovery in thesoma and the dendrites (Fig. 5D). Especially in the Sm-den-drites, BI-D or FMK profoundly suppressed the pHi recoveryrate, which was only �25% of the normoxic basal levels (p 0.05; Fig. 5D). These data suggest that p90RSK pathways play amore dominant role in pHi regulation in the dendrites than inthe soma.Last, the pHi recovery rates in all three regions were deter-

mined in the presence of 21 mM bicarbonate under normoxiaandOGD/REOX conditions (Fig. 5E). The results are similar tothose in the absence of physiological bicarbonate. These datasuggest that the role for bicarbonate-dependent ion transport-ers in regulation of pHi is negligible under these conditions.

To further test NHE-1-mediated Na� overload, we exam-inedwhether there was a differential Na� overload in soma andSm-dendrites when Na�/K�-ATPase was blocked by ouabain(0.1 mM) during REOX. As shown in supplemental Fig. 1A,

FIGURE 5. NHE-1-mediated Na� entry in soma and dendrites following OGD/REOX. A, representative SBFIpseudocolored images of changes in [Na�]i in normoxic and OGD/REOX-treated neurons. Arrowhead, low levelsof [Na�]i; arrow, localized increases in [Na�]i. B, summary data of [Na�]i in soma, Lg-dendrites, and Sm-den-drites of neurons under normoxic conditions and 2-h OGD/45-min REOX. Data are mean � S.E. (error bars), n �3–5. *, p 0.05 versus normoxia; #, p 0.05 versus OGD/REOX; �, p 0.05 versus soma OGD/REOX. C, HOE642-sensitive change in [Na�]i. Data are mean � S.E., n � 3–5. *, p 0.05 versus corresponding normoxia; #, p 0.05 versus soma under OGD/REOX. D, pHi recovery rates were determined. Normoxia and OGD/REOX data arefrom the experiments in Fig. 3B. BI-D1870 (1 �M), FMK (3 �M), or HOE 642 (1 �M) was present only during REOX.Data are mean � S.E., n � 3. *, p 0.05 versus normoxia; #, p 0.05 versus OGD/REOX; �, p 0.05 versus somaOGD/REOX; �, p 0.05 versus Sm-dendrite normoxia. E, pHi recovery rates were determined in the presence of21 mM HCO3

�. HOE 642 (1 �M) was present only during REOX. Data are mean � S.D., n � 4 – 8 cells. *, p 0.05versus normoxia; #, p 0.05 versus OGD/REOX.

Dendritic Damage

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35161

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Na�i clearly loaded faster in Sm-dendrites than in soma under

these conditions. Moreover, at 30 min of REOX, blockingNHE-1 activity decreased the Na� influx in Sm-dendrites (sup-plemental Fig. 1B). A similar trend was also observed in Lg-dendrites (supplemental Table 1).Delayed Dysregulation of Ca2� Depends on a Concurrent

Activation of NHE-1 and NCXrev—One consequence of Na�i

overload is to triggerNCXrev andCa2� entry. To investigate thepossible concerted activation of NHE-1 and NCXrev, we firstmonitored changes of local [Ca2�]i following REOX. Thesomata [Ca2�]i was 68 � 9 nM under normoxic control condi-tions and increased modestly to 104 � 8 nM after 2 h of OGD(p 0.05), which remained unchanged over the initial 35-minREOX (Fig. 6,A andB). In contrast, 2 h ofOGDcaused a slightlyhigher elevation in the dendritic [Ca2�]i (157 � 21 nM, p 0.05). Twenty min of REOX triggered a secondary rise in thedendritic [Ca2�]i that initiated from local “hot spots” and thenspread toward the soma over time (Fig. 6, A and C). The ampli-tude of the dendritic Ca2� dysregulation was significantlyhigher than the soma. By 45min of REOX, the dendritic [Ca2�]irose dramatically to 1206 � 440 nM and spread to the soma,which exhibited slightly lower levels (766 � 300 nM). Sustainedelevation in [Ca2�]i during REOX (60–100min) resulted in celldeath, as reflected by a sudden loss of the dye (data not shown).We speculated that NCXrev contributes to this Ca2� dys-

regulation as a result of the robust NHE-1 activation and Na�

overload. First, we investigated whether inhibition of NHE-1activity with HOE 642 could block the delayed rise in [Ca2�]i.As shown in Fig. 6 (B and C), when HOE 642 was present onlyduring 60 min of REOX, no secondary rise in [Ca2�]i occurredin the soma and the dendrites following REOX. These dataimply that NHE-1 activation is involved in the secondary loss ofCa2�

i homeostasis during REOX. To establish whether thisCa2� rise results from activation of NCXrev, we conducted theexperiments in the presence of SEA0400 (1�M), a potent inhib-itor ofNCXrev. As shown in Fig. 6,B andC, REOX failed to elicitthe secondary elevation in [Ca2�]i in both the soma and thedendrites. The results were similar to those of the HOE 642-treated cells. This led us to conclude that concerted activationof NHE-1 and NCXrev contributed to the delayed Ca2� dys-regulation in soma and dendrites following in vitro ischemia.However, these inhibitors did not affect basal levels of [Na�]iand [Ca2�]i under normoxic control conditions (supplementalTable 2).Changes of Ca2�

m and �m in Soma and Dendrites followingOGD/REOX—Excessive dendriticNa�

i andCa2�ioverloadwill

affect mitochondrial Ca2� homeostasis and mitochondrialfunction. To investigate this, we first monitored changes ofCa2�

m in the soma and the dendrites during 0–60 min REOX.There was a slow and progressive elevation in the somataCa2�

m starting by 10 min REOX and reaching a plateau value(�2.5-fold control) by 40minREOX (Fig. 7A). Interestingly, thesomata �m did not decrease significantly, whereas Ca2�

m wasincreasing during early REOX. However, �m depolarized sig-nificantly after Ca2�

m reached its plateau levels, and it wasreduced to 47 � 6% of control at 60 min of REOX (Fig. 7A).By 10min of REOX, Ca2�

m levels had increased significantlyin the Sm-dendrites but not in the Lg-dendrites (Fig. 7, B and

C). The rate of Ca2�m increase in the Sm-dendrites during early

REOX was significantly faster than in the soma (0.018 versus0.003 relative change/min, p 0.05). At 60 min REOX, Ca2�

mwas increased by �4.5-fold in the Sm-dendrites and �3.2-fold

FIGURE 6. Changes in [Ca2�]i in soma and dendrites following OGD/REOX.A, representative fura-2 pseudocolored images of changes in [Ca2�]i in neu-rons at 0, 20, 30, 40, and 45 min REOX. Arrowhead, low levels of [Ca2�]i. Arrow,localized increases in [Ca2�]i. B and C, summarized changes of [Ca2�]i in soma(B) or Sm-dendrites (C). In the HOE 642 or SEA0400 studies, the drugs (1 �M)were present only during 1 h of REOX. Data are mean � S.E. (error bars), n � 4.*, p 0.05 versus 0 min REOX; #, p 0.05 HOE versus non-treated OGD/REOX;�, p 0.05 HOE or SEA versus non-treated OGD/REOX.

Dendritic Damage

35162 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

in the Lg-dendrites. Ca2� accumulation in the mitochondriaremained elevated throughout the neuron until �100 min,when a sudden loss in the rhod-2 dye signal occurred, probablydue to a collapse of mitochondrial function (data not shown).The earlier rise in Ca2�

m in the dendrites was accompaniedby a faster decrease of �m (Fig. 7, B andC). �m in the dendrites(particularly in the Sm-dendrites) depolarized at a rate twicethat of the somata (1.2 versus 0.7%/min). �m in the Sm-den-drites dropped to the lowest level (17� 4%of control) by 40minof REOX. The kinetics of the �m collapse in the dendritesexhibited a significant negative correlation with the dendriticCa2�

m accumulation (Pearson product moment correlation

coefficient � �0.964, p 0.001). Thus, compared with thesoma, the dendrites show two characteristics: earlier onset timeand larger magnitude in the loss of mitochondrial Ca2� home-ostasis and �m. This demonstrates that the dendritic mito-chondria are more sensitive to OGD/REOX damage than thesoma. These changes are consistent with the earlier loss of Na�

and Ca2� homeostasis in the dendrites.Interestingly, inhibition of NHE-1 activity with 1 �M HOE

642 prevented the REOX-mediated changes of Ca2�m and �m

in soma. In the presence of 1 �M HOE 642, there were no sig-nificant increases in Ca2�

m in the Lg-dendrites and the somata.A slow accumulation in Ca2�

m was detected in the Sm-den-drites, which was not statistically significant from 0min REOX.Moreover, the delayed depolarization of the somata �m during50–60 min of REOX was absent with the HOE 642 treatment(Fig. 7A). Strong protective effects of HOE 642 on�mwere alsofound in the Lg- and the Sm-dendrites. The small dendritic�mdepolarized to�44% of control (instead of 17% of control) at 60min of REOX when NHE-1 activity was inhibited (p 0.05).Thus, preservation of�m by NHE-1 inhibitionmay result fromdecreased mitochondrial Ca2� loading.To determine the role of NHE-1 in mitochondrial dysfunc-

tion followingOGD/REOX, we examined whether FMK (3�M)would prevent mitochondrial damage in the soma and the den-drites. As shown in Fig. 8,A andB, inhibition of theNHE kinasep90RSK during REOX attenuated loss of �m following OGD/REOX in the soma and the Lg-dendrites. In the Sm-dendrites,FMK effectively prevented depolarization of �m as early as 10min of REOX (Fig. 8C). Interestingly, these effects were similarto the direct inhibition of NHE-1mediated by HOE 642. Takentogether, we can firmly conclude that blocking either NHE-1 orp90RSK significantly preserves mitochondrial function in ische-mic neurons.In a parallel study, in order to confirm the reliability of the

�m determination with JC-1, we also used the cationic dyeTMRE to monitor changes of �m at 0 and 60 min of REOX(supplemental Fig. 2, A and B). Both JC-1 and TMREmeasure-ments indicated that the REOX-induced decrease in �m wasmore profound in the Lg- and the Sm-dendrites. Additionally,inhibition of NHE-1 with HOE 642 during REOX significantlyattenuated theOGD/REOX-induceddecrease in�m in all areasof the cells (p 0.05; supplemental Fig. 2B).Role of Mitochondrial Uniporter in Mitochondrial Ca2�

Accumulation—To investigate the role of the uniporter inmitochondrial Ca2� loading following OGD/REOX, we deter-mined changes of Ca2�

m when the mitochondrial uniporterwas inhibited with 10 �M RU360 during 60 min of REOX. Inhi-bition of the uniporter prevented accumulation of Ca2�

m in themitochondria in all regions of the neuron (Fig. 9).Dendritic Damage Is Reduced by Inhibition of NHE-1 follow-

ing OGD/REOX—We further investigated dendritic damage bymonitoring the dendritic beading (varicosities) and membraneintegrity changes followingOGD/REOX.As shown in Fig. 10,Aand B, 2 h of OGD caused some swelling in the dendrites with-out formations of varicosities (arrow). Varicosities developed inthe dendrites over 30–60min REOX (Fig. 10C, arrowhead). At60min of REOX, the bead density increased by13 times com-paredwith the normoxic neurons (Fig. 10,C and E). In contrast,

FIGURE 7. Changes in Ca2�m and �m in soma and dendrites following

OGD/REOX. Changes in Ca2�m and �m were monitored in soma (A), Lg-den-

drites (B), and Sm-dendrites (C). Blue lines, �m data determined with JC-1 andexpressed as the percentage of the maximal FCCP-induced change in the JC-1ratio of normoxic controls. Red lines, Ca2�

m determined by the relativechange in rhod-2 fluorescence. In the HOE study, 1 �M HOE 642 was presentonly during 0 –1 h REOX. Data are mean � S.E. (error bars), n � 3– 4. *, p 0.05versus 0 min REOX; #, p 0.05 versus OGD/REOX.

Dendritic Damage

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35163

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

in the presence of 1 �M HOE 642 during REOX, the bead for-mation increased only by�3 times (Fig. 10,D and E, arrow; p0.05). These data suggest that inhibition of NHE-1 activity withHOE 642 not only reduces the dendritic ionic dysregulation butalso decreases dendritic swelling following OGD/REOX.

DISCUSSION

Robust NHE-1 Activity in Dendrites—In the current study,we characterized NHE-1 activity in the somata and the Lg- and

the Sm-dendrites under normoxic and OGD/REOX condi-tions. We observed that pHi regulation in the Lg- and the Sm-dendrites was �90–300% faster than in the somata under nor-moxic conditions. The differential pHi regulation rates betweenthe dendrites and the soma were abolished when they werecorrected by the differences in the A/V ratios. Therefore, thedata illustrate that the dendrites can change pHi more rapidlythan the soma due to the small cytosolic volume comparedwithits surface area. Robust NHE activity has previously beendetected in the hippocampal nerve terminals following intra-cellular acidification under normoxic conditions (21). In our

FIGURE 8. Effects of FMK on changes of �m in soma and dendrites follow-ing OGD/REOX. Changes in �m were determined with JC-1 and expressed asthe percentage of the maximal FCCP-induced change in the JC-1 ratio ofnormoxic controls in soma (A), Lg-dendrites (B), and Sm-dendrites (C). In theFMK study, 3 �M FMK was present during 0 –1 h REOX. Data are mean � S.E.(error bars), n � 4. *, p 0.05 versus 0 min REOX; #, p 0.05 versus OGD/REOX.

FIGURE 9. Effects of mitochondrial uniporter inhibitor RU360 on Ca2�m in

soma and dendrites following OGD/REOX. Ca2�m was assessed by the rel-

ative change in rhod-2 fluorescence when the mitochondrial uniporter inhib-itor, RU360 (10 �M), was present during 60-min REOX in soma (A), Lg-den-drites (B), and Sm-dendrites (C). Data were plotted against Ca2�

m data fromFig. 7 for OGD/REOX and OGD/REOX � HOE. Data are mean � S.E. (error bars),n � 3– 4. *, p 0.05 versus 0 min REOX; #, p 0.05 versus OGD/REOX.

Dendritic Damage

35164 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

study, following OGD/REOX, the Na�-dependent H� extru-sion activity was further elevated in the soma (264%) and theLg-dendrites (218%), whereas the A/V ratios remainedunchanged.TheH� extrusionmechanisms in the soma and the dendrites

have not been well defined. In this study, we concluded that thesomata pHi regulation under HCO3

�-free conditions is exclu-sively mediated by NHE-1 activity, which is consistent with ourprevious findings using bothHOE 642-mediated pharmacolog-ical inhibition andNHE-1 genetic knockdown approaches (11).Moreover, the significance of NHE-1 in neuronal ionic regula-tion is further highlighted by the abundant expression ofNHE-1 compared with NHE-2, NHE-3, or NHE-5 in the neu-rons. This is consistent with the earlier reports on the preemi-nence of both NHE-1 mRNA and protein expression in brainsover the isoforms NHE-2 to -4 and the abundance of NHE-3 inthe cerebellum (19, 22).We report here that NHE-1 plays a dominant role in the

regulation of dendritic pHi (60%). The remaining pHi regula-tion in the dendrites depended on the functions of the lessabundant NHE isoforms (NHE-2 and NHE-5) and vacuolarH�-ATPases. The H� pumps are highly expressed in the vesi-

cles of synaptic terminals and responsible for acid loading andaccumulation of neurotransmitters (23). Although the H�

pumps are typically expressed inmembranes of organelles, theyhave been detected in the plasma membrane of hippocampalastrocytes and are active in pHi regulation under Na�- andHCO3

�-free conditions (24, 25). Our findings of H� pumpactivity in the dendritic plasma membrane suggest that den-drites are equipped withmultiple H� extrusionmechanisms tocounteract the robust Ca2�-dependent intracellular acidifica-tion during synchronous neural activity (26).Recently, spatial nonuniformity in pHi has been reported in

the proximal and distal dendrites of oligodendrocytes (27). Thealkaline microdomains in the perikaryon and proximal den-drites of the oligodendrocytes were attributed to localizedincreases in NHE activity, whereas the acidic pHi in the distaldendrites may be the result of Na�/HCO3

� cotransporter-me-diated HCO3

� extrusion (27). Thus, the pHi microdomain andregulatory mechanisms in the oligodendrocyte distal dendritesappear to be different from those in the Sm-dendrites of neu-rons. These findings suggest that different cell types expressdifferent pHi-regulating mechanisms in regulating microdo-main pHi.Lack of NHE-1 Activation in Sm-dendrites following OGD/

REOX—The basal level of NHE-1-mediated H� extrusion washigh in the Sm-dendrites compared with soma. OGD/REOXdid not further stimulate it. This suggests that, given their largeA/V ratio and their higher basal JH�, Sm-dendrites are able tomaintain pHi without significant further elevation of NHEactivity. This may also be a reflection of the sophisticated pHiregulatory mechanisms expressed in Sm-dendrites, preventingoverstimulation of H� extrusion. On the other hand, this Sm-dendritic JH� phenomenon may be unique under the circum-stances of culturemodels in plastic plates but not characteristicof dendrites under in vivo conditions. Indeed, we have observedmuch higher A/V ratios and faster pHi recovery rates in neu-rites grown in microfluidic devices, a model that mimics theslow diffusion and convection of the in vivomicroenvironment(28). It remains to be investigated whether OGD/REOX affectsNHE-1 function differently in the Sm-dendrites using themicrofluidic device model.p90RSK-mediated Stimulation of NHE-1 Activity following

OGD/REOX—In the current study, OGD/REOX-mediatedstimulation of NHE-1 activity was abolished in both the somaand the dendrites when NHE-1 kinase p90RSK was inhibited byits potent inhibitors BI-D1870 and FMK. Activation of p90RSKand NHE-1 phosphorylation is downstream of the ERK1/2 sig-naling pathways (8). FMK is a novel, specific inhibitor forp90RSK isoforms 1 and 2 (17). FMKblocks the�1-adrenocepter-mediated NHE-1 phosphorylation and stimulation in rat ven-tricular myocytes (29). On the other hand, BI-D1870 has beenshown to be a potent ATP-competitive inhibitor of all p90RSKisoforms (20). In this study, we found that FMK and BI-D1870were equally effective in inhibiting OGD/REOX-mediatedNHE-1 activation, implying a role for p90RSK isoforms 1 and 2.Moreover, the p90RSK inhibitors reduced the NHE-1 activity tobelow the base-line levels. This suggests that p90RSK is alsoinvolved in the regulation of basal NHE-1 activity in all threeregions.

FIGURE 10. Dendritic beading in neurons following OGD/REOX. A–D, den-dritic beading formation in DiO-loaded cells was detected at 0, 30, 45, and 60min of REOX. In the HOE studies, 1 �M HOE 642 was present only during 0 –1 hof REOX. Arrow, swelling in dendrites without formations of varicosities;arrowhead, varicosities. E, summary data of dendritic beading. The data ofnormoxia control and normoxia plus HOE 642 (red) groups overlap. Data aremean � S.E. (error bars), n � 3– 4. *, p 0.05 versus 0 min REOX. #, p 0.05versus OGD/REOX.

Dendritic Damage

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35165

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

NHE-1-mediated Na� Entry following OGD/REOX—Dis-ruption of dendritic ionic homeostasis occurs during early cer-ebral ischemia and may play a role in irreversible dendriticdamage. Excessive Na� influx via ionotropic glutamate recep-tors or tetrodotoxin-sensitive Na� channels leads to neuronaldeath under excitotoxic or hypoxic conditions (4, 30, 31). How-ever, subsequent studies have suggested that hypoxia-inducedNa� influx could be through pathways other than ionotropicglutamate receptors or tetrodotoxin-sensitive channels (32). Incultured hippocampal neurons, Na� entry immediately afteranoxia results from activation of NHE and a Gd3�-sensitivepathway (33). Recent reports demonstrate that dendritic dam-age following brief in vivo ischemia (34) or axonal morpholog-ical changes following in vitro hypoxia (35) are independent ofionotropic glutamate receptor activation. In the present study,we observed that OGD/REOX triggered an �3-fold increase in[Na�]i (�50mM). TheNa� accumulationwas eliminatedwhenNHE-1 activity was inhibited by its inhibitor HOE 642 or thep90RSK kinase inhibitor BI-D1870. Thus, the elevated NHEactivity in the dendrites not only accelerates pHi recovery afterOGD/REOX but also intensifies disruption of Na� ionic home-ostasis and causes dendritic vulnerability to ischemic damage.These findings also suggest that Na�/K�-ATPase function isnot sufficient to maintain Na�

i homeostasis following OGD/REOXwhen there is an increase inNa� influx. BlockingNHE-1activity would decrease the need for Na� extrusion via Na�/K�-ATPase and preserve cellular ATP levels (Fig. 11). Thisimbalance between Na� extrusion via Na�/K�-ATPase andNHE-1-mediated Na� influx can also have a significant impacton [Na�]i in Sm-dendrites, particularly because of their largeA/V ratio and high basal JH�, even without further elevation ofNHE-1 activity following OGD/REOX. Taken together, weconclude that theOGD/REOX-mediated stimulation ofNHE-1plays a dominant role in dendritic Na� overload.

We failed to detect elevation of NHE-1-mediated H� extru-sion in the Sm-dendrites after OGD/REOX. The causes for thediscrepancy between the NHE-1-mediated H� extrusion andNa� overload in the Sm-dendrites are not apparent. One pos-sible explanation is that a subtle increase inNHE-1 activitymaynot be detected with the instantaneous measurement of H�

extrusion (dpHi/dt), whereas its impact on Na� overload overtime (at a steady-state level) can be revealed. This speculation isalso supported by HOE 642-sensitive effects on mitochondrialdysfunction and Ca2� dysregulation. Future study is needed tofurther address this issue.NHE-1-dependent Changes in Dendritic Ca2� following

OGD/REOX—Activation of NHE activity in hippocampalnerve terminals following intracellular acidification is accom-panied with an elevation in [Na�]i and [Ca2�]i as well asincreased postsynaptic currents (21). The authors attributethese changes to the concurrent activation of NHE andNCXrevin the nerve terminals (21). These findings suggest that NHEand NCXrev could play a coordinating role in the regulation ofdendritic Na� and Ca2� homeostasis and affect Ca2�-depen-dent release of neurotransmitters. In dendrites and dendriticspines close to postsynaptic localities, all three isoforms ofNCX(NCX-1 to -3) are preferentially expressed, suggesting a role forNCX in Ca2� signaling at the excitatory postsynaptic sites (36).

In this study, 2-h OGD triggered a moderate elevation indendritic Ca2�

i, but, during 60 min of REOX, a delayed accel-erated Ca2�

i rise occurred. The OGD/REOX-induced Ca2�

deregulationwas initiated in the dendrites and then propagatedto the soma. Interestingly, the secondary Ca2�

i deregulationcan be prevented when either NHE-1 activity was inhibited byHOE 642 or NCXrev was blocked by SEA0400. These findingsimply that a coupled NHE-1 and NCXrev function is a majorcontributor to Ca2�

i deregulation in the dendrites of culturedcortical neurons. Moreover, we believe that NCX-1 is the dom-inant NCX isoform in this study because SEA0400 has a highaffinity against the reverse mode function of NCX-1 (IC50 �50nM) as compared with NCX-2 (IC50 �1 �M) and is ineffectiveagainst either NCX-3 or NCKX-2 (37).The role of NCXrev in dendritic Ca2�

i deregulation has alsobeen examined during sustained NMDA exposure. DelayedCa2� deregulation in CA1 neurons of acute hippocampal slicesdepended on Na� loading but was not prevented by the non-specific NCX inhibitor KB-R7943 (38). However, when Na�

loading is potentiated with low levels of ouabain (30 �M),NMDA can trigger secondary Ca2� deregulation in dendrites,which is completely blocked by KB-R7943 (39). These findingsfurther suggest that activation of NCXrev requires excessiveNa� loading. Our previous thermodynamic analysis predictsthat NCXrev occurs when [Na�]i is elevated to �20 mM in cor-tical neurons at a resting membrane potential of �60 mV (40).The vulnerability of neuronal dendrites is characterized by

the initial membrane depolarization, mitochondrial structurecollapse, and dendritic beading in the dendrites and the subse-quent propagation toward the soma during hypoxia and activa-tion of NMDA receptors (1, 6). Our current study illustratesthat blocking of NHE-1 activity attenuated many similarchanges in the dendrites following OGD/REOX. Therefore,concerted activation of NHE-1 andNCXrev may also play a role

FIGURE 11. Illustration of dendritic ionic disruption and mitochondrialdysregulation in ischemic neurons. Following ischemia, activation ofNHE-1 causes an increase in dendritic [Na�]i, which triggers NCXrev and leadsto increases in [Ca2�]i. The [Na�]i overload also causes increased consump-tion of ATP by Na�/K�-ATPase to maintain dendritic ionic homeostasis. Onthe other hand, the [Ca2�]i overload stimulates Ca2�

m uptake by theuniporter (UP) and formation of the mitochondrial permeability transitionpore (PTP). Blocking NHE-1 and NCXrev would reduce disruption of Na� andCa2� homeostasis and preserve mitochondrial bioenergetics and dendriticmembrane integrity.

Dendritic Damage

35166 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

in dendritic injury in conditions such as glutamate-mediatedneurotoxicity, epilepsy, etc.Changes of Dendritic �m and Ca2�

m following OGD/REOX—In the current study, depolarization of �m in the Sm-dendritesoccurred 40 min earlier than the soma following OGD/REOX.The loss of �m in the dendrites closely correlated with theCa2�

m accumulation. Mitochondria are capable of sequester-ing large amounts of Ca2� under various pathological condi-tions (41). Increases in free mitochondrial Ca2� would occurwhen Ca2� entry into themitochondria exceeds the capacity ofmitochondrial Ca2� extrusion and the mitochondrial robustphosphate buffering system (42). The Ca2�

m accumulationreported in this study probably reflectsCa2� entry via a voltage-dependentCa2�uniporter before the collapse of�m (41). Smallincreases in Ca2�

m stimulate Ca2�-dependent dehydrogenasesand mitochondrial metabolism, but massive Ca2� loading ofmitochondria leads to depolarization of �m (6, 42). Sustainedloss of �mwill eventually trigger the opening of the permeabil-ity transition pore and release of Ca2�

m and collapse of mito-chondria bioenergetics (6). In the current study, most of theCa2�

m accumulation occurred before�m decreased below 50%and remained at a sustained level when �m was �45% of con-trol in the soma and�20% of control in the Sm-dendrites. Thisimplies that a residual level of�m (20%) for a short period (�40min) is sufficient to maintain high mitochondrial Ca2� levels.However, when low�m and high Ca2�

mwere extended past 60min, there was a sudden loss of residual�m that coincided withthe loss of the rhod-2 signal, suggesting a release of Ca2�

m fromthe permeability transition pore under these conditions.Interestingly, when either NHE-1, p90RSK, or NCXrev was

inhibited during 60-min REOX, loss of�m andCa2�m accumu-

lation in dendrites was significantly reduced. This finding isconsistent with the earlier reports on NHE-1 inhibition-in-duced attenuation of the mitochondrial Ca2� overload andmitochondrial permeability transition pore opening in car-diomyocytes and in ischemic/reperfused rat hearts (43–45).Taken together, these studies demonstrate a conserved role oftheNHE-1 signalingmechanism in ischemic reperfusion injuryamongmultiple cell types.Moreover, it has been suggested thatNHE-1 inhibitors, including HOE 642, have a direct effect onreactive oxygen species production and mitochondrial perme-ability transition pore formation (13). In the current study, it isunknown whether any protective effects mediated by HOE 642result from its direct actions on mitochondria. We speculatethat such a possibility is low in light of the similar protectiveeffects offered by inhibition of p90RSK or NCXrev in thisstudy as well as the protection observed in NHE-1 transgenicknock-out neurons (11). Moreover, in general, mild acidosiscan inhibit neurotransmission, whereas alkaline pHi stimu-lates excitability. Thus, we cannot rule out that HOE 642may protect neurons in part via directly correcting NHE-1-mediated alkalinization.In summary (Fig. 11), the current study reports that NHE-1-

mediated Na� entry and subsequent stimulation of NCXrevactivity contribute to the selective vulnerability of dendritesfollowing in vitro ischemia. A newly emerging hypothesis spec-ulates that dendritic Na� overload and the subsequent activa-tion of Na�/K�-ATPase would consume more ATP and fur-

ther collapse mitochondrial biogenesis (6). However, to date,themechanisms underlying the excessive Na� influx andmito-chondrial dysfunction are not well defined. Our current studydemonstrates that activation of NHE-1 in dendrites presents amajor pathway for Na� overload. NHE-1 inhibition preventsNa� accumulation, which is required for dendritic beading.Blocking NHE-1 function also attenuates loss of the dendritic�m and Ca2�

m homeostasis and preserves mitochondrialbioenergetics and dendritic membrane integrity.

Acknowledgments—We thank Dr. Eugene B. Chang for providing theNHE-1 and NHE-2 antibodies and for helpful discussions. We alsothank Dr. Yong Wang for assistance with the immunoblotting assays.

REFERENCES1. Greenwood, S. M., and Connolly, C. N. (2007) Neuropharmacology 53,

891–8982. Hasbani, M. J., Viquez, N.M., and Goldberg, M. P. (2001)Neuroreport 12,

2731–27353. Park, J. S., Bateman, M. C., and Goldberg, M. P. (1996) Neurobiol. Dis. 3,

215–2274. Hasbani, M. J., Hyrc, K. L., Faddis, B. T., Romano, C., and Goldberg, M. P.

(1998) Exp. Neurol. 154, 241–2585. Vander Jagt, T. A., Connor, J. A., and Shuttleworth, C. W. (2008) J. Neu-

rosci. 28, 5029–50396. Greenwood, S. M., Mizielinska, S. M., Frenguelli, B. G., Harvey, J., and

Connolly, C. N. (2007) J. Biol. Chem. 282, 26235–262447. Avkiran, M. (2001) Basic Res. Cardiol. 96, 306–3118. Luo, J., Kintner, D. B., Shull, G. E., and Sun, D. (2007) J. Biol. Chem. 282,

28274–282849. Karmazyn, M., Sawyer, M., and Fliegel, L. (2005) Curr. Drug Targets Car-

diovasc. Haematol. Disord. 5, 323–33510. Kintner, D. B., Su, G., Lenart, B., Ballard, A. J.,Meyer, J.W., Ng, L. L., Shull,

G. E., and Sun, D. (2004) Am. J. Physiol. Cell Physiol. 287, C12–C2111. Luo, J., Chen, H., Kintner, D. B., Shull, G. E., and Sun, D. (2005) J. Neurosci.

25, 11256–1126812. Friedman, J. E., and Haddad, G. G. (1994) Brain Res. 641, 57–6413. Garciarena, C. D., Caldiz, C. I., Correa, M. V., Schinella, G. R., Mosca,

S. M., Chiappe de Cingolani, G. E., Cingolani, H. E., and Ennis, I. L. (2008)J. Appl. Physiol. 105, 1706–1713

14. Kintner, D. B., Luo, J., Gerdts, J., Ballard, A. J., Shull, G. E., and Sun, D.(2007) Am. J. Physiol. Cell Physiol. 292, C1113–C1122

15. Nicholls, D. G. (2006) J. Biol. Chem. 281, 14864–1487416. Musch,M.W., Bookstein, C., Rocha, F., Lucioni, A., Ren,H., Daniel, J., Xie,

Y., McSwine, R. L., Rao, M. C., Alverdy, J., and Chang, E. B. (2002) Am. J.Physiol. Gastrointest. Liver Physiol. 283, G975–G985

17. Cohen,M. S., Zhang, C., Shokat, K.M., andTaunton, J. (2005) Science 308,1318–1321

18. Kleeman, H. W., and Weichert, A. G. (1999) IDrugs 2, 1009–102519. Ma, E., and Haddad, G. G. (1997) Neuroscience 79, 591–60320. Sapkota, G. P., Cummings, L., Newell, F. S., Armstrong, C., Bain, J., Frodin,

M., Grauert, M., Hoffmann, M., Schnapp, G., Steegmaier, M., Cohen, P.,and Alessi, D. R. (2007) Biochem. J. 401, 29–38

21. Trudeau, L. E., Parpura, V., and Haydon, P. G. (1999) J. Neurophysiol. 81,2627–2635

22. Douglas, R. M., Schmitt, B. M., Xia, Y., Bevensee, M. O., Biemesderfer, D.,Boron, W. F., and Haddad, G. G. (2001) Neuroscience 102, 217–228

23. Moriyama, Y., Maeda, M., and Futai, M. (1990) J. Biochem. 108, 689–69324. Philippe, J. M., Dubois, J. M., Rouzaire-Dubois, B., Cartron, P. F., Vallette,

F., and Morel, N. (2002) Glia 37, 365–37325. Pappas, C. A., and Ransom, B. R. (1993) Glia 9, 280–29126. Makani, S., and Chesler, M. (2007) J. Neurosci. 27, 7438–744627. Ro, H. A., and Carson, J. H. (2004) J. Biol. Chem. 279, 37115–3712328. Vitzthum, L., Chen, X., Kintner, D. B., Huang, Y., Chiu, S. Y., Williams, J.,

and Sun, D. (2010) Integr. Biol. 2, 58–64

Dendritic Damage

NOVEMBER 5, 2010 • VOLUME 285 • NUMBER 45 JOURNAL OF BIOLOGICAL CHEMISTRY 35167

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

29. Cuello, F., Snabaitis, A. K., Cohen, M. S., Taunton, J., and Avkiran, M.(2007)Mol. Pharmacol. 71, 799–806

30. Choi, D.W.,Maulucci-Gedde,M., and Kriegstein, A. R. (1987) J. Neurosci.7, 357–368

31. Raley-Susman, K. M., Kass, I. S., Cottrell, J. E., Newman, R. B., Chambers,G., and Wang, J. (2001) J. Neurophysiol. 86, 2715–2726

32. Muller, M., and Somjen, G. G. (2000) J. Neurophysiol. 84, 1869–188033. Sheldon, C., Diarra, A., Cheng, Y.M., andChurch, J. (2004) J. Neurosci. 24,

11057–1106934. Murphy, T. H., Li, P., Betts, K., and Liu, R. (2008) J. Neurosci. 28,

1756–177235. Underhill, S. M., and Goldberg, M. P. (2007) Neurobiol. Dis. 25, 284–29036. Minelli, A., Castaldo, P., Gobbi, P., Salucci, S., Magi, S., and Amoroso, S.

(2007) Cell Calcium 41, 221–23437. Iwamoto, T., Kita, S., Uehara, A., Imanaga, I., Matsuda, T., Baba, A., and

Katsuragi, T. (2004) J. Biol. Chem. 279, 7544–7553

38. Shuttleworth, C. W., and Connor, J. A. (2001) J. Neurosci. 21,4225–4236

39. Dietz, R.M., Kiedrowski, L., and Shuttleworth, C.W. (2007)Hippocampus17, 1049–1059

40. Chen, X., Kintner, D. B., Luo, J., Baba, A., Matsuda, T., and Sun, D. (2008)J. Neurochem. 106, 1563–1576

41. Spat, A., Szanda, G., Csordas, G., and Hajnoczky, G. (2008) Cell Calcium44, 51–63

42. Szabadkai, G., and Duchen, M. R. (2008) Physiology 23, 84–9443. Teshima, Y., Akao,M., Jones, S. P., andMarban, E. (2003)Circulation 108,

2275–228144. Toda, T., Kadono, T., Hoshiai,M., Eguchi, Y., Nakazawa, S., Nakazawa,H.,

Higashijima, N., and Ishida, H. (2007) Am. J. Physiol. Heart Circ. Physiol.293, H3517–H3523

45. Javadov, S., Huang, C., Kirshenbaum, L., and Karmazyn, M. (2005) J. Mol.Cell Cardiol. 38, 135–143

Dendritic Damage

35168 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 45 • NOVEMBER 5, 2010

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Taunton and Dandan SunAkemichi Baba, Toshio Matsuda, Mike Cohen, John Orlowski, Shing-Yan Chiu, Jack

Douglas B. Kintner, Xinzhi Chen, Julia Currie, Vishal Chanana, Peter Ferrazzano, Ischemiain VitroHomeostasis and Mitochondrial Dysfunction following

2+ and Ca+ Exchange in Disruption of Dendritic Na+/H+Excessive Na

doi: 10.1074/jbc.M110.101212 originally published online September 3, 20102010, 285:35155-35168.J. Biol. Chem. 

  10.1074/jbc.M110.101212Access the most updated version of this article at doi:

 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/2010/09/03/M110.101212.DC1

  http://www.jbc.org/content/285/45/35155.full.html#ref-list-1

This article cites 45 references, 16 of which can be accessed free at

by guest on May 19, 2020

http://ww

w.jbc.org/

Dow

nloaded from