Caspase Inhibitor z-DEVD-fmk Attenuates Calpain and NecroticCell Death in Vitro and After Traumatic Brain Injury

*Susan M. Knoblach, *Daniel A. Alroy, *Maria Nikolaeva, *Ibolja Cernak, *Bogdan A. Stoica,and *†‡Alan I. Faden

*Departments of Neuroscience, †Pharmacology, and ‡Neurology, Georgetown University Medical Center, Washington, DC

Summary: In studies designed to evaluate the therapeuticwindow for treatment of traumatic brain injury, the caspase 3inhibitor z-DEVD-fmk improved neurologic function and re-duced lesion volumes when administered at 1 but not at 4, 8, or24 hours after injury. Moreover, neither caspase 3 nor PARP, acaspase 3 substrate, were cleaved in injured, untreated cortexfrom 1 to 72 hours after injury. Few cortical neurons expressedactive caspase 3 or were TUNEL positive from 6 to 24 hoursafter injury, and TUNEL staining was primarily Type I (ne-crotic). Nissl staining revealed extensive neuronal necrosis inthe injured cortex from 6 to 24 hours after impact. Consideredtogether, these data suggested that z-DEVD-fmk may reduceneuronal necrosis, so we used an in vitro model of necrotic celldeath induced by maitotoxin to test this further and explore the

potential mechanism(s) involved. Z-DEVD-fmk (1 nM-100�M) significantly attenuated maitotoxin induced neuronal celldeath and markedly reduced expression of the 145 kD calpain-mediated �-spectrin breakdown product after maitotoxin in-jury. Neither the 120 kD caspase-mediated �-spectrin cleavageproduct nor cathepsin B were expressed after maitotoxin injury.In a cell free assay, z-DEVD-fmk reduced hydrolysis of caseinby purified calpain I. Finally, z-DEVD-fmk reduced expressionof the 145 kD calpain-mediated �-spectrin cleavage fragmentafter traumatic brain injury in vivo. These data suggest thatneuroprotection by z-DEVD-fmk may, in part, reflect inhibi-tion of calpain-related necrotic cell death. Key Words: Trau-matic brain injury—Calpain—Caspase—Necrosis—Maitotox-in—Apoptosis.

Traumatic brain injury (TBI) may cause either ne-crotic or apoptotic cell death. Apoptosis and necrosishave been classically distinguished by morphologic cri-teria (Clarke, 1990; Kerr et al., 1972). They have alsobeen defined biochemically, with apoptosis often consid-ered an active process that requires energy and proteinsynthesis and necrosis viewed as a passive process re-flecting loss of ionic homeostasis secondary to energydepletion (Ankarcrona et al., 1995; Pang et al., 2003;Volbracht et al., 1999). However, hybrid forms of celldeath showing mixed apoptotic/necrotic features haverecently been identified, and some have coined the termaponecrosis to reflect a continuum between these pheno-types (Formigli et al., 2000; Kitanaka and Kuchino, 1999).

Studies suggest that both the type and temporal courseof cell death after central nervous system (CNS) injurydepends upon a variety of factors, including type andseverity of the insult, as well as age, species, and sex ofthe subject (Kupina et al., 2003; Martin et al., 1998; Pohlet al., 1999; Prins and Hovda, 2003). Apoptotic neuronaldeath has been documented after head injury in humans,and in several of the most widely used models of TBI,including injury induced by lateral fluid-percussion(LFP) or controlled cortical impact (CCI) (Colicos et al.,1996; Conti et al., 1998; Fox et al., 1998; Newcombet al., 1999; Rink et al., 1995). After LFP, DNA frag-mentation (as detected by TUNEL staining) was ob-served in neurons with either “classic” apoptotic (TypeII) or necrotic (Type I) features (Rink et al., 1995). Suchneuronal apoptosis occurred both in the injured cortex, aswell as in adjacent regions (thalamus and hippocampus),at later time points (Conti et al., 1998; Rink et al., 1995).After CCI, TUNEL-labeled neurons were Type II orType I, and localized to the ipsilateral cortex (Fox et al.,1998; Newcomb et al., 1999), or to cortex and hippo-campus, depending upon injury severity (Colicos et al.,1996; Fox et al., 1998; Newcomb et al., 1999).

Received February 23, 2004; final version received May 11, 2004;accepted June 21, 2004.

This work was supported by NIH R01 NS36527 and Department ofDefense Contract DAMD17-99-2-9007 to A.I.F.

Address correspondence and reprint requests to Dr. Knoblach,Georgetown University Medical Center, Department of Neuroscience,EP16 New Research Building, 3970 Reservoir Rd. NW, Washington,DC, 20057; e-mail: sknobl01@georgetown.edu

Journal of Cerebral Blood Flow & Metabolism24:1119–1132 © 2004 The International Society for Cerebral Blood Flow and MetabolismPublished by Lippincott Williams & Wilkins, Baltimore

1119 DOI: 10.1097/01.WCB.0000138664.17682.32

Calpain and caspase 3 cysteine proteases have re-ceived considerable attention with regards to their poten-tial role(s) in the pathophysiologic response to CNS in-jury. Both calpain and caspase 3 are activated afterhuman head injury and after experimental TBI, whereinhibitors of these proteases have also been shown toimprove neurologic outcome and reduce tissue degrada-tion (Beer et al., 2000; Buki et al., 1999, 2000; Clarket al., 1999, 2000; Harter et al., 2001; Kampfl et al.,1996, 1997; Keane et al., 2001; Knoblach et al., 2002;Kupina et al., 2001; McCracken et al., 1999; Posmanturet al., 1997; Saatman et al., 1996a,b; Yakovlev et al.,1997, 2001a; Zhao et al., 1998).

Active caspase 3 has been colocalized with positiveTUNEL staining and DNA fragmentation in neuronalsoma after CCI or LFP and is also increased within theaxonal compartment in regions of diffuse injury associ-ated with LFP (Beer et al., 2000; Buki et al., 2000; Clarket al., 2000; Knoblach et al., 2002; Yakovlev et al., 1997,2001a). These studies, together with others showing ac-tivation of caspase-3 after cerebral hypoxia or ischemia,support a role for caspase-dependent apoptosis in acuteneurodegeneration (Blomgren et al., 2001; Chen et al.,1998; Fink et al., 1998).

Z-DEVD-fmk is a tetrapeptide caspase inhibitor thatis considered relatively selective for caspase-3 (Garcia-Calvo et al., 1998; Thornberry et al., 1997) and has beenwidely used in in vitro and in vivo models of acute injuryto delineate roles for caspase 3 in neuronal cell death.Intracerebroventricular injections of z-DEVD-fmk im-proved function after LFP (Yakovlev et al., 1997). Intra-parenchymal infusion of z-DEVD-fmk over several daysafter combined CCI and hypoxia reduced lesion size,although functional outcome was not significantly im-proved in this model (Clark et al., 2000).

Because we have previously described necrosis, apo-ptosis, and caspase 3 activation after CCI in the mouse(Clark et al., 1997; Fox et al., 1998; Newcomb et al.,1999), the present study evaluated the therapeutic win-dow for z-DEVD-fmk treatment in this model. Z-DEVD-fmk was administered intracerebroventricularly, ratherthan intraparenchymally, to avoid the possibility of ob-taining a local treatment effect and because in previousexperiments intracerebroventricular injections improvedbehavioral outcomes (Knoblach et al., 2002; Yakovlevet al., 1997). Surprisingly, we observed substantial neu-roprotective effects only after very early administraion(1 hour), at a time in which little apoptosis or caspaseactivation was present, and neuronal cell death was pri-marily necrotic. Additional in vivo and in vitro stud-ies demonstrated that z-DEVD-fmk was a potent inhibi-tor of calpain and that improvement observed after treat-ment with z-DEVD-fmk may reflect, at least in part,this action.

MATERIALS AND METHODS

CCI injury and administration of z-DEVD-fmkMale C57Bl/6 mice (20–25 g) (Taconic Farms, NY, U.S.A.)

were maintained at constant temperature (22 ± 2°C) and a 12hour light/dark cycle with food and water available ad libitumfor 1 week before procedures. Details on the injury device, CCIinjury model, and behavioral tests were previously reported(Fox et al., 1998). Briefly, animals were anesthetized with iso-flurane in oxygen and placed on a heating pad to maintain corebody temperature at 38 ± 0.2°C. A 4-mm craniotomy wasperformed on the central aspect of the left parietal bone, keep-ing the dura mater intact. The dura mater was moistened withsterile saline (37.5°C) and injury was induced (6.0 m/s velocity,1 mm tissue deformation) with a pneumatic injury device. Afterinjury, the incision was closed, anesthesia was discontinued,and mice were placed in a heated cage to maintain normother-mia for 45 minutes. Animals were then monitored for 4 hoursafter surgery and daily thereafter. For treatment with z-DEVD-fmk or vehicle after CCI, mice were reanesthetized with iso-flurane at various times after injury, placed in a stereotaxicapparatus, and the CCI wound was reopened for intracerebro-ventricular injection. Either Z-DEVD-fmk (160 ng in 2 �LDMSO) (ICN Biomedicals, Aurora, OH, U.S.A.), or DMSOvehicle (Sigma, St. Louis, MO, U.S.A.) was injected over a5-minute period.

Live animal procedures were approved by the GeorgetownUniversity Animal Care and Use Committee and performedaccording to principles detailed in the Guide for the Careand Use of Laboratory Animals (DHEW publication NIH85-23-2985).

Evaluation of motor and cognitive functionRecovery of motor function was evaluated using a beam

walking task (Fox et al., 1998). To perform the task, micetraversed a narrow wooden beam (6 mm wide, 120 mm long)suspended 30 cm above a thick foam rubber pad. The numberof footfaults (slips) made by the right hindlimb was recordedover 50 steps counted in either direction on the beam. A basallevel of competence (<10 faults over 50 steps) was establishedbefore surgery. Cognitive function (spatial learning) was as-sessed using a Morris water maze paradigm. To perform themaze, mice were required to locate a resting platform hidden ata fixed location within a pool of opaque water, using extramazevisual cues. Mice experienced four trials per day, each 30 min-utes apart, for 4 consecutive days. For each trial, mice wereplaced into the water at one of four randomly chosen locations,separated by 90°. The length of time to find the platform (la-tency) was recorded for each mouse. After 90 seconds, micethat failed to find the platform were assisted to it. Animals wereallowed to remain on the platform for 15 seconds on the firsttrial and 10 seconds on subsequent trials. After completion ofall trials, a black, clearly visible platform was placed into thepool at a new location. Mice were then reintroduced to thewater tank, and the latency to locate the visible platform wasrecorded over three trials. This test was included as a control toverify the visuomotor status of injured mice. All behavioraldata were expressed as means ± SEM, for each group.

StudiesFor the study with behavioral/magnetic resonance imaging

(MRI) outcomes, mice were subjected to CCI and then treatedwith z-DEVD-fmk either 1 hour (n � 19), 4 hours (n � 20),8 hours (n � 16), or 24 hours (n � 16) later. Controls weretreated with equivolume DMSO (n � 20) 1 hour after CCI. Forthe study to evaluate �-spectrin breakdown after z-DEVD-fmk

S. KNOBLACH ET AL.1120

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

treatment, animals were injured and then treated with eitherz-DEVD-fmk (n � 4) or vehicle (n � 3) 1 hour later asdescribed above. Naive (n � 4) and sham-injured animals(n � 2) were also included. To assess expression of activecaspase 3 subunits or PARP cleavage via immunoblot, un-treated mice were killed at 1, 6, 15, 24, or 72 hours after injury(n � 5 for all groups except 72 hours, where n � 2), and theirbrains were removed and dissected on ice. Control corticalextracts were obtained from naive (n � 2) and sham-injuredanimals (n � 3). For the histologic studies, animals were sub-jected to CCI without treatment and then killed at 6, 12, 24, and48 hours (n � 3/group) after injury. Sham-injured animalsserved as controls (n � 2). Study numbers were based onprevious experience with each method and expected variability.

T2-weighted MRI to determine lesion volumeWe have previously shown that the T2-weighted MRI signal

days after TBI significantly correlates with histologic assess-ments of lesion volume taken from the same animals (Albensiet al., 2000; Faden et al., 2003). In addition, the protocol forT2-weighted imaging after TBI has been detailed (Albensiet al., 2000). Briefly, 21 days after TBI, animals were anesthe-tized with isoflurane (5% for induction, 1.5% for maintainence,in 70%/30% nitrous/oxygen) and placed in a 72 mm 1H bird-cage resonator on a 37°C heating pad in a plexiglass MR probe.The probe was inserted into the bore of a horizontal 7T/21 cmBiospec-Avance magnet (Bruker, Germany) and positionedso that the animal’s head was in the center of the magnet. Arepiratory monitor was used for respiratory gating to reducemotion artifact. Field homogeneity across the brain wasoptimized and a sagittal scout image acquired (RARE image,FOV � 4 × 4 cm, 128 × 128 resolution, TR/TE � 1,500/10 mswith a rare factor of 8 making the effective TE � 40 ms). Eightcontiguous multi-slice T2-weighted images were then acquiredbeginning from the end of the olfactory bulb and extending tothe posterior base of the brain (FOV � 3 × 3 cm, slice thick-ness � 2 mm, 128 × 128 resolution, TR/TE � 1,500/20 ms,4 echo images, and 2 averages). Lesion volume was estimatedfrom the summation of areas of hyperintensity on each slice,multiplied by slice thickness, for the ipsilateral and contralat-eral hemispheres. Data are expressed as mean ± SEM (�L)lesion volume for each group.

Immunoblot AnalysesCortical tissue samples were homogenized in RIPA buffer

(60 mM Tris-HCl, pH 7.8, containing 150 mM NaCl, 5 mMEDTA, 10% glycerol, 2 mM Na3VO4, 25 mM NaF, 10 �g/mLleupeptin (Sigma), 10 �g/mL aprotinin (Sigma), 1 mM AEBSF(Sigma), 1 mM Pepstatin (Sigma), 0.1% SDS (Gibco BRL),0.5% Deoxycholic Acid (Sigma) and 1% Triton X-100 (Cal-biochem, La Jolla, CA, U.S.A.), or in extraction buffer K240-100 (BioVision, Mountainview, CA, U.S.A.). The supernatentwas removed, aliquoted, and frozen at −80°C. Tissue culturesamples were pooled (15 wells/group), and homogenized inHepes buffer (10 mM Hepes/KOH, pH 7.4, 3 mM EDTA, 1%Chaps, 5 mM DTT, 10 �g/mL aprotinin, 1 mM AEBSF). Thesupernatant was removed, aliquoted and frozen at −80°C. Ali-quots from either mouse cortex or tissue culture were thawed asneeded, and the protein concentration in each determined witha Micro-BCA protein assay kit (Pierce), according to manufac-turer’s instructions. For caspase 3, poly-(ADP ribose) polymer-ase (PARP), and �-spectrin breakdown products, equal proteinaliquots (50–100 mg) were resolved on 16% or 4% Tris-glycinegels in a 25 mM TRIS/192 mM Glycine/ 0.01% SDS, pH 8.3running buffer. Proteins were transferred to Hybond-C Super

nitrocellulose membranes (Amersham, Arlington Heights, IL,U.S.A.) in a 20% solution of methanol in running buffer,blocked for 16 to 24 hours in blocking buffer (5% BSA or 5%powdered milk /0.02% Tween in phosphate buffered saline[PBS]), and then probed with specific primary antibodies asfollows: a) rabbit polyclonal antibody specific for the 17 and 19kDa fragments of active caspase-3 (#9661) from Cell SignalingTechnology (New England Biolabs, Beverly, MA, U.S.A.);b) rabbit polyclonal antibody specific for the 85 kDa caspase-mediated cleavage fragment of PARP (containing the carboxy-terminus) (sc-7150) from Santa Cruz Biotechnology; c) mousemonoclonal antibody to �-fodrin (�II-spectrin) from AffinitiResearch Products Ltd. (Exeter, U.K.); or d) rabbit polyclonalantibody to cathepsin B (Upstate Cell Signaling Solutions,Lake Placid, NY, U.S.A.). Immune complexes were detectedusing horseradish peroxidase-linked secondary antibodies(Amersham), chemiluminescence reagent (Super Signal West-Dura, Pierce, Rockford, IL, U.S.A.), and Kodak Biomax MR-1film (Sigma). Immunoblots were scanned with a GS-710 Im-aging Densitometer (Bio-Rad, Hercules, CA, U.S.A.) and semi-quantitative analysis was performed with Image Quant soft-ware. Data were expressed as the percentage of control(naive/sham-injury) relative density units.

Histologic proceduresMice were anesthetized and transcardially perfused with sa-

line and 4% paraformaldehyde. Their brains were removed,stored in fresh 4% paraformaldehyde overnight, protected insucrose, frozen in O.C.T. media, sectioned (40 �m), and storedfree-floating in 2% paraformaldehyde. Later, sections wereplaced in 4% paraformaldehyde for 5 minutes, rinsed in PBSand incubated with 10% goat serum/0.3% triton for 1 hour atroom temperature. Primary antibody specific for the active sub-unit of caspase 3 (Cell Signaling, Beverly, MA) was diluted1:100 in 2% goat serum/0.2% triton/PBS and applied overnightat 4°C. For cell-type specific double-labeling experiments, an-tineuronal nuclear protein (NeuN) (Chemicon, Temecula, CA,U.S.A.) was diluted 1:100 in solution with anti-caspase primaryantibody. The next day, sections were rinsed twice in PBS andincubated with secondary antibodies (1:50) for 1 hour at roomtemperature. Goat anti-rabbit FITC (Sigma) and goat anti-mouseTexas Red (Accurate Chemicals, Westbury, NY, U.S.A.) wereused as secondary antibodies. After incubation with secondaryantibody, sections were rinsed twice in cold PBS, mounted onglass slides, coverslipped with antifade mounting medium, andvisualized with an Olympus IX-70 laser confocal scanning mi-croscope. Standardized parameters were used to allow validcomparisons between experimental groups. For negative con-trols, either primary or secondary antibody was omitted. Insections colabeled with TUNEL, Tdt reaction mix was appliedafter primary antibodies were rinsed off. Tdt reaction mix con-tained biotinylated dNTPs, Tdt buffer, and Tdt enzyme (Gibco,Gaithersburg, MD, U.S.A.) in ultrapure H2O. After 1 hour(37°C), slides were rinsed in 0.1 M EDTA, pH 8.0 for 2 to5 minutes, PBS briefly, and then incubated with secon-dary antibody plus DN-Avidin-FITC (Vector Labs, Burlin-game, CA, U.S.A.) for 1 hour at room temperature. Finally,sections were rinsed in PBS, coverslipped, and viewed as de-scribed previously in this report. For cresyl violet staining,sections were dehydrated in ethanol and chloroform, stainedwith cresyl violet (10 minutes), rinsed and cleared in ethanoland xylenes, coverslipped with Permount (Fisher Scientific,Hanover Park, IL, U.S.A.), and viewed with an Olympus BH-2light microscope.

Z-DEVD-FMK REDUCES NECROSIS AND INHIBITS CALPAIN 1121

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

Neuronal glial cocultures and maitotoxin injuryThe preparation of neuronal glial cocultures has previously

been detailed (Mukhin et al., 1998). Briefly, cortices from 1- to5-day-old Sprague-Dawley (Taconic, Germantown, NY,U.S.A.) rat were freed from the meninges and triturated inmagnesium and calcium free Hank’s Balanced Salt Solutionsupplemented with 1% 1 M HEPES (pH 7.0) and 1% 100 mMsodium pyruvate. Dissociated cells were seeded on 96-well cell+ plates (Sarstedt, Newton, NC, U.S.A.) in minimal essentialmedium (MEM) with Earle’s salts that contained 10% each offetal bovine and equine serums, 25 mM HEPES (pH 7.2), 2mM glutamine, 20 mM glucose, 10 ng/mL murine epidermalgrowth factor (EGF), and 1% antibiotic/antimycotic. At 9 daysin vitro (DIV), cortices of day 17 to 19 gestation rat embryoswere dissociated and seeded onto plates in media as describedpreviously in this report, except that serums were reduced to5%, and no EGF was added. Cultures were fed twice weeklythereafter until used for experiments on 18 to 21 DIV. Cytosinearabinoside (3 mM) was added at the first feeding, and bovineserum was deleted from all feeding media. After 11 DIV, all

serum was excluded from the feeding medium. For maitotoxininjury, cultures were washed with MEM plus 10 mM HEPES(pH 7.2–7.4), 1 mM glutamine, and 20 mM glucose, and thenpreincubated with drugs for 15 minutes before addition of mai-totoxin. Maitotoxin (Alexis Biochemicals, San Diego, CA,U.S.A.) was added to a final concentration of 0.1 nM. One hourlater, cultures were washed, and fresh media, with or withoutdrugs, was added as appropriate. At various times after injury,as described for each experiment, media was removed for lac-tate dehydrogenase (LDH) measurement using a CytoTox 96cytotoxicity assay kit (Promega, Madison, WI, U.S.A.). Tostain nuclei, Hoechst 33258 was added to a final concentrationof 0.2 ng/�L, and nuclear morphology was visualized with aNikon TE3000 inverted fluorescent microscope. Some cultureswere washed and then incubated with 0.3 �M staurosporine for16 hours to serve as positive controls for apoptotic morphology.Data were expressed as the percentage of LDH release in treatedcultures compared to that induced by injury alone, with LDHfrom uninjured controls subtracted from all values. Bars rep-resent the means ± SEM for n � 12 to 20 wells per condition.

FIG. 1. Early z-DEVD-fmk treatment improved mo-tor and cognitive function after traumatic CNS in-jury induced by severe controlled cortical impact(CCI) in the mouse. Mice were subjected to CCIand then treated with z-DEVD-fmk (160 ng, intra-cerebroventricularly) at 1 hour (n = 19), 4 hours(n = 20), 8 hours (n = 16), or 24 hours (n = 16) afterinjury. Controls received equivolume DMSO ve-hicle at 1 hour (n = 20) after injury. (A) Motor func-tion: footfaults (missteps) made by injured micewere counted as they traversed a narrow beam 1,2, 3, 7, 14, or 21 days after injury as indicated onthe x-axis. Dots represent the mean ± SEM numberof footfaults for each group as shown in the legend.The 0 indicates footfaults made by all mice beforeinjury. *P < 0.05 (at 7 and 14 days) or *P < 0.01 (at21 days) for 1 hour treatment vs. control. +P < 0.05for 4 hours treatment vs. control (at 21 days). (B)Cognitive function: animals experienced four train-ing trials 30 minutes apart on a Morris water mazetest on days 14 to 17 after injury as indicated on thex-axis. Dots represent the mean ± SEM time re-quired to find a hidden platform (goal latency) av-eraged over four trials on each day. *P < 0.05 for1 hour treatment vs. control. Data were analyzedby two-way ANOVA with time or group as Factorsfollowed by post hoc Tukey’s test.

S. KNOBLACH ET AL.1122

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

Cell-free digestion assayLyophilized BODIPY TR-X conjugated casein was recon-

stituted in 100 mM NaHCO3, pH 8.3 to a 10 �g/mL stock solu-tion (4°C). The stock was diluted 1:200 in Tris-HCL buffer, pH7.8 that contained 1 U of purified calpain I with or withoutz-DEVD-fmk (0.01–100 �M). Triplicates of each conditionwere incubated in the dark for 1 hour at room temperature, andfluorescence was measured with a Cytofluor Series 4000 fluo-rometer (Perseptive Biosystems, Framingham, MA, U.S.A.) us-ing ex/em of 590/645 nm. The assay was performed twice, anddata from one assay were expressed as the percentage of cal-pain activity observed in the absence of z-DEVD-fmk.

Data analysisBeam walking and Morris water maze data were analyzed by

two-way ANOVA with time or group as factors, followed bypost hoc Tukey’s pairwise comparison of each time of treat-ment versus control. MRI data were analyzed by one-wayANOVA, followed by post hoc Tukey’s pairwise comparisonof each treatment time versus control. LDH and casein diges-tion data were analyzed by one-way ANOVA, followed by posthoc t-tests of each concentration versus control, with Bonfer-roni correction for multiple comparisons. Immunoblot data (%)were subjected to square root transformation and then analyzedby t-test with Bonferroni correction. Individual data pointsgreater than 3 standard errors from the mean were consideredoutliers and not included for statistical analysis. For all statis-tics, P < 0.05 was considered statistically significant.

RESULTS

Early treatment with z-DEVD-fmk improvedneurologic function and reduced lesion volume

Z-DEVD-fmk (160 ng) was injected intracerebroven-tricularly at either 1, 4, 8, or 24 hours after CCI. Toassess motor recovery, mice were tested for the ability to

FIG. 2. Early z-DEVD-fmk treatment reduced lesion volume aftertraumatic CNS injury induced by severe controlled cortical impact(CCI) in the mouse. (A) Lesion volume of the injured/ipsilateralcortex at 21 days after CCI as assessed by T2 weighted MRI.Bars represent the mean lesion volume for each z-DEVD-fmktreatment group, as shown on the x-axis. *P < 0.05 vs. control.Data were analyzed by 1-way ANOVA followed by post hocTukey’s test. (B) Representative MRI scans from rats treatedeither with vehicle or z-DEVD-fmk, as indicated. (C) Photomicro-graph of a cresyl violet stained section showing a representativelesion from an untreated mouse.

FIG. 3. Cleavage of caspase 3 (A), or of poly-ADP-ribose-polymerase (PARP) was not detected by immunoblot analysis ofthe injured cortex at 1, 6, 12, 24, or 72 hours after CCI. Proteinextracts from sham control or traumatized mouse cortex at indi-cated times after TBI were subjected to SDS-PAGE and immu-noblot analysis. In (A) the positive control is from cleaved ratrecombinant caspase 3. A representative �-actin control for pro-cedures is shown directly below. In (B), the positive control for the89 kDa fragment of cleaved PARP is from Fas-stimulated Jurkatcells. Representative blot is from n = 5 per group, except 72hours, where n = 2.

Z-DEVD-FMK REDUCES NECROSIS AND INHIBITS CALPAIN 1123

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

traverse a narrow, suspended beam during recovery overa 21-day period (Fig. 1A). Mice treated 1 hour after CCIperformed significantly better than did vehicle controlson days 7, 14, and 21 after injury. Mice treated 4 hoursafter CCI performed significantly better than controlsonly on day 21 after injury, but this was an isolatedobservation, as they did not show a trend toward betterperformance compared with other treatment groups onany other testing day. Groups treated at other times didnot differ significantly from controls on any testing day.Cognitive recovery was assessed with a Morris watermaze (MWM) test. Mice treated 1 hour after CCI per-formed significantly better than vehicle controls on thethird and fourth days of the MWM test (days 16 and 17after injury) (Fig. 1B). Groups treated at other times didnot differ significantly from controls. At the end of the21-day recovery period, mice were evaluated via T2-

weighted MRI to determine the volume of the lesion inthe hemisphere ipsilateral to impact. Treatment at 1 hourafter CCI significantly reduced lesion volume, in com-parision with controls (Figs. 2A and 2B). Groups treatedat other times were not significantly different than controls.

Limited caspase 3 activation in injured cortexand marked neurodegeneration

A separate group of animals was concurrently pre-pared to evaluate caspase-3 activity and apoptotic celldeath after CCI. Neither cleaved caspase 3 subunits norcleaved PARP fragments were detected in immunoblotassays of injured cortical homogenates from these animalsat 1, 5, 12, 24, or 72 hours after CCI (Figs. 3A and 3B).

Because immunoblotting procedures may lack the sen-sitivity to detect limited expression in selected cells, ad-ditional studies used cell-type specific double-labeling

FIG. 4. Active caspase 3 was not widely expressed in cortical neurons at 6 or 12 hours after CCI. Double-label immunocytochemicalstaining for anti-active caspase 3 (FITC/green), anti-neuronal nuclear protein (NeuN) (Texas Red/red), or merged images of both proteinsare shown. Antiactive caspase 3 positive cells were thinly scattered throughout the injured cortex. Some of these were neuronal, asindicated by coexpression of NeuN. Anti–NeuN-positive cells were distinctly absent from some areas of injured cortex at 6 hours afterinjury (arrows on 6 hour images). By 12 hours after injury, these regions had expanded throughout the contused cortex. Images arerepresentative of data obtained from n = 3 animals per timepoint.

S. KNOBLACH ET AL.1124

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

methods to qualitatively assess whether caspase 3 acti-vation and DNA fragmentation via TUNEL labeling. Afew cells in the injured cortex and hippocampus (dentategyrus) expressed active caspase 3 at 6, 12, and 24 hoursafter injury (Figs. 4 and 5) (data for 24 hours not shown).Some of these cells were colabeled with anti-NeuN, in-dicating that they were neurons (Figs. 4 and 5). TUNELlabeling had a similar pattern, although TUNEL-positivecells were more frequently colabeled with NeuN thanwere anti-active caspase 3 positive cells (Fig. 6A).TUNEL labeling was predominately Type I (necrotic)(Rink et al., 1995), as very few TUNEL-positive cellsshowed any evidence of nuclear condensation or frag-mentation. There was less NeuN staining in the injuredcortex, compared with the contralateral side. Patchyclumps of reduced staining were evident as early as 6hours after injury and had expanded to include most ofthe ipsilateral cortex by 12 hours after injury.

To characterize degeneration that was indirectly evi-dent as reduced staining with anti-NeuN, alternate sec-tions were stained with cresyl violet and visualized bylight microscopy. Using this method, marked neurode-generation was observed thoughout the cortex 6 hoursafter CCI (Fig. 6B), as well as at 12 or 24 hours afterinjury. Most remaining cells were swollen with con-

densed nuclei (necrotic), but some cells with reducedcytoplasmic volumes and punctate or fragmented nuclei(apoptotic) were also present.

Z-DEVD-fmk reduces cell death and inhibits calpainin a model of in vitro necrosis and a cell free assay

Maitotoxin evokes necrotic cell death in neurons andglia through calcium influx and activation of calpain(Gusovsky and Daly, 1990; Legrand and Bagnis, 1984).Nevertheless, z-DEVD-fmk (1 nM to 100 �M) was neu-roprotective in this model, as it both reduced LDH re-lease and attenuated morphologic evidence of neuronalcell death (Figs. 7A–7D). Staining with Hoechst 33258confirmed that cells injured with maitotoxin had roundednuclei and showed no indication of DNA fragmentationor apoptotic bodies (Figs. 7E and 7F). Calpain inhibitorsE64D, ALLN, and calpastatin were used as positive con-trols and were neuroprotective in this model, as well(data not shown). Calpain and caspase cleave �-spectrinat unique sites, generating enzyme-specific �-spectrinbreakdown products (SBDs) that indicate calpain orcaspase 3 activation (Vanderklish and Bahr, 2000).Therefore, additional cultures were injured with maito-toxin in the presence or absence of z-DEVD-fmk, andLDH release, active caspase 3 fragments, and SBDs were

FIG. 5. Active caspase 3 was expressed in the dentate gyrus of the hippocampus at 24 (A, B) and 48 (C, D) hours after CCI. Double-labelimmunocytochemical staining for anti-active caspase 3 (FITC/green) (B, D) or anti-neuronal nuclear protein (NeuN) (Texas Red/red)(A, C) are shown. Note positive staining for active caspase 3 in areas where NeuN staining is reduced. Images are representative of dataobtained from n = 3 animals per timepoint.

Z-DEVD-FMK REDUCES NECROSIS AND INHIBITS CALPAIN 1125

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

assessed over time. Initial elevations in LDH releasewere observed 1 and 3 hours after maitotoxin injury andwere not altered by z-DEVD-fmk (Fig. 8A). However,LDH release at 6, 9, and 12 hours after maitotoxin wassubstantially higher than that observed initially, and thisdelayed release was significantly reduced in the presenceof z-DEVD-fmk. The calpain-mediated 145 kD SBD in-creased from 1 to 12 hours after maitotoxin injury butwas abolished or reduced at all times in the presence ofz-DEVD-fmk (Fig. 8B). In contrast, the caspase-medi-ated 120 kDa SBD was not increased after maitotoxinnor altered by z-DEVD-fmk treatment. In agreementwith this observation, caspase 3 cleavage fragmentscould not be detected at any time after maitotoxin injury(Fig. 8C). However, analysis of PARP cleavage or ofcaspase 3 activation at the cellular level wasnot performed.

Z-DEVD-fmk has recently been shown to inhibit ca-thepsin B (Schotte et al., 1999). Therefore, expression ofthe active form of this enzyme was evaluated, but it was

not increased by maitotoxin-induced cell death, nor al-tered by z-DEVD-fmk (Fig. 8D).

A cell-free casein digestion assay was also used toassess inhibition of calpain by z-DEVD-fmk directly. Inthis experiment, calpain I activity was reduced more than95% by 100 or 10 uM z-DEVD-fmk, and more than50% by 1 uM z-DEVD-fmk (Fig. 9). Z-DEVD-fmk didnot inhibit calpain at concentrations less than or equal to0.01 uM.

Z-DEVD-fmk treatment inhibitscalpain activity after TBI in vivo

To test whether z-DEVD-fmk may inhibit calpain ac-tivity after TBI in vivo, mice were injected with z-DEVD-fmk or vehicle 1 hour after CCI, as in the originalexperiment which determined the optimal therapeuticwindow for treatment with z-DEVD-fmk, and killed1 hour later. Expression of SBD in the injured cortex wasthen analyzed by immunoblot. The 145 kDa SBD wassubstantially increased after injury in vehicle controls,

FIG. 6. Few TUNEL labeled neurons were observed in the injured cortex, despite cell loss. (A) Colocalization of TUNEL labeling(FITC/green) and anti-NeuN staining (Texas Red/red) in injured mouse cortex 24 hours after CCI. Images shown represent the outerinjured cortex or contralateral, uninjured side, as indicated. In injured cortex, approximately 30% of anti-NeuN positive cells were alsoTUNEL positive (merged frame). In most of these cells TUNEL staining was diffuse and cytoplasmic (Type I), rather than condensed orpunctate/fragmented (Type II). There were many more anti-NeuN stained cells in the contralateral (uninjured) cortex than in injured cortex.(B) Nissl stain of injured mouse cortex at 6 and 24 hours after injury. At low magnification, patchy areas of extensive cell loss werevisualized at 6 and 24 hours after injury. High magnification of these areas revealed degenerating neurons with necrotic (wide arrows)and apoptotic (narrow arrow) features.

S. KNOBLACH ET AL.1126

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

but such expression was completely abolished in animalstreated with z-DEVD-fmk (Figs. 10A and 10B). In con-trast, injury had no effect upon expression of the caspase-mediated 120 kDa SBD band, indicating that no caspaseactivation was present at this time.

DISCUSSION

The present studies were designed to evaluate thetherapeutic window for z-DEVD-fmk treatment in amouse model of CCI. We found that z-DEVD-fmk im-proved neurologic function and reduced tissue damage atan injury severity that showed predominantly necroticneuronal cell death with minimal evidence of caspase 3activation. The therapeutic window for treatment withz-DEVD-fmk in our model was less than 4 hours andthus most consistent with an antinecrotic mechanism.This conclusion was supported by both biochemical and

histologic outcome measures: active caspase 3 subunitsor caspase 3-mediated cleavage fragments of PARP wereundetectable in immunoblots of cortical extracts takenfrom 1 to 72 hours after injury; TUNEL staining wasprimarily Type I (necrotic); active caspase 3 expressionwas limited in the injured cortex and adjacent hippocam-pus; and Nissl staining of the injured cortex revealedprimarily necrotic morphology. Moreover, effectivetreatment with z-DEVD-fmk was associated with re-duced calpain-mediated �-spectrin degradation. Z-DEVD-fmk was also neuroprotective, at concentrationslower than those routinely used to inhibit caspase 3, in anin vitro model of necrotic neuronal cell death induced bymaitotoxin. After maitotoxin injury, as in the in vivostudy, neuroprotection was associated with inhibition ofcalpain rather than caspase 3, which was not activated inthis model. Finally, in a cell-free assay, z-DEVD-fmkdirectly inhibited calpain-mediated hydrolysis of caseinat a concentration range that extended below 1 uM, and

FIG. 7. Z-DEVD-fmk prevented cell death in neuronal glial cocultures exposed to maitotoxin, an inducer of necrotic cell death.(A) Representative phase-contrast images of cocultures 16 hours after injury induced by incubation with 0.1 nM maitotoxin for 1 hour or(B) maitotoxin in the presence of z-DEVD-fmk (10 µM). (C) Control cultures received identical procedures but no drugs. (D) LDH release14 to 16 hours after maitotoxin injury in the presence of increasing concentrations (as shown on x-axis) of z-DEVD-fmk. DMSO servedas the vehicle control (0 µM). Bars represent the mean ± SEM (n = 12–15 wells per condition) from one experiment of two performed.*P < 0.01 vs. control by ANOVA and post hoc t-tests with Bonferroni correction.

Z-DEVD-FMK REDUCES NECROSIS AND INHIBITS CALPAIN 1127

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

was similar to the concentration range that protectedagainst maitotoxin injury.

Previous work by our group and others using the ratLFP model suggests that caspase activation and apo-ptotic cell death are important delayed events after TBI(Conti et al., 1998; Knoblach et al., 2002; Rink et al.,1995; Yakovlev et al., 1997, 2001a,b). In addition, initialcharacterization of the impact device used in the presentstudy indicated that delayed apoptosis (24 hours), pri-marily confined to the ipsilateral cortex, was a feature ofmouse CCI (Fox et al., 1998). To date, however, caspaseinhibition strategies have only been evaluated in prein-jury or repetitive administration paradigms. Therefore,

the goals of the present study were both to extend pre-vious observations to a mouse model of CCI and toevaluate the therapeutic window for caspase inhibition.Although the results we obtained were unexpected, bothin terms of injury severity and in response to z-DEVD-fmk treatment, they are consistent with the view that celldeath mechanisms and phenotypes may vary with spe-cies, model, and injury severity. In general, the CCImodel tends to have a larger necrotic component thandoes LFP, and the proportion of necrosis after injury ineither model has been shown to increase with severity(Kochanek et al., 1995; Newcomb et al., 1999; Rinket al., 1995).

FIG. 8. Z-DEVD-fmk reduced cell deathand calpain-mediated proteolysis of�-spectrin in neuronal glial cocultures ex-posed to maitotoxin. (A) LDH release overtime after maitotoxin injury. Media wassampled from different cultures over timeas indicated on the x-axis. *P < 0.05 vs.vehicle treatment at same time by Stu-dent’s t-test comparison. (B) Expression of�-spectrin breakdown products (SBDs)over time from cultures represented in A.Time course is indicated above the blot.D, z-DEVD-fmk; V, equivalent volume ofDMSO vehicle, ctrl, uninjured control. The150 kD SBD is produced by both calpainand caspase 3, the 145 kD SBD is specifi-cally generated by calpain, and the 120 kDSBD is specifically generated by caspase3. (C) No expression of active caspase 3over time in cultures represented in A andB. Active recombinant caspase 3 servedas a positive control. (D) No expression ofcathepsin B over time in cultures repre-sented in A, B, and C. Ctrl, uninjured cul-tures; +Ctrl, mouse spleen. Data shown inA–D represent one of four total experi-ments performed: two with 100 µM z-DEVD-fmk and two with 50 µM z-DEVD-fmk. Results from all experiments wereidentical.

S. KNOBLACH ET AL.1128

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

Calpains are activated by micromolar (calpain I) ormillimolar (calpain II) elevations of calcium, whereascaspases are activated in response to specific signalsfrom external (cell death receptors) or internal (mito-chondria, endoplasmic reticulum) sources (for review seeChan and Mattson, 1999; Ray and Banik, 2003). Caspase3 activation has been used as a confirmatory marker forongoing apoptosis. In contrast, calpains are frequentlyassociated with necrotic cell death, although they mayalso be involved in neuronal apoptosis (Pike et al., 1998).Calpain inhibitors have been shown to improve neuro-logic outcome and reduce neurodegeneration in severalmodels of TBI, although such benefits did not alwayscorrelate precisely with calpain inhibition (Kupina et al.,2001; Saatman et al., 1996b, 2000). Caspase inhibitionstrategies may also be beneficial. Repeated injection ofz-DEVD-fmk from just before injury until 24 hours laterimproved function after LFP (Yakovlev et al., 1997).Intracerebroventricular injection of the pan-caspase in-hibitor z-VAD-fmk just after injury also improved func-tional recovery in this model, as well (Knoblach et al.,2002). Although these studies did not address the effectof treatment upon lesion volume, other work supportsthat use of caspase inhibitors may be neuroprotective.For example, intracerebroventricular pretreatment withz-VAD-fmk, or the caspase 1 selective inhibitor Ac-YVAD-cmk, reduced lesion volume after CCI (Finket al., 1999). Intraparenchymal infusion of z-DEVD-fmkover several days after combined CCI + hypoxia reducedlesion size, although functional outcome was not im-proved (Clark et al., 2000). In contrast to these results,intracerebroventricular pretreatment with z-Boc-fmk, an-other pan-caspase inhibitor, was reported to show no ef-fect upon lesion volume after CCI (Sullivan et al., 2002).

To circumvent complex issues of injury model, sever-ity and mechanism(s) related to in vivo studies, z-DEVD-fmk was tested in an in vitro model of neuronal cell deathcaused by maitotoxin. Even mild injury induced by mai-totoxin produces relatively pure necrotic cell death(Kutty et al., 1989; McGinnis et al., 2003; Wang et al.,1996). This is characterized by a rapid increase in cyto-solic free calcium, opening of a cytolytic pore on theplasma membrane, ATP depletion, calpain activation,and release of cytochome c, but not activation of caspase3 or generation of caspase 3-specific cleavage fragmentsof PARP or �-spectrin (Kutty et al., 1989; McGinniset al., 2003; Wang et al., 1996). Thus, the demonstrationthat z-DEVD-fmk is neuroprotective in this model sup-ports a role for it as an inhibitor of caspase-independentor necrotic cell death. In addition, the data showing thatexpression of calpain-mediated SBD is abolished or re-duced by z-DEVD-fmk after maitotoxin injury stronglysupport inhibition of calpain as the mechanism of pro-tection, particularly as there was no increase in caspase-mediated SBD (consistent with the model). Although the

FIG. 9. BODIPY-labeled casein assay directly shows inhibition ofcalpain I by z-DEVD-fmk. BODIPY-labeled fluorescent caseinwas incubated with 10 units of calpain I and increasing concen-trations of z-DEVD-fmk as indicated on the x-axis. Bars representthe mean ± SEM from triplicates prepared at each concentrationof z-DEVD-fmk. *P < 0.05 vs. 0 µM z-DEVD-fmk, by one-wayANOVA followed by t-tests with Bonferroni correction. Dataare representative of one experiment from two repetitions.

FIG. 10. Z-DEVD-fmk treatment reduces accumulation of the145 kD SBD 2 hours after CCI. Mice were subjected to CCI andtreated with 160 ng z-DEVD-fmk 1 hour later as detailed in Fig. 1.Two hours after injury (1 hour after z-DEVD-fmk treatment), theinjured cortex was removed and homogenized, and �-spectrinproteolysis was assessed by immunoblotting methods. (A) Rep-resentative blot. (B) Semiquantitative analysis of 145 kD bandfrom n = 6 sham or naive mice, n = 4 injured, z-DEVD-fmk treatedmice, or n = 3 injured, vehicle control mice. *P < 0.05 vs. vehicle.Bars represent the percent of control relative density units. Per-cent data were subjected to square root transformation and thenanalyzed by t-tests with Bonferroni correction.

Z-DEVD-FMK REDUCES NECROSIS AND INHIBITS CALPAIN 1129

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

present studies focused on calpain as a potential mecha-nism, the data do not preclude additional effects of z-DEVD-fmk on other mechanisms or factors. For ex-ample, lysosomal proteases have historically beenconsidered to be passively involved in necrotic celldeath, and z-DEVD-fmk inhibits cathepsin B (Schotteet al., 1999). However, active cathepsin B was not de-tected after maitotoxin injury; therefore, it is unlikelythat the neuroprotective effect of z-DEVD-fmk involvesthis protease.

A direct effect of z-DEVD-fmk on calpain was sup-ported by data from the BODIPY-labeled casein diges-tion assay, where the activity of purified calpain I wasinhibited 57% by 1 uM z-DEVD-fmk. In seperate ex-periments, we have confirmed that z-DEVD-fmk inhibitscalpain I at a Ki of 1 uM (Knoblach et al., unpublisheddata; August, 2004). Notably, z-DEVD-fmk showed neu-roprotection against maitotoxin at even lower concentra-tions, suggesting the possible existence of additional celldeath factors that are activated by maitotoxin and inhib-ited by z-DEVD-fmk.

Despite their efficacy in improving outcomes in mod-els of ischemia and CNS trauma, it has been known forsome time that most presently available calpain inhibi-tors have poor cell permeability. Many also lack speci-ficity and inhibit other cysteine proteases, such as ca-thepsins (B and L), as well as the proteosome (Inoueet al., 2003). Tetrapeptide-based caspase inhibitors likez-DEVD-fmk have generally been considered to be se-lective for caspase cysteine proteases, albeit only rela-tively selective for individual caspases (Garcia-Calvoet al., 1998; Thornberry et al., 1997). For this reason,they have been used extensively as “proof of principle”tools to implicate caspase activation or ongoing apo-ptotic cell death in physiologic or pathophysiologic pro-cesses. Recently, however, data have emerged that indi-cate that these drugs effectively inhibit other cysteineproteases of the papain superfamily. Specifically, inhibi-tion of calpain(s) and cathepsins by z-VAD-fmk is sup-ported by several studies from different groups, as isinhibition of cathepsin B by z-DEVD-fmk (Blomgrenet al., 2001; Schotte et al., 1999; Vancompernolle et al.,1998; Waterhouse et al., 1998; Wolf et al., 1999). To ourknowledge, the present data are the first to show inhibi-tion of calpains by z-DEVD-fmk. It is important to pointout that the concentration of z-DEVD-fmk that showedneuroprotection in the maitotoxin model and that in-hibited calpain directly was much lower than that whichis widely used to study caspase-mediated cell deathin vitro.

Ultimately, it may be most effective clinically to de-velop broad-spectrum protease inhibition therapeuticstrategies, rather than to use selective approaches, be-cause both calpain and caspase may contribute criticallyto the injury response. In addition, these proteases par-

ticipate in cross-talk pathways that are still not under-stood (Nakagawa and Yuan, 2000; Neumar et al., 2003).Further, inhibition of one cell death mechanism (or pro-tease), in general, may cause cells to switch to another(Lemaire et al., 1998; Schwab et al., 2002). Finally, bothcalpain and caspase cleave and inactivate PARP, and thismay influence noncaspase mediated apoptotic cell deathinduced by apoptosis inducing factor, which appears tobe both activated by PARP and involved in cell deathafter TBI (McGinnis et al., 1999; Yu et al., 2002; Zhanget al., 2002) (Clark et al., 1997; Kampfl et al., 1997).

CONCLUSION

The present data show that treatment with z-DEVD-fmk improves behavioral recovery, reduces tissue dam-age and prevents accumulation of calpain-mediated�-spectrin breakdown products when administered notlater than 1 hour after injury in a TBI model that primar-ily shows necrosis. Z-DEVD-fmk also reduces necroticneuronal cell death in vitro, and such neuroprotection isassociated with inhibition of calpain, but not caspase 3 orcathepsin B. In addition, z-DEVD-fmk reduces calpain-mediated hydrolysis of casein, which indicates that z-DEVD-fmk can directly inhibit calpain. This nonspecificproperty of z-DEVD-fmk may account, at least in part,for its neuroprotective actions.

Acknowledgment: The authors thank Dr. Xiao Di andMs. Sadia Aden for expert technical assistance.

REFERENCES

Albensi BC, Knoblach SM, Chew BG, O’Reilly MP, Faden AI, PekarJJ (2000) Diffusion and high resolution MRI of traumatic braininjury in rats: time course and correlation with histology. ExpNeurol 162:61–72

Ankarcrona M, Dypbukt JM, Bonfoco E, Zhivotovsky B, Orrenius S,Lipton SA, Nicotera P (1995) Glutamate-induced neuronal death:A succession of necrosis or apoptosis depending on mitochondrialfunction. Neuron 15:961–973

Beer R, Franz G, Srinivasan A, Hayes RL, Pike BR, Newcomb JK,Zhao X, Schmutzhard E, Poewe W, Kampfl A (2000) Temporalprofile and cell subtype distribution of activated caspase-3 follow-ing experimental traumatic brain injury. J Neurochem 75:1264–1273

Blomgren K, Zhu C, Wang X, Karlsson JO, Leverin AL, Bahr BA,Mallard C, Hagberg H (2001) Synergistic activation of caspase-3by m-calpain after neonatal hypoxia- ischemia: A mechanism of‘pathological apoptosis‘? J Biol Chem 276:10191–10198

Buki A, Siman R, Trojanowski JQ, Povlishock JT (1999) The role ofcalpain-mediated spectrin proteolysis in traumatically inducedaxonal injury. J Neuropathol Exp Neurol 58:365–375

Buki A, Okonkwo D, Wang K, Povlishock J (2000) Cytochrome crelease and caspase activation in traumatic axonal injury. J Neu-rosci 20: 2825–2834

Chan SL, Mattson MP (1999) Caspase and calpain substrates: Roles insynaptic plasticity and cell death. J Neurosci Res 58:167–190

Chen J, Nagayama T, Jin K, Stetler RA, Zhu RL, Graham SH, SimonRP (1998) Induction of caspase-3-like protease may mediate de-layed neuronal death in the hippocampus after transient cerebralischemia. J Neurosci 18:4914–4928

Clark RS, Kochanek PM, Dixon CE, Chen M, Marion DW, HeinemanS, DeKosky ST, Graham SH (1997) Early neuropathologic effects

S. KNOBLACH ET AL.1130

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

of mild or moderate hypoxemia after controlled cortical impactinjury in rats. J Neurotrauma 14:179–189

Clark RS, Kochanek PM, Chen M, Watkins SC, Marion DW, Chen J,Hamilton RL, Loeffert JE, Graham SH (1999) Increases in Bcl-2and cleavage of caspase-1 and caspase-3 in human brain after headinjury. FASEB J 13:813–821

Clark RS, Kochanek PM, Watkins SC, Chen M, Dixon CE, SeidbergNA, Melick J, Loeffert JE, Nathaniel PD, Jin KL, Graham SH(2000) Caspase-3 mediated neuronal death after traumatic braininjury in rats. J Neurochem 74:740–753

Clarke PG (1990) Developmental cell death: Morphological diversityand multiple mechanisms. Anat Embryol (Berl) 181:195–213

Colicos MA, Dixon CE, Dash PK (1996) Delayed, selective neuronaldeath following experimental cortical impact injury in rats: Pos-sible role in memory deficits. Brain Res 739:111–119

Conti AC, Raghupathi R, Trojanowski JQ, McIntosh TK (1998) Ex-perimental brain injury induces regionally distinct apoptosis duringthe acute and delayed post-traumatic period. J Neurosci 18:5663–5672

Faden AI, Fox GB, Di X, Knoblach SM, Cernak I, Mullins P, Ni-kolaeva M, Kozikowski AP (2003) Neuroprotective and nootropicactions of a novel cyclized dipeptide after controlled cortical im-pact injury in mice. J Cereb Blood Flow Metab 23:355–363

Fink K, Zhu J, Namura S, Shimizu-Sasamata M, Endres M, Ma J,Dalkara T, Yuan J, Moskowitz MA (1998) Prolonged therapeuticwindow for ischemic brain damage caused by delayed caspaseactivation. J Cereb Blood Flow Metab 18:1071–1076

Fink KB, Andrews LJ, Butler WE, Ona VO, Li M, Bogdanov M,Endres M, Khan SQ, Namura S, Stieg PE, Beal MF, MoskowitzMA, Yuan J, Friedlander RM (1999) Reduction of post-traumaticbrain injury and free radical production by inhibition of thecaspase-1 cascade. Neuroscience 94:1213–1218

Formigli L, Papucci L, Tani A, Schiavone N, Tempestini A, OrlandiniGE, Capaccioli S, Orlandini SZ (2000) Aponecrosis: Morphologi-cal and biochemical exploration of a syncretic process of cell deathsharing apoptosis and necrosis. J Cell Physiol 182:41–49

Fox GB, Fan L, Levasseur RA, Faden AI (1998) Sustainedsensory/motor and cognitive deficits with neuronal apoptosis fol-lowing controlled cortical impact brain injury in the mouse.J Neurotrauma 15:599–614

Garcia-Calvo M, Peterson EP, Leiting B, Ruel R, Nicholson DW,Thornberry NA (1998) Inhibition of human caspases by peptide-based and macromolecular inhibitors. J Biol Chem 273:32608–32613

Gusovsky F, Daly JW (1990) Maitotoxin: A unique pharmacologicaltool for research on calcium-dependent mechanisms. BiochemPharmacol 39:1633–1639

Harter L, Keel M, Hentze H, Leist M, Ertel W (2001) Caspase-3activity is present in cerebrospinal fluid from patients with trau-matic brain injury. J Neuroimmunol 121:76–78

Inoue J, Nakamura M, Cui YS, Sakai Y, Sakai O, Hill JR, Wang KK,Yuen PW (2003) Structure-activity relationship study and drugprofile of N-(4-fluorophenylsulfonyl)-L-valyl-L-leucinal(SJA6017) as a potent calpain inhibitor. J Med Chem 46:868–871

Kampfl A, Posmantur R, Nixon R, Grynspan F, Zhao X, Liu SJ, New-comb JK, Clifton GL, Hayes RL (1996) Mu-calpain activation andcalpain-mediated cytoskeletal proteolysis following traumaticbrain injury. J Neurochem 67:1575–1583

Kampfl A, Posmantur RM, Zhao X, Schmutzhard E, Clifton GL, HayesRL (1997) Mechanisms of calpain proteolysis following traumaticbrain injury: Implications for pathology and therapy: A review andupdate. J Neurotrauma 14:121–134

Keane RW, Kraydieh S, Lotocki G, Alonso OF, Aldana P, Dietrich WD(2001) Apoptotic and antiapoptotic mechanisms after traumaticbrain injury. J Cereb Blood Flow Metab 21:1189–1198

Kerr JF, Wyllie AH, Currie AR (1972) Apoptosis: A basic biologicalphenomenon with wide-ranging implications in tissue kinetics.Br J Cancer 26:239–257

Kitanaka C, Kuchino Y (1999) Caspase-independent programmed celldeath with necrotic morphology. Cell Death Differ 6:508–515

Knoblach SM, Nikolaeva M, Huang X, Fan L, Krajewski S, Reed JC,Faden AI (2002) Multiple caspases are activated after traumatic

brain injury: Evidence for involvement in functional outcome.J Neurotrauma 19:1155–1170

Kochanek PM, Marion DW, Zhang W, Schiding JK, White M, PalmerAM, Clark RS, O’Malley ME, Styren SD, Ho C, et al. (1995)Severe controlled cortical impact in rats: Assessment of cerebraledema, blood flow, and contusion volume. J Neurotrauma 12:1015–1025

Kupina NC, Nath R, Bernath EE, Inoue J, Mitsuyoshi A, Yuen PW,Wang KK, Hall ED (2001) The novel calpain inhibitor SJA6017improves functional outcome after delayed administration in amouse model of diffuse brain injury. J Neurotrauma 18:1229–1240

Kupina NC, Detloff MR, Bobrowski WF, Snyder BJ, Hall ED (2003)Cytoskeletal protein degradation and neurodegeneration evolvesdifferently in males and females following experimental head in-jury. Exp Neurol 180:55–73

Kutty RK, Singh Y, Santostasi G, Krishna G (1989) Maitotoxin-induced liver cell death involving loss of cell ATP following influxof calcium. Toxicol Appl Pharmacol 101:1–10

Legrand AM, Bagnis R (1984) Effects of highly purified maitotoxinextracted from dinoflagellate Gambierdiscus toxicus on action po-tential of isolated rat heart. J Mol Cell Cardiol 16:663–666

Lemaire C, Andreau K, Souvannavong V, Adam A (1998) Inhibition ofcaspase activity induces a switch from apoptosis to necrosis. FEBSLett 425:266–270

Martin LJ, Al-Abdulla NA, Brambrink AM, Kirsch JR, Sieber FE,Portera-Cailliau C (1998) Neurodegeneration in excitotoxicity,global cerebral ischemia, and target deprivation: A perspective onthe contributions of apoptosis and necrosis. Brain Res Bull 46:281–309

McCracken E, Hunter AJ, Patel S, Graham DI, Dewar D (1999) Cal-pain activation and cytoskeletal protein breakdown in the corpuscallosum of head-injured patients. J Neurotrauma 16:749–761

McGinnis KM, Gnegy ME, Park YH, Mukerjee N, Wang KK (1999)Procaspase-3 and poly(ADP)ribose polymerase (PARP) are cal-pain substrates. Biochem Biophys Res Commun 263:94–99

McGinnis KM, Gnegy ME, Falk N, Nath R, Wang KK (2003) Cyto-chrome c translocation does not lead to caspase activation in mai-totoxin-treated SH-SY5Y neuroblastoma cells. Neurochem Int42:517–523

Mukhin AG, Ivanova SA, Allen JW, Faden AI (1998) Mechanicalinjury to neuronal/glial cultures in microplates: Role of NMDAreceptors and pH in secondary neuronal cell death. J Neurosci Res51:748–758

Nakagawa T, Yuan J (2000) Cross-talk between two cysteine proteasefamilies: Activation of caspase-12 by calpain in apoptosis. J CellBiol 150:887–894

Neumar RW, Xu YA, Gada H, Guttmann RP, Siman R (2003) Cross-talk between calpain and caspase proteolytic systems during neu-ronal apoptosis. J Biol Chem 278:14162–14167

Newcomb JK, Zhao X, Pike BR, Hayes RL (1999) Temporal profile ofapoptotic-like changes in neurons and astrocytes following con-trolled cortical impact injury in the rat. Exp Neurol 158:76–88

Pike BR, Zhao X, Newcomb JK, Wang KK, Posmantur RM, Hayes RL(1998) Temporal relationships between de novo protein synthesis,calpain caspase 3-like protease activation, and DNA fragmentationduring apoptosis in septo-hippocampal cultures. J Neurosci Res 52:505–520

Posmantur R, Kampfl A, Siman R, Liu J, Zhao X, Clifton GL, HayesRL (1997) A calpain inhibitor attenuates cortical cytoskeletal pro-tein loss after experimental traumatic brain injury in the rat. Neu-roscience 77:875–888

Pang Z, Bondada V, Sengoku T, Siman R, Geddes JW (2003) Calpainfacilitates the neuron death induced by 3-nitropropionic acid andcontributes to the necrotic morphology. J Neuropathol Exp Neurol62:633–643

Pohl D, Bittigau P, Ishimaru MJ, Stadthaus D, Hubner C, Olney JW,Turski L, Ikonomidou C (1999) N-Methyl-D-aspartate antagonistsand apoptotic cell death triggered by head trauma in developing ratbrain. Proc Natl Acad Sci U S A 96:2508–2513

Prins ML, Hovda DA (2003) Developing experimental models to ad-dress traumatic brain injury in children. J Neurotrauma 20:123–137

Z-DEVD-FMK REDUCES NECROSIS AND INHIBITS CALPAIN 1131

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004

Ray SK, Banik NL (2003) Calpain and its involvement in the patho-physiology of CNS injuries and diseases: therapeutic potential ofcalpain inhibitors for prevention of neurodegeneration. Curr DrugTarget CNS Neurol Disord 2:173–189

Rink A, Fung KM, Trojanowski JQ, Lee VM, Neugebauer E, McIntoshTK (1995) Evidence of apoptotic cell death after experimentaltraumatic brain injury in the rat. Am J Pathol 147:1575–1583

Saatman KE, Bozyczko-Coyne D, Marcy V, Siman R, McIntosh TK(1996a) Prolonged calpain-mediated spectrin breakdown occursregionally following experimental brain injury in the rat. J Neuro-pathol Exp Neurol 55:850–860

Saatman KE, Murai H, Bartus RT, Smith DH, Hayward NJ, Perri BR,McIntosh TK (1996b) Calpain inhibitor AK295 attenuates motorand cognitive deficits following experimental brain injury in therat. Proc Natl Acad Sci U S A 93:3428–3433

Saatman KE, Zhang C, Bartus RT, McIntosh TK (2000) Behavioralefficacy of posttraumatic calpain inhibition is not accompanied byreduced spectrin proteolysis, cortical lesion or apoptosis. J CerebBlood Flow Metab 20:66–73

Schotte P, Declercq W, Van Huffel S, Vandenabeele P, Beyaert R(1999) Non-specific effects of methyl ketone peptide inhibitors ofcaspases. FEBS Lett 442:117–121

Schwab BL, Guerini D, Didszun C, Bano D, Ferrando-May E, Fava E,Tam J, Xu D, Xanthoudakis S, Nicholson DW, Carafoli E, Nico-tera P (2002) Cleavage of plasma membrane calcium pumps bycaspases: A link between apoptosis and necrosis. Cell Death Differ9:818–831

Sullivan PG, Keller JN, Bussen WL, Scheff SW (2002) Cytochrome crelease and caspase activation after traumatic brain injury. BrainRes 949:88–96

Thornberry NA, Rano TA, Peterson EP, Rasper DM, Timkey T, Gar-cia-Calvo M, Houtzager VM, Nordstrom PA, Roy S, VaillancourtJP, Chapman KT, Nicholson DW (1997) A combinatorial approachdefines specificities of members of the caspase family and gran-zyme B. Functional relationships established for key mediators ofapoptosis. J Biol Chem 272:17907–17911

Vancompernolle K, Van Herreweghe F, Pynaert G, Van de Craen M,De Vos K, Totty N, Sterling A, Fiers W, Vandenabeele P, GrootenJ (1998) Atractyloside-induced release of cathepsin B, a proteasewith caspase-processing activity. FEBS Lett 438:150–158

Vanderklish PW, Bahr BA (2000) The pathogenic activation of calpain:A marker and mediator of cellular toxicity and disease states. IntJ Exp Pathol 81:323–339

Volbracht C, Leist M, Nicotera P (1999) ATP controls neuronal apo-ptosis triggered by microtubule breakdown or potassium depriva-tion. Mol Med 5:477–489

Wang KK, Nath R, Raser KJ, Hajimohammadreza I (1996) Maitotoxininduces calpain activation in SH-SY5Y neuroblastoma cells andcerebrocortical cultures. Arch Biochem Biophys 331:208–214

Waterhouse NJ, Finucane DM, Green DR, Elce JS, Kumar S, AlnemriES, Litwack G, Khanna K, Lavin MF, Watters DJ (1998) Calpainactivation is upstream of caspases in radiation-induced apoptosis.Cell Death Differ 5:1051–1061

Wolf BB, Goldstein JC, Stennicke HR, Beere H, Amarante-MendesGP, Salvesen GS, Green DR (1999) Calpain functions in a caspase-independent manner to promote apoptosis-like events during plate-let activation. Blood 94:1683–1692

Yakovlev AG, Knoblach SM, Fan L, Fox GB, Goodnight R, Faden AI(1997) Activation of CPP32-like caspases contributes to neuronalapoptosis and neurological dysfunction after traumatic brain in-jury. J Neurosci 17:7415–7424

Yakovlev AG, Ota K, Wang G, Movsesyan V, Bao WL, Yoshihara K,Faden AI (2001) Differential expression of apoptotic protease-activating factor-1 and caspase-3 genes and susceptibility to apo-ptosis during brain development and after traumatic brain injury.J Neurosci 21:7439–7446

Yakovlev AG, Di X, Movsesyan V, Mullins PG, Wang G, Boulares H,Zhang J, Xu M, Faden AI. (2001) Presence of DNA fragmentationand lack of neuroprotective effect in DFF45 knockout mice sub-jected to traumatic brain injury. Mol Med 7:205–216

Yu SW, Wang H, Poitras MF, Coombs C, Bowers WJ, Federoff HJ,Poirier GG, Dawson TM, Dawson VL (2002) Mediation of poly-(ADP-ribose) polymerase-1-dependent cell death by apoptosis-in-ducing factor. Science 297:259–263

Zhang X, Chen J, Graham SH, Du L, Kochanek PM, Draviam R, GuoF, Nathaniel PD, Szabo C, Watkins SC, Clark RS (2002) Intra-nuclear localization of apoptosis-inducing factor (AIF) and largescale DNA fragmentation after traumatic brain injury in rats and inneuronal cultures exposed to peroxynitrite. J Neurochem 82:181–191

Zhao X, Posmantur R, Kampfl A, Liu SJ, Wang KK, Newcomb JK,Pike BR, Clifton GL, Hayes RL (1998) Subcellular localizationand duration of mu-calpain and m-calpain activity after traumaticbrain injury in the rat: a casein zymography study. J Cereb BloodFlow Metab 18:161–167

S. KNOBLACH ET AL.1132

J Cereb Blood Flow Metab, Vol. 24, No. 10, 2004