Delayed treatment with a p53 inhibitor enhances recovery in stroke brain

Delayed Treatment with a p53 InhibitorEnhances Recovery in Stroke Brain

Yu Luo, PhD,1 Chi-Chung Kuo, MD,1 Hui Shen, MD,1 Jenny Chou, BS,1 Nigel H. Greig, PhD,2

Barry J. Hoffer, MD, PhD,1 and Yun Wang, MD, PhD1

Objective: Cerebral ischemia can activate endogenous reparative processes, such as proliferation of endogenous neural progen-itor cells (NPCs) in the subventricular zone (SVZ). Most of these new cells die shortly after injury. The purpose of this studywas to examine a novel strategy for treatment of stroke at 1 week after injury by enhancing the survival of ischemia-inducedendogenous NPCs in SVZ.Methods: Adult rats were subjected to a 90-minutes middle cerebral artery occlusion. A p53 inhibitor pifithrin-� (PFT-�) wasadministered to stroke rats from days 6 to 9 after middle cerebral artery occlusion. Locomotor behavior was measured using anactivity chamber. Proliferation, survival, migration, and differentiation of endogenous NPCs were examined using quantitativereverse transcription polymerase chain reaction, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling, andimmunohistochemistry.Results: PFT-� enhanced functional recovery as assessed by a significant increase in multiple behavioral measurements. DelayedPFT-� treatment had no effect on the cell death processes in the lesioned cortical region. However, it enhanced the survival ofSVZ progenitor cells, and promoted their proliferation and migration. PFT-� inhibited the expression of a p53-dependentproapoptotic gene, termed PUMA (p53-upregulated modulator of apoptosis), within the SVZ of stroke animals. The enhance-ment of survival/proliferation of NPCs was further found in SVZ neurospheres in tissue culture. PFT-� dose-dependentlyincreased the number and size of new neurosphere formation.Interpretation: Delayed treatment with a p53 inhibitor PFT-� is able to modify stroke-induced endogenous neurogenesis andimprove the functional recovery in stroke animals.

Ann Neurol 2009;65:520–530

Current treatment strategies for stroke primarily focuson reducing the size of ischemic damage and on rescu-ing dying cells early after occurrence. Treatments, suchas the use of thrombolytic agents, are often limited bya narrow therapeutic time window, and no pharmaco-logical agent has been shown to enhance the recoveryprocess when therapy is initiated in the chronic or sub-acute phase, that is, 6 days, after stroke in patients.1,2

Recent studies have indicated that transplantation ofneural stem cells (NSCs) and neural progenitor cells(NPCs) is beneficial in animal models of stroke. Forexample, grafted human NSCs survived and migratedto the lesioned site 1 month after transplantation instroke rats.3 The clinical application of this approachmay be limited by ethical and logistic concerns. On theother hand, cerebral ischemia can also activate endog-enous repair processes. De novo neurogenesis wasfound in the subventricular zone (SVZ) of adult mam-

malian brain after stroke.4 Most of these cells diewithin weeks after injury,5 and their ability to restorebehavioral deficits in the stroke animals are minimal.

p53 is a major mediator of ischemia-induced celldeath in brain. Cerebral ischemia activates p53 expres-sion. Administration of the small synthetic p53 inhibi-tor, pifithrin-� (PFT-�),6 30 minutes before7 or 1 hourafter middle cerebral artery occlusion (MCAO) reducedp53 activity, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) labeling,cerebral infarction in the ischemic brain, and motor dis-ability in rodents.8 The protective effects of PFT-� werenot significant if given 3 hours after MCAO. These datasuggest that programmed cell death at the lesioned site isinitiated rapidly after ischemia, and the effectiveness of ap53 inhibitor to rescue these injured cells is limited to anarrow therapeutic window (ie, hours after ischemia),

From the 1National Institute on Drug Abuse, and 2National Insti-tute of Aging, National Institutes of Health Intramural ResearchProgram, Baltimore, MD.

Address correspondence to Dr Wang, National Institute on DrugAbuse, I.R.P., Neural Protection and Regeneration Section, 251Bayview Boulevard, Baltimore, MD 21224.E-mail: [email protected]

Potential conflict of interest: Nothing to report.

Additional Supporting Information may be found in the online ver-sion of this article.

Received Jul 28, 2008, and in revised form Oct 31. Accepted forpublication Oct 31, 2008.

Published in Wiley InterScience (www.interscience.wiley.com).DOI: 10.1002/ana.21592

520 © 2009 American Neurological Association

because the cell death biochemical cascade quickly passesbeyond the point of involvement of p53.9

In addition to its involvement in early cell death inthe ischemic penumbra, p53 plays important roles inother nonischemic regions. A recent study indicatedthat p53 protein is highly expressed in the NPC area inSVZ, compared with surrounding brain regions. Neu-rospheres from the lateral ventricle wall of adult p53null animals proliferated faster and incorporated morebromodeoxyuridine (BrdU) than wild-type control an-imals.10 These data suggest that the survival/prolifera-tion of adult NPCs from the SVZ is influenced byp53. It is also possible that inhibition of p53 may ex-tend the proliferation and survival of endogenousNPCs, and alter the biological outcomes days afterstroke.

In this study, instead of rescuing cells in the penum-bra, we attempted to enhance the survival of ischemia-induced endogenous NPCs in SVZ using a p53 inhib-itor. Because activation of NPC proliferation in SVZoccurs at a later stage,11 compared with the cell deaththat occurs in ischemic core or penumbra, pharmaco-logical intervention to promote NPC survival may beeffective late (ie, a week) after the onset of stroke.

Materials and MethodsTransient Focal Ischemia ModelAdult male Sprague–Dawley rats were anesthetized withchloral hydrate (400mg/kg intraperitoneally). The right mid-dle cerebral artery was ligated and bilateral common carotidswere clamped for 90 minutes to generate focal ischemia inthe cerebral cortex.12–14 Core body temperature was main-tained at 37°C.

Administration of Pifithrin-� or VehicleRats were anesthetized 6 days after MCAO. PFT-� (0.4�g/�l � 20�l) or vehicle (20�l, 10% DMSO in saline) wasadministered intracerebroventricularly (anteroposterior [AP]:0.8mm; lateral: 1.5mm to bregma; dorsoventral: �3.7mm).Animals then received PFT-� (0.2mg/100gm, intraperitone-ally twice a day) or vehicle for three consecutive days.

Injection of BromodeoxyuridineThree injection protocols were used: (1) the kinetics of pro-liferation of NPCs after injury—BrdU was injected systemi-cally on days 2, 4, 6, 8, 10, 12, or 14 after MCAO at thedose of 50mg/kg � 4 at 2-hour interval; (2) proliferation ofNPCs induced by PFT-� (see Fig 3A)—BrdU was adminis-tered daily for 3 days (50mg/kg � 2) after PFT-� injection(ie, days 7–9 after MCAO); and (3) migration and differen-tiation of NPCs (see Fig 4A)—BrdU (50mg/kg � 4) wasgiven before PFT-� injection (ie, day 5 after MCAO).

Behavioral MeasurementsEach animal was individually placed in a 42 � 42 � 31cmactivity monitor for 24 hours (Omnitech Electronic, Colum-bus, OH). Water and food were provided in the chambers.

The monitor contained 16 horizontal and 8 vertical infraredsensors spaced 2.5cm apart. Locomotor activity was calcu-lated using the number of infrared beams broken by the an-imals (see supplement 3).

ImmunohistochemistryBrdU and proliferating cell nuclear antigen (PCNA) immu-nostaining were performed as described elsewhere15,16 (fordetails, see also supplement 1). For double-labeling immuno-staining, brain sections were incubated in solutions of anti-bodies against various markers (Nestin, microtubule-associated protein 2, neuronal nuclear antigen [NeuN], orglial fibrillary acidic protein [GFAP]). For double-labelingimmunocytochemistry, confocal images were obtained by us-ing UltraView confocal microscopic system (Perkin-Elmer,Oak Brook, IL).

Terminal Deoxynucleotidyl Transferase-MediateddUTP Nick-End Labeling HistochemistryA standard TUNEL procedure for frozen tissue sections wasperformed as described previously17 and detailed in Supple-ment 1.

Quantitative Reverse Transcription Polymerase Chain Reaction.The SVZ was dissected out. Total RNA (1�g) was treatedwith RQ-1 Rnase-free Dnase I and reverse transcribed intocomplementary DNA using random hexamers by AMV re-verse transcriptase (Roche, Nutley, N.J.). Complemen-tary DNA levels for hypoxanthine guanine phosphoribo-syl transferase (HPRT), phosphoglycerate kinase 1 (PGK1),Bax, Bcl2, apoptotic peptidase activating factor 1 (Apaf1),p21, and p53-upregulated modulator of apoptosis (PUMA)were determined by specific universal probe Library primerprobe sets (Roche) by quantitative reverse transcriptionpolymerase chain reaction (qRT-PCR).18 Primers and (6-carboxyfluorescein) FAM-labeled probes for each gene arelisted in Table 1of Supplement 2.

Subventricular Zone Cultures. SVZ cells were collected frommouse embryos and dissociated by triturating. Cells wereplated in neurobasal medium (Invitrogen, La Jolla, CA) sup-plemented with B27 (Invitrogen), 2mM L-glutamine (In-vitrogen), antibiotic/antimycotic (Invitrogen), and heparin(Sigma). Equal numbers of cells were plated and kept in me-dium containing 20ng/ml epidermal growth factor. Cellswere plated at 10,000 cells/cm2. PFT-� or vehicle was addedat different concentrations at day 1 in vitro (DIV1) orDIV3. The number and size (diameter) of neurospheres werecounted 7 days after plating.

Statistical AnalysisStatistical analysis was performed using Student’s t test, andone- or two-way analysis of variance (ANOVA) as appropri-ate, with Newman–Keuls post hoc tests. p values less than0.05 were considered significant.

Luo et al: p53 Inhibition and Stroke 521

ResultsKinetics of Subventricular Zone Progenitor CellProliferation and Cortical Cell Apoptosisafter IschemiaDepending on the stroke model used, the activation ofNPC proliferation in the SVZ and cell death in theischemic core/penumbra varies after ischemia.11 A1-day BrdU injection (50mg/kg � 4, every 2 hours)was given systemically on days 2, 4, 6, 8, 10, 12, or 14after MCAO in 45 rats to examine the kinetics ofNPC proliferation in the stroke model used in thisstudy. The density of immunoreactive BrdU (�) cellswas examined 12 hours after the last injection. We ob-served a robust increase of BrdU immunoreactivity inSVZ as early as 2 days after MCAO (Figs 1A, B). Theincrease of BrdU immunoreactivity was sustainedthrough 4 days after MCAO, started to decline be-tween days 6 and 8, and returned to basal levelsaround day 10 (see Figs 1A, B). The anterior SVZ (AP�0.2mm to bregma) showed more BrdU incorpora-tion than the posterior SVZ (AP �0.4mm; see Figs1A, B [p � 0.001, two-way ANOVA). In another setof rats (n � 10), the kinetics of apoptosis/necrosis inthe ischemic side cortex was examined on days 2, 6,and 9 after MCAO. The density of TUNEL labelingpeaked at day 2 after MCAO; it returned to basal lev-els (compared with sham surgery) on day 6 (see Figs1C, D), suggesting differential temporal windows ofcell proliferation in the SVZ and cell death in the isch-emic cortex. Furthermore, PFT-�, given at days 6 to 9after MCAO, had no effect on the cell death processesin cortical regions (see Figs 1C, E; p � 0.266, t test).These data suggest that, for “therapeutic efficacy,” thesuppression of apoptosis in the ischemic cortex shouldtake place earlier after MCAO, whereas rescue of theNPCs in SVZ can occur after day 6.

Delayed Treatment with Pifithrin-� Improves theBehavioral Recovery after StrokePFT-� or vehicle was given from days 6 to 9 afterMCAO or sham surgery (Fig 2A). Administration ofPFT-� did not alter locomotor activity in nonstrokerats (n � 12, data not shown). In the stroke animals(n � 29), PFT-� treatment significantly enhanced mo-tor function, as demonstrated by increases in horizontalactivity (see Fig 2B; p � 0.001, two-way ANOVA),total distance traveled (see Fig 2C; p � 0.001, two-wayANOVA), number of movements (see Fig 2D; p �0.009, two-way ANOVA), and vertical activity (see Fig2E; p � 0.045, two-way ANOVA) on days 14 and 21after MCAO (see Table 2 in Supplement 3). These re-sults demonstrate that delayed PFT-� treatment is ef-fective in improving the functional outcome in strokerats.

Delayed Treatment with Pifithrin-� Enhances theProliferation/Survival of Progenitor Cells in theSubventricular Zone in Stroke RatsBecause p53 negatively regulates self-renewal of NPCsin SVZ of adult brain,10 the behavioral improvementinduced by p53 antagonist may thus be attributed tothe upregulation of endogenous NPCs in SVZ. Wenext tested whether PFT-� altered the density of NPCsin SVZ in 18 stroke rats. PFT-� (n � 10) or vehicle(n � 8) was administered daily from days 6 to 9; BrdUwas given from days 7 to 9 after MCAO (Fig 3A).Animals were killed at 12 hours after the last injectionof BrdU. Most of the BrdU-labeled progenitor cellswere present in the SVZ. PFT-� significantly enhancedthe number of BrdU-positive cells in SVZ (see Fig 3B;p � 0.008, two-way ANOVA) in stroke rats (see Fig3C).

Because p53 can regulate cell proliferation, survival,or both, we next examined whether the increase inNPC density in SVZ after delayed PFT-� treatment ismediated through these two mechanisms. PCNA, amarker for proliferating cells, was used to study the ef-fect of proliferation19 in five vehicle-treated and sevenPFT-�–treated rats (see timeline in Fig 3A). Significantincrease in PCNA (�) cell density (see Fig 3E; p �0.003, t test) was found in the rostral SVZ (AP: 0.2 tobregma) in PFT-�–treated rats on day 10 afterMCAO. Typical PCNA immunostaining is demon-strated in Figure 3D. These data suggest that PFT-�increases the proliferation of NPCs in SVZ of strokerats. The effect of PFT-� on cell survival was examinedin seven rats. Delayed PFT-� treatment significantlysuppressed TUNEL labeling, a marker for apoptosis/necrosis, in the SVZ on day 10 after MCAO (see Fig3F; p � 0.010, two-way ANOVA). Taken together,these data suggest that delayed PFT-� treatment en-hanced both proliferation and survival of NPCs inSVZ in stroke rats.

Treatment with Pifithrin-� Promotes the Survivalof Progenitor Cells from Subventricular Zoneby a p53-Upregulated Modulator ofApoptosis–Mediated PathwayTen days after MCAO, SVZ tissues were collectedfrom rats treated with PFT-� (n � 8) or vehicle (n �9) for qRT-PCR analysis (see Fig 3G). PFT-� did notalter the expression of Bcl2 (p � 0.95, t test) or Bax(p � 0.97, t test), but it significantly reduced mRNAlevels for PUMA (p � 0.028, t test). The difference inp21 expression was not significant (p � 0.21, t test;see Fig 3G). These data suggest that PFT-� exerts anantiapoptotic effect by inhibiting the p53-regulatedproapoptotic gene PUMA.

522 Annals of Neurology Vol 65 No 5 May 2009

Migration and Differentiation of NeuralProgenitor CellsRats (n � 29) were given BrdU systemically on day 5 andthen treated with PFT-� (n � 15) or vehicle (n � 14)

from days 6 to 9 after MCAO (Fig 4A). In the SVZ (seeFig 4B), a majority of BrdU-positive cells were present onday 10 (p � 0.001, two-way ANOVA); few cells werefound on day 21 after stroke. In the lesioned cortex (see

Fig 1. Kinetics of proliferation of neural progenitor cells (NPCs) in subventricular zone (SVZ) and apoptosis in the lesioned cortexafter ischemia. (A, B) Bromodeoxyuridine (BrdU) was injected systemically on days 2, 4, 6, 8, 10, 12, and 14 after a transientfocal ischemia induced by middle cerebral artery occlusion (MCAO) at the dose of 50mg/kg � 4 at 2-hour interval in 45 adultrats. Rats were killed 12 hours after the last dose of BrdU. Two sections (anterior [red line]: anteroposterior [AP] �0.2mm frombregma; posterior [black line]: AP �0.4mm from bregma) from each animal were counted. (A) Line graph represents the BrdUimmunoreactivity (optical density) of SVZ in all animals studied. Ischemia/reperfusion induced BrdU labeling both in the anteriorand posterior SVZ. The anterior is more active in BrdU incorporation than the posterior SVZ. In the anterior SVZ, the increase inBrdU immunoreactivity was sustained through 4 days after MCAO, started to decline at between days 6 and 8, and returned tobasal levels around day 10. (B) Photomicrographs demonstrate that BrdU immunoreactivity is differentially enhanced in the ante-rior and posterior SVZ from days 2 to 14 after MCAO. (C–E) Levels of apoptosis were measured by terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL) staining at the ischemic site in cortex on days 2, 6, and 9 after MCAO.(C) TUNEL labeling was enhanced on day 2 and returned to basal level less than 6 days after MCAO. Data are normalized tothe maximum response of TUNEL staining (2 days after MCAO); dotted line represents the level of TUNEL activity in sham-operated rats (n � 3). Administration of pifithrin-� (PFT-�; open circle) from days 6 to 9 did not alter TUNEL activity in theischemic cortex on day 9 (p � 0.266, t test). Solid circles represent vehicle. (D) Representative histology of TUNEL labeling inthe lesioned cortex after MCAO or sham surgery at 2 or 6 days after MCAO. An increase in TUNEL density was found at 48hours after MCAO. (E) Photomicrographs demonstrate that PFT-�, given from days 6 to 9, did not alter TUNEL activity in theischemic cortex on day 9. Scale bar � 200�m (B); 100�m (D, E).


Fig 4C), however, a significantly higher BrdU-positive celldensity was found on day 21 compared with day 10 (p �0.002, two-way ANOVA), suggesting the migration ofNPCs from SVZ to the lesioned cortex between days 10and 21. Almost no BrdU cells were present in the nonle-sioned side cortex on days 10 and 21 in all animals (datanot shown). Administration of PFT-� significantly in-creased density of BrdU-positive cell in the lesioned cortexon day 21 (see Fig 4C; p � 0.05, two-way ANOVA) and

in SVZ on day 10 (see Fig 4B; p � 0.05, two-wayANOVA). Typical BrdU labeling in the lesioned cortexon day 21 is shown in Figure 4E. Taken together, ourdata suggest that delayed treatments with PFT-� fromdays 6 to 9 facilitated the proliferation of NPCs in SVZ asseen on day 10 and resulted in an increase in migration ofthe NPCs to the injured cortex as seen on day 21.

We found that there is a significant correlation be-tween the functional outcomes and number of surviving

Fig 2. Delayed pifithrin-� (PFT-�) treatment improved locomotor behavior in stroke rats. (A) Timeline indicates that rats weresubjected to a 90-minute middle cerebral artery occlusion (MCAO) surgery on day 0. Vehicle or PFT-� was given from day 6 for4 consecutive days. Behavioral measurements were conducted in automated activity chambers for 24 hours at 2 (before administra-tion of PFT-� or vehicle), 14, and 21 days after MCAO. No difference was found before PFT-� or vehicle injection on day 2.Rats receiving poststroke PFT-� treatment showed enhanced recovery in motor function demonstrated by significant increases in (B)horizontal activity (total number of beam interruptions in the horizontal sensors), (C) total distance traveled, (D) number of move-ments, and (E) vertical activity (total number of beam interruptions in the vertical sensors), compared with the control group.(�p � 0.05, two-way analysis of variance; *p � 0.05, post hoc Newman–Keuls test). icv � intracerebroventricular; ip � intra-peritoneal. Red circles represent PFT-�; black triangles represent vehicle.


BrdU-positive cells in the lesioned cortex. BrdU wasgiven to 11 rats on day 5 after MCAO. Of these ani-mals, five were treated with vehicle and six were treatedwith PFT-� (see Fig 4A). Horizontal activity was mea-sured on days 14 and 21. Density of BrdU-positive cellsin the lesioned cortex was counted on day 21. The im-provement in horizontal activity (HACTV) on days 14and 21 was measured using the following equation: %improvement � (HACTV on day 14 or 21 � HACTVon day 2)/(HACTV on day 2) � 100%. There is a lin-ear correlation between the number of BrdU cells in thelesioned side cortex and HACTV improvement on days14 (r � 0.828, p � 0.002, black tracing) and 21 (r �0.644, p � 0.032, red tracing) after MCAO (see Fig4D). These data suggest that NPCs migrated to, and sur-vived in, the lesioned target (ie, cortex), contributing tothe improvement of locomotor function in stroke rats.

The differentiation of the NPCs in the lesionedcortex was examined using double-label immunostain-ing. Stroke rats treated with PFT-� (n � 6) or vehi-

cle (n � 6) were killed on day 21 after MCAO. Con-focal microscopic examination indicated that BrdU-positive cells in the lesioned side cerebral cortexexpressed the neuronal markers nestin, NeuN, ormicrotubule-associated protein 2 (Fig 5). Because ofthe diffuse staining pattern of GFAP (see Fig 5), itwas difficult to determine colocalization of BrdU andGFAP. PFT-� treatment significantly increased thedensity of BrdU cells coexpressing nestin ( p � 0.011,t test), but not microtubule-associated protein 2 ( p �0.17, t test), in the lesioned cortex. There was a mar-ginal increase in NeuN/BrdU cell density ( p �0.075, t test) after PFT-� administration. Using moreanimals may lead to a significant change.

Pifithrin-� Increases the Self-Renewal ofSubventricular Zone Progenitor Cell ProliferationIn VitroSVZ cells from rodent embryos were isolated and cul-tured in epidermal growth factor–containing media to

Fig 3. Delayed pifithrin-� (PFT-�) treatment promotes the proliferation of neural progenitor cells (NPCs) and inhibits apoptosis inthe subventricular zone (SVZ) in stroke rats. (A). Rats received PFT-� or vehicle from days 6 to 9 after middle cerebral arteryocclusion (MCAO). Bromodeoxyuridine (BrdU) was administered daily for 3 days after PFT-� injection. (B) PFT-� treatmentincreased the density of BrdU immunoreactivity (optical density) in SVZ (�p � 0.05, two-way analysis of variance [ANOVA]).The increase in BrdU labeling was more prominent in the anterior SVZ (*p � 0.05, post hoc Newman–Keuls test, two-wayANOVA). (C) Representative photomicrographs of the anterior SVZ indicate that BrdU labeling was enhanced in the stroke ani-mals treated with PFT-� (red circles). Black triangles represent vehicle. Calibration � 100�m. (D). Representative photomicro-graphs of the anterior subependymal zone indicate that density of proliferating cell nuclear antigen (PCNA)–positive cell increasedin the stroke animals treated with PFT-�. (panels 2 and 4, from the far left) compared with vehicle-treated animals (panels 1 and3) on day 10 after MCAO. Panels 3 and 4 are images at higher magnification. Scale bar � 50�m. (E) PFT-� significantly en-hanced the number of PCNA-positive cells at the anterior portion of the subependymal zone (*p � 0.05, t test). (F) DelayedPFT-� treatment significantly suppressed terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) in theSVZ on day 10 after MCAO (�p � 0.05, two-way ANOVA). (G) PFT-� suppresses the expression of p53-upregulated modulatorof apoptosis (PUMA). SVZ tissues were harvested from PFT-�– or vehicle-treated stroke rats on the fourth day of PFT-� treat-ment. p53-regulated genes (messenger RNA [mRNA] levels) were determined by quantitative reverse transcriptase polymerase chainreaction (qRT-PCR). All the genes were normalized to the housekeeping gene phosphoglycerate kinase 1 (PGK1) mRNA level.PFT-� did not change the level of Bcl2 or Bax genes. However, it decreased the mRNA levels of PUMA (*p � 0.05, t test).Black bars represent vehicle; red bars represent PFT-�. icv � intracerebroventricular; ip � intraperitoneal.


form neurospheres. Cells were plated at postnatal day2. PFT-� or vehicle was added either on DIV1 orDIV3 at various concentrations; the total number ofneurospheres was counted on DIV7. PFT-� on DIV1significantly (Fig 6A; p � 0.05, one-way ANOVA)and dose-dependently increased the number of neu-rosphere formations ( p � 0.001, r � 0.528, linearregression). There was no significant change in thedensity of neurospheres when PFT-� was given onDIV3, suggesting a time window for the trophic ef-fect of PFT-� in cultured SVZ progenitor cells (seeFig 6A). A significant difference was also observed be-

tween the size of spheres receiving PFT-� and vehicletreated from DIV1 (see Figs 6B, C; p � 0.05, one-way ANOVA).

DiscussionIn this study, we have demonstrated a delayed treat-ment strategy for stroke. Administration of PFT-�, ap53 inhibitor, starting from the sixth day after MCAO,enhanced the proliferation and migration of endoge-nous NPCs in the SVZ, and more importantly, it im-proved motor function after ischemia.

Although PFT-� can pass through the blood–brain

Fig 4. Delayed pifithrin-� (PFT-�) treatment increased the survival and migration of newly generated neural progenitor cells(NPCs) in the lesioned cortex. (A) Rats were subjected to middle cerebral artery occlusion (MCAO) on day 0. Newly generated cellswere prelabeled with BrdU on day 5 after stroke. PFT-� or vehicle was given 1 day after the BrdU labeling. Density of BrdUimmunoreactivity in (B) subventricular zone (SVZ) or (C) cortex on days 10 and 21 after MCAO was normalized to the level ofBrdU activity in vehicle group on day 10. (B) In the SVZ, BrdU-labeled cells were mainly present on day 10 after MCAO. FewBrdU cells were found on day 21. PFT-� treatment significantly increased the density of BrdU cells on day 10 in SVZ. (C) In thelesioned cortex, fewer BrdU cells were found on day 10. Significantly increased BrdU labeling was found on day 21. PFT-� treat-ment enhanced BrdU labeling in the lesioned cortex on days 10 and 21 after MCAO. (D) Correlation of the functional outcomeswith migration of NPCs to the lesioned cortex in stroke rats. The behavioral improvement in horizontal activity (HACTV) on days14 and 21 was calculated using the following equation: % improvement � (HACTV on day 14 or 21 � HACTV on day 2)/(HACTV on day 2) � 100%. BrdU (�) cells in the lesioned side cortex was examined on day 21 after MCAO. There is linercorrelation between the number of BrdU cells in cortex and HACTV improvement at 14 (r � 0.828, p � 0.002, black tracing)and 21 days (r � 0.644, p � 0.032, red tracing) after MCAO. (E) Photomicrographs indicate that PFT-� treatment increasedBrdU-labeled cells in the lesioned cortex on day 21 after MCAO. Calibration � 50�m. icv � intracerebroventricular; ip � in-traperitoneal. Black bars and black circles represent vehicle; red bars and red circles represent PFT-�.


barrier when given systemically,7 we found that sys-temic administration of PFT-� alone, at the dose of0.2mg/100gm (twice a day intraperitoneally), was notenough to provide significant behavioral recovery inour pilot study. A loading intracerebroventricular doseof PFT-�, which increases PFT-� levels in the brainrapidly without dramatically increasing PFT-� levelsperipherally, was required to induce significant behav-ioral recovery.

Ischemic brain injury activates apoptotic cascades atthe lesioned site.20 We found that the density ofTUNEL labeling in the ischemic cortex peaked onday 2 after MCAO. Similarly, it has been reportedthat p53 mRNA or protein is upregulated shortly af-ter stroke, which leads to p53-dependent pro-grammed cell death in the penumbra.21,22 Less focaldamage is found in stroke animals pretreated with ap53 inhibitor7 or in p53 knock-out mice.23 Thesedata suggest that p53 is an important mediator ofprogrammed cell death in the ischemic region. As the

initiation of programmed cell death in the ischemicregion occurs rapidly after stroke, Leker and col-leagues8 reported that the effectiveness of p53 inhibi-tion is observed only when an inhibitor is given be-fore or shortly after MCAO. PFT-� given 3 hours orlater after MCAO had no beneficial effect on neuro-nal cell loss.8 Using the same treatment regimen thatLeker and colleagues8 described, we also found thatPFT-�, given at 3 hours after MCAO as a single dosedid not alter locomotor behaviors on day 2 afterMCAO. There was also no difference in BrdU label-ing in the SVZ between the stroke animals treatedwith PFT-� and vehicle at 24 hours after the lastdose of BrdU injection. This temporal restriction haslargely limited the clinical potential of the drug. Con-sistent with these observations, we also found thatTUNEL labeling in the ischemic cortex returned tobasal levels on day 6 after MCAO, suggesting that thewindow of apoptosis/necrosis in the ischemic core/penumbra is less than 6 days. Because PFT-� was

Fig 5. Differentiation of endogenous neural progenitor cells (NPCs) in the lesioned cortex in stroke rats receiving pifithrin-�(PFT-�) treatment. Animals were killed on day 21 after middle cerebral artery occlusion (MCAO). Double immunostaining forbromodeoxyuridine (BrdU) (middle) and markers for neuronal nuclear antigen (NeuN), microtubule-associated protein 2 (MAP2),nestin, as well as glial fibrillary acidic protein (GFAP) (top) were analyzed by confocal microscopy. The newly generated cells la-beled by BrdU coexpressed NeuN. BrdU-positive nuclei were also surrounded by the cytosol, which contains MAP2 or nestin (ar-rows, bottom). Calibration � 100�m.


given from day 6 in this study, the improvement ofmotor function is not attributable to the suppressionof apoptosis at the lesioned site in stroke rats; this isfurther supported by the finding that delayed treat-ment with PFT-� from days 6 to 9 did not alterTUNEL labeling in the ischemic cortex, as seen inFigure 1E.

Ischemic brain injury induces SVZ neurogenesis inthe adult brain.5,24 We observed that an increase inBrdU labeling in SVZ was sustained from days 2 to 8after MCAO and returned to basal levels on day 10.These data suggest that the window of new cell pro-liferation at SVZ is longer than that of apoptosis inischemic penumbra. Despite the increase in new NPCproliferation in SVZ, without p53 inhibition, the ma-jority of these cells die before differentiating into a

functional phenotype. It is thought that the lesionedbrain tissue does not provide a suitable environmentfor the survival of these newly generated cells.20

We administered PFT-� from days 6 to 9 afterMCAO, a temporal window when ischemia-inducedNPC proliferation in SVZ, but not apoptosis in thepenumbra, is still ongoing. This allows differentiationof the effect of p53 inhibition at the lesioned penum-bra compared with SVZ. We found that deliveringPFT-� enhanced behavioral outcome in multiple pa-rameters of locomotor activity in the stroke animals atboth 1 and 2 weeks after treatment. Because PFT-�did not augment locomotor behavior in nonstroke rats,the effect of PFT-� on promoting behavioral recoveryappears specific to the stroke animals. Previous studiesindicated that PFT-�, given at 3 hours after MCAO,did not reduce cell death in stroke rats.8 The behav-ioral improvement induced by delayed PFT-� treat-ment in this study is thus probably not because of res-cue of the injured cells at the penumbra, but rather toaugmenting the proliferation or survival of NPCs inthe SVZ cells.

We found that BrdU labeling is enhanced in strokeanimals receiving delayed PFT-� treatment, suggest-ing that PFT-� enhanced new cell proliferation in theSVZ after ischemic insults. It has been suggested thatBrdU can be transferred to other cells over time byphagocytosis of dying cells. For example, Coyne andcoworkers25 report BrdU labeling can be transferredfrom prelabeled viable or control nonviable donormarrow stromal cells, derived from green fluorescentprotein transgenic rats, to phagocytes, astrocytes, andneurons in host brain of Sprague–Dawley rats. In-flammation was found at the transplantation site. Nu-merous blood vessels in the surrounding graft regionswere dilated, with cells exhibiting monocytic, lym-phocytic, and granular nuclear morphology. These re-sults suggest that exogenous marrow stromal cellstransplanted to the intact adult brain were rejected byan inflammatory response. Coyne and coworkers25

thus caution against the continued use of BrdU asdonor cell labels. However, in this study, the increasein BrdU labeling in the SVZ observed in animalstreated with PFT-� is not due to the phagocytosis ofdying cells based on the following four reasons: (1) Incontrast with the transplantation of exogenous cells,we enhanced proliferation of endogenous progenitorcells by PFT-� in this study. Because no exogenouscells were grafted, almost no inflammation was foundin the SVZ. (2) Brain trauma, such as intracerebro-ventricular injection, can induce inflammation. Wefound that PFT-� did not alter BrdU labeling nearthe needle tract in cerebral cortex (unpublished ob-servation). (3) The proliferation of SVZ was also con-firmed by another marker. We found that the immu-noreactivity of PCNA, an endogenous protein that is

Fig 6. Pifithrin-� (PFT-�) increased the proliferation/survivalof subventricular zone (SVZ) progenitor cells in vitro. (A)PFT-� was added to the SVZ progenitor cell culture on day 1in vitro (DIV1; black circles) or DIV3 (gray triangles). Den-sity of neurospheres was counted on DIV7. Administration ofPFT-� on DIV1, but not on DIV3, dose-dependently in-creased number of neurospheres as compared with the non-treatment controls (*p � 0.05, one-way analysis of variance[ANOVA] � post hoc Newman–Keuls test). (B) The spheresreceiving PFT-� treatment on DIV1 also grew significantlylarger compared with vehicle-treated sphere cultures (*p �0.05, one-way ANOVA � post hoc Newman–Keuls test). (C)Representative images of neurospheres treated with PFT-� orvehicle on DIV1. Calibration � 100�m.


specifically expressed during the cell proliferation cy-cle, was enhanced after PFT-� treatment in strokerats (see Fig 3E). (4) There is a functional correlationbetween BrdU incorporation and behavioral recovery.

We found that administration of PFT-� increasedNPC proliferation and survival of NPCs in the SVZafter MCAO. Selective regulation of p53 target genes iscritical for the different cellular outcomes mediated byp53, and under physiological/pathological conditions,p53 can regulate different apoptosis-related and cell-cycle arrest genes.26 It has been shown that PFT-� in-hibits p53 function in vivo by blockade of the translo-cation of p53 to the nucleus and mitochondria,8 andby subsequent suppression of the expression of pro-apoptotic genes.27 In this study, the expression of pro-apoptotic genes in the SVZ is, likewise, modulated byPFT-� in stroke rats. We examined the gene expres-sion level in SVZ for known p53-regulated genes thatare involved in neuronal apoptosis, such as Bax/Bcl2,9

apoptotic protease-activating factor 1,28 and PUMA.29

PFT-� inhibited the p53-dependent proapoptoticgene, PUMA, a gene that is upregulated by hypoxia30

and can lead to caspase activation.29 Because deletionof PUMA attenuated cell death and improved physio-logical function after ischemia-reperfusion injury,30 itis possible that PFT-� promotes the survival of SVZprogenitor cells, at least in part, through the inhibitionof expression of PUMA.

p21 is another downstream target gene for p53. De-ficiencies in p21 affect stem cell proliferation inmice.31 p21 gene expression is upregulated after isch-emia.31,32 Deficiencies in p21 increased SVZ progeni-tor cell proliferation in mice after ischemic injury.31

However, in our study, there was not a significant dif-ference in the p21 expression level in SVZ after PFT-�treatment in stroke animals.

Endogenous NPCs can migrate from the SVZ tothe damaged areas after MCAO.33 Our data demon-strated that the migration of NPCs to the lesionedcortex is associated with functional recovery. We alsoobserved a differential distribution of BrdU-positivecells in the SVZ and cortex on days 10 and 21 afterstroke. A higher density of BrdU cells was present inthe SVZ on day 10, whereas increased BrdU immu-noreactivity was found in the lesioned cortex on day21. This difference may be attributed to the migra-tion of NPCs from SVZ to the lesioned cortex be-tween days 10 and 21. Loss of p53 in p53 knock-out(�/�) mice confers a proliferative advantage tothe cells in the adult SVZ and favors the progressiontoward a differentiated phenotype.32 We also foundthat PFT-� enhanced proliferation of NPCs and in-hibited cell death/necrosis in the SVZ on day 10,which may secondarily facilitate the migration anddifferentiation of NPCs as the densities of BrdU,Nestin/BrdU, and NeuN/BrdU (this was marginal,

p � 0.075) cells in the lesioned cortex were increasedby PFT-� on day 21 after MCAO.

We found that PFT-� dose- and time-dependentlyincreased the number and size of new neurosphereformations in in vitro cultures. Spheres that receivedPFT-� treatment on DIV1 grew larger comparedwith vehicle-treated sphere cultures, consistent withthe report that SVZ cells extracted from p53�/� miceshowed an increased number of neurosphere formations,as well as increased size in the spheres formed.34 Thesedata suggest that the positive effect of PFT-� on NPCsfrom SVZ is reflected in survival and growth in vitro.

In conclusion, our results support an effect ofPFT-� in enhancing neurogenesis by promoting pro-liferation and survival of the endogenous NPCs in astroke model. In contrast with previous studies usingPFT-� at a much earlier and narrower time frame,our results demonstrate a significant behavioral recov-ery at much later time points, days after the ischemiaoccurs. Our results suggest a new potential treatmentstrategy for stroke patients, enabling a longer treat-ment window after onset of stroke, a disorder fre-quently difficult to treat immediately after occur-rence.

This study was supported by the Intramural Research Programs atthe NIH (National Institute on Drug Abuse; National Institute ofAging).

We thank Dr T. Hayashi for his technical assistance.

References1. Goldstein LB. Acute ischemic stroke treatment in 2007. Circu-

lation 2007;116:1504–1514.2. Gilman S. Pharmacologic management of ischemic stroke: rel-

evance to stem cell therapy. Exp Neurol 2006;199:28–36.3. Darsalia V, Kallur T, Kokaia Z. Survival, migration and neu-

ronal differentiation of human fetal striatal and cortical neuralstem cells grafted in stroke-damaged rat striatum. Eur J Neu-rosci 2007;26:605–614.

4. Abrahams JM, Gokhan S, Flamm ES, Mehler MF. De novoneurogenesis and acute stroke: are exogenous stem cells reallynecessary? Neurosurgery 2004;54:150–156.

5. Arvidsson A, Collin T, Kirik D, et al. Neuronal replacementfrom endogenous precursors in the adult brain after stroke. NatMed 2002;8:963–970.

6. Zhu X, Yu QS, Cutler RG, et al. Novel p53 inactivators withneuroprotective action: syntheses and pharmacological evalua-tion of 2-imino-2,3,4,5,6,7-hexahydrobenzothiazole and2-imino-2,3,4,5,6,7-hexahydrobenzoxazole derivatives. J MedChem 2002;45:5090 –5097.

7. Culmsee C, Zhu X, Yu QS, et al. A synthetic inhibitor of p53protects neurons against death induced by ischemic and excito-toxic insults, and amyloid beta-peptide. J Neurochem 2001;77:220–228.

8. Leker RR, Aharonowiz M, Greig NH, Ovadia H. The role ofp53-induced apoptosis in cerebral ischemia: effects of the p53inhibitor pifithrin alpha. Exp Neurol 2004;187:478–486.

9. Culmsee C, Mattson MP. p53 in neuronal apoptosis. BiochemBiophys Res Commun 2005;331:761–777.


10. Meletis K, Wirta V, Hede SM, et al. p53 suppresses the self-renewal of adult neural stem cells. Development 2006;133:363–369.

11. Felling RJ, Levison SW. Enhanced neurogenesis followingstroke. J Neurosci Res 2003;73:277–283.

12. Wang Y, Chang CF, Morales M, et al. Diadenosine tetraphos-phate protects against injuries induced by ischemia and6-hydroxydopamine in rat brain. J Neurosci 2003;23:7958–7965.

13. Chen ST, Hsu CY, Hogan EL, et al. A model of focal ischemicstroke in the rat: reproducible extensive cortical infarction.Stroke 1986;17:738–743.

14. Tomac AC, Agulnick AD, Haughey N, et al. Effects of cerebralischemia in mice deficient in Persephin. Proc Natl Acad SciU S A 2002;99:9521–9526.

15. Chou J, Harvey BK, Chang CF, et al. Neuroregenerative effectsof BMP7 after stroke in rats. J Neurol Sci 2006;240:21–29.

16. Hoglinger GU, Rizk P, Muriel MP, et al. Dopamine depletionimpairs precursor cell proliferation in Parkinson disease. NatNeurosci 2004;7:726–735.

17. Deng X, Wang Y, Chou J, Cadet JL. Methamphetamine causeswidespread apoptosis in the mouse brain: evidence from usingan improved TUNEL histochemical method. Brain Res MolBrain Res 2001;93:64–69.

18. Chou J, Luo Y, Kuo CC, et al. Bone morphogenetic protein-7reduces toxicity induced by high doses of methamphetamine inrodents. Neuroscience 2008;151:92–103.

19. Curtis MA, Penney EB, Pearson AG, et al. Increased cell pro-liferation and neurogenesis in the adult human Huntington’sdisease brain. Proc Natl Acad Sci U S A 2003;100:9023–9027.

20. Wiltrout C, Lang B, Yan Y, et al. Repairing brain after stroke:a review on post-ischemic neurogenesis. Neurochem Int 2007;50:1028–1041.

21. Li Y, Chopp M, Zhang ZG, et al. p53-immunoreactive proteinand p53 mRNA expression after transient middle cerebral ar-tery occlusion in rats. Stroke 1994;25:849–856.

22. Li Y, Chopp M, Powers C, Jiang N. Apoptosis and proteinexpression after focal cerebral ischemia in rat. Brain Res 1997;765:301–312.

23. Crumrine RC, Thomas AL, Morgan PF. Attenuation of p53expression protects against focal ischemic damage in transgenicmice. J Cereb Blood Flow Metab 1994;14:887–891.

24. Okano H, Sakaguchi M, Ohki K, et al. Regeneration of thecentral nervous system using endogenous repair mechanisms.J Neurochem 2007;102:1459–1465.

25. Coyne TM, Marcus AJ, Woodbury D, Black IB. Marrow stro-mal cells transplanted to the adult brain are rejected by an in-flammatory response and transfer donor labels to host neuronsand glia. Stem Cells 2006;24:2483–2492.

26. Das S, Boswell SA, Aaronson SA, Lee SW. P53 promoter selection:choosing between life and death. Cell Cycle 2008;7:154–157.

27. Duan W, Zhu X, Ladenheim B, et al. p53 inhibitors preservedopamine neurons and motor function in experimental parkin-sonism. Ann Neurol 2002;52:597–606.

28. Fortin A, Cregan SP, MacLaurin JG, et al. APAF1 is a keytranscriptional target for p53 in the regulation of neuronal celldeath. J Cell Biol 2001;155:207–216.

29. Puthalakath H, Strasser A. Keeping killers on a tight leash:transcriptional and post-translational control of the pro-apoptotic activity of BH3-only proteins. Cell Death Differ2002;9:505–512.

30. Toth A, Jeffers JR, Nickson P, et al. Targeted deletion of Pumaattenuates cardiomyocyte death and improves cardiac functionduring ischemia-reperfusion. Am J Physiol Heart Circ Physiol2006;291:H52–H60.

31. Qiu J, Takagi Y, Harada J, et al. Regenerative response in isch-emic brain restricted by p21cip1/waf1. J Exp Med 2004;199:937–945.

32. Tomasevic G, Shamloo M, Israeli D, Wieloch T. Activation ofp53 and its target genes p21(WAF1/Cip1) and PAG608/Wig-1in ischemic preconditioning. Brain Res Mol Brain Res 1999;70:304–313.

33. Zhang R, Zhang Z, Wang L, et al. Activated neural stem cellscontribute to stroke-induced neurogenesis and neuroblast mi-gration toward the infarct boundary in adult rats. J CerebBlood Flow Metab 2004;24:441–448.

34. Gil-Perotin S, Marin-Husstege M, Li J, et al. Loss of p53 induceschanges in the behavior of subventricular zone cells: implicationfor the genesis of glial tumors. J Neurosci 2006;26:1107–1116.


Delayed treatment with a p53 inhibitor enhances recovery in stroke brain

Documents

Transcript of Delayed treatment with a p53 inhibitor enhances recovery in stroke brain