Neurogenesis in the adult dentate gyrus after cortical infarcts: Effects of infarct location, N-methyl-d-aspartate receptor blockade and anti-inflammatory treatment

NCNA

MCa

Ab

U

AigddppwlaAwtbsanwoapNdinipidNtr

Kp

*EADcncc

Neuroscience 135 (2005) 723–735

0d

EUROGENESIS IN THE ADULT DENTATE GYRUS AFTERORTICAL INFARCTS: EFFECTS OF INFARCT LOCATION,-METHYL-D-ASPARTATE RECEPTOR BLOCKADE
ND ANTI-INFLAMMATORY TREATMENT
TrN(dnHfise2m2cle2eav((wtpesidlhst

eaD

aNmm(Fng

. M. KLUSKA,a O. W. WITTE,a J. BOLZb AND. REDECKERa*

Department of Neurology, Friedrich-Schiller-University, Erlangerllee 101, D-07747 Jena, Germany

Department of Zoology and Animal Physiology, Friedrich-Schiller-niversity, Erbertstrasse 1, D-07743 Jena, Germany

bstract—Stimulation of cell proliferation and neurogenesisn the adult dentate gyrus has been observed after focal andlobal brain ischemia but only little is known about the un-erlying mechanisms. We here analyzed neurogenesis in theentate gyrus after small cortical infarcts leaving the hip-ocampal formation and subcortical regions intact. Using thehotothrombosis model in adult rats, focal ischemic infarctsere induced in different cortical areas (sensorimotor fore-

imb and hindlimb cortex) and proliferating cells were labeledt days 3–14 after infarct induction with bromodeoxyuridine.t 2, 4, and 10 weeks after ischemia, immunocytochemistryas performed with immature neuronal (doublecortin), ma-

ure neuronal (neuronal nuclei antigen) and glial (calcium-inding protein beta S100�) markers. When compared withham-operated controls, animals with infarcts in the forelimbs well as hindlimb cortex revealed an increase in survival ofewborn progenitor cells at four and 10 weeks after the insultith predominance at the ipsilateral side. Triple immunoflu-rescence and confocal laser scanning microscopy revealedn increase in neurogenesis in all groups that was moreronounced 10 weeks after the infarct. Application of the-methyl-D-aspartate (NMDA)-receptor antagonist MK-801uring lesion induction significantly enhanced neurogenesis

n the dentate gyrus. An even stronger increase in newborneurons was observed after anti-inflammatory treatment with

ndomethacine during the first 16 days of the experiment. Theresent study demonstrates that small cortical infarcts leav-

ng subcortical structures intact increase neurogenesis in theentate gyrus and that these processes can be stimulated by-methyl-D-aspartate receptor blockade and anti-inflamma-

ory treatment. © 2005 IBRO. Published by Elsevier Ltd. Allights reserved.

ey words: focal cerebral ischemia, indomethacine, MK-801,hotothrombosis, progenitor cells, stroke.

Corresponding author. Tel: �49-3641-9323430; fax: �49-3641-9323402.-mail address: [email protected] (C. Redecker).bbreviations: BrdU, bromodeoxyuridine; COX, cyclooxygenase;CX, doublecortin; DG, dentate gyrus; FL, forelimb representationortex; HL, hindlimb representation cortex; MK-801, dizocilpine; NeuN,euronal nuclei antigen; NMDA, N-methyl-D-aspartate; Oc2, occipital

2ortex; PFA, paraformaldehyde; SGZ, subgranular zone; S100�, cal-ium-binding protein beta; TBS, Tris-buffered saline.

306-4522/05$30.00�0.00 © 2005 IBRO. Published by Elsevier Ltd. All rights reseroi:10.1016/j.neuroscience.2005.06.082

723

he mammalian dentate gyrus (DG) is one of two brainegions where adult neurogenesis occurs (Gage, 2000).eural progenitor cells arising from the subgranular zone

SGZ) migrate into the granule cell layer (GCL) where theyifferentiate into new neurons and integrate into the localetwork (Kee et al., 2001; van Praag et al., 2002; Schmidt-ieber et al., 2004). The physiological function of the newly

ormed neurons has not been completely clarified but ev-dence shows that they might play an important role inpatial learning and memory (Shors et al., 2001; Dobrossyt al., 2003; Ambrogini et al., 2004; Bruel-Jungerman et al.,005). Physiological factors such as enriched environ-ent, running, and learning (Ehninger and Kempermann,003; Leuner et al., 2004) influence the dynamic pro-esses of neurogenesis in the SGZ but also pathophysio-

ogical factors such as global and focal brain ischemia (Jint al., 2001; Sharp et al., 2002), epileptic seizures (Parent,003; Scharfman, 2004), and traumatic brain injury (Ricet al., 2003) strongly increase neurogenesis (Arvidsson etl., 2001; Scharfman, 2004). Most ischemia models in-olve cortical and subcortical brain regions with directHerguido et al., 1999; Sasaki et al., 2004) or indirectWang et al., 2004) hippocampal damage. It is unclearhether DG neurogenesis is also stimulated by pure cor-

ical infarcts without direct or indirect damage to the hip-ocampal formation and whether the location of the isch-mic insult influences the proliferative response. In thistudy we induced circumscribed small cortical infarcts us-

ng the low invasive photothrombosis model in rats, firstescribed by Watson et al. (1985). These infarcts were

ocated in distinct cortical areas either affecting parts of theindlimb representation cortex (HL), the occipital, and theomatosensory cortex or the forelimb representation cor-ex (FL) in another group of animals.

The underlying mechanisms causing adult neurogen-sis under physiological and pathophysiological conditionsre widely unknown. Several studies indicate that N-methyl--aspartate (NMDA) receptors are involved (Cameron etl., 1995; Arvidsson et al., 2001). However, the role ofMDA receptors seems to be ambiguous. Following phar-acological blockade of NMDA receptors both, enhance-ent (Nacher et al., 2001, 2003) as well as reduction

Arvidsson et al., 2001) of neurogenesis was observed.urthermore, inflammatory processes modulate the rate ofeurogenesis by changing the microenvironment of pro-enitor cells in the SGZ (Monje et al., 2003; Ekdahl et al.,
003). We here analyzed the effects of pharmacological
ved.

beithl

cntgatlBasm

I

TraHataiaSwarbGd

o(4ItdmtaawamocbptaTttsddm

Ap

P(b5eoapmc

T

S

S

P

A

a

M. M. Kluska et al. / Neuroscience 135 (2005) 723–735724

lockade of NMDA receptors on infarct-induced neurogen-sis in the DG by application of MK-801 during lesion

nduction. We further examined whether anti-inflammatoryreatment with the unselective cyclooxygenase (COX) in-ibitor indomethacine influences adult neurogenesis fol-

owing cortical infarcts.The present study demonstrates that even small corti-

al infarcts leaving the hippocampal formation intact sig-ificantly enhance neurogenesis and long-term survival of

hese newborn cells in the SGZ. This increase in neuro-enesis occurred after infarcts in different cortical areas,lthough the extent of this stimulation was stronger when

he cortical infarcts were located in the sensorimotor hind-imb cortex compared with more frontal infarct positions.lockade of NMDA-dependent neurotransmission as wells anti-inflammatory treatment with indomethacinetrongly increased post-ischemic neurogenesis in thisodel.

EXPERIMENTAL PROCEDURES

nduction of focal cortical infarcts

he experiments were carried out on a total of 74 male Wistarats (280 –310 g, 12–14 weeks). The study was performed inccordance with the guidelines of the National Institutes ofealth (Revised 1987) and all experimental procedures werepproved by the German Animal Care and Use Committee. Aotal of 44 animals received photothrombotic infarcts and 30nimals served as controls. Cortical infarcts were induced us-

ng the photothrombosis model initially described by Watson etl. (1985) with only minor modifications (Redecker et al., 2002;hanina et al., 2005). Briefly, the animals were anesthetizedith 2.5–3.5% enflurane in a mixture of O2/N2O. Body temper-ture was kept constant throughout the surgery at 36.5 °C. Theats were placed in a stereotactic frame and an optic fiberundle mounted on a cold light source (Schott KL 1500, Jena,ermany) was positioned at defined skull-coordinates. For in-

able 1. Total BrdU-positive cell numbers and colocalization of BrdU.

urgery 2 Weeks

IpsilateralMean�SD

ContralateralMean�SD

(n�4)ham BrdU� total 3370.0�299.3 3052.0�282.8

NeuN� (%) 9.0�5.2 9.5�7.2Neither (%) 90.5�5.5 87.5�6.8S100�� (%) 0.5�1.0 3.0�1.1

(n�5)osterior BrdU� total 3444.0�1931.2 2774.4�1243.8

NeuN� (%) 7.6�5.3 8.0�5.1Neither (%) 91.2�4.8 92.8�4.1S100�� (%) 1.2�1.8 1.2�1.8

nterior BrdU� total —�— —�—NeuN� (%) —�— —�—Neither (%) —�— —�—S100�� (%) —�— —�—

Values represent mean�standard deviation. Statistical significasterisks (* P�0.05 or ** P�0.01).

ucing posterior ischemic lesions, affecting parts of the HL, the b

ccipital cortex (Oc2) and the somatosensory cortex (Par1)Zilles, 1985), the optic fiber bundle was positioned exactlymm posterior to bregma and 4 mm lateral to midline (n�29).

n another group of animals anterior ischemic lesions, affectinghe FL cortex, were induced by positioning the optic fiber bun-le exactly 0.3 mm anterior to bregma and 4 mm lateral toidline (n�15). Immediately following onset of illumination (to-

al time: 20 min), Rose Bengal (1.3 mg/100 mg body weight atconcentration of 10 mg/ml in 0.9% NaCl) was injected throughfemoral vein catheter. Following this surgery, the animals

ere returned to their cages with free access to ground, foodnd water. Sham-operated animals received the same treat-ent without illumination of the brain (n�30). In a further groupf 10 animals (photothrombotic infarcts: n�5, sham-operatedontrols: n�5) the NMDA receptor antagonist MK-801 (2 mg/kgody weight, Sigma-Aldrich, Taufkirchen, Germany) was ap-lied intraperitoneally 30 min prior to illumination. Additionally,he blockade of inflammatory processes was analyzed by i.p.pplication of the COX inhibitor indomethacine (Sigma-Aldrich,aufkirchen, Germany) in another group of 11 animals (photo-

hrombotic infarcts: n�6; sham-operated controls: n�5). Tohis purpose indomethacine (2.5 mg/kg body weight) was dis-olved in lightly alkaline saline and injected twice daily for 16ays beginning 1 day prior to infarct induction. A detailedescription of the different numbers of animals in the experi-ental and control groups is shown in Table 1 and Table 2.

pplication of bromodeoxyuridine (BrdU) and tissuerocessing

roliferating cells were marked using the thymidine analog BrdUSigma-Aldrich, Taufkirchen, Germany) dissolved in 0.9% Tris-uffered saline (TBS). All rats received a daily i.p. injection of0 mg/kg 5-bromodeoxyuridine (BrdU) at days 3–14 after isch-mia or sham surgery. Animals were allowed to recover for 2, 4,r 10 weeks after infarct induction. The rats were then deeplynesthetized with diethylether and perfused with 4% ice-coldhosphate buffered paraformaldehyde (PFA). Brains were re-oved immediately after perfusion and postfixed in 4% PFA. For

ryoprotection the tissue was then transferred to a phosphate-

nd S100�

s 10 Weeks

alD




(n�6) (n�10)358.8 1983.3�176.8 2016.0�767.5 1919.2�629.78.7 63.4�6.3 74.6�8.7 71.8�8.08.7 36.5�6.3 25.2�9.0 28.2�8.00 0�0 0.2�0.6 0�0

(n�7) (n�6)738.4* 2128.8�543.1 2853.3�178.8* 2685.3�273.8*8.3 59.3�11.1 85.7�5.7** 82.7�12.8*8.6 40.5�11.7 12.6�5.3** 17�13*1.5 0.3�0.7 1.7�2.0* 0.3�0.8

(n�7) (n�8)491.5** 2344.6�565.4 2399.0�765.2 2176.5�748.811.6 58.7�11.6 83.4�6.2* 82.3�4.1**11.4 40.7�11.1 14.6�6.2* 18.5�5.4**1.0 0.5�0.8 2.0�2.3* 1.1�1.6*

ween infarcted animals and appropriate controls is indicated by

NeuN a

4 Week

IpsilaterMean�S

1083.3�

62.0�

38.0�

0�

2632.4�

61.0�

38.0�

1.0�

2683.4�

60.0�

38.2�

0.8�

nce bet

uffered saline containing 30% sucrose for 24 h and stored at

�i

I

BadfirbX(Liaaf(

3dgs

pdanri(0wXdgaP

T

S

S

P

( indomet

Fh1ds

M. M. Kluska et al. / Neuroscience 135 (2005) 723–735 725

75 °C for further processing. The experimental design and BrdUnjection protocol are illustrated in Fig. 1.

mmunohistochemistry

rains were cut on a freezing microtome into 40 �m coronal slicesnd collected in a 0.1 mol/l phosphate-buffered saline. For peroxi-ase staining, the free-floating sections were denaturated in 2 N HClor 30 min at 37 °C, rinsed in 0.1 M boric acid (pH 8.5) for 10 min andncubated in 1% H2O2 for 30 min. Between these steps sections wereinsed three times in TBS. The sections were then incubated inlocking solution for 30 min (3% donkey serum and 0.1% Triton-100 in TBS) followed by incubation with the primary antibody

monoclonal rat anti-BrdU IgG, 1: 500, Harlan Sera-Laboratory,oughborough, UK) in blocking solution at 4 °C overnight. Before

ncubation in the secondary antibody (Biotin-Sp-conjugated donkeynti-rat IgG, Dianova, Hamburg, Germany) for 2 h at room temper-ture, sections were rinsed in TBS and incubated in blocking serumor 30 min. After rinsing the slices were treated with ABC solutionVectastain Elite Kid, Vector Laboratories, Burlingame, CA, USA) for

able 2. Total BrdU-positive cell numbers and colocalization of BrdU,

urgery MK-801


(n�

ham BrdU� total 2126.4�204.1NeuN� (%) 84.8�4.1Neither (%) 35.2�4.1S100�� (%) 0.0�0.0

(n�

osterior BrdU� total 3337.8�748.4**NeuN� (%) 70.8�10.3Neither (%) 28.8�9.6S100�� (%) 0.4�0.9

Values represent mean�standard deviation. Statistical significance* P�0.05 or ** P�0.01). Statistical significance between MK-801 and

ig. 1. Schematic illustration of the experimental design and BrdU inindlimb and visual cortex (posterior lesions) and sham-operated contr0 weeks). (B) Animals with photothrombotic infarcts in the forelimb and
ivided into two groups with survival periods of 4 and 10 weeks. All rats receivurgery.
0 min at room temperature. Sections were then processed usingiaminobenzidine hydrochloride (DAB; Sigma-Aldrich) as chromo-en and mounted onto gelatin-coated slides, air-dried and cover-lipped (Entellan; Merck, Darmstadt, Germany).

For triple immunofluorescence staining the brain slices wereretreated in the same way as described for the immunoperoxi-ase staining. Sections were then incubated for 36 h with 1:500 ratnti-BrdU (Harlan Sera-Laboratory), 1:500 mouse anti-neuronaluclei antigen (NeuN) (Chemicon, Temecula, CA, USA), 1:2500abbit anti-calcium-binding protein beta (S100�) (Swant, Bell-nzona, Switzerland) and 1:200 goat anti-doublecortin (DCX)Santa Cruz, CA, USA). After rinsing with TBS and treatment with.1% Triton X-100 and 3% donkey serum for 30 min, sectionsere incubated for 2 h with the secondary antibodies Rhodamine-conjugated donkey anti-rat F(ab=)2, (Dianova), Cy5-conjugatedonkey anti-mouse F(ab=)2 (Dianova), Alexa Fluor 488-conju-ated donkey anti-rabbit (Molecular Probes, Leiden, Netherlands)nd Alexa Fluor 488-conjugated donkey anti-goat (Molecularrobes) in a mixture of 0.1% Triton X-100 and 3% donkey serum

nd S100�

Indomethacine

tralateralan�SD



(n�5)5.6�150.2 2351.2�780.8 2277.6�932.19.6�4.3 65.6�3.8 64.8�10.50.0�4.7 34.4�3.8 34.0�10.90.4�0.9 0.0�0.0 0.8�1.1

(n�6)5.6�453.5* 4468.7�1369.3*‡ 2051.3�568.15.0�5.7 71.0�9.0*‡ 64.3�6.94.4�5.7 29.0�9.0 35.3�7.20.0�0.0 0.0�0.0 0.3�0.8

infarcted animals and appropriate controls is indicated by asteriskshacine-treated animals is marked with a double-cross (‡ P�0.05).

otocol. (A) Animals with photothrombotic infarcts in the sensorimotorbeen divided into three groups with different survival periods (2, 4, andmotor cortex (anterior lesions) and sham-operated controls have been

NeuN a

ConMe

5)212

54

5)270

63

between

jection prols have

primary
ed a daily i.p. injection of BrdU at days 3–14 after ischemia or sham

fTc

Q

CfGstBtw(

f5cm

uig(taP

M

Atpagmaadneew4ii1dwci(dmsa

En

PB

paoo4s(pvcscPtbts

pcn(BauTDbcto6ai(w1s8c7cdDs(ona

ithnct


or 2 h at room temperature. Finally, sections were rinsed again inBS and coverslipped with Moviol 40-88-DABCO (1,4-diazabicy-lo [2.2.2]octane 98%) medium (Sigma-Aldrich).

uantification and statistical analysis

ell counting of peroxidase-stained BrdU-positive cells was per-ormed with an Axioplan 2 imaging microscope (Zeiss, Jena,ermany) using a 20� magnification. Every fourth section wastained and the total number of BrdU-positive cells was counted inhe complete ipsi- and contralateral DG. The resulting number ofrdU-positive cells was then multiplied by a factor four to assess

he total number of BrdU-positive cells in the entire DG. Imagesere taken with an AxioCam HR and Axio Vision 3.1 software

Zeiss, Jena, Germany).The phenotypes of BrdU-positive cells were analyzed by con-

ocal laser scanning microscopy (LSM 510, Zeiss). For each DG,0 BrdU-positive cells were phenotyped using a Z-scan to verify alear co-localization of the BrdU-stained nucleus with cell-specificarkers. The percentage of co-labeled cells was then calculated.

For statistical analysis we first tested for normal distributionsing the Shapiro-Wilk test. Statistical significance of differences

n cell counts between the different experimental and controlroups was then calculated using a one-way analysis of varianceANOVA) followed by an unpaired two-tailed t-test and the multipleest procedure of Bonferroni-Holm. Significant differences weressumed with P�0.05 and highly significant changes with�0.01.

RESULTS

orphology of the photothrombotic infarcts

ll animals in the experimental subgroups (n�44) hadypical cortical infarcts. The photothrombotic infarcts im-aired all cortical layers while the subcortical white matterppeared to be intact in the majority of animals. In oneroup of animals (posterior lesion) the infarct affectedainly the sensorimotor HL and visual Oc2 cortex. Innother experimental group (anterior lesion) the infarctffected the FL and primary motor cortex (Fr1). The me-iolateral diameter of the infarct, as measured on immu-ohistochemical sections taken at the center of the isch-mic lesion, showed no significant differences between thexperimental groups. For animals with posterior lesions itas 2.8�0.1 mm (n�5) at 2 weeks, 1.9�0.4 mm (n�7) atweeks, and 1.8�0.3 mm (n�6) at 10 weeks after the

nfarct. Animals with anterior lesions also showed a declinen lesion diameter from 2.4�0.1 mm (n�7) at 4 weeks to.8�0.2 mm (n�8) at 10 weeks. The reduction of lesioniameter is probably due to scar formation processes asell as to shifts of surrounding tissue toward the lesionenter. Animals which received MK-801 prior to lesionnduction showed slightly larger lesion diameters2.4�0.5 mm; n�5) compared with untreated controls butid not reach significance. In animals treated with indo-ethacine the lesion diameter was 2.5�0.2 mm (n�6). No

tructural abnormalities were found in the sham-operatednimals (n�30).

ffects of focal photothrombotic ischemic lesions oneurogenesis in the DG

eroxidase-immunocytochemistry with antibodies against
rdU initially revealed no differences in number of BrdU- w
ositive cells in the ipsi- and contralateral DG at 2 weeksfter the photothrombotic infarct compared with the sham-perated animals. In the following two weeks the numberf BrdU-positive cells then strongly declined (ipsilateral:3.8%, P�0.001; contralateral: 35.0%, P�0.001) in theham-operated controls with no further loss until week 10Table 1, Fig. 2 and Fig. 3). In contrast, in animals withhotothrombotic lesions in the sensorimotor hindlimb andisual cortex (posterior location) this loss of BrdU-positiveells was reduced resulting in an increase in the number ofurviving cells at 4 weeks (ipsilateral: 39.0%, P�0.047;ontralateral: 7.3%, n.s.) and 10 weeks (ipsilateral: 41.5%,�0.021; contralateral: 39.9%, P�0.014) compared with

he sham-operated controls (Fig. 3). This increase in num-er of surviving BrdU-positive cells was most prominent inhe ipsilateral cortex but also present on the contralateralide (Fig. 3).

In order to determine the phenotypes of the BrdU-ositive cells, we performed double- and triple-immuno-ytochemistry with antibodies against BrdU as well aseuronal (DCX and NeuN) and glial markers (S100�)Fig. 4). At 2 weeks after the surgery, the percentage ofrdU-positive neurons did not differ between infarctednd sham-operated animals (Table 1) and showed val-es between 6 –10% in the ipsi- and contralateral DG.here was also no difference in co-labeling of BrdU andCX at this time point. DCX staining was observedilaterally in approximately 80% of the BrdU-positiveells in infarcted animals as well as sham-operated con-rols (Fig. 5). At 4 weeks after the infarct the percentagef BrdU/NeuN-positive cells increased to approximately0% and was similar in both experimental groups. Onlynimals with photothrombotic infarcts showed a slight

ncrease in number of DCX-BrdU co-labeled cells12.0%), not reaching statistical significance comparedith sham-operated controls (7.0%, Fig. 5). However, at0 weeks after the surgery lesioned animals showed aignificant increase in BrdU-positive neurons (ipsilateral:6.7%, P�0.009; contralateral: 82.7%, P�0.053, n.s.)ompared with controls (ipsilateral: 74.6%; contralateral:1.8%). Applied to the total number of BrdU-positiveells in the DG this resulted in approximately 49% ad-itional newborn neurons (P�0.041) in the ipsilateralG 10 weeks after focal cortical infarcts (Fig. 6). Aimilar increase (45%) occurred in the contralateral DGP�0.083, n.s.). Co-labeling of BrdU and DCX was notbserved at 10 weeks in all experimental groups. Theumber of S100�-positive glial cells was between 0 –3%t all time points (Tables 1 and 2).

Taking these findings together, focal ischemic infarctsncreased the number of surviving BrdU-positive cells inhe DG and increased neurogenesis not only by an en-ancement of surviving cells but also by promoting neuro-al differentiation. The number of BrdU-positive neuronsontinued to increase between 4 and 10 weeks indicatinghat the differentiation process was not completed at 4
eeks after the surgery.

I

Isoti

ssetnnw

Bps

ls6wb(Psnatcba(t

Fia r of anima2 B), respe


nfluence of lesion location on neurogenesis

n order to analyze whether the increase in progenitor cellurvival and neurogenesis was dependent on the locationf the lesion, we induced photothrombotic infarcts affectinghe FL and the primary motor cortex (Fr) (anterior lesions)n a second series of experiments (Fig. 7).

At 4 weeks after anterior photothrombotic lesions alight increase in number of BrdU-positive cells was ob-erved in the DG (ipsilateral: 41.7%, P�0.007; contralat-ral: 18.2%, n.s.) compared with the sham-operated con-rols. This enhancement of surviving cells was less pro-ounced at 10 weeks after the infarct (ipsilateral: 19.0%,.s.; contralateral: 13.4%, n.s.) compared with the animalsith posterior lesions (Table 1; Fig. 7).

Triple-immunocytochemistry with antibodies againstrdU, NeuN and S100� revealed no difference in theercentage of BrdU-positive neurons at 4 weeks after the

ig. 2. Distribution of BrdU-positive cells in the ipsilateral DG after pmmunohistochemically stained with antibodies against the proliferativefter infarct induction. (B) Higher magnifications of the granule cell laye, 4, and 10 weeks post-surgery. Scale bars�200 �m (A) or 20 �m (

urgery between animals with anterior photothrombotic w

esions (ipsilateral: 60.8%; contralateral: 59.3%) andham-operated controls (ipsilateral: 62.0%; contralateral:3.4%; Table 1). At 10 weeks after the surgery animalsith anterior photothrombotic infarcts showed a significantilateral increase in percentage of BrdU-positive neuronsipsilateral: 83.4%, P�0.036; contralateral; 82.3%,�0.006) compared with the sham-operated animals (ip-ilateral: 74.6%; contralateral: 71.8%). Applied to the totalumber of BrdU-positive cells in the DG this resulted inpproximately 35% (n.s.) additional newborn neurons inhe ipsilateral DG and approximately 32% (n.s.) in theontralateral DG. However, the comparison of total num-ers of BrdU-positive neurons between operated animalsnd sham-operated controls did not reach significanceFig. 8). Anterior photothrombotic lesions did not influencehe number of S100�-positive glial cells.

These findings indicated that neurogenesis in the DG

mbotic lesions and sham surgery. Coronal sections through the DGrdU. (A) Overview of BrdU-labeled cells in the ipsilateral DG 2 weeksls with posterior photothrombotic lesions and sham-operated controlsctively.

hotothromarker B

as also facilitated by photothrombotic lesions in the fore-

ltwc

En

Tvmtitbt

nbBNtite

Bhp

iAtrtT0N

Ei

RccbaaoSwa(ni

nt

FMcwi rviving Ba .E.M. As


imb and frontal motor cortex (anterior location) althoughhis increase was less pronounced compared with animalsith cortical infarcts in the sensorimotor HL and visual Oc2ortex (posterior location).

ffects of NMDA receptor blockade by MK-801 oneurogenesis

o determine whether the increase in progenitor cell sur-ival and neurogenesis following small cortical infarcts isediated through excitatory NMDA-dependent neuro-

ransmission we treated the animals 30 min prior to lesionnduction (posterior lesion location) with a single i.p. injec-ion of the NMDA receptor antagonist MK-801 (3 mg/kgody weight) and analyzed cell proliferation and differen-iation 4 weeks after the surgery.

MK-801 treatment did not reduce the lesion-inducedumber of BrdU-positive cells in animals with photothrom-otic infarcts (Fig. 9A). In contrast, an enhancement ofrdU-positive cells was observed after application of theMDA receptor antagonist MK-801. Infarcted animals

reated with MK-801 showed a significant bilateral increasen BrdU-positive cells (ipsilateral: 57.0%; P�0.008; con-ralateral: 27.3%, P�0.026) compared with the sham-op-rated controls receiving the NMDA receptor antagonist.

Triple-immunocytochemistry with antibodies againstrdU, NeuN, and S100� revealed a slight bilateral en-ancement in the percentage of BrdU-positive neurons in

ig. 3. Increase in number of BrdU-positive cells in the DG after corticorphology of photothrombotically induced infarcts. The small well-de

oronal Nissl-stained section through the lesion (marked with a circle)hite matter was mostly intact. (B) Diagram of the total number of Brd

nfarct or sham surgery. Note the significant increase in number of sut 10 weeks on the contralateral side. Bars represent mean values�S

hotothrombotic animals (Table 2) that was not observed i

n non-treated animals 4 weeks after the lesion (Table 1).pplied to the total number of BrdU-positive cells in the DG

his caused a significant bilateral increase in newborn neu-ons in these animals compared with the appropriate con-rols and non-treated photothrombotic animals (Fig. 9B).he number of S100�-positive glial cells was between–1.1% in both experimental groups which received theMDA receptor antagonist.

ffects of anti-inflammatory treatment withndomethacine on neurogenesis

ecent studies have demonstrated that inflammatory pro-esses in the microenvironment of endogenous progenitorells in the SGZ decreased neurogenesis following cere-ral radiation and that this reduction could be blocked bypplication of the COX inhibitor indomethacine (Monje etl., 2003). A widespread inflammatory response was alsobserved after focal brain ischemia (Stoll et al., 1998;chroeter et al., 1999). In order to test the hypothesishether inflammatory processes inhibit DG neurogenesisfter cortical infarcts we treated a further group of animalsphotothrombotic infarcts: n�6; sham-operated controls:�5) with indomethacine beginning one day prior to infarct

nduction until day 14 (2.5 mg/kg body weight, twice daily).Indomethacine treatment strongly enhanced the

umber of BrdU-positive cells in animals with photo-hrombotic infarcts (Table 2; Fig. 10A). Interestingly, this

in the sensorimotor hindlimb and visual cortex (posterior lesions). (A)ion with a diameter of 1.9�0.4 mm is indicated with an arrowhead. A

beneath. The infarct affected all cortical layers while the subcorticale cells in the ipsi- and contralateral DG 2, 4, and 10 weeks after the

rdU-positive cells in the ipsilateral DG of infarcted animals as well asterisks indicate significant differences (P�0.05).

al infarctsfined lesis shownU-positiv

ncrease was only observed in the ipsilateral DG. Com-

pbet(tcSp�ps

(bDeSe

Tc

FsNZoai( 20 �m.


ared with the indomethacine-treated controls the num-er of BrdU-positive cells was nearly doubled (ipsilat-ral: �90.0%; P�0.013), but also comparison with non-reated infarcted animals revealed a significant increaseipsilateral: �69.7%; P�0.01). In sham-operated con-rols no alterations were found. Triple-immunocyto-hemistry with antibodies against BrdU, NeuN, and100� revealed a slight bilateral enhancement in theercentage of BrdU-NeuN-positive cells (ipsilateral:16.6%, P�0.043; contralateral: �8.4%, P�0.345) inhotothrombotic animals (Table 2) that was not ob-

ig. 4. Immunofluorescence coronal sections through the granule cections through the granule cell layer stained with antibodies againseuN (blue). To analyze the colocalization of these different markers w-stack scans (left column). In the middle and right columns, representaf double-labeled cells 2 weeks following photothrombotic infarcts anppear yellow. (B) Double immunofluorescence coronal sections thro

mmature neuronal marker DCX (green). Coronal sections of DG 2 agreen) -labeled cells. Immature neurons appear yellow. Scale bars�

erved in non-treated animals 4 weeks after the lesion i

Table 1). These changes resulted in a significant dou-ling of the number of newborn neurons in the ipsilateralG after indomethacine treatment of animals (ipsilat-ral: 109.9%; P�0.005; Fig. 10B). The number of100�-positive glial cells was not altered in the differentxperimental groups.

DISCUSSION

he present study provides further evidence for an in-rease in neurogenesis in the adult DG following focal

of infarcted animals and sham-operated controls. (A) Triple-staineded), the glial marker S100� (green) and the mature neuronal marker

the immunohistochemical phenotype of the BrdU-positive cells usingocal laser scanning microscope images are shown depicting examplessurgery. BrdU-labeled neurons appear purple and BrdU-labeled gliaranule cell layer stained with antibodies against BrdU (red) and theks after lesion induction depicting examples of BrdU (red) and DCX

ell layert BrdU (re definedtive confd shamugh the gnd 4 wee

schemia. We have demonstrated that even small focal

cta

rnE

Fepwe ificant ev rcted anim

F(sr


ortical ischemic infarcts leaving the hippocampal forma-ion intact increase survival of newborn cells in the SGZnd facilitate neuronal differentiation. This increase in neu-

ig. 5. Differentiation of BrdU-positive cells in the DG after cortical innhancement in DCX/BrdU double-stained cells at 4 weeks after foercentage of BrdU-positive cells colabeling NeuN (circle), S100� (trianith an infarct and open symbols illustrate sham-operated animals (pnhancement of DCX/BrdU-positive cells and at 10 weeks a signalues�S.E.M. Asterisks indicate significant differences between infa

ig. 6. Total number of BrdU-positive neurons in the DG after cortposterior lesions). Diagram of the numbers of BrdU-positive neurons
urgery. Note the significantly increased number of BrdU-positive neurons in thepresent mean values�S.E.M. Asterisks indicate significant differences (P�0
ogenesis was observed bilaterally although the number ofewborn neurons tends to be higher in the ipsilateral DG.ven when the infarct was induced in different brain re-

he sensorimotor hindlimb and visual cortex (posterior lesions). Slightcal infarcts (posterior lesion) in the ipsilateral DG. Diagram of theDCX (rectangle) at the ipsilateral DG. Filled symbols illustrate animalssion). At 4 weeks after the infarct there is a slight but not significant

nhancement in NeuN/BrdU-positive cells. Graphs represent meanals and sham-operated controls (P�0.05).

othrombotic infarcts in the sensorimotor hindlimb and visual cortexsi- and contralateral DG 2, 4, and 10 weeks after the infarct or sham

farcts in tcal cortigle) and

osterior le

ical photin the ip

e infarcted animals in the ipsilateral DG 10 weeks after surgery. Bars.05).

gH

fv

FlNwo g BrdU-pv

FBo


ions a neurogenic response in the DG was observed.owever, this effect seemed to be more pronounced after

ig. 7. Increase in number of BrdU-positive cells in the DG after cortiesions). (A) Morphology of the anterior infarcts. The small well-definedissl-stained section through the cortical lesion (marked with a circle)hite matter was mostly intact. (B) Diagram of total number of BrdU-pr sham surgery. Note the significant increase in number of survivinalues�S.E.M. Asterisks indicate significant differences (P�0.05).

ig. 8. Total number of BrdU-positive neurons in the DG after cortica
rdU-positive neurons in the ipsi- and contralateral DG 4 and 10 weeks afterbserved in all groups this reached no statistical significance. Bars represent me
ocal cortical infarcts in the sensorimotor hindlimb andisual cortex (posterior location) compared with more fron-

thrombotic infarcts in the forelimb and primary motor cortex (anteriorth a diameter of 1.8�0.2 mm is indicated with an arrowhead. A coronal

beneath. The infarct affected all cortical layers while the subcorticalells in the ipsi- and contralateral DG 4 and 10 weeks after the infarctositive cells in the ipsilateral DG at 4 weeks. Bars represent mean

rombotic infarcts (anterior lesions). Diagram of the numbers of total

cal photolesion wiis shownositive c

l phototh
the infarct or sham surgery. Although a slight bilateral increase wasan values�S.E.M. Asterisks indicate significant differences (P�0.05).

tr

al

FhtBci tion, whiv

FswBns(


al infarcts in the forelimb and primary motor cortex (ante-ior location). The study further demonstrates that block-

ig. 9. Effects of MK-801 on total numbers of BrdU-positive cells (A) aindlimb and visual cortex (posterior lesions). (A) Diagram of the numhe infarct or sham surgery in MK-801-treated animals and untreardU-positive cells in the infarcted animals after MK-801 applicationontralateral DG 4 weeks after the infarct or sham surgery in MK-801n number of new neurons in the infarcted animals after MK-801 applicaalues�S.E.M. Asterisks indicate significant differences (P�0.05).

ig. 10. Effects of indomethacine on total numbers of BrdU-positiveensorimotor hindlimb and visual cortex (posterior lesions). (A) Diagraeeks after the infarct or sham surgery in indomethacine-treated animrdU-positive cells in the ipsilateral DG in the infarcted animals aftereurons in the ipsi- and contralateral DG 4 weeks after the infarct or sh
ignificantly increased number of new neurons in the ipsilateral DG. Bars reprP�0.05).
de of NMDA receptors by application of MK-801 beforeesion induction increased the number of newborn neurons

positive neurons (B) after photothrombotic infarcts in the sensorimotortal BrdU-positive cells in the ipsi- and contralateral DG 4 weeks afterrols. Note the bilateral significant increase in number of survivingram of the numbers of total BrdU-positive neurons in the ipsi- andnimals and untreated controls. Note the bilateral significant increase

ch was not observed in sham-operated animals. Bars represent mean

and BrdU-positive neurons (B) after photothrombotic infarcts in thetotal number of BrdU-positive cells in the ipsi- and contralateral DG 4ntreated controls. Note the significantly increased number of survivinghacine application. (B) Diagram of the total number of BrdU-positivery in indomethacine-treated animals and untreated controls. Note the

nd BrdU-ber of toted cont. (B) Diag-treated a

cells (A)m of theals and uindometam surge

esent mean values�S.E.M. Asterisks indicate significant differences

ivflds

diit1t(Tsodhdb(2aseasawdsbEn2lBaptebigngaafcadescparT

vics

BerddwittsB

gscatbteacscipft3inpe2(snNHMi(dttrtifaseb


n the DG of infarcted animals. Furthermore our data pro-ide evidence that blocking of activated microglia and in-ammatory processes with the anti-inflammatory drug in-omethacine results in a strong enhancement of postle-ional neurogenesis.

An increase in neurogenesis in the adult DG has beenescribed in different models of global and focal brain

schemia (Kokaia and Lindvall, 2003). After global cerebralschemia, progenitor cells in the SGZ of the DG augmentheir proliferative activity (Liu et al., 1998; Takagi et al.,999; Yagita et al., 2001; Iwai et al., 2001) and migrate intohe granule cell layer and differentiate into new neuronsKee et al., 2001; Sharp et al., 2002; Tanaka et al., 2004).hese processes occurred bilaterally (Jin et al., 2001) andeemed to be stronger in young animals compared withlder ones (Yagita et al., 2001). Liu et al. (1998) furtheremonstrated that neurogenesis even increased when theippocampal formation did not show any morphologicallyetectable ischemic damage in the CA1 region. Focalrain ischemia also stimulates neurogenesis in the DGArvidsson et al., 2001; Jin et al., 2001; Komitova et al.,002; Zhu et al., 2004). Using transient middle cerebralrtery occlusion as a model for focal ischemic stroke,everal groups reported a bilateral increase in neurogen-sis which was more prominent in the ipsilateral DG butlso present on the contralateral side (Jin et al., 2001). Ourtudy indicates that this increase in adult neurogenesislso occurred after small focal ischemic infarcts whichere confined to the cerebral cortex. Compared with mid-le cerebral artery occlusions, photothrombotic infarcts aremaller with less secondary damage to the surroundingrain tissue (Redecker et al., 2002; Frahm et al., 2004).ven these small infarcts cause a bilateral increase in DGeurogenesis. In accordance with the literature (Jin et al.,001) this enhancement was most prominent on the ipsi-

ateral side. At day 14 after the infarct the numbers ofrdU-positive cells only slightly differ between infarctednimals and sham-operated controls. In both groups ap-roximately 80% of the BrdU-positive cells already expresshe immature neuronal marker DCX. The number of DCXxpressing BrdU-positive cells then declined accompaniedy an increase in mature NeuN-BrdU-positive cells. This

ncrease in number of NeuN-BrdU-positive cells was stron-er in infarcted animals resulting in a higher number of neweurons 10 weeks after the infarct. The present data sug-est that this enhancement of neurogenesis is caused bylonger survival of newborn cells in the DG as well as byfacilitated neuronal differentiation. There is no evidence

or an increased proliferation of endogenous progenitorells in the present study. To address this issue and tonalyze the early proliferative response after the infarct,ifferent BrdU injection protocols have to be used. How-ver, stimulation of neurogenesis in the DG after circum-cribed cortical infarcts seems to be a long-lasting pro-ess. Lesioning of the entorhinal cortex, which gives rise toowerful projections via the perforant path to the DG (Li etl., 1993) leads to an enhancement of hippocampal neu-ogenesis (Cameron et al., 1995; Gama Sosa et al., 2004).
he entorhinal cortex of the rat receives inputs from a H
ariety of cortical regions (Amaral and Witter, 1995) includ-ng the FL, primary somatosensory and the Oc2. Theseonnections might, at least in part, mediate the proliferativeignals to the DG after photothrombotic lesions.

There is also evidence that dying neurons incorporaterdU and express proliferation markers after brain isch-mia (Kuan et al., 2004). Kuan and colleagues (2004) eveneported that hypoxia–ischemia induces DNA synthesis inying mature CA1 neurons which showed apoptotic celleath and no cell proliferation. However, this observationas limited to the CA1 region and absent 26 days after the

nsult. No BrdU-positive apoptotic neurons were found inhe SGZ of the DG. These findings make it implausible thathe increase in BrdU-positive neurons described in ourtudy 10 weeks after the cortical infarct is caused byrdU-uptake of dying mature neurons.

The mechanisms causing the increase in DG neuro-enesis following brain ischemia are only poorly under-tood. Several studies have demonstrated that NMDA re-eptors are involved in the mediation of cell proliferationnd neurogenesis in the DG. Under physiological condi-ions, stimulation of NMDA receptors decreased the num-er of newborn neurons in the DG whereas the blockade ofhese receptors by MK-801 increased neurogenesis (Cam-ron et al., 1995; Bernabeu and Sharp, 2000; Nacher etl., 2003; Okuyama et al., 2004). However, Kitamura andolleagues (2003) demonstrated that in mice lacking the 1ubunit of the NMDA receptor, wheel running did not in-rease neurogenesis indicating that this receptor subunit is

mportant for the activity-dependent increase in hippocam-al neurogenesis. The present findings now disclose thatollowing brain ischemia the role of NMDA-dependentransmission is ambiguous as well. Application of MK-8010 min prior to the induction of focal ischemic cortical

nfarcts significantly increased the number of BrdU-positiveeurons in the DG bilaterally. This finding is in contrast withrevious studies on global and focal brain ischemia usingither the four vessel occlusion model (Bernabeu et al.,000) or transient middle cerebral artery occlusionsArvidsson et al., 2001), respectively. In both models aingle application of MK-801 significantly reduced theumber of newborn neurons in the DG indicating thatMDA receptors facilitate post-ischemic neurogenesis.owever, in both models direct neuroprotective effects ofK-801 have been reported leading to a reduction of

nfarct size in the middle cerebral artery occlusion modelBertorelli et al., 1998) and to a lesser extent of CA1amage after global ischemia (Gill and Kemp, 1989). Sincehe amount of tissue salvageable by MK-801 depends onhe size of the partially ischemic penumbral tissue sur-ounding the lesion (Yao et al., 1993), which is very thin inhe photothrombotic model, MK-801 failed to reduce thenfarct size compared with untreated controls. These dif-erences might at least in part explain the obvious discrep-ncies between the infarct models. Furthermore, MK-801trongly reduces cytokine expression (Jander et al., 2000),specially TNF-alpha (Bertorelli et al., 1998), which haseen shown to induce post-ischemic cell loss (Chao and
u, 1994; Bertorelli et al., 1998).

m(feTtaidswnTekcbiBhliriitdsu

AiaiMaF

A

A

A

B

B

B

C

C

D

E

E

F

G

G

G

G

G

G

H

I

J

J

K

K

K

K

K


It is also well known that inflammatory processes in theicroenvironment of stem cells inhibit neurogenesis

Monje et al., 2003; Ekdahl et al., 2003). Both global andocal ischemia lead to microglial pro-inflammatory cytokinexpression (Stoll et al., 1998; Yamaguchi et al., 2000).reatment with the common nonsteroidal anti-inflamma-

ory drug indomethacine resulted in a strong decrease ofctivated microglia by inhibition of COX, an enzyme that is

nvolved in the synthesis of prostaglandins from arachi-onic acid (Monje et al., 2003; Ekdahl et al., 2003). Atrong microglial activation after photothrombotic infarctsas observed in the remote ipsilateral and contralateraleocortical areas (Stoll et al., 1998; Schroeter et al., 1999).he detrimental effect of activated microglia on neurogen-sis is presumably caused by cytokines such as interleu-in-1�, interleukin-6, nitric oxide and tumor necrosis factorwhich disturb the microenvironment of endogenous stem

ells (Giulian and Vaca, 1993; Giulian et al., 1994; Ge-icke-Haerter, 2001). In the present study application of

ndomethacine caused a strong ipsilateral increase inrdU-positive neurons 4 weeks after infarct induction andad no effect on the sham-operated controls. Following the

iterature, our study provides further evidence that post-schemic inflammatory processes inhibit hippocampal neu-ogenesis. Blocking of these processes might thereforemprove the efficacy of hippocampal neurogenesis follow-ng stroke. These findings suggest that both NMDA recep-ors and inflammatory processes are involved in the me-iation of hippocampal neurogenesis following brain in-ults but additional studies are needed to further clarify thenderlying mechanisms.

cknowledgments—The authors would like to express their grat-tude to Dr. G. Kempermann for his help and continuous supportnd to Dr. M. Bastmeyer and A. Güllmar for excellent assistance

n confocal laser scanning microscopy. The authors also thank N.anthei, C. Sommer, I. Ingrisch and J. Heyder for great technicalssistance. This work was supported by grants of the Deutscheorschungsgemeinschaft Re 1315/3–1 and IZKF Jena (TP 1.7).

REFERENCES

maral DG, Witter MP (1995) Hippocampal formation. In: The ratnervous system (Paxinos G, ed), pp 443–493. Sydney: AcademicPress.

mbrogini P, Orsini L, Mancini C, Ferri P, Ciaroni S, Cuppini R (2004)Learning may reduce neurogenesis in adult rat dentate gyrus.Neurosci Lett 359:13–16.

rvidsson A, Kokaia Z, Lindvall O (2001) N-methyl-D-aspartate recep-tor-mediated increase of neurogenesis in adult rat dentate gyrusfollowing stroke. Eur J Neurosci 14:10–18.

ernabeu R, Sharp FR (2000) NMDA and AMPA/kainate glutamatereceptors modulate dentate neurogenesis and CA3 synapsin-I innormal and ischemic hippocampus. J Cereb Blood Flow Metab20:1669–1680.

ertorelli R, Adami M, Di Santo E, Ghezzi P (1998) MK 801 anddexamethasone reduce both tumor necrosis factor levels and in-farct volume after focal cerebral ischemia in the rat brain. NeurosciLett 246:41–44.

ruel-Jungerman E, Laroche S, Rampon C (2005) New neurons in thedentate gyrus are involved in the expression of enhanced long-term memory following environmental enrichment. Eur J Neurosci
21:513–521.
ameron HA, McEwen BS, Gould E (1995) Regulation of adult neu-rogenesis by excitatory input and NMDA receptor activation in thedentate gyrus. J Neurosci 15:4687–4692.

hao CC, Hu S (1994) Tumor necrosis factor-alpha potentiates glu-tamate neurotoxicity in human fetal brain cell cultures. Dev Neu-rosci 16:172–179.

obrossy MD, Drapeau E, Aurousseau C, Le Moal M, Piazza PV,Abrous DN (2003) Differential effects of learning on neurogenesis:learning increases or decreases the number of newly born cellsdepending on their birth date. Mol Psychiatry 8:974–982.

hninger D, Kempermann G (2003) Regional effects of wheel runningand environmental enrichment on cell genesis and microglia pro-liferation in the adult murine neocortex. Cereb Cortex 13:845–851.

kdahl CT, Claasen JH, Bonde S, Kokaia Z, Lindvall O (2003) Inflam-mation is detrimental for neurogenesis in adult brain. Proc NatlAcad Sci U S A 100:13632–13637.

rahm C, Haupt C, Weinandy F, Siegel G, Bruehl C, Witte OW (2004)Regulation of GABA transporter mRNA and protein after photo-thrombotic infarct in rat brain. J Comp Neurol 478:176–188.

age FH (2000) Mammalian neural stem cells. Science 287:1433–1438.

ama Sosa MA, Wen PH, De Gasperi R, Perez GM, Senturk E,Friedrich VL Jr, Elder GA (2004) Entorhinal cortex lesioning pro-motes neurogenesis in the hippocampus of adult mice. Neuro-science 127:881–891.

ebicke-Haerter PJ (2001) Microglia in neurodegeneration: molecularaspects. Microsc Res Tech 54:47–58.

ill R, Kemp JA (1989) Protection of CA1 pyramidal cell function byMK-801 following ischaemia in the gerbil. Neurosci Lett105:101–106.

iulian D, Li J, Leara B, Keenen C (1994) Phagocytic microglia re-lease cytokines and cytotoxins that regulate the survival of astro-cytes and neurons in culture. Neurochem Int 25:227–233.

iulian D, Vaca K (1993) Inflammatory glia mediate delayed neuronaldamage after ischemia in the central nervous system. Stroke24:I84–I90.

erguido MJ, Carceller F, Roda JM, Avendano C (1999) Hippocampalcell loss in transient global cerebral ischemia in rats: a criticalassessment. Neuroscience 93:71–80.

wai M, Hayashi T, Zhang WR, Sato K, Manabe Y, Abe K (2001)Induction of highly polysialylated neural cell adhesion molecule(PSA-NCAM) in postischemic gerbil hippocampus mainly dissoci-ated with neural stem cell proliferation. Brain Res 902:288–293.

ander S, Schroeter M, Stoll G (2000) Role of NMDA receptor signal-ing in the regulation of inflammatory gene expression after focalbrain ischemia. J Neuroimmunol 109:181–187.

in K, Minami M, Lan JQ, Mao XO, Batteur S, Simon RP, GreenbergDA (2001) Neurogenesis in dentate subgranular zone and rostralsubventricular zone after focal cerebral ischemia in the rat. ProcNatl Acad Sci U S A 98:4710–4715.

ee NJ, Preston E, Wojtowicz JM (2001) Enhanced neurogenesisafter transient global ischemia in the dentate gyrus of the rat. ExpBrain Res 136:313–320.

itamura T, Mishina M, Sugiyama H (2003) Enhancement of neuro-genesis by running wheel exercises is suppressed in mice lackingNMDA receptor epsilon 1 subunit. Neurosci Res 47:55–63.

okaia Z, Lindvall O (2003) Neurogenesis after ischaemic brain in-sults. Curr Opin Neurobiol 13:127–132.

omitova M, Perfilieva E, Mattsson B, Eriksson PS, Johansson BB(2002) Effects of cortical ischemia and postischemic environmentalenrichment on hippocampal cell genesis and differentiation in theadult rat. J Cereb Blood Flow Metab 22:852–860.

uan CY, Schloemer AJ, Lu A, Burns KA, Weng WL, Williams MT,Strauss KI, Vorhees CV, Flavell RA, Davis RJ, Sharp FR, Rakic P(2004) Hypoxia-ischemia induces DNA synthesis without cell pro-liferation in dying neurons in adult rodent brain. J Neurosci
24:10763–10772.

L

L

L

M

N

N

O

P

R

R

S

S

S

S

S

S

S

S

T

T

v

W

W

Y

Y

Y

Z

Z


euner B, Mendolia-Loffredo S, Kozorovitskiy Y, Samburg D, Gould E,Shors TJ (2004) Learning enhances the survival of new neuronsbeyond the time when the hippocampus is required for memory.J Neurosci 24:7477–7481.

i D, Field PM, Starega U, Li Y, Raisman G (1993) Entorhinal axonsproject to dentate gyrus in organotypic slice co-culture. Neuro-science 52:799–813.

iu J, Solway K, Messing RO, Sharp FR (1998) Increased neurogen-esis in the dentate gyrus after transient global ischemia in gerbils.J Neurosci 18:7768–7778.

onje ML, Toda H, Palmer TD (2003) Inflammatory blockade restoresadult hippocampal neurogenesis. Science 302:1760–1765.

acher J, Alonso-Llosa G, Rosell DR, McEwen BS (2003) NMDAreceptor antagonist treatment increases the production of newneurons in the aged rat hippocampus. Neurobiol Aging24:273–284.

acher J, Rosell DR, Alonso-Llosa G, McEwen BS (2001) NMDAreceptor antagonist treatment induces a long-lasting increase inthe number of proliferating cells, PSA-NCAM-immunoreactivegranule neurons and radial glia in the adult rat dentate gyrus. EurJ Neurosci 13:512–520.

kuyama N, Takagi N, Kawai T, Miyake-Takagi K, Takeo S (2004)Phosphorylation of extracellular-regulating kinase in NMDA recep-tor antagonist-induced newly generated neurons in the adult ratdentate gyrus. J Neurochem 88:717–725.

arent JM (2003) Injury-induced neurogenesis in the adult mammalianbrain. Neuroscientist 9:261–272.

edecker C, Wang W, Fritschy JM, Witte OW (2002) Widespread andlong-lasting alterations in GABA(A)-receptor subtypes after focalcortical infarcts in rats: mediation by NMDA-dependent processes.J Cereb Blood Flow Metab 22:1463–1475.

ice AC, Khaldi A, Harvey HB, Salman NJ, White F, Fillmore H,Bullock MR (2003) Proliferation and neuronal differentiation ofmitotically active cells following traumatic brain injury. Exp Neurol183:406–417.

asaki T, Kitagawa K, Yamagata K, Takemiya T, Tanaka S, Omura-Matsuoka E, Sugiura S, Matsumoto M, Hori M (2004) Ameliorationof hippocampal neuronal damage after transient forebrain isch-emia in cyclooxygenase-2-deficient mice. J Cereb Blood FlowMetab 24:107–113.

charfman HE (2004) Functional implications of seizure-induced neu-rogenesis. Adv Exp Med Biol 548:192–212.

chmidt-Hieber C, Jonas P, Bischofberger J (2004) Enhanced synap-tic plasticity in newly generated granule cells of the adult hip-
pocampus. Nature 429:184–187.
chroeter M, Jander S, Witte OW, Stoll G (1999) Heterogeneity of themicroglial response in photochemically induced focal ischemia ofthe rat cerebral cortex. Neuroscience 89:1367–1377.

hanina EV, Redecker C, Reinecke S, Schallert T, Witte OW (2005)Long-term effects of sequential cortical infarcts on scar size, brainvolume and cognitive function. Behav Brain Res 158:69–77.

harp FR, Liu J, Bernabeu R (2002) Neurogenesis following brainischemia. Brain Res Dev Brain Res 134:23–30.

hors TJ, Miesegaes G, Beylin A, Zhao M, Rydel T, Gould E (2001)Neurogenesis in the adult is involved in the formation of tracememories. Nature 410:372–376.

toll G, Jander S, Schroeter M (1998) Inflammation and glial re-sponses in ischemic brain lesions. Prog Neurobiol 56:149–171.

akagi Y, Nozaki K, Takahashi J, Yodoi J, Ishikawa M, Hashimoto N(1999) Proliferation of neuronal precursor cells in the dentate gyrusis accelerated after transient forebrain ischemia in mice. Brain Res831:283–287.

anaka R, Yamashiro K, Mochizuki H, Cho N, Onodera M, Mizuno Y,Urabe T (2004) Neurogenesis after transient global ischemia in theadult hippocampus visualized by improved retroviral vector. Stroke35:1454–1459.

an Praag H, Schinder AF, Christie BR, Toni N, Palmer TD, Gage FH(2002) Functional neurogenesis in the adult hippocampus. Nature415:1030–1034.

ang W, Redecker C, Bidmon HJ, Witte OW (2004) Delayed neuronaldeath and damage of GDNF family receptors in CA1 following focalcerebral ischemia. Brain Res 1023:92–101.

atson BD, Dietrich WD, Busto R, Wachtel MS, Ginsberg MD (1985)Induction of reproducible brain infarction by photochemically initi-ated thrombosis. Ann Neurol 17:497–504.

agita Y, Kitagawa K, Ohtsuki T, Takasawa K, Miyata T, Okano H,Hori M, Matsumoto M (2001) Neurogenesis by progenitor cells inthe ischemic adult rat hippocampus. Stroke 32:1890–1896.

amaguchi M, Saito H, Suzuki M, Mori K (2000) Visualization ofneurogenesis in the central nervous system using nestin promoter-GFP transgenic mice. Neuroreport 11:1991–1996.

ao H, Ginsberg MD, Watson BD, Prado R, Dietrich WD, Kraydieh S,Busto R (1993) Failure of MK-801 to reduce infarct volume inthrombotic middle cerebral artery occlusion in rats. Stroke24:864–870.

hu DY, Lau L, Liu SH, Wei JS, Lu YM (2004) Activation of cAMP-response-element-binding protein (CREB) after focal cerebralischemia stimulates neurogenesis in the adult dentate gyrus. ProcNatl Acad Sci U S A 101:9453–9457.

illes K (1985) The cortex of the rat. Berlin-Heidelberg: Springer
Verlag.
(Accepted 22 June 2005)(Available online 8 September 2005)

Neurogenesis in the adult dentate gyrus after cortical infarcts: Effects of infarct location, N-methyl-d-aspartate receptor blockade and anti-inflammatory treatment

Documents