Rn

ED

h

���

a

ARRAA

KPr6CTA

1

aada

0h

Toxicology Letters 213 (2012) 332– 344

Contents lists available at SciVerse ScienceDirect

Toxicology Letters

jou rn al h om epa ge: www.elsev ier .com/ locate / tox le t

osiglitazone, a PPAR-� agonist, protects against striatal dopaminergiceurodegeneration induced by 6-OHDA lesions in the substantia nigra of rats

un Young Lee1, Jeong Eun Lee1, Jae Hyeon Park, In Chul Shin, Hyun Chul Koh ∗

epartment of Pharmacology, College of Medicine, Hanyang University, 133-791 Seoul, Republic of Korea

i g h l i g h t s

Intranigral 6-OHDA injection induced striatal TH fiber loss.COX-2 involved in 6-OHDA-mediated TH fiber degeneration.PPAR� agonist prevents 6-OHDA-induced neuronal damage.

r t i c l e i n f o

rticle history:eceived 23 September 2011eceived in revised form 13 June 2012ccepted 19 July 2012vailable online 27 July 2012

eywords:eroxisome proliferator-activatedeceptor-�-OHDAyclooxygenase-2yrosine hydroxylasestrocyte

a b s t r a c t

Rosiglitazone is a commonly prescribed insulin-sensitizing drug with selective agonistic activity at theperoxisome proliferator-activated receptor-� (PPAR�). Previously, rosiglitazone was shown to attenu-ate dopaminergic cell loss in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mouse model ofParkinson’s disease (PD), an effect attributed to its anti-inflammatory properties. To elucidate the neuro-protective mechanisms of rosiglitazone, we investigated the effects of rosiglitazone on the expressionsof striatal tyrosine hydroxylase (TH), cyclooxygenase-2 (COX-2) and glial fibrillary acidic protein (GFAP)in a 6-OHDA-lesioned rat PD model. Rosiglitazone (3 mg/kg) was administered intraperitoneally at 24 hand 30 min prior to the creation of an intranigral 6-OHDA lesion. A reduction in TH protein expressionbegan at 3 days and a prominent decrease was observed at 7 days post-lesion, and decreases of dopamine(DA) levels began at 1 day post-lesion. In contrast, GFAP expression was significantly increased at 3 daysand preserved for up to 7 days post-lesion and the patterns of GFAP expression was inversely correlatedto changes in TH expression. Furthermore, COX-2 expression in the rostral striatum showed a significantincrease at 6 h post-lesion while that of the caudal striatum was increased at 12 h. In the 6-OHDA-lesionedmodel, the activation of PPAR� by rosiglitazone significantly prevented TH protein expression reductions,and inhibited 6-OHDA-induced microglia activation in striatum. In addition, rosiglitazone attenuated in

production of both COX-2 and TNF-� expression. In contrast, rosiglitazone pretreatment led to greaterincreases in striatal GFAP expression than 6-OHDA alone and changes in the expression of this proteinpreceded the changes that were seen with TH expression. These results suggest that the neuroprotectionobserved with rosiglitazone treatment may be partially due to the attenuation of COX-2 production andthe strengthening of astrocyte function. Our results provide insight into the neuroprotective mechanismsof rosiglitazone against 6-OHDA-induced neuronal damages.

Crow

. Introduction

Parkinson’s disease (PD) is a neurological disorder that affectspproximately 1–3% of the population (Bennett et al., 1996; Lang

nd Lozano, 1998). The primary pathology is significant loss ofopaminergic neurons in the substantia nigra (SN), which lead to

significant loss of striatal dopamine. The driving force behind the

∗ Corresponding author. Tel.: +82 2 2220 0653; fax: +82 2 2292 6686.E-mail address: hckoh@hanyang.ac.kr (H.C. Koh).

1 These authors contributed equally to this work.

378-4274/$ – see front matter. Crown Copyright © 2012 Published by Elsevier Ireland Lttp://dx.doi.org/10.1016/j.toxlet.2012.07.016

n Copyright © 2012 Published by Elsevier Ireland Ltd. All rights reserved.

progressive nature of neurodegenerative disease such as PD hasremained elusive, but recent evidence point to chronic neuroin-flammation as a key feature in disease progression (Rogers et al.,2007; McGeer and McGeer, 2008).

The analysis of brains from patients with PD has revealedthe presence of activated microglia as well as increased levels ofneuroinflammatory mediators in the SN, indicating that inflam-mation plays a role in the disease (McGeer et al., 1988). The

inflammation associated with PD can also be modeled in exper-imental animals by injecting either bacterial lipopolysaccharide(LPS; Gao et al., 2002) or neurotoxins such as 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP; McGeer et al., 2003).

td. All rights reserved.

Lette

6moOtt22(

tizhni(iSeti2

mcbMsrb2rcWianiea2

o6alfi

2

2

taAo

2

a�PfS

E.Y. Lee et al. / Toxicology

-Hydroxydopamine (6-OHDA) is known to induce the activation oficroglia in the SN of rats (Marinova-Mutafchieva et al., 2009). One

f the earliest characterizations of glial activation in response to 6-HDA was demonstrated by Akiyama and McGeer (1989) following

he introduction of a lesion in the SN. Glia activation has been fur-her characterized following partial striatal lesions (Henning et al.,008; Ambrosi et al., 2010), full-striatal lesions (Cicchetti et al.,002), nigral lesions (Iczkiewicz et al., 2007) and axonal lesionsHenning et al., 2008; Marinova-Mutafchieva et al., 2009).

Cyclooxygenase (COX) is a rate-limiting enzyme involved inhe production of various prostaglandins and thromboxanes ands known to exist in two isoforms, COX-1 and COX-2. COX isoen-ymes, which are generally expressed only in the peripheral organs,ave recently been found to be up-regulated in the brain followingeuronal insults. Of these two isoforms, evidence has revealed the

nvolvement of the COX-2 isoform in neuropathologic conditionsTeismann et al., 2003; Minghetti, 2004). COX-2 is a key enzymen the inflammatory process and the up-regulation of COX-2 in theN was found in both PD patient and animal models of PD (Liangt al., 2007). Studies from other groups have demonstrated protec-ive actions of COX-2 inhibitors in various neurological disordersncluding epilepsy, stress and PD related pathologies (Dhir et al.,007; Hunter et al., 2007; Akula et al., 2008).

Peroxisome proliferator-activated receptor-gamma (PPAR-�), aember of the nuclear receptor family that regulates lipid and

arbohydrate homeostasis, exerts anti-inflammatory actions inoth the periphery and CNS (Heneka et al., 2007; Bernardo andinghetti, 2008). PPAR-� displays a restricted pattern of expres-

ion in the CNS, with basal ganglia nuclei being enriched in thiseceptor. Moreover, the presence of PPAR-� has been detected inoth neuronal and glial cells (Cristiano et al., 2001; Moreno et al.,004). By binding to the DNA of promoter regions, PPAR-� canegulate the expression of several genes, including inflammatoryytokines, TNF-�, COX and iNOS (Bernardo and Minghetti, 2006;

oster and Combs, 2007; Chaturvedi and Beal, 2008). In vivo stud-es have shown that PPAR-� agonists are effective in a variety ofnimal models of neurodegenerative disorders, suggesting that theeuroprotection of these agents might be mediated by their anti-

nflammatory properties (Yan et al., 2003; Kiaei et al., 2005; Drewt al., 2006; Luo et al., 2006). However, limited information is avail-ble to date on the effects of PPAR-� agonists in PD (Breidert et al.,002; Dehmer et al., 2004; Quinn et al., 2008).

In this study, we investigated the effects of PPAR-� agonistsn striatal tyrosine hydroxylase (TH) fiber loss from intranigral-OHDA-lesioned rats, We then investigated the effects of thesegents on the expressions of COX-2, TNF-�, Iba-1 and glial fibril-ary acidic protein (GFAP) in the setting of 6-OHDA-induced THber degeneration.

. Materials and methods

.1. Animals

Adult male Sprague-Dawley rats weighing 225–250 g were housed in a roomhat was kept at a constant temperature (18 ± 2 ◦C) and humidity (50 ± 10%) withn automatic 12 h/12 h light/dark cycle. Food and water were available ad libitum.ll surgical procedures were performed in accordance with regulations for the usef laboratory animals set by the Ethical Committee of Hanyang University.

.2. Reagents and antibodies

Rosiglitazone (Sigma–Aldrich, MO, USA) was dissolved in DMSO. TH and COX-2

ntibodies were purchased from Cell Signaling Technology (Boston, MA, USA). TNF-

was purchased from Biovision (CA, USA) and Iba-1 was purchased from Wakoure Chemical Industries (JAPAN). In addition, GFAP antibodies were purchasedrom MP Biomedicals (Solon, OH, USA). All other chemicals were obtained fromigma–Aldrich.

rs 213 (2012) 332– 344 333

2.3. 6-OHDA lesioning and rosiglitazone administration

All 6-OHDA injections were performed under anesthesia that was induced byinjecting Rompun plus ketamine in a 7:3 ratio (Apoteksbolaget, Sweden) using astereotaxic frame (Stoelting, Wood Dale, IL, USA) with an attached Hamilton syringe.The animals received 3 �l of 6-OHDA (Sigma–Aldrich AB, Sweden) injected into theSN (8 mg/ml free base in 0.02% ascorbic acid in saline), causing a lesion in the nigros-triatal pathway. The following coordinates relative to the bregma were used: AP,−5.3 mm; ML, −2.4 mm; DV, −7.6 mm with the incisor bar positioned at 2.4 mmbelow the interaural line (Paxinos and Watson, 1986). The injection speed was1.0 �l/min and the syringe was kept in place for an additional 5 min before it wasslowly retracted. To minimize mechanical damage, all injections were performedwith a 26 gauge needle on a Hamilton syringe. Rosiglitazone (3 mg/kg) was dis-solved in vehicle (10% DMSO) while vehicle-treated animals received 10% DMSO. Alltreatments were administered intraperitoneally at 24 h and 30 min prior to surgeryaccording to Hyong et al. (2008).

2.4. Tissue preparation

For immunostaining, mice (n = 5) were perfused intracardially with 0.1 ml ofheparin (20 IU) and 20 ml of phosphate buffered saline (PBS; pH 7.4), followed by50 ml of 4% paraformaldehyde (PFA) in 0.1 M PBS. The brain tissues were removedand post-fixed for 2 h in 4% PFA. Then, the fixed brain tissues were soaked in 30%sucrose for two days. The brain tissues were frozen in Tissue-Teck® (Sakura FinetekUSA, Torrance, CA, USA) solution and cut into 25 �m coronal sections using a cryostatmicrotome (CM1850, Leica, Wetzlar, Germany) at −20 ◦C. The tissue sections werestored in 50% glycerol in PBS at −20 ◦C until the immunohistochemistry analysis wasperformed. Coronal sections from two rostro-caudal levels through the striatum: arostral striatum (+1.20 mm relative to bregma) and a caudal striatum (−1.40 mmrelative to bregma), were divided into the dorsal and ventral subregions, respec-tively (Paxinos and Watson, 1986). In the sections stained for TH, GFAP and Iba-1,the SN pars reticulate was distinguished from the pars compacta based on tissuearchitecture. The half of the brain tissues was used for determination of dopamine(DA) by HPLC (n = 5). After removing the whole brain from skull, subregions of stria-tum were rapidly isolated using a rodent brain matrix (RBMS-300 C, World PrecisionInstruments, FL, USA). The tissue were immediately stored at −80 ◦C until used forassay.

For Western blot analyses of the rats (n = 5), after removing the whole brainfrom the skull, the striata were rapidly isolated by with a rodent brain matrix (RBMS-200 C, World Precision Instruments, FL, USA) and a standard rats brain atlas (Paxinosand Watson, 1986). The striatum was collected as a fragment that was approximately1 mm × 1 mm × 1 mm in size. Tissues were immediately stored at −80 ◦C until needfor the assay.

2.5. Immunohistochemistry

Free-floating 25 �m sections were collected and processed for immunohisto-chemistry. These sections were fixed in 100% methanol for 5 min and washed threetimes in PBS/BSA (0.1 M PBS containing 0.1% BSA). The sections were then post-fixedin 4% PFA/0.15% picric acid in 0.1 M PBS for 20 min at room temperature (RT) andwashed three times in PBS/BSA. After fixation, the sections were blocked with 10%normal goat serum (NGS; Gibco BRL, NY, USA) and 0.3% Triton-X 100 in PBS/BSAfor 1 h and incubated overnight with the antibody (TH, 1:250, GFAP, 1:400, Iba-1,1:200) at 4 ◦C. After rinsing, sections were incubated in biotinylated anti-mouse IgGsecondary antibody (1:200; Pel-Freez Biological, Rogers, AR, USA) for 1 h at RT. Afterwashing, the sections were incubated in horseradish peroxidase streptavidin (1:400,Vector Laboratories, Inc., Burlingame, CA, USA) for 2 h at RT. The sections werewashed three times and visualization was performed using 3,3’-diaminobenzidinetetrahydrochloride (SK-4100 DAB kit; Vector) for 40 s. Section development washalted with distilled water. The sections were mounted on slides, air-dried, anddehydrated by passage through an increasing ethanol series (70, 80, 90, 100, and100% for 2 min each). The sections were cleared in xylene and coverslipped with per-mount (Fisher Scientific, Inc., IL, USA). The densities of TH immunoreactive fibers inthe striatum were determined using a computerized image analysis system (imageanalysis software). Images from sections were collected on a Leica DMRXE fluores-cent microscope equipped with a F-View II camera. The densities of TH-IR fibers inthe striatum were measured at the dorsal and ventral subregions of the rostral andcaudal striatum. ‘Mean density’ values from the two areas were averaged. The opticaldensity of the white matter (corpus callosum) was used as a background value andwas subtracted from each mean density. The optical density value from the lesionedregion was expressed as a percentage of the control group on the same section. Toquantification of TH-positive neurons in the SNpc, Only TH-positive neurons withnuclei were counted.

2.6. Quantification of DA

The determination of DA was determined as described (Mori et al., 2005). Thestriatal tissue was homogenized in 0.2 N perchloric acid containing 0.1 M EDTA withinternal standard, centrifuged at 13,000 rpm for 10 min, and filtered though min-ispin filters for an additional 5 min at 13,000 rpm. Samples (20 �l) were injected

3 Lette

wu(oEvaEKticn

2

amR1Idcuen11t(uPC

2

S

F(Rot

34 E.Y. Lee et al. / Toxicology

ith a Rheodyne injector for separation on a reverse phase �-Bondapak C18 col-mn (150 mm × 3.0 mm, Eicom, Japan) maintained at 30 ◦C with a column heaterWaters, Cotland, NY). The mobile phase consisted of 0.05 M citric acid, 0.05 M dis-dium phosphate (pH3.1), 3.2 mM 1-octanesulphonic acid (sodium salts), 0.3 mMDTA and 12% methanol pumped at a flow rate of 0.5 ml/min with a Waters sol-ent delivery system. Electroactive compounds were analyzed at +750 mV usingn analytical cell with an amperometric detector (Eicom, Model ECD-300, Japan).lution peaks were processed using the DS Chromatographic Software (Donam,orea). DA and its metabolites peaks were corrected by comparison to the respec-

ive internal standard, and concentrations were calculated from external standardnjected immediately before and after each experiment. For further analysis, con-entrations of DA was corrected for sample protein concentration and expressed asg/mg protein.

.7. Western blot analysis of TH, COX-2, GFAP and TNF-˛

After the rats were sacrificed by decapitation, each brain was rapidly removednd snap-frozen for use in Western blot analysis. Striatum from control and experi-ental rats was dissected from the whole brain and the tissue was homogenized in

IPA lysis buffer (1% NP-40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate (SDS),50 mM NaCl and 50 mM Tris at pH 7.4) with a protease inhibitor cocktail (Roche,ndianapolis, IN, USA) and 100 �M phenylmethylsulfonyl fluoride (PMSF). Cellularebris was prepared by centrifugation at 12,000 g for 20 min at 4 ◦C. Protein con-entration was determined by the Bradford method. Total proteins extracts (30 �g)nderwent electrophoresis in a 10% SDS gradient polyacrylamide gel and were thenlectrotransferred to nitrocellulose membranes. Membranes were blocked in 5%on-fat milk in 15 mM Tris–HCl, pH7.4, 150 mM NaCl and 0.05% Tween 20 (TBST) for

h, followed by overnight incubation with the primary antibody (COX-2, 1:500, TH,: 1000, GFAP, 1:500, TNF-�, 1:2000) in 5% non-fat milk in TBST at 4 ◦C. After washinghree times with TBST, the membranes were incubated in horseradish peroxidaseHRP)-conjugated secondary antibody (1:1000). Immunoreactivity was visualizedsing enhanced chemiluminescence (Amersham, Buckinghamshire, England, UK).rotein bands were quantified with a densitometer (Molecular Devices, VERSAmax,A, USA).

.8. Statistical analysis

For statistical analysis, the SPSS PC computer program (Statistical Package forocial Science 12.0) was used. Data were expressed as mean ± standard error of

ig. 1. Time-dependent loss of TH fibers in the striatum, following intranigral injection oPaxinos and Watson, 1986), including the rostral and caudal striata at the level of +1.20 mmepresentative photomicrographs of brain coronal sections containing the 6-OHDA-injectf 6-OHDA. (Original magnification, 2×; scale bar 2 mm). RD, dorsal subregion of the rosthe caudal striatum; CV, ventral subregion of the caudal striatum; L, lesion; U, unlesion.

rs 213 (2012) 332– 344

mean (S.E.M.) and the significance of the differences was assessed using the unpairedStudent’s t test at a p value < 0.05. The significance of the differences among meanvalues was assessed using two-way ANOVA coupled with Duncan’s multiple rangetests at a p value of < 0.05.

3. Results

3.1. Time course effects of 6-OHDA on TH fiber loss in foursubregions of the striata

To clarify the regional effects of striatal TH fiber degeneration,we examined striatal TH fibers loss following the infusion of 6-OHDA unilaterally into the SNpc. TH fiber loss was assessed atselected time points (0, 1, 3 and 7 days) by TH immunostaining andWestern blot analysis of striatal subregions. The dorsal and ven-tral striata were analyzed separately because projections to thoseareas differ, with the SNpc projecting predominantly to the dorsalstriatum. In addition, the rostral and caudal striata were analyzed,because the extent of TH neuron denervation was most prominentin the caudal parts while more fibers were seen in the rostral partof the striatum (Rosenblad et al., 2000). Fig. 1A is a diagram of theisolated brain tissues regions (Paxinos and Watson, 1986), includ-ing rostral and caudal striata at the level +1.20 mm and −1.40 mmrelative to the bregma, respectively.

No detectable decrease in TH immunostaining of the ipsilateralstriatum was observed until 1 day post-lesion. Interestingly, thereduction in TH positive immunoreactivity started at 3 days anda prominent decrease was observed at 7 days post-lesion (Fig. 1B).To quantify the striatal TH fibers loss, we performed Western blot

analysis in striatal tissue. At 3 days post-lesion, TH expression inthe rostrodosal (RD) and rostroventral (RV) striata of the lesionedhemisphere were about 36.59 and 65.79% of the corresponding stri-atal subregions in the vehicle group, respectively (Fig. 2). However,

f 6-hydroxydopamine (6-OHDA). (A) Diagrams of the isolated brain tissues regions and −1.40 mm relative to the bregma respectively. (B) TH immunohistochemistry.

ed striatum from animals sacrificed at 1, 3 and 7 days following intranigral injectionral striatum; RV, ventral subregion of the rostral striatum; CD, dorsal subregion of

E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344 335

Fig. 2. Striatal TH immunohistochemistry and Western blot analysis at 1, 3 and 7 days post-lesion, where n = 5 per group. (A) Pretreatment with rosiglitazone attenuated THfiber loss, as characterized by TH-positive immunoreactivity in the rostral striatum of the 6-OHDA-lesioned model. (B) Pretreatment with rosiglitazone blocked decreases inT riatumt s postt

cavshssi

gSw

3s

TrOtpt

H protein expression, as characterized by TH Western blot analysis in the rostral stwo rostral subregions of the striatum (RD and RV) were determined at 1, 3 and 7 dayo animals treated with control groups.

audodorsal (CD) and caudoventral (CV) striatal subregions werebout 52.94 and 62.50% of the corresponding subregions in theehicle group, respectively (Fig. 3). At 7 days post-lesion, TH expres-ion in the RD and RV striata subregions of the 6-OHDA-lesionedemisphere were about 12.68 and 42.11% of the correspondingtriatal subregions, respectively (Fig. 2). However, the CD and CVtriatal subregions were about 29.41 and 41.07% of the correspond-ng subregions in the vehicle group, respectively (Fig. 3).

We have therefore demonstrated that the RD striatal subre-ion was more susceptible to the infusion of 6-OHDA into theNpc than ventral regions (RV and CV) at each time point thatas examined.

.2. The effects of rosiglitazone pretreatment on 6-OHDA-inducedtriatal TH fiber degeneration in striatal subregions

To determine the preventive effect of rosigliatazone on striatalH fiber degeneration in the 6-OHDA-lesioned rats, we treated theats with rosiglitazone twice before the intranigral injection of 6-

HDA. As a result, rosiglitazone pretreatment significantly blocked

he loss of TH fibers in both rostral and caudal striata at each timeoint that was examined (Figs. 2 and 3). Interestingly, the protec-ive effects of rosiglitazone were more prominent in the rostral

of the 6-OHDA-lesioned model. Quantitative analyses of TH protein expression in-lesion. Each column represents the mean ± S.E.M., **p < 0.01, *p < 0.05 as compared

striatum than the caudal striatum at each time point that was exam-ined (Figs. 2 and 3). The group that was treated with rosiglitazonealone did not exhibit any differences in TH protein expression ateach time point when compared with the vehicle group. Moreover,to confirm the effects of rosiglitazone in protection of DA neuronalloss, we investigated TH-positive cell bodies in SNpc. The results ofthe TH-immunohistochemistry in the SN at pre-lesion and at 0, 1, 3and 7 days post-lesion are shown in Fig. 4A. At the SN regions, infu-sion of 6-OHDA induced a progressive loss of TH-immunoreactivecells. A mild lesion was already apparent at 1 day after post-lesion.Seven days after unilateral 6-OHDA infusion, a prominent decreasewas observed at 7 days post-lesion. The quantification of DA neu-ron loss in the SN at each time point is presented in Fig. 4B. The lossof TH positive cells reached 21% at 1 day, 46% at 3 days, and 87%at 7 days after injection. In contrast, loss of TH positive cell bod-ies were dramatically decreased in substantia nigra pars compecta(SNpc) at any time point after rosiglitazone treatment before theintranigral injection of 6-OHDA (Supplementary Fig. 1A and B).

3.3. Determination of DA levels in striatal subresions

Levels of DA were analyzed at selected time point (0, 1, 3 and7 days) after the last injection of 6-OHDA. As shown in Fig. 4,

336 E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344

Fig. 3. Striatal TH immunohistochemistry and Western blot analysis at 1, 3 and 7 days post-lesion, where n = 5 per group. (A) Pretreatment with rosiglitazone attenuated THfiber loss, as characterized by TH-positive immunoreactivity in the caudal striatum of the 6-OHDA-lesioned model. (B) Pretreatment with rosiglitazone blocked decreases inTH protein expression, as characterized by TH Western blot analysis in the caudal striatum of the 6-OHDA-lesioned model. Quantitative analyses of TH protein expression int s postt

iaetbwbtoa

3G

sbeGaG1

wo caudal subregions of the striatum (CD and CV) were determined at 1, 3 and 7 dayo animals treated with control groups.

nfusion of 6-OHDA significantly reduced DA levels 1 dayfter exposure to 6-OHDA at RD, RV and CD. DA lev-ls gradually decreased after infusion of 6-OHDA. In con-rast, the group of treated with rosiglitazone significantlylocked in DA levels at each time point when comparedith time-matched 6-OHDA-treated groups. However, DA level

egan to decrease after 3 days in CD striatal subresion andhere are no statistical significant between pre-treatmentf rosiglitazone before 6-OHDA infusion and 6-OHDA-treatednimals.

.4. The effects of rosiglitazone pretreatment on 6-OHDA-inducedFAP expression in striatal subregions and SNpc

GFAP expression was clearly detectable in the four striatalubregions and there were no differences in GFAP expressionetween these subregions. In the four striatal subregions, GFAPxpression was unchanged up to 1 day post-lesion. Thereafter,

FAP expression was significantly increased in the striatum at 3nd 7 days post-lesion (Fig. 5A and B). At 3 days post-lesion, thePAP expressions of RD, RV, CD and CV striata were 2.36, 2.33,.29 and 2.07 fold of the corresponding expression in the intact

-lesion. Each column represents the mean ± S.E.M., **p < 0.01, *p < 0.05 as compared

striatum, respectively. There was no time difference in the GFAPexpression between the rostral and caudal striata in 6-OHDA-lesioned striatum at each time point examined (Fig. 5A and B).Interestingly, when treatment with rosiglitazone was paired withthe intranigral 6-OHDA injections, significant increases in GFAPexpression were seen at day 1 post-lesion and sustained at day7 (Fig. 5A and B). At 1 day post-lesion, the GFAP expressions ofrosiglitazone-treated RD, RV, CD and CV striata were 1.81, 2.43,2.01 and 1.56 fold of the corresponding expression in the striatumof 6-OHDA-lesioned rats, respectively. Changes in GFAP expressionpreceded those of TH loss in 6-OHDA-lesioned rats. In addition,rosiglitazone treatment caused rapid increases in GFAP expres-sion compared with 6-OHDA-lessoned rats. In comparison withthe vehicle group, the rosiglitazone alone group did not exhibitany differences in GFAP protein expression at each time point thatwas examined. Following 6-OHDA injection, astrocytes were exam-ined with antibody against GFAP to assess gliosis in the SNpc.At 1 day after 6-OHDA injection, there was an increased in GFAP

immunofluorescence in the SNpc compared with saline-treatedanimals. In addition, following 6-OHDA infusion, the stained astro-cytes in resioned SNpc displayed a characteristic reactive cellmorphology and gradually increased GFAP expression until 7 days.

E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344 337

CV

Do

pam

ine (

ng

/mg

pro

tein

)0

2

4

6

8

10

12

Vehicle

6-OHDA

RSG + 6-OHDA

RV

Do

pam

ine (

ng

/mg

pro

tein

)

0

2

4

6

8

10

12

14

16

Vehicle

6-OHDA

RSG + 6-OHDA

1 3 7 1 3 7 (day)

#

****

RD

Do

pam

ine (

ng

/mg

pro

tein

)

0

2

4

6

8

10

12

14

16

18

Vehicle

6-OHDA

RSG + 6-OHDA

1 3 7 1 3 7 (day)

**

**

#

**

##

##

CD

Do

pam

ine (

ng

/mg

pro

tein

)

0

2

4

6

8

10

12

14

Vehicle

6-OHDA

RSG + 6-OHDA

1 3 7 1 3 7 (day)

**

**

* #

# N.S

#

N.S

N.S

N.S

* N.S N.S

**

**

1 3 7 1 3 7 (day)

Fig. 4. Changes in the concentration of striatal dopamine (DA) induced at 1, 3 and 7 days following intranigral injection of 6-OHDA. Pretreatment with rosiglitazone attenuatedd ote th#

Mro2

3is

gs2psdffCtGtaOattaisiTi

ecreases of DA levels in the striatum of the 6-OHDA-lesioned model. Error bars denp < 0.05, ##p < 0.01 with respect to the 6-OHDA-injected animals (n = 5).

oreover, this increase in GFAP expression observed followingosiglitazome pre-treatment was more powerful compare to thatbserved in rat treated with 6-OHDA, shown in Supplementary Fig..

.5. The effects of rosiglitazone pretreatment on 6-OHDA-inducednflammation related gene such as COX-2 and TNF- ̨ expression intriatal subregions

As can be seen in Fig. 6A, in comparison with the vehicleroup the COX-2 expression of the rostral striatum exhibited aignificantly increase at 12 h post-lesion. Thereafter, the COX-

expression of the rostral striatum decreased significantly 24 host-lesion. Interestingly, the COX-2 expression of the caudaltriatum exhibited a significant increase at 6 h post-lesion butecreased to normal levels at 12 h and thereafter was preservedor up to 24 h (Fig. 6B). We found that there was a time dif-erence between the rostral and caudal striata with respect toOX-2 expression in response to 6-OHDA injections. In addi-ion, changes in COX-2 expression preceded changes in TH andFAP expression in 6-OHDA-lesioned rats. Rosiglitazone pre-

reatment significantly blocked increases in COX-2 expressiont each time point that was examined after the creation of 6-HDA lesions both in rostral and caudal striata. The rosiglitazonelone group did not exhibit any difference in the COX-2 pro-ein expression in comparison with the vehicle group at eachime point that was examined. TNF-� was significantly inducedt 1 day following 6-OHDA treatment when compared to controlsn post-lesions (Fig. 7). Especially, as shown in Fig. 7A, expres-

ion of TNF-� in RD was continued until 7 days after 6-OHDAnjections while other subregions showed transient expression ofNF-�. Rosiglitazone pretreatment significantly blocked increasesn TNF-� expression at each time point that was examined

e standard error of the mean (SE). *p < 0.05, **p < 0.01 with the respect to the control.

after the creation of 6-OHDA lesions in both rostral and caudalstriata.

3.6. The effects of rosiglitazone pretreatment on 6-OHDA-inducedIba-1 expression in striatal subregions and SNpc

Iba-1 expression was clearly detectable in the four striatal sub-regions. In control animals, Iba-1 positive microglia was observedin small number, exhibiting a non-reactive morphology (Fig. 8).In the four striatal subregions, 6-OHDA triggered increase in totalnumber of Iba-1 and reactive microglia was significantly detectedat 1 day post-lesion and continued up to 7 days post-lesion in RDand RV (Fig. 8A and B). In caudal striata in 6-OHDA-lesioned stria-tum, activated microglia was returned to basal level (Fig. 8C andD). Interestingly, when treatment with rosiglitazone was pairedwith the intranigral 6-OHDA injections, significant reduction inIba-1 expression were seen at day 3 post-lesion and sustained atday 7 (Fig. 8). In 6-OHDA-lesioned SNpc, microglia was increasedand activated at 1day post-lesion and continued up to 7 days. Andalso, rosiglitazone significantly reduced microglia activation at eachtime point (Supplementary Fig. 3).

4. Discussion

The results of this study provide evidence that intranigral 6-OHDA injection elicited the loss of TH fibers and the up-regulationof GFAP expression and preceded the increase in COX-2 expres-sion in the striatum. In addition, we have demonstrated that as aPPAR-� agonist, rosiglitazone prevented the striatal TH fiber loss inthese models. The protective effects of rosiglitazone are most likely

attributable to anti-inflammatory effects with increases in astro-cyte function. These findings offer an additional explanation of thepreviously reported neuroprotective effects of PPAR-� agonists inmodels of PD.

338 E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344

Fig. 5. Striatal GFAP Western blot analysis at 1, 3 and 7 days post-lesion, where n = 5 per group. Intranigral 6-OHDA injection led to significant increases in the GFAP expressionof the rostral striatum and rosiglitazone pretreatment led to greater increases in GFAP expression compared to 6-OHDA alone. (A) Representative Western blots for RD andR wo rop ared t

namswApst6

V striatum are shown. (B) Quantitative analyses of GFAP protein expression in tost-lesion. Each column represents the mean ± S.E.M., **p < 0.01, *p < 0.05 as comp

We found that intranigral 6-OHDA injections produced promi-ent decreases in the TH-positive immunoreactivities of SNpcnd the dorsal part of striatum (RD and CD) in a time-dependentanner, although the ventral part of the striatum (RV and CV)

howed only slight decreases. To clarify the degree of TH fiber loss,e performed Western blot analyses for four striatal subregions.

decrease in striatal TH expression was observed at 3 days

ost-lesion and TH expression in the rostral part of the striatumhowed a prominent decrease at 7 days when compared withhe caudal part. A similar report demonstrated that intranigral-OHDA injections led to decreases in TH expression, as measured

stral subregions of the striatum (RD and RV) were determined at 1, 3 and 7 dayso animals treated with control groups.

by an immunostaining of the rostral striatal sections (Hanrottet al., 2008). In contrast, Rosenblad et al. (2000) reported that in6-OHDA-lesioned rats, the extent of striatal TH fiber denervationwas most prominent in the caudal part of the striatum comparedwith the rostral parts. Interestingly, we found that the intranigral6-OHDA lesion was the most prominent in the RD striatum com-pared to the four striatal subregions at each time point examined.

These results are the first evidence for a regional effect of 6-OHDAin this model. In fact, the nigrostriatal tract connects the dopaminecells of the SNpc (A9 dopaminergic neuron) with the dorsolateralstriatum, while the mesolimbic part of the tract connects the

E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344 339

Fig. 6. Striatal COX-2 Western blot analysis at 6, 12 and 14 h post-lesion, where n = 5 per group. Intranigral 6-OHDA injection led to significant increases in the COX-2expression of the striatum and rosiglitazone pretreatment blocked the increase in COX-2 expression. (A) Representative Western blots for RD and RV striatum are shown.Quantitative analyses of COX-2 protein expression in two rostral subregions of the striatum (RD and RV) were determined at 6, 12 and 24 h post-lesion. Each column representsthe mean ± S.E.M., **p < 0.01, *p < 0.05 as compared to 6-OHDA-injected animals and pre-treatment of rosiglitazone before 6-OHDA injection. (B) Representative Westernb xpres1 .05 as6

nmVaw2ot

l

lots for CD and CV striatum are shown. Quantitative analyses of COX-2 protein e2 and 24 h post-lesion. Each column represents the mean ± S.E.M., **p < 0.01, *p < 0-OHDA injection.

eurons of the VTA (A10 dopaminergic neuron) with the ventro-edial striatum, nucleus accumbens, olfactory tubercle, and layerI of the neocortex (van Domburg and ten Donkelaar, 1991). Inddition, phenotypes of motor deficits show a difference patternhen A9/A10 dopaminergic neurons are degenerated (Moore et al.,

001). Therefore, we hypothesize that the anatomical distribution

f TH neuron may contribute to the different response of TH fiberso 6-OHDA.

To elucidate the involvement of inflammation in TH fibeross, we measured striatal COX-2 expression in intranigral

sion in two caudal subregions of the striatum (CD and CV) were determined at 6, compared to 6-OHDA-injected animals and pre-treatment of rosiglitazone before

6-OHDA-lesioned models. Herein we found an acute up-regulationof COX-2 protein expression, which occurred in the rostral partof the striatum at 12 h post-lesion. Interestingly, the caudal stria-tum, however, showed the increase in COX-2 expression as early as6 h post-lesion. In addition, there was no time lag between dorsaland ventral COX-2 expression. A similar report demonstrated that

intrastriatal injection with LPS showed an increase in COX-2 proteinexpression as measured by an immunostaining and Western blot ofstriatal tissues (Hunter et al., 2007). In addition, Gupta et al. (2009)reported that COX-2 inhibition produces neuroprotective effects

340 E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344

Fig. 7. Striatal TNF-� Western blot analysis at 1, 3 and 7 days post-lesion, where n = 5 per group. Intranigral 6-OHDA injection led to significant increases in the TNF-�expression of the striatum and rosiglitazone pretreatment blocked the increase in TNF-� expression. (A) Representative Western blots for RD and RV striatum are shown.Quantitative analyses of TNF-� protein expression in two rostral subregions of the striatum (RD and RV) were determined at 1, 3 and 7 days post-lesion. Each columnrepresents the mean ± S.E.M., **p < 0.01, *p < 0.05 as compared to 6-OHDA-injected animals and pre-treatment of rosiglitazone before 6-OHDA injection. (B) RepresentativeWestern blots for CD and CV striatum are shown. Quantitative analyses of TNF-� protein expression in two caudal subregions of the striatum (CD and CV) were determineda , *p <

b

ihMeLt

t 1, 3 and 7 days post-lesion. Each column represents the mean ± S.E.M., **p < 0.01efore 6-OHDA injection.

n MPTP-induced striatal lesions in rats. In contrast, other reportsave asserted that striatal COX-2 proteins were not changed in LPS,

PTP and 6-OHDA-inudced PD models, although the COX proteins

xpression of the substantia nigra was increased (de Meira Santosima et al., 2006). Although the exact mechanism is unknown, theime lag in COX-2 expression in the rostral and caudal striatum can

0.05 as compared to 6-OHDA-injected animals and pre-treatment of rosiglitazone

be attributed to the following observed effects: the regional dif-ferences in striatal TH density, the differences in striatal TH fiber

degeneration between the rostral and caudal parts (Rosenblad et al.,2000), and other unknown effects. Furthermore, in present study,TNF-� levels were significantly elevated after injection with 6-OHDA. These results agree with previous reports. Mogi et al. (1994)

E.Y. Lee et al. / Toxicology Letters 213 (2012) 332– 344 341

Fig. 8. Striatal Iba-1 immunohistochemistry at 1, 3 and 7 days post-lesion, following intranigral injection of 6-hydroxydopamine (6-OHDA), where n = 5 per group. (Originalmagnification, 5×; scale bar 1 mm). (A and B) Intranigral 6-OHDA injection led to significant increases in the Iba-1 expression of the rostral striatum and rosiglitazonepretreatment attenuated Iba-1 expression compared to 6-OHDA alone. (C and D) Intranigral 6-OHDA injection led to significant increases in the Iba-1 expression of therostral striatum. Pretreatment with rosiglitazone blocked increased in Iba-1 expression in the caudal striatum of the 6-OHDA-lesioned model.

3 Lette

ro2edPkteOotSotWmTreesefsvTrtTttLni

�ieeehLtwrt

T6yahsafSiGwbott

c

42 E.Y. Lee et al. / Toxicology

eported that a significant increase in the level of TNF-� mRNA wasbserved in 6-OHDA-lesioned striatum, with maximal value after4 h comparing to the control. In addition, several cytokines lev-ls, especially TNF-�, were found to be increased in the striatalopaminergic regions in MPTP-induced parkinsoonism mice andD patients (Mogi et al., 1994; Sriram et al., 2002). TNF-� play aey role in the interaction between the nervous and immune sys-em, our results indicate the involvement of the immunologicalvents during the process of the neurodegeneration induced by 6-HDA in rats. Some evidences to suggest that enhanced expressionf TNF-� associated with microglia activation. This reaction is con-ributed to damage of dopaminergic neurons (Mogi et al., 1994;treit et al., 2005). In accordance with previous reports, expressionf Iba-1 significantly increased after 6-OHDA injection. Rosigli-azone prevented microglia activation via attenuation of TNF-�.

hen we administered rosiglitazone prior to 6-OHDA, the pretreat-ent blocked the 6-OHDA-induced TH fiber loss in the striatum.

his treatment ultimately protected against the toxin. A similareport demonstrated that PPAR-� agonists had neuroprotectiveffects in the substantia nigra of an MPTP mouse model of PD (Quinnt al., 2008; Schintu et al., 2009). In addition, Hunter et al. (2008)uggested that pioglitazone, a PPAR-� agonist, showed a protectiveffects against an intrastriatal LPS model. However, our results haveor the first time shown that intranigral 6-OHDA lesions provokedtriatal fiber degeneration of TH neurons and that PPAR-� had a pre-entive effect against the degeneration of TH fibers in the striatum.hese results are the first evidence for a neuroprotective effect ofosiglitazone in the striatum. Furthermore, we found that rosigli-azone pretreatment attenuated the increase in COX-2 as well asNF-� protein expression in the striatum when compared withhose injected with 6-OHDA alone. A similar report demonstratedhat a PPAR-� agonist abolished the COX-2 expression induced byPS in the cortical neurons (Kim et al., 2002). And also, PPAR-� ago-ist decreased the TNF-� expression as well as microglia activation

nduced by MPTP in the SNpc (Carta et al., 2011).Our results, together with other reports suggest that PPAR-

agonists have anti-inflammatory effects that these anti-nflammatory effects are involved in preventing the neurodegen-ration caused by 6-OHDA toxin. In fact, the regulation of COX-2xpression through PPAR agonists is specific to cell type. In cornealpithelial cells, PPAR-� activator increased COX-2 expression byypoxia, but PPAR-� exerted a decrease in COX-2 expression byPS stimulation in fetal hepatocytes. Combs et al. (2000) showedhat amyloid stimulated COX-2 expression in primary microgliaas inhibited by PPAR-� agonists. In contrast, Staels et al. (1998)

eported that PPAR-� ligands, and not PPAR-� ligands, inhibitedhe production of COX-2, which were induced by interleukin-1.

In order to examine the potential relationship between striatalH loss and astrocyte function, astrocytic responses to intranigral-OHDA-lesioning were evaluated using GFAP Western blot anal-sis and immunohistochemistry. Reactive gliosis is a response ofstrocytes to a variety of brain insults that is characterized byypertrophy of the cell bodies and processes, altered gene expres-ion, an increase in the expression of GFAP (Ridet et al., 1997),nd proliferation bordering the wound that occurs in a gradatedashion in relation to the severity of the injury (Sofroniew, 2009;ofroniew and Vinters, 2010). In the present study, we found thatntranigral 6-OHDA lesions induced increases in SNpc and striatalFAP expression and that the regional effects of GFAP expressionere not observed in the four striatal subregions. Similarly, it has

een reported that astrocyte activation parallels the time coursef dopaminergic cell loss in the SN as well as the striatum and

hat the expression of GFAP remains up-regulated even after MPTPreatment (Watanabe et al., 2008; Walsh et al., 2011).

Interestingly, we observed that rosiglitazone pretreatment,ompared with 6-OHDA injection alone, produced more increases

rs 213 (2012) 332– 344

in GFAP expression in the SNpc and striatum. In addition, changesin GFAP expression preceded changes in TH fiber loss and rosigli-tazone caused rapid increases in GFAP expression when comparedwith 6-OHDA-lessoned rats. These results suggest that the neuro-protection of PPAR-� might be involved in the increases in astrocytefunction that are seen in these models. Astrocytes express PPAR-�and accumulating evidence indicates that PPAR-� agonists mod-ulate astrocyte functions (Cullingford et al., 1998; Cristiano et al.,2001).

However, when astrocytes are up-regulated, whether the astro-cytes are beneficial or detrimental remains controversial. In theinitial stages of brain injury, reactive astrocytes have a neuroprotec-tive effect, enhancing neuronal survival. In advanced stages of braininjury, these cells have also been shown to inhibit neuronal regen-eration. In addition, reactive astrocytes are ubiquitous in damagedCNS tissue and they are often regarded as uniformly harmful, caus-ing toxic edema, provoking inflammation and releasing cytotoxins(Barreto et al., 2011). Furthermore, astrocytes produce several fac-tors that may be important in the inflammatory reaction that occursin the substantia nigra in PD. The amount of GFAP-positive astro-cytes correlates inversely to the amounts of dopaminergic cell loss(Kato et al., 2004). They respond particularly to pro-inflammatorycytokines such as IL-1� and TNF-�, and it is believed that thesecytokines participate in astrocyte activation after CNS damage(Stoll and Jander, 1999; Vila et al., 2001). Astrocytes however, areconsidered to be the main targets for neuronal protection follow-ing brain insults. For instance, Endo (2005) reported that galectin-1induces transforming growth factor beta expression in astrocytes,and this is associated with reduced neuronal death. In addition,antioxidant therapy using metallothionein-1, a protein stronglyexpressed by astrocytes after injury, decreased oxidative stress andproapoptotic signaling, thus enhancing neuronal survival followingbrain injury (Chung et al., 2008; Leung et al., 2010).

In conclusion, we demonstrated that pretreatment with therosiglitazone attenuates striatal dopaminergic neurodegenerationin the intranigral 6-OHDA-lesioned models of PD by a mechanisminvolving anti-inflammatory activity and increases in astrocytefunction. We also highlighted the importance of clinical applica-tions of PPAR-� agonists for the treatment of PD.

Conflict of interest statement

The authors state that they have no financial interest in theproducts mentioned within this article.

Acknowledgements

This work was supported by a grant from the Korea Science andEngineering Foundation (2011-0028269) through the MRC for Reg-ulation of Stem Cell Behaviors at Hanyang University College ofMedicine, Republic of Korea.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.toxlet.2012.07.016.

References

Akiyama, H., McGeer, P.L., 1989. Microglial response to 6-hydroxydopamine-induced substantia nigra lesions. Brain Research 489, 247–253.

Akula, K.K., Dhir, A., Kulkarni, S.K., 2008. Rofecoxib a selective cyclooxygenase-2(COX-2) inhibitor increases pentylenetetrazol seizure threshold in mice: possi-ble involvement of adenosinergic mechanism. Epilepsy Research 78, 60–70.

Ambrosi, G., Armentero, M.T., Levandis, G., Bramanti, P., Nappi, G., Blandini, F., 2010.Effects of early and delayed treatment with an mGluR5 antagonist on motor

Lette

B

B

B

B

B

C

C

C

C

C

C

C

d

D

D

D

E

G

G

H

H

H

H

H

E.Y. Lee et al. / Toxicology

impairment nigrostriatal damage and neuroinflammation in a rodent model ofParkinson’s disease. Brain Research Bulletin 82, 29–38.

arreto, G.E., Gonzalez, J., Torres, Y., Morales, L., 2011. Astrocytic-neuronal crosstalk:implications for neuroprotection from brain injury. Neuroscience Research 71,107–113.

ennett, D.A., Beckett, L.A., Murray, A.M., Shannon, K.M., Goetz, C.G., Pilgrim, D.M.,Evans, D.A., 1996. Prevalence of parkinsonian signs and associated mortality ina community population of older people. New England Journal of Medicine 334,71–76.

ernardo, A., Minghetti, L., 2006. PPAR-gamma agonists as regulators of microglialactivation and brain inflammation. Current Pharmaceutical Design 12, 93–109.

ernardo, A., Minghetti, L., 2008. Regulation of glial cell functions by PPAR-gammanatural and synthetic agonists. PPAR Research 2008, 1–10.

reidert, T., Callebert, J., Heneka, M.T., Landreth, G., Launay, J.M., Hirsch, E.C., 2002.Protective action of the peroxisome proliferator-activated receptor-gammaagonist pioglitazone in a mouse model of Parkinson’s disease. Journal of Neuro-chemistry 82, 615–624.

arta, A.R., Frau, L., Pisanu, A., Wardas, J., Spiga, S., Carboni, E., 2011. Rosiglita-zone decreases peroxisome proliferator receptor-gamma levels in microglia andinhibits TNF-alpha production: new evidences on neuroprotection in a progres-sive Parkinson’s disease model. Neuroscience 27, 250–261.

haturvedi, R.K., Beal, M.F., 2008. PPAR: a therapeutic target in Parkinson’s disease.Journal of Neurochemistry 106, 506–518.

hung, R.S., Penkowa, M., Dittmann, J., King, C.E., Bartlett, C., Asmussen, J.W., Hidalgo,J., Carrasco, J., Leung, Y.K., Walker, A.K., Fung, S.J., Dunlop, S.A., Fitzgerald, M.,Beazley, L.D., Chuah, M.I., Vickers, J.C., West, A.K., 2008. Redefining the role ofmetallothionein within the injured brain: extracellular metallothioneins play animportant role in the astrocyte-neuron response to injury. Journal of BiologicalChemistry 283, 15349–15358.

icchetti, F., Brownell, A.L., Williams, K., Chen, Y.I., Livni, E., Isacson, O., 2002.Neuroinflammation of the nigrostriatal pathway during progressive 6-OHDAdopamine degeneration in rats monitored by immunohistochemistry and PETimaging. European Journal of Neuroscience 15, 991–998.

ombs, C.K., Johnson, D.E., Karlo, J.C., Cannady, S.B., Landreth, G.E., 2000. Inflamma-tory mechanisms in Alzheimer’s disease: inhibition of beta-amyloid-stimulatedproinflammatory responses and neurotoxicity by PPARgamma agonists. Journalof Neuroscience 20, 558–567.

ristiano., L., Bernardo, A., Cerù, M.P., 2001. Peroxisome proliferator-activated recep-tors (PPARs) and peroxisomes in rat cortical and cerebellar astrocytes. Journalof Neurocytology 30, 671–683.

ullingford, T.E., Bhakoo, K., Peuchen, S., Dolphin, C.T., Patel, R., Clark, J.B., 1998.Distribution of mRNAs encoding the peroxisome proliferator-activated receptoralpha, beta, and gamma and the retinoid X receptor alpha, beta, and gamma inrat central nervous system. Journal of Neurochemistry 70, 1366–1375.

e Meira Santos Lima, M., Braga Reksidler, A., Marques Zanata, S., Bueno Machado,H., Tufik, S., Vital, M.A., 2006. Different parkinsonism models produce a time-dependent induction of COX-2 in the substantia nigra of rats. Brain Research1101, 117–125.

ehmer, T., Heneka, M.T., Sastre, M., Dichgans, J., Schulz, J.B., 2004. Protection bypioglitazone in the MPTP model of Parkinson’s disease correlates with I kappaB alpha induction and block of NF kappa B and iNOS activation. Journal of Neu-rochemistry 88, 494–501.

hir, A., Naidu, P.S., Kulkarni, S.K., 2007. Neuroprotective effect of nimesulide, apreferential COX-2 inhibitor, against pentylenetetrazol (PTZ)-induced chem-ical kindling and associated biochemical parameters in mice. Seizure 16,691–697.

rew, P.D., Xu, J., Storer, P.D., Chavis, J.A., Racke, M.K., 2006. Peroxisome proliferator-activated receptor agonist regulation of glial activation: relevance to CNSinflammatory disorders. Neurochemistry International 49, 183–189.

ndo, T., 2005. Glycans and glycan-binding proteins in brain: galectin-1-inducedexpression of neurotrophic factors in astrocytes. Current Drug Targets 6,427–436.

ao, H.M., Jiang, J., Wilson, B., Zhang, W., Hong, J.S., Liu, B., 2002. Microglialactivation-mediated delayed and progressive degeneration of rat nigraldopaminergic neurons: relevance to Parkinson’s disease. Journal of Neurochem-istry 81, 1285–1297.

upta, A., Dhir, A., Kumar, A., Kulkarni, S.K., 2009. Protective effect of cyclooxygenase(COX)-inhibitors against drug-induced catatonia and MPTP-induced striatallesions in rats. Pharmacology, Biochemistry, and Behavior 94, 219–226.

anrott, K., Murray, T.K., Orfali, Z., Ward, M., Finlay, C., O’Neill, M.J., Wonnacott, S.,2008. Differential activation of PKC delta in the substantia nigra of rats followingstriatal or nigral 6-hydroxydopamine lesions. European Journal of Neuroscience27, 1086–1096.

eneka, M.T., Landreth, G.E., Hüll, M., 2007. Drug insight: effects mediated by perox-isome proliferator-activated receptor-gamma in CNS disorders. Nature ClinicalPractice. Neurology 3, 496–504.

enning, J., Strauss, U., Wree, A., Gimsa, J., Rolfs, A., Benecke, R., Gimsa, U., 2008.Differential astroglial activation in 6-hydroxydopamine models of Parkinson’sdisease. Neuroscience Research 62, 246–253.

yong, A., Jadhav, V., Lee, S., Tong, W., Rowe, J., Zhang, J.H., Tang, J., 2008. Rosiglita-zone, a PPAR gamma agonist, attenuates inflammation after surgical brain injury

in rodents. Brain Research 1215, 218–224.

unter, R.L., Dragicevic, N., Seifert, K., Choi, D.Y., Liu, M., Kim, H.C., Cass, W.A., Sul-livan, P.G., Bing, G., 2007. Inflammation induces mitochondrial dysfunction anddopaminergic neurodegeneration in the nigrostriatal system. Journal of Neuro-chemistry 100, 1375–1386.

rs 213 (2012) 332– 344 343

Hunter, R.L., Choi, D.Y., Ross, S.A., Bing, G., 2008. Protective properties affordedby pioglitazone against intrastriatal LPS in Sprague-Dawley rats. NeuroscienceLetters 432, 198–201.

Iczkiewicz, J., Rose, S., Jenner, P., 2007. Osteopontin expression in activated glial cellsfollowing mechanical- or toxin-induced nigral dopaminergic cell loss. Experi-mental Neurology 207, 95–106.

Kato, H., Kurosaki, R., Oki, C., Araki, T., 2004. Arundic acid, an astrocyte-modulatingagent, protects dopaminergic neurons against MPTP neurotoxicity in mice. BrainResearch 1030, 66–73.

Kiaei, M., Kipiani, K., Chen, J., Calingasan, N.Y., Beal, M.F., 2005. Peroxisomeproliferator-activated receptor-gamma agonist extends survival in transgenicmouse model of amyotrophic lateral sclerosis. Experimental Neurology 191,331–336.

Kim, E.J., Kwon, K.J., Park, J.Y., Lee, S.H., Moon, C.H., Baik, E.J., 2002. Effects of perox-isome proliferator-activated receptor agonists on LPS-induced neuronal deathin mixed cortical neurons: associated with iNOS and COX-2. Brain Research 941,1–10.

Lang, A.E., Lozano, A.M., 1998. Parkinson’s disease. First of two parts. New EnglandJournal of Medicine 339, 1044–1053.

Leung, Y.K., Pankhurst, M., Dunlop, S.A., Ray, S., Dittmann, J., Eaton, E.D., Palumaa,P., Sillard, R., Chuah, M.I., West, A.K., Chung, R.S., 2010. Metallothionein inducesa regenerative reactive astrocyte phenotype via JAK/STAT and RhoA signallingpathways. Experimental Neurology 221, 98–106.

Liang, X., Wu, L., Wang, Q., Hand, T., Bilak, M., McCullough, L., Andreasson, K., 2007.Function of COX-2 and prostaglandins in neurological disease. Journal of Molec-ular Neuroscience 33, 94–99.

Luo, Y., Yin, W., Signore, A.P., Zhang, F., Hong, Z., Wang, S., Graham, S.H., Chen, J.,2006. Neuroprotection against focal ischemic brain injury by the peroxisomeproliferator-activated receptor-gamma agonist rosiglitazone. Journal of Neuro-chemistry 97, 435–448.

Marinova-Mutafchieva, L., Sadeghian, M., Broom, L., Davis, J.B., Medhurst, A.D.,Dexter, D.T., 2009. Relationship between microglial activation and dopamin-ergic neuronal loss in the substantia nigra: a time course study in a6-hydroxydopamine model of Parkinson’s disease. Journal of Neurochemistry110, 966–975.

McGeer, P.L., Itagaki, S., Boyes, B.E., McGeer, E.G., 1988. Reactive microglia are pos-itive for HLA-DR in the substantia nigra of Parkinson’s and Alzheimer’s diseasebrains. Neurology 38, 1285–1291.

McGeer, P.L., McGeer, E.G., 2008. Glial reactions in Parkinson’s disease. MovementDisorders 23, 474–483.

McGeer, P.L., Schwab, C., Parent, A., Doudet, D., 2003. Presence of reactivemicroglia in monkey substantia nigra years after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine administration. Annals of Neurology 54, 599–604.

Minghetti, L., 2004. Cyclooxygenase-2 (COX-2) in inflammatory and degenerativebrain diseases. Journal of Neuropathology and Experimental Neurology 63,901–910.

Mogi, M., Harada, M., Riederer, P., Narabayashi, H., Fujita, K., Nagatsu, T., 1994.Tumor necrosis factor-alpha (TNF-alpha) increases both in the brain and inthe cerebrospinal fluid from parkinsonian patients. Neuroscience Letters 165,208–210.

Moore, A.E., Cicchetti, F., Hennen, J., Isacson, O., 2001. Parkinsonian motor deficitsare reflected by proportional A9/A10 dopamine neuron degeneration in the rat.Experimental Neurology 171, 363–376.

Moreno, S., Farioli-Vecchioli, S., Cerù, M.P., 2004. Immunolocalization of peroxisomeproliferator-activated receptors and retinoid X receptors in the adult rat CNS.Neuroscience 123, 131–145.

Mori, A., Ohashi, S., Nakai, M., Moriizumi, T., Mitsumoto, Y., 2005. Neural mechanismsunderlying motor dysfunction as detected by the tail suspension test in MPTP-treated C57BL/6 mice. Neuroscience Research 51, 265–274.

Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates. AcademicPress, San Diego.

Quinn, L.P., Crook, B., Hows, M.E., Vidgeon-Hart, M., Chapman, H., Upton, N., Med-hurst, A.D., Virley, D.J., 2008. The PPARgamma agonist pioglitazone is effective inthe MPTP mouse model of Parkinson’s disease through inhibition of monoamineoxidase B. British Journal of Pharmacology 154, 226–233.

Ridet, J.L., Malhotra, S.K., Privat, A., Gage, F.H., 1997. Reactive astrocytes: cellular andmolecular cues to biological function. Trends Neuroscience 20, 570–577.

Rogers, J., Mastroeni, D., Leonard, B., Joyce, J., Grover, A., 2007. Neuroinflammation inAlzheimer’s disease and Parkinson’s disease are microglia pathogenic in eitherdisorder? International Review of Neurobiology 82, 235–246.

Rosenblad, C., Kirik, D., Björklund, A., 2000. Sequential administration of GDNF intothe substantia nigra and striatum promotes dopamine neuron survival andaxonal sprouting but not striatal reinnervation or functional recovery in thepartial 6-OHDA lesion model. Experimental Neurology 161, 503–516.

Schintu, N., Frau, L., Ibba, M., Caboni, P., Garau, A., Carboni, E., Carta, A.R., 2009. PPAR-gamma-mediated neuroprotection in a chronic mouse model of Parkinson’sdisease. The European Journal of Neuroscience 29, 954–963.

Sofroniew, M.V., 2009. Molecular dissection of reactive astrogliosis and glial scarformation. Trends Neuroscience 32, 638–647.

Sofroniew, M.V., Vinters, H.V., 2010. Astrocytes: biology and pathology. Acta Neu-ropathologica 119, 7–35.

Sriram, K., Matheson, J.M., Benkovic, S.A., Miller, D.B., Luster, M.I., O′Callaghan, J.P.,2002. Mice deficient in TNF receptors are protected against dopaminergic neu-rotoxicity: implications for Parkinson’s disease. FASEB Journal 16, 1474–1476.

Staels, B., Koenig, W., Habib, A., Merval, R., Lebret, M., Torra, I.P., Delerive, P., Fadel,A., Chinetti, G., Fruchart, J.C., Najib, J., Maclouf, J., Tedgui, A., 1998. Activation

3 Lette

S

S

T

v

44 E.Y. Lee et al. / Toxicology

of human aortic smooth-muscle cells is inhibited by PPARalpha but not byPPARgamma activators. Nature 393, 790–793.

toll, G., Jander, S., 1999. The role of microglia and macrophages in the pathophysi-ology of the CNS. Progress in Neurobiology 58, 233–247.

treit, W.J., Conde, J.R., Fendrick, S.E., Flanary, B.E., Mariani, C.L., 2005. Role ofmicroglia in the central nervous system’s immune response. NeurologicalResearch 27, 685–691.

eismann, P., Tieu, K., Choi, D.K., Wu, D.C., Naini, A., Hunot, S., Vila,M., Jackson-Lewis, V., Przedborski, S., 2003. Cyclooxygenase-2 is instru-mental in Parkinson’s disease neurodegeneration. Proceedings of theNational Academy of Sciences of the United States of America 100, 5473–

5478.

an Domburg, P.H., ten Donkelaar, H.H., 1991. The human substantia nigra andventrak tegmental area. A neuroanatomical study with notes on aging andaging disease. Advances in Anatomy, Embryology, and Cell Biology 121,1–132.

rs 213 (2012) 332– 344

Vila, M., Jackson-Lewis, V., Guégan, C., Wu, D.C., Teismann, P., Choi, D.K., Tieu, K.,Przedborski, S., 2001. The role of glial cells in Parkinson’s disease. Current Opin-ion in Neurology 14, 483–489.

Walsh, S., Finn, D.P., Dowd, E., 2011. Time-course of nigrostriatal neurodegenerationand neuroinflammation in the 6-hydroxydopamine-induced axonal and termi-nal lesion models of Parkinson’s disease in the rat. Neuroscience 175, 251–261.

Watanabe, Y., Kato, H., Araki, T., 2008. Protective action of neuronal nitric oxidesynthase inhibitor in the MPTP mouse model of Parkinson’s disease. MetabolicBrain Disease 23, 51–69.

Woster, A.P., Combs, C.K., 2007. Differential ability of a thiazolidinedionePPARgamma agonist to attenuate cytokine secretion in primary microglia and

macrophage-like cells. Journal of Neurochemistry 103, 67–76.

Yan, Q., Zhang, J., Liu, H., Babu-Khan, S., Vassar, R., Biere, A.L., Citron, M., Landreth,G., 2003. Anti-inflammatory drug therapy alters beta-amyloid processing anddeposition in an animal model of Alzheimer’s disease. Journal of Neuroscience23, 7504–7509.