SEARCH

SEARCH BY CITATION

Keywords:

  • diabetes;
  • gene expression;
  • hypertension;
  • PPAR-γ;
  • stroke;
  • thiazolidinedione

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Thiazolidinediones (TZDs) are synthetic agonists of the ligand-activated transcription factor peroxisome proliferator-activated receptor-γ (PPARγ). TZDs are known to curtail inflammation associated with peripheral organ ischemia. As inflammation precipitates the neuronal death after stroke, we tested the efficacy of TZDs in preventing brain damage following transient middle cerebral artery occlusion (MCAO) in adult rodents. As hypertension and diabetes complicate the stroke outcome, we also evaluated the efficacy of TZDs in hypertensive rats and type-2 diabetic mice subjected to transient MCAO. Pre-treatment as well as post-treatment with TZDs rosiglitazone and pioglitazone significantly decreased the infarct volume and neurological deficits in normotensive, normoglycemic, hypertensive and hyperglycemic rodents. Rosiglitazone neuroprotection was not enhanced by retinoic acid × receptor agonist 9-cis-retinoic acid, but was prevented by PPARγ antagonist GW9662. Rosiglitazone significantly decreased the post-ischemic intercellular adhesion molecule-1 expression and extravasation of macrophages and neutrophils into brain. Rosiglitazone treatment curtailed the post-ischemic expression of the pro-inflammatory genes interleukin-1β, interleukin-6, macrophage inflammatory protein-1α, monocyte chemoattractant protein-1, cyclooxygenase-2, inducible nitric oxide synthase, early growth response-1, CCAAT/enhancer binding protein-β and nuclear factor-kappa B, and increased the expression of the anti-oxidant enzymes catalase and copper/zinc-superoxide dismutase. Rosiglitazone also increased the expression of the anti-inflammatory gene suppressor of cytokine signaling-3 and prevented the phosphorylation of the transcription factor signal transducer and activator of transcription-3 after focal ischemia. Thus, PPARγ activation with TZDs might be a potent therapeutic option for preventing inflammation and neuronal damage after stroke with promise in diabetic and hypertensive subjects.

Abbreviations used
15-D-PGJ2

15-deoxy-Δ-12,14-prostaglandin J2

anova

analysis of variance

ATF

activating transcription factor

C/EBP

CCAAT/enhancer binding protein

COX

cyclooxygenase

CREB

cAMP response element binding protein

CREM

cAMP response element modulator

db

diabetic allele

DMSO

dimethyl sulfoxide

ECA

external carotid artery

Egr1

early growth response-1

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GFAP

glial fibrillary acidic protein

HIF

hypoxia inducible factor

HRP

horse radish peroxidase

IL

interleukin

iNOS

inducible nitric oxide synthase

IRF1

interferon regulatory factor-1

MCAO

middle cerebral artery occlusion

MCP

monocyte chemoattractant protein

MIP

macrophage inflammatory protein

NF-IL6

nuclear factor-IL6

NF-κB

nuclear factor kappa B

NGS

normal goat serum

PPAR

peroxisome proliferator-activated receptor

PPREs

peroxisome proliferator response elements

pSTAT3

phosphorylated STAT3

rCBF

regional cerebral blood flow

RXR

retinoic acid × receptor

SD

Sprague-Dawley

SHR

spontaneously hypertensive

SOCS3

suppressor of cytokine signaling-3

SOD

superoxide dismutase

STAT3

signal transducer and activator of transcription-3

TBS

Tris-buffered saline

TBS-T

TBS with Triton-X 100

TNF

tumor necrosis factor

TTC

2,3,5-triphenyl tetrazolium chloride

TZD

thiazolidinedione

Stroke (transient focal cerebral ischemia) is a devastating condition that causes a significant neurological dysfunction in most of the survivors. Inflammation that starts within minutes and is sustained for days after an ischemic attack is a major contributor of the secondary neuronal death after stroke (Zheng et al. 2003). Although therapies that can minimize inflammation and neuronal death can help stroke patients, no satisfactory options are currently available.

Many transcription factors are known to be induced during the acute phase after focal ischemia (Vemuganti et al. 2002; Lu et al. 2003; MacManus et al. 2004; Rickhag et al. 2006). Some of them like nuclear factor-kappa beta (NF-κB), activating transcription factor-2, early growth response-1 (Egr1), CCAAT/enhancer binding protein-β (C-EBPβ), interferon regulatory factor-1 and signal transducer and activator of transcription-3 (STAT3) were shown to promote post-ischemic inflammation and neuronal damage (Iadecola et al. 1999; Hu et al. 2000; Stephenson et al. 2000; Bowen et al. 2005; Kapadia et al. 2006; Satriotomo et al. 2006). On the other hand, transcription factors like hypoxia inducible factor-1, cyclic AMP response element binding protein, c-fos and P53 are known to prevent ischemic brain damage (Bergeron et al. 1999; Tanaka et al. 2000; Cho et al. 2001; Maeda et al. 2001).

Peroxisome proliferator-activated receptor-γ (PPARγ) is a ligand-activated transcription factor of the nuclear hormone receptor superfamily. PPAR isoforms (α, γ, and δ/β) modulate multiple cellular functions including glucose absorption, lipid metabolism, cell growth and differentiation, apoptosis, and inflammation (Escher and Wahli 2000; Blanquart et al. 2003; Argmann et al. 2005; Kiec-Wilk et al. 2005). When a ligand binds to PPARγ, it dimerizes with retinoid × receptor (R × R) to form a heterodimeric complex that binds to the cis-acting sequences (peroxisome proliferator response elements; PPREs) on DNA to initiate or trans-repress the transcription of target genes (Escher and Wahli 2000). 15-deoxy-Δ-12,14-prostaglandin J2 (15-D-PGJ2) is the natural agonist, and thiazolidinediones (TZDs; troglitazone, ciglitazone, rosiglitazone, and pioglitazone) are potent synthetic agonists of PPARγ. Of these, troglitazone was removed from the market due to hepatotoxicity, while rosiglitazone and pioglitazone are currently FDA approved for type-2 diabetes treatment (Cheng-Lai and Levine 2000; Tolman and Chandramouli 2003). In addition to their blood glucose lowering effects, PPARγ agonists were shown to control inflammation associated with gut, myocardial, lung and cerebral ischemia (Ichikawa et al. 2002; Okada et al. 2002; Wayman et al. 2002; Sundararajan et al. 2005).

We currently evaluated the therapeutic efficacy of rosiglitazone and pioglitazone in preventing neuronal damage and neurological dysfunction after transient focal ischemia in adult rats and mice. As hypertension and diabetes are potent precipitators of post-stroke brain damage, we also studied the effectiveness of TZDs in preventing ischemic brain damage in spontaneously hypertensive (SHR) rats and type-2 diabetic db/db (hyperglycemic) mice. We tested if the post-ischemic neuroprotection induced by PPARγ agonist rosiglitazone can be (a) enhanced by combining with a RXR agonist 9-cis-retinoic acid (9-cis-RA), and (b) prevented by PPARγ antagonist GW9662. To identify the mechanism of action of TZD neuroprotection, we studied the effect of rosiglitazone on post-ischemic leukocyte infiltration, inflammatory gene expression, anti-oxidant enzyme expression and suppressor of cytokine signaling-3 (SOCS3) and STAT3 activation.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals

All the surgical procedures were approved by the Research Animal Resources and Care Committee of the University of Wisconsin-Madison and the animals were cared for in accordance with the Guide for the Care and Use of Laboratory Animals, U.S. Department of Health and Human Services Publication Number. 86-23 (revised). We used 208 C57BL/6 mice, 23 SV129 mice (129 S2/SvPasCrl developed by Dr L.C. Stevens of Jackson Laboratories), 62 db/db (C57BLKS/J-m+/+Leprdb/db; type-2 diabetic) mice, 62 db/+ (normoglycemic genetic controls of db/db) mice, 32 SHR rats and 20 Sprague–Dawley (SD) rats. The db/db and db/+ mice were obtained from the Jackson Laboratories (Bar Harbor, ME, USA) and all other rodents were obtained from the Charles River Laboratories (Wilmington, MA, USA).

Drug treatment

Rosiglitazone potassium salt, pioglitazone potassium salt, 15d-PGJ2 in a solution in methyl acetate and GW9662 were purchased from Cayman Chemicals (Ann Arbor, MI, USA). The RXR agonist 9-cis-RA was obtained from Sigma Chemicals (St Louis, MO, USA). The rosiglitazone and pioglitazone potassium salts were dissolved in phosphate buffered saline (PBS; pH 7.2). In the case of 15-d-PGJ2 the methyl acetate was evaporated under a gentle stream of nitrogen and the neat oil was immediately dissolved in PBS. GW9662 and 9-cis-RA were dissolved in dimethyl sulfoxide (DMSO) and diluted in PBS to decrease the DMSO concentration to 3%. All drugs were prepared just before use and injected intraperitoneal (i.p.) at various time points (4 h before ischemia to 6 h of reperfusion) and doses (0.5–6 mg/kg) as indicated in the results section. The controls were injected with either PBS or 3% DMSO or both based on the experiment. In some experiments, the db/db and db/+ mice were fed for 3 weeks on a rodent chow enriched with 100 ppm rosiglitazone (Research Diets, New Brunswick, NJ, USA) or with metformin (200 mg/kg/day; Sigma Chemicals) in drinking water.

Focal ischemia

Transient middle cerebral artery occlusion (MCAO) was induced by the intraluminal suture method as described earlier (Dhodda et al. 2004; Tureyen et al. 2004; Vemuganti et al. 2004; Kapadia et al. 2006; Satriotomo et al. 2006; Yan et al. 2006). In brief, a rat or a mouse was anesthetized with halothane (induction: 2%; maintenance: 1.2% in an oxygen and nitrous oxide 50 : 50 mixture), placed in a stereotaxic frame fitted with a nose cone with halothane anesthesia. The left femoral artery was cannulated for continuous monitoring of arterial blood pressure and to obtain the measurements of pH, PaO2, PaCO2, hemoglobin and blood glucose concentration (i-STAT; Sensor Devices, Waukesha, WI, USA). The rectal temperature was controlled at 37.0 ± 0.5°C during surgery with a feedback-regulated heating pad. After a midline skin incision, the left external carotid artery (ECA) was exposed, and its branches were coagulated. A surgical monofilament nylon suture blunted at the end (3-0 for rats and 6-0 for mice) was introduced into the ECA lumen and gently advanced to the internal carotid artery until the regional cerebral blood flow (rCBF) was reduced to 10–16% of the baseline (recorded by laser Doppler flowmeter; Vasamedics, LLC, St Paul, MN, USA) as described earlier (Tureyen et al. 2004; Vemuganti et al. 2004). The MCA was occluded for 1 h in rats, 2 h in the C57BL/6 and SV129 mice, and 45 min in the db/db and db/+ mice. The restoration of the blood flow was confirmed by laser Doppler following the withdrawal of the suture. After suturing the wound, the animals were allowed to recover from anesthesia and returned to the cage with ad libitum access to food and water. During the surgery, the animals were under spontaneous respiration. The animals were sacrificed either at 6 h of reperfusion (mouse gene expression studies) or 1 day (rat gene expression studies) or 3 days (for infarct volume estimation and immunohistochemistry) of reperfusion.

Infarct volume estimation

Infarct volume was measured as described earlier using either the 2,3,5-triphenyl tetrazolium chloride monohydrate (TTC) staining or Cresyl violet staining (Dhodda et al. 2004; Tureyen et al. 2004; Vemuganti et al. 2004; Kapadia et al. 2006; Satriotomo et al. 2006). For TTC staining, each mouse brain was placed in ice-cold saline for 10 min, and then cut into 1-mm coronal slices in a mouse brain matrix (Activational Systems, Warren, MI, USA). The slices were stained with 2% TTC for 20 min at 37°C, scanned on a flat-bed scanner and the infarct volume was estimated by NIH Image J program version 1.29x (written by Dr Wayne Rasband and can be downloaded at no cost from http://rsb.info.nih.gov/ij). For Cresyl violet staining, the animal was perfused transcardially with buffered paraformaldehyde, the brain was post-fixed, cryoprotected and sectioned (coronal; 40 µm thick at an interval of 320 µm). The serial sections were stained with Cresyl violet and scanned using the NIH Image program. The volume of the ischemic lesion was computed by the numeric integration of data from serial sections (5–6 in case of mouse and 8–9 in case of rat) in respect to the sectional interval. To account for edema and differential shrinkage resulting from tissue processing, the injury volumes were corrected by using the Swanson formula: corrected injury volume = contralateral hemisphere volume – (ipsilateral hemisphere volume-measured injury volume) (Swanson et al. 1990).

Neurological evaluation

Post-ischemic neurological deficits were evaluated on a 6-point scale before transient MCAO, and at 3 days of reperfusion by an investigator blinded to the study groups as described earlier (Zea Longa et al. 1989; Vemuganti et al. 2001, 2004; Kapadia et al. 2006). A score of 0 suggests no neurological deficit (normal), 1 suggests mild neurological deficit (failure to extend right forepaw fully), 2 suggests moderate neurological deficit (circling to the right), 3 suggests severe neurological deficit (falling to the right), and 4 suggests very severe neurological deficit (the rat could not walk spontaneously; depressed level of consciousness).

Immunohistochemistry

Parallel sets of 40 µm coronal brain sections from each rat were immunostained with antibodies against intercellular adhesion molecule-1 (ICAM1), ED1, OX42, cyclooxygenase-2 (COX2), inducible nitric oxide synthase (iNOS), catalase and phosphorylated STAT3 (pSTAT3) using the standard immunohistochemical procedures. The following antibodies were used: mouse anti-rat ED1 (Cd68; 1 : 500; Serotec, Oxford, UK), mouse anti-rat OX42 (CD11b; 1 : 1000; BD Pharmingen, San Jose, CA, USA), monoclonal ICAM1 (mouse IgG1, Clone 1A29; 1 : 2000; BD Pharmingen), rabbit polyclonal COX2 (1 : 1000; Cayman Chemicals), rabbit polyclonal iNOS (1 : 1000; Chemicon, Temecula, CA, USA), rabbit polyclonal catalase (1 : 1000; Abcam, Cambridge, MA, USA) and polyclonal pSTAT3 (pTyr705) (1 : 500; Cell Signaling Technologies, Beverly, MA, USA). In brief, the sections were rinsed in 0.1 mol/L Tris-buffered saline with 0.1% Triton-X100 (TBS-T; 3 × 5 min), incubated in 1% H2O2 for 30 min and washed in TBS-T (3 × 5 min). The sections were then incubated in the desired primary antibody overnight at 4°C. The sections were washed in TBS-T (3 × 5 min), incubated for 1 h in anti-rat or anti-mouse biotinylated secondary antibodies from Vector Labs (1 : 200; Burlingame, CA, USA). Conjugation with avidin-biotin complex (1 : 100; Vecstatin Elite ABC kit; Vector Labs) was followed by visualization with 3,3′-diaminobenzidine-hydrogen peroxidase (Vector Labs). The sections were dehydrated, cleared and mounted in Permount. Sections incubated without primary or secondary antibodies served as negative controls. A parallel set of sections was subjected to Hematoxylin and Eosin (H&E; Fisher Diagnostics, Middletown, VA, USA) staining to identify neutrophils infiltrated into brain parenchyma by their polymorphonuclear structure as described earlier (Kapadia et al. 2006). The ICAM1 immunopositive vessels, ED1 positive macrophages, OX42 positive microglia and the neutrophils were counted in four to six X300 fields in the ipsilateral cortex of each rat. To ensure that homologous areas of injury were sampled between animals, parallel sets of sections from +0.5 to −1.5 mm from Bregma were used for the H&E, ICAM1, ED1, and OX42 staining. The sections from this coordinates showed consistent infarct in all rats.

Real-time PCR analysis

Gene expression analysis was conducted using quantitative real-time PCR as described earlier (Dhodda et al. 2004; Vemuganti et al. 2004, 2006; Vemuganti and Dempsey 2005). The following transcripts known to be altered after focal ischemia were estimated: interleukin (IL)1β, IL6, macrophage inflammatory protein-1α (MIP-1α), monocyte chemoattractant protein-1 (MCP1), ICAM1, Egr1, C-EBPβ, NF-κB, iNOS, COX2, catalase, copper/zinc-superoxide dismutase (Cu/Zn-SOD) and SOCS3. Cohorts of SHR rats subjected to transient MCAO and treated with either rosiglitazone (2 mg/kg at 1 min and 2 h of reperfusion) or vehicle were sacrificed at 1 day of reperfusion (n = 4 per group). Total RNA was extracted from the ipsilateral cortex of each rat using the Trizol reagent (Invitrogen, Carlsbad, CA, USA). 1 μg of RNA from each rat was reverse transcribed with oligo(dT)15 and random hexamer primers using M-MuLV reverse transcriptase (Life Technologies, Rockville, MD, USA). As much as 10 ng of cDNA and gene-specific primers were added to SYBR Green PCR Master Mix (SYBR Green I Dye, AmpliTaq DNA polymerase, dNTPs with dUTP and optimal buffer components; Applied Biosystems) and subjected to PCR amplification in a Perkin-Elmer TaqMan 5700 Sequence Detection System (1 cycle at 50°C for 2 min, 1 cycle at 95°C for 10 min, and 40 cycles at 95°C for 15 s and 60°C for 1 min). PCR reactions were conducted in duplicate. The amplified transcripts were quantified with the comparative CT method using 18S rRNA and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the internal controls as described earlier (Dhodda et al. 2004; Vemuganti et al. 2006). The real-time PCR primers were designed using the Primer Express software (Applied Biosystems) based on the GenBank accession numbers. The primer sequences were given in our recent papers (Dhodda et al. 2004; Vemuganti et al. 2004, 2006; Naylor et al. 2005). In a separate experiment, total RNA was extracted from the ipsilateral cortex of 16 mice subjected to 45 min transient MCAO and 6 h reperfusion (four each of db/db rosiglitazone chow, db/db control chow, db/+ rosiglitazone chow and db/+ control chow). RNA samples extracted from the cortex of 8 sham-operated rats (two each of db/db rosiglitazone chow, db/db control chow, db/+ rosiglitazone chow and db/+ control chow) were used as controls. Total RNA from each mouse was converted to cDNA and subjected to real-time PCR analysis using primers specific for mouse IL6 (NM_031168), IL1β (NM_008361), MCP1 (AF065933), ICAM1 (BC008626), P-selectin (NM_011347) and 18 s rRNA. The procedures were essentially same as described above for the rat experiments.

Statistical analysis

The data are expressed as mean ± SD. Comparisons among groups were performed by one-way anova with Tukey–Kramer multiple comparisons post-test.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Thiazolidinediones decreased post-ischemic infarction in a dose- and time-dependent manner

In normotensive mice (C57BL/6), a two dose pre-treatment (at 4 h before MCAO and 1 min of reperfusion) with either rosiglitazone (2 or 4 mg/kg) or pioglitazone (2 or 4 mg/kg) significantly decreased the transient MCAO-induced infarct volume compared to vehicle control (rosi: by 59–66% and pio: by 47–57%; < 0.05; n = 7 to 10/group) (Fig. 1a). There was a significant decrease in the infarct volume even when mice were post-treated (at 1 min and 2 h of reperfusion) with 2 or 4 mg/kg with rosiglitazone (by 49–54%; < 0.05; n = 7/group) and pioglitazone (by 36–43%; < 0.05; n = 7/group) (Fig. 1a). A single injection of rosiglitazone at 2 h reperfusion also induced a dose-dependent decrease in infarct volume from 1 mg/kg (by 37%; < 0.05; n = 5/group) to 4 mg/kg (by 58%; < 0.05; n = 5/group) compared with vehicle control (Fig. 1b). Lower doses of rosiglitazone (0.25 and 0.5 mg/kg given at 2 h reperfusion) did not induce any neuroprotection; and a higher dose of 6 mg/kg had no further benefit than the 4 mg/kg dose (Fig. 1b). In addition, a single injection of 1 mg/kg pioglitazone given at 2 h reperfusion (n = 5) induced no significant decrease in the infract volume compared with vehicle control (data not shown). We evaluated the window of neuroprotection with a single dose of 2 mg/kg rosiglitazone given at different reperfusion time points after transient MCAO. The infarct volume was significantly less when rosiglitazone was given at 1 min (by 53%, < 0.05; n = 5), 2 h (by 40%, < 0.05; n = 5) and 3 h (by 21%< 0.05; n = 5) of reperfusion compared with vehicle control (Fig. 1c). Rosiglitazone (2 mg/kg) given at 4 or 6 h of reperfusion induced no neuroprotection (Fig. 1c).

image

Figure 1.  Thiazolidinediones decreased post-ischemic infarction in adult mouse brain. A 2-dose pre-treatment as well as a 2-dose post-treatment with either rosiglitazone or pioglitazone at a dose of 2 or 4 mg/kg significantly decreased the transient middle cerebral artery occlusion-induced infarction (a). A 1-dose rosiglitazone given at 2 h of reperfusion dose-dependently decreased the infarct volume from 1 to 4 mg/kg (b). A 1-dose 2 mg/kg rosiglitazone decreased infarct volume when administered between 1 min and 3 h of reperfusion following transient middle cerebral artery occlusion (c). In (a), the number of mice used were 10 for the vehicle (five pre-treated and five post-treated with no significant difference between them in infarct volume), nine each for the 2 mg/kg pre-treatment groups and seven each for the other groups. In (b) and (c), five mice were used for each group. Bars in the histograms represent mean ± SD. *< 0.05 compared with the respective vehicle control by one-way anova followed by Tukey–Kramer multiple comparisons post-test. PBS was injected to vehicle controls as rosiglitazone and pioglitazone potassium salts were dissolved in PBS. The compounds were injected i.p. In all these studies, MCA was occluded for 2 h and the infarct volume was estimated using the TTC stained brain sections at 3 days of reperfusion.

Download figure to PowerPoint

Thiazolidinediones prevented post-ischemic neurological dysfunction without altering rCBF

In adult mice subjected to transient MCAO, the two dose post-treatment (at 1 min and 2 h reperfusion) with 2 mg/kg rosiglitazone or pioglitazone significantly reduced the neuroscores measured at 3 days of reperfusion reflecting an improved neurological function (Fig. 2a). Moderate to severe neurological dysfunction was observed in the vehicle-treated mice (average neuroscore of 2.74; n = 10) (Fig. 2a), whereas only mild neurological dysfunction was seen in rats treated with rosiglitazone (average neuroscore: 1.33; n = 9) and pioglitazone (average neuroscore: 1.54; n = 9) (Fig. 2a). Decreased neuroscores were also observed in the other groups of mice pre- and post-treated with TZDs (data not shown). Figure 2b shows TTC-stained coronal brain sections from representative mice of the three groups (vehicle, 2 mg/kg rosiglitazone and 2 mg/kg pioglitazone at 1 min and 2 h of reperfusion) indicating smaller infarcts in the drug groups. The rCBF measured as a function of occlusion and reperfusion was not significantly different between the vehicle and 2 mg/kg rosiglitazone or pioglitazone post-treatment (at 1 min and 2 h) groups (Fig. 2c). The physiological parameters (pH, PaO2, paCO2, hemoglobin and blood glucose) measured before, during and after MCAO were also not significantly different between the vehicle, rosiglitazone and pioglitazone groups (data not shown).

image

Figure 2.  Thiazolidinediones prevented neurological dysfunction without altering regional cerebral blood flow. The neuroscores measured at 3 days of reperfusion after transient middle cerebral artery occlusion for each mouse of the vehicle, rosiglitazone (2 mg/kg at 2 h) and pioglitazone (2 mg/kg at 2 h) groups are shown in (a); (b) shows TTC stained brain sections from the representative mice of each group; (c) shows that the regional cerebral blood flow measured using laser-Doppler before, during and after middle cerebral artery occlusion was not significantly different between the three groups. The values are mean ± SD (n = 10 for vehicle group and nine each for the rosiglitazone and pioglitazone groups).

Download figure to PowerPoint

Thiazolidinedione neuroprotection was not enhanced by RXR agonist; but prevented by peroxisome proliferator-activated receptor-γ antagonist

Ligand-activated PPARγ dimerizes with ligand-activated RXR and the complex translocates into the nucleus to induce the expression of PPRE-containing genes. Hence, we tested if supplementing with the RXR agonist 9-cis-RA enhances the neuroprotective effect of rosiglitazone after focal ischemia. 9-cis-RA (2 mg/kg at 2 h reperfusion) alone had no significant effect on the transient MCAO-induced infarct volume compared to DMSO (vehicle) control (n = 5/group; Fig. 3). Furthermore, there was no significant difference in the decrease in the infarct volume between the groups treated with rosiglitazone alone (2 mg/kg at 2 h) and a combination of rosiglitazone and 9-cis-RA (both 2 mg/kg at 2 h) compared with their respective vehicle controls (n = 5/group; Fig. 3). The infarct volumes were not significantly different between the three control groups (PBS, 3% DMSO and PBS + DMSO; n = 5/group) (Fig. 3). Pre-treating mice with the PPARγ antagonist GW9662 (4 mg/kg at 90 min of reperfusion) prevented the neuroprotection induced by 2 mg/kg rosiglitazone (given at 2 h of reperfusion) by 95% (< 0.05; n = 5/group).

image

Figure 3.  Retinoic acid × receptor agonist had no effect while peroxisome proliferator-activated receptor-γ antagonist prevented the rosiglitazone neuroprotection. Retinoic acid × receptor agonist 9-cis-RA alone or in combination with rosiglitazone had no significant effect on the infarct volume following transient middle cerebral artery occlusion. On the other hand, pre-treating mice with peroxisome proliferator-activated receptor-γ antagonist GW9662 at a dose of 4 mg/kg prevented the rosiglitazone-induced decrease in infarct volume. Bars represent mean ± SD of n = 5 mice per group. *< 0.05 compared with the respective vehicle control (PBS in case of rosiglitazone and 3% DMSO in case of 9-cis-RA and GW9662) by one-way anova followed by Tukey–Kramer multiple comparisons post-test. The MCA occlusion was for 2 h and the reperfusion was for 3 days in all animals.

Download figure to PowerPoint

Peroxisome proliferator-activated receptor-γ ligand-induced neuroprotection is not strain dependent

We tested the efficacy of 2 mg/kg rosiglitazone, 2 mg/kg pioglitazone and 1 mg/kg 15-d-PGJ2 (all given at 2 h reperfusion) in SV129 mice in comparison with C57BL/6 mice subjected to transient MCAO. The infarct volume observed at 3 days of reperfusion following transient MCAO was similar between the vehicle-treated SV129 and C57BL/6 mice (n = 5/group) (Fig. 4). Compared with their respective vehicle-treated control groups, both strains showed similar decrease in the infarct volume when treated with rosiglitazone (by 39% in SV129 and by 45% in C57BL/6 < 0.05; n = 5/group), pioglitazone (by 34% in SV129 and by 35% in C57BL/6; < 0.05; n = 5/group) and 15-d-PGJ2 (by 49% in SV129 and by 43% in C53BL/6; < 0.05; n = 5/group) (Fig. 4).

image

Figure 4.  peroxisome proliferator-activated receptor-γ agonist-induced neuroprotection is strain independent. A single injection of rosiglitazone (2 mg/kg) or pioglitazone (2 mg/kg) or 15-d-PGJ2 (1 mg/kg) given i.p. at 2 h of reperfusion decreased the infarct volume in the SV129 mice in addition to C57BL/6 mice subjected to transient middle cerebral artery occlusion. Bars represent mean ± SD of n = 5 mice per group for the drug groups and eight mice for the vehicle group. As there was no significant difference between the PBS and 3% DMSO injected controls, the two vehicle groups (n = 4 each) were combined. *< 0.05 compared with the respective vehicle group by one-way anova followed by Tukey–Kramer multiple comparisons post-test.

Download figure to PowerPoint

Effect of rosiglitazone treatment on ischemic brain damage in type-2 diabetic mice

Millions of type-2 diabetics currently use TZDs to control the blood glucose and lipid balance. Diabetics are at an increased risk of stroke incidence and stroke causes more damage in diabetics compared with normoglycemic individuals. Hence, we tested the efficacy of rosiglitazone in adult db/db mice (a model of type-2 diabetes with high blood sugar level of 433 ± 61 mg/dL; n = 54) in comparison with db/+ (normoglycemic littermates with a blood sugar level of 122 ± 14 mg/dL; n = 54). Following a 2 h transient MCAO (which was used in the above described normoglycemic mouse studies), the mortality rate in the first 24 h of reperfusion was observed to be 75% for the db/db mice (six of eight died) compared with 13% for the db/+ mice (one of eight died). Hence, we used a 45 min transient MCAO followed by 3 days of reperfusion to assess the effect of rosiglitazone on infarct volume in db mice. Rosiglitazone treatment (2 mg/kg at 2 h reperfusion) decreased the infarct volume by 54% in the db/+ mice, but only by 21% in the db/db mice compared with respective vehicle control (< 0.05; n = 7/group) (Fig. 5a). Even pre-treatment (at −4 h and 1 min reperfusion) with a higher dose of rosiglitazone (4 mg/kg) decreased the infarct volume by only 30% in the db/db mice compared with vehicle control (< 0.05; n = 7/group) (Fig. 5a). In the normoglycemic db/+ mice the 4 mg/kg two-dose pre-treatment decreased the infarct volume by 68% (< 0.05; n = 7) compared with vehicle control (Fig. 5a). The acute pre- or post-MCAO treatment with rosiglitazone had no significant effect on the blood glucose levels measured at 6 h, 1 day and 3 days of reperfusion compared with pre-MCAO levels in either the db/+ or the db/db mice (Table 1). In human type-2 diabetics, there is a lag phase of c. 3 weeks before the beneficial blood glucose lowering effect can be seen after starting the TZD treatment. Hence, we fed a cohort of db/db and db/+ mice with a chow fortified with rosiglitazone for 3 weeks before inducing transient MCAO. Feeding the normoglycemic db/+ mice on rosiglitazone chow had no significant effect on the blood glucose levels at 14 or 21 days or after MCAO (up to 3 days of reperfusion) compared with either the pre-treatment levels or the time-matched normal chow-fed controls (Table 1). On the other hand, the db/db mice fed on rosiglitazone chow showed a significant decrease in the blood glucose levels at both 14 days (by 21%; < 0.05) and 21 days (by 32%; < 0.05) compared with the pre-treatment levels (Table 1). In db/+ mice feeding rosiglitazone chow showed no significant benefit in decreasing the infarct volume compared with a bolus pre- or post-treatment with 2 mg/kg rosiglitazone (Fig. 5a). However, in db/db mice, the long-term oral pre-treatment with rosiglitazone induced a significantly better neuroprotection (47% decrease in infarct volume compared with normal chow fed db/db controls; < 0.05; n = 7/group) than bolus injections (Fig. 5a and b).

image

Figure 5.  Rosiglitazone induced neuroprotection in type-2 diabetic mice. A 2-dose pre-treatment (at −4 h and 1 min of reperfusion with 4 mg/kg) or a 1 dose post-treatment (at 2 h of reperfusion with 2 mg/kg) with rosiglitazone significantly decreased the infarct volume in hyperglycemic (db/db) as well as normoglycemic (db/+) mice subjected to transient middle cerebral artery occlusion (MCAO) (a). In a separate experiment, cohorts of db/db and db/+ mice subjected to transient MCAO following feeding on rosiglitazone-fortified chow for 3 weeks showed significantly smaller infarcts compared with db/db and db/+ mice fed on normal chow (a). Mice subjected to transient MCAO following feeding with metformin for 3 weeks had no significant effect on the infarct volume in either db/db or db/+ mice (a). The Cresyl violet stained brain sections from representative db/db mice fed with control chow, rosiglitazone chow and metformin were shown in (b). For estimating the infarct volume, the mice were subjected to 45 min of transient MCAO and 3 days of reperfusion. Transient MCAO (45 min) and 6 h of reperfusion significantly increased the expression of several inflammatory genes in the ipsilateral cortex of db/db and db/+ mice fed on control chow compared with respective sham controls (c). The db/db and db/+ mice fed on rosiglitazone chow showed significantly curtailed expression of the inflammatory genes compared with control chow fed groups (c). For each genotype, the sham-operated rats fed on control and rosiglitazone chows were pooled as they showed no significant difference in any of the transcripts analyzed. Bars in the histogram represent mean ± SD of n = 7 mice per group in (a) and n = 4 mice per group in (c). *< 0.05 compared with the respective control (in a), a< 0.05 compared with the sham group (in c) and b< 0.05 compared with the MCAO/vehicle group (in c) by one-way anova followed by Tukey–Kramer multiple comparisons post-test.

Download figure to PowerPoint

Table 1.   Blood glucose levels in the db/+ and db/db mice subjected to transient middle cerebral artery occlusion (MCAO)
 db/+db/db
VehicleRosi-preRosi-postVehicleRosi-preRosi-post
  1. The blood glucose levels are given as mg/dL. The value is mean ± SD (n). The 0 h (pre-MCAO) values in the acute experiments and the before treatment values in the chronic experiments were pooled.

  2. *p < 0.05 compared with the respective before treatment group (by anova followed by Tucky–Kramer multiple comparisons post-test).

0 h (pre-MCAO)120 ± 14 (21)432 ± 56 (21)
6 h reperfusion133 ± 16 (7)126 ± 17 (7)124 ± 16 (7)433 ± 74 (7)419 ± 81 (7)436 ± 77 (7)
24 h reperfusion117 ± 19 (7)113 ± 23 (7)116 ± 18 (7)387 ± 97 (7)371 ± 92 (7)391 ± 84 (7)
72 h reperfusion113 ± 20 (7)115 ± 19 (7)113 ± 17 (7)379 ± 97 (7)364 ± 89 (7)368 ± 89 (7)
 Normal chowRosi chowMetformin treatedNormal chowRosi chowMetformin treated
Before treatment127 ± 10 (26) 122 ± 14 (7)447 ± 76 (26)433 ± 81 (7)
14 days121 ± 15 (13)116 ± 14 (13)119 ± 15 (7)449 ± 96 (13) 353 ± 67* (13)360 ± 61* (7)
21 days (pre-MCAO)127 ± 12 (13)121 ± 18 (13)124 ± 9 (7)433 ± 92 (13)303 ± 63* (13)304 ± 52* (7)
6 h reperfusion123 ± 14 (11)117 ± 15 (7)120 ± 14 (7)451 ± 81 (11)293 ± 64* (11)300 ± 63* (7)
24 h reperfusion115 ± 13 (7)110 ± 19 (7)115 ± 15 (7)423 ± 73 (7)290 ± 59* (7)290 ± 67* (7)
72 reperfusion113 ± 14 (7)107 ± 16 (7)103 ± 15 (7)398 ± 81 (7)286 ± 63* (7)293 ± 64* (7)

In the hyperglycemic db/db mice, the long-term oral treatment with metformin (200 mg/kg/day in drinking water) significantly decreased the blood glucose levels at 14 days (by 17%; < 0.05) and 21 days (by 29%; < 0.05) compared with pre-treatment levels (Table 1). Whereas, metformin had no significant effect of blood glucose levels in the normoglycemic db/+ mice (Table 1). Metformin treatment had no significant effect on the infarct volume compared with vehicle control in either the db/+ or the db/db mice (Fig. 5a and b).

To understand if the rosiglitazone neuroprotection in diabetic mice involves prevention of inflammation, we estimated the expression of five proinflammatory transcripts viz. IL6, IL1β, MCP1, ICAM1 and E-selectin using the mRNA extracted from the ipsilateral cortex of db/db and db/+ mice subjected to transient MCAO and 6 h of reperfusion following feeding on rosiglitazone and control chow for 3 weeks. Both the genotypes (db/db and db/+) fed on control chow showed significantly elevated mRNA expression of all these transcripts following transient MCAO compared with the sham (Fig. 5c). However, the post-ischemic fold increases for all the transcripts were significantly higher for the db/db group compared with db/+ group (Fig. 5c). Feeding mice on rosiglitazone chow significantly prevented the post-ischemic induction of all these transcripts by c. 60 to 90% in both the db/db and db/+ mice (Fig. 5c).

Thiazolidinediones induced neuroprotection in hypertensive rats

As hypertension significantly increases stroke complications and infarct size (McCulloch 1996;Dogan et al. 1998; Amenta et al. 2003), we tested the efficacy of TZDs after transient MCAO in hypertensive SHR rats in comparison with the normotensive SD rats. We presently observed that SHR rats show 21% bigger infarcts than SD rats following 1 h transient MCAO (Fig. 6). In both strains, a 2-dose pre-treatment (at 1 min and 2 h reperfusion) with 2 mg/kg rosiglitazone significantly decreased the infarct volume by 45–56% (< 0.05; n = 5/group) compared with respective vehicle controls (Fig. 6). A 1 dose post-treatment with 2 mg/kg rosiglitazone (at 2 h reperfusion) significantly decreased the infarct volume (by 39% in SD and by 47% in SHR; < 0.05; n = 5/group) compared with the respective vehicle controls (Fig. 6). In both strains, 2 mg/kg pioglitazone given at 2 h reperfusion also significantly decreased the infarct volume compared with the vehicle controls (by 30% in SD and by 42% in SHR; < 0.05; n = 5/group) (Fig. 6). Furthermore, the post-ischemic neurological deficits estimated on a 6-point scale were none to mild in all the TZD groups, and moderate to severe in the vehicle groups (data not shown). The rCBF, arterial blood pressure and the other physiological parameters measured during the MCAO and reperfusion were not significantly different between the vehicle, rosiglitazone and pioglitazone groups (Table 2).

image

Figure 6.  Thiazolidinediones induced neuroprotection in hypertensive rats. Adult SD and SHR rats subjected to a 1 h transient middle cerebral artery occlusion and 3 days of reperfusion showed significantly smaller infarcts when pre-treated with rosiglitazone (2 mg/kg at −4 h and 1 min of reperfusion) or post-treated with rosiglitazone or pioglitazone (both 2 mg/kg at 2 h reperfusion). Bars represent mean ± SD of n = 5 rats per group. *< 0.05 compared with the vehicle group by one-way anova followed by Tukey–Kramer multiple comparisons post-test.

Download figure to PowerPoint

Table 2.   Physiological parameters in rats subjected to transient middle cerebral artery occlusion (MCAO)
 SHR ratsSD rats
MABP mm HgrCBF % of baselineGlucose mg/dLMABP mm HgrCBF % of baselineGlucose mg/dL
  1. Values are given as mean ± SD (n = 5/group).

  2. rCBF, regional cerebral blood flow. Rosi (pre) group was treated with 2 mg/kg rosiglitazone at 4 h and 1 min of reperfusion. Rosi (post) and pio (post) groups were treated with 2 mg/kg rosiglitazone and 2 mg/kg pioglitazone at 2 h of reperfusion. In all the rats, the other physiological parameters were observed to be in the following ranges before, during and after MCAO (PaO2: 117 ± 16, PaCO2: 32 ± 6, pH: 7.39 ± 0.4, brain temperature (°C): 36.5 ± 0.4 and body temperature (°C): 36.9 ± 0.5.

  3. There were no statistically significant differences between the different treatment groups [vehicle, rosi (pre), rosi (post) and pio (post)] at a specific period or within a treatment group at different periods (pre-ischemia, during ischemia and post-ischemia) in any of the parameters.

Pre-ischemia (30 min before start of occlusion)
 Vehicle135 ± 9101 ± 674 ± 697 ± 699 ± 781 ± 5
 Rosi (pre)138 ± 10104 ± 781 ± 793 ± 6105 ± 877 ± 7
 Rosi (post)133 ± 799 ± 572 ± 796 ± 7102 ± 682 ± 6
 Pio (post)141 ± 9103 ± 780 ± 8100 ± 897 ± 877 ± 7
During Ischemia (30 min after start of occlusion)
 Vehicle132 ± 1113 ± 279 ± 596 ± 816 ± 375 ± 7
 Rosi (pre)129 ± 1014 ± 376 ± 499 ± 715 ± 278 ± 6
 Rosi (post)135 ± 811 ± 374 ± 5102 ± 917 ± 477 ± 8
 Pio (post)136 ± 1113 ± 378 ± 597 ± 816 ± 482 ± 8
Post-ischemia (30 min after start of reperfusion)
 Vehicle141 ± 9111 ± 577 ± 495 ± 9105 ± 884 ± 5
 Rosi (pre)139 ± 10114 ± 681 ± 3103 ± 6109 ± 876 ± 6
 Rosi (post)142 ± 8115 ± 776 ± 5105 ± 9108 ± 582 ± 7
 Pio (post)138 ± 11107 ± 677 ± 5100 ± 8113 ± 779 ± 7

Rosiglitazone decreased post-ischemic ICAM1 expression and inflammatory cell infiltration

Following focal ischemia, increased expression of the adhesion molecules promotes extravasation of neutrophils and macrophages into brain parenchyma that promote inflammation. We observed several ICAM1 immunopositive capillaries (Fig. 7a and b), ED1+ macrophages (Fig. 7c), OX42+ activated microglia (Fig. 7d), and neutrophils (Fig. 7e) in the ipsilateral cortex of the MCAO/vehicle group of rats at 3 days of reperfusion. Rosiglitazone (2 mg/kg at 2 h of reperfusion) treatment significantly curtailed the post-ischemic increase of ICAM1 positive vessels (Fig. 7f and g), ED1+ macrophages (Fig. 7h), OX42+ activated microglia (Fig. 7i), and the neutrophils (Fig. 7j) compared with vehicle controls. The sham-operated rats showed no appreciable immunostaining of ICAM1 (Fig. 7k), ED1 (Fig. 7l) and OX42 (Fig. 7m), or extravasated neutrophils (Fig. 7n). The cell numbers in the MCAO/vehicle and MCAO/rosiglitazone groups are given in Fig. 7o. The mRNA expression of the adhesion molecules ICAM1, E-selectin and P-selectin was significantly lower in the MCAO/rosiglitazone group compared to MCAO/vehicle group (n = 4/group; < 0.05 in each case) at 1 day of reperfusion (Fig. 7p).

image

Figure 7.  Rosiglitazone treatment prevented post-ischemic adhesion molecule expression and inflammatory cell infiltration. Following transient middle cerebral artery occlusion (MCAO), the cortex of the vehicle treated rats showed several ICAM1 immunopositive small capillaries (arrow in a), large capillaries (arrow in b), ED1 positive macrophages (c), OX42 positive activated microglia (d) and neutrophils (arrows in e). The cortex of the rosiglitazone treated (2 mg/kg at 2 h reperfusion) mice subjected to transient MCAO showed significantly lower number of ICAM1 immunopositive capillaries (f and g), ED1 positive macrophages (h), OX42 positive activated microglia (i) and neutrophils (j). The sham-operated rats showed no ICAM1 immunostaining (k) or ED1 positive cells (l) or OX42 positive cells (m) or neutrophils (n). Arrowheads in N indicate neurons. The histogram on the lower left (o) shows the cells number in the vehicle and rosiglitazone treated ischemic groups. The images shown in this figure were from the ischemic core within the ipsilateral cortex. The scale bar in K, L, M and N is 100 µm. Rosiglitazone treatment also significantly prevented the post-ischemic mRNA expression of the adhesion molecules ICAM1, E-selectin and P-selectin which showed significant up-regulation in the vehicle group (p). The bars represent mean ± SD of n = 5 rats per group for the immunostaining studies (o) and n = 4 rats per group for the gene expression studies (p). a< 0.05 compared with the sham group and b< 0.05 compared with the MCAO/vehicle group by one-way anova followed by Tukey–Kramer multiple comparisons post-test.

Download figure to PowerPoint

Rosiglitazone prevented post-ischemic induction of COX2 and iNOS

Induction of the proinflammatory enzymes COX2 and iNOS leads to formation of the neurotoxic oxygen and nitrogen radicals after focal ischemia. In the ipsilateral cortex of the MCAO/vehicle group, the mRNA levels COX2 and iNOS increased by 5.3 and 4.2-fold, respectively over sham (Fig. 8a). In the MCAO/rosiglitazone group, mRNA induction of both COX2 and iNOS was significantly curtailed compared with MCAO/vehicle group (by 84% and 81%, respectively; < 0.05; n = 4/group) (Fig. 8a). In the ipsilateral cortex of the MCAO/vehicle group many neurons showed intense COX2 immunostaining (Fig. 8b) and many astroglia-like cells showed iNOS immunostaining (Fig. 8c). Post-ischemic COX2 immunostaining and iNOS immunostaining (Fig. 8d and e) was significantly curtailed in the MCAO/rosiglitazone group (2 mg/kg at 2 h reperfusion) compared with MCAO/vehicle group (Fig. 8b and c). In the cortex of the sham-operated rats, no cells were observed to be immunostained for COX2 (Fig. 8j) or iNOS (Fig. 8k).

image

Figure 8.  Effect of rosiglitazone on inflammatory and oxidative mechanisms. Real-time PCR analysis showed significantly increased mRNA expression of the pro-inflammatory genes COX2 and iNOS, anti-inflammatory gene SOCS3 and anti-oxidant enzymes catalase and Cu/Zn-SOD in the ipsilateral cortex of the middle cerebral artery occlusion (MCAO)/vehicle group at 1 day of reperfusion compared with sham (a). The MCAO/rosiglitazone group (2 mg/kg at 2 h reperfusion) showed significantly curtailed expression of COX2 and iNOS mRNAs and significantly enhanced expression of catalase, Cu/Zn-SOD and SOCS3 mRNAs compared with MCAO/vehicle group (a). The bars represent mean ± SD (n = 4 rats per group). a< 0.05 compared with the sham group and b< 0.05 compared with the MCAO/vehicle group by one-way anova followed by Tukey–Kramer multiple comparisons post-test. In the MCAO/vehicle group, several neurons showed COX2 immunostaining (b) and astrocytes showed iNOS immunostaining (c). MCAO/rosiglitazone group showed significantly less COX2 immunostaining (d) and absence of iNOS immunostaining (e). The MCAO/vehicle group showed faint catalase immunostaining (f) and the MCAO/rosiglitazone group showed several intensely catalase positive neurons in the ischemic penumbra (g). The MCAO/vehicle group also showed several pSTAT3 immunopositive cells (h) which were significantly less in the MCAO/rosiglitazone group (i). No immunostaining of COX2 (j), iNOS (k), catalase (l) or pSTAT3 (m) was observed in the sham-operated rats cortex. The COX2, iNOS and catalase immunostained images were from the peri-infarct area and the pSTAT3 images were from the ischemic core of the ipsilateral cortex. The scale bar is 100 μm in all cases.

Download figure to PowerPoint

Rosiglitazone induced the expression of the antioxidant enzymes

The PPARγ down-stream enzymes catalase and Cu/Zn-SOD can prevent post-ischemic oxidative neuronal damage. Following transient MCAO and 24 h reperfusion, both catalase and Cu/Zn-SOD mRNA expression increased significantly in the ipsilateral cortex of the vehicle and rosiglitazone (2 mg/kg at 1 min and 2 h reperfusion) treated rats over sham controls. However, the fold increases were significantly higher for both the transcripts in the MCAO/rosiglitazone group over the MCAO/vehicle group (by 113–141%, < 0.05; n = 4/group) (Fig. 8a). Furthermore, catalase immunostained neurons in the peri-infarct cortex were higher in number and more intensely stained in the MCAO/rosiglitazone group (Fig. 8g) than the MCAO/vehicle group (Fig. 8f). Sham-operated rat cortex showed no catalase immunostaining (Fig. 8l).

Rosiglitazone increased SOCS3 expression and prevented STAT3 phosphorylation

The inflammatory actions of the cytokine IL6 are amplified by Janus kinase (JAK)-STAT signaling pathways. Our recent studies showed that following transient focal ischemia, increased STAT3 phosphorylation mediates the neuronal damage and induction of SOCS3 prevents JAK and STAT phosphorylation and thereby acts as a negative modulator of cytokine signaling (Vemuganti et al. 2002; Satriotomo et al. 2006). We currently observed that the post-ischemic SOCS3 mRNA expression was significantly higher (by 98%; < 0.05; n = 4/group) in the MCAO/rosiglitazone group over MCAO/vehicle group (Fig. 8a). Furthermore, the pSTAT3 phosphorylation was significantly lower in the MCAO/rosiglitazone group (Fig. 8i) compared with MCAO/vehicle group (Fig. 8h). The sham-operated rat cortex showed no pSTAT3 immunostaining (Fig. 8m).

Rosiglitazone prevented post-ischemic inflammatory gene expression in SHR rats

Real-time PCR analysis showed increased mRNA levels of the pro-inflammatory cytokines IL6 and IL1β (by 15.4-fold and 11.7-fold, respectively; < 0.05) and chemokines MIP1α and MCP1 (by 11.6-fold and 15.6-fold, respectively; < 0.05) in the ipsilateral cortex of MCAO/vehicle group compared with sham (Fig. 9). Rosiglitazone (2 mg/kg at 1 min and 2 h reperfusion) significantly curtailed the post-ischemic induction of both the cytokines (IL6 by 81% and IL1β by 87%; < 0.05; n = 4/group) and the chemokines (MIP1α by 71% and MCP1 by 66%; < 0.05; n = 4/group) compared with vehicle treatment (Fig. 9). The MCAO/vehicle group showed increased expression of the pro-inflammatory transcription factors Egr1, C/EBPβ and NF-κB (by 4.6-fold, 8.4-fold and 4.3-fold over sham, respectively; < 0.05) (Fig. 9) and the MCAO/rosiglitazone group showed significantly lower expression of these transcription factors (Egr1 by 67%, C/EBPβ by 79% and NF-κB by 72%, < 0.05; n = 4/group) compared with the MCAO/vehicle group (Fig. 9).

image

Figure 9.  Rosiglitazone treatment prevented inflammatory gene expression. In the middle cerebral artery occlusion (MCAO)/vehicle group of rats, there was a significant increase in the mRNA expression of cytokines, chemokines and pro-inflammatory transcription factors estimated by real-time PCR at 1 day of reperfusion compared with sham. The MCAO/rosiglitazone group showed a significantly curtailed expression of IL6, IL1β, MIP1α, MCP1, Egr1, C/EBPβ and NF-κB compared with the MCAO/vehicle group. The bars represent mean ± SD (n = 4 rats per group). a< 0.05 compared with the sham group and b< 0.05 compared with the MCAO/vehicle group by one-way anova followed by Tukey–Kramer multiple comparisons post-test.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In brief, the present study showed that TZDs induce neuroprotection after stroke (a) in different strains of rats and mice, (b) in the setting of hypertension as well as type-2 diabetes, (c) even when administered at 3 h of reperfusion, and (d) by acting through PPARγ. Furthermore, TZD treatment had no effect on the rCBF during ischemia and reperfusion, and the neuroprotection induced by TZDs was not enhanced by exogenous supplementation of RXR agonist 9-cis-RA. The results also showed that the TZD-induced neuroprotection after stroke might be mediated by preventing macrophage and neutrophil infiltration and curtailing inflammatory gene expression. TZD treatment induced the expression of anti-oxidant enzymes catalase and Cu/Zn-SOD which might prevent the oxidative damage in the post-ischemic brain. Furthermore, TZD-induced neuroprotection after stroke might also be mediated by stimulation the cytokine signaling negative modulator SOCS3 expression and prevention of STAT3 phosphorylation that dampens IL6 signaling.

Peroxisome proliferator-activated receptor-γ ligands were shown to induce beneficial effects in many neuroinflammatory disorders. In particular, pioglitazone was shown to relieve the symptoms in animal models of Parkinson’s disease and amyotrophic lateral sclerosis (Breidert et al. 2002; Schutz et al. 2005). PPARγ natural ligand 15-d-PGJ2 administration was reported to decrease the disease severity of experimental autoimmune encephalopathy (Diab et al. 2004). Rosiglitazone treatment was shown to preserve cognition in early Alzheimer’s disease patients (Watson et al. 2005). A positive correlation of increased plasma 15-d-PGJ2 levels to neurological outcome was reported in 552 stroke patients indicating an active role of PPARγ in post-ischemic brain damage (Blanco et al. 2005).

Recent studies showed that TZDs (troglitazone, pioglitazone and rosiglitazone) decrease infarct volume following transient focal ischemia in adult normotensive and normoglycemic rodents (Shimazu et al. 2005; Sundararajan et al. 2005; Luo et al. 2006; Pereira et al. 2006; Zhao et al. 2006a). In addition, Zhao et al. (2006a) showed that PPARγ natural agonist 15-d-PGJ2 decreases neurological deficits following intracerebral hemorrhage (ICH) in adult rats. The present study extends these studies by showing that TZDs prevent post-ischemic brain damage in hypertensive and type-2 diabetic rodents in addition to normotensive/normoglycemic rodents.

In our studies, both pre-treatment as well as post-treatment with pioglitazone and rosiglitazone induced neuroprotection after focal ischemia, and these two TZDs have comparable efficacy. Rosiglitazone-induced neuroprotection was observed when given as late as 3 h of reperfusion which is very beneficial as a stroke therapy. A previous study using in vitro cultured microglia indicated that TZDs can act directly without activating PPARγ under certain conditions (Park et al. 2003). However, the present in vivo studies observed that pre-treating animals with a PPARγ antagonist prevents rosiglitazone-induced neuroprotection after transient focal ischemia. This is supported by the observation that 15-d-PGJ2 induces a rapid activation of PPARγ that correlates with prevention of NF-κB activation within an hour after ICH (Zhao et al. 2006a).

We currently observed that the effective neuroprotective dose of rosiglitazone and pioglitazone after focal ischemia is similar. This is puzzling as the Kd of rosiglitazone and pioglitazone to PPARγ are c. 45 and c. 500 nmol/L, respectively. Although pioglitazone binds to PPARγ with 10-fold lower affinity than rosiglitazone, pioglitazone has been shown to cross the BBB and accumulate in brain more efficiently (Maeshiba et al. 1997). Therefore, higher doses of rosiglitazone may be needed to achieve sufficient levels in brain to provide neuroprotection. Furthermore, TZDs including pioglitazone and rosiglitazone exert PPARγ independent as well as PPARγ dependent effects, and in most cases pioglitazone was shown to be equally or more effective than rosiglitazone (Feinstein et al. 2005). TZDs can increase astrocyte glucose uptake and lactate production in a PPARγ independent manner (Dello Russo et al. 2003), effects which could contribute to the neuroprotective actions observed.

Similar to PPARγ, RXR is also a ligand activated transcription factor of nuclear hormone receptor superfamily. RXR contains a highly conserved central DNA-binding domain and a less conserved ligand-binding domain and hence serves as a common heterodimeric, permissive partner for many nuclear receptors including PPAR. When a ligand binds to PPAR, it dimerizes with RXR to form a complex that binds to PPREs on DNA to initiate or trans-repress the transcription of target genes (Escher and Wahli 2000). Previous studies showed that supplementing RXR ligand 9-cis-RA increases the beneficial potential of TZDs in experimental autoimmune encephalomyelitis (EAE) (Diab et al. 2004). However, the present study observed no added benefit of treating post-ischemic rodents with rosiglitazone and 9-cis-RA together over rosiglitazone alone. This indicates that endogenous RXR ligands might be depleted in chronic neuroinflammatory conditions like EAE, but not in acute neuroinflammatory conditions like stroke.

Prevention of inflammation is one of the major mechanisms thought to underlie the TZD neuroprotection after focal ischemia (Sundararajan et al. 2005; Luo et al. 2006). In support, we observed a significant dampening of post-ischemic induction of many pro-inflammatory genes including cytokines (IL1β and IL6) and chemokines (MIP1α and MCP1) in rosiglitazone-treated animals. Following focal ischemia, increased expression of ICAM1 and selectins induces extravasation of blood cells like neutrophils and macrophages into brain parenchyma. These cells release a host of neurotoxic substances including free radicals and lipid peroxidation products to precipitate secondary neuronal damage in the post-ischemic brain. Prevention of ICAM1 expression and decreased infiltration of these blood-borne cells into brain is neuroprotective after focal ischemia. We currently observed a significant prevention of post-ischemic induction of ICAM1, E-selectin and P-selectin expression followed by curtailed infiltration of macrophages and neutrophils in rosiglitazone-treated rats. We also observed that diabetic mice fed for 3 weeks with rosiglitazone-fortified chow showed significantly less induction of IL6, IL1β, MCP1, ICAM1 and P-selectin following transient MCAO compared with control chow fed controls.

By virtue of being upstream to inflammatory genes, transcription factors like NF-κB, Egr1, C/EBPβ and STAT3 contribute to the post-ischemic inflammatory gene expression and thus brain damage. Focal ischemia stimulates a rapid and sustained expression of these transcription factors (Stephenson et al. 2000; Vemuganti et al. 2002; Kapadia et al. 2006; Satriotomo et al. 2006). Therapeutic paradigms that prevent NF-κB activation are known to induce neuroprotection after focal ischemia (Stephenson et al. 2000; Williams et al. 2003). We currently observed that rosiglitazone treatment significantly curtailed post-ischemic NF-κB expression which might be one of the neuroprotective mechanisms of TZD activation. In support, Zhao et al. (2006a) suggested that 15-d-PGJ2 induces neuroprotection after ICH by preventing NF-κB activation.

Egr1 and C/EBPβ are the oxygen-sensitive transcription factors up-regulated after focal ischemia (Bowen et al. 2005; Kapadia et al. 2006). C/EBPβ over-expression was reported to promote inflammatory gene expression in cultured neurons (Cortes-Canteli et al. 2004). Furthermore, Egr1 induction was thought to be vital for the onset of inflammatory gene expression in peripheral organs (Yan et al. 2000). Okada et al. (2002) showed that PPARγ activation minimizes inflammation by preventing Egr1 induction following lung ischemia. Our recent studies showed curtailed inflammation and smaller infarcts in Egr1 and C/EBPβ knockout mice following transient MCAO suggesting a deleterious role for these transcription factors induced in the post-ischemic brain (Bowen et al. 2005; Kapadia et al. 2006). We currently observed that the mRNA expression of Egr1 and C/EBPβ was prevented by rosiglitazone and this might be an important mechanism of TZD neuroprotection after stroke.

The pro-inflammatory cytokine IL6 formed in excess during the acute phase after focal ischemia is known to promote inflammatory neuronal damage. IL6 binding to its receptors induces transphosphorylation of the receptor-associated JAKs and the down-stream STAT transcription factor family members. Phosphorylated STATs dimerize, translocate into nucleus and promote further cytokine gene expression to sustain inflammation. SOCS3 expressed down-stream to STAT acts as a feed-back inhibitor of JAK-STAT phosphorylation to curtail IL6 signal transduction. Our recent studies showed that focal ischemia induces STAT3 phosphorylation and its prevention results in significant neuroprotection (Vemuganti et al. 2002; Satriotomo et al. 2006). In interferon-γ stimulated cultured microglia, PPARγ ligands were shown to suppress JAK-STAT signaling by inducing SOCS3 expression (Park et al. 2003). Our studies show that rosiglitazone treatment increases post-ischemic SOCS3 mRNA expression and prevents the STAT3 phosphorylation which might be an important mechanism underlying the TZD-induced neuroprotection.

Although normal levels of STAT3 activation are essential for many cellular functions, its excess phosphorylation after focal ischemia is neurotoxic (Satriotomo et al. 2006). Increased levels of IL-6 might be responsible for the over-activation of STAT3 which promotes further cytokine gene expression following focal ischemia. The pSTAT3 also stimulates SOCS3 expression which dampens the cytokine signal tranduction by preventing STAT3 phosphorylation and knockdown of post-ischemic SOCS3 induction exacerbates ischemic neuronal damage (Vemuganti et al. 2002). However, the endogenous SOCS3 induction might only partially prevent the IL-6-JAK-STAT inflammatory signaling. Recent preliminary studies from our laboratory showed that focal ischemia following adenoviral-mediated over-expression of SOCS3 prevented the STAT3 phosphorylation and infarction (Vemuganti et al. 2005). Furthermore, recent studies showed that two other families of proteins viz., the SH2 containing phosphotases and the protein inhibitors of activated STATs (PIAS) act co-operatively with the SCOS proteins to efficiently inhibit cytokine signal transduction. In addition, protein modifiers like ubiquitin and small ubiquitin-like modifier SUMO play a significant role in the function of SOCS and PIAS to repress cytokine signaling (see the review by Rakesh and Agarwal 2005).

Oxygen free radicals formed in excess after focal ischemia are major mediators of ischemic neuronal death. Induction of the anti-oxidant enzymes catalase and SOD serves as the endogenous defense mechanism that prevents oxidative neuronal damage (Saito et al. 2005). We currently observed that treating ischemic rats with rosiglitazone significantly increased the catalase and Cu/Zn-SOD mRNA levels, and catalase protein expression in neurons in the peri-infarct area. Our observations are concurrent to studies that showed increased catalase expression in rodents treated with 15-d-PGJ2 following ICH (Zhao et al. 2006a) and increased Cu/Zn-SOD activity in pioglitazone-treated rodents subjected to transient MCAO (Shimazu et al. 2005). Induction of the antioxidant enzymes seems to be a contributing mechanism of PPARγ neuroprotection. Activation of pro-inflammatory enzymes iNOS and COX2 promotes progression of neuronal damage by forming nitric oxide and super oxide radicals, and their inhibition is neuroprotective after stroke (Graham and Hickey 2003; Iadecola and Gorelick 2005; Rodrigo et al. 2005). Following transient MCAO, the infarct grows up to 3 days of reperfusion and some of the surviving neurons in the peri-infarct area (infarct boundary zone) positive for COX2 at day 1 die by day 3 if there was no therapeutic intervention. We presently show that rosiglitazone treatment decreases iNOS and COX2 expression in the peri-infarct neurons leading to survival of the infarct boundary zone resulting in smaller infarcts. Previous studies also showed that troglitazone and pioglitazone prevents post-ischemic iNOS and COX2 induction (Sundararajan et al. 2005; Zhao et al. 2006b). Thus, TZDs induce neuroprotection by preventing formation (by inhibiting iNOS and COX2) and promoting disposal (by inducing catalase and SOD) of free radicals.

In conclusion, our studies show that the neuroprotective effects of TZDs after focal ischemia are (a) independent of species and strain, (b) can be seen in hypertensive as well as type-2 diabetic subjects, (c) do not need RXR ligand supplementation and (d) are not due to vascular changes leading to altered CBF. Mechanistically, TZDs induce neuroprotection by preventing inflammation and minimizing oxidative stress. As activation of other PPAR isoforms (PPARα and PPARβ/δ) also induces neuroprotection following stroke (Deplanque et al. 2003; Arsenijevic et al. 2006), testing dual agonists that can activate multiple PPAR isoforms might pave the way to new drug development for CNS inflammatory diseases.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Funded by NIH RO1 NS044173 (R. Vemuganti), NIH RO1 NS049448 (R. Vemuganti) and American Heart Association Grant-in-Aid 0350164 N (R. Vemuganti) and NIH RO1 NS044945 (D.L. Feinstein). The authors wish to thank Dr A. Hazell for critical comments.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Amenta F., Di Tullio M. A. and Tomassoni D. (2003) Arterial hypertension and brain damage-evidence from animal models. Clin. Exp. Hypertens. 25, 359380.
  • Argmann C. A., Cock T. A. and Auwerx J. (2005) Peroxisome proliferator-activated receptor gamma: the more the merrier? Eur. J. Clin. Invest. 35, 8292.
  • Arsenijevic D., De Bilbao F., Plamondon J., Paradis E., Vallet P., Richard D., Langhans W. and Giannakopoulos P. (2006) Increased infarct size and lack of hyperphagic response after focal cerebral ischemia in peroxisome proliferator-activated receptor beta-deficient mice. J. Cereb. Blood Flow Metab. 26, 433445.
  • Bergeron M., Yu A. Y., Solway K. E., Semenza G. L. and Sharp F. R. (1999) Induction of hypoxia-inducible factor-1 (HIF-1) and its target genes following focal ischemia in rat brain. Eur. J. Neurosci. 11, 41594170.
  • Blanco M., Moro M. A., Davalos A., Leira R., Castellanos M., Serena J., Vivancos J., Rodriguez-Yanez M., Lizasoain I. and Castillo J. (2005) Increased plasma levels of 15-deoxyDelta prostaglandin J2 are associated with good outcome in acute atherothrombotic ischemic stroke. Stroke 36, 11891194.
  • Blanquart C., Barbier O., Fruchart J. C., Staels B. and Glineur C. (2003) Peroxisome proliferator-activated receptors: regulation of transcriptional activities and roles in inflammation. J. Steroid Biochem. Mol. Biol. 85, 267273.
  • Bowen K. K., Tureyen K., Kalluri H., Svaren J. and Vemuganti R. (2005) Egr1 null mice show curtailed inflammatory gene expression and significant neuroprotection following transient focal ischemia. Soc. Neurosci. Abst. 31, 97.1.
  • Breidert T., Callebert J., Heneka M. T., Landreth G., Launay J. M. and Hirsch E. C. (2002) Protective action of the peroxisome proliferator-activated receptor-γ agonist pioglitazone in a mouse model of Parkinson’s disease. J. Neurochem. 82, 615624.
  • Cheng-Lai A. and Levine A. (2000) Rosiglitazone: an agent from the thiazolidinedione class for the treatment of type 2 diabetes. Heart Dis. 2, 326333.
  • Cho S., Park E. M., Kim Y., Liu N., Gal J., Volpe B. T. and Joh T. H. (2001) Early c-Fos induction after cerebral ischemia: A possible neuroprotective role. J. Cereb. Blood Flow Metab. 21, 550556.
  • Cortes-Canteli M., Wagner M., Ansorge W. and Perez-Castillo A. (2004) Microarray analysis supports a role for CCAAT/enhancer-binding protein-beta in brain injury. J. Biol. Chem. 279, 14 40914 417.
  • Dello Russo C., Gavrilyuk V., Weinberg G., Almeida A., Bolanos J. P., Palmer J., Pelligrino D., Galea E. and Feinstein D. L. (2003) Peroxisome proliferator-activated receptor gamma thiazolidinedione agonists increase glucose metabolism in astrocytes. J. Biol. Chem. 278, 58285836.
  • Deplanque D., Gele P., Petrault O. et al. (2003) Peroxisome proliferator-activated receptor-alpha activation as a mechanism of preventive neuroprotection induced by chronic fenofibrate treatment. J. Neurosci. 23, 62646271.
  • Dhodda V. K., Sailor K. A. and Vemuganti R. (2004) Putative endogenous mediators of preconditioning-induced ischemic tolerance in rat brain identified by genomic and proteomic analysis. J. Neurochem. 89, 7389.
  • Diab A., Hussain R. Z., Lovett-Racke A. E., Chavis J. A., Drew P. D. and Racke M. K. (2004) Ligands of the peroxisome proliferators-activated receptor-gamma and retinoid X receptor exert additive anti-inflammatory effects on experimental autoimmune encephalopathy. J. Neuroimmunol. 148, 116126.
  • Dogan A., Baskaya M. K., Rao V. L., Rao A. M. and Dempsey R. J. (1998) Intraluminal suture occlusion of the middle cerebral artery in Spontaneously Hypertensive rats. Neurol. Res. 20, 265270.
  • Escher P. and Wahli W. (2000) Peroxisome proliferators-activated receptors: insights into multiple cellular functions. Mutation Res. 448, 121138.
  • Feinstein D. L., Spagnolo A., Akar C., Weinberg G., Murphy P., Gavrilyuk V. and Dello Russo C. (2005) Receptor-independent actions of PPAR thiazolidinedione agonists: is mitochondrial function the key? Biochem. Pharmacol. 70, 177188.
  • Graham S. H and Hickey R. W (2003) Cyclooxygenases in central nervous system diseases: a special role for cyclooxygenase 2 in neuronal cell death. Arch. Neurol. 60, 628630.
  • Hu B. R., Liu C. L. and Park D. J. (2000) Alteration of MAP kinase pathways after transient forebrain ischemia. J. Cereb. Blood Flow Metab. 20, 10891095.
  • Iadecola C. and Gorelick P. B. (2005) The Janus face of cyclooxygenase-2 in ischemic stroke: shifting toward downstream targets. Stroke 36, 182185.
  • Iadecola C., Salkowski C. A., Zhang F., Aber T., Nagayama M., Vogel S. N. and Ross M. E. (1999) The transcription factor interferon regulatory factor-1 is expressed after cerebral ischemia and contributes to ischemic brain injury. J. Exp. Med. 189, 719727.
  • Ichikawa H., Naito Y., Takagi T., Tomatsuri N., Yoshida N. and Yoshikawa T. (2002) A specific peroxisome proliferator-activated receptor-gamma (PPAR-γ) ligand, pioglitazone, ameliorates gastric mucosal damage induced by ischemia and reperfusion in rats. Redox Rep. 7, 343346.
  • Kapadia R., Tureyen K., Bowen K., Kalluri H., Johnson P. F. and Vemuganti R. (2006) Decreased brain damage and curtailed inflammation in transcription factor CCAAT/enhancer binding protein beta knockout mice following transient focal cerebral ischemia. J. Neurochem. (in press).
  • Kiec-Wilk B., Dembinska-Kiec A., Olszanecka A., Bodzioch M. and Kawecka-Jaszcz K. (2005) The selected pathophysiological aspects of PPARs activation. J. Physiol. Pharmacol. 56, 149162.
  • Lu A., Tang Y., Ran R., Clark J. F., Aronow B. J. and Sharp F. R. (2003) Genomics of the periinfarction cortex after focal cerebral ischemia. J. Cereb. Blood Flow Metab. 23, 786810.
  • Luo Y., Yin W., Signore A. P., Zhang F., Hong Z., Wang S., Graham S. H. and Chen J. (2006) Neuroprotection against focal ischemic brain injury by the peroxisome proliferators-activated receptor-γ agonist rosiglitazone. J. Neurochem. 97, 435448.
  • MacManus J. P., Graber T., Luebbert C., Preston E., Rasquinha I., Smith B. and Webster J. (2004) Translation-state analysis of gene expression in mouse brain after focal ischemia. J. Cereb. Blood Flow Metab. 24, 657667.
  • Maeda K., Hata R., Gillardon F. and Hossman K. A. (2001) Aggravation of brain injury after focal cerebral ischemia in p53-deficient mice. Mol. Brain Res. 88, 5461.
  • Maeshiba Y., Kiyota Y., Yamashita K., Yoshimura Y., Motohashi M. and Tanayama S. (1997) Disposition of the new antidiabetic agent pioglitazone in rats, dogs, and monkeys. Arzneimittelforschung 47, 2935.
  • McCulloch J. (1996) Neuroprotective drug development in stroke: blood pressure and its impact. J. Hypertens. 14, S1318.
  • Naylor M., Bowen K. K., Sailor K. A., Dempsey R. J. and Vemuganti R. (2005) Preconditioning-induced ischemic tolerance stimulates growth factor expression and neurogenesis in adult rat hippocampus. Neurochem. Int. 47, 565572.
  • Okada M., Yan S. F. and Pinsky D. J. (2002) Peroxisome proliferator-activated receptor-gamma (PPAR-gamma) activation suppresses ischemic induction of Egr-1 and its inflammatory gene targets. FASEB J. 16, 18611868.
  • Park E. J., Park S. Y., Joe E. H. and Jou I. (2003) 15d-PGJ2 and rosiglitazone suppress Janus kinase-STAT inflammatory signaling through induction of suppressor of cytokine signaling 1 (SOCS1) and SOCS3 in glia. J. Biol. Chem. 278, 14 74714 752.
  • Pereira M. P., Hurtado O., Cardenas A. et al. (2006) Rosiglitazone and 15-deoxy-Delta12,14-prostagliandin J2 cause potent neuroprotection after experimental stroke through noncompletely overlapping mechanisms. J. Cereb. Blood Flow Metab. 26, 218229.
  • Rakesh K. and Agarwal D. K. (2005) Controlling cytokine signaling by constitutive inhibitors. Biochem. Pharmacol. 70, 649657.
  • Rickhag M., Wieloch T., Gido G. et al. (2006) Comprehensive regional and temporal gene expression profiling of the rat brain during the first 24 h after experimental stroke identifies dynamic ischemia-induced gene expression patterns, and reveals a biphasic activation of genes in surviving tissue. J. Neurochem. 96, 1429.
  • Rodrigo J., Fernandez A. P., Serrano J., Peinado M. A. and Martinez A. (2005) The role of free radicals in cerebral hypoxia and ischemia. Free Radic. Biol. Med. 39, 2650.
  • Saito A., Maier C. M., Narasimhan P. et al. (2005) Oxidative stress and neuronal death/survival signaling in cerebral ischemia. Mol. Neurobiol. 31, 105116.
  • Satriotomo I., Bowen K. and Vemuganti R. (2006) JAK2 and STAT3 activation contributes to neuronal damage following transient focal cerebral ischemia. J. Neurochem. 98, 13531368.
  • Schutz B., Reimann J., Dumitrescu-Ozimek L., Kappes-Horn K., Landreth G. E., Schurmann B., Zimmer A. and Heneka M. T. (2005) The oral antidiabetic pioglitazone protects from neurodegeneration and amyotrophic lateral sclerosis-like symptoms in superoxide dismutase-G93A transgenic mice. J. Neurosci. 25, 78057812.
  • Shimazu T., Inoue I., Araki N., Asano Y., Sawada M., Furuya D., Nagoya H. and Greenberg J. H. (2005) A peroxisome proliferators-activated receptor-gamma agonist reduces infarct size in transient but not in permanent ischemia. Stroke 36, 353359.
  • Stephenson D., Yin T., Smalstig E. B., Hsu M. A., Panetta J., Little S. and Clemens J. (2000) Transcription factor nuclear factor-kappa B is activated in neurons after focal cerebral ischemia. J. Cereb. Blood Flow Metab. 20, 592603.
  • Sundararajan S., Gamboa J. L., Victor A. N., Wanderi E. W., Lust W. D. and Landreth G. E. (2005) Peroxisome proliferation-activated receptor-γ ligands reduce inflammation and infarction size in transient focal ischemia. Neuroscience 130, 685696.
  • Swanson R. A., Morton M. T., Tsao-Wu G., Savalos R. A., Davidson C. and Sharp F. R. (1990) A semiautomated method for measuring brain infarct volume. J. Cereb. Blood Flow Metab. 10, 290293.
  • Tanaka K., Nogawa S., Ito D., Suzuki S., Dembo T., Kosakai A. and Fukuchi Y. (2000) Activated phosphorylation of cyclic AMP response element binding protein is associated with preservation of striatal neurons after focal cerebral ischemia in the rat. Neuroscience 100, 345354.
  • Tolman K. G. and Chandramouli J. (2003) Hepatotoxicity of the thiazolidinediones. Clin. Liver Dis. 7, 369379.
  • Tureyen K., Vemuganti R., Sailor K., Bowen K. K. and Dempsey R. J. (2004) Transient focal cerebral ischemia-induced neurogenesis in the dentate gyrus of the adult mouse. J. Neurosurg. 101, 799805.
  • Vemuganti R. and Dempsey R. J (2005) Carotid atherosclerotic plaques from symptomatic stroke patients share the molecular fingerprints to develop in a neoplastic fashion: a microarray analysis study. Neuroscience 131, 359374.
  • Vemuganti R., Dogan A., Bowen K. K. and Dempsey R. J. (2001) Ornithine decarboxylase knockdown exacerbates transient focal cerebral ischemia-induced neuronal damage in rat brain. J. Cereb. Blood Flow Metab. 21, 945954.
  • Vemuganti R., Bowen K. K., Dhodda V. K., Song G., Franklin J. L., Gavva N. R. and Dempsey R. J. (2002) Gene expression analysis of spontaneously hypertensive rat cerebral cortex following transient focal cerebral ischemia. J. Neurochem. 83, 10721086.
  • Vemuganti R., Dempsey R. J. and Bowen K. K. (2004) Inhibition of ICAM-1 protein expression by antisense oligonucleotides is neuroprotective after transient middle cerebral artery occlusion in rat. Stroke 35, 179184.
  • Vemuganti R., Satriotomo I. and Bowen K. K. (2005) Role of STAT/SOCS signaling in controlling stroke-induced inflammation and brain damage. J. Neurochem. 94, 117.
  • Vemuganti R., Kalluri H., Bowen K. K., Yi J. H. and Hazell A. S. (2006) Gene expression changes in thalamus and inferior colliculus associated with inflammation, cellular stress, metabolism and structural damage in thiamine deficiency. Eur. J. Neurosci. 23, 11721188.
  • Watson G. S., Cholerton B. A., Reger M. A. et al. (2005) Preserved cognition in patients with early Alzheimer disease and amnesic mild cognitive impairment during treatment with rosiglitazone: a preliminary study. J. Geriatr. Psychiatry 13, 950958.
  • Wayman N. S., Hattori Y., McDonald M. C., Mota-Filipe H., Cuzzocrea S., Pisano B., Chatterjee P. K. and Thiemermann C. (2002) Ligands of the peroxisome proliferator-activated receptors (PPAR-γ and PPAR-α) reduce myocardial infarct size. FASEB J. 16, 10271040.
  • Williams A. J., Hale S. L., Moffett J. R., Dave J. R., Elliott P. J., Adams J. and Tortella F. C. (2003) Delayed treatment with MLN519 reduces infarction and associated neurologic deficit caused by focal ischemic brain injury in rats via anti-inflammatory mechanisms involving nuclear factor-kappaB activation, gliosis, and leukocyte infiltration. J. Cereb. Blood Flow Metab. 23, 7587.
  • Yan S. F., Fujita T., Lu J., Okada K., Shan Zou Y., Mackman N., Pinsky D. J. and Stern D. M. (2000) Egr-1, a master switch coordinating upregulation of divergent gene families underlying ischemic stress. Nature Med. 6, 13551361.
  • Yan Y. P., Sailor K. A., Vemuganti R. and Dempsey R. J. (2006) Insulin-like growth factor-1 is an endogenous mediator of focal ischemia-induced neural progenitor proliferation. Eur. J. Neurosci. 24, 4554.
  • Zea Longa E., Weinstein P. R., Carlson S. and Cummins R. (1989) Reversible middle cerebral artery occlusion without craniectomy in rat. Stroke 20, 8491.
  • Zhao X., Zhang Y., Strong R., Grotta J. C. and Aronowski J. (2006a) 15d-Prostaglandin J2 activates peroxisome proliferators-activated receptor-γ, promotes expression of catalase, and reduces inflammation, behavioral dysfunction, and neuronal loss after intracerebral hemorrhage in rats. J. Cereb. Blood Flow Metab. 26, 811820.
  • Zhao Y., Patzer A., Herdegen T., Gohlke P. and Culman J. (2006b) Activation of cerebral peroxisome proliferator-activated receptors gamma promotes neuroprotection by attenuation of neuronal cyclooxygenase-2 overexpression after focal cerebral ischemia in rats. FASEB J. 20, 11621175.
  • Zheng Z., Lee J. E. and Yenari M. A. (2003) Stroke: Molecular mechanisms and potential targets for treatment. Curr. Mol. Med. 3, 361372.