15d-Prostaglandin J 2 Protects Brain From Ischemia-Reperfusion Injury

【摘要】  Objective- Brain expresses abundant lipocalin-type prostaglandin (PG) D 2 (PGD 2 ) synthase but the role of PGD 2 and its metabolite, 15-deoxy- 12,14 PGJ 2 (15d-PGJ 2 ) in brain protection is unclear. The aim of this study is to assess the effect of 15d-PGJ 2 on neuroprotection.

Methods and Results- Adenoviral transfer of cyclooxygenase-1 (Adv-COX-1) was used to amplify the production of 15d-PGJ 2 in ischemic cortex in a rat focal infarction model. Cortical 15d-PGJ 2 in Adv-COX-1-treated rats was increased by 3-fold over control, which was correlated with reduced infarct volume and activated caspase 3, and increased peroxisome proliferator activated receptor- (PPAR ) and heme oxygenase-1 (HO-1). Intraventricular infusion of 15d-PGJ 2 resulted in reduction of infarct volume, which was abrogated by a PPAR inhibitor. Rosiglitazone infusion had a similar effect. 15d-PGJ 2 and rosiglitazone at low concentrations suppressed H 2 O 2 -induced rat or human neuronal apoptosis and necrosis and induced PPAR and HO-1 expression. The anti-apoptotic effect was abrogated by PPAR inhibition.

Conclusion- 15d-PGJ 2 suppressed ischemic brain infarction and neuronal apoptosis and necrosis in a PPAR dependent manner. 15d-PGJ 2 may play a role in controlling acute brain damage induced by ischemia-reperfusion.

Adv-COX-1 gene transfer increased 15d-PGJ 2 in ischemic brain accompanied by reduced infarct volume, activated caspase 3, and enhanced heme oxygenase-1 and peroxisome proliferator activated receptor (PPAR ) expression. 15d-PGJ 2 and rosiglitazone inhibited neuronal apoptosis and necrosis in a PPAR -dependent manner.

【关键词】  COX dPGJ PPAR apoptosis stroke

Introduction

Prostaglandin (PG) H synthase-1 (also known as cyclooxygenase-1 [COX-1]) is constitutively expressed in almost all mammalian cells. 1 It is a bifunctional enzyme with a cyclooxygenase activity that converts arachidonic acid to PG G 2 (PGG 2 ) and a peroxidase activity that converts PGG 2 to PGH 2. 2 PGH 2 is converted to diverse prostanoids by specific enzymes. COX-1 plays an important role in maintaining physiological homeostasis and protecting brain tissues from ischemia-reperfusion (I/R) injury. COX-1 deleted mice are highly susceptible to ischemic brain infarction, 3 whereas COX-1 overexpression protects brain from I/R damage, which is abrogated by a selective COX-1 inhibitor. 4 COX-1 overexpression in ischemic brain augments the production of PGI 2, PGD 2, and PGE 2, and suppresses leukotriene B 4 (LTB 4 ) and LTC 4. As LTB 4 and LTC 4 have been shown to be detrimental to brain tissue, whereas PGI 2 is protective, 5-7 COX-1 overexpression tilts the eicosanoid balance toward tissue protection. PGD 2 is elevated in COX-1 overexpressed brain tissues but its role in brain I/R injury is unclear. Brain is enriched in lipocalin-type PGD synthase (L-PGDS), which catalyzes the formation of abundant PGD 2. 8 The role of PGD 2 in I/R brain injury is unclear. As 15-deoxy- 12,14; PGJ 2 (15d-PGJ 2 ), a nonenzymatic product of PGD 2, was shown to possess anti-inflammatory properties through activation of peroxisome proliferator activated receptor- (PPAR ), 9-13 PGD 2 has been implicated in tissue protection. However, it has recently been argued that the tissue 15d-PGJ 2 level is too low to elicit an anti-inflammatory action in vivo, especially in vascular tissues. 14 In view of abundant expression of L-PGDS and PGD 2 in brain, we postulated that 15d-PGJ 2 contributes to cerebral protection. Our experimental findings show a considerable amount of 15d-PGJ 2 in ischemia brain, which was enhanced by adenoviral COX-1 gene transfer. 15d-PGJ 2 and rosiglitazone reduced brain infarct volume, inhibited brain and neuronal apoptosis, suppressed NF- B activation, and upregulated heme oxygenase-1 (HO-1) in a PPAR -dependent manner.

Methods

Stroke Model

The rat focal cerebral infarction model has been described previously. 4 In brief, male Long-Evans rats were anesthetized, right middle cerebral artery (MCA) was ligated reversibly with a 10-0 suture, and both common carotid arteries were occluded with aneurysm clips. At the indicated time point, the aneurysm clips and the suture were removed and blood flow in all 3 arteries was restored. The animals were kept in an air-ventilated incubator at 24.0±0.5°C for 24 hours and then euthanized under anesthesia. Brains were quickly removed and the ischemic or contralateral cerebral cortex was isolated and frozen. Infarct volume was measured by incubating coronally dissected brain slices with 2,3,5-triphenyltetrazolium chloride as previously described. 4 All procedures were performed in accordance with the Public Health Service Guide for the Care and Use of Laboratory Animals and approved by the Academia Sinica Animal Studies Committee.

Cell Culture and H 2 O 2 -Mediated Oxidative Stress

Rat primary cortical neuron cultures were prepared from 14- to 15-day-old fetus according to procedures previously described. 15,16 All the experiments were performed on cultured neurons after 12 to 14 days in vitro. More than 90% of cells stained positive for microtubal associated protein-2. Human BE(2)-C neuroblastoma cells (American Type Culture Collection) were grown to 70% confluence in a 1:1 mixture of DMEM and Ham?s F-12K medium in a humidified 5% CO 2 atmosphere. H 2 O 2, 15d-PGJ 2 (Cayman), bisphenol A diglyceryl ether (BADGE) (Fluka), rosiglitazone (Cayman), zVADfmk, and zDEVDfmk (Biovision) were added to serum-deprived BE(2)-C cells either alone or in various combinations for 12 hours. Extent of cytotoxicity was assessed by lactate dehydrogenase (LDH) and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assays according to manufacturer?s instructions (Roche).

Preparation of Replication-Defective Recombinant Adenoviral Vectors

Viral vectors were prepared as previously described. 17 We constructed in the replication-defective recombinant adenoviral (rAd) vector a human phosphoglycerate kinase (PGK) promoter to drive COX-1 (Adv-COX-1), green fluorescent protein (Adv-GFP), or PGK alone to serve as the control (Adv-PGK).

Intracerebral Ventricular Infusion of Adenoviral Constructs and 15d-PGJ 2

The procedure was performed as previously described. 4 Briefly, anesthetized rats were placed in a stereotaxic apparatus; 10 µL of artificial cerebrospinal fluid containing rAd at 10 8 plaque-forming units (pfu) or 10 µL of 15d-PGJ 2 (1 to 50 pg) were infused into the right lateral ventricle at a rate of 5 µL/min at the following coordinates: Anterior, 2.5 mm caudal to bregma; Right, 2.8 mm lateral to midline; and Ventral, 3.0 mm ventral to dural surface. Periodic confirmation of proper placement of the needle was performed with infusion of fast green. To delineate the distribution of the transgene expression, we infused Adv-GFP into the right lateral ventricle for 72 hours and GFP was visualized under microscopy. GFP was detected in the lining of ependymal cells and cells surrounding the right ventricle in all 8 coronal brain slices but not in the left ventricle (Figure I, available online at http://atvb.ahajournals.org).

Measurements of Brain Tissue PGD 2 and 15d-PGJ 2

Brain tissues for PGD 2 and 15d-PGJ 2 analysis were prepared as previously described. 4 Briefly, the ischemic cortex was homogenized in 1 mL ice-cold buffer and centrifuged at 55 000 g for 1 hour. Eicosanoids were extracted with a Sep-Pak C18 cartridge and analyzed by enzyme immunoassays using reagents from Cayman for PGD 2 and R&D Systems for 15d-PGJ 2.

Western Blot Analysis

Analysis of proteins in the cortex and BE(2)-C cells by Western blotting was performed as described previously, 4 using antibodies for COX-1 (Cayman, 1:1000), COX-2 (Cayman, 1:1000), PPAR (Santa Cruz, 1:500), HO-1 (ABR, 1:2000), GAPDH (BD Pharmingen, 1:10000), active caspase-3 (Cell signaling, 1:500), Poly (ADP-ribose) polymerase (PARP) (Cell signaling, 1:1000), and I B- (Santa Cruz, 1;1000). Protein bands were visualized by an enhanced chemiluminescence system (Pierce).

Nuclear Extract Preparation and Electrophoretic Mobility Shift Assay

Nuclear extracts of brain were prepared and electrophoretic mobility shift assay (EMSA) was performed as previously described. 18 Cerebral cortex was gently homogenized and centrifuged at 3300 g for 15 minutes. Nuclei pellets were resuspended and centrifuged at 25 000 g for 30 minutes at 4°C and the supernatant was dialyzed; 20 µg of nuclear extracts were incubated with 5 ng of a 32 P-labeled B-containing probe at room temperature for 30 minutes. The mixture was run on a 5% nondenaturing polyacrylamide gel. For supershift experiments, antibodies against p65 or p50 (Santa Cruz) were added before adding labeled probes. Gels were dried under vacuum and exposed to Kodak X-Omat/XB-1 films. The radioactive bands were quantified by densitometry.

Analysis of Apoptosis by Flow Cytometry

Cells were fixed in ice-cold 70% ethanol for 30 minutes at 4°C. After centrifugation, cells were resuspended and incubated in phosphate-buffered saline containing 1 mg/mL RNase A and 10 µg/mL propidium iodide (PI) at room temperature for 30 minutes and analyzed in a fluorescence-activated-cell sorter (FACS) caliber flow cytometer (BD Bioscience). Percentages of cells with hypodiploid DNA (sub-G 0 /G 1 ) were measured.

Statistical Analysis

Analysis of variance (ANOVA) was used to compare the temporal expression of proteins, infarct volumes, and eicosanoid levels. The level of significance for differences between groups was further analyzed with post-hoc Fisher?s protected t tests by GB-STAT 5.0.4 (Dynamic Microsystem, Inc, Silver Springs, Md). P <0.05 was considered significant.

Results

Adenoviral COX-1 Gene Transfer Increased 15d-PGJ 2 in Ischemic Brain

Adv-COX-1 infusion 72 hours before a 50-minute MCA occlusion resulted in increased COX-1 proteins, PGD 2, and 15d-PGJ 2 levels accompanied by a reduction in infarct volume ( Figure 1 a). The extent of 15d-PGJ 2 increase (&3-fold) correlated with that of COX-1 increase (&3-fold) ( Figure 1 b). COX-2 was markedly suppressed whereas PPAR and HO-1 were elevated by Adv-COX-1 treatment ( Figure 1 b). EMSA analysis reveals increased p50/p65 NF- B DNA binding in ischemic cortex as previously reported, 19 which was abrogated by Adv-COX-1 (Figure II, available online at http://atvb.ahajournals.org).

Figure 1. Effects of COX-1 gene transfer. Adv-COX-1 was infused 72 hours before a 50-minute MCA occlusion. a, Reduction of infarct volume (A) accompanied by increased PGD 2 (B) and 15d-PGJ 2 (C). * P <0.05 and ** P <0.01. b, Representative Western blots of 5 experiments. COX-1, HO-1, and PPAR were significantly increased, whereas COX-2 was reduced.

15d-PGJ 2 and Rosiglitazone Reduced Infarct Volume

To confirm the protective effect of 15d-PGJ 2, we infused 15d-PGJ 2 24 hours before a 50-minute MCA occlusion. 15d-PGJ 2 at 1 pg 50% (Figure III, available online at http://atvb.ahajournals.org). However, when 15d-PGJ 2 was infused immediately after the 50 minutes of ischemia, it failed to reduce infarct volume significantly even when the dose was increased to 50 pg (Figure III). 15d-PGJ 2 infusion immediately after ischemia reduced the infarct size in right cortex when the MCA occlusion was shortened to 30 minutes (Figure IV, available online at http://atvb.ahajournals.org) in a concentration-dependent manner ( Figure 2 a). It remained effective when administered 2 hours after ischemia but was no longer effective at 3 hours after ischemia ( Figure 2 a). To determine whether the anti-infarct action of 15d-PGJ 2 is mediated via PPAR, we infused GW9662, a selective inhibitor of PPAR. GW9662 completely abrogated the protective effect of 15d-PGJ 2 ( Figure 2 b). Infusion of rosiglitazone (50 ng) immediately after the 30-minute ischemia reduced the 80% and remained effective when infused 2 hours, but not 3 hours, after ischemia. 15d-PGJ 2 and rosiglitazone reduced I B degradation to a similar extent (Figure V, available online at http://atvb.ahajournals.org). These results suggest that 15d-PGJ 2 reduced infarct volume via a PPAR -dependent pathway.

Figure 2. Attenuation of ischemia-induced brain infarct volume by 15d-PGJ 2. a, 15d-PGJ 2 was infused immediately (0) or 2 or 3 hours after a 30-minute transient occlusion (n=4 to 14). b, 15d-PGJ 2 (50 pg) or vehicle (0.1% DMSO) was infused with GW 9662 (165 ng) immediately after a 30-minute occlusion (n=5). ** P <0.01.

PPAR Ligands Upregulated PPAR Protein Levels

The PPAR protein level in ischemic brain was higher than control ( Figure 3 a), which was concentration-dependently enhanced by 15d-PGJ 2 ( Figure 3 a), and rosiglitazone (data not shown). 15d-PGJ 2 upregulated PPAR in normal brains in a time-dependent manner, with a 3-fold increase in the protein level 12 hours after infusion ( Figure 3 b).

Figure 3. Upregulation of PPAR by 15d-PGJ 2. a, 15d-PGJ 2 was infused immediately after a 30-minute ischemia. PPAR in ischemic cortex was analyzed by Western blotting. The densitometric comparison is shown at the lower panel (n=3). b, 15d-PGJ 2 (10 pg) was infused into normal rat brains (n=3) and at the indicated time points, PPAR was analyzed by Western blots. ** P <0.01.

Adv-COX-1, 15d-PGJ 2, and Rosiglitazone Inhibited Brain Tissue Caspase 3 Activation

To determine whether the tissue protective effects are mediated by blocking apoptosis, we measured activated caspase-3 in ischemic cortex treated with Adv-COX-1 or PPAR ligands. Activated caspase-3 was increased in ischemic cortex, which was abrogated by Ad-COX-1, 15d-PGJ 2 ( Figure 4a and 4 b), or rosiglitazone infusion (data not shown). 15d-PGJ 2 at 50 pg or rosiglitazone at 50 ng reduced activated caspase-3 to the basal level and Adv-COX-1 suppressed caspase 3 activation to a similar extent.

Figure 4. Suppression of caspase 3 activation. a, Adv was infused 72 hour before a 50-minute ischemia. b, 15d-PGJ 2 was infused immediately after a 30-minute ischemia. The upper panel shows a representative blot and the lower panel densitometric comparison (n=5 per group). ** P <0.01.

15d-PGJ 2 Suppressed Neuronal Apoptosis and Necrosis

The in vivo data reveal that 15d-PGJ 2 protected brain tissues from I/R-induced cell death. To ascertain its effect on protecting neuronal survival, we evaluated the effect of 15d-PGJ 2 on H 2 O 2 -induced neuronal cell death in human BE(2)-C and rat primary cortical neuron culture. Because H 2 O 2 has been reported to be a major mediator of I/R-induced neural cell apoptosis and necrosis, 20,21 we measured H 2 O 2 -induced LDH release and MTT reduction and several apoptotic changes. H 2 O 2 exerts a similar concentration-dependent increase in LDH release from cultured rat neurons (Figure VIa, available online at http://atvb.ahajournals.org) and BE(2)-C cells (Figure VIb) with a reciprocal MTT reduction. 15d-PGJ 2 at 100 to 200 nM effectively suppressed H 2 O 2 -induced LDH release from rat and human neurons ( Figure 5 a), which was abrogated by BADGE, a PPAR inhibitor ( Figure 5 a). Paradoxically, 15d-PGJ 2 5 µmol/L) induced LDH release from BE(2)-C cells ( Figure 5 b). Similar to 15d-PGJ 2, rosiglitazone 50%, which was abrogated by BADGE (Figure VIIa, available online at http://atvb.ahajournals.org), but paradoxically induced LDH release at higher concentrations ( 10 µmol/L) (Figure VIIb). Caspase inhibitors blocked H 2 O 2 -induced LDH release and restored the inhibitory action of 15d-PGJ 2 even in the presence of BADGE ( Figure 5 c).

Figure 5. Control of H 2 O 2 -induced LDH release. a, BE(2)-C were treated with H 2 O 2 in the presence or absence of 15d-PGJ 2 and BADGE. b, BE(2)-C cells were treated with 15d-PGJ 2 at increasing concentrations. c, BE(2)-C cells were treated with H 2 O 2 and a combination of compounds as indicated. Each bar in a to c refers to mean±SD (n=3). * P <0.05. ** P <0.01.

To confirm that 15d-PGJ 2 protects neurons from H 2 O 2 -induced apoptosis, we measured apoptotic cells by flow cytometry. 15d-PGJ 2 inhibited H 2 O 2 50%, which was abrogated by BADGE ( Figure 6 a). Moreover, 15d-PGJ 2 inhibited H 2 O 2 -induced caspase 3 activation and PARP cleavage in BE(2)-C cells, which were also abrogated by BADGE ( Figure 6 b).

Figure 6. Inhibition of neuronal apoptosis. a, BE(2)-C cells treated with H 2 O 2, 15d-PGJ 2, and/or BADGE for 12 hours. Sub-G 0 apoptotic cells were analyzed by flow cytometry. Upper panel shows percentages of Sub-G 0 in representative flow profiles and the lower panel mean±SD (n=3). b, BE(2)-C cells were treated with or without H 2 O 2, 15d-PGJ 2, and/or BADGE for 12 hours. Caspase 3 was determined as described in Methods. The upper panels show representative Western gels and the lower panels, mean±SD of densitometry of 3 experiments.

15d-PGJ 2 increased PPAR protein levels by &2-fold, which was not influenced by H 2 O 2 treatment (Figure VIIIa, available online at http://atvb.ahajournals.org). Because HO-1 is an important mediator of cell survival, we determined whether HO-1 level is altered by H 2 O 2 and 15d-PGJ 2 treatment. 15d-PGJ 2 increased HO-1 protein levels by &10-fold (Figure VIIIb). H 2 O 2 did not alter HO-1 level, nor did it interfere with the stimulatory action of 15d-PGJ 2 (Figure VIIIb). By contrast, the HO-1 stimulatory action of 15d-PGJ 2 was blocked by BADGE (Figure VIIIb).

Discussion

Results from this study indicate that a considerable quantity of 15d-PGJ 2 is generated in I/R-injured cortex, which is augmented by COX-1 overexpression. Elevation of 15d-PGJ 2 was correlated 80% reduction in activated caspase 3. 15d-PGJ 2 has a potent effect on controlling the expansion of the infarct size as administration of 15d-PGJ 2 intraventricularly at 1 pg was sufficient to induce &50% reduction in infarct volume. These results suggest that the endogenously generated 15d-PGJ 2 may be involved in suppressing I/R-induced infarct expansion. However, the causal relationship between the endogenous generation of 15d-PGJ 2 and reduction in infarct volume is not fully established in our study as we have not performed time-course experiments to demonstrate that 15d-PGJ 2 generation precedes the reduction in the infarct size. Work is in progress to characterize the relationship between endogenous 15d-PGJ 2, infarct size, and biochemical markers.

15d-PGJ 2 is highly effective in reducing infarct volume when administered before a 50-minute MCA occlusion but loses its activity when administered after occlusion. Its window of effectiveness extends to 2 hours, but not 3 hours, when administered after a shorter (30 minutes) and therefore a milder ischemic insult. These results suggest that 15d-PGJ 2 acts on early pathophysiological events after ischemic injury. It has been proposed that acute cerebral injury immediately after transient focal ischemia is attributed to energy failure and excitotoxicity that results in neuronal necrosis. 22 Acute ischemic injury also results in mitochondrial damage, which may lead to apoptosis. The extent of neuronal necrosis and apoptosis is influenced by the duration of MCA occlusion. A severe insult after a prolonged MCA occlusion causes predominantly neuronal necrosis, whereas a mild insult such as a 30-minute MCA occlusion incurs predominantly apoptosis. 23,24 Our results show that 15d-PGJ 2 is capable of suppressing neuronal necrosis and apoptosis, thereby restricting the infarct development after a 50-minute or 30-minute MCA occlusion, albeit with a different window of effectiveness. Evidence supporting the action of 15d-PGJ 2 includes: (1) pretreatment of rat or human neurons with 15d-PGJ 2 prevented H 2 O 2 -induced cytotoxicity as detected by LDH leakage and reduction in MTT staining; (2) 15d-PGJ 2 prevented H 2 O 2 -induced neuronal apoptosis; and (3) administration of 15d-PGJ 2 intraventricularly abrogated caspase 3 activation in the ischemic cortex. Because necrosis develops rapidly after a severe ischemic injury, 15d-PGJ 2 is ineffective in controlling infarct size unless it is administered before I/R injury. By contrast, neuronal apoptosis induced by shorter ischemia takes place not as rapidly and therefore is responsive to 15d-PGJ 2 inhibition even after tissue damage has occurred.

Several pieces of evidence support the requirement of PPAR activation for the protective action of 15d-PGJ 2. First, the effect of 15d-PGJ 2 on reducing infarct volume in vivo was abrogated by a PPAR inhibitor, GW9662. Second, the effect of 15d-PGJ 2 on protecting neuronal cytotoxicity and apoptosis was abrogated by another PPAR inhibitor, BADGE. Third, known PPAR ligands such as rosiglitazone inhibited I/R-induced brain infarction in vivo and H 2 O 2 -induced neuronal apoptosis and cytotoxicity in vitro. Activation of PPAR has been shown to upregulate HO-1 expression 25,26 and suppress the expression of an array of genes by blocking the transcriptional activity of transactivators such as NF- B. 12,13 Our data confirm that 15d-PGJ 2 upregulates HO-1 expression, inhibits NF- B activation, and suppresses COX-2 expression in ischemic cortex and cultured neurons by a PPAR -dependent pathway. HO-1 possesses tissue protective properties, whereas NF- B mediated genes, such as COX-2, 13,27 which aggravates tissue damage by producing pro-inflammatory prostanoids and reactive oxygen species. Upregulation of the protective HO-1 coupled with suppression of COX-2 and other NF- B-dependent genes via PPAR activation represents a major mechanism by which 15d-PGJ 2 and rosiglitazone protect against I/R-induced neuronal necrosis and apoptosis and thereby limit the expansion of the infarct size.

We observed that 15d-PGJ 2 and rosiglitazone exhibit a concentration-dependent paradoxical effect on cytotoxicity. 15d-PGJ 2 induces LDH release at 5 µmol/L but protects neurons from necrosis and apoptosis at 1 µmol/L. This observation may explain the conflicting data reported in the literature. 15d-PGJ 2 was reported to induce apoptosis of several cell types including neurons 28-32 and protect cerebellar granular cells from apoptosis. 33,34 A detailed review of those reports reveals that the paradoxical effects of 15d-PGJ 2 may be caused by use of different 15d-PGJ 2 concentrations. Studies using 15d-PGJ 2 10 µmol/L reported a pro-apoptotic effect, whereas those using concentrations 1 µmol/L were anti-apoptotic. The reason for this paradoxical action is unclear. Because rosiglitazone has a similar concentration-dependent paradoxical effect, it is tempting to speculate that PPAR is involved. Further studies are needed to resolve this puzzle.

Our results reveal that PPAR expression in rat brain tissues is increased by I/R and further enhanced by 15d-PGJ 2 and rosiglitazone. These results are consistent with an autoregulation of PPAR by its ligands. It is unclear how 15d-PGJ 2 upregulates PPAR. 15d-PGJ 2 may enhance PPAR at the transcriptional level or, alternatively, at the level of protein stability. Because PPAR upregulation by 15d-PGJ 2 in BE(2)-C cells was blocked by BADGE, it is reasonable to conclude that ligand-induced PPAR upregulation requires PPAR activation, which creates a positive feedback loop for tissue protection.

Thiazolidinediones (TZD) have been shown to protect brain from I/R injury 35-37 and reduce myocardial infarction. 38 Pioglitazone and troglitazone administered intraperitoneally reduced infarct volume and improve neurological function accompanied by suppression of COX-2 and IL-1ß in the rat stroke model. 35 Oral administration of pioglitazone for 4 days improved blood flow and reduced infarct volume in the rat model. 36 Our results show that intraventricular administration of rosiglitazone 2 hours after ischemia was effective in controlling infarct size. However, the effect was abated when administered after 3 hours. Thus, TZD drugs that are now commonly used clinically for treating diabetes may be useful in preventing I/R-induced tissue injury in humans. Although TZD alone has a narrow therapeutic window, its combination with PGI 2 analogs may prolong the therapeutic window and achieve a synergistic effect as PGI 2 protects against tissue damage by different mechanisms. 39,40

Acknowledgments

We thank Susan Mitterling for editorial assistance. The work is supported by grants from National Institutes of Health to K.K.W. (P50 NS-23327 and RO1 HL-50675) and by grants to T.N.L. from National Science Council of Taiwan and Academia Sinica of Taiwan.

【参考文献】
  Wu KK. Inducible cyclooxygenase and nitric oxide synthase. Adv Pharmacol. 1995; 33: 179-207.

Smith WL, DeWitt DL, Garavito RM. Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem. 2000; 69: 145-182.

Iadecola C, Sugimoto K, Niwa K, Kazama K, Ross ME. Increased susceptibility to ischemic brain injury in cyclooxygenase-1-deficient mice. J Cereb Blood Flow Metab. 2001; 21: 1436-1441.

Lin H, Lin T. -N., Cheung, W.M., Nian, G.M., Tseng, P.H., Chen, S.F., Chen, J.J., Shyue, S.K., Liou, J.Y., Wu, C.W., Wu, K.K. Cyclooxygenase-1 (COX-1) and Bicistronic COX-1/Prostacyclin synthase gene transfer protect against ischemic cerebral infarction. Circulation. 2002; 105: 1962-1969.

Hsu CY, Liu TH, Xu J, Hogan EL, Chao J, Sun G, Tai HH, Beckman JS, Freeman BA. Arachidonic acid and its metabolites in cerebral ischemia. Ann NY Acad Sci. 1989; 559: 282-295.

Narumiya S, Sugimoto Y, Ushikubi F. Prostanoid receptors: structures, properties, and functions. Physiol Rev. 1999; 79: 1193-1226.

Hao CM, Redha R, Morrow J, Breyer MD. Peroxisome proliferator-activated receptor delta activation promotes cell survival following hypertonic stress. J Biol Chem. 2002; 277: 21341-21345.

Igarashi M, Nagata A, Toh H, Urade Y, Hayaishi O. Structural organization of the gene for prostaglandin D synthase in the rat brain. Proc Natl Acad Sci U S A. 1992; 89: 5376-5380.

Forman BM, Tontonoz P, Chen J, Brun RP, Spiegelman BM, Evans RM. 15-Deoxy-delta 12, 14-prostaglandin J2 is a ligand for the adipocyte determination factor PPAR gamma. Cell. 1995; 83: 803-812.

Kliewer SA, Lenhard JM, Willson TM, Patel I, Morris DC, Lehmann JMA. prostaglandin J 2 metabolite binds peroxisome proliferator-activated receptor gamma and promotes adipocyte differentiation. Cell. 1995; 83: 813-819.

Gilroy DW, Colville-Nash PR, Willis D, Chivers J, Paul-Clark MJ, Willoughby DA. Inducible cyclooxygenase may have anti-inflammatory properties. Nature Med. 1999; 5: 698-701.

Ricote M, Li AC, Willson TM, Kelly CJ, Glass CK. The peroxisome proliferator-activated receptor-gamma is a negative regulator of macrophage activation. Nature. 1998; 391: 79-82.

Jiang C, Ting AT, Seed, B. PPAR. -gamma agonists inhibit production of monocyte inflammatory cytokines. Nature. 1998; 391: 82-86.

Bell-Parikh LC, Ide T, Lawson JA, McNamara P, Reilly, M. FitzGerald, G.A. Biosynthesis of 15-deoxy-delta12,14-PGJ 2 and the ligation of PPARgamma. J Clin Invest. 2003; 112: 945-955.

Choi DW, Maulucci-Gedde M, Kriegstein AR. Glutamate neurotoxicity in cortical cell culture. J Neurosci. 1987; 7: 357-368.

Shih CL, Chi SI, Chiu TH, Sun GY, Lin TN. Ethanol effects on nitric oxide production in cerebral pial cultures. Alcohol Clin Exp Res. 2001; 25: 612-618.

Zoldhelyi P, McNatt J, Xu XM, Loose-Mitchell D, Meidell RS, Clubb FJ Jr, Buja LM, Willerson JT, Wu KK. Prevention of arterial thrombosis by adenovirus-mediated transfer of cyclooxygenase gene. Circulation. 1996; 93: 10-17.

An G, Lin T -N, Liu JS, Xue JJ, He YY, Hsu CY. Expression of c-fos and c-jun family genes after focal cerebral ischemia. Ann Neurol. 1993; 33: 457-464.

O?Neill LA, Kaltschmidt CNF. -kappa B: a crucial transcription factor for glial and neuronal cell function. Trends Neurosci. 1997; 20: 252-258.

Chan PH. Reactive oxygen radicals in signaling and damage in the ischemic brain. J Cereb Blood Flow Metab. 2001; 21: 2-14.

Hyslop PA, Zhang Z, Pearson DV, Phebus LA. Measurement of striatal H 2 O 2 by microdialysis following global forebrain ischemia and reperfusion in the rat: correlation with the cytotoxic potential of H 2 O 2 in vitro. Brain Res. 1995; 671: 181-186.

Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci. 1999; 22: 391-397.

Gobbel GT, Chan PH. Neuronal death is an active, caspase-dependent process after moderate but not severe DNA damage. J Neurochem. 2001; 76: 520-531.

Ruan YW, Ling GY, Zhang JL, Xu ZC. Apoptosis in the adult striatum after transient forebrain ischemia and the effects of ischemic severity. Brain Res. 2003; 982: 228-240.

Juan SH, Lee TS, Tseng KW, Liou JY, Shyue SK, Wu KK, Chau LY. Adenovirus-mediated heme oxygenase-1 gene transfer inhibits the development of atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2001; 104: 1519-1525.

Alvarez-Maqueda M, El Bekay R, Alba G, Monteseirin J, Chacon P, Vega A, Martin-Nieto J, Bedoya FJ, Pintado E, Sobrino F. 15-Deoxy-delta 12,14-prostaglandin J 2 induces heme oxygenase-1 gene expression in a reactive oxygen species-dependent manner in human lymphocytes. J Biol Chem. 2004; 279: 21929-21937.

Subbaramaiah K, Lin DT, Hart JC, Dannenberg AJ. Peroxisome proliferator-activated receptor gamma ligands suppress the transcriptional activation of cyclooxygenase-2. Evidence for involvement of activator protein-1 and CREB-binding protein/p300. J Biol Chem. 2001; 276: 12440-12448.

Kondo M, Shibata T, Kumagai T, Osawa T, Shibata N, Kobayashi M, Sasaki S, Iwata M, Noguchi N, Uchida K. 15-Deoxy-Delta(12,14)-prostaglandin J 2 : the endogenous electrophile that induces neuronal apoptosis. Proc Natl Acad Sci U S A. 2002; 99: 7367-7372.

Bishop-Bailey D, Hla T. Endothelial cell apoptosis induced by the peroxisome proliferator-activated receptor (PPAR) ligand 15-deoxy-Delta12, 14-prostaglandin J2. J Biol Chem. 1999; 274: 17042-17048.

Padilla J, Kaur K, Cao HJ, Smith TJ, Phipps RP. Peroxisome proliferator activator receptor-gamma agonists and 15-deoxy-Delta(12,14)(12,14)-PGJ(2) induce apoptosis in normal and malignant B-lineage cells. J Immunol. 2000; 165: 6941-6948.

Piva R, Gianferretti P, Ciucci A, Taulli R, Belardo G, Santoro MG. 15-Deoxy-delta 12,14-prostaglandin J2 induces apoptosis in human malignant B cells: an effect associated with inhibition of NF-kappa B activity and down-regulation of antiapoptotic proteins. Blood. 2005; 105: 1750-1758.

Ward C, Dransfield I, Murray J, Farrow SN, Haslett C, Rossi AG. Prostaglandin D2 and its metabolites induce caspase-dependent granulocyte apoptosis that is mediated via inhibition of I kappa B alpha degradation using a peroxisome proliferator-activated receptor-gamma-independent mechanism. J Immunol. 2002; 168: 6232-6243.

Heneka MT, Feinstein DL, Galea E, Gleichmann M, Wullner U, Klockgether T. Peroxisome proliferator-activated receptor gamma agonists protect cerebellar granule cells from cytokine-induced apoptotic cell death by inhibition of inducible nitric oxide synthase. J Neuroimmuno. 1999; 100: 156-168.

Levonen AL, Dickinson DA, Moellering DR, Mulcahy RT, Forman HJ, Darley-Usmar VM. Biphasic effects of 15-deoxy-delta(12,14)-prostaglandin J 2 on glutathione induction and apoptosis in human endothelial cells. Arterioscler Thromb Vasc Biol. 2001; 21: 1846-1851.

Sundararajan S, Gamboa JL, Victor NA, Wanderi EW, Lust WD, Landreth GE. Peroxisome proliferator-activated receptor-gamma ligands reduce inflammation and infarction size in transient focal ischemia. Neuroscience. 2005; 130: 685-696.

Shimazu T, Inoue I, Araki N, Asano Y, Sawada M, Furuya D, Nagoya H, Greenberg JHA. peroxisome proliferator-activated receptor-gamma agonist reduces infarct size in transient but not in permanent ischemia. Stroke. 2005; 36: 353-359.

Uryu S, Harada J, Hisamoto M, Oda T. Troglitazone inhibits both post-glutamate neurotoxicity and low-potassium-induced apoptosis in cerebellar granule neurons. Brain Res. 2002; 924: 229-236.

Yue TL, Chen J, Bao W, Narayanan PK, Bril A, Jiang W, Lysko PG, Gu JL, Boyce R, Zimmerman DM, Hart TK, Buckingham RE, Ohlstein EH. In vivo myocardial protection from ischemia/reperfusion injury by the peroxisome proliferator-activated receptor-gamma agonist rosiglitazone. Circulation. 2001; 104: 2588-2594.

Matsuda S, Wen TC, Karasawa Y, Araki H, Otsuka H, Ishihara K, Sakanaka M. Protective effect of a prostaglandin I2 analog, TEI-7165, on ischemic neuronal damage in gerbils. Brain Res. 1997; 769: 321-328.

Cui Y, Kataoka Y, Satoh T, Yamagata A, Shirakawa N, Watanabe Y, Suzuki M, Yanase H, Kataoka K, Watanabe Y. Protective effect of prostaglandin I(2) analogs on ischemic delayed neuronal damage in gerbils. Biochem Biophys Res Commun. 1999; 265: 301-304.

作者单位：From Neuroscience Division (T.-N.L., W.-M.C., J.-S.W., J.-J.C., H.L., J.-J.Chen, S.-K.S.), Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan.; Vascular Biology Research Center (J.-Y.L., K.K.W.), Institute of Molecular Medicine and Division of Hematology, University of Texas-Houston H