Cyclooxygenase-2 Does Not Contribute to Postischemic Production of Reactive Oxygen Species

Abstract

We sought to determine whether reactive oxygen species (ROS) derived from cyclooxygenase-2 (COX-2) are involved in ischemic brain injury. Focal cerebral ischemia was induced by transient middle cerebral artery occlusion in C57BL/6 mice. The time course of neocortical ROS production was assessed in vivo using hydroethidine as a marker. The same brain sections were used for infarct volume measurements. Transient middle cerebral artery occlusion led to a biphasic increase in ROS production with peaks 2 and 72 h after reperfusion. The COX-2 inhibitor NS398 (10 mg/kg) attenuated the production of COX-2-derived prostaglandin E₂ and reduced brain injury, but did not affect ROS production at 2 and 72 h. Similarly, ROS production was not reduced in COX-2-null mice. In contrast, ROS production and brain injury were reduced in mice lacking the nox2 subunit of the superoxide-producing enzyme nicotinamide adenine dinucleotide phosphate (reduced form) oxidase. The data suggest that COX-2 is not a major source of oxygen radicals after cerebral ischemia and raise the possibility that other COX-2 reaction products, including prostanoids or nonoxygen-based radicals, mediate the COX-2-dependent component of the injury.

Keywords

hydroethidine mice NADPH oxidase nox2 reactive oxygen species

Introduction

Cyclooxygenase-2 (COX-2), the rate-limiting enzyme for prostanoid synthesis, has been implicated in the pathogenesis of a wide variety of neurologic diseases, including ischemic stroke (Iadecola and Gorelick, 2005; Minghetti, 2004). Cyclooxygenase-2 catalyzes the conversion of arachidonic acid to prostaglandin (PG) H₂ (Simmons et al, 2004). Prostaglandin H₂ is then converted into five prostanoids (PGE₂, PGI₂, PGD₂, PGF_2α, and thromboxane A₂) by cell-specific synthases (Hata and Breyer, 2004). In addition to prostanoids, COX-2 also generates free radical species (Armstead et al, 1988; Chan and Fishman, 1980; Jiang et al, 2004; Kontos et al, 1980; Torres et al, 2004). Pharmacological inhibition or genetic inactivation of COX-2 attenuates the brain damage resulting for focal and global cerebral ischemia (Iadecola et al, 2001a; Nakayama et al, 1998; Nogawa et al, 1997; Sasaki et al, 2004). Furthermore, mice overexpressing COX-2 in the brain are more susceptible to the brain damage produced by focal cerebral ischemia (Dore et al, 2003). These observations have raised the possibility that COX-2 inhibitors could be valuable tools in the treatment of ischemic brain injury (Minghetti, 2004).

The therapeutic potential of COX-2 inhibitors has been recently challenged by clinical reports that long-term treatment with these agents increases the incidence of cardiovascular events, including myocardial infarction and stroke (Topol, 2004). The mechanisms underlying these complications have not been elucidated, but it has been suggested that COX-2 inhibitors downregulate PGI₂ (prostacyclin) and upregulate thromboxane A₂ resulting in a prothrombotic state (FitzGerald, 2003). Therefore, COX-2 activity is not uniformly deleterious and treatments based on inhibiting the COX-2 pathway have to focus on the mediators responsible for the deleterious effects of the enzyme sparing those mediating the beneficial actions (Iadecola and Gorelick, 2005; Rocca, 2006).

Unfortunately, the COX-2 reaction products mediating the neurotoxicity of COX-2 have not been defined. Although COX-2-derived prostanoids could mediate brain injury, COX-2-derived radicals can lead to oxidative stress, a major pathogenic factor in ischemic brain injury (Armstead et al, 1988; Chan, 2001; Torres et al, 2004). Recent evidence suggests that in models of glutamate neurotoxicity prostanoids, not oxidative stress, mediate the deleterious effects of COX-2 (Carlson, 2003; Kawano et al, 2006; Manabe et al, 2004). However, the mediators responsible for COX-2-dependent injury in models of cerebral ischemia have not been defined. In particular, COX-2 is markedly upregulated in inflammatory cells invading the postischemic brain, which are a major source of oxidative stress (Iadecola et al, 1999; Traystman et al, 1991). Therefore, COX-2-derived radical species could contribute to the toxicity exerted by postischemic inflammation.

In the present study, we used a mouse model of transient focal cerebral ischemia to examine the contribution of COX-2-derived radicals in the early and late phases of ischemic brain damage. The data indicate that although reactive oxygen species (ROS) play a pathogenic role in ischemic injury, COX-2 does not contribute to postischemic oxidative stress. Rather, a nox2-containing nicotinamide adenine dinucleotide phosphate (reduced form) (NADPH) oxidase is a major source of ROS produced in the late stage of cerebral ischemia.

Materials and methods

Animals

All experimental procedures were approved by the Institutional Animal Care and Use Committee of Weill Cornell Medical College. Experiments were performed in C57BL/6 mice, COX-2-null mice, and mice lacking the gp91^phox subunit of NADPH oxidase (nox2) (Morham et al, 1995; Pollock et al, 1995). All mice (weight 20 to 22 g) were male and were obtained from in-house colonies (Iadecola et al, 2001a; Park et al, 2004b). Cyclooxygenase-2 and nox2-null mice were congenic with the C57BL/6 strain. Wild-type littermates were used as controls in all experiments.

Transient Middle Cerebral Artery Occlusion

Procedures for middle cerebral artery (MCA) occlusion were identical to those described previously (Cho et al, 2005; Park et al, 2004a) and are only summarized. Mice were anesthetized with a mixture of isoflurane (2% to 2.5%), oxygen, and nitrogen. For continuous monitoring of cerebral blood flow in the center of the ischemic territory, a fiber optic probe was glued to the parietal bone (2 mm posterior and 5 mm lateral to the bregma) and connected to a laser-Doppler flowmeter (Periflux System 5010; Perimed, Järfälla, Sweden). For MCA occlusion, a heat-blunted 6-0 monofilament surgical thread was inserted into the exposed external carotid artery, advanced into the internal carotid artery and wedged into the Circle of Willis at the origin of the MCA. Obstruction of the MCA was confirmed by an immediate cerebral blood flow reduction. The monofilament was left in place for 25 mins and then withdrawn. Only mice with a cerebral blood flow reduction by >85% during MCA occlusion and a recovery to at least 80% of baseline cerebral blood flow within 10 mins of reperfusion were included in our study. This procedure leads to reproducible cortical and striatal infarcts similar in size to those reported by others using transient MCA occlusion of comparable duration (Borsello et al, 2003; Boutin et al, 2001; Plesnila et al, 2001). Rectal temperature was kept constant (37.0°C ±0.5°C) during the surgical procedure and in the recovery period until the animals regained consciousness.

Determination of Reactive Oxygen Species Production

Reactive oxygen species production was determined using in vivo hydroethidine microfluorography (Kondo et al, 1997), as previously described (Cho et al, 2005; Manabe et al, 2004). Hydroethidine is a cell-permeable dye that is oxidized to ethidium by superoxide (Bindokas et al, 1996). Ethidium is trapped intracellularly by intercalating with DNA (Rothe and Valet, 1990). The fluorescence signal attributable to ethidium reflects cumulative ROS production during the period between administration of hydroethidine and killing of the animal. Hydroethidine (10 mg/kg) was injected into the jugular vein in isoflurane-anesthetized mice 1 h before the animal is killed. Brains were removed, frozen, and cut in a cryostat (thickness 20 μm), collected at 600-μm intervals. The sections were analyzed with a Nikon E800 fluorescence microscope (Melville, NY, USA) equipped with a custom filter set (Chroma Technology, Rockingham, VT, USA). Images were acquired by a computer-controlled digital monochrome camera (Coolsnap; Roper Scientific, Trenton, NJ, USA) attached to the microscope. The analysis of ROS production was performed in a masked manner using IPLab software package (Scantalytics, Fairfax, VA, USA) (Cho et al, 2005; Manabe et al, 2004). After subtracting the camera dark current, pixel intensities of ethidium signals were assessed in the ischemic territory. Fluorescence intensities were measured in five serial sections per animal (rostrocaudal levels +1.6, +1.0, +0.4, −0.2, −0.8 mm from bregma). The sum of the fluorescence intensity for each region was divided by the total number of pixels analyzed and expressed as relative fluorescence units (Cho et al, 2005; Manabe et al, 2004). Reactive oxygen species production was assessed 1, 2, 4, 8, 24, 48, and 72 h after MCA occlusion (Figure 1A).

Infarct Volume Measurement

Infarct volume measurements were performed using the same brain sections in which ROS production was assessed. After staining the sections with cresyl violet, the infarct volume was measured using an image analyzer (MCID; Imaging Research, St Catharines, Ontario, Canada). To eliminate the contribution of postischemic edema to the volume of injury, infarct volume measurements were corrected for swelling according to the method of Lin et al. (1993) as previously described (Zhang and Iadecola, 1994).

Treatment with NS398

Some mice were treated with the COX-2 inhibitor NS398 (10 mg/kg; ip) or vehicle (H₂O, pH 12), as previously described (Sugimoto and Iadecola, 2003). Mice sacrificed 2 h after MCA occlusion received one injection of NS398 10 mins after reperfusion. Animals that survived for 72 h were treated 10 mins and 6 h after reperfusion, and thereafter twice daily at days 1 and 2, and once at day 3.

Prostaglandin Assay

PGE₂ concentration in the ischemic tissue was measured 2 h and 72 h after MCA occlusion. These time points correspond to the peak of ROS production (Figure 1A). Samples of the ischemic neocortex were dissected, weighted, and frozen in liquid nitrogen. Samples were homogenized and centrifuged for 10 mins at 12,000 r.p.m. at 4°C. PGE₂ concentration was measured in the supernatant using an enzyme-linked immunosorbent assay-based assay kit (Cayman Chemical Co., Ann Arbor, MI, USA) (Manabe et al, 2004; Nogawa et al, 1997). PGE₂ concentration was expressed as ng/g of wet tissue. Although brain edema could lead to underestimation of the final PGE₂ concentration, we have previously shown that NS398 does not influence brain edema (Nogawa et al, 1997).

Statistical Analysis

Data are presented as mean±s.e.m. Comparisons between two groups were statistically evaluated by the Student's t test. Multiple comparisons were evaluated by analysis of variance followed by Bonferroni's multiple comparison test. Differences were considered significant at P<0.05.

Results

Temporal Profile of Reactive Oxygen Species Production and Tissue Injury After Middle Cerebral Artery Occlusion

We first examined the time course of ROS production in the postischemic brain. Transient MCA occlusion increased ROS production in the affected hemisphere (Figure 1A). The increase was first observed 1 to 2 h after MCA occlusion (P<0.05). Reactive oxygen species production stabilized at a lower level between 4 and 24 h, and increased again 48 to 72 h after MCA occlusion (Figure 1A). Injury volume developed rapidly between 1 and 2 h after MCA occlusion, corresponding to the first peak of ROS production (Figures 1A and 1B). Thus, 2 h after MCA occlusion the volume of injury was 44% of the volume at 72 h (Figure 1B). Lesion volume rose steadily between 4 and 48 h after MCA occlusion, reaching a plateau at 48 to 72 h (Figure 1B).

Cyclooxygenase-2 Contribution to Reactive Oxygen Species Production 2 and 72 h After Middle Cerebral Artery Occlusion

We next sought to define the contribution of COX-2 to postischemic oxidative stress. Analyses were performed at 2 and 72 h because these times correspond to the peaks of ROS production (Figure 1A). Administration of NS398 reduced the postischemic elevation of the COX-2-derived prostanoid PGE₂ (P<0.05; Figure 2A). The effect was associated with a reduction in infarct volume 72 h after MCA occlusion (P<0.05; Figure 2B). However, NS398 did not influence postischemic ROS production 2 or 72 h after MCA occlusion (Figures 2C and 2D). Similarly, infarct volume, but not ROS production, was attenuated in COX-2-null mice (Figures 2B, 2C and 2D).

Reactive Oxygen Species Production and Ischemic Brain Injury in nox2-null Mice

Nicotinamide adenine dinucleotide phosphate (reduced form) oxidase is a potential source of ROS in the late phase of cerebral ischemia (Park et al, 2004a; Vallet et al, 2005) and contributes to ischemic injury (Walder et al, 1997). Therefore, we used mice lacking nox2 to assess the contribution of NADPH oxidase to ROS production and ischemic injury 72 h after MCA occlusion. Nox2-null mice exhibited a 44% reduction in ROS (P<0.05) in the postischemic brain at 72 h compared with wild-type mice (Figure 3A). In addition, the infarct volume at 72 h in nox2-null mice was 61% (P<0.05) smaller than in wild-type mice (Figure 3B).

Discussion

It is well established that ROS are produced after cerebral ischemia and play a critical role in the resulting tissue damage (Chan, 2001). However, the sources of postischemic ROS have not been elucidated. Cyclooxygenase-2 plays a deleterious role in ischemic brain injury and is also a potential source of oxidative stress (Armstead et al, 1988; Chan and Fishman, 1980; Kontos et al, 1980; Minghetti, 2004). Therefore, we sought to determine whether COX-2-derived ROS contribute to the brain damage induced by transient focal ischemia. We found that MCA occlusion produced a marked increase in ROS production, which peaked 2 and 72 h after ischemia. Cyclooxygenase-2 inhibition with NS398 attenuated PGE₂ production and reduced injury volume, but did not attenuate postischemic ROS production. Similarly, ROS production was not altered in COX-2-null mice. In contrast, ROS production and brain injury were reduced in mice lacking nox2. These data, collectively, demonstrate that COX-2 does not play a role in postischemic ROS production. In particular, COX-2 is not involved in the oxidative stress produced by inflammatory cells in the late stages of cerebral ischemia. Furthermore, the evidence suggests that a nox2-containing NADPH oxidase is a significant source of ROS in the late postischemic period.

The fact that ROS production was unchanged after treatment with NS398 cannot be attributed to insufficient inhibition of COX-2, because PGE₂ production was attenuated and injury volume was reduced. Moreover, postischemic ROS production was not affected in null mice lacking COX-2. It is also unlikely that the increased fluorescent signal observed in the postischemic brain reflects oxidation of hydroethidine by factors other than ROS, because ROS production was attenuated in mice lacking nox2, in which NADPH oxidase-dependent ROS production is reduced (Park et al, 2004b). Cytochrome c released from mitochondria as a result of brain injury could also oxidate hydroethidine (Papapostolou et al, 2004). However, this is unlikely to be the case in our model because COX-2 inhibition or gene inactivation reduced the tissue damage, and presumably cytochrome c release, but did not affect the fluorescent signal. Therefore, the findings of the present study cannot be attributed to confounding effects related to the experimental preparation or the method used to assess ROS production.

The identity of the radical species produced by COX-2 has not been established. Studies in vivo suggest that COX activity can generate oxygen radicals, such as superoxide (Armstead et al, 1988; Chan and Fishman, 1980; Kontos et al, 1980; Torres et al, 2004). However, in vitro investigations have provided evidence that COX-2 activity is coupled to generation of carbon centered radicals that leads to lipid peroxidation (Jiang et al, 2004). Although our data suggest that COX-2 is not a source of postischemic oxygen radicals, we cannot rule out the participation of other radical species derived from COX-2.

To our knowledge, this is the first study to correlate the time course of postischemic ROS production with the maturation of cerebral infarction. Analysis of the temporal profile of postischemic ROS production revealed a biphasic pattern with an early peak at 2 h and a late increase beginning 48 h after MCA occlusion. The fact that the ROS peak at 2 h in parallel with rapid expansion of the lesion suggests a contribution of this early ROS burst to the evolution of ischemic brain injury. The slightly lower ROS production from 4 to 24 h could be related to the cell death resulting from the early ROS burst, whereas the second peak of ROS activity could result from the inflammatory reaction that involves the ischemic brain at this time (Iadecola et al, 2004).

Cyclooxygenase-2 contributes both to the early and late phases of ischemic brain injury. In the early phase, evidence suggests that the contribution of COX-2 is related to the neurotoxicity exerted by glutamate receptors, which are critical for the initiation of ischemic brain injury (Lo et al, 2003). Thus, COX-2 plays a role in the brain damage produced by activation of glutamate receptors (Carlson, 2003; Hewett et al, 2000; Iadecola et al, 2001a; Manabe et al, 2004). Our findings that COX-2-derived ROS do not contribute to the ROS production in the early phase of cerebral ischemia is consistent with our previous demonstration that pharmacological inhibition or genetic inactivation of COX-2 does not attenuate the ROS production induced by N-methyl-d-aspartate receptor activation (Manabe et al, 2004). Mitochondria are a likely source of ROS in the early stages of cerebral ischemia (Chan, 2001).

The contribution of COX-2 to the late phase of ischemic damage is likely to be related to its role in postischemic inflammation. Thus, COX-2 is expressed in infiltrating inflammatory cells and delayed treatment with COX-2 inhibitors attenuates the injury (Iadecola et al, 1999; Miettinen et al, 1997; Sugimoto and Iadecola, 2003). The present study suggests that COX-2 does not contribute to the ROS production that occurs in the late phase of ischemic injury. Rather, our data clearly indicate that a nox2-containing NADPH oxidase is a substantial source of ROS and contributes to ischemic brain injury. Although the role of NADPH oxidase in ischemic brain injury was previously established (Walder et al, 1997), nox2-dependent ROS production in the late postischemic period had not been demonstrated. However, the fact that ROS production in nox2-null mice is not completely abolished suggests additional sources of ROS, including NADPH oxidases containing other nox homologues, mitochondrial enzymes, lipoxygenases, p450 enzymes, xanthine oxidase, as well as NOS uncoupling from substrate and cofactor depletion (Chan, 2001; Gottlieb, 2003; Munzel et al, 2005; Traystman et al, 1991; Vallet et al, 2005). Although COX-1 is also a potential source of radicals, inhibition of COX-1 enlarges the infarct produced by MCA occlusion, suggesting a protective role (Iadecola et al, 2001b). Therefore, it is unlikely that COX-1-derived radicals are a major factor in the tissue damage produced by ischemia.

There is increasing evidence that COX-2 exerts its neurotoxic effect through prostanoids, PGE₂ in particular (Carlson, 2003; Kawano et al, 2006; Manabe et al, 2004). It has been shown that the COX-2-dependent component of excitotoxic brain injury is mediated by PGE₂ via activation of EP1 receptors (Kawano et al, 2006). Activation of EP1 receptors disrupts intracellular Ca²⁺ homeostasis by impairing Na⁺–Ca²⁺ exchange and contributes to excitotoxic cell death (Kawano et al, 2006). However, it is unclear whether EP1 receptor-dependent mechanisms are also involved in the inflammatory component of the damage in which COX-2 also plays a role (Hata and Breyer, 2004). Irrespective of these mechanisms, the data presented here suggest that COX-2-derived ROS do not contribute to the late inflammatory phase of the injury. However, we cannot exclude the participation of other radical species derived from COX-2, such as carbon centered radicals, which can lead to lipid peroxidation (Jiang et al, 2004).

The COX-2 pathway is an important therapeutic target for acute ischemic stroke and other neurologic diseases (FitzGerald, 2003; Minghetti, 2004). However, recent reports that longterm treatment with COX-2 inhibitors increases the incidence of stroke and myocardial infarction have led to a reevaluation of the therapeutic potential of these drugs (Iadecola and Gorelick, 2005; Topol, 2004). These vascular complications have been attributed to the inhibition of the vasoprotective effects of COX-2-derived PGI₂ (FitzGerald, 2003). Thus, therapeutic interventions based on the COX-2 pathway need to target the downstream effectors of the injury sparing the beneficial vascular actions of COX-2 (Rocca, 2006). We have shown here that COX-2-derived ROS do not contribute to ischemic damage. Although the data are consistent with the hypothesis that COX-2-derived prostanoids are important mediators of injury and a potential therapeutic target (Ahmad et al, 2006; Carlson, 2003; Kawano et al, 2006), we cannot rule out a pathogenic role of COX-2-derived carbon-centered radicals.

In conclusion, we have shown that focal cerebral ischemia is associated with a biphasic pattern of ROS production that correlates with the development of tissue damage. We found that COX-2 inhibition or genetic inactivation reduces ischemic brain injury without altering postischemic ROS production. Reactive oxygen species production and ischemic injury are attenuated in mice lacking the nox2 subunit of NAPDH oxidase. The findings suggest that COX-2 is not a major source of oxygen radicals after cerebral ischemia and support the hypothesis that other COX-2 reaction products, including prostanoids or nonoxygen-based radicals, mediate the COX-2-dependent component of the injury.

References

Ahmad Saleem

Ahmad

Dore

(2006) Prostaglandin EP1 receptor contributes to excitotoxicity and focal ischemic brain damage. Toxicol Sci 89:265–70

Armstead

Mirro

Busija

Leffler

(1988) Postischemic generation of superoxide anion by newborn pig brain. Am J Physiol 255:H401–3

Bindokas

Jordan

Lee

Miller

(1996) Superoxide production in rat hippocampal neurons: selective imaging with hydroethidine. J Neurosci 16:1324–36

Borsello

Clarke

Hirt

Vercelli

Repici

Schorderet

Bogousslavsky

Bonny

(2003) A peptide inhibitor of c-Jun N-terminal kinase protects against excitotoxicity and cerebral ischemia. Nat Med 9:1180–6

Boutin

LeFeuvre

Horai

Asano

Iwakura

Rothwell

(2001) Role of IL-1alpha and IL-1beta in ischemic brain damage. J Neurosci 21:5528–34

Carlson

(2003) Neuroprotection of cultured cortical neurons mediated by the cyclooxygenase-2 inhibitor APHS can be reversed by a prostanoid. J Neurosci Res 71:79–88

Chan

(2001) Reactive oxygen radicals in signaling and damage in the ischemic brain. J Cereb Blood Flow Metab 21:2–14

Chan

Fishman

(1980) Transient formation of superoxide radicals in polyunsaturated fatty acidinduced brain swelling. J Neurochem 35:1004–7

Cho

Park

Febbraio

Anrather

Park

Racchumi

Silverstein

Iadecola

(2005) The class B scavenger receptor CD36 mediates free radical production and tissue injury in cerebral ischemia. J Neurosci 25:2504–12

10.

Dore

Otsuka

Mito

Sugo

Hand

Hurn

Traystman

Andreasson

(2003) Neuronal overexpression of cyclooxygenase-2 increases cerebral infarction. Ann Neurol 54:155–62

11.

FitzGerald

(2003) COX-2 and beyond: approaches to prostaglandin inhibition in human disease. Nat Rev Drug Discov 2:879–90

12.

Gottlieb

(2003) Cytochrome P450: major player in reperfusion injury. Arch Biochem Biophys 420:262–7

13.

Hata

Breyer

(2004) Pharmacology and signaling of prostaglandin receptors: multiple roles in inflammation and immune modulation. Pharmacol Ther 103:147–66

14.

Hewett

Uliasz

Vidwans

Hewett

(2000) Cyclooxygenase-2 contributes to N-methyl-D-aspartate-mediated neuronal cell death in primary cortical cell culture. J Pharmacol Exp Ther 293:417–25

15.

Iadecola

Cho

Feuerstein

Hallenbeck

(2004) Cerebral ischemia and inflammation. In: Stroke: pathophysiololgy, diagnosis, and management ( Mohr

Choi

Grotta

, eds), Philadelphia: Churchill Livingstone, 883–93

16.

Iadecola

Forster

Nogawa

Clark

Ross

(1999) Cyclooxygenase-2 immunoreactivity in the human brain followingcerebral ischemia. Acta Neuropathol (Berlin) 98:9–14

17.

Iadecola

Gorelick

(2005) The Janus face of cyclooxygenase-2 in ischemic stroke: shifting toward downstream targets. Stroke 36:182–5

18.

Iadecola

Niwa

Nogawa

Zhao

Nagayama

Araki

Morham

Ross

(2001a) Reduced susceptibility to ischemic brain injury and N-methyl-D-aspartate-mediated neurotoxicity in cyclooxygenase-2-deficient mice. Proc Natl Acad Sci USA 98:1294–9

19.

Iadecola

Sugimoto

Niwa

Kazama

Ross

(2001b) Increased susceptibility to ischemic brain injury in cyclooxygenase-1-deficient mice. J Cereb Blood Flow Metab 21:1436–41

20.

Jiang

Borisenko

Osipov

Martin

Chen

Shvedova

Sorokin

Tyurina

Potapovich

Tyurin

Graham

Kagan

(2004) Arachidonic acid-induced carbon-centered radicals and phospholipid peroxidation in cyclo-oxygenase-2-transfected PC12 cells. J Neurochem 90:1036–49

21.

Kawano

Anrather

Zhou

Park

Wang

Frys

Kunz

Cho

Orio

Iadecola

(2006) Prostaglandin E(2) EP1 receptors: downstream effectors of COX-2 neurotoxicity. Nat Med 12:225–9

22.

Kondo

Reaume

Huang

Carlson

Murakami

Chen

Hoffman

Scott

Epstein

Chan

(1997) Reduction of CuZn-superoxide dismutase activity exacerbates neuronal cell injury and edema formation after transient focal cerebral ischemia. J Neurosci 17:4180–9

23.

Kontos

Wei

Povlishock

Dietrich

Magiera

Ellis

(1980) Cerebral arteriolar damage by arachidonic acid and prostaglandin G2. Science 209:1242–5

24.

Lin

Khan

Hsu

(1993) Effect of brain edema on infarct volume in a focal cerebral ischemia model in rats. Stroke 24:117–21

25.

Dalkara

Moskowitz

(2003) Mechanisms, challenges and opportunities in stroke. Nat Rev Neurosci 4:399–415

26.

Manabe

Anrather

Kawano

Niwa

Zhou

Ross

Iadecola

(2004) Prostanoids, not reactive oxygen species, mediate COX-2-dependent neurotoxicity. Ann Neurol 55:668–75

27.

Minghetti

(2004) Cyclooxygenase-2 (COX-2) in inflammatory and degenerative brain diseases. J Neuropathol Exp Neurol 63:901–10

28.

Miettinen

Fusco

Yrjanheikki

Keinanen

Hirvonen

Roivainen

Narhi

Hokfelt

Koistinaho

(1997) Spreading depression and focal brain ischemia induce cyclooxygenase-2 in cortical neurons through N-methyl-D-aspartic acid-receptors and phospholipase A2. Proc Natl Acad Sci USA 94:6500–5

29.

Morham

Langenbach

Loftin

Tiano

Vouloumanos

Jennette

Mahler

Kluckman

Ledford

Lee

Smithies

(1995) Prostaglandin synthase 2 gene disruption causes severe renal pathologyin the mouse. Cell 83:473–82

30.

Munzel

Daiber

Ullrich

Mulsch

(2005) Vascular consequences of endothelial nitric oxide synthase uncoupling for the activity and expression of the soluble guanylyl cyclase and the cGMP-dependent protein kinase. Arterioscler Thromb Vasc Biol 25:1551–7

31.

Nakayama

Uchimura

Zhu

Nagayama

Rose

Stetler

Isakson

Chen

Graham

(1998) Cyclooxygenase-2 inhibition prevents delayed death of CA1 hippocampal neurons followingglobal ischemia. Proc Natl Acad Sci USA 95:10954–9

32.

Nogawa

Zhang

Ross

Iadecola

(1997) Cyclo-oxygenase-2 gene expression in neurons contributes to ischemic brain damage. J Neurosci 17:2746–55

33.

Papapostolou

Patsoukis

Georgiou

(2004) The fluorescence detection of superoxide radical using hydroethidine could be complicated by the presence of heme proteins. Anal Biochem 332:290–8

34.

Park

Cho

Frys

Racchumi

Zhou

Anrather

Iadecola

(2004a) Interaction between inducible nitric oxide synthase and poly(ADP-ribose) polymerase in focal ischemic brain injury. Stroke 35:2896–901

35.

Park

Anrather

Zhou

Frys

Wang

Iadecola

(2004b) Exogenous NADPH increases cerebral blood flow through NADPH oxidase-dependent and -independent mechanisms. Arterioscler Thromb Vasc Biol 24:1860–5

36.

Plesnila

Zinkel

Amin-Hanjani

Qiu

Chiarugi

Thomas

Kohane

Korsmeyer

Moskowitz

(2001) BID mediates neuronal cell death after oxygen/glucose deprivation and focal cerebral ischemia. Proc Natl Acad Sci USA 98:15318–23

37.

Pollock

Williams

Gifford

Fisherman

Orkin

Doerschuk

Dinauer

(1995) Mouse model of X-linked chronicgranulomatous disease, an inherited defect in phagocyte superoxide production. Nat Genet 9:202–9

38.

Rocca

(2006) Targeting PGE2 receptor subtypes rather than cyclooxygenases: a bridge overtroubled water? Mol Interv 6:68–73

39.

Rothe

Valet

(1990) Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidine and 2′,7′-dichlorofluorescin. J Leukoc Biol 47:440–8

40.

Sasaki

Kitagawa

Yamagata

Takemiya

Tanaka

Omura-Matsuoka

Sugiura

Matsumoto

Hori

(2003) Amelioration of hippocampal neuronal damage after transient forebrain ischemia in cyclooxygenase-2-deficient mice. J Cereb Blood Flow Metab 24:107–13

41.

Simmons

Botting

Hla

(2004) Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacol Rev 56:387–437

42.

Sugimoto

Iadecola

(2003) Delayed effect of administration of COX-2 inhibitor in mice with acute cerebral ischemia. Brain Res 960:273–6

43.

Topol

(2004) Failing the public health—rofecoxib, Merck, and the FDA. N Engl J Med 351:1707–9

44.

Torres

Anderson

Marro

Mishra

Delivoria-Papadopoulos

(2004) Cyclooxygenase-mediated generation of free radicals during hypoxia in the cerebral cortex of newborn piglets. Neurochem Res 29:1825–30

45.

Traystman

Kirsch

Koehler

(1991) Oxygen radical mechanisms of brain injury followingischemia and reperfusion. J Appl Physiol 71:1185–95

46.

Vallet

Charnay

Steger

Ogier-Denis

Kovari

Herrmann

Michel

Szanto

(2005) Neuronal expression of the NADPH oxidase NOX4, and its regulation in mouse experimental brain ischemia. Neuroscience 132:233–8

47.

Walder

Green

Darbonne

Mathias

Rae

Dinauer

Curnutte

Thomas

(1997) Ischemic stroke injury is reduced in mice lacking a functional NADPH oxidase. Stroke 28:2252–8

48.

Zhang

Iadecola

(1994) Infarct measurement methodology. J Cereb Blood Flow Metab 14:697–8