Pathogenic Importance of Cysteinyl Leukotrienes inDevelopment of Gastric Lesions Induced byIschemia/Reperfusion in Mice

Yuka Nakamori, Yoshino Komatsu, Tohru Kotani, Shinji Kojima, and Koji TakeuchiDivision of Pathological Sciences, Department of Pharmacology and Experimental Therapeutics, Kyoto PharmaceuticalUniversity, Misasagi, Yamashina, Kyoto, Japan

Received October 9, 2009; accepted December 29, 2009

ABSTRACTWe examined the role of cysteinyl leukotrienes (CysLTs) in thegastric ulcerogenic response to ischemia/reperfusion (I/R) inmice. Experiments were performed in male C57BL/6J miceafter 18-h fasting. Under urethane anesthesia, the celiac arterywas clamped for 30 min, and then reperfusion was achieved byremoving the clamp. The stomach was examined for lesions 60min thereafter. The severity of I/R-induced gastric damage wasreduced by prior administration of pranlukast [CysLT receptortype 1 (CysLT1R) antagonist] as well as 1-[[5�-(3�-methoxy-4�-ethoxycarbonyl-oxyphenyl)-2�,4�-pentadienoyl]aminoethyl]-4-diphenylmethoxypiperidine [TMK688; 5-lipoxygenase (5-LOX)inhibitor]. On the contrary, these lesions were markedly wors-ened by pretreatment with indomethacin, and this responsewas abrogated by the coadministration of TMK688 or pran-lukast. The gene expression of CysLT1R but not 5-LOX was

up-regulated in the stomach after I/R, but both expressionswere increased under I/R in the presence of indomethacin. I/Rslightly increased the mucosal CysLT content of the stomach,yet this increase was markedly enhanced when the animalswere pretreated with indomethacin. The increased CysLT bio-synthetic response to indomethacin during I/R was attenuatedby TMK688. Indomethacin alone caused a slight increase ofCysLT1R expression and markedly up-regulated 5-LOX expres-sion in the stomach. We concluded that I/R up-regulated theexpression of CysLT1R in the stomach; CysLTs play a role inthe pathogenesis of I/R-induced gastric damage through theactivation of CysLT1R; and the aggravation by indomethacin ofthese lesions may be brought about by the increase of CysLTproduction and the up-regulation of 5-LOX expression, in ad-dition to the decreased prostaglandin production.

Ischemia/reperfusion (I/R) injury is caused by interruptionof the blood supply to an organ tissue followed by the rein-troduction of blood into the affected area. The phenomenon ofI/R injury is a major clinical problem after stroke, infarction,shock, and organ transplantation (Boros et al., 1989; Stein etal., 1990). The depletion of ATP and disturbance of intracel-lular Ca2� homeostasis have been suggested as the majorpathophysiological mechanism during ischemia, leading toloss of cell viability (Farber et al., 1981; Cheung et al., 1986).Reperfusion of ischemic tissues paradoxically exacerbatesthe injury process and leads to the release of reactive oxygenspecies and proinflammatory mediators, and the attraction of

inflammatory cells infiltrating the tissues (Chamoun et al.,2000; Piper et al., 2003). Indeed, many studies showed that axanthine oxidase inhibitor as well as radical scavengers pro-tect the stomach against injury during I/R, confirming theimportance of free radicals in the pathogenesis of I/R-inducedgastric injury (Perry et al., 1986; Yasue et al., 1992).

Prostaglandins (PGs), cyclooxygenase (COX) products, areessential to maintain the mucosal integrity of the gastroin-testinal tract, and a deficiency of endogenous PGs shows apropensity to damage the mucosa (Kunikata et al., 2001;Suzuki et al., 2001; Takeuchi et al., 2007) and worsens theulcerogenic response to noxious stimuli in the gastrointesti-nal tract, including I/R (Takeuchi et al., 1986; Kotani et al.,2006; Tanaka et al., 2007). Leukotrienes (LTs), lipoxygenase(5-LOX) products, are also known to play a role in the inflam-matory response (Steinhilber, 1999). Thus, it is possible thatLTs are also involved in the pathogenic mechanism of I/R-induced gastric lesions, in addition to PGs. Among the 5-LOX

This work was supported in part by the Kyoto Pharmaceutical University’s“21st Century COE” program [Grant K 26] and the “Open Research” Program[Grant 040019] from the Ministry of Education, Science and Culture of Japan.

Article, publication date, and citation information can be found athttp://jpet.aspetjournals.org.

doi:10.1124/jpet.109.162578.

ABBREVIATIONS: I/R, ischemia/reperfusion; PG, prostaglandin; COX, cyclooxygenase; LT, leukotriene; 5-LOX, 5-lipoxygenase; CysLT, cysteinylleukotriene; CysLT1R, CysLT receptor type 1; TMK688, 1-[[5�-(3�-methoxy-4�-ethoxycarbonyl-oxyphenyl)-2�,4�-pentadienoyl]aminoethyl]-4-diphe-nylmethoxypiperidine; MPO, Myeloperoxidase; RT-PCR, reverse transcription-polymerase chain reaction; GAPDH, glyceraldehyde-3-phosphatedehydrogenase; i.d., intraduodenally; NS-398, N-[2-(cyclohexyloxyl)-4-nitrophenyl]-methane sulfonamide; TXA2, thromboxane A2.

0022-3565/10/3331-91–98$20.00THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 333, No. 1Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics 162578/3567146JPET 333:91–98, 2010 Printed in U.S.A.

91

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

products, LTC4, LTD4, and LTE4, referred to as cysteinyl LTs(CysLTs), are recognized as potent inflammatory mediators(Murakami et al., 1995; Gronert et al., 2001). CysLTs aregenerated from cell membrane phospholipid-associated ara-chidonic acids by 5-LOX and are reported to play a patho-genic role in the development of I/R-induced damage in kid-ney, liver, and bladder (Takamatsu et al., 2004; Sener et al.,2006, 2007). However, it remains unknown whether CysLTsare involved in the development of gastric lesions under I/R.

Recently, two subtypes of CysLT receptors, CysLT1R andCysLT2R, have been characterized, both of which are classicG protein-coupled receptors with seven transmembrane do-mains (Brink et al., 2003). It has been reported that CysLT1Ris highly expressed in spleen, small intestine, colon, heart,pancreas, kidney, liver, and placenta (Lynch et al., 1999;Sarau et al., 1999) and detected in mast cells, mononuclearcells, and microvascular endothelia by in situ hybridizationor immunohistochemistry (Lynch et al., 1999; Zhang et al.,2004). Several studies showed that the administration of a5-LOX inhibitor or a selective CysLT1R antagonist reducedthe severity of ethanol- or nonsteroidal anti-inflammatorydrug-induced gastric injury by inhibiting neutrophil adhe-sion (Sala and Folco, 2001; Dengiz et al., 2007), suggestingthe pathogenic importance of CysLTs/CysLT1R in gastriclesions. Because CysLT2R, found in pulmonary vein prepa-rations by pharmacological assays, has now been reported toexpress in various tissues (Brink et al., 2003; Kanaoka andBoyce, 2004; Moos et al., 2008), it is possible that CysLT2Rmay also play a role in the pathogenesis of gastric lesions asinduced by I/R. However, none of the selective CysLT2Rantagonists is available for in vivo studies.

In the present study, we thus focused on CysLTs andCysLT1R and examined the effects of a 5-LOX inhibitor anda selective CysLT1R antagonist on I/R-induced gastric dam-age in mice and investigated the roles of CysLTs/CysLT1R inmucosal defense during I/R. In addition, we examined themechanism by which indomethacin aggravates I/R-inducedlesions, in relation to CysLTs/CysLT1R as well as 5-LOX inthe stomach.

Materials and MethodsAnimals. Male C57BL/6J mice (3 months old; SLC, Shizuoka,

Japan) were used. The animals were deprived of food but allowedfree access to tap water for 18 h before the experiments. All experi-ments were carried out using five to eight mice under urethaneanesthesia, unless otherwise specified. The experimental procedureswere approved by the Experimental Animal Research Committee ofKyoto Pharmaceutical University.

Induction of Ischemia and Reperfusion-Induced GastricDamage. Acute gastric mucosal lesions were produced by I/R inmice, according to a modified method reported by Wada et al. (1996).In brief, under urethane anesthesia (1.25 g/kg i.p.), the celiac arterywas clamped with a small clamp (disposable vascular clip; holdingforce 40, g; BEAR Medical Corporation, Chiba, Japan), and 30 minlater reperfusion was achieved through removal of the clamp. In apreliminary study, we found that macroscopically visible lesionswere consistently observed in the stomach 60 min after the onset ofreperfusion following ischemia for 30 min. The effect of varioustreatments on the formation of lesions was assessed 60 min after thereperfusion, according to a previous study (Yoshikawa et al., 1989).After reperfusion for 60 min, the stomach was excised, inflated byinjecting 0.4 ml of 2% formalin for 10 min to fix the tissue walls, andopened along the greater curvature. Pranlukast (a CysLT1R antag-

onist; 0.1–10 mg/kg; Nishio et al., 2007), montelukast (a CysLT1Rantagonist; 3 mg/kg), or TMK688 (a 5-LOX inhibitor; 3–30 mg/kg;Konaka et al., 1999) was administered orally 60 min before ischemia.Indomethacin (a nonselective COX inhibitor; 5 mg/kg) was givensubcutaneously 30 min before ischemia. The area (square millime-ters) of hemorrhagic lesions developed in the stomach was measuredunder a dissecting microscope (Olympus, Tokyo, Japan) with asquare grid (�10). The person measuring the lesions did not knowthe treatments given to the animals. In some cases, the gastricmucosa was examined with a light microscope after I/R. The animalswere killed after I/R treatment, and the stomachs were excised. Thetissue samples were then immersed in 2% formalin-saline, embeddedin paraffin, sectioned at 4 �m, and stained with hematoxylin andeosin.

Determination of Myeloperoxidase Activity. Myeloperoxi-dase (MPO) activity in the gastric mucosa was measured after I/Rtreatment in mice, according to a modified version of the method ofKrawisz et al. (1984). At 60 min after I/R treatment, the animalswere sacrificed by the withdrawal of blood from the heart throughperfusion with saline, and the stomach was excised. After rinsing ofthe tissue with cold saline, the mucosa was scraped with glass slides,weighed, and homogenized in a 50 mM phosphate buffer containing0.5% hexadecyl-trimethyl-ammonium bromide, pH 6.0 (Sigma-Al-drich, St. Louis, MO). The homogenized samples were subjected tofreezing and thawing three times, and centrifuged at 2000g for 10min at 4°C. MPO activity in the supernatant was determined byadding 50 �l of the supernatant to 50 �l of 10 mM phosphate buffer,pH 6.0, and 50 �l of 1.5 M o-dianisidine hydrochloride (Sigma-Aldrich) containing 0.0005% (w/v) hydrogen peroxide. The changesin absorbance at 450 nm were recorded on a microplate reader(VERSAmax; Molecular Devices, Sunnyvale, CA). Sample proteincontent was estimated by spectrophotometric assay (protein assaykit; Pierce Chemical, Rockford, IL), and the MPO activity was ob-tained from the slope of the reaction curve, based on the followingequation: specific activity (micromoles of H2O2 per minute per mil-ligram of protein � (OD/min)/OD/�mol H2O2 � mg protein), whereOD is optical density. Pranlukast (3 mg/kg) or TMK688 (30 mg/kg)was administered orally 60 min before ischemia, whereas indometh-acin was given subcutaneously 30 min before ischemia.

Measurement of Mucosal CysLT Levels. Levels of CysLT inthe gastric mucosa were measured after I/R treatment in mice.TMK688 (30 mg/kg) or pranlukast (3 mg/kg) was administered orallyalone or 60 min before ischemia, whereas indomethacin was givensubcutaneously 30 min before ischemia. The animals were killedunder deep ether anesthesia after the 60-min reperfusion period, andthe gastric mucosa was isolated, weighed, and placed in a tubecontaining 100% methanol. Then, the tissues were homogenized by aPolytron homogenizer (IKA, Tokyo, Japan) and centrifuged at12,000g for 10 min at 4°C. After the supernatant of each sample hadbeen evaporated with N2 gas, the residue was resolved in assaybuffer and used for the determination of CysLTs. Concentrations ofCysLTs were measured using a CysLT enzyme immunoassay kit(Cayman Chemical, Ann Arbor, MI).

Analyses of Expression of 5-LOX and CysLT1R mRNAs byRT-PCR. Animals were killed under deep ether anesthesia 60 minafter I/R treatment in the absence or presence of indomethacin (5mg/kg) given subcutaneously 30 min before the ischemia, and thestomachs were removed, frozen in liquid nitrogen, and stored at�80°C before use. In some cases, the animals were given TMK688(30 mg/kg) or indomethacin (5 mg/kg) alone without I/R treatmentand killed 60 min later. Tissue samples were pooled from two tothree mice for extraction of total RNA, which was prepared by asingle-step acid phenol-chloroform extraction procedure by use ofSepasol RNA-I (Nacalai Tesque, Kyoto, Japan). Total RNA primed byrandom hexadeoxy ribonucleotide was reverse-transcribed with theSuperScript preamplification system (Invitrogen, Carlsbad, CA). Thesequences of sense and antisense primers for mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 5-LOX, and CysLT1R are

92 Nakamori et al.

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

listed in Table 1. An aliquot of the reverse transcription reactionproduct served as a template in 35 cycles of PCR with 1 min ofdenaturation at 95°C and 1 min of annealing at 68°C by Advantage-2PCR kit (BD Biosciences, Palo Alto, CA) on a thermal cycler. Aportion of the PCR mixture was electrophoresed in a 1.5% agarosegel in Tris-EDTA-acetic acid buffer, and the gel was stained withethidium bromide and photographed. Images were analyzed with theImageJ version 1.39 (National Institutes of Health, Bethesda, MD),and the semiquantitative measurement of mRNA expression waspresented as a ratio compared with GAPDH.

Determination of Gastric Acid Secretion. Acid secretion wasmeasured in pylorus-ligated mice. Animals were deprived of food for18 h and water for 2 h before the operation. Under light etheranesthesia, the abdomen was opened and the pylorus was ligated.Then, the animals were then allowed to recover from the anesthesia.TMK688 or pranlukast was given intraduodenally immediately afterpylorus ligation. Two hours after pylorus ligation, the animals weresacrificed with an overdose of ether, and the gastric contents werecollected and analyzed for volume and acidity. Acidity was measuredby automatic titration of the gastric contents against 0.01 M/l NaOH,pH 7.0, with a titrator (Comtite COM-550; Hiranuma, Tokyo, Ja-pan). Gastric acid output (volume � acidity) was expressed as mi-croequivalents per 1 h. Control animals were administered vehiclealone.

Preparation of Drugs. The drugs used were urethane (TokyoKasei, Tokyo, Japan), TMK688 (Banyu Pharm. Co. Ltd., Tokyo, Japan),pranlukast (4-oxo-8-[4-(4-phenylbutoxy)benzoyl amino]-2)-2-(tetrazol-5-yl)-4H-1-benzopyran hemi-hydrate; Ono Pharm. Co. Ltd., Osaka, Ja-pan), montelukast (2-[1-[[(1R)-1-[3-[2-(7-chloroquinolin-2-yl)ethenyl-]phenyl]-3-[2-(2-hydroxypropan-2-yl)phenyl]propyl]sulfanylmethyl]cyclopropyl]acetic acid; Cayman Chemical), and indomethacin (Sigma-Aldrich). Indomethacin, TMK688, pranlukast, and montelukast weresuspended in a hydroxypropylcellulose solution (Wako Pure Chemicals,Osaka, Japan). Each agent was prepared immediately before use andadministered orally, intraduodenally, or subcutaneously in a volume of

0.1 ml/20 g body weight. Control animals received hydroxypropylcellu-lose as the vehicle.

Statistical Analysis. Data are presented as the mean � S.E. forfive to eight rats per group. Statistical analyses were performedusing a one-way analysis of variance and Student’s t test or Dun-nett’s multiple comparison test where appropriate. Values of P 0.05 are considered significant.

ResultsEffects of TMK688 and Pranlukast on I/R-Induced

Gastric Damage in Mice. Laparotomy without clamping ofthe gastric artery (sham operation) did not produce any dam-age in the gastric mucosa. In the animals subjected to I/Rtreatment (30 min of ischemia followed by reperfusion for 60min), however, multiple hemorrhagic lesions were observedin the gastric mucosa, with the lesion score being 11.2 � 1.4mm2 (Fig. 1A). Ischemia for 30 min did not induce any mac-roscopically visible damage in the mucosa (data not shown).Pretreatment of the animals with TMK688 (3–30 mg/kg p.o.),a 5-LOX inhibitor, dose-dependently prevented the I/R-in-duced development of gastric lesions, with the inhibition at30 mg/kg being 49.1%, which is significant compared with thecontrol.

To further investigate the involvement of CysLTs in thepathogenesis of I/R-induced gastric lesions, we examined theeffect of pranlukast, a CysLT1R antagonist, on the gastriculcerogenic response to I/R. As shown in Fig. 1B, the severityof the gastric lesions was reduced by the prior administrationof pranlukast (0.1–10 mg/kg p.o.), in a dose-dependent manner,and a significant effect was observed at 1 mg/kg or greater, withthe inhibition at 1, 3, and 10 mg/kg being 44.4, 72.2, and 74.0%,respectively. A significant protection was similarly obtained bymontelukast (3 mg/kg), another CysLT1R antagonist, the inhi-bition being 73.6% (3.9 � 0.7 versus 14.8 � 1.8 mm2).

In a sham-operated animal without I/R treatment, no dam-age was detected even by histological observation (Fig. 2A).By contrast, severe damage was observed histologically inthe stomach after I/R treatment; most of the damage wasrestricted to the surface epithelium, but some damage oc-curred deep in the mucosa, extending to the region of pits andglands (Fig. 2B). When the animal was pretreated withTMK688 (30 mg/kg p.o.) or pranlukast (3 mg/kg p.o.), theseverity of the histological damage was markedly reduced,

TABLE 1PCR primer sequences and product size

Molecule Primer Sequence ProductSize

bpGAPDH Forward, 5�-AACGACCCC TTCATTGAC-3� 191

Reverse, 5�-TCCACGACATACTCAGCAC-3�CysLT1R Forward, 5�-CAACGAACTATCCACCTTCACC-3� 163

Reverse, 5�-AGCCTTCTCCTAAAGTTTCCAC-3�5-LOX Forward, 5�-CCCCCGAAGCTCCCAGTGACC-3� 482

Reverse, 5�-TCCCGGGCCTTAGTGTTGATA-3�

bp, base pairs.

Fig. 1. Effects of TMK688 (A) and pranlukast(B) on I/R-induced gastric lesions in mice. Un-der urethane anesthesia, the celiac arterywas clamped (ischemia), and reperfusion wasachieved 30 min later by removal of theclamp. After a 60-min reperfusion period, gas-tric lesions were evaluated. TMK688 (3–30mg/kg) or pranlukast (0.1–10 mg/kg) wasgiven orally 60 min before ischemia. Data arepresented as the mean � S.E. for five to eightmice. �, P 0.05, significant difference fromcontrol.

Role of CysLTs in I/R-Induced Gastric Injury 93

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

and only slight damage was observed in the surface epithe-lium (Fig. 2, C and D).

Effects of TMK688 and Pranlukast on I/R-InducedGastric Damage in Mice Pretreated with Indometha-cin. It has been shown that COX inhibition stimulates theproduction of LTs by inducing a shift in balance from protec-tive PGs to proulcerogenic CysLTs (Peskar, 1991) and thatI/R-induced gastric injury was markedly aggravated by prioradministration of indomethacin, a nonselective COX inhibi-tor (Kotani et al., 2006). We therefore examined the effects of5-LOX inhibitor and CysLT1R antagonist on I/R-induced gas-tric ulceration in the mice pretreated with indomethacin.

When the stomach was subjected to I/R in the presence ofindomethacin (5 mg/kg s.c.), the severity of gastric lesionswas significantly increased, with the lesion score being20.3 � 2.3 mm2, approximately 2.5 times greater than that(8.3 � 1.5 mm2) for I/R alone (Fig. 3). This effect of indometh-acin was significantly abrogated by cotreatment with eitherTMK688 (30 mg/kg p.o.) or pranlukast (3 mg/kg p.o.), withthe inhibition being 40.4 and 70.0%, respectively.

Effects of TMK688 and Pranlukast on MPO Activityin Mouse Stomachs after I/R. It is known that I/R-inducedgastric injury is accompanied by an increase in neutrophilrecruitment in the mucosa (Kotani et al., 2006, 2007). In thepresent study, the severity of the I/R-induced gastric lesionswas decreased by prior administration of TMK688 and pran-lukast. To confirm that the mucosal inflammatory responseduring I/R is also affected by these treatments, we examinedthe effects of TMK688 and pranlukast on the changes inMPO activity in the gastric mucosa after I/R.

Gastric mucosal MPO activity in sham-operated mice wasless than 0.02 �mol H2O2/min/mg protein. The MPO activitywas markedly increased after I/R, with the values being0.042 � 0.007 �mol H2O2/min/mg protein (Fig. 4). This re-

sponse was significantly abrogated by the prior administra-tion of pranlukast (3 mg/kg p.o.), with the inhibition being59.5%. TMK688 (30 mg/kg p.o.) also suppressed the increaseof MPO activity induced by I/R, although the effect was notstatistically significant.

We further examined the effects of TMK688 and pran-lukast on the changes in MPO activity in the gastric mucosaafter I/R in the presence of indomethacin. The mucosal MPO

Fig. 4. Effects of TMK688 and pranlukast on changes in gastric mucosalMPO activity induced by I/R in mice. Under urethane anesthesia, theceliac artery was clamped (ischemia), and then reperfusion was achieved30 min later by removal of the clamp. After a 60-min reperfusion period,gastric MPO activity was measured. TMK688 (30 mg/kg) or pranlukast (3mg/kg) was given orally 60 min before ischemia. Data are presented asthe mean � S.E. for five to six mice. Significant difference at �, P 0.05from sham and #, from control (I/R alone).

Fig. 2. Microscopical observation of the gastric mucosa subjected to shamand I/R treatment. Under urethane anesthesia, the celiac artery wasclamped (ischemia), and reperfusion was achieved 30 min later by re-moval of the clamp. After a 60-min reperfusion period, the stomachs wereexcised. TMK688 (30 mg/kg) or pranlukast (1–10 mg/kg) was given orally60 min before ischemia. Shown are sham treatment (A), I/R treatment(B), I/R plus TMK688 (30 mg/kg p.o.; C), and I/R plus pranlukast (3 mg/kgp.o.; D). Scale bar, 100 �m.

Fig. 3. Effects of TMK688 and pranlukast on I/R-induced gastric lesionsin mice pretreated with indomethacin. Under urethane anesthesia, theceliac artery was clamped (ischemia), and reperfusion was achieved 30min later by removal of the clamp. After a 60-min reperfusion period,gastric lesions were evaluated. Indomethacin (5 mg/kg) was given subcu-taneously 30 min before ischemia, whereas TMK688 (30 mg/kg) or pran-lukast (3 mg/kg) was given orally 60 min before. Data are presented asthe mean � S.E. for five to seven mice. Significant difference at �, P 0.05 from control and #, from vehicle.

94 Nakamori et al.

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

activity in control mice was 0.044 � 0.006 �mol H2O2/min/mg protein when determined 60 min after I/R treatment(Fig. 5). The I/R-induced increase of MPO activity was mark-edly enhanced in the animals pretreated with indomethacin(5 mg/kg s.c.), with the values being 0.11 � 0.013 �molH2O2/min/mg protein. This response was significantly sup-pressed by the prior administration of TMK688 (30 mg/kgp.o.) or pranlukast (3 mg/kg p.o.), with the inhibition being56.4 and 58.2%, respectively.

Changes in Mucosal CysLT Content of Mouse Stom-ach after I/R with or without Indomethacin. To furtherinvestigate the role of CysLTs in the I/R-induced develop-ment of gastric lesions, we measured the mucosal CysLTlevels in the mouse stomach in the absence or presence ofindomethacin.

The levels of CysLTs in the gastric mucosa were notchanged after I/R treatment compared with the sham oper-ation, with the values being 3.3 � 0.2 ng/mg tissue in thesham-operated group and 2.4 � 0.3 ng/mg tissue in theI/R-treated group. Pretreatment with TMK688 (30 mg/kgp.o.) inhibited the production of CysLTs in the gastric mucosasubjected to I/R treatment and significantly decreased thelevels to approximately half of control values (Fig. 6). Pre-treatment of the animals with indomethacin (5 mg/kg s.c.),however, significantly increased the mucosal levels of Cys-LTs to 7.0 � 1.3 ng/mg tissue under I/R, and this responsewas almost totally suppressed by cotreatment with TMK688(30 mg/kg), with the inhibition being 90.6%. Pranlukast (3mg/kg p.o.) had no effect on the mucosal CysLT levels in thestomach after I/R treatment, in the absence or presence ofindomethacin.

Expression of CysLT1R and 5-LOX mRNAs in MouseStomach after I/R. GAPDH mRNA, the housekeeping gene,was clearly detectable in the stomach of control mice, and its

expression remained unaffected following I/R. AlthoughCysLT1R was expressed in the gastric mucosa subjected tothe sham operation, the expression was markedly up-regu-lated after I/R treatment when examined 1 h after reperfu-sion (Fig. 7). Ischemia alone had no effect on the CysLT1Rexpression (data not shown). Indomethacin (5 mg/kg s.c.)alone slightly up-regulated the expression of CysLT1R in thegastric mucosa but did not further enhance the increasedexpression after I/R.

5-LOX mRNA was also observed in the normal stomachafter the sham operation and remained unchanged after I/Rtreatment when examined 1 h after reperfusion. Ischemiaalone did not affect the 5-LOX expression in the gastricmucosa (data not shown). However, the mucosal expressionof 5-LOX under I/R was significantly increased by pretreat-ment of the animals with indomethacin (5 mg/kg) comparedwith the control mouse stomach subjected to I/R alone (Fig.8). It was also found that indomethacin by itself caused amarked up-regulation of 5-LOX expression in the stomach,and the increased expression by indomethacin was not fur-ther enhanced by additional I/R treatment. TMK688 (30mg/kg p.o.) did not affect the expression of 5-LOX as well asCysLT1R in the gastric mucosa with or without I/R treatment(data not shown).

Effects of TMK688 and Pranlukast on Gastric AcidSecretion. The acid output in normal mouse stomachs was1.00 � 0.12 �Eq/h when determined by a 2-h pylorus ligation(Table 2). Neither TMK688 (30 mg/kg i.d.) nor pranlukast (3mg/kg i.d.) had any effect on the basal acid output in mice.

DiscussionThe supply of blood to ischemic tissues paradoxically exac-

erbates the injury process and leads to the production of freeradicals and proinflammatory mediators, and the attractionof inflammatory cells infiltrating the tissues (Farber et al.,1981; Piper et al., 2003). Although there is a substantial body

Fig. 5. Effects of TMK688 and pranlukast on changes in gastric mucosalMPO activity induced by I/R in mice pretreated with indomethacin.Under urethane anesthesia, the celiac artery was clamped (ischemia),and then reperfusion was achieved 30 min later by removal of the clamp.After a 60-min reperfusion period, gastric MPO activity was measured.Indomethacin (5 mg/kg s.c.) was given 30 min before ischemia, whereasTMK688 (30 mg/kg) or pranlukast (3 mg/kg) was given orally 60 minbefore. Data are presented as the mean � S.E. for five to six mice.Significant difference at �, P 0.05 from control (I/R alone) and #, fromvehicle (I/R plus indomethacin).

Fig. 6. Changes in mucosal CysLT levels in the mouse stomach subjectedto sham and I/R treatment. Under urethane anesthesia, the celiac arterywas clamped (ischemia), and then reperfusion was achieved 30 min laterby removal of the clamp. After a 60-min reperfusion period, the mucosalCysLT levels were measured. Indomethacin (5 mg/kg) was given subcu-taneously 30 min before ischemia, whereas pranlukast (3 mg/kg) orTMK688 (30 mg/kg) was given orally alone or 60 min before. Data arepresented as the mean � S.E. for six to eight mice. Significant differenceat $, P 0.05 from sham; �, from control; and #, from vehicle.

Role of CysLTs in I/R-Induced Gastric Injury 95

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

of experimental data characterizing the factors that reduceor enhance the severity of I/R-induced gastric lesions, theroles of CysLTs in the pathogenesis remain unknown. Thepresent study demonstrated that I/R markedly up-regulatedCysLT1R expression in the gastric mucosa with a slight in-crease of CysLT production and that both a 5-LOX inhibitor anda CysLT1R antagonist suppressed the development of gastriclesions in response to I/R, strongly suggesting the involvementof CysLTs as a pathogenic factor in I/R-induced gastric damage.In addition, we observed a marked aggravation of these lesionson pretreatment with indomethacin, consistent with our previ-ous observations (Takeuchi et al., 1986; Kotani et al., 2006), andwe further showed the underlying mechanism of this responseto be associated with the up-regulation of 5-LOX expression andCysLT production in the stomach.

The arachidonic acids released by phospholipase A2 serveas substrates for the production of a group lipid mediatorsknown as LTs, which induce proinflammatory signaling

through the activation of CysLTs and LTB4 receptors. 5-LOXis the key enzyme in the biosynthesis of LTs from arachidonicacids. LTs, potent mediators of inflammatory and allergicreactions, are locally released from leukocytes and other5-LOX-expressing cells, and they exert their effects by bind-ing to specific membrane receptors (Maekawa et al., 2002).CysLT receptors exist in two subtypes, CysLT1R andCysLT2R, and they are expressed in various tissues andinflammatory cells (Lynch et al., 1999; Sarau et al., 1999;Brink et al., 2003). Sala and Folco (2001) reported that aselective CysLT1R antagonist reduced the severity of etha-nol- or nonsteroidal anti-inflammatory drug-induced gastriclesions by inhibiting neutrophil adhesion, suggesting thepathogenic importance for CysLT1R. Certainly, CysLT2Rmay also play a role in the pathogenesis of gastric lesions.However, none of the selective CysLT2R antagonists is avail-able for in vivo studies. In the present study, we thus focusedon CysLTs/CysLT1R and investigated the effects of a 5-LOXinhibitor and a selective CysLT1R antagonist on I/R-inducedgastric damage.

First, we found that both pranlukast, a CysLT1R antago-nist, and TMK688, a 5-LOX inhibitor, significantly reducedthe severity of I/R-induced gastric lesions. CysLTs are report-edly known to increase vascular permeability, neutrophilinvasion, and inflammatory cytokine production (Sala andFolco, 2001; Maekawa et al., 2002; Profita et al., 2008). Be-cause I/R-induced damage is a neutrophil-dependent re-sponse (Zimmerman and Granger, 1990), it is assumed thatthe protective effect of pranlukast is brought about, at least

TABLE 2Effects of TMK688 or pranlukast on gastric acid secretion in normalmiceAcid secretion was measured in pylorus-ligated mice for 2 h under normal conditions.TMK688 (30 mg/kg) or pranlukast (3 mg/kg) was given intraduodenally immediatelyafter the pylorus ligation. Data are presented as the mean � S.E.

Group Dose No. of Mice Gastric Acid Output

mg/kg � Eq/h

Control 5 1.00 � 0.12Pranlukast 3 5 0.88 � 0.30TMK688 30 5 1.01 � 0.50

Fig. 7. A, mucosal expression of CysLT1R mRNA in themouse stomach given indomethacin alone or subjected toI/R treatment in the absence or presence of indometha-cin. Indomethacin (IM; 5 mg/kg) was given subcutane-ously 30 min before ischemia. The expression ofCysLT1R mRNA was analyzed by RT-PCR 2 h afterindomethacin alone or 1 h after reperfusion. B, densito-metric quantitation was performed with ImageJ soft-ware, and the results are expressed as the ratio ofCysLT1R to GAPDH. Data are presented as the mean �S.E. for five mice. Significant difference at �, P 0.05from sham and #, from indomethacin alone.

Fig. 8. A, mucosal expression of 5-LOX mRNA in themouse stomach given indomethacin alone or sub-jected to I/R treatment in the absence or presence ofindomethacin. Indomethacin (IM; 5 mg/kg) wasgiven subcutaneously 30 min before ischemia. Theexpression of 5-LOX mRNA was analyzed by RT-PCR 2 h after indomethacin alone or 1 h after reper-fusion. B, densitometric quantitation was performedwith ImageJ software, and the results are expressedas the ratio of 5-LOX to GAPDH. Data are presentedas the mean � S.E. for five mice. Significant differ-ence at �, P 0.05 from sham and #, from control.

96 Nakamori et al.

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

partly, by the suppression of these inflammatory responsessuch as neutrophil infiltration, through the blockade ofCysLT1R. Anderson et al. (2009) reported that montelukast,a CysLT1R antagonist, inhibits the proinflammatory actionof neutrophils in vitro. In the present study, we confirmed theneutrophil infiltration in the stomach during I/R treatment,as represented by a marked increase of MPO activity, andfurther observed that pranlukast as well as TMK688 atten-uated the increase of MPO activity after I/R treatment. Theseall results strongly suggest the involvement of CysLTs/CysLT1R in the pathogenesis of I/R-induced gastric injury.We also found that the expression of CysLT1R mRNA in thegastric mucosa was significantly increased under I/R condi-tions, although the mucosal CysLT content remained un-changed before and after I/R treatment. In other organs suchas brain and kidney, the expression of CysLT1R mRNA wasreportedly up-regulated by ischemia alone or I/R treatment(Fang et al., 2007; Matsuyama et al., 2008). It is assumedthat the increased sensitivity to CysLT in the mouse stomachunder I/R conditions is, at least in part, accounted for by theup-regulation of CysLT1R expression, although I/R did notincrease CysLT production in the stomach. By contrast, in-domethacin alone slightly up-regulated the CysLT1R expres-sion but did not further enhance the increased expressioncaused by I/R, excluding a possibility of the up-regulation ofCysLT1R responsible for the aggravating mechanism by in-domethacin of the I/R-induced gastric injury.

We reported previously that I/R caused gastric lesions withthe up-regulation of COX-2 expression and that the damagewas significantly aggravated by indomethacin as well as theselective COX-2 inhibitor rofecoxib but not the selective COX-1inhibitor SC-560, suggesting the involvement of COX-2/PGs inmucosal defense during I/R (Kotani et al., 2006). Because I/Rinjury is a neutrophil-dependent response (Zimmerman andGranger, 1990), it is assumed that selective COX-2 inhibitorspromote the adherence of leukocyte to the vascular endotheliumduring I/R, thereby resulting in aggravation of the lesions in thestomach (Muscara et al., 2000). Similar results were reportedby Maricic et al. (1999) who showed that I/R-induced gastricdamage was aggravated by the administration of selective COX-2inhibitors such as NS-398 and 5,5-dimethyl-3-(3-fluorophenyl)-4-(4-methylsulfonyl)phenyl-2(5H)-furanone. Because several stud-ies showed that CysLT production in the stomach or the smallintestine was increased by pretreatment with indomethacin(Wallace et al., 1990; Nishio et al., 2007), it is possible that themetabolism of arachidonic acid leads to the production of LTsvia 5-LOX when another pathway to PGs is blocked by COXinhibition. In the present study, we observed that both CysLTproduction and 5-LOX mRNA expression in the stomach weresignificantly increased in the presence of indomethacin underI/R conditions. It should also be noted that indomethacin alone,but not I/R, markedly up-regulated the expression of 5-LOX inthe gastric mucosa. Moreover, the coadministration of TMK688,a 5-LOX inhibitor, markedly suppressed the increase of CysLTproduction and the aggravation of I/R-induced gastric lesionsunder indomethacin pretreatment, without any effect on theexpression of CysLT1R. These results together with previousfindings (Kotani et al., 2006) suggest that the aggravation byindomethacin of I/R-induced gastric damage is brought aboutby not only a decrease of PG production but also an increase of5-LOX expression/CysLT production. At present, however, the

mechanism for how indomethacin up-regulates 5-LOX expres-sion in the stomach remains unknown.

Although a selective COX-2 inhibitor, rofecoxib, increasedthe severity of I/R-induced gastric injury, similar to indo-methacin (Kotani et al., 2006), the mechanisms for aggrava-tion seem to differ in these cases. The worsening effect ofthese two agents was significantly prevented by the coadmin-istration of prostacyclin (Kotani et al., 2006). We have re-cently found that the aggravation by rofecoxib, but not indo-methacin, of I/R-induced gastric injury was significantlyabrogated by ozagrel, the inhibitor of thromboxane A2 (TXA2)synthesis as well as seratrodast, the TXA2 receptor antago-nist (Nakamori et al., 2009). These results suggest that theworsening effect of rofecoxib may be partly accounted for bythe imbalance of TXA2/prostacyclin ratio in the stomach andis apparently different from that of indomethacin, which maybe related to the increase of CysLT production and 5-LOXexpression, in addition to PG deficiency.

It has been reported that gastric acid secretion was sub-stantially decreased after ischemia and remained decreasedfor several hours even after reperfusion (Takeuchi et al.,1986; Nakamoto et al., 1998). However, Kitano et al. (1997)reported that cimetidine, a histamine H2 receptor antagonist,had a protective effect against I/R-induced gastric damagethrough the suppression of acid secretion. In a preliminarystudy, we observed that lansoprazole, a proton pump inhib-itor, also significantly reduced the severity of I/R-inducedgastric damage in mice, suggesting the partial involvementof gastric acid in the pathogenesis of these lesions. In thepresent study, however, because basal gastric acid secretionin mice was significantly affected by neither pranlukast norTMK688, the protective effects of these agents against I/R-induced gastric ulceration were accounted for by actionsother than the inhibition of acid secretion. These resultssupport an idea that the prophylactic effects of pranlukast orTMK688 may be brought about by the suppression of theinflammatory responses such as neutrophil infiltration andcytokine production during I/R, through the inhibition ofCysLT production.

Given the findings of the present study, we concluded thatI/R treatment up-regulated the expression of CysLT1R in thestomach and that CysLTs play a crucial role in the patho-genesis of I/R-induced gastric lesions through the activationof CysLT1R. In addition, the aggravation by indomethacin ofI/R-induced gastric injury may be brought about by the in-crease of CysLT production as well as the up-regulation of5-LOX expression, in addition to the decreased PG produc-tion. The present study cannot exclude a possibility thatCysLT2R may play a role in the gastric ulcerogenic responseto I/R, in addition to CysLT1R. Further study using a selec-tive CysLT2R antagonist is needed to clarify this point.

ReferencesAnderson R, Theron AJ, Gravett CM, Steel HC, Tintinger GR, and Feldman C (2009)

Montelukast inhibits neutrophil pro-inflammatory activity by a cyclic AMP-dependent mechanism. Br J Pharmacol 156:105–115.

Boros M, Kaszaki J, and Nagy S (1989) Oxygen free radical-induced histaminerelease during intestinal ischemia and reperfusion. Eur Surg Res 21:297–304.

Brink C, Dahlen SE, Drazen J, Evans JF, Hay DW, Nicosia S, Serhan CN, ShimizuT, and Yokomizo T (2003) International Union of Pharmacology XXXVII. Nomen-clature for leukotriene and lipoxin receptors. Pharmacol Rev 55:195–227.

Chamoun F, Burne M, O’Donnell M, and Rabb H (2000) Pathophysiologic role ofselectins and their ligands in ischemia reperfusion injury. Front Biosci 5:E103–E109.

Cheung JY, Bonventre JV, Malis CD, and Leaf A (1986) Calcium and ischemicinjury. N Engl J Med 314:1670–1676.

Role of CysLTs in I/R-Induced Gastric Injury 97

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from

Dengiz GO, Odabasoglu F, Halici Z, Suleyman H, Cadirci E, and Bayir Y (2007)Gastroprotective and antioxidant effects of amiodarone on indomethacin-inducedgastric ulcers in rats. Arch Pharmacol Res 30:1426–1434.

Fang SH, Zhou Y, Chu LS, Zhang WP, Wang ML, Yu GL, Peng F, and Wei EQ (2007)Spatio-temporal expression of cysteinyl leukotriene receptor-2 mRNA in rat brainafter focal cerebral ischemia. Neurosci Lett 412:78–83.

Farber JL, Chien KR, and Mittnacht S Jr (1981) Myocardial ischemia: the patho-genesis of irreversible cell injury in ischemia. Am J Pathol 102:271–281.

Gronert K, Martinsson-Niskanen T, Ravasi S, Chiang N, and Serhan CN (2001)Selectivity of recombinant human leukotriene D(4), leukotriene B(4), and lipoxinA4 receptors with aspirin-triggered 15-epi-LXA(4) and regulation of vascular andinflammatory responses. Am J Pathol 158:3–9.

Kanaoka Y and Boyce JA (2004) Cysteinyl leukotrienes and their receptors: cellulardistribution and function in immune and inflammatory responses. J Immunol173:1503–1510.

Kitano M, Wada K, Kamisaki Y, Nakamoto K, Kishimoto Y, Kawasaki H, and ItohT (1997) Effects of cimetidine on acute gastric mucosal injury induced by ischemia-reperfusion in rats. Pharmacology 55:154–164.

Kobata A, Kotani T, Komatsu Y, Amagase K, Kato S, and Takeuchi K (2007) Dualaction of nitric oxide in the pathogenesis of ischemia/reperfusion-induced mucosalinjury in mouse stomach. Digestion 75:188–197.

Konaka A, Nishijima M, Tanaka A, Kunikata T, Kato S, and Takeuchi K (1999)Nitric oxide, superoxide radicals and mast cells in pathogenesis of indomethacin-induced small intestinal lesions in rats. J Physiol Pharmacol 50:25–38.

Kotani T, Kobata A, Nakamura E, Amagase K, and Takeuchi K (2006) Roles ofcyclooxygenase-2 and prostacyclin/IP receptors in mucosal defense against isch-emia/reperfusion injury in mouse stomach. J Pharmacol Exp Ther 316:547–555.

Krawisz JE, Sharon P, and Stenson WF (1984) Quantitative assay for acute intes-tinal inflammation based on myeloperoxidase activity: assessment of inflamma-tion in rat and hamster models. Gastroenterology 87:1344–1350.

Kunikata T, Araki H, Takeeda M, Kato S, and Takeuchi K (2001) Prostaglandin Eprevents indomethacin-induced gastric and intestinal damage through differentEP receptor subtypes. J Physiol Paris 95:157–163.

Lynch KR, O’Neill GP, Liu Q, Im DS, Sawyer N, Metters KM, Coulombe N, Abramo-vitz M, Figueroa DJ, Zeng Z, et al. (1999) Characterization of the human cysteinylleukotriene CysLT1 receptor. Nature 399:789–793.

Maekawa A, Austen KF, and Kanaoka Y (2002) Targeted gene disruption reveals therole of cysteinyl leukotriene 1 receptor in the enhanced vascular permeability ofmice undergoing acute inflammatory responses. J Biol Chem 277:20820–20824.

Maricic N, Ehrlich K, Gretzer B, Schuligoi R, Respondek M, and Peskar BM (1999)Selective cyclo-oxygenase-2 inhibitors aggravate ischaemia-reperfusion injury inthe rat stomach. Br J Pharmacol 128:1659–1666.

Matsuyama M, Funao K, Kawahito Y, Sano H, Chargui J, Touraine JL, Takemoto Y,Nakatani T, and Yoshimura R (2008) Study of cysteinyl leukotriene-1 receptor inrat renal ischemia-reperfusion injury. Transplant Proc 40:2149–2151.

Moos MP, Mewburn JD, Kan FW, Ishii S, Abe M, Sakimura K, Noguchi K, ShimizuT, and Funk CD (2008) Cysteinyl leukotriene 2 receptor-mediated vascular per-meability via transendothelial vesicle transport. FASEB J 22:4352–4362.

Murakami M, Asagoe K, Dekigai H, Kusaka S, Saita H, and Kita T (1995) Productsof neutrophil metabolism increase ammonia-induced gastric mucosal damage. DigDis Sci 40:268–273.

Muscara MN, Vergnolle N, Lovren F, Triggle CR, Elliott SN, Asfaha S, and WallaceJL (2000) Selective cyclo-oxygenase-2 inhibition with celecoxib elevates bloodpressure and promotes leukocyte adherence. Br J Pharmacol 129:1423–1430.

Nakamori Y, Komatsu Y, Kotani T, Amagase K, and Takeuchi K (2009) Pathogenicimportance of thromboxane A2 in ischemia/reperfusion-induced gastric injury inmice. Gastroenterology 136 (Suppl 1): w1559.

Nakamoto K, Wada K, Kitano M, Kishimoto Y, Ashida K, Kamisaki Y, Kawasaki H,and Itoh T (1998) The role of endogenous acid in the development of acute gastriculcer induced by ischemia-reperfusion in the rat. Life Sci 62:PL63–PL69.

Nishio H, Hayashi Y, Terashima S, and Takeuchi K (2007) Protective effect ofpranlukast, a cysteinyl-leukotriene receptor 1 antagonist, on indomethacin-induced small intestinal damage in rats. Inflammopharmacology 15:266–272.

Perry MA, Wadhwa S, Parks DA, Pickard W, and Granger DN (1986) Role of oxygenradicals in ischemia-induced lesions in the cat stomach. Gastroenterology 90:362–367.

Peskar BM (1991) Lipoxygenase products in gastric damage and protection. AdvProstglandin Thromboxane Leukot Res 21B:753–760.

Piper HM, Meuter K, and Schafer C (2003) Cellular mechanisms of ischemia-reperfusion injury. Ann Thorac Surg 75:S644–S648.

Profita M, Sala A, Bonanno A, Siena L, Ferraro M, Di Giorgi R, Montalbano AM,Albano GD, Gagliardo R, and Gjomarkaj M (2008) Cysteinyl leukotriene-1 receptoractivation in a human bronchial epithelial cell line leads to signal transducer andactivator of transcription 1-mediated eosinophil adhesion. J Pharmacol Exp Ther325:1024–1030.

Sala A and Folco G (2001) Neutrophils, endothelial cells, and cysteinyl leukotrienes:a new approach to neutrophil-dependent inflammation? Biochem Biophys ResCommun 283:1003–1006.

Sarau HM, Ames RS, Chambers J, Ellis C, Elshourbagy N, Foley JJ, Schmidt DB,Muccitelli RM, Jenkins O, Murdock PR, et al. (1999) Identification, molecularcloning, expression, and characterization of a cysteinyl leukotriene receptor. MolPharmacol 56:657–663.

Sener G, Sehirli O, Toklu H, Ercan F, and Alican I (2007) Montelukast reducesischaemia/reperfusion-induced bladder dysfunction and oxidant damage in the rat.J Pharm Pharmacol 59:837–842.

Sener G, Sehirli O, Velioglu-Ogunc A, Cetinel S, Gedik N, Caner M, Sakarcan A, andYegen BC (2006) Montelukast protects against renal ischemia/reperfusion injuryin rats. Pharmacol Res 54:65–71.

Stein B, Pfenninger E, Grunert A, Schmitz JE, and Hudde M (1990) Influence ofcontinuous haemofiltration on haemodynamics and central blood volume in exper-imental endotoxic shock. Intensive Care Med 16:494–499.

Steinhilber D (1999) 5-Lipoxygenase: a target for anti-inflammatory drugs revisited.Curr Med Chem 6:71–85.

Suzuki K, Araki H, Mizoguchi H, Furukawa O, and Takeuchi K (2001) ProstaglandinE inhibits indomethacin-induced gastric lesions in rats through EP-1 receptors.Digestion 63:92–101.

Takamatsu Y, Shimada K, Chijiiwa K, Kuroki S, Yamaguchi K, and Tanaka M(2004) Role of leukotrienes on hepatic ischemia/reperfusion injury in rats. J SurgRes 119:14–20.

Takeuchi K, Ohno T, and Okabe S (1986) Variations of gastric transmucosal poten-tial difference and lesion formation during hemorrhagic shock in the rat. Gastro-enterology 91:1113–1122.

Takeuchi K, Tanaka A, Kato S, Aihara E, and Amagase K (2007) Effect of (S)-4-(1-(5-chloro-2-(4-fluorophenyoxy)benzamido)ethyl) benzoic acid (CJ-42794), a selec-tive antagonist of prostaglandin E receptor subtype 4, on ulcerogenic and healingresponses in rat gastrointestinal mucosa. J Pharmacol Exp Ther 322:903–912.

Tanaka A, Hatazawa R, Takahira Y, Izumi N, Filaretova L, and Takeuchi K (2007)Preconditioning stress prevents cold restraint stress-induced gastric lesions inrats: roles of COX-1, COX-2 and PLA2. Dig Dis Sci 52:478–87.

Wada K, Kamisaki Y, Kitano M, Kishimoto Y, Nakamoto K, and Itoh T (1996) A newgastric ulcer model induced by ischemia-reperfusion in the rat: role of leukocyteson ulceration in rat stomach. Life Sci 59:PL295–PL301.

Wallace JL, McKnight GW, Keenan CM, Byles NI, and MacNaughton WK (1990)Effects of leukotrienes on susceptibility of the rat stomach to damage and inves-tigation of the mechanism of action. Gastroenterology 98:1178–1186.

Yasue N, Chan ET, Kaplowitz N, and Guth PH (1992) Effect of phorone and allo-purinol on ischemia-reperfusion injury in gastrointestinal mucosa of the rat.Pharmacology 44:334–343.

Yoshikawa T, Ueda S, Naito Y, Takahashi S, Oyamada H, Morita Y, Yoneta T, andKondo M (1989) Role of oxygen-derived free radicals in gastric mucosal injuryinduced by ischemia or ischemia-reperfusion in rats. Free Radic Res Commun7:285–291.

Zhang WP, Hu H, Zhang L, Ding W, Yao HT, Chen KD, Sheng WW, Chen Z, and WeiEQ (2004) Expression of cysteinyl leukotriene receptor 1 in human traumaticbrain injury and brain tumors. Neurosci Lett 363:247–251.

Zimmerman BJ and Granger DN (1990) Reperfusion-induced leukocyte infiltration:role of elastase. Am J Physiol 259:H390–H394.

Address correspondence to: Dr. Koji Takeuchi, Division of PathologicalSciences, Department of Pharmacology and Experimental Therapeutics, KyotoPharmaceutical University, Misasagi, Yamashina, Kyoto 607-8414, Japan.E-mail: takeuchi@mb.kyoto-phu.ac.jp

98 Nakamori et al.

at ASPE

T Journals on A

ugust 23, 2018jpet.aspetjournals.org

Dow

nloaded from