Effect of ischemic acute renal damage on the expression of COX-2and oxidative stress-related elements in rat kidneySandra Villanueva, Carlos Cespedes, Alexis A. Gonzalez, Carlos P. Vio, and Victoria VelardeDepartamento de Fisiolog a, Ponticia Universidad Catolica de Chile, Santiago, ChileSubmitted 30 August 2006; accepted in nal form 15 January 2007VillanuevaS, CespedesC, GonzalezAA, VioCP, VelardeV.EffectofischemicacuterenaldamageontheexpressionofCOX-2andoxidative stress-relatedelements inrat kidney. AmJ PhysiolRenal Physiol 292: F1364F1371, 2007. First published January 23,2007; doi:10.1152/ajprenal.00344.2006.Acuterenal failure(ARF)is a clinical syndrome characterized by deterioration of renal functionover aperiodof hours or days. Theprincipal causes of ARFareischemic and toxic insults that can induce tissue hypoxia. Transcrip-tionalresponsestohypoxiacanbeinammatoryoradaptivewiththe participation of the hypoxia-inducible factor 1 and the expres-sionofspecicgenesrelatedtooxidativestress. Theproductionofperoxynitrites and protein nitrotyrosylation are sequelae of oxidativestress.Inseveralclinicalandexperimentalconditions,inammatoryresponses have been related to cyclooxygenase (COX)-2, suggestingthat its activation might play an important role in the pathogenesis andprogressionofnephropathiessuchasARF.Inthekidney,reninandbradykininparticipateontheregulationofCOX-2synthesis.Withthehypothesis that in ARF there is an increase in the expression of agentsinvolved in adaptive and inammatory responses, the distribution patternand abundance of COX-2, its regulators renin, kallikrein, bradykinin B2receptor, and oxidative stress elements, heme oxygenase-1 (HO-1), eryth-ropoietin (EPO), inducible nitric oxide synthase (iNOS), and nitrotyrosy-lated residues were studied by immunohistochemistry and immunoblotanalysisinrat kidneysafterbilateral ischemia. InkidneyswithARF,important initial damage was demonstrated by periodic acid-Schiff stain-ing and by the induction of the damage markers -smooth muscle actinand ED-1. Coincident with the major damage, an increase in the abun-dance of EPO, HO-1, and iNOS and an increase in renin and bradykininB2 receptor were observed. Despite the B2 receptor induction, weobservedanimportant decreaseinCOX-2intheischemic-reperfusedkidney. These results suggest that COX-2 does not participate in inam-matory responses induced by hypoxia.cyclooxygenase-2; acute renal failureACUTE RENAL FAILURE (ARF) is a clinical syndrome that isassociated with high morbidity and mortality rates (30). ARFhas an initiating phase, characterized by a reduced renal bloodow, that causes epithelial and vascular cell injury and a rapiddecreaseinglomerular ltrationrate; theinjurydisruptstheability of renal tubular epithelial cells and renal vascularendothelialcellstomaintainnormalrenalfunction, initiatingsignalingcascadesthatcontributetoinammationandorgandysfunction (34). The initiating phase is immediately followedby an extension phase, which is characterized by the followingtwomajor events: persistent hypoxia andinammatoryre-sponse. During the extension phase, multiple interrelated pro-cesses exacerbate epithelial and endothelial cell injury and celldeath,primarilyinthecorticomedullaryregionofthekidney(20). The maintenance phase is a period of injury stabilizationduringwhichcorrectiveeventsfacilitatecellularrepair, divi-sion, and redifferentiation (34).Acutetubularnecrosis(ATN)withprerenal diseaseisthemost commoncause of ARF, accountingfor two-thirds ofintrinsic causes (8). The major cause of ATNis hypoxiainducedbyischemia-reperfusion(I/R),whichcanbeinducedbyclinical conditions, suchashemorrhagicshockor sepsis(18). Hypoxia could increase oxidative stress in acute renal I/R(24). Several data show that hydroxyl radical-like activities aregenerated from peroxynitrites. This latter compound is emerg-ing as one of the important sequelae of oxidative andnitrosativestresses. Peroxynitriteisproducedbythereactionbetweennitricoxide(NO)andsuperoxide. Inaddition, tran-scriptional responses to hypoxia can be adaptive or inamma-tory (36). Adaptive responses are controlled primarily throughthe nuclear accumulation of the heterodimeric hypoxia-induc-ible factor (HIF-1) that regulates the expression of a numberof adaptive genes coding for angiogenic, glycolytic, and otherproteins, supportingtissue survival inhypoxia (6). Thesetarget genes, particularlyerythropoietin(EPO) andhemeoxygenase-1 (HO-1), are central to both local and systemicresponses to hypoxia and to cellular responses to alteredglucose, energy metabolism, and probably oxidative stress.Westenfelder et al. (41) havedemonstratedthat theEPOreceptor is expressedthroughout thekidney, particularlyintubular epithelial cells, mesangial cells, and the glomerulus. Inaddition, it has been demonstrated that EPO stimulates endo-thelialcellmitogenesisandangiogenesis(3), whichimprovetissueoxygenation.ThisevidencesupportsaroleforEPOinprotecting the kidney against I/R injury.On the other hand, HO-1 is part of the integrated response tooxidativestress. Theexpressionofthisproteinincreasesinin-ammatory cells and may be associated with the resolution phaseof acute inammation. Invarious models of oxidative tissueinjuries, theinductionofHO-1confersprotectionfromfurtherdamages by removing the prooxidant heme group or by virtue ofthe antioxidative, anti-inammatory, and/or antiapoptotic actionsof one or more of the products of heme metabolism, i.e., carbonmonoxide, biliverdin, bilirubin, and iron by the HO reaction (13).However,thelevels,togetherwiththelocalizationofEPOandHO-1, have not been evaluated in any I/R model.The kidney produces several vasoactive substances that playimportant rolesinthemaintenanceof renal bloodowandglomerular ltration. The renin-angiotensin system(RAS),with the production of ANG II, has important vasoconstrictor,antinatriuretic, and antidiuretic effects. On the other hand, theAddress for reprint requests and other correspondence: S. Villanueva,Departamento de Fisiolog a, Ponticia Universidad Catolica de Chile, Casilla114-D, Santiago, Chile (e-mail: svillanu@bio.puc.cl).The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.Am J Physiol Renal Physiol 292: F1364F1371, 2007.First published January 23, 2007; doi:10.1152/ajprenal.00344.2006.0363-6127/07 $8.00 Copyright 2007 the American Physiological Society http://www.ajprenal.org F1364kallikrein-kininsystem(KKS),withtheproductionofbrady-kinin (BK), induces vasodilation, natriuresis, and diuresis. BKeffects are mediated by prostaglandins derived from cyclooxy-genases (COX). COX-1 is found in the glomerulus and afferentarteriole, whereas COX-2 is expressed in podocytes, thickascending limb of the loop of Henle, macula densa, andafferent arteriole. Upregulation of COX-2 has been describedinseveral clinical andexperimental conditionscharacterizedbyinammation, suggestingthat activationofCOX-2mightplay an important role in the pathogenesis and progression ofnephropathies, such as ARF. In addition, although pharmaco-logical inhibition of COX-2 can protect from I/R (9), it can alsoleadtoseriousadverseeventsinkidney, includingARFandhyperkalemia(5). Wehypothesizethat inamodel of ARFinducedbyI/R, whichis characterizedbya brief hypoxiccondition, the levels of several vasoactive agents found in thekidneyarealteredbecauseofadecreaseinkallikreinlevelsevenafterseveraldaysofreperfusion.Thiswouldbeassoci-ated with an increase in RAS and with a specic regulation ofCOX-2andnitricoxidesynthase(NOS).Forthisreason,weanalyzedtherelativeabundanceof COX-2, induciblenitricoxide synthase (iNOS), renin, kallikrein, and related elementsof oxidative stress at different time periods after reperfusion.MATERIALSANDMETHODSAnimals. Adult male Sprague-Dawley rats (220250 g, n 7 for eachI/R group: 24, 48, 72, and 96 h; n 5 for each sham-operated group)were housed in a 12:12-h light-dark cycle. Control and treated animalswereweighedatthetimeofinitiationofbilateralischemicinjuryandaftercompletionofexperiments. Theanimalshadfoodandwateradlibitum and were maintained at the university animal care facilities. Allexperimental procedures were in accordance with institutional and inter-national standardsforthehumanecareanduseoflaboratoryanimals(Animal Welfare Assurance Publication A5427-01, Ofce for Protectionfrom Research Risks, Division of Animal Welfare, National Institutes ofHealth). In this study, the protocol for the use of animals was reviewedand approved by the institutional and independent ethical committee ofthe Ponticia Universidad Catolica de Chile.Renal I/Rinjury. Anestablishedmodel of renal I/Rinjurythatresemblesstructuralandfunctionalconsequencesofrenalischemia,includingapoptotictubularepithelial cellswasperformed(4). Ani-mals were anesthetized with ketamine-xylazine (25:2.5 mg/kg ip), andbodytemperaturewasmaintainedat 37C. Bothkidneyswereex-posed by a ank incision, and both renal arteries were occluded withanontraumaticvascular clampfor 30min. Next, clampswerere-moved,renalbloodowwasreestablished,andbothincisionsweresutured. Rats were allowed to recover in a warm room with water andfoodadlibitum. Sham-operatedrats were submittedtothe sameFig. 1. Evidenceofhistological renal damageinhypoxickidneyinducedbyischemia-reperfusion(I/R). Immunohistochemistrywasperformedinkidneysamples obtained at 24 h after I/R (n 7 for each I/R group). A clear induction of renal damage markers such as macrophages (ED-1; B) or -smooth muscleactin(SMA;C)canbeobservedintheinterstitialspacefromtheinnerandoutermedulla. A:histologicaldamageevaluatedbyperiodicacid-Schiff(PAS)staining. Brush border, epithelial attening, and mitosis are shown, indicating the renal damage induced by I/R. Kidneys from sham rats were also used to detectthese markers (DF); n 5 for each control group. Scale bar 100 m. Arrows, localization of the markers.F1365 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.orgsurgical procedure and conditions, without clamping the renal arteries.Rats were killed under anesthesia (ketamine-xylazine) 24, 48, 72, and96 h after reperfusion; both kidneys were removed and processed forimmunohistochemistry and Western blotting.Tissue processing and immunohistochemical analysis. Tissue pro-cessing for immunohistochemical studies in paraplast-embedded sec-tions was carried out according to methods previously described (40).Immunolocalization studies were performed using an indirectimmunoperoxidasetechnique, aspreviouslydescribed(40). Briey,tissue sections were dewaxed, rehydrated, rinsedin0.05MTris-phosphate-saline buffer (pH7.6), andincubatedwiththe primaryantibodyovernight at 22C. Afterward, sectionswerewashedthreetimes for 5 min each, followed by a 30-min incubation at 22C withthecorrespondingsecondaryantibodyandwiththeperoxidase-anti-peroxidase (PAP) complex. Immunoreactive sites were revealed using0.1%(wt/vol)3,3-diaminobenzidineand0.03%(vol/vol)hydrogenperoxide solution.Antibodies and chemicals. The following primary antibodies wereused: rabbit polyclonal antibodies against renin, B2receptor, andkallikrein were prepared as described previously (37). The monoclo-nal antibodies against macrophages (clone ED-1) were obtained fromBiosource (Camarillo, CA), against -smooth muscle actin (-SMA;clone 1A4) from Sigma (St. Louis, MO), and against iNOS (catalog39120) from Transduction Laboratories (Lexington, KY); the poly-clonal antibodiesagainst COX-2(catalog160126)wereobtainedfromCayman(AnnArbor,MI),againstEPO(catalogN-19)fromSantaCruzBiotechnology(SantaCruz, CA), against HO-1fromStressGen Biotechnologies (Victoria, Canada), and againstnitrotyrosine residues (PNT; clone 1A6) fromUpstate (Charlotts-ville, VA).Secondary antibodies and the corresponding PAP complexes werepurchasedfromICNPharmaceuticals-Cappel (Aurora, OH). TritonX-100, 3,3-diaminobenzidine, carrageenan, TrisHCl, hydrogen per-oxide, phosphate salts, andother chemicals were purchasedfromSigma.Immunoblotting. For immunoblotting analysis, inner/outer medullakidneysections (1mmthick) werehomogenizedwithanUltra-Turrax homogenizer in buffer containing 0.05 M EDTA, 1PBS, andproteaseinhibitorcocktail (Pierce, Rockford, IL). Theproteincon-centrationwasdeterminedthroughtheBradfordmethod(Bio-Rad,Richmond, CA). Westernblottingwas performedas describedbyHarlowandLane(11). Protein(60g) was mixedwithanequalvolume of SDS-PAGE sample buffer (100 mM TrisHCl, pH 6.8, 200mMdithiothreitol,4%SDS,0.2%bromphenolblue,and20%glyc-erol) andboiledfor 3min. Proteinswereseparatedon12%SDS-polyacrylamide gels and transferred to nitrocellulose membranes.Blockingwas carriedout byincubationinblockingsolution(8%nonfat dry milk in Tris-buffered saline-0.1% Tween) for 2 h at roomtemperature.Afterbeingblocked,membraneswereprobedwiththecorrespondingantibodyfor18hat 4C, washedwithTris-bufferedsaline-Tween, and incubated with the appropriate horseradish perox-idase-conjugatedsecondaryantibodyfor 1hat roomtemperature.Proteins were detected using enhanced chemiluminescence techniques(PerkinElmer, Life Sciences, Boston, MA).The blots were scanned, and densitometric analysis was performedusingthepublicdomainNational InstitutesofHealth(NIH)Imageprogramversion1.61(U.S. NIH, http://rsb.info.nih.gov/nih-image).The expression of tubulin was used to correct for variation in sampleloading.Determinationoftissuedamageandimmunohistochemicalquan-tication. Tissue damage was evaluated through periodic acid-Schiff(PAS) staining. Immunolocalization of ED-1 and interstitial-SMAwere used as markers of tissue damage.Fig. 2. Immunolocalization of oxidative stress-related elements in hypoxic kidney induced by I/R. Maximum staining intensity for erythropoietin (EPO; A), hemeoxygenase (HO-1; B), inducible nitric oxide synthase (iNOS; C), and nitrotyrosine residues (PNT; D) was observed at 24 h after I/R (n 7 for each I/R group).The staining for each marker was also performed at similar time periods in kidneys from sham-operated rats (DG); n 5 for each control group. The stainingfor HO-1 was localized in proximal tubules, iNOS was localized in papilla and inner medulla ducts, and EPO and PNT were localized mainly in the cortex ofthe kidney. Scale bar 100 m. Arrows, localization of the markers.F1366 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.orgThe immunoreactive area in each image was determined by imageanalysis using Simple PCI software (Compix). Total immunostained(brown) cells were averagedandexpressedas the meanabsolutevalues or the mean percentage of stained cell area per eld aspreviously described (38) with minor modications.Statistical analysis. Data fromdifferent groups were assessedwiththeparametricStudentst-test whencomparingtwogroupsand ANOVA for multiple comparisons with Fishers post hoc testwhen comparing more than two groups. The signicance level wasP0.05.RESULTSDeterminationof histological renal damage. PASstainingrenal sectionsafter theinduceddamageshowedclear alter-ations inkidneymorphologyconsistent withATN, suchasbrush borders and epithelial attening (Fig. 1A). The followingtwo indicators of renal damage were used after I/R: the pres-ence of macrophages (ED-1) and interstitial -SMA. Immuno-staining in postischemic rats showed an increase of bothmarkers in interstitial space (-SMA: 9.1 2.3 m2in I/R ratsvs. 2.7 1.7 m2in sham rats and ED-1: 42 5 m2in I/Rrats vs. 10 2 m2in sham rats; P 0.05; Fig. 1, B and C).Nochanges wereobservedinshamanimals (Fig. 1, DF).Additionally, in PAS-stained slides from24- to 48-h I/Rkidneys, an important number of mitosis could be observed inproximal tubular cells (Fig. 1A) that was not observed incontrol rats (Fig. 1D).Increasedlevels of EPO, HO-1, andiNOSat 24hafterischemia. We previously demonstrated that HIF-1 levelpeaked at 24 h after I/R (38). Thus we wanted to analyze theexpression pattern of the following three target genes of HIF-1: EPO, HO-1 (32), and iNOS(24). In normal kidneys,immunostainingfor iNOSwasobservedmainlyinmedullarstructures (Fig. 2G), and HO-1 was observed in proximaltubules (Figs. 2F and 3, C and D), similar to what was reportedpreviously (12, 28). In addition, kidneys from rats killed 24 hafter I/R showed an increased staining for both proteins com-pared with their controls (for HO-1 sham: 11,306 2,675 m2vs. I/R: 15,667 1,438m2, P not signicant). Althoughthe intensity was higher, the localization was similar to the oneobserved in normal kidneys (Fig. 2, B and C).AscanbeobservedinFig.4, AC,thelevelsofthethreeproteinsweremaximalat24h, returningtocontrollevelsat48 h (n 4, P 0.05). This increase was consistent with theincrease inHIF-1previouslyobserved(38) andcouldbeproduced in response to this transcription factor.Immunostaining for EPO was not observed in normal kidneys(Fig. 2D); on the other hand, EPO immunostaining was present indiscrete cells in kidneys fromI/Ranimals after 24 h (Fig. 2A). Thedistribution was mainly peritubular close to proximal tubules andloops of Henle, identied by their immunoreactivity to GP330 andTamm-Horsfall, respectively (Fig. 3, A, B, and D).After reperfusion (24 h), iNOS was increased and returnedtocontrol levels at 48h(Fig. 4A). EPOshowedasimilarbehavior (Fig. 4B), returningtocontrol levelsat 72hafterreperfusion. Similarly, HO-1 was maximally increased at 24 h,returning to control levels at 48 h (Fig. 4C).ProteinnitrosylationisincreasedinkidneysfromI/Rrats.Oneof theconsequencesof I/Ristheincreaseinoxidativestress. As mentioned before, iNOS levels were increased afterFig. 3. Localizationof HO-1andEPOonkidney tubular structures. Sections werestained for Tamm-Horsfall protein as amarker of thick ascending loop of Henle (A),EPO(B), HO-1 (C), and GP330 (D) as amarker of proximal tubules. Scalebar100m. Arrows, proximal tubules; arrowheads,thick ascending loops of Henle. Brown stain-ing indicates positive reaction for eachprotein.F1367 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.org24 h of I/R, so it was possible to expect an increase in the freeradical peroxynitrite. Thiscompoundreactswithproteinstoformnitrotyrosineresidues.Todeterminetheamountofper-oxynitrite, weevaluatedthepresenceofnitrotyrosilatedpro-teins. As observedinFig. 2D, nitrotyrosine reactivitywaspresent inalmost all tubularstructuresandwasincreasedinkidneys from I/R animals after 24 h (14,792 2,793 m2; P 0.05) compared with control (7,025 1,229 m2; n 5; Fig.2H). In addition, when evaluated by Western blot, the patternof nitrosylated proteins was different at 24 h after reperfusioncompared with control, although the levels were not changed.In addition, protein nitrosylation was decreased at 48 h withoutreturning to control levels, even at 96 h, suggesting an activedegradation and resynthesis of proteins (Fig. 4D).Thelevelsofcomponentsofvasoactivesystemsaremodi-ed in I/R. For the kidney to have a proper function, the KKSand the RASmust be in balance (29). We evaluated thedistribution and levels of certain components of these systemsin kidneys after I/R (Figs. 5 and 6).At24hafterI/R,immunostainedtissuesectionsshowedadecreased amount of immunoreactive cells to COX-2, but theintensitywas similar (Fig. 5, AandE). Inaddition, renin-immunoreactive cells were increased (Fig. 5, B and F) on theafferent arteriole, and kallikrein immunoreactivity was de-creased not only in cell number but also in intensity (Fig. 5, CandG). Finally, BKB2receptor immunoreactivitywas in-creased in kidneys 24 h after I/R(1,299 141 m2vs.74960 m2; P 0.036; Fig. 5, D and H).When analyzed by Western blot, we observed that 24 h afterreperfusion renin was increased and returned to control levelsat 48h(P0.05; Fig. 6A). Interestingly, COX-2, whichis expressedonlyinafewcellsincontrol conditions, wasdecreased at 24 h (Fig. 6B) without returning to control levels,even at 96 h after reperfusion. Kallikrein was maintainedrelatively constant up to 72 h, where its levels decreased (Fig.6C). Finally, the B2 receptor levels showed a biphasic behav-ior, slightlyincreasingat 24h, returningtonormal level at4872 h, but increasing again at 96 h (Fig. 6D).DISCUSSIONThemajor ndingsof thisstudyarethat, althoughEPO,HO-1, and iNOS are increased 24 h after ischemia and returntobasal 48hlater, correlatingwiththepreviouslyobservedbehavior for HIF-1, components of the renin-angiotensin, thekallikrein-kinin, andtheCOX-prostaglandinsystemshaveadifferentialregulationintheI/Rmodelduringthesametimeperiod.I/RinratsiswidelyutilizedtostudyARF.ARFinratsisreversible, andtherecoveryresponseischaracterizedbytherestoration of glomerular ltration rate and remodeling of therenal tubularsystem(4). However, therearepersistent alter-ations in renal function postischemic injury that lead to apermanent compromiseinurinary-concentratingabilityasso-ciated with a reduction in renal medullary tonicity and perma-nent vascular damage. We studied the effects of ischemia after24, 48, 72, and 94 h of reperfusion to followhowthesepersistent alterations areestablished. Weusedtwomethod-ological approaches, immunohistochemistry to evaluate theproteinlocalizationandWesternblottohaveasemiquanti-cation of the levels of each protein.Werst analyzedproteinsthat areregulatedbyhypoxia,such as HO-1, EPO, and iNOS, followed by the evaluation ofproteins that are involved in the regulation of urine concentra-tion and pressure regulation such as kallikrein, renin, andCOX-2.AlthoughthreeHOisoformshavebeenreported, onlytheinducible isoform, HO-1, and the constitutively expressedisoform, HO-2, have a bona de heme oxygenase activity (5).Inductionof HO-1occursasanadaptiveandbenecial re-Fig. 4. Immunoblot for oxidative stress-related elements in acute renal failure (ARF)induced by I/R. Oxidative stress markersiNOS (A), EPO(B), HO(C), and nitro-tyrosine (D) were analyzed in kidneys fromshamanimalsandat 24, 48, 72, and96hafter 30-min ischemia. Tubulin was used asloading control (n 4 for each group).*P0.05.F1368 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.orgsponsetoseveral injury-signalingprocesses, andit hasbeenimplicated in many clinically relevant disease states, includingacute renal injury. Increased HO-1 expression protects kidneysfromoxidative injuries (22), rhabdomyolysis (23), cisplatinnephrotoxicity(2), ARF(33), andI/R-mediatedtissueinjury(26), probably through the generation of reaction productsfrom heme degradation, iron, carbon monoxide, biliverdin, andin particular bilirubin, which exert important antioxidant, anti-Fig. 5. Immunolocalization of vasoactive substances in hypoxic kidney induced by I/R. Staining for cyclooxygenase (COX)-2 (A), renin (B), kallikrein (C), andB2 receptor (D) was observed at 24 h after ischemia (n 7 for each I/R group). The staining for each marker was also performed at similar time periods inkidneys from sham-operated rats (EH); n 5 for each control group. The expression of COX-2 and B2 was mainly observed in thick ascending limb, reninwas observed in afferent arteriole, and kallikrein was observed in connecting tubule cells localized in the cortex of the kidney. Scale bar 100 m. Arrows,marker localization.Fig. 6. Immunoblot for vasoactiveproteinsinARFinducedbyI/Randvasoactivepro-teins renin(A), COX-2(B), kallikrein(C),andB2receptor (D) wereanalyzedinkid-neys from sham animals and 24, 48, 72, and96 h after 30-min ischemia. Tubulin was usedas loadingcontrol; n4for eachgroup.*P 0.05.F1369 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.orginammatory, and cytoprotective functions (13). In our model,HO-1expressionafterI/RcorrelateswithHIF-1activation(38). The hypothesis that HIF-1 can induce HO-1 expressionis supported by results from Ockaili et al. (25), who observedan increase in HO-1 in rabbit hearts treated with HIF-1 activator,theprolylhydroxylaseinhibitordimethyloxalylglycine.Inaddi-tion, inHIF-1-decient Hepa-1cells, thehypoxia-inducedincreaseinHO-1wasabolished(15), conrmingthepartici-pation of HIF-1 on the induction of HO-1.A hypoxia-responsive enhancer (HRE) has been identied inthe 3-anking region of the Epo gene in a region required forEPOtranscriptional activation(19). The HREcontains theconsensussequencefor thebindingof HIF-1. It has beenproposed that Epo has protective effects in retina and renaltubule cells. The mechanismproposedfor this actionin-volves the activation of Jak-2, Akt, and multiple targets withantiapoptotic effects (31). In our study, we have observed anincrease in Epo immunoreactivity at 24 h by Western blot andimmunocytochemistry, which is consistent with the increase inHIF-1 observed previously. This increase could be related tothe molecular machinery elicited in the protection response tothe I/R damage.The promoter region of the iNOSgene also contains asequencehomologytotheHRE.Inmurinemacrophages,theinduction of HIF-1by the iron chelator desferrioxaminestimulatestheexpressionofthemRNAforiNOS. iNOShasbeenpostulatedas apromoter of apoptosis inI/R(39). Inaddition, several in vivo and in vitro investigations havesuggested that NO generated by iNOS contributes to renal I/Rinjury(7), possiblybythegenerationofperoxynitriteduetothereactionof NOwithsuperoxideradical, andconsequentprotein tyrosine nitrosylation. An increased production of NOforprolongedperiodsoftimeinmanypathological statesisknowntocontributetooxidativedamageof critical cellularmacromolecules, includingproteins, whichoftenshowele-vatedlevels of 3-nitrotyrosine (1, 14, 21, 27). The resultsshown by other authors in more aggressive models of I/R areconsistent withour data andsupport the idea that lackofoxygenisnotonlyinducingproteinsthatdamagethekidneybutalsotheinductionofotherproteinsthatprotecttherenaltissue. In this regard, we wanted to determine the effects of I/Ron the expression of vasoactive systems of the kidney.We previously reported that kallikrein was one of the genesdownregulated 35 days after recovery from bilateral I/R injury(4). In the present study, we observed a decrease in kallikreinas early as 72 h. Kallikrein has been involved in the protectionagainst I/R-induced myocardial injury (43). Kallikrein cleaveskininogentoproduceBK, whichbindstoitsreceptor (B2),activatingintracellular signaling. WhenBKconcentrationisincreased, the receptor is downregulated by internalizationfollowed by degradation. We speculate that the cyclic behaviorof theBKreceptor couldbebecauseof adecreaseinBK,produced by the decrease in kallikrein at longer periods, withthe consequent upregulation of the receptor.Matsuyama et al. (17) demonstrated that COX-2 was max-imallyexpressedina rat model at 4hafter I/Rinjury,decliningthereafter; but his observations were restrictedto24 h. In addition, it has been proposed that COX-1 and/or -2blockade ameliorates the renal tissue damage triggered by I/Rinjury(9). Althoughwedidnot repeat thetimepointsmea-suredbyMatsuyamaet al., wecanpostulatethat COX-2issensitive to tissue damage, so when tubules are well preserved,as happens in the rst hours after reperfusion, COX-2 can beinduced; but later, when a maximal tissue damage is observed,COX-2decreases.AdecreaseinCOX-2activitywithacon-comitant increase in HO-1 activity has also been observed in adifferent model such as endothelial cells (10, 16, 42), suggest-ing that the heme-HO system can function as a cellular regu-lator of the expression of COX-2.In summary, in kidneys with ARF, an important initialdamageisobservedat 24hafterI/R. Coincidentlywiththemajor damage, there is an increase in EPO, HO-1, and iNOS.On the other hand, components of the renal vasoactive systemssuchasrenin, kallikrein, BKB2receptor, andCOX-2havedissimilar behaviors. Althoughreninwas increased, theB2receptor had a biphasical response, whereas kallikrein andCOX-2decreasedintheI/Rkidney,suggestingadifferentialregulationoftheseproteinsduringthistypeofdamage, andcould explain, in part, the increased propensity to damage laterin life.ACKNOWLEDGMENTSWethankMariaAlcoholadofortechnical assistanceintissueprocessingand Dr. Ricardo Moreno for a critical reading of the manuscript.GRANTSThisworkwassupportedbyFondecyt Grants3050075(S. Villanueva),1050977 (C. P. Vio), and 1040809 (V. Velarde).REFERENCES1. Abe K, Pan LH, Watanabe M, Konno H, Kato T, Itoyama Y.Upregulationof protein-tyrosinenitrationintheanterior horncells ofamyotrophic lateral sclerosis. Neurol Res 19: 124128, 1997.2. Agarwal A, Balla J, Alam J, Croatt AJ, Nath KA. Induction of hemeoxygenaseintoxicrenalinjury:aprotectiveroleincisplatinnephrotox-icity in the rat. Kidney Int 48: 12981307, 1995.3. AshleyRA, DubuqueSH, DvorakB, WoodwardSS, WilliamsSK,KlingPJ. Erythropoietinstimulatesvasculogenesisinneonatalratmes-enteric microvascular endothelial cells. Pediatr Res 51: 472478, 2002.4. BasileDP, FredrichK, AlausaM, VioCP, LiangM, RiederMR,GreeneAS, CowleyJrAW. Identicationofpersistentlyalteredgeneexpressioninkidneyfollowingfunctionalrecoveryfromischemicacuterenal failure. Am J Physiol Renal Physiol 288: F253F263, 2005.5. BradenGL, OSheaMH, MulhernJG, GermainMJ. Acute renalfailureandhyperkalaemiaassociatedwithcyclooxygenase-2inhibitors.Nephrol Dial Transplant 19: 11491153, 2004.6. BunnHF, PoytonRO. Oxygensensingandmolecular adaptationtohypoxia. Physiol Rev 76: 839885, 1996.7. Chatterjee PK, Patel N, Kvale E, Cuzzocrea S, Brown P, Stewart N,Mota-FilipeH, ThiemermannC. Inhibitionof induciblenitricoxidesynthasereducesrenalischemia/reperfusioninjury. KidneyInt61:862871, 2002.8. Esson ML, Schrier RW. Diagnosis and treatment of acute tubularnecrosis. Ann Intern Med 137: 744752, 2002.9. Feitoza CQ, Camara NO, Pinheiro HS, Goncalves GM, Cenedeze MA,Pacheco-Silva A, Santos OF. Cyclooxygenase 1 and/or 2 blockadeameliorates the renal tissue damage triggered by ischemia and reperfusioninjury. Int Immunopharmacol 5: 7984, 2005.10. Haider A, Olszanecki R, Gryglewski R, Schwartzman ML, Lianos E,Kappas A, Nasjletti A, Abraham NG. Regulation of cyclooxygenase bythe heme-heme oxygenase system in microvessel endothelial cells. J Phar-macol Exp Ther 300: 188194, 2002.11. HarlowE, Lane D. Antibodies: ALaboratory Manual. Cold SpringHarbor, NY: Cold Spring Harbor Laboratory, 1998.12. Holmqvist B, Olsson CF, Svensson ML, Svanborg C, Forsell J, Alm P.Expression of nitric oxide synthase isoforms in the mouse kidney: cellularlocalization and inuence by lipopolysaccharide and Toll-like receptor 4.J Mol Histol 36: 499516, 2005.F1370 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.org13. Kirkby KA, Adin CA.Productsofhemeoxygenaseandtheirpotentialtherapeuticapplications. AmJPhysiol Renal Physiol 290: F563F571,2006.14. KooyNW, Lewis SJ, Royall JA, YeYZ, KellyDR, BeckmanJS.Extensive tyrosine nitration in human myocardial inammation: evidencefor the presence of peroxynitrite. Crit Care Med 25: 812819, 1997.15. Lee PJ, Jiang BH, Chin BY, Iyer NV, Alam J, Semenza GL, Choi AM.Hypoxia-inducible factor-1 mediates transcriptional activation of the hemeoxygenase-1 gene in response to hypoxia. J Biol Chem 272: 53755381,1997.16. Li Volti G, SetaF, SchwartzmanML, Nasjletti A, AbrahamNG.Heme oxygenase attenuates angiotensin II-mediated increase in cycloox-ygenase-2activityinhumanfemoralendothelialcells.Hypertension41:715719, 2003.17. Matsuyama M, Yoshimura R, Hase T, Kawahito Y, Sano H, NakataniT. Study of cyclooxygenase-2 in renal ischemia-reperfusion injury. Trans-plant Proc 37: 370372, 2005.18. MehtaR. Outcomesresearchinacuterenal failure. SeminNephrol 23:283294, 2003.19. MelilloG, Taylor LS, Brooks A, MussoT, CoxGW, VaresioL.Functionalrequirementofthehypoxia-responsiveelementintheactiva-tionoftheinduciblenitricoxidesynthasepromoterbytheironchelatordesferrioxamine. J Biol Chem 272: 1223612243, 1997.20. Molitoris BA, Sutton TA. Endothelial injury and dysfunction: role in theextension phase of acute renal failure. Kidney Int 66: 496499, 2004.21. Moriel P, Abdalla DS. Nitrotyrosine bound to beta-VLDL-apoproteins: abiomarker of peroxynitrite formation in experimental atherosclerosis.Biochem Biophys Res Commun 232: 332335, 1997.22. Morimoto K, Ohta K, Yachie A, Yang Y, Shimizu M, Goto C, TomaT, Kasahara Y, Yokoyama H, Miyata T, Seki H, Koizumi S. Cytopro-tective role of heme oxygenase (HO)-1 in human kidney with various renaldiseases. Kidney Int 60: 18581860, 2001.23. NathKA, BallaG, Vercellotti GM, BallaJ, JacobHS, LevittMD,Rosenberg ME. Induction of heme oxygenase is a rapid, protectiveresponse in rhabdomyolysis in the rat. J Clin Invest 90: 267270, 1992.24. Noiri E, Nakao A, Uchida K, Tsukahara H, Ohno M, Fujita T,BrodskyS, GoligorskyMS. Oxidativeandnitrosativestress inacuterenal ischemia. Am J Physiol Renal Physiol 281: F948F957, 2001.25. Ockaili R, Natarajan R, Salloum F, Fisher BJ, Jones D, Fowler 3rdAA, Kukreja RC. HIF-1 activation attenuates postischemic myocardialinjury: role for heme oxygenase-1 in modulating microvascular chemokinegeneration. Am J Physiol Heart Circ Physiol 289: H542H548, 2005.26. Ogawa T, Nussler AK, Tuzuner E, Neuhaus P, Kaminishi M, MimuraY, BegerHG. Contributionof nitricoxidetotheprotectiveeffectsofischemicpreconditioninginischemia-reperfusedratkidneys.JLabClinMed 138: 5058, 2001.27. Petruzzelli S, Puntoni R, Mimotti P, PuleraN, BalivaF, Fornai E,Giuntini C. Plasma 3-nitrotyrosine in cigarette smokers. Am J Respir CritCare Med 156: 19021907, 1997.28. RodriguezF, KempR, BalazyM, Nasjletti A. Effectsof exogenoushemeonrenal function: roleof hemeoxygenaseandcyclooxygenase.Hypertension 42: 680684, 2003.29. SchmaierAH. Thekallikrein-kininandtherenin-angiotensinsystemshave a multilayered interaction. Am J Physiol Regul Integr Comp Physiol285: R1R13, 2003.30. SchrierRW, WangW, PooleB, MitraA. Acuterenalfailure:deni-tions, diagnosis, pathogenesis, and therapy. J Clin Invest 114: 514, 2004.31. Sharples EJ, Patel N, Brown P, Stewart K, Mota-Philipe H, Sheaff M,Kieswich J, Allen D, Harwood S, Raftery M, Thiemermann C,YaqoobMM. Erythropoietinprotectsthekidneyagainsttheinjuryanddysfunction caused by ischemia-reperfusion. J Am Soc Nephrol 15: 22402241, 2004.32. Sharples EJ, Thiemerrmann C, Yaqoob M. Mechanism of disease: celldeath in acute renal failure and emerging evidence for a protective role oferythropoietin. Nat Clin Pract 1: 8797, 2005.33. Shimizu H, Takahashi T, Suzuki T, Yamasaki A, Fujiwara T, OdakaY, Hirakawa M, Fujita H, Akagi R. Protective effect of heme oxygenaseinductioninischemicacuterenal failure. Crit CareMed28: 809817,2000.34. SuttonTA,FisherCJ,MolitorisBA.Microvascularendothelialinjuryand dysfunction during ischemic acute renal failure. Kidney Int 62:15391549, 2002.35. Takahashi T, MoritaK, Akagi R, SassaS. Protectiveroleof hemeoxygenase-1 in renal ischemia. Antioxid Redox Signal 6: 867877, 2004.36. TaylorCT, ColganSP. Therapeutictargetsfor hypoxia-elicitedpath-ways. Pharm Res 16: 14981505, 1999.37. Velarde V, Humphreys J, Figueroa CD, Vio CP. Postnatal maturationof tissuekallikrein-producingcells (connectingtubulecells) intheratkidney:amorphometricandimmunohistochemicalstudy. AnatEmbryol192: 407414, 1995.38. Villanueva S, Cespedes C, Vio CP. Ischemic acute renal failure inducestheexpressionof awiderangeof nephrogenicproteins. AmJPhysiolRegul Integr Comp Physiol 290: R861R870, 2006.39. Vin as JL, Sola A, Genesca M, Alfaro V, Pi F, Hotter G. NO and NOSisoformsinthedevelopmentofapoptosisinrenalischemia/reperfusion.Free Radic Biol Med 40: 9921003, 2006.40. VioCP, AnSJ, Cespedes C, McGiff JC, Ferreri NR. Inductionofcyclooxygenase-2inthickascendinglimbcellsbyadrenalectomy.JAmSoc Nephrol 12: 649658, 2001.41. WestenfelderC, BiddleDL, BaranowskiRL. Human, rat, andmousekidneycellsexpressfunctional erythropoietinreceptors. KidneyInt 55:808820, 1999.42. YangL, QuanS, Nasjletti A, Laniado-SchwartzmanM, AbrahamNG. Heme oxygenase-1 gene expression modulates angiotensin II-induced increase in blood pressure. Hypertension 43: 12211226, 2004.43. YinH, ChaoL, ChaoJ. Kallikrein/kininprotectsagainst myocardialapoptosisafterischemia/reperfusionviaAkt-glycogensynthasekinase-3andAkt-Bad.14-3-3signalingpathways. JBiol Chem280: 80228030,2005.F1371 EFFECT OF ARF ON COX-2 AND OXIDATIVE STRESS-INDUCED ELEMENTSAJP-Renal Physiol VOL292 MAY2007 www.ajprenal.org