.Joarnol af Nrurachoxurtry,3004.9,917-951 de1n,11110.1471-414304位4线x Lack of mitochondrial nitric oxide production in the mouse brain Zsomhor Lacza,+Thomas F.W.Hom,James A.Snipes,*Jie Zhang,*Sanjoy Roychowdhury, Eszter M.Ilorvath,Jorge P.Figueroa,+Mirk Kollai,Csaba Szabot and David W.Busija* Depariment of Phsiolo/Phurwueology Wake Foren Uiverairy School af Meicine,Winto-Slew,Nore Curolint,USA te of futom Phyriningy d Clnieal Eaperiaental Rer Srmmelweis Linirrity Badapst.Hugrry etitute fur Mfedicul Nouroliolug:Ono-von-Gueriche Universin Magdebwg.Germauy Abstract ct the whole brain NOS activity kvel,which may be attnbuted Based on our inhal firding that the nitnc ade (NO)sorsitve to ctramitochondrial contaminaticn Extensie immunoblot. fluorochrome daminofuuorescen (DAF)was localized to ting and immunoprecipitaton experiments faled to show the milochondria in oultured primary neurons,we investigaled presence of endothdlial neuronsl,or inducble NOS in mouse whcther bran mpochondria produce NO through a mto- brain mitochondris using a varcty of primary antibodes. chondrial NO syrthase (mNOS)onzyme.Isolated brain Arginino.calmoduln or 2,5-ADP atfinty purfcation protoools mitochondria ware loaded with DAF and subjecind to tlow successhuly conoontratod oNOS and nNOS trom tull brain cyfometry analysis.Nather the application of NOS inhbitars tissue hut tailed to show any signal in mtochandria.We nor the genetic cfsnaption of ather NOS gene ciminished the condude that mouse hrain mitochondria do not contain NOS DAF-llcrescence.Homever,perooyitrile 8c8venger8 isoforms,nor do they produce NO through a NOS-dependent roduced the mhochondral DAF tluorescenoe,indicating that mechansm. the DAF signal is not specitie to ND.Chomiluminascence Keywords:daminoducrescain,FP15.L-NAME mitochon detection in the head space gas and a Clark-type NO-sersi- drial nitric cride synthase,mitochondrion,pamxynitrite. thve dlectrode in the soltion faied to deledt NO release in Nawecham200490,942-951, brain milochondria.NOS adivity in milocndria was only 1% Nimric axide (ND)and pemxyntrre (ONOC)are considered mcochoedrial peoteins,inclodine moost elements of complex to be pachological fictors in a range of nerologieal disurdkers I (Murray ef af 20103).Beeause NO is involved i so many such as Pnrkinson's disease,Alzheimer's disease,mmltiple nspects of mitochondrinl function,it is an intriguing hypo sclemsis,stroke and amyotrophic lateral sclerosis (Hrown thesis that mitochoedna themselves are capable of NO 1997.Heales al 1999).The hasic mechanisms by which production. these nitmgen radicals induce celbalr damage involve the The existenee of a distinet mitochondrial NO synhuase inhibilion of the mlochundrial respirakry chain (Rubb ef af. (mINOS)tas bocn sugpeslad in several papers (Bales ef of. 1999.Muray e al.2003)Low NO concentrations effoct- ively compele with axygen at the binding sile of cytochrome c oxidase,which may serve as a pbysiological regulatory Beceived Docenber 18.200:nevised mmscript received March 19. mecanism (Bruwn 1999,2000)Also,cylochrome c osilse 204,3pld3L,2004 w酒shuwn lo be the cnzyme心ponsibk for the climination ddess comegpordece3 nd Tepeint水n乙arebor Lac刀， of NO in miochondria-rich cells (Pearce et al.2002). lipdilale uf Hatan Phyiuluay and Clnisdl Expcrittnal Roench, Another mitochondrial pathway of NO metabolism is its Seramelueis Unhersiy.0lM们h1G灯pet,Ien2y. E1udca面w用 nonenzymatic reaction with superoxide to form ONOO. Anesnoor ase:DAF-2-DA.4daminouorescen discerzte: This potent labile oxidant can also inhibit mitochondrial DAF-FM 4nis-S-mtlry hnin2'?'ddiflateuflaetcir DIC.dif respimtion in an ireversible manner at complexes ll nd Il ferentlal intererense comtrast ogris eNos.endothelial nme coode and it is involved in the induction of cell necrosis and aytle Hh,baglbi;iN0S.ndubl:ritrir uxitk:rtleor apoptosis (Liaudet et al.2000.Valdez er al.2000).Further LNAME,Naitro-toginine methyl ester L-NMMA.N.methyH. arginine:mNOS.mnochondrisl mitric coide symchase:nNOS.neuromal more,ONOO can nitrate the tyrosine residues of ntric uside xyuthase;N05,mrie uside xyathiac. 942 2004 Sucisy fur.I.(2004)90,942-951
Lack of mitochondrial nitric oxide production in the mouse brain Zsombor Lacza,*, Thomas F. W. Horn, James A. Snipes,* Jie Zhang,* Sanjoy Roychowdhury, Eszter M. Horva´th,*, Jorge P. Figueroa,* Ma´rk Kollai, Csaba Szabo´ and David W. Busija* *Department of Physiology/Pharmacology, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Budapest, Hungary Institute for Medical Neurobiology, Otto-von-Guericke University, Magdeburg, Germany Abstract Based on our initial finding that the nitric oxide (NO) sensitive fluorochrome diaminofluorescein (DAF) was localized to mitochondria in cultured primary neurons, we investigated whether brain mitochondria produce NO through a mitochondrial NO synthase (mtNOS) enzyme. Isolated brain mitochondria were loaded with DAF and subjected to flow cytometry analysis. Neither the application of NOS inhibitors nor the genetic disruption of either NOS gene diminished the DAF-fluorescence. However, peroxynitrite scavengers reduced the mitochondrial DAF fluorescence, indicating that the DAF signal is not specific to NO. Chemiluminescence detection in the head space gas and a Clark-type NO-sensitive electrode in the solution failed to detect NO release in brain mitochondria. NOS activity in mitochondria was only 1% of the whole brain NOS activity level, which may be attributed to extramitochondrial contamination. Extensive immunoblotting and immunoprecipitation experiments failed to show the presence of endothelial, neuronal, or inducible NOS in mouse brain mitochondria using a variety of primary antibodies. Arginine, calmodulin or 2,5-ADP affinity purification protocols successfully concentrated eNOS and nNOS from full brain tissue but failed to show any signal in mitochondria. We conclude that mouse brain mitochondria do not contain NOS isoforms, nor do they produce NO through a NOS-dependent mechanism. Keywords: diaminofluorescein, FP15, L-NAME, mitochondrial nitric oxide synthase, mitochondrion, peroxynitrite. J. Neurochem. (2004) 90, 942–951. Nitric oxide (NO) and peroxynitrite (ONOO– ) are considered to be pathological factors in a range of neurological disorders such as Parkinson’s disease, Alzheimer’s disease, multiple sclerosis, stroke and amyotrophic lateral sclerosis (Brown 1997; Heales et al. 1999). The basic mechanisms by which these nitrogen radicals induce cellular damage involve the inhibition of the mitochondrial respiratory chain (Robb et al. 1999; Murray et al. 2003). Low NO concentrations effectively compete with oxygen at the binding site of cytochrome c oxidase, which may serve as a physiological regulatory mechanism (Brown 1999, 2000). Also, cytochrome c oxidase was shown to be the enzyme responsible for the elimination of NO in mitochondria-rich cells (Pearce et al. 2002). Another mitochondrial pathway of NO metabolism is its nonenzymatic reaction with superoxide to form ONOO– . This potent, labile oxidant can also inhibit mitochondrial respiration in an irreversible manner at complexes II and III and it is involved in the induction of cell necrosis and apoptosis (Liaudet et al. 2000; Valdez et al. 2000). Furthermore, ONOO– can nitrate the tyrosine residues of mitochondrial proteins, including most elements of complex I (Murray et al. 2003). Because NO is involved in so many aspects of mitochondrial function, it is an intriguing hypothesis that mitochondria themselves are capable of NO production. The existence of a distinct mitochondrial NO synthase (mtNOS) has been suggested in several papers (Bates et al. Received December 18, 2003; revised manuscript received March 19, 2004; accepted March 31, 2004. Address correspondence and reprint requests to Zsombor Lacza, Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, U¨ llo¨i u´t 78/a, 1082 Budapest, Hungary. E-mail: zlacza@mac.com Abbreviations used: DAF-2-DA, 4,5-diaminofluorescein diacetate; DAF-FM, 4-amino-5-methylamino-2¢,7¢-difluorofluorescein; DIC, differential interference contrast optics; eNOS, endothelial nitric oxide synthase; Hb, hemoglobin; iNOS, inducible nitric oxide synthase; L-NAME, N-nitro-L-arginine methyl ester; L-NMMA, N-methyl-Larginine; mtNOS, mitochondrial nitric oxide synthase; nNOS, neuronal nitric oxide synthase; NOS, nitric oxide synthase. Journal of Neurochemistry, 2004, 90, 942–951 doi:10.1111/j.1471-4159.2004.02553.x 942 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
Minchondria do not produce ritc oxide 94G 1995:Gmulivi 1998:Cimulivi er af.1998:Tatoyan and Ciiulivi The purity of the mitochondrial peeparation was tested by turo 1998 Amaix al 1999:Gihafourifar sd Richter 1999: independent metbods.Frst electron microscopic observatioas Kanai e ad 2001:Laeza er of.2001:Boveris er uf 7002: shownd very littke rnnainttion nf the mitochredria prparartion Elfcring er ai.2002:Ribo er of.2002:Giulivi 2003:Schild hy bmoken mitnchondris nr lysoarmes Seoond.the purity of the ef of.2003).Some of the initial findings.however.were not prcpanun was tsessed by weslen bloging:The enduplasmic reproducible in other laboratories challenging the physiolo ecalum mker e山ticulin wi peewen i由whule tioue gical relevance of mtNOS (Brown 1999;Lacza e al.2003a). preparatone bit was roduced to msipificant amomts in the purified michoadria.In contrast,the mitochondnial marker cylochrome c The majority of the sluics imstigating mitochondria in the idaseas sigificantly enriched is the michna perpratis. conlext of'NO production were conducted on rat or mouse Special care was taken In enuir:the hext possible reitnchondrin liver preparations (Giulivi er al.1998.Taloyan and Giulivi prepartor and purity testing wus cpplied regularty %o maimale the 1998:Ghafourifar et a!1999.Locza et al.2001:Elfering 6 omparbility of山hee3as er al.2002:Locza er af.2003a:Schid e al.2003 Our previos experiments with liver mitochondri hmve shomn an C通cult up-regulation of mtNO follorwring a hrief hypoxic episode Printaty tal coelical teuruto were cullured fiet El8 Wilir rat indesting that mtNOS may play a role in the pathomech- etuses as deser bed peeviuly (Kis af af.2003)The cutical picces anism of hypoxia-reoygenation (Lacza er af.2001:Schild tom the fetuses were washed twice in Dulbocco's modified Eagle's er al.2005).In case this mechanism is also present in the medum suppkmented wih penicillin (Sigma,I间UL)国d stmpleeyein (Sim,I00 up'ml)then were incboned with dispise brain,one may hypothesize that it compnses a major factoe I(2 U/mL,Roche,Masnbeit,Gerany)for 35 min ut 37C.Cells in the pathogenests of neurodegenemtive disesses. weve dixsociuted by two series of geetle triteation ad plaled oto So far there are oely a few studies that have amempted ta poly olysine cooned coverslgs for comfocal micmoscogic analysis. identify brain mINOS and they rewalkal in cuntradicting dala. Afer cell attachment,the plating mediu was replaood by In an carly puper,Bales er al.(1995)decribed cNOS Neurubreil mediam (Giibeo BRL)supplmesled with B27 (Gibeu immunoreactivity in beain milochondria.Later.Riobo and BRL2%)L-glutmine(Siga05 mv)B-mereapocthandl (iibeu colleagues measured NO release fiom rat hrain mitochoedria BRL,55 uM)and potssium chonde (Sigma 25 mMk.Cultires and describod developmental changes in mNOS activity were gro4nt37℃鱼wmiified wmnoophese o3al山nag%COa: (Riobo e ol.2002).In cuntrast.Henrich ef of.(2002)found in air.ad the medtm was changnd om every third dry.Cultees NOS immunoructivity only in the ckese proximity of the cnreisted of mere than 9%of neimmes verified hy poetive miochondria.but not in the organelks themselves,ques imsusostiaing foe microtulrale-roucialed protein-2 (Bectun tioning the existence of mNOS in neural tissues.This Dickirun),t negiive immurtairing Sor glal fibrillary acidi proem (Chemicon.Temecala,CA.USA)Experiments were observation is sapported by a study of Rothe and colleagues peromed n 7-4.day-old cubures.a time perid dunng which in which they found NADPH diaphomase nctvity,a marker of netrens expeess fineticrul NMDA.AMPA and ksinale merpees NO prodaction,associabed with the outer membrane of the miochondria but not in the cristae (Rotse e al.1999).Direct Floocesceet confocal microscopy and tow cytometry measarements of mitochondrial NO production or attempts Freshly isled modri wee dbpersed in 'uero to purify mtNOS from the bmin have noc been publiseed in ing 125 mM KC1,2 mM Kl 5 mM MgCk.10 mM IIEPES. amy species or preparations Therefore,the peesent study 10 JM EGTA at pli 7J and planod on poly D-lysie coated aimed to investigate t也eexi对ence of mtNO5dmh cuverlies.Maxchondia wer eergil by te aditin of sdim ondrial NO production with several overfapping methods in malate mM)and sodium glutamate (5 mM)and were visualized the brain. usng a Zebss scanning confocal microscope (Auovert 100 M)with diffrrertisll inlerferrner cntrae apticx (DIC)and a funmserin rhndamine filter set (excitition:48%am,cmisonn 505-50 rm; excitalion:543 am,emission.560 n,repectively).The fluot- Experimental procedures oplioe:s usol were MioFluoe Rol (1 us)and 4-mino-s-cictlyl amino-2'.'-diftaoroftuarescen (DAP+M.J.Molecular Probes, Mitochoadria preparation Eupene,UR.USA)Mitochondral uorescence wis observed under All peooures wae uppuved by the Animl Cae and Ue curtrol conditons and is the pesence of ON0O snengers Comnmimoe of Wake Forest University.Mitochandria were peepared tetrakis 2-tricthykme glycol mnmnmetkyl ctherpyrid porpkyrin tom halthane aneshetized mouse brain using the discontiuous (2-T(PEG3)PyP)(FP15,100 yt,lnotek Corp,Beverly,MA,USAI Prreoll tradiest method a deazihed pevioudy (acza rr ai ISeubo av.2002:Loezz e av.2003:Pucher er af.2003)o 5.10, 200%)The sexpiration cnstrel rtio of the prrpininn was 15.20-tctrakis(4-snHfonainghemdporpkyrinato imr(lll)chlnride 494+0.4o起eμus6:e of ghtnate and ml山k(54each IFeTIP'S.I00 jv Calmochem.San Diego.CA.USA)Bckgmmd T C5?HUR CNOS KO 16 12912-Nod The following mruse strams were uoed in the cxp flucrecenoe uf the pmubes or the miluchondki slune was below .iNO路KO delectice lim 4,5-diamirafluorexccis dicctale (DAF-2-DA,7 pst, B6lP2-Nu2u4一；cumol fur te cfiN0sK0CS7ELy 6 000664:NOS KO B6.12954-Nus]r colrul foe the nNOS Molecular Probes)was used in cell cnloures (Lopez.Figacroa er ai. 2000)sd DAF-FM in isolatrd milnchnndria.wher the necrweiry KOC37 BLb J (Jacks国，r Harbor,E.UsA1. 2004 lalemrimnal Socicty fur Narochambtry,Mwuchemt (2004)90,942-051
1995; Giulivi 1998; Giulivi et al. 1998; Tatoyan and Giulivi 1998; Arnaiz et al. 1999; Ghafourifar and Richter 1999; Kanai et al. 2001; Lacza et al. 2001; Boveris et al. 2002; Elfering et al. 2002; Riobo et al. 2002; Giulivi 2003; Schild et al. 2003). Some of the initial findings, however, were not reproducible in other laboratories challenging the physiological relevance of mtNOS (Brown 1999; Lacza et al. 2003a). The majority of the studies investigating mitochondria in the context of NO production were conducted on rat or mouse liver preparations (Giulivi et al. 1998; Tatoyan and Giulivi 1998; Ghafourifar et al. 1999; Lacza et al. 2001; Elfering et al. 2002; Lacza et al. 2003a; Schild et al. 2003). Our previous experiments with liver mitochondria have shown an up-regulation of mtNOS following a brief hypoxic episode, indicating that mtNOS may play a role in the pathomechanism of hypoxia-reoxygenation (Lacza et al. 2001; Schild et al. 2003). In case this mechanism is also present in the brain, one may hypothesize that it comprises a major factor in the pathogenesis of neurodegenerative diseases. So far there are only a few studies that have attempted to identify brain mtNOS and they resulted in contradicting data. In an early paper, Bates et al. (1995) described eNOS immunoreactivity in brain mitochondria. Later, Riobo and colleagues measured NO release from rat brain mitochondria and described developmental changes in mtNOS activity (Riobo et al. 2002). In contrast, Henrich et al. (2002) found NOS immunoreactivity only in the close proximity of the mitochondria, but not in the organelles themselves, questioning the existence of mtNOS in neural tissues. This observation is supported by a study of Rothe and colleagues in which they found NADPH diaphorase activity, a marker of NO production, associated with the outer membrane of the mitochondria but not in the cristae (Rothe et al. 1999). Direct measurements of mitochondrial NO production or attempts to purify mtNOS from the brain have not been published in any species or preparations. Therefore, the present study aimed to investigate the existence of mtNOS and mitochondrial NO production with several overlapping methods in the brain. Experimental procedures Mitochondria preparation All procedures were approved by the Animal Care and Use Committee of Wake Forest University. Mitochondria were prepared from halothane anesthetized mouse brain using the discontinuous Percoll gradient method as described previously (Lacza et al. 2003b). The respiration control ratio of the preparation was 4.94 ± 0.46 in the presence of glutamate and malate (5 mM each). The following mouse strains were used in the experiments: wild type C57BL/6; eNOS KO B6.129P2-Nos3tm1UNC, iNOS KO B6.129P2-Nos2tm1Lau; control for the e/iNOS KO C57BL/ 6 000664; nNOS KO B6.129S4-Nos1tm1Plh; control for the nNOS KO C57BL/6 J (Jackson, Bar Harbor, ME, USA). The purity of the mitochondrial preparation was tested by two independent methods. First, electron microscopic observations showed very little contamination of the mitochondria preparation by broken mitochondria or lysosomes. Second, the purity of the preparations was assessed by western blotting: The endoplasmic reticulum marker calreticulin was present in the whole tissue preparations but was reduced to insignificant amounts in the purified mitochondria. In contrast, the mitochondrial marker cytochrome c oxidase was significantly enriched in the mitochondria preparations. Special care was taken to ensure the best possible mitochondria preparations and purity testing was applied regularly to maintain the comparability of the results. Cell culture Primary rat cortical neurons were cultured from E18 Wistar rat fetuses as described previously (Kis et al. 2003). The cortical pieces from the fetuses were washed twice in Dulbecco’s modified Eagle’s medium supplemented with penicillin (Sigma, 100 U/mL) and streptomycin (Sigma, 100 lg/mL) then were incubated with dispase I (2 U/mL, Roche, Mannheim, Germany) for 35 min at 37C. Cells were dissociated by two series of gentle trituration and plated onto poly D-lysine coated coverslips for confocal microscopic analysis. After cell attachment, the plating medium was replaced by Neurobasal medium (Gibco BRL) supplemented with B27 (Gibco BRL, 2%), L-glutamine (Sigma, 0.5 mM), b-mercaptoethanol (Gibco BRL, 55 lM) and potassium chloride (Sigma, 25 mM). Cultures were grown at 37C in humidified atmosphere containing 5% CO2 in air, and the medium was changed on every third day. Cultures consisted of more than 98% of neurons verified by positive immunostaining for microtubule-associated protein-2 (BectonDickinson), and negative immunostaining for glial fibrillary acidic protein (Chemicon, Temecula, CA, USA). Experiments were performed in 7–9-day-old cultures, a time period during which neurons express functional NMDA, AMPA and kainate receptors. Fluorescent confocal microscopy and flow cytometry Freshly isolated mitochondria were dispersed in K+ -buffer containing 125 mM KCl, 2 mM K2HPO4, 5 mM MgCl2, 10 mM HEPES, 10 lM EGTA at pH 7.0 and plated on poly D-lysine coated coverslips. Mitochondria were energized by the addition of sodium malate (5 mM) and sodium glutamate (5 mM) and were visualized using a Zeiss scanning confocal microscope (Axiovert 100 M) with differential interference contrast optics (DIC) and a fluorescein/ rhodamine filter set (excitation: 488 nm, emission: 505–530 nm; excitation: 543 nm, emission: > 560 nm, respectively). The fluorophores used were MitoFluorRed (1 lM) and 4-amino-5-methylamino-2¢,7¢-difluorofluorescein (DAF-FM, 7 lM, Molecular Probes, Eugene, OR, USA). Mitochondrial fluorescence was observed under control conditions and in the presence of ONOO– scavengers tetrakis 2-triethylene glycol monomethyl ether(pyridil porphyrin (2-T(PEG3)PyP) (FP15, 100 lM, Inotek Corp, Beverly, MA, USA) (Szabo et al. 2002; Lacza et al. 2003; Pacher et al. 2003) or 5, 10, 15, 20-tetrakis(4-sulfonatophenyl)porphyrinato iron(III) chloride (FeTPPS, 100 lM; Calbiochem, San Diego, CA, USA). Background fluorescence of the probes or the mitochondria alone was below detection limit. 4,5-diaminofluorescein diacetate (DAF-2-DA, 7 lM, Molecular Probes) was used in cell cultures (Lopez-Figueroa et al. 2000) and DAF-FM in isolated mitochondria, where the necessary Mitochondria do not produce nitric oxide 943 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
944 Z.Lacza er n. enzyetes to elave the disoetne fiom D.AF-2-DA are no preacut.In Wurld Precison lnstnenenb.Berlin.Gemamy)as deseribed earlier organells-fioe solutiun both DAF-2 tnd DAF-FM showed simu ISdhild e a.2003)The NO-semive electrode wus cbruted by sensievrity to various NO donors (not shonl. penenating soichineetrc NO standrds from dofned amtomnts of the Flow-cytometry was perfommod on freshby isoled mitochondris NO donoe SNAP injocted into fiod volrmes of I M CuCl:.The in K'-buffer in a simiar fadhiun lo the confocal measmemests. liweal linil of detoction was foand tu be i 16 nM SNAP which Furwand-scak,sik-atler ind凸e2 eenee (FLI-用wee3a- cumegosds in a total af 32 pmmal of No in a 2-ml.smle. ded by a BD-FacsCalibur tlow-cytomteter from 100 000 evemts in Measurertents of the NO ocncentration i e mitochondria coch peeparatiou..Miochondra-free baffer and butfer coutaining prepartons were perfocmed in the presece of different respiratory Pereoll were ted as negitive cnrtrele.Tlata were evalroned by the sbstrates (ax idicaled in the fimme legende)At the end of every CHIOsrP'ao mnftwart. experimet an ecess of the NO emmger oyhemogintin (h)was lka山eupk.Ouly if Hb eaoed t dectee i由由eOe teadng sach dilference wus cursiderrd tu be du:tu feee NO. No kvels in the mikchordnal samples of mouse beain tissue were deterinod iing an NJ-onalyzer INOA 280.SIEVERS.Boulder, NOS acthiry assay erihed (Taeta w al 2001)The ascry was carried For aralysis of froe NO in the sarpks.the peactiou vessel was o鱼daplicales,ong100 of isolapod michoadrie (protein tilled with wuer.The NO scarried fiout the seaction veel to mples were incabaled with 75 pl of C-bbeed -arginne in Tris buffer.incuhned for min at 325C. ally hbelel arginine(pecificuctivily 331 mCanml, s phokmalliplier.The isslnmeat was calbnbod by injectiun of New Eagland Nuclear,Boson.MA.USA)win parfol fhroug a Dowex iX (aoetate tom)column and diluted in bufter containig that generaned slochiomeri NO statdards from the reaction 66.7 mM Tris,27 mM NADPH and 3.33 gM calcum chlride at pH 74.Cofacter wrne added at a firodl cnrcentratien of 2 uM fre 2KN0+2X1+2s04→20++2H:0+2K:50 FMN.2 JM fre FAD :nd li pM fre RHa Arginoe activity was inibited by otrithine at a dooe of I aM Sepuratioet of irginine fiom cimulline was perfumoll by in-exchatge duomrogiaply with 知ax*0Wsm《s3diim3m,0×8+o,Signal After w7shg ncaction vessel (2 mL total vhme)was purpod with N:for 10 min the un the citie was collectd ad before the expcriment. meawered int a scintillition cnanler.The peoei cuncemration nf the Fur meseaing free NO in the mitochdria.feshly ialale nrigital pleswa meun by Bio-Rad DC protein aoay kil miochodria were dispersed is buffer (d125 mM KCl Bio-Rad Laborores,Heroules,CA USA)ond the enzyme octivity 2 mmol K.HPO.5 mM MaCl 10 M HEPES.I uM Cxtla al W色expuessed in pob of citrulline per mill国of protein pe pH 7.0)in presence of malme'glutamane (5 mM each)and ADP 0 min.Camodnl dependenoe was eahited hy the addition of (2mM)at a protein cnomtration of I mgml.Afer 5 min ad culnideolum ul a cucet山：un of I00 M.Backgoutd cuuts 20 min the smplas were transfemed to an air-ght eoyen-fice vial were meoured in the ubornce uf cofucles ud in the psnce of fulluwed bry mrourtieie of head spuce NO. IG间μNnimo-L-er知nire methy ester (L-NAME)and were To rule out自e possiblity that the bck of NUS o的cloes is respoasible forthe lack of NO prodaction we added such subslances In the miochendia spencon.Foe this pupene,K'ber (fo Wextern hle and immunoprecipitaiom cumpoditinn see Flureescent rnnfecal micmmnpy and dow cytum- Protein wis extrueed from miochondria by the addition of boilig etry)oas misod wah 100 av L-againe (Aleis Biocbenicals,San lyas bafTer (ooulainag 1%w I M Tris aid 1%wy sodian Diego,CA.USA)0.2 s NADPH (Sigm SL Louis,MO.LSA). dodecy对l.The smples were soncaed,heated at5℃fg 10 jrv temrahydnbinpeerin (11:Alexis Binchemicals)5 uM fwvin 3 mim and centnfupod3r20mint2g孩4℃.1h adene dimclootide FAD)5 po flavin adenine monomackonide superaatinl was ined fiur immmublotting.Peoein cunoetration [FXN)10 upinL cilmodulin (Alexis Biochemicals),10 uM CaCl wa measured hry a Bio-Ra DC prutein asery kit (Biu-Rady Equal in aeed vial coninng a linyisimer and purged with amounts of poooein were separtod on 24%to 20%i gradient min gel nitrogen gas to remove O Cne bundred microlrers of theshly IBio-Rad),and tardemed to polyvinyldene ditluocide membrine. isoad mitochondrial suspersion Iprotein cortent 10 mgi'mL)was Afer hlcking with 5%milk.priary antibndy uas applied dded mn this vial with a tight syringr (pnerinud with Na)Hrad mermghe fnilnued by horseradsh pemxcidoe conjmgated serndiry spce mmples (0f l)were drarwn from these viaks befnme. tlibudly.Chemnilaninscece was tol to viualize the bands. imolialdy ufer ail 30 afer mtocoudkial uldition uid mjected Molecular weiallt markers (Bio-Ra)were included ou cach lilot. ito the acactiun chumbes (filked wih wale)of a Sievetx Nuawin Specificity of the method uas tested by omitting the primary analyzer for detectiom of NO.Samples were analyzed in duplicates 2nio女om the pocd,c resued■the disappn of the spocific hane山 The fillewing antbodies wrre used:seti-nivencl NOS (puly- Ampemmetric NO measurement with an NO clectrede cloral Affinity Bioreuzents.Golen,CO,USA):anti-eNOS 1030- NO production in the miochoudrial preparons wus so mesured 1209 (monnclenal,Traredncrian Laboratrics,San Diegn.CA. usag an NO.sersitive Cik-type dlexmode (250-NO-METER. D204 aterrotirnal Soci与for Neumnchemat与y.了wehw2004地94-4l
enzymes to cleave the diacetate from DAF-2-DA are not present. In organelle-free solutions both DAF-2 and DAF-FM showed similar sensitivity to various NO donors (not shown). Flow-cytometry was performed on freshly isolated mitochondria in K+ -buffer in a similar fashion to the confocal measurements. Forward-scatter, side-scatter and fluorescence (FL1-H) were recorded by a BD-FacsCalibur flow-cytometer from 100 000 events in each preparation. Mitochondria-free buffer and buffer containing Percoll were used as negative controls. Data were evaluated by the CELLQUESTPRO software. Chemiluminescence NO detection NO levels in the mitochondrial samples of mouse brain tissue were determined using an NO-analyzer (NOA 280, SIEVERS, Boulder, USA) in conjunction with the computerized data analysis program NOAWIN as described before (Roychowdhury et al. 2001). For analysis of free NO in the samples, the reaction vessel was filled with water. The NO was carried from the reaction vessel to the analysis chamber by a steady flow of N2. Chemiluminescence that resulted from the reaction of ozone with NO was measured via a photomultiplier. The instrument was calibrated by injection of constant volumes of head space samples from a reaction mixture that generated stoichiometric NO standards from the reaction: 2KNO2 þ 2KI þ 2H2SO4 ! 2NO þ I2 þ 2H2O þ 2K2SO4 The lowest detection limit was found to be 50 pmol NO generated in 2 mL reaction mixture. To avoid any contamination of air, the reaction vessel (2 mL total volume) was purged with N2 for 10 min before the experiment. For measuring free NO in the mitochondria, freshly isolated mitochondria were dispersed in buffer (composed of 125 mM KCl, 2 mmol K2HPO4, 5 mM MgCl2, 10 M HEPES, 1 lM CaCl2 at pH 7.0) in presence of malate/glutamate (5 mM each) and ADP (2 mM) at a protein concentration of 1 mg/mL. After 5 min and 20 min the samples were transferred to an air-tight oxygen-free vial followed by measurement of head space NO. To rule out the possibility that the lack of NOS cofactors is responsible for the lack of NO production we added such substances to the mitochondria suspension. For this purpose, K+ buffer (for composition see Fluorescent confocal microscopy and flow cytometry) was mixed with 100 lM L-arginine (Alexis Biochemicals, San Diego, CA, USA), 0.2 mM NADPH (Sigma, St. Louis, MO, USA), 10 lM tetrahydrobiopterin (THB; Alexis Biochemicals), 5 lM flavin adenine dinucleotide (FAD), 5 lM flavin adenine mononucleotide (FMN), 10 lg/mL calmodulin (Alexis Biochemicals), 10 lM CaCl2 in a sealed vial containing a tiny magnetic stirrer and purged with nitrogen gas to remove O2. One hundred microliters of freshly isolated mitochondrial suspension (protein content 10 mg/mL) was added to this vial with a gas tight syringe (prerinsed with N2). Head space samples (100 lL) were drawn from these vials before, immediately after and 30 after mitochondrial addition and injected into the reaction chamber (filled with water) of a Sievers Noawin analyzer for detection of NO. Samples were analyzed in duplicates and the experiment was performed three times. Amperometric NO measurement with an NO electrode NO production in the mitochondrial preparations was also measured using an NO-sensitive Clark-type electrode (ISO-NO-METER, World Precision Instruments, Berlin, Germany) as described earlier (Schild et al. 2003). The NO-sensitive electrode was calibrated by generating stoichiometric NO standards from defined amounts of the NO donor SNAP injected into fixed volumes of 0.1 M CuCl2. The lowest limit of detection was found to be at 16 nM SNAP, which corresponds to a total of 32 pmol of NO in a 2-mL sample. Measurements of the NO concentration in the mitochondria preparations were performed in the presence of different respiratory substrates (as indicated in the figure legends). At the end of every experiment an excess of the NO scavenger oxyhemoglobin (Hb) was added to the sample. Only if Hb caused a decrease in the NO meter reading such difference was considered to be due to free NO. NOS activity assay NOS activity was evaluated by labeled arginine to citrulline conversion as described (Lacza et al. 2001). The assay was carried out in duplicates, using 100 lL of isolated mitochondria (protein concentration 1–4 lg/lL). The samples were incubated with 75 lL of 14C-labeled L-arginine in Tris buffer, incubated for 30 min at 32C. The 14C universally labeled arginine (specific activity 331 mCi/mmol, New England Nuclear, Boston, MA, USA) was purified through a Dowex iX-8 (acetate form) column and diluted in buffer containing 66.7 mM Tris, 2.7 mM NADPH and 3.33 mM calcium chloride at pH 7.4. Cofactors were added at a final concentration of 2 lM for FMN, 2 lM for FAD and 10 lM for BH4. Arginase activity was inhibited by ornithine at a dose of 1 mM. Separation of arginine from citrulline was performed by ion-exchange chromatography with Dowex-50 W resin (sodium form, 50 · 8-400, Sigma). After washing the column the eluent containing the 14C-citrulline was collected and measured in a scintillation counter. The protein concentration of the original samples was measured by a Bio-Rad DC protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA) and the enzyme activity was expressed in pmols of citrulline per milligram of protein per 30 min. Calmodulin dependence was evaluated by the addition of calmidazolium at a concentration of 100 lM. Background counts were measured in the absence of cofactors and in the presence of 100 lM N-nitro-L-arginine methyl ester (L-NAME) and were subtracted from each measurement. Western blot and immunoprecipitaion Protein was extracted from mitochondria by the addition of boiling lysis buffer (containing 1% v/v 1 M Tris and 1% w/v sodium dodecyl sulfate). The samples were sonicated, heated at 95C for 5 min and centrifuged for 20 min at 12 000 g at 4C. The supernatant was used for immunoblotting. Protein concentration was measured by a Bio-Rad DC protein assay kit (Bio-Rad). Equal amounts of protein were separated on a 4% to 20% gradient mini gel (Bio-Rad), and transferred to polyvinylidene difluoride membrane. After blocking with 5% milk, primary antibody was applied overnight followed by horseradish peroxidase conjugated secondary antibody. Chemiluminescence was used to visualize the bands. Molecular weight markers (Bio-Rad) were included on each blot. Specificity of the method was tested by omitting the primary antibody from the procedure, which resulted in the disappearance of the specific bands. The following antibodies were used: anti-universal NOS (polyclonal, Affinity Bioreagents, Golden, CO, USA); anti-eNOS 1030– 1209 (monoclonal, Transduction Laboratories, San Diego, CA, 944 Z. Lacza et al. 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
Mitnchondrin dn not pmdnon sitric ncide USAl:arn-nNOS 10N5-1239 (monodanal ard polyclnal.Trns- doction Laboratoriesk:anti-nNUS 251-27U (polyclonal,Sigmal: anti-nNOS 1383-1398 (polyckaal,Trmsdectiun Laberakries): anti-nNOS 1-181 (mnmoclmal,Sigma'k anti-nNOS 1400-1429 (polyclonal Sigmak aminNOS polycknal SC-648 (Santa Cnz Biccochnolgies);arti-iNOS 961-1144 (monocknal,Simak ani iNOS 7?2-78?(reoroclnral Sigm) Imminoprocipitatin uas performed on freshly isolsted mitnch- ondiinl poias both in sative and destured cusditiots.Equal smount of peotein were incubuted wit uti-NOS polyeltal antibodies at ovemight in a volame of 750 HL.The artibodics were precpitated by the addnon of 40 uL peoten Ai coupled to headed ape and cemifuped a0g fur 10The peecipite was tesed with wextem hlotting as descrihad ahove. Amnity purification The hypochctical mitochondnal nitnc osde synthase was punhied usng oofactors of NO immmobilizod on beadod sepharose. Michondia and f tsue hompees were lysod wih the addnon of 1%NP.40 (Sigma)to preserve the native foems of the prooeis.The lysaes were peepunted based on arginine binding by Irdne onto I ml.argmnine-epharcar colrns (Signa)ard cluting with excras arginie The chutes weme firther perifed by NADPH bisding usiig 2,5-ADP-epluno culaun (Sigaa)o by clnod ulis binlng tig a calmdalin-sepooe columt (Signa)The colrmns were washed and the bound protcins were denatired by copy.The rod channdl shows he mlochondra using MtoTrackerRed adding Lacmmli sample buffer (Bi-Rad)and heanod to 95C foe stoining.while the geen channd dopicta the NO dependent DAF.2 5 min.Uncrejugited seph:rcee wits used as a seyative cotrul.The ucrescance.The merped image thows fe colcaliration of te mhochondrisl and the DAF-2 signals.Nole that nat all of the Mto- TrckerRod labeled mrochondria ire abe strined with DAF-2 Results Freshly isolated respiring mitochondria were visualized by confocal microscopy and the quantification of the 凡orescent measu代cs Huorescence was achaeved by flow cytometry in separate Mitochendrial NO prodoction was assessed by using the experiments.The miocboedria showed very stong green NO-sensitive fluorescent dyes.DAF-FM or DAF-2-DA. fuorescence in the presence of DAF-FM while the These compounds have a weak green fuorescence,which is auofluocescence of the mitochondria was below the detec greatly incresed in the presence of NO.Howrever,previous tion limit.DAP-FM fhoorescence was observed in all three stodies have shown that the ircrease of the dye's fhoorescence NOS knockout stmins with coefocal microscopy and u was was much greater when the NO donors were applied together not signifcantly different when quantified by flow- with ONOO.Therefire we used DAF funresoence as an cylometry (Fig,2)The DAF-FM fluorescence in the indicntor of nitrogen mdical productian,that inclades both isolated mitochondrin was um fTicted by the applicstion of NO and ONOO. the pharmacolugical NO5 inhibilors like L-NAME or L- Misochondrial ratrogen mdical production was investi- NMMA (up to I ms)(Fg 3).In coarrast,the minochond- eaed in culnued primary neurons wirh laser seamning rial mitrogen radical signal was signineamly hlocked by the confical mieroseopy The cell were kaded with MitoTrack- spplication of the ONOO deooerposition eatalysls FPI5 o erRed tu labdl mitochondria and wih DAF-2-DA lo munitor FcTPPS (Fig.3)Prulonged tralrnl of'the animals with nitrugen radical production.The DAF-2 fluorescence was I mgimL L-NAME for 3 days in the drinking waler and completely colocalizod with the mitochocdria,although not supplementing LNAME in the preparation bffers also every mikchondrion was bbelad with DAF-2 (Fig 1). failed to attenuate the DAF.FM signal Moreover,the signal inersingly.the mitochundril uuria inend was insensitive to Ca'or to the withdrawal of the substrate when the pruparatioes were exposod lo the stanning laser L-anginine (Fig.31. light for a loog time.This observation reflects that the main sources of nitrogen radicals are the mitochondria.The Chemiluminescent NO detection subsequent experiments were coodacted on isolated brain The results with DAF forescence showed prominent mnochondna preparatoes. mnochondnal nitrogen radical production.However,this 2004 Socicty fur,Mwochent (2004)90,942-851
USA); anti-nNOS 1095–1289 (monoclonal and polyclonal, Transduction Laboratories); anti-nNOS 251–270 (polyclonal, Sigma); anti-nNOS 1383–1398 (polyclonal, Transduction Laboratories); anti-nNOS 1–181 (monoclonal, Sigma); anti-nNOS 1409–1429 (polyclonal, Sigma); anti-nNOS polyclonal SC-648 (Santa Cruz Biotechnologies); anti-iNOS 961–1144 (monoclonal, Sigma); antiiNOS 772–787 (monoclonal, Sigma). Immunoprecipitation was performed on freshly isolated mitochondrial proteins both in native and denatured conditions. Equal amounts of protein were incubated with anti-nNOS polyclonal antibodies at 4C overnight in a volume of 750 lL. The antibodies were precipitated by the addition of 40 lL protein A/G coupled to beaded agarose and centrifuged at 1200 g for 30 s. The precipitate was tested with western blotting as described above. Affinity purification The hypothetical mitochondrial nitric oxide synthase was purified using cofactors of NOS immobilized on beaded sepharose. Mitochondria and full tissue homogenates were lysed with the addition of 1% NP-40 (Sigma) to preserve the native forms of the proteins. The lysates were prepurified based on arginine binding by loading onto 1 mL arginine-sepharose columns (Sigma) and eluting with excess arginine. The eluates were further purified by NADPH binding using a 2,5-ADP-sepharose column (Sigma) or by calmodulin binding using a calmodulin-sepharose column (Sigma). The columns were washed and the bound proteins were denatured by adding Laemmli sample buffer (Bio-Rad) and heated to 95C for 5 min. Unconjugated sepharose was used as a negative control. The purified proteins were probed with western blotting as described above. Results Fluorescent measurements Mitochondrial NO production was assessed by using the NO-sensitive fluorescent dyes, DAF-FM or DAF-2-DA. These compounds have a weak green fluorescence, which is greatly increased in the presence of NO. However, previous studies have shown that the increase of the dye’s fluorescence was much greater when the NO donors were applied together with ONOO– . Therefore, we used DAF fluorescence as an indicator of nitrogen radical production, that includes both NO and ONOO– . Mitochondrial nitrogen radical production was investigated in cultured primary neurons with laser scanning confocal microscopy. The cells were loaded with MitoTrackerRed to label mitochondria and with DAF-2-DA to monitor nitrogen radical production. The DAF-2 fluorescence was completely colocalized with the mitochondria, although not every mitochondrion was labeled with DAF-2 (Fig. 1). Interestingly, the mitochondrial fluorescent signal increased when the preparations were exposed to the scanning laser light for a long time. This observation reflects that the main sources of nitrogen radicals are the mitochondria. The subsequent experiments were conducted on isolated brain mitochondria preparations. Freshly isolated respiring mitochondria were visualized by confocal microscopy and the quantification of the fluorescence was achieved by flow cytometry in separate experiments. The mitochondria showed very strong green fluorescence in the presence of DAF-FM while the autofluorescence of the mitochondria was below the detection limit. DAF-FM fluorescence was observed in all three NOS knockout strains with confocal microscopy and it was not significantly different when quantified by flowcytometry (Fig. 2). The DAF-FM fluorescence in the isolated mitochondria was unaffected by the application of the pharmacological NOS inhibitors like L-NAME or LNMMA (up to 1 mM) (Fig. 3). In contrast, the mitochondrial nitrogen radical signal was significantly blocked by the application of the ONOO– decomposition catalysts FP15 or FeTPPS (Fig. 3). Prolonged treatment of the animals with 1 mg/mL L-NAME for 3 days in the drinking water and supplementing L-NAME in the preparation buffers also failed to attenuate the DAF-FM signal. Moreover, the signal was insensitive to Ca2+ or to the withdrawal of the substrate L-arginine (Fig. 3). Chemiluminescent NO detection The results with DAF fluorescence showed prominent mitochondrial nitrogen radical production. However, this Fig. 1 Confocal microscopy of cultured primary neurons. Two representative live cells are visualized by laser scanning confocal microscopy. The red channel shows the mitochondria using MitoTrackerRed staining, while the green channel depicts the NO-dependent DAF-2 fluorescence. The merged image shows the colocalization of the mitochondrial and the DAF-2 signals. Note that not all of the MitoTrackerRed labeled mitochondria are also stained with DAF-2. Mitochondria do not produce nitric oxide 945 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
46之（习开风 methodology is ahle to detect s0 pmol NO in a 2 mL. reacrion vessel (as ealihrated wirh NO donors.see Tahle 1). which is significantly lower than the lowest physiologically rdevanl NO cunccnlratiun.All samples drawn from the headspuce of'the milochondria suspensions at iny of the investigated conditions showed no detectable NO levels even in the preserce of exogenous Larginine (Table 1)Also the addition of cofaclurs of the NOS cnzyme did no nosult in Fig.2 DAF.I Mfuemscneen in NC kneckourt animals Cuartisearion deleclable NO-levels. hs自rn5ncnn85lsn9rah一ochan防rt博a移csnd fow cyoomery and the progartiona were viauaizod by confocdl micracopy (narts).The lt panel ahows that cNO9 KO miochon. Electrochemical NO detection dra have a bright DAF-FM fuorescenoe as scen in the microscoplc A previous study has shown NO pcoductian by isolated heart image.Flow cytometric analysis contirms that the fluorescenoe level is mitochoodria with an electrochemical electrode.This method simlar in th prrol and the KO arains The centeal panel showo a nllows the contmmous measarement of NO in the solution mpraeentsttan aspariment with rNOS knockour minchrnda,whil and genume NO can be conrimed hy the addition of the night pand showa an cxpermene with NOB knockout animal.The hemoglobin in the end of each experiment.In our hands.this miochondrial DAF-FM fluonesoence i unatlectod by the lock of cither method also failed to detect any signal,which was attribut- 405n85 able to NO in the bmin mitochondria preparations (Fig.4) a L NOS activity asay Isulsial brain miluchondria had a low NO5 wclivily.which amountod to kss than 1%of the full brain NOS activity (Fig.5).Such low mitochondrial NOS activity was partially inhihitad by ether I-NAMF or the calmodalin mhihitor caimibzolium (Fig.5)Milochondra isolsied fiom cNOS, iNOS a阳nNOS k5k减animals had a comparabl actiity us the respective wil-tpri匹FigS1As卷cU世oarison, NOS activity was also measured in full beain tissue homogenates froen the same animals as used for the c) mnchoodria peeparatios.The measured fulltissue ctivity wns comparable all strains,and it was greatly reduced in nNOS knockous (Fig.5). Wester blotting and immunoprecipitation Previous studies have demonstmted inmanoreactive bands in mitochoedrin preparations hen Inbeled with specific NOS antibodies,while other stukdies reported negative results using the same antibo:ies (for a review see Lneza er af. 2003a)We have chosen l strocn the prepuraliuns from brain mitochondria with a set of 10 different NOS antibodies Flg.3 DAF-FM fuorscence in te presence of NCS inbbtors or to test the anrigeniry of mtNO5.To identiry the respective ONOO scavengers.Sold Iine shows the contiol reoordings in the persrce al rgnin (1 mul and Ca"(1 pul.(s)and shorw that eNOS,nNOS or iNOS bands we used known positive ithdrswnl ol L-aririna and Ca+and adrftnn of tha competrten NOS controls ohlained from tissues of wild-type as well as NOS inhibiters L NAME or LNMMA (1 m oach)cannot block the DAF.FM knockout animals fuorescenoe in the miochondria (c)and (d]show that wwo diterenl Michondria fom noral mouse brain tissue did not ONOO decompor州5mca水p15 and FeTPP:51G0w对hl show any immunoceactive proteins tested with two different markodly detrasse the fluorescence evon in the prosenco cf L-angin- anti-iNOS antbodies.an anti-eNOS antibody or a universal NOS anlibody (Fig.6).However,there were other miloch- ondrially enrichod bands when probed with several anti- method cannot diferentiate between NO,ONOO,or other nNOS antibodies.These bands had a lower molecular weight NO derived reactive species.Therefore,we atempoed to (70 110 kDa)than expected for nNOS isoforms and were measre gemine NO coecentration with the SIEVERS also present in the nNOS KO animals (Fig.7).Because chemiluminescence NO Analyzer in samples from the nNOS is the only NOS with more than one knoan protein headspace gas of isolated respiring hmain mitochondria.This splice varinnts nnd the exnn2 knockout used in the present 2004 lalenaliteul Socisy fur Neuruchematry,f Mwuohe.(2004)90,042-051
method cannot differentiate between NO, ONOO– , or other NO-derived reactive species. Therefore, we attempted to measure genuine NO concentration with the SIEVERS chemiluminescence NO Analyzer in samples from the headspace gas of isolated respiring brain mitochondria. This methodology is able to detect 50 pmol NO in a 2 mL reaction vessel (as calibrated with NO donors, see Table 1), which is significantly lower than the lowest physiologically relevant NO concentration. All samples drawn from the headspace of the mitochondria suspensions at any of the investigated conditions showed no detectable NO levels even in the presence of exogenous L-arginine (Table 1). Also the addition of cofactors of the NOS enzyme did not result in detectable NO-levels. Electrochemical NO detection A previous study has shown NO production by isolated heart mitochondria with an electrochemical electrode. This method allows the continuous measurement of NO in the solution and genuine NO can be confirmed by the addition of hemoglobin in the end of each experiment. In our hands, this method also failed to detect any signal, which was attributable to NO in the brain mitochondria preparations (Fig. 4). NOS activity assay Isolated brain mitochondria had a low NOS activity, which amounted to less than 1% of the full brain NOS activity (Fig. 5). Such low mitochondrial NOS activity was partially inhibited by either L-NAME or the calmodulin inhibitor calmidazolium (Fig. 5). Mitochondria isolated from eNOS, iNOS and nNOS knockout animals had a comparable activity as the respective wild-type strains (Fig. 5). As a comparison, NOS activity was also measured in full brain tissue homogenates from the same animals as used for the mitochondria preparations. The measured full tissue activity was comparable in all strains, and it was greatly reduced in nNOS knockouts (Fig. 5). Western blotting and immunoprecipitation Previous studies have demonstrated immunoreactive bands in mitochondria preparations when labeled with specific NOS antibodies, while other studies reported negative results using the same antibodies (for a review see Lacza et al. 2003a). We have chosen to screen the preparations from brain mitochondria with a set of 10 different NOS antibodies to test the antigenity of mtNOS. To identify the respective eNOS, nNOS or iNOS bands we used known positive controls obtained from tissues of wild-type as well as NOS knockout animals. Mitochondria from normal mouse brain tissue did not show any immunoreactive proteins tested with two different anti-iNOS antibodies, an anti-eNOS antibody or a universal NOS antibody (Fig. 6). However, there were other mitochondrially enriched bands when probed with several antinNOS antibodies. These bands had a lower molecular weight (70–110 kDa) than expected for nNOS isoforms and were also present in the nNOS KO animals (Fig. 7). Because nNOS is the only NOS with more than one known protein splice variants and the exon2 knockout used in the present Fig. 2 DAF-FM fluorescence in NOS knockout animals. Quantification of the fluorescence in isolated brain mitochondria was achieved by flow cytometry and the preparations were visualized by confocal microscopy (inserts). The left panel shows that eNOS KO mitochondria have a bright DAF-FM fluorescence as seen in the microscopic image. Flow cytometric analysis confirms that the fluorescence level is similar in the control and the KO strains. The central panel shows a representative experiment with nNOS knockout mitochondria, while the right panel shows an experiment with iNOS knockout animals. The mitochondrial DAF-FM fluorescence is unaffected by the lack of either NOS genes. Fig. 3 DAF-FM fluorescence in the presence of NOS inhibitors or ONOO– scavengers. Solid line shows the control recordings in the presence of L-arginine (1 mM) and Ca++ (1 lM). (a) and (b) show that withdrawal of L-arginine and Ca++ and addition of the competitive NOS inhibitors L-NAME or L-NMMA (1 mM each) cannot block the DAF-FM fluorescence in the mitochondria. (c) and (d) show that two different ONOO– decomposition catalysts FP15 and FeTPPS (100 lM each) markedly decrease the fluorescence even in the presence of L-arginine (1 mM) and Ca++ (1 lM). 946 Z. Lacza et al. 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
Table 1 Head space measurement of NO using the Eiever's chemluminesoent NO andlyzer Butler Buffor Miochondria Miochondn Milochondria ehem过ie Miochendria 80 nu SNAP 0 nv BNAP moio ADP malale mak 10的vL-arginine plutsmate glutamate guo■0 200 uv NADPH AOP ADP 10日凡 10 mu Ag 5yu FAD 5uN FMN 10 pofeel oalmodufin 10 jM CaCl 57.3±128 151.7±27.0nd n.d n.c. nd. ad. mikingi unlikely that these prutcins share significam homology with nNO5. In order to roduce non-specific binding to the amibodies, next we used immuncprecipitation protocols.Two polyclonal atibodies,one igainst the oxygenase domnain and cne against the reductase domain,were used to precipitate nNOS in separte experiments.The isolated proteins were probed with westem blotting using other anti-nNOS antibodies raised against different epitopes.This approach resubed in the concentration of nNOSx and nNOSB in full brain tissue (Fig.7).However,no mitochondrial proteins were sound (Fig 7),not even the secondary handks,which were seen with westem bltting (see ahove),making it very unlikely that these kwer molecular weight bandl represent unknown plice variants of nNOS. 10 Affinity purification 释0 Brin mikchondris sed full brain lixse xamples were Fig.4 No detectable nitric axide production was measured in bolated loxdad onto aginine-phar culurms and cluted with hain mtn中rt的时h an Necimchemicn nincteodie (a)ghmw the ecess arginine.The elucnt frum the full tissu,but not from the miochoodria coctained nNOS is shown with mixochondria preparaion.Malle and plutarnone (5 ms 100 s)were western blotting (Fig.7)Subsequent affinity purification odded at te boginning lollowod try Ie addiion of micchondrin (0.5- with the NADPH analog 2.5-ADP resuhed in a further 10 mg protein 200 5).Subsequenty.ADP (2 j 400 s)was added concentralion of nNOS in the brain.still without any nNOS- Io induce the state 3 resplration Fnaly.in arder to detoct amy niric ke protein in the mitochondria (Fig.7).Similar results nerle in tha arlution hamoghin七t岔0ndeo动atae were obtained with calmodulin-sepbarose (Fig.7).Probing hich did nol xrow arty signifrart drop in Ia xigl pring thant them of the affinity purfied proteins with an amti-eNOS antibody is no detectoble NO in te medum.Data iepresenta mean SEM al also resalted in a strong band in the full tissae samples with ree dicrent年oriments.[b)Verication al the.o水nsthty no mitochondrial peoteins at all.Pmhing with an anti-iN Addion af diterent cancertratione of the NO denor SNAP (in nano molar concentr3n5adh85呀海endent rse in oumo-t antibody did not show any immunoreactve hands in the Adcion Hb0 pv)inmedialely crss th SNAP-induoed prepairalions te由Dicidline waltc. Discussion and other snides lscks only the most ahundant nNOS There are two mai销observatioes鱼he present study.First, isofurm,the possibility remained open that these other bands we showed that brain mitochondria do not produce NO via reprocnl a milchundrial splice varinnt of'nNOS.In onder lu the arginine-o-citrulline cunverson pathway.Seodwe did test this hypothesis we screened the mitocbondria with six not find significant kvels of NOS ezymes in the mitchon- different anti-nNOS atibodies.None of these lower molecu- dria.Both of these obeervations indicate that mNOS is not lar weight bands reacted with more than one antibody, present in hrain mitochondr. o24 htemrimnal号exie时r Nerochemitry./uht(2m)维42-sl
and other studies lacks only the most abundant nNOSa isoform, the possibility remained open that these other bands represent a mitochondrial splice variant of nNOS. In order to test this hypothesis we screened the mitochondria with six different anti-nNOS antibodies. None of these lower molecular weight bands reacted with more than one antibody, making it unlikely that these proteins share significant homology with nNOS. In order to reduce non-specific binding to the antibodies, next we used immunoprecipitation protocols. Two polyclonal antibodies, one against the oxygenase domain and one against the reductase domain, were used to precipitate nNOS in separate experiments. The isolated proteins were probed with western blotting using other anti-nNOS antibodies raised against different epitopes. This approach resulted in the concentration of nNOSa and nNOSb in full brain tissue (Fig. 7). However, no mitochondrial proteins were found (Fig. 7), not even the secondary bands, which were seen with western blotting (see above), making it very unlikely that these lower molecular weight bands represent unknown splice variants of nNOS. Affinity purification Brain mitochondria and full brain tissue samples were loaded onto arginine-sepharose columns and eluted with excess arginine. The eluent from the full tissue, but not from the mitochondria contained nNOS as shown with western blotting (Fig. 7). Subsequent affinity purification with the NADPH analog 2,5-ADP resulted in a further concentration of nNOS in the brain, still without any nNOSlike protein in the mitochondria (Fig. 7). Similar results were obtained with calmodulin-sepharose (Fig. 7). Probing of the affinity purified proteins with an anti-eNOS antibody also resulted in a strong band in the full tissue samples with no mitochondrial proteins at all. Probing with an anti-iNOS antibody did not show any immunoreactive bands in the preparations. Discussion There are two main observations in the present study. First, we showed that brain mitochondria do not produce NO via the arginine-to-citrulline conversion pathway. Second, we did not find significant levels of NOS enzymes in the mitochondria. Both of these observations indicate that mtNOS is not present in brain mitochondria. Table 1 Head space measurement of NO using the Siever’s chemiluminescent NO analyzer Buffer 80 nM SNAP Buffer 320 nM SNAP Mitochondria malate glutamate Mitochondria ADP Mitochondria malate glutamate ADP Mitochondria malate glutamate ADP 10 mM Arg Mitochondria 100 lM L-arginine 200 lM NADPH 10 lM BH4 5lM FAD 5lM FMN 10 lg/ml calmodulin 10 lM CaCl2 57.3 ± 12.8 nM 151.7 ± 37.0 nM n.d. n.d. n.d. n.d. n.d. All experiments were performed at least three times with three different mitochondria preparations. n.d., no detectable NO was measured. Fig. 4 No detectable nitric oxide production was measured in isolated brain mitochondria with an electrochemical electrode. (a) shows the average recording traces of the electrode submersed in a brain mitochondria preparation. Malate and glutamate (5 mM, 100 s) were added at the beginning followed by the addition of mitochondria (0.5– 1.0 mg protein, 200 s). Subsequently, ADP (2 lM, 400 s) was added to induce the state 3 respiration. Finally, in order to detect any nitric oxide in the solution hemoglobin (Hb; at 600 and 660 s) was added, which did not show any significant drop in the signal proving that there is no detectable NO in the medium. Data represents mean ± SEM of three different experiments. (b) Verification of the electrode sensitivity. Addition of different concentrations of the NO donor SNAP (in nanomolar concentrations) resulted in a dose dependent rise in current. Addition of Hb (350 lM) immediately decreases the SNAP-induced rise to the baseline values. Mitochondria do not produce nitric oxide 947 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
Mk 7 [3r23 rr al 兰 改 (b) brain mtsshordra Fig.6 Wesiom boming wih anti-NOS anbodea Ful brain tiaaue from the same animals was used as a poitive cortrol for eNOS and nC5nd&c寸sted mac"gp33eai00m8c5时was uced as a posiiv comol ler iNO&.All thres ielbodica mcogrized th nuapctiw hand in the contol but not in the minchandri.Tha poly- clonal anti-nNOS ontbody (Trareduction Laboralories,610606) recogrizes several non-spocito bands.Equal amounts of protcin were k新s8h时 NOSKD O6 KO HMOS TNDSKD in the presert study.Sevennl other groups hane puhlished similr resulls in various preparalions indicaling that the localization of DAF forescen o the mitochondrion is a 000 gencral observation (Lopez-Figuerua er af.2000:Dennis and Bennett 2003)However,withdrawal of arginine and C from the solulion Gaild lo deenease miluchondlrial DAF fuorescenoe.The signal was not inhibited by the application of clssical,competitive NOS inhibiturs like L-NAME or L NMMA.even in high doses (I mst)or prolocged appli cation (3 days poetreatment before mitchondria prepar- ntion).Furthermore,the DAF fuorescerxce uns similar in minochoedria prepared from all three NOS knockou strains. whDs NOS 0 NdS KD rNCG nY15 KO These observations reflect that the mitochondrial conversion 0图 ccngol of DAF to a fluorescent product is not the reslt of a classical Fig.5 Mlochondiol NOS activiy messurod by the argirine-o-ctl. NOS enzyme activity.Our experiments with ONOO lne conversion assay.NOS actMty is eopressed as pg cltrulline pro decomposition camlysts sggest that ocher nitrogen radicals duction per mg prclein per 30 min.(al shows results obtained in brain besides NO may be responsible for mitochondnal DAF milechendris nl wild typa mice.Applearion ol L-NAJE (100 jul cr Iht fuorescence.Furthemore,DAF staining of the mitochcexdlria camodulin nhiciar calidarcium (CMZ:100 makedly mfuced wi very hekcrupencuus both in cullured cell snd in isolted the actviy.showa results oatained in brain miochondria al NOS n5必3nar=s aed the88t特nrtr够.Th4m%可 organclles.Further sudis are nodkd lo udkrstand the 与nticant dtereno语nngh5n5G间hoa感G5aty chemistry of DAF betore we can form a conclusion of the marurod in full broin tiasue.The nNO8 KO atain has a signilicandy nature of the mitochondrial DAF fluorescence One possible rodueid activty compard o he comrel Nole tha the mtNO&activily working hypothesis is that mitochondria generate NO from is approxdmatey 100 times lower than that af te ful brain Six to 10 nitrite via a mchanism that do not involve the NO aniels were used in each group:'p<0.06,p<0.01. synthaases (Nohl et al 2000),followed by a rapid reaction of NO with superuxide (produced by the mitochondrial oxida- Based on the growing literature of mtNOS and cur tive processes).Amother possibility is that the DAF fluores previos experiments with liver mitochondra,we expected cence,in mitochondria,is related to some peculiarity of the to see significant mitochondrial NO)production in the brain. probe.when reacting with eactive oygen spocies,which First,we visualized the miochondria in cultured neurons and can also be neutralized by the porphyrinic antioxidants used found that the NO-sensitive DAF foorescence colocalizes in our studies. with the mitochoedria.This observation was confmed in Since the specificity of DAF to NO has been questioned in isolated mitochondria and was quantified by fow cytometry other preparations also,we tried to detect mitochondrial NO 2004 Sucisy fur.I.(2004)90,942-951
Based on the growing literature of mtNOS and our previous experiments with liver mitochondria, we expected to see significant mitochondrial NO production in the brain. First, we visualized the mitochondria in cultured neurons and found that the NO-sensitive DAF fluorescence colocalizes with the mitochondria. This observation was confirmed in isolated mitochondria and was quantified by flow cytometry in the present study. Several other groups have published similar results in various preparations indicating that the localization of DAF fluorescence to the mitochondrion is a general observation (Lopez-Figueroa et al. 2000; Dennis and Bennett 2003). However, withdrawal of arginine and Ca2+ from the solution failed to decrease mitochondrial DAF fluorescence. The signal was not inhibited by the application of classical, competitive NOS inhibitors like L-NAME or L-NMMA, even in high doses (1 mM) or prolonged application (3 days pretreatment before mitochondria preparation). Furthermore, the DAF fluorescence was similar in mitochondria prepared from all three NOS knockout strains. These observations reflect that the mitochondrial conversion of DAF to a fluorescent product is not the result of a classical NOS enzyme activity. Our experiments with ONOO– decomposition catalysts suggest that other nitrogen radicals besides NO may be responsible for mitochondrial DAF fluorescence. Furthermore, DAF staining of the mitochondria was very heterogeneous both in cultured cells and in isolated organelles. Further studies are needed to understand the chemistry of DAF before we can form a conclusion of the nature of the mitochondrial DAF fluorescence. One possible working hypothesis is that mitochondria generate NO from nitrite via a mechanism that does not involve the NO synthases (Nohl et al. 2000), followed by a rapid reaction of NO with superoxide (produced by the mitochondrial oxidative processes). Another possibility is that the DAF fluorescence, in mitochondria, is related to some peculiarity of the probe, when reacting with reactive oxygen species, which can also be neutralized by the porphyrinic antioxidants used in our studies. Since the specificity of DAF to NO has been questioned in other preparations also, we tried to detect mitochondrial NO Fig. 5 Mitochondrial NOS activity measured by the arginine-to-citrulline conversion assay. NOS activity is expressed as pg citrulline production per mg protein per 30 min. (a) shows results obtained in brain mitochondria of wild type mice. Application of L-NAME (100 lM) or the calmodulin inhibitor calmidazolium (CMZ; 100 lM) markedly reduced the activity. (b) shows results obtained in brain mitochondria of NOS knockout animals and the respective control strains. There was no significant difference among the groups. (c) shows NOS activity measured in full brain tissue. The nNOS KO strain has a significantly reduced activity compared to the control. Note that the mtNOS activity is approximately 100 times lower than that of the full brain. Six to 10 animals were used in each group; *p < 0.05, ** p < 0.01. Fig. 6 Western blotting with anti-NOS antibodies. Full brain tissue from the same animals was used as a positive control for eNOS and nNOS, and activated macrophage lysate (macro) was used as a positive control for iNOS. All three antibodies recognized the respective band in the control, but not in the mitochondria. The polyclonal anti-nNOS antibody (Transduction Laboratories, 610606) recognizes several non-specific bands. Equal amounts of protein were loaded in each lane. 948 Z. Lacza et al. 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
Michondria do not produce im oxide 4特1t b时 simmatices isolted respirng hrain mitochondria did ot r中4n produce detertble amoumts of NO cither in the solution oe in the headbpace gs.We altered the respiralory so 道 suppiemenlod the availbality of substrac or potential cofactoes bul still no NO pruduction was obescrved.These 题 22 observations are in cootrast to several studies in the lterature. Riobo and colleagues used hemoglbin oxidation to indi 1 reclly meissure NO produdion in brain mochondrial membranes (Riobo e a.2002).In the present study.we used a direct NO detection method and we did not observe ay changes.which would be inhibitable by hemodlobin in brain mitochondria.One explanation of these divergent data can be that mtNOS is only active at the early stages in brin 29 development,and the adult mrNOS levels are just insigni- nermt remnants of the emhryomnie stage,with some variarims among species and animal strains. e 义礼在红 If there is no detectable NO prodction i the mitochon- dria,why do we measre arginire conversioe althomgh at very modest levels?The aginine to citrulline comversion assy optimizes the conditions for the NO5 enzymes the suberale and the cfaclurs are sburdanl ted wther armine- melabolizing enzymes are inhibited.Under these conditions. we were able o detect a very weak mitochondrial NOS setivity signal,which amounted to less than 1%of the fiall ” lisue aclivily.As even the best milochandri preprations contain 1-4%nun-miluchondrial manbranes (Sims 1990).it is impossible to distinguish whether this very k NOS activity belongs to the mitochondria or the contaminant membranes.It is also possible that arginine conversicn is the Ng NADP Ag+Car nAg+CP每C result of ocher enzymes like arginase Il or nrginine decarb oxylase,both od which are ahundant minochondrial enzymes. t491t的 Therefore.the mitochondrial NOS activity meassurements in Fig.7 Search lor nNOS splloe varlants in miochondria Pand A the present study cannot pcove the existence of a distinct hGw8 a tepresentali情s对am hint wih音=mnrW-hn核 mNOS enyae. -th布，whhc灯r女a%al krown rNOS splira waiants.Ful bran Extensive screemang of the mitochondrial proteans using a toe contaira nNOS,which i miaaing in te nNOS KO arimnal.An set of 11 NOS nntibodies faaled to show the presence of any opprcoimnely 75-kDa bond is enrichod in he milochondrin compused specific NOSs in hrain mitochondna.We used full bmin lo the ful brain.This band is simlarfy present in mitochondria trom both control and nNOS KO mice.indicating that it cannot be a deg homogenaex as positive cuntrols amd the respoctive NOS ion pmdua of炉S2间owx Iha=mumprecipirton cl knuckoul srains as negative cuntruls Immaoprecipilation hoin And mitncnn中pmparsticrs wih来pa士a-tnoS protocols were applied to concentrate nNOS in the mito. abad,e preciprale w色oted wi鱼a dferent monoclonal anti chondra.This procedure snecessfuilly eliminated the non- body apainst te sane epiope.Two nNOS spice vaants (nNCSa specifie hands and concentral the known nNO5 splice and rNOSl were identted in the full brain.but no mitochondral varianl(宝，Y)in the brain bu域阔in mitochondris.We proleins wore lound.cl shaws the resuts of the attinity purtication al also tried to purify mINOS busod on its aflirity to the nNOS frem bnin and Erain mrochendria.Argnine (Argl.25-ADP substrale or cofaclors of NOS.This procedure very effect- (ADP)or camodulin (Cam]atnity purlt cason etlectively ooncentrates ively concentrated eNOS and nNOS in fill brain tissue, the nNOS form the brain tissue,however,no mhochandrial bands however.mnochocdria preparations did not coetain any NOS immmunorcaclive prulcins cven afler aflinity purilicatin. Futherore,we also triod to find NOS immunureactivity Sonatice by direet methods (Jourd'heail 2002:Roychowdh- in the mitochondria with immunogold electron mcroscpy. ury er af.2002.Zhang 2002)Both the chemilumines- without any success (Lacza er af.,unpublished observations). ccnee and the amperoenctrie deleclion tre sensitive methods This finding is also supported by the lack of micochondrial to measune NO rekuse froen mituchondria as dornestralod in tansport tngs in the sequences of either NOS peotein,which liver tissue (Schald er af.20013)Under these well-controlled we published recently (Laczn er of.2003a). 2004 lalemrimnal Socicty fur Nerochmbtry,Mwuchemt (2004)90,942-051
formation by direct methods (Jourd’heuil 2002; Roychowdhury et al. 2002; Zhang et al. 2002). Both the chemiluminescence and the amperometric detection are sensitive methods to measure NO release from mitochondria as demonstrated in liver tissue (Schild et al. 2003). Under these well-controlled situations isolated respiring brain mitochondria did not produce detectable amounts of NO either in the solution or in the headspace gas. We altered the respiratory status or supplemented the availability of substrate or potential cofactors but still no NO production was observed. These observations are in contrast to several studies in the literature. Riobo and colleagues used hemoglobin oxidation to indirectly measure NO production in brain mitochondrial membranes (Riobo et al. 2002). In the present study, we used a direct NO detection method and we did not observe any changes, which would be inhibitable by hemoglobin in brain mitochondria. One explanation of these divergent data can be that mtNOS is only active at the early stages in brain development, and the adult mtNOS levels are just insigni- ficant remnants of the embryonic stage, with some variations among species and animal strains. If there is no detectable NO production in the mitochondria, why do we measure arginine conversion, although at very modest levels? The arginine to citrulline conversion assay optimizes the conditions for the NOS enzymes: the substrate and the cofactors are abundant and other argininemetabolizing enzymes are inhibited. Under these conditions, we were able to detect a very weak mitochondrial NOS activity signal, which amounted to less than 1% of the full tissue activity. As even the best mitochondria preparations contain 1–4% non-mitochondrial membranes (Sims 1990), it is impossible to distinguish whether this very low NOS activity belongs to the mitochondria or the contaminant membranes. It is also possible that arginine conversion is the result of other enzymes like arginase II or arginine decarboxylase, both of which are abundant mitochondrial enzymes. Therefore, the mitochondrial NOS activity measurements in the present study cannot prove the existence of a distinct mtNOS enyzme. Extensive screening of the mitochondrial proteins using a set of 11 NOS antibodies failed to show the presence of any specific NOSs in brain mitochondria. We used full brain homogenates as positive controls and the respective NOS knockout strains as negative controls. Immunoprecipitation protocols were applied to concentrate nNOS in the mitochondria. This procedure successfully eliminated the nonspecific bands and concentrated the known nNOS splice variants (a, b, c) in the brain, but not in mitochondria. We also tried to purify mtNOS based on its affinity to the substrate or cofactors of NOS. This procedure very effectively concentrated eNOS and nNOS in full brain tissue, however, mitochondria preparations did not contain any NOS immunoreactive proteins even after affinity purification. Furthermore, we also tried to find NOS immunoreactivity in the mitochondria with immunogold electron microscopy, without any success (Lacza et al., unpublished observations). This finding is also supported by the lack of mitochondrial transport tags in the sequences of either NOS protein, which we published recently (Lacza et al. 2003a). (a) (c) (b) Fig. 7 Search for nNOS splice variants in mitochondria. Panel A shows a representative western blot with a monoclonal anti-nNOS antibody, which recognizes all known nNOS splice variants. Full brain tissue contains nNOS, which is missing in the nNOS KO animal. An approximately 75-kDa band is enriched in the mitochondria compared to the full brain. This band is similarly present in mitochondria from both control and nNOS KO mice, indicating that it cannot be a degradation product of nNOSa. (b) shows the immunoprecipitation of brain and mitochondria preparations with a polyclonal anti-nNOS antibody, the precipitate was blotted with a different monoclonal antibody against the same epitope. Two nNOS splice variants (nNOSa and nNOSb) were identified in the full brain, but no mitochondrial proteins were found. (c) shows the results of the affinity purification of nNOS from brain and brain mitochondria. Arginine (Arg), 2,5-ADP (ADP) or calmodulin (Cam) affinity purification effectively concentrates the nNOS form the brain tissue, however, no mitochondrial bands were found. Mitochondria do not produce nitric oxide 949 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
950 Z.Lacza er w. One possible esplanation for the divergent data in the Frardorn U.Inez-Pigrem M and Helleen Y.(1996)I ncalization nf conled of mIN05 is that oclalar NOS is allachod to the niic atide syuhbe in hanen skektl mocl.Bxkn.Bo球h以. milochondria in a similar manner like gNOS is locsliad to 可C1wmw227.8-91 Gile oufir P.ud Richik C.l9g升Milclae止l nir utidke syu the cavoole in the endotbelial cell membrane.Indeed.several hw晚dates reitechondr3i本ll.dA3相.125 stodies descnbed NOS immunoceactivity or NADPIl 108. diaphorase activity in the proximity of the mitochoodria. Ghorner P.Schenk U..Klein S D and Richer C.(1999)Mitrch- but not in the arganelles themselves (Frandsen1996; u止airi线iyDg3白u山6unu5kG Rothe er al.1999;Henrich et ml.2002)The present study release om isnled mtochendria:eviderce foe intrametochondnal hcws位t mitochondria are not cap:止e of NO production Gidlivi C.(1998)Fuaclital inplicaluts uf tilric exid:prodated by when they are separated from other cellalar elemects,no they coctain a mtNOS,however,there may be a functional 670 attnchment of NOS emyzmes to the outer miochondnal Girlin C (2003)Chacasnerganion ard frneton of mnochordnal nitnc. membrane in the intsct cell.This phenomenon may account unitk:.Fn..Mod 34.397-405. Gmliv C..Poderoso J.J.ad Boweris A.(1598)Prodiction of mitc for the positive results in some earlier studies (Rates er af. utitk:by mibchuadr∠.品k.n2r3,1038-1l043. 1995:Rothe et al 1999.Lacza ef af.2001).Further,our data Heales S.,Bons人.P.Stewant￥C-Brookes PS.Lnd1.Mand underine the hypochesis that mtNOS plays n role in the Clatk J.B.(199)Niric usit,miuchondria and nesrubgiral pathomechanism of neurolgical disorders. disease.Blochin.Bfogys.n 1410.215-225. Hetrich M.,Haffirans K.,Korig P.,Grass M,Fiscibach T.,Godake A..Hemtpelmonn G and Kummer W.12002)Sensory meurons Acknowledgements p园hpaw尚NO prution scaited wi油mi chondna.Mal Ced Mavooef.20.307-322. The authoes are yraleful for Ken Cinnt for hi nvaltible heip with Jourd'heil D.(202)lacreanod nirie oxide-dpanda ritrueybeic uf confocal and electran mcroscopy ard Dr Bela Kis for cultaring 4.5-daminoflarrrin by oxidate inelictims fre the msmre. neurons.We also thank Dr Martha Alexander-Miler for ber ment of imrooelulor ahni oode.Free Rafke.Mol.Mfod.33.676 einen f-eyiometry This sy was ord hy erats 684 fiom the American Heart Asncintion (Mid-Atantic Grant Kara A.J.Peorce L.L.Clemems P.R.Birder LA.VanBiber M.M. 99512724,Bugher Fuudaion Aud 0270114 N)and the NIH Chni S.Y..de Crret W.C.and Peterrn J.(201)ldentifcation nf HL30260,HL46S58.HL50587.HD38964%the Hangian OTKA ing clectechenial dete国.me.Nadl Acad名a星4 1D-4593,1-02916份，147831g7w603ndT2482G 14126-14131. 24线2士and自e Geman DHG W0474WI-4.sA2wTA两xG Kis B..Snipes J.A.looe T..Nagy K.ard Bupa D.W.(00%)Patative and NBL301220107 Fbw.h2头12行-122 Lc4Z,Pu域uM,Fiacrou人B,Zhong J.民ajupakse N.ad Busija References Aniic S.L.Connel M.F.ad Buveris A.(1990)Nire uide. octive and b fusctisnuly upregalaod i lypoxia.Free Raie Biw. Cdi,1c0-1615 superocide,and hydopen peroside pioductin in brain mro chotaria afler hbpraol tratrem.Nune Oude 3,235-243. LZ,eva由E.MKK,Heb4oi工，S2aoC.l Bes T.E..Loesch A.Bimstock Ci.and Clrk J.B.(1945)Immuro. Basii D.W.(20031 FARP inhibitien impowes the effectiveress of clchamanl evidmnee Sr a miochdally Ieealed niric cxide 人.d12.L3-159 symthase in brain and lher.Biochevt Bisis.Res.Cams 213. lac 2.Sripes I.A Zhang I.Honath R.M.F-gurmo I.P.SxahoC Bevers A..Amaiz S.L.Bustzmeme J..AMarez 5.Valdez L.Bovens n Busija D.W.(2003ul Mitchoorial nizie oxide syullas:i net eNOS,nNnS or iNOS Fr Badir Miol led 35.1217-1228 A.D.in2 Nnam A.12002)Muracnngicl regalatian of ril- ochondnal nimic ooode syuthase.MfeMor Errvo.359.328-339. Lacz Z.Supes I.A,Kis B.,Szabo C.Giruver G.ad Busja D.W. 13hi Ime中ion od th是sahurrit com写ogtion and the h mrochoncral resperodion:irplcations for iflmmalory.nearo. mkgy of the mikcbska ATP-perda3-）ad拉e degrneraive and iochaertie pocholrgix Mn Co Bnckr 194. mh.备o5.92-6 19-192. Liauel L-Sorianu F.G.u Sabu C.[2000)Biology uf niric eoide Rirn 0.C.(1909)Nirir cocide and nitchardrinl mpiration Avone sipoling Cret Care LMd 18.N37-N52 Lupet-Fisurut M.O.,Caanat C..Muratu M L,Rinn L.C.,Akil H. %w4.4w141L31-369. Rirn G.C(Nitric mide as a crempetitive inhibiter n ncypen and Warsen S.1 (2000 Direct evderce of nitic oxide presence cnesamptns in the mitnchendrial respinaiery chain.AeM Arynol willin anclusdra fiuchen Broyire.Br.Conmn 272 129 antL16.667-674 135 ern1 and Remett J.P」.20)eractins amar吧nitric oside and Muray J.,Tayhe S.W.,Zhang B.Ghooh S.S.aal Capeldi R.A.[20031 Oailalie darape lo milchenrial tumpex I de w penoynarile: Bel-family peoteis afr MPP+expooue of SH-SYSY neanl oels M+incnoes mitnchnndnal城)nd Hax metrin了am &.T276-8 m.7k37223-37230 Flfiring S L.Sarkea T M and Gialivi C:(Riochemietry of Nohl IL.Sranck K..Sobhan B..Batram S..Recl 11.and Keolov A.V. m山阳4lie-wkye6.人.Bi.Chs起277,3切9- (2000)Mhodria nyck silrik:bck the bitegiae sitri m3mw6e.A形.N.7,13-921 D204 aterrotirnal Soci与for Neumnchemat与y.了wehw2004地94-4l
One possible explanation for the divergent data in the context of mtNOS is that cellular NOS is attached to the mitochondria in a similar manner like eNOS is localized to the caveole in the endothelial cell membrane. Indeed, several studies described NOS immunoreactivity or NADPH diaphorase activity in the proximity of the mitochondria, but not in the organelles themselves (Frandsen et al. 1996; Rothe et al. 1999; Henrich et al. 2002). The present study shows that mitochondria are not capable of NO production when they are separated from other cellular elements, nor they contain a mtNOS, however, there may be a functional attachment of NOS enyzmes to the outer mitochondrial membrane in the intact cell. This phenomenon may account for the positive results in some earlier studies (Bates et al. 1995; Rothe et al. 1999; Lacza et al. 2001). Further, our data undermine the hypothesis that mtNOS plays a role in the pathomechanism of neurological disorders. Acknowledgements The authors are grateful for Ken Grant for his invaluable help with confocal and electron microscopy and Dr Bela Kis for culturing neurons. We also thank Dr Martha Alexander-Miller for her suggestions on flow-cytometry. This study was supported by grants from the American Heart Association (Mid-Atlantic Grant 99512724, Bugher Foundation Award 0270114 N) and the NIH (HL30260, HL46558, HL50587, HD38964); the Hungarian OTKA (D-45933, T-029169, T-037885, T-037386) and ETT (248/2003, 249/2003); and the German DFG WO 474/11–4, SA 2997 A/088G and NBL301220107. References Arnaiz S. L., Coronel M. F. and Boveris A. (1999) Nitric oxide, superoxide, and hydrogen peroxide production in brain mitochondria after haloperidol treatment. Nitric Oxide 3, 235–243. Bates T. E., Loesch A., Burnstock G. and Clark J. B. (1995) Immunocytochemical evidence for a mitochondrially located nitric oxide synthase in brain and liver. Biochem. Biophys. Res. Commun. 213, 896–900. Boveris A., Arnaiz S. L., Bustamante J., Alvarez S., Valdez L., Boveris A. D. and Navarro A. (2002) Pharmacological regulation of mitochondrial nitric oxide synthase. Methods Enzymol. 359, 328–339. Brown G. C. (1997) Nitric oxide inhibition of cytochrome oxidase and mitochondrial respiration: implications for inflammatory, neurodegenerative and ischaemic pathologies. Mol. Cell Biochem. 174, 189–192. Brown G. C. (1999) Nitric oxide and mitochondrial respiration. Biochim. Biophys. Acta 1411, 351–369. Brown G. C. (2000) Nitric oxide as a competitive inhibitor of oxygen consumption in the mitochondrial respiratory chain. Acta Physiol. Scand. 168, 667–674. Dennis J. and Bennett J. P. J. (2003) Interactions among nitric oxide and Bcl-family proteins after MPP+ exposure of SH-SY5Y neural cells I: MPP+ increases mitochondrial NO and Bax protein. J. Neurosci. Res. 72, 76–88. Elfering S. L., Sarkela T. M. and Giulivi C. (2002) Biochemistry of mitochondrial nitric-oxide synthase. J. Biol. Chem. 277, 38079– 38086. Frandsen U., Lopez-Figueroa M. and Hellsten Y. (1996) Localization of nitric oxide synthase in human skeletal muscle. Biochem. Biophys. Res. Commun. 227, 88–93. Ghafourifar P. and Richter C. (1999) Mitochondrial nitric oxide synthase regulates mitochondrial matrix pH. Biol. Chem. 380, 1025– 1028. Ghafourifar P., Schenk U., Klein S. D. and Richter C. (1999) Mitochondrial nitric-oxide synthase stimulation causes cytochrome c release from isolated mitochondria: evidence for intramitochondrial peroxynitrite formation. J. Biol. Chem. 274, 31185–31188. Giulivi C. (1998) Functional implications of nitric oxide produced by mitochondria in mitochondrial metabolism. Biochem. J. 332, 673– 679. Giulivi C. (2003) Characterization and function of mitochondrial nitricoxide synthase. Free Radic. Biol. Med. 34, 397–408. Giulivi C., Poderoso J. J. and Boveris A. (1998) Production of nitric oxide by mitochondria. J. Biol. Chem. 273, 11038–11043. Heales S. J., Bolanos J. P., Stewart V. C., Brookes P. S., Land J. M. and Clark J. B. (1999) Nitric oxide, mitochondria and neurological disease. Biochim. Biophys. Acta 1410, 215–228. Henrich M., Hoffmann K., Konig P., Gruss M., Fischbach T., Godecke A., Hempelmann G. and Kummer W. (2002) Sensory neurons respond to hypoxia with NO production associated with mitochondria. Mol. Cell Neurosci. 20, 307–322. Jourd’heuil D. (2002) Increased nitric oxide-dependent nitrosylation of 4,5-diaminofluorescein by oxidants: implications for the measurement of intracellular nitric oxide. Free Radic. Biol. Med. 33, 676– 684. Kanai A. J., Pearce L. L., Clemens P. R., Birder L. A., VanBibber M. M., Choi S. Y. , de Groat W. C. and Peterson J. (2001) Identification of a neuronal nitric oxide synthase in isolated cardiac mitochondria using electrochemical detection. Proc. Natl Acad. Sci. USA 98, 14126–14131. Kis B., Snipes J. A., Isse T., Nagy K. and Busija D. W. (2003) Putative cyclooxygenase-3 expression in rat brain cells. J. Cereb. Blood Flow Metab. 23, 1287–1292. Lacza Z., Puskar M., Figueroa J. P., Zhang J., Rajapakse N. and Busija D. W. (2001) Mitochondrial nitric oxide synthase is constitutively active and is functionally upregulated in hypoxia. Free Radic Biol. Med. 31, 1609–1615. Lacza Z., Horvath E. M., Komjati K., Hortobagyi T., Szabo C. and Busija D. W. (2003) PARP inhibition improves the effectiveness of neural stem cell transplantation in experimental brain trauma. Int. J. Mol. Med. 12, 153–159. Lacza Z., Snipes J. A., Zhang J., Horvath E. M., Figueroa J. P., Szabo C. and Busija D. W. (2003a) Mitochondrial nitric oxide synthase is not eNOS, nNOS or iNOS. Free Radic Biol. Med. 35, 1217–1228. Lacza Z., Snipes J. A., Kis B., Szabo C., Grover G. and Busija D. W. (2003b) Investigation of the subunit composition and the pharmacology of the mitochondrial ATP-dependent K(+) channel in the brain. Brain Res. 994, 27–36. Liaudet L., Soriano F. G. and Szabo C. (2000) Biology of nitric oxide signaling. Crit. Care Med. 28, N37–N52. Lopez-Figueroa M. O., Caamano C., Morano M. I., Ronn L. C., Akil H. and Watson S. J. (2000) Direct evidence of nitric oxide presence within mitochondria. Biochem. Biophys. Res. Commun. 272, 129– 133. Murray J., Taylor S. W., Zhang B., Ghosh S. S. and Capaldi R. A. (2003) Oxidative damage to mitochondrial complex I due to peroxynitrite: identification of reactive tyrosines by mass spectrometry. J. Biol. Chem. 278, 37223–37230. Nohl H., Staniek K., Sobhian B., Bahrami S., Redl H. and Kozlov A. V. (2000) Mitochondria recycle nitrite back to the bioregulator nitric monoxide. Acta Biochim. Pol. 47, 913–921. 950 Z. Lacza et al. 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951
Milclandris do nol pruduo:silti:oxak:951 Pacher P.Lindet I..llai P rr al (0)Prtere metiinperpiryrin per- od深am格aaee percaytitrie rather than ritn:0ikw exyritrik dooumpostion culalyst protects taast the develpmcn 38103-14 of dnerrubicin-indeend cankac dysfinctinn.Cuwao 117.K05- Schid L Renberkd T..Keioer M.Hem T.F.W.Wolf G ad 04. Augustin W.(2003)Nitk ooxide prodaced ia rat liver mitochondne Pearee LL.Katai A.1 Birder 1.A.,Pin B.k.and Peiron J.(2002) cns onidhtive enoa and ipairert of respiralin aner netecn The catabohe tre of nimc oxkde:the mmo oxide oxidase and P00a0d.17.214-2201 Peruynirile udivili uf yuchrume usic.Bd Sutto N..(19901 Kapid isulalion uf mctabusczlly aclive itilochantkn 271.134-1362 fmm rat hean ard sahnegiene ung Peroll dnsty gradient cer- Riobo N.A.Mdeni M.Sarjun N.Fisann M.L.,Grviell M.C.. he0人aeht55698J0m Camera M.C.Cadenas E.and Prdeman J.I (20002)The ahaC、山ry I.G,Mneler S.M%imareich k,cher P.Vinag modltan c milncherdrial niricuside rthes asviy in E L,Sariasp F.G,anDg上H,Millimns W,Salzan A.L国d brain developmen.dln.177,42447-424535. Gos人.T.2g2)nn上Pathopenenic le peroxyitme in th树 Rbb8.J,Chupen L D,g山K.J.Thuma A卫d Cutnor1R devdupmenl uf dsbds and diabi:vocelar tumpliclins (1999)Infuence of nic nocile on celnar and metochendrsl andies with IP15.a roel potet pemorynemme decorrpognon L.h.d&5刃-50 n-I7h Tatryan A.and Gulivi (1998)Prrificrim and checterization afa Rothe F..Hug P.L and Wolr G.(1999)Uliraspuctanal lealoatien siti-oside synchne from ret Iiver mbxoehondria.Biot Chent 273.11044-11043 rucleus of wild-rype and mise.94. Valdez L B.Aharez S,Araiz S.L..Schoprer F.Cameras M.C. 193-01. Aco0J.J.artl Borerb A.2000R▣uf purusyuirie i田 Roychowdry S,wfG,Kei发，0hD3d1mT.(2m1) the miochordrol momx /kee Rodv.Anl.Mvd.19.349-356. Poulosiua of primery glal oclb by grupe sced pruartloyaridin Zhag X.,Kim W.S..Halcher N.Pogicter K.Moouz LL,Gillete B xt2e时nmgo由twa守wg.r0d失17-149 and Sweedler J.V (20412)Ieterfering with ntric oxide meagare- Royehowdiury S.,Luhe A.KeiDoff G.,Wolf Gi.and Fom T.F.(2002) meats:4,5-diomnoflaooesccin sears with dcty Croxscoesic acid ad8hiei.【8ln17力，442-4% 2004 Intematonal Sociery foe Ngarocheristry.Mwochew.12004)90.942-95]
Pacher P., Liaudet L., Bai P. et al. (2003) Potent metalloporphyrin peroxynitrite decomposition catalyst protects against the development of doxorubicin-induced cardiac dysfunction. Circulation 107, 896– 904. Pearce L. L., Kanai A. J., Birder L. A., Pitt B. R. and Peterson J. (2002) The catabolic fate of nitric oxide: the nitric oxide oxidase and peroxynitrite reductase activities of cytochrome oxidase. J. Biol. Chem. 277, 13556–13562. Riobo N. A., Melani M., Sanjuan N., Fiszman M. L., Gravielle M. C., Carreras M. C., Cadenas E. and Poderoso J. J. (2002) The modulation of mitochondrial nitric-oxide synthase activity in rat brain development. J. Biol. Chem. 277, 42447–42455. Robb S. J., Gaspers L. D., Wright K. J., Thomas A. P. and Connor J. R. (1999) Influence of nitric oxide on cellular and mitochondrial integrity in oxidatively stressed astrocytes. J. Neurosci. Res. 56, 166–176. Rothe F., Huang P. L. and Wolf G. (1999) Ultrastructural localization of neuronal nitric oxide synthase in the laterodorsal tegmental nucleus of wild-type and knockout mice. Neuroscience 94, 193–201. Roychowdhury S., Wolf G., Keilhoff G., Bagchi D. and Horn T. (2001) Protection of primary glial cells by grape seed proanthocyanidin extract against nitrosative/oxidative stress. Nitric Oxide 5, 137–149. Roychowdhury S., Luthe A., Keilhoff G., Wolf G. and Horn T. F. (2002) Oxidative stress in glial cultures: detection by DAF-2 fluorescence used as a tool to measure peroxynitrite rather than nitric oxide. Glia 38, 103–114. Schild L., Reinheckel T., Reiser M., Horn T. F. W., Wolf G. and Augustin W. (2003) Nitric oxide produced in rat liver mitochondria causes oxidative stress and impairment of respiration after transient hypoxia. FASEB J. 17, 2194–2201. Sims N. R. (1990) Rapid isolation of metabolically active mitochondria from rat brain and subregions using Percoll density gradient centrifugation. J. Neurochem. 55, 698–707. Szabo C., Mabley J. G., Moeller S. M., Shimanovich R., Pacher P., Virag L., Soriano F. G., van Duzer J. H., Williams W., Salzman A. L. and Groves J. T. (2002) Part I: Pathogenetic role of peroxynitrite in the development of diabetes and diabetic vascular complications: studies with FP15, a novel potent peroxynitrite decomposition catalyst. Mol. Med. 8, 571–580. Tatoyan A. and Giulivi C. (1998) Purification and characterization of a nitric-oxide synthase from rat liver mitochondria. J. Biol. Chem. 273, 11044–11048. Valdez L. B., Alvarez S., Arnaiz S. L., Schopfer F., Carreras M. C., Poderoso J. J. and Boveris A. (2000) Reactions of peroxynitrite in the mitochondrial matrix. Free Radic. Biol. Med. 29, 349–356. Zhang X., Kim W. S., Hatcher N., Potgieter K., Moroz L. L., Gillette R. and Sweedler J. V. (2002) Interfering with nitric oxide measurements: 4,5-diaminofluorescein reacts with dehydroascorbic acid and ascorbic acid. J. Biol. Chem. 277, 48472–48478. Mitochondria do not produce nitric oxide 951 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 942–951