Brightness and shadows of mitochondrial ROS in the brain

Neu obiology o Disease 184 (2023) 106199 A ailable online 14 June 2023 0969-9961/© 2023 Published by Else ie Inc. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). B igh ness and shadows o mi ochond ial ROS in he b ain Daniel Jimenez-Blasco a , b , c , * , Angeles Almeida a , b , * , Juan P. Bola˜ nos a , b , c , * a Ins i u o de Biología Funcional y Gen´ omica (IBFG), Uni e sidad de Salamanca, CSIC, Salamanca, Spain b Ins i u o de In es igaci´ on Biom´ edica de Salamanca (IBSAL), Hospi al Uni e si a io de Salamanca, Uni e sidad de Salamanca, CSIC, Salamanca, Spain c Cen o de In es igaci´ on Biom´ edica en Red de F agilidad y En ejecimien o Saludable (CIBERFES), Mad id, Spain ARTICLE INFO Keywo ds: As ocy es Reac i e oxygen species Mi ochond ia Supe complexes Neu ons Glu a hione ABSTRACT Mi ochond ial eac i e oxygen species (mROS) ha e been gene ally conside ed ha m ul byp oduc s wan ed o clea when ele a ed o a oid b ain damage. Howe e , he abundance o mROS in as ocy es is e y high -abou one o de o magni ude abo e ha in neu ons-, despi e hey a e essen ial o p ese e cell me abolism and animal beha io . He e, we ha e ocused on his appa en ambigui y by discussing (i) he in insic mechanisms ac- coun ing o he highe p oduc ion o mROS by he mi ochond ial espi a o y chain in as ocy es han in neu ons, (ii) he speci ic molecula a ge s o as ocy ic bene icial mROS, and (iii) how dec eased as ocy ic mROS causes excess neu onal mROS leading o cellula and o ganismal damage. We hope ha his mini- e iew se es o cla i ying he appa en con o e sy on he bene icial e sus dele e ious aces o ROS in he b ain om molecula o highe -o de o ganismal le els. 1. P oduc ion o mi ochond ial ROS by as ocy es Mi ochond ia a e conside ed he main p oduce s o ROS wi hin he cell (Mu phy, 2009). Amongs he ple ho a o ROS sou ces wi hin he mi ochond ia (B and, 2010), i appea s ha a la ge pool comes om he mi ochond ial espi a o y chain (MRC) du ing oxida i e phospho yla- ion (OXPHOS) (Mu phy, 2009). Thus, MRC complexes couple he oxida ion o educed co ac o s (NADH o FADH 2 ) wi h p o on pumping o he mi ochond ial in e memb ane space, gene a ing he elec o- chemical g adien used by ATP syn hase o gene a e ATP (En iquez, 2016). To egula e he e iciency o his sys em, MRC complexes I, III and IV can be o ganized and assembled in o highe -o de s uc u es called supe complexes (SCs) (Chaban e al., 2014; En iquez, 2016). No ably, he supe -assembly o complexes I and III -a di e en s oi- chiome ies- inc eases elec on lux in o he MRC likely enhancing i s e iciency (Lapuen e-B un e al., 2013). Con e sely, disassembly o complex I om III leading o a highe p opo ion o ee complex I esul s in lowe MRC e iciency (Lapuen e-B un e al., 2013). The MRC may hus adop a ange o s uc u al con o ma ions wi hin a cell o issue o adap agains speci ic me abolic ci cums ances (Gua as e al., 2016). Fo ins ance, wi hin he b ain, neu ons show a cons i u i e p e e ence o supe -assemble complex I and III; in con as , as ocy es display a highe p opo ion o disassembled complex I (Lopez-Fabuel e al., 2016). The e o e, his di e en MRC con o ma ion is likely a e lec ion o dis inc me abolic si ua ions in neu ons and as ocy es, mo e OXPHOS dependen in he o me (Ala ian e al., 2011; Ku e e al., 2019; Almeida e al., 2001; Almeida e al., 2004; Bola˜ nos e al., 1995; He - e o-Mendez e al., 2009). In mammals, complex I consis s o 45 subuni s wi h a o al molecula weigh o ~980 KDa (Ca oll e al., 2003; Ca oll e al., 2006). All hese subuni s a e assembled o o m h ee unc ional modules, namely he N module, which is placed in he pe iphe al a m o he complex and ac- cep s elec ons om NADH, he Q module, which eleases elec ons o ubiquinone, and he P module, placed in he memb ane-embedded a m, whe e i is esponsible o he ansloca ion o p o ons in o he in e - memb ane space (Ca oll e al., 2003; Ca oll e al., 2006; Ma hiesen and Hage hall, 2002; Mimaki e al., 2012). No ably, complex I can d i e elec ons no only o ubiquinone, bu also o O 2 , o ming supe oxide anion (O 2 •- , he o iginal ROS), a unc ion ha is known o ake place a he N-module. Using di e en p opo ions o MRC complexes in Abb e ia ions: CB 1 , cannabinoid-1 ecep o ; Cys-Gly, cys einyl-glycine; G6PD, glucose 6-phospha e dehyd ogenase; HDAC4, his one deace ylase-4; HIF1 α , hyp- oxia-inducible ac o -1 α ; m CB1, mi ochond ial cannabinoid-1 ecep o ; mROS, mi ochond ial eac i e oxygen species; MRC, mi ochond ial espi a o y chain; NDUFS4, NADH:ubiquinone oxido educ ase subuni S4; NRF2, nuclea e y h oid- ela ed ac o 2; 3NP, 3-ni op opionic acid; OXPHOS, mi ochond ial oxida i e phospho yla ion sys em; PPP, pen ose-phospha e pa hway; ROS, eac i e oxygen species; SCs, supe complexes; THC, Δ 9 - e ahyd ocannabinol.. * Co esponding au ho s a : Ins i u o de Biología Funcional y Gen´ omica (IBFG), Uni e sidad de Salamanca, CSIC, Salamanca, Spain. E-mail add esses: [email p o ec ed] (D. Jimenez-Blasco), [email p o ec ed] (A. Almeida), [email p o ec ed] (J.P. Bola˜ nos). Con en s lis s a ailable a ScienceDi ec Neu obiology o Disease jou nal homepage: www.else ie .com/loca e/ynbdi h ps://doi.o g/10.1016/j.nbd.2023.106199 Recei ed 12 Feb ua y 2023; Recei ed in e ised o m 10 June 2023; Accep ed 12 June 2023 Neu obiology o Disease 184 (2023) 106199 2 liposomes, i was ound ha SCs o ma ion dec eased ROS p oduc ion by complex I (Ma anzana e al., 2013). This obse a ion was in e p e ed o be consequence o he lowe p obabili y o elec ons o be ans e ed om NADH o O 2 ( o ming O 2 •- ) by complex I, gi en he e icien lux o elec ons om NADH o ubiquinone in supe -assembled MRC (Ma - anzana e al., 2013). In good ag eemen wi h his hypo hesis, as ocy es -which show a highe p opo ion o disassembled complex I han neu- ons-, na u ally p oduce mo e ROS han neu ons (Lopez-Fabuel e al., 2016). I should be no ed ha his pa allelism is no a simple co ela ion, because he le els o ROS abundances in bo h cell ypes we e e e sed by gene ically al e ing he abili y o complex I o assemble wi h complex III (Lopez-Fabuel e al., 2016). Thus, he assembly o complex I wi h complex III de e mines complex I-media ed ROS p oduc ion, which in as ocy es occu s a one o de o magni ude abo e ha in neu ons (Fig. 1). 2. Molecula a ge s o as ocy ic mi ochond ial ROS A key ques ion ega ding he in iguing high mROS o ma ion in as ocy es would be unde s anding hei impac on cellula and animal wel a e. Add essing his ques ion implies he implemen a ion o a sys- em ha is able o swi ch o mROS abundance in as ocy es in i o o assess i s consequences. This was achie ed by gene ically enginee ing an inducible mouse model ha exp esses he an ioxidan enzyme ca alase in he mi ochond ial ma ix o as ocy es in adul hood (Vicen e- Gu ie ez e al., 2019). E icien a enua ion o mROS in as ocy es dis- played a p o ound al e a ion in he ansc ip ional inge p in o as o- cy es and in he b ain me abolomics signa u e. Essen ially, i was ound inc eased he ibose-5-phospha e/6-phosphoglucona e a io and dec eased he py u a e/glucose a io, indica ing inc eased pen ose- phospha e pa hway (PPP) and dec eased glycolysis, a me abolic swi ch ha was con i med by lux analysis (Vicen e-Gu ie ez e al., 2019). In e es ingly, exp ession o glucose-6-phospha e dehyd ogenase (G6PD), he i s and a e-limi ing s ep o he PPP, was inc eased in he mROS-ab oga ed as ocy es (Vicen e-Gu ie ez e al., 2019), indica ing ha his enzyme migh be a di ec o indi ec mROS a ge . G6PD is known o be ansc ip ionally ep essed by he mic o-RNA 206 (miR- 206) (Coda e al., 2015), he exp ession o which is in u n ep essed by he nuclea abundance o he edox senso ansc ip ional modula o , his one deace ylase-4 (HDAC4) (Winbanks e al., 2011). No ably, miR- 206 le els in mROS-a enua ed as ocy es we e dec eased (Vicen e- Gu ie ez e al., 2019), an e ec ha was escued by o e -exp ession o miR-206 (Vicen e-Gu ie ez e al., 2019). Mo eo e , nuclea abundance o HDAC4 was highe in mROS-dec eased as ocy es, indica ing ha he educed (ac i e) o m o HDAC4 was p edominan in hese cells. Finally, inhibi ing HDAC ac i i y wi h ichos a in A p e en ed he inc ease in G6PD in mROS-a enua ed as ocy es. Taken oge he , hese obse a- ions sugges ha a physiological signaling ole o mROS in as ocy es would be o keep HDAC4 oxidized and seques e ed in he cy osol o allow miR-206 exp ession, which des abilizes G6PD and dec eases PPP ac i i y, in u n sus aining glycolysis (Vicen e-Gu ie ez e al., 2019) (Fig. 1). Amongs o he unc ions, PPP main ains he NADPH(H + )/NADP + a io, which was ound o be dec eased in mROS-impo e ished as o- cy es, sugges ing inc eased NADPH(H + ) consump ion (Vicen e-Gu ie - ez e al., 2019). NADPH(H + ) oxidases (NOXs), which p oduce supe oxide anion (O 2 •- ), a e highly ac i e NADPH(H + ) consuming en- zymes (Naye nia e al., 2014). NOX-1 and -2 we e ound inc eased in he ansc ip omic analysis o he mROS-lowe ed as ocy es (Vicen e- Gu ie ez e al., 2019). In e es ingly, hese NOX-1 and -2 a e localized in he plasma memb ane wi h hei ac i e si e acing o he ex acellula space. Indeed, as ocy es dep i ed o mROS show enhanced ex acel- lula O 2 •- (Vicen e-Gu ie ez e al., 2019). Hence, as ocy ic mROS physiologically down egula e NOXs exp ession and keeps ex acellula O 2 •- p oduc ion low. Howe e , how mROS downmodula es NOXs exp ession? I should be men ioned ha NOX-1 and -2 exp ession a e down egula ed by he nuclea ac o -e y h oid 2- ela ed ac o -2 (NRF2) (Ko ac e al., 2015; Pendyala and Na a ajan, 2010), a ansc ip ion ac o known o be ac i a ed by ROS o go e n he exp ession o a wide ba e y o an ioxidan genes aimed o p o ec agains cellula edox s ess (Hayes and McMahon, 2009; Tebay e al., 2015). NRF2 is ep essed by Kelch-like ECH-associa ed p o ein 1 (KEAP1)-Cullin 3 Fig. 1. Physiological e sus pa hological oles o mi ochond ial ROS (mROS) in as ocy es and neu ons. Le . Unde physiological condi ions, mROS gene a ion by he mi ochond ial espi a o y chain o as ocy es is high, in pa due o he disassembly o complex I (I) om complex III (III). Howe e , in neu ons, complex I is mo e assembled wi h complex III esul ing in lowe mROS gene a ion. High as ocy ic mROS sus ains ac i e ansc ip ion ac o s HIF1 (hypoxia-inducible ac o -1), which con ibu e o p ese e glycolysis and NRF2 (nuclea ac o -e y h oid 2- ela ed ac o -2), which p omo es he exp ession o γ-glu amyl-cys eine syn he ase (γ-GCS) and hence glu a hione (GSH) biosyn hesis; no ably, by ep essing he exp ession o NADH oxidases-1 and -2 (NOX-1, −2), as ocy ic mROS down egula es ex a- cellula O 2 •- . Glycoly ically- eleased lac a e and GSH-de i ed Cys-Gly can be aken up by neu ons, whe e hey a e used o uel OXPHOS (oxida i e phospho yla ion) and p ese e he an ioxidan s a us, espec i ely, main aining o ganismal wel a e. Righ . Complex I des abiliza ion in as ocy es lead o mROS educ ion. Ei he by his mechanism o by selec i e sca enging o mROS in as ocy es, he educ ion o as ocy ic mROS leads o he e e sal o HIF1 and NRF2 ansc ip ional ac i i y, causing educed lac a e and Cys-Gly elease, and inc eased O 2 •- . In addi ion, complex I disassembly om complex III in neu ons, which has been obse ed in Pa - kinson’s disease, in insically may con ibu e o pa hological high neu onal mROS and OXPHOS ine iciency. Al oge he , hese ac o s con ibu e o neu onal ene gy ailu e and edox s ess, causing o ganismal damage. D. Jimenez-Blasco e al. Neu obiology o Disease 184 (2023) 106199 3 (CUL3)-RING-box p o ein 1 (Rbx1), an E3 ubiqui in ligase ha a ge s NRF2 o p o easomal deg ada ion (Tebay e al., 2015). In e es ingly, he p o ein exp ession o CUL3 and KEAP1 a e e y low in as ocy es when compa ed wi h neu ons (Bell e al., 2015; Jimenez-Blasco e al., 2015), which explains he cons i u i ely highe ac i i y o NRF2 in as- ocy es (Bax e e al., 2015; Jimenez-Blasco e al., 2015). Mo eo e , ROS oxidizes KEAP1 o inac i e i , p omo ing NRF2 ansc ip ional ac- i i y (Tebay e al., 2015). The e o e, i is emp ing o specula e ha he na u ally high mROS abundance in as ocy es con ibu e o p ese e NRF2 ansc ip ional ac i i y, hence explaining he inc eased NOX-1 and -2 exp ession in he mROS-a enua ed as ocy es (Vicen e-Gu ie - ez e al., 2019). Indeed, exp essing a cons i u i ely ac i e NRF2 iso o m in mROS-impo e ished as ocy es was su icien o escue he inc eased NOX-1 and -2 mRNA abundances and he enhanced ex acellula O 2 •- elease. Al oge he , hese obse a ions s ongly sugges ha mROS in as ocy es physiologically downmodula e he elease o po en ially damaging O 2 •- in o he ex acellula space (Fig. 1). 3. Neu opa hological consequences o dec eased as ocy ic mROS In good ag eemen wi h he obse a ions ha mROS p oduc ion by as ocy es igge s he ac i a ion o molecula mechanisms aimed o p ese e he me abolic and edox s a us o hese cells, loss o mROS causes b ain damage. In essence, dis up ion o physiological mROS in as ocy es weakens he signaling pa hways ha sus ain NRF2- an ioxidan sys em and glycolysis (Jimenez-Blasco e al., 2020; Vice- n e-Gu ie ez e al., 2019). In consequence, neighbo neu ons become dys unc ional e en ually leading o damage and highe -o de al e a ions. By ab oga ing as ocy ic mROS using he mi ochond ial-speci ic exp ession o ca alase, NRF2-media ed an ioxidan sys em is down- egula ed esul ing in he impai men o glu a hione (GSH) biosyn hesis (Vicen e-Gu ie ez e al., 2019). GSH, which is e y abundan in as o- cy es (Bola˜ nos e al., 1996; Maka e al., 1994), is clea ed in hese cells in o he dipep ide cys einyl-glycine (Cys-Gly) by he ac ion o he ec oenzyme γ-glu amyl- anspep idase o elease he dipep ide in o he ex acellula space (Saga a e al., 1993). Cys-Gly can hen be aken up by neu ons, which, being unable o ake up cys eine (Saga a e al., 1993), use he dipep ide and glu ama e o syn hesize GSH (D ingen e al., 1999) and hence o sus ain neu onal an ioxidan p o ec ion (D ingen e al., 1999; Jimenez-Blasco e al., 2015). Acco dingly, when as ocy ic mROS a e ab oga ed, he NRF2-media ed GSH shu ling om as ocy es o neu ons is impai ed, esul ing in neu onal GSH loss, edox s ess, cellula damage and cogni i e impai men in mice (Vicen e-Gu ie ez e al., 2019). F om hese obse a ions, i could be concluded ha , unde a physiological poin o iew, as ocy es p ese e a na u ally occu ing high mROS o ma ion aimed o sus ain he an ioxidan machine y, whe eas neu ons a e weak mROS p oduce s, hence equi ing he con inuous supply o an ioxidan s om neighbo as ocy es. This is suppo ed by indings ob ained using he mi ochond ial- agged ca alase gene ic exp ession in as ocy es o in neu ons in i o upon a edox s ess s imulus. Thus, p e en ing he inc ease in mROS in as ocy es does no a oid he mo o discoo dina ion caused by he in ape i oneal injec ion o he neu o oxin 3-ni op opionic acid (3NP); howe e , p e en ing he inc ease in mROS in neu ons ully a oids 3NP-media ed mo o dis- coo dina ion (Vicen e-Gu ie ez e al., 2021). Hence, high mROS in as- ocy es is physiological, so i s loss causes neu opa hology; in con as , low mROS in neu ons is physiological, so i s inc ease causes damage (Fig. 1). O no e, complex I disassembly om complex III has been obse ed in Pa kinson’s disease pa ien s and in neu ons om a mouse model o his diso de (Lopez-Fabuel e al., 2017) indica ing ha he co ec MRC assembly in neu ons impac s on OXPHOS e iciency and damaging mROS o ma ion. Ano he piece o e idence ha con i ms he physiological impo - ance o high o ma ion o mROS by as ocy es is shown upon ch onic adminis a ion o cannabinoids. Thus, pe sis en cannabis consump ion is well known o cause loss o social in e ac ion, amongs o he psy- choses (Busque s-Ga cia e al., 2022). Howe e , he molecula mecha- nism ha igge s his e ec has long been elusi e. The key obse a ion ha cannabinoid ecep o -1 (CB1) a e exp essed in mi ochond ia (m CB1), whe e hei ac i a ion causes inhibi ion o mi ochond ial complex I (Bena d e al., 2012), led o ollowing s udy showing ha , by inhibi ing complex I, adminis a ion o cannabinoid agonis s, such as Δ 9 - e ahyd ocannabinol (THC), igge ed cogni i e impai men in mice (Hebe -Cha elain e al., 2016). In e es ingly, m CB1 ecep o s a e also p esen in as ocy es (Gu ie ez-Rod iguez e al., 2018), he ac i a ion o which also leads o complex I inhibi ion (Jimenez-Blasco e al., 2020). The mechanism causing complex I inhibi ion was un eiled using na i e elec opho esis o he MRC complexes and supe complexes (Jimenez- Blasco e al., 2020). In essence, i was ound ha pe sis en m CB1 ac i a ion in as ocy es causes loss o he N module o complex I. Mechanis ically, m CB1 ecep o s, by inhibi ing p o ein kinase A, causes dephospho yla ion o many p o eins, including he complex I N-module subuni , NDUFS4 (NADH:ubiquinone oxido educ ase subuni S4) (Jimenez-Blasco e al., 2020). In ac , p o ein kinase A was p e iously known o phospho yla e NDUFS4 subuni a i s Se 173 posi ion as an essen ial s ep in he assembly o he N-module (De Rasmo e al., 2008). Consequen ly, ch onic ac i a ion o m CB1 in as ocy es no only in- hibi s he abili y o complex I o educe ubiquinone -hence inhibi ing he ene gy conse ing ole o complex I -, bu i also abla es i s abili y o physiologically gene a e ROS (Jimenez-Blasco e al., 2020). This leads o dec eased mROS gene a ion ha esul s in loss o ac i a ion o hypoxia- inducible ac o -1 α (HIF1 α ), known o be s abilized by ROS o p omo e he ansc ip ional ac i i y o HIF1 (Pa en e al., 2010). Since HIF1 is an impo an con ibu ion ac o ha sus ains he ansc ip ional exp es- sion o he glycoly ic machine y (Semenza e al., 1994), i s inhibi ion a e ROS a enua ion con ibu es o a dec ease in he glycoly ic lux (Jimenez-Blasco e al., 2020). Impo an ly, he impai ed glycoly ic lux a ec s he elease o lac a e, which canno be p ope ly aken up by neu ons, acco ding o he as ocy e-neu ons lac a e shu le model (Machle e al., 2016; Pelle in and Magis e i, 1994; Suzuki e al., 2011). E en ually, his scena io causes neu onal dys unc ion ha , upon THC adminis a ion is ma e ialized as a loss o social in e ac ion (Jimenez-Blasco e al., 2020). Indeed, in ace eb al injec ion o lac a e e icien ly escues he loss o social in e ac ion caused by ch onic THC adminis a ion in mice (Jimenez-Blasco e al., 2020), in good ag eemen wi h p e ious esul s showing an idep essan e ec s by pe iphe al adminis a ion o lac a e (Ca a d e al., 2018). These indings indica e ha , by dec easing he na u al high abundance o mROS, m CB 1 e- cep o pe sis en ac i a ion impai s as oglial glucose me abolism impac ing on neu onal ac i i y, which leads o a pa hological ci cum- s ance like loss o social in e ac ion. 4. Concluding ema ks and u u e pe spec i es In conclusion, his ocused e iew aims o con ibu e o ou be e unde s anding on he molecula mechanisms esponsible o he physi- ological gene a ion and signaling oles o mROS by as ocy es, ac- co ding o in i o wo k. Ob iously, gi en he pleio opic e ec s o ROS, he e would be many o he speci ic a ge s ha could accoun o he p ese a ion o cellula , issula and o ganismal unc ions by as ocy ic mROS, which su ely will be elucida ed in he u u e. By ma ching mo- lecula mROS a ge s wi h neu onal unc ions, i will be possible o dissec ou speci ic pa hways ha link molecules wi h o ganismal highe -o de beha io . Fu he mo e, la ge p oduc ion o sus ained mROS may bo h be bene icial o dele e ious, a balance ha is de e - mined by he cellula ype and he me abolic condi ion (Fig. 1). In e - es ingly, as ocy ic mROS u he inc ease du ing physiological hypoxia as pa o a possible pO 2 sensing mechanism (Angelo a e al., 2015). The e o e, accu a ely unde s anding how high physiological mROS le els exe s bene icial e ec s is impo an when designing an ioxidan D. Jimenez-Blasco e al. Neu obiology o Disease 184 (2023) 106199 4 s a egies o comba diseases in which bu s s o mROS -likely happening in neu ons- a e dele e ious, since sca enging o such sudden inc eases in mROS indisc imina ely -i.e., a ec ing as ocy ic mROS- may pa adoxi- cally esul in loss o an ioxidan p e en ion a he long e m conse- quen ly causing delayed neu al damage. Fu u e wo k would be necessa y o design cell-speci ic he apeu ic s a egies o sca enge mROS in neu ons ansien ly wi hou a ec ing he na u al p oduc ion o mROS by neighbo as ocy es. Funding in o ma ion Au ho s we e suppo ed by he Nex Gene a ionEU/PRTR and Agencia Es a al de In es igaci´ on (10.13039/501100011033; PID2019- 105699RB-I00; PID2022-138813OB-I00; PDC2021–121013-I00; RED2018–102576-T o JPB; RED2022–134407-T o AA); Plan Nacional de D ogas (2020I028 o JPB); Ins i u o de Salud Ca los III (CB16/10/ 00282 o JPB; PI21/00727; PMP22/00084 and RD21/0006/0005 co- unded by he Eu opean Union, o AA); Ma ie Sklodowska-Cu ie Ac- ion HORIZON-MSCA-2021-DN-01 (101072759 o JPB and AAP), and Jun a de Cas illa y Le´ on (CS/151P20 co- unded by P.O. FEDER o AA; Apoyo Regional a la Compe i i idad Emp esa ial, ICE 04/18/LE/0017 o JPB, and Escale a de Excelencia CLU-2017-03 o JPB and AA). DJB is a ecipien o a Juan de la Cie a-Inco po aci´ on con ac (IJC2020–044230-I). Au ho con ibu ion JPB concep ualiza ion. DJB Da a cu a ion, W i ing - o iginal d a . JPB and AA w i ing - e iew & edi ing. Decla a ion o Compe ing In e es The au ho s decla e ha hey ha e no con lic s o in e es wi h he con en s o his a icle. Da a a ailabili y No da a was used o he esea ch desc ibed in he a icle. Re e ences Ala ian, K.N., Li, H., Collis, L., Bonanni, L., Zeng, L., Sacche i, S., e al., 2011. Bcl-xL egula es me abolic e iciency o neu ons h ough in e ac ion wi h he mi ochond ial F1FO ATP syn hase. Na . Cell Biol. 13, 1224–1233. h ps://doi.o g/10.1038/ ncb2330. Almeida, A., Almeida, J., Bola˜ nos, J.P., Moncada, S., 2001. Di e en esponses o as ocy es and neu ons o ni ic oxide: he ole o glycoly ically-gene a ed ATP in as ocy e p o ec ion. P oc. Na l. Acad. Sci. U. S. A. 98, 15294–15299. h ps://doi. o g/10.1073/pnas.261560998. Almeida, A., Moncada, S., Bola˜ nos, J.P., 2004. Ni ic oxide swi ches on glycolysis h ough he AMP p o ein kinase and 6-phospho uc o-2-kinase pa hway. Na . Cell Biol. 6, 45–51. h ps://doi.o g/10.1038/ncb1080. Angelo a, P.R., Kasymo , V., Ch is ie, I., Sheikhbahaei, S., Tu o sky, E., Ma ina, N., e al., 2015. Func ional oxygen sensi i i y o as ocy es. J. Neu osci. 35, 10460–10473. h ps://doi.o g/10.1523/JNEUROSCI.0045-15.2015. Bax e , P.S., Bell, K.F., Hasel, P., Kaindl, A.M., F icke , M., Thomson, D., e al., 2015. Synap ic NMDA ecep o ac i i y is coupled o he ansc ip ional con ol o he glu a hione sys em. Na . Commun. 6, 6761. h ps://doi.o g/10.1038/ncomms7761. Bell, K.F., Al-Muba ak, B., Ma el, M.A., McKay, S., Wheelan, N., Hasel, P., e al., 2015. Neu onal de elopmen is p omo ed by weakened in insic an ioxidan de ences due o epigene ic ep ession o N 2. Na . Commun. 6, 7066. h ps://doi.o g/10.1038/ ncomms8066. Bena d, G., Massa, F., Puen e, N., Lou enco, J., Bellocchio, L., So ia-Gomez, E., e al., 2012. Mi ochond ial CB(1) ecep o s egula e neu onal ene gy me abolism. Na . Neu osci. 15, 558–564. h ps://doi.o g/10.1038/nn.3053. Bola˜ nos, J.P., Heales, S.J.R., Land, J.M., Cla k, J.B., 1995. E ec o pe oxyni i e on he mi ochond ial espi a o y chain: di e en ial suscep ibili y o neu ones and as ocy es in p ima y cul u es. J. Neu ochem. 64, 1965–1972. h ps://doi.o g/ 10.1046/j.1471-4159.1995.64051965.x. Bola˜ nos, J.P., Heales, S.J.R., Peuchen, S., Ba ke , J.E., Land, J.M., Cla k, J.B., 1996. Ni ic oxide-media ed mi ochond ial damage: a po en ial neu op o ec i e ole o glu a hione. F ee Radic. Biol. Med. 21, 995–1001. h ps://doi.o g/10.1016/s0891- 5849(96)00240-7. B and, M.D., 2010. The si es and opology o mi ochond ial supe oxide p oduc ion. Exp. Ge on ol. 45, 466–472. h ps://doi.o g/10.1016/j.exge .2010.01.003. Busque s-Ga cia, A., Bolanos, J.P., Ma sicano, G., 2022. Me abolic Messenge s: endocannabinoids. Na . Me ab. 4, 848–855. h ps://doi.o g/10.1038/s42255-022- 00600-1. Ca a d, A., Elsayed, M., Ma gineanu, M., Bou y-Jamo , B., F agnie e, L., Meylan, E.M., e al., 2018. Pe iphe al adminis a ion o lac a e p oduces an idep essan -like e ec s. Mol. Psychia y 23, 488. h ps://doi.o g/10.1038/mp.2016.237. Ca oll, J., Fea nley, I.M., Shannon, R.J., Hi s , J., Walke , J.E., 2003. Analysis o he subuni composi ion o complex I om bo ine hea mi ochond ia. Mol. Cell. P o eomics 2, 117–126. h ps://doi.o g/10.1074/mcp.M300014-MCP200. Ca oll, J., Fea nley, I.M., Skehel, J.M., Shannon, R.J., Hi s , J., Walke , J.E., 2006. Bo ine complex I is a complex o 45 di e en subuni s. J. Biol. Chem. 281, 32724–32727. h ps://doi.o g/10.1074/jbc.M607135200. Chaban, Y., Boekema, E.J., Dudkina, N.V., 2014. S uc u es o mi ochond ial oxida i e phospho yla ion supe complexes and mechanisms o hei s abilisa ion. Biochim. Biophys. Ac a 1837, 418–426. h ps://doi.o g/10.1016/j.bbabio.2013.10.004. Coda, D.M., Lingua, M.F., Mo ena, D., Foglizzo, V., Be sani, F., Ala, U., e al., 2015. SMYD1 and G6PD modula ion a e c i ical e en s o miR-206-media ed di e en ia ion o habdomyosa coma. Cell Cycle 14, 1389–1402. h ps://doi.o g/ 10.1080/15384101.2015.1005993. De Rasmo, D., Panelli, D., Sa danelli, A.M., Papa, S., 2008. cAMP-dependen p o ein kinase egula es he mi ochond ial impo o he nuclea encoded NDUFS4 subuni o complex I. Cell. Signal. 20, 989–997. h ps://doi.o g/10.1016/j. cellsig.2008.01.017. D ingen, R., P ei e , B., Hamp ech , B., 1999. Syn hesis o he an ioxidan glu a hione in neu ons: supply by as ocy es o CysGly as p ecu so o neu onal glu a hione. J. Neu osci. 19, 562–569. h ps://doi.o g/10.1523/JNEUROSCI.19-02-00562.1999. En iquez, J.A., 2016. Sup amolecula O ganiza ion o Respi a o y Complexes. Annu. Re . Physiol. 78, 533–561. h ps://doi.o g/10.1146/annu e -physiol-021115- 105031. Gua as, A., Pe ales-Clemen e, E., Cal o, E., Acin-Pe ez, R., Lou ei o-Lopez, M., Pujol, C., e al., 2016. The CoQH2/CoQ a io se es as a senso o espi a o y chain e iciency. Cell Rep. 15, 197–209. h ps://doi.o g/10.1016/j.cel ep.2016.03.009. Gu ie ez-Rod iguez, A., Bonilla-Del Rio, I., Puen e, N., Gomez-U quijo, S.M., Fon aine, C.J., Egana-Hugue , J., e al., 2018. Localiza ion o he cannabinoid ype-1 ecep o in subcellula as ocy e compa men s o mu an mouse hippocampus. Glia. 66, 1417–1431. h ps://doi.o g/10.1002/glia.23314. Hayes, J.D., McMahon, M., 2009. NRF2 and KEAP1 mu a ions: pe manen ac i a ion o an adap i e esponse in cance . T ends Biochem. Sci. 34, 176–188. h ps://doi.o g/ 10.1016/j. ibs.2008.12.008. Hebe -Cha elain, E., Desp ez, T., Se a , R., Bellocchio, L., So ia-Gomez, E., Busque s- Ga cia, A., e al., 2016. A cannabinoid link be ween mi ochond ia and memo y. Na u e. 539, 555–559. h ps://doi.o g/10.1038/na u e20127. He e o-Mendez, A., Almeida, A., Fe nandez, E., Maes e, C., Moncada, S., Bolanos, J.P., 2009. The bioene ge ic and an ioxidan s a us o neu ons is con olled by con inuous deg ada ion o a key glycoly ic enzyme by APC/C-Cdh1. Na . Cell Biol. 11, 747–752. h ps://doi.o g/10.1038/ncb1881. Jimenez-Blasco, D., San o imia-Cas ano, P., Gonzalez, A., Almeida, A., Bolanos, J.P., 2015. As ocy e NMDA ecep o s’ ac i i y sus ains neu onal su i al h ough a Cdk5-N 2 pa hway. Cell Dea h Di e . 22, 1877–1889. h ps://doi.o g/10.1038/ cdd.2015.49. Jimenez-Blasco, D., Busque s-Ga cia, A., Hebe -Cha elain, E., Se a , R., Vicen e- Gu ie ez, C., Ioannidou, C., e al., 2020. Glucose me abolism links as oglial mi ochond ia o cannabinoid e ec s. Na u e. 583, 603–608. h ps://doi.o g/ 10.1038/s41586-020-2470-y. Ko ac, S., Angelo a, P.R., Holms om, K.M., Zhang, Y., Dinko a-Kos o a, A.T., Ab amo , A.Y., 2015. N 2 egula es ROS p oduc ion by mi ochond ia and NADPH oxidase. Biochim. Biophys. Ac a 1850, 794–801. h ps://doi.o g/10.1016/j. bbagen.2014.11.021. Ku e , K.Z., Olech, T., Denche , N.A., 2019. Inc eased ene ge ic demand suppo ed by mi ochond ial elec on ans e chain and as ocy e assis ance is essen ial o main ain he compensa o y abili y o he dopamine gic neu ons in an animal model o ea ly Pa kinson’s disease. Mi ochond ion 47, 227–237. h ps://doi.o g/10.1016/ j.mi o.2018.12.002. Lapuen e-B un, E., Mo eno-Loshue os, R., Acin-Pe ez, R., La o e-Pellice , A., Colas, C., Balsa, E., e al., 2013. Supe complex assembly de e mines elec on lux in he mi ochond ial elec on anspo chain. Science. 340, 1567–1570. h ps://doi.o g/ 10.1126/science.1230381. Lopez-Fabuel, I., Le Douce, J., Logan, A., James, A.M., Bon en o, G., Mu phy, M.P., e al., 2016. Complex I assembly in o supe complexes de e mines di e en ial mi ochond ial ROS p oduc ion in neu ons and as ocy es. P oc. Na l. Acad. Sci. U. S. A. 113, 13063–13068. h ps://doi.o g/10.1073/pnas.1613701113. Lopez-Fabuel, I., Ma in-Ma in, L., Resch-Beushe , M., Azkona, G., Sanchez- Pe nau e, R., Bola˜ nos, J.P., 2017. Mi ochond ial espi a o y chain diso ganiza ion in Pa kinson’s disease- ele an PINK1 and DJ1 mu an s. Neu ochem. In . 109, 101–105. h ps://doi.o g/10.1016/j.neuin .2017.03.023. Machle , P., Wyss, M.T., Elsayed, M., S oba , J., Gu ie ez, R., on Fabe -Cas ell, A., e al., 2016. In i o e idence o a lac a e g adien om as ocy es o neu ons. Cell Me ab. 23, 94–102. h ps://doi.o g/10.1016/j.cme .2015.10.010. Maka , T.K., Nede gaa d, M., P euss, A., Gelba d, A.S., Pe umal, A.S., Coope , A.J.L., 1994. Vi amin E, asco ba e, glu a hione, glu a hione disul ide, and enzymes o glu a hione me abolism in cul u es o chick as ocy es and neu ones: e idence ha as ocy es play an impo an ole in an ioxida i e p ocesses in he b ain. J. Neu ochem. 62, 45–53. h ps://doi.o g/10.1046/j.1471-4159.1994.62010045.x. D. Jimenez-Blasco e al. Neu obiology o Disease 184 (2023) 106199 5 Ma anzana, E., Ba be o, G., Falasca, A.I., Lenaz, G., Geno a, M.L., 2013. Mi ochond ial espi a o y supe complex associa ion limi s p oduc ion o eac i e oxygen species om complex I. An ioxid. Redox Signal. 19, 1469–1480. h ps://doi.o g/10.1089/ a s.2012.4845. Ma hiesen, C., Hage hall, C., 2002. T ansmemb ane opology o he NuoL, M and N subuni s o NADH:quinone oxido educ ase and hei homologues among memb ane- bound hyd ogenases and bona ide an ipo e s. Biochim. Biophys. Ac a 1556, 121–132. h ps://doi.o g/10.1016/s0005-2728(02)00343-2. Mimaki, M., Wang, X., McKenzie, M., Tho bu n, D.R., Ryan, M.T., 2012. Unde s anding mi ochond ial complex I assembly in heal h and disease. Biochim. Biophys. Ac a 1817, 851–862. h ps://doi.o g/10.1016/j.bbabio.2011.08.010. Mu phy, M.P., 2009. How mi ochond ia p oduce eac i e oxygen species. Biochem. J. 417, 1–13. h ps://doi.o g/10.1042/BJ20081386. Naye nia, Z., Jaque , V., K ause, K.H., 2014. New insigh s on NOX enzymes in he cen al ne ous sys em. An ioxid. Redox Signal. 20, 2815–2837. h ps://doi.o g/10.1089/ a s.2013.5703. Pa en, D.A., La leu , V.N., Robi aille, G.A., Chan, D.A., Giaccia, A.J., Richa d, D.E., 2010. Hypoxia-inducible ac o -1 ac i a ion in nonhypoxic condi ions: he essen ial ole o mi ochond ial-de i ed eac i e oxygen species. Mol. Biol. Cell 21, 3247–3257. h ps://doi.o g/10.1091/mbc.E10-01-0025. Pelle in, L., Magis e i, P.J., 1994. Glu ama e up ake in o as ocy es s imula es ae obic glycolysis: a mechanism coupling neu onal ac i i y o glucose u iliza ion. P oc. Na l. Acad. Sci. U. S. A. 91, 10625–10629. h ps://doi.o g/10.1073/pnas.91.22.10625. Pendyala, S., Na a ajan, V., 2010. Redox egula ion o Nox p o eins. Respi . Physiol. Neu obiol. 174, 265–271. h ps://doi.o g/10.1016/j. esp.2010.09.016. Saga a, J., Miu a, K., Bannai, S., 1993. Main enance o neu onal glu a hione by glial cells. J. Neu ochem. 61, 1672–1676. h ps://doi.o g/10.1111/j.1471-4159.1993. b09802.x. Semenza, G.L., Ro h, P.H., Fang, H.M., Wang, G.L., 1994. T ansc ip ional egula ion o genes encoding glycoly ic enzymes by hypoxia-inducible ac o 1. J. Biol. Chem. 269, 23757–23763. Suzuki, A., S e n, S.A., Bozdagi, O., Hun ley, G.W., Walke , R.H., Magis e i, P.J., Albe ini, C.M., 2011. As ocy e-neu on lac a e anspo is equi ed o long- e m memo y o ma ion. Cell. 144, 810–823. h ps://doi.o g/10.1016/j.cell.2011.02.018. Tebay, L.E., Robe son, H., Du an , S.T., Vi ale, S.R., Penning, T.M., Dinko a-Kos o a, A. T., Hayes, J.D., 2015. Mechanisms o ac i a ion o he ansc ip ion ac o N 2 by edox s esso s, nu ien cues, and ene gy s a us and he pa hways h ough which i a enua es degene a i e disease. F ee Radic. Biol. Med. 88, 108–146. h ps://doi. o g/10.1016/j. ee adbiomed.2015.06.021. Vicen e-Gu ie ez, C., Bono a, N., Bobo-Jimenez, V., Jimenez-Blasco, D., Lopez- Fabuel, I., Fe nandez, E., e al., 2019. As ocy ic mi ochond ial ROS modula e b ain me abolism and mouse beha iou . Na . Me ab. 1, 201–211. h ps://doi.o g/ 10.1038/s42255-018-0031-6. Vicen e-Gu ie ez, C., Bono a, N., Jimenez-Blasco, D., Lopez-Fabuel, I., Ba es, G., Mu phy, M.P., e al., 2021. Ab oga ing mi ochond ial ROS in neu ons o as ocy es e eals cell-speci ic impac on mouse beha iou . Redox Biol. 41, 101917 h ps://doi. o g/10.1016/j. edox.2021.101917. Winbanks, C.E., Wang, B., Beye , C., Koh, P., Whi e, L., Kan ha idis, P., G ego e ic, P., 2011. TGF-be a egula es miR-206 and miR-29 o con ol myogenic di e en ia ion h ough egula ion o HDAC4. J. Biol. Chem. 286, 13805–13814. h ps://doi.o g/ 10.1074/jbc.M110.192625. D. Jimenez-Blasco e al.