Prolyl Hydroxylase-dependent Modulation of Eukaryotic Elongation Factor 2 Activity and Protein Translation under Acute Hypoxia

P olyl Hyd oxylase-dependen Modula ion o Euka yo ic Elonga ion Fac o 2 Ac i i y and P o ein T ansla ion unde Acu e Hypoxia * □ S Recei ed o publica ion, Augus 30, 2011, and in e ised o m, Janua y 14, 2012 Published, JBC Pape s in P ess, Feb ua y 3, 2012, DOI 10.1074/jbc.M111.299180 An onio Rome o-Ruiz ‡1,2 , Lucía Bau is a ‡1 , Vi ginia Na a o § , An onio He as-Ga ín ‡ , Rosana Ma ch-Díaz ‡ , An onio Cas ellano ‡ , Raquel Gómez-Díaz ‡ , Ma ía J. Cas o ‡ , Edu ne Be a § , José López-Ba neo ‡3 , and Albe o Pascual ‡4 F om he ‡ Ins i u o de Biomedicina de Se illa, Hospi al Uni e si a io Vi gen del Rocío/CSIC/Uni e sidad de Se illa, 41013 Se ille and he § Cen o de In es igación Coope a i a en Biociencias, CIC bioGUNE, Pa que Tecnológico de Bizkaia, 48160 De io, Spain Backg ound: T ansla ional a es is a classical cellula esponse o hypoxia, he unde lying mechanisms o which a e unknown. Resul s: Inhibi o y phospho yla ion o euka yo ic elonga ion ac o 2 by acu e hypoxia depends on oxygen-sensi i e p olyl hyd oxylases (PHDs). Conclusion: The elonga ion phase o p o ein syn hesis is egula ed by PHDs. Signi icance: This wo k un a els a no el cellula p ocess con olled by PHDs, po en ial pha macological a ge s in se e al human diseases. Ea ly adap i e esponses o hypoxia a e essen ial o cell su - i al, bu hei na u e and unde lying mechanisms a e poo ly known. We ha e s udied he pos - ansc ip ional changes in he p o eome o mammalian cells elici ed by acu e hypoxia and ound ha phospho yla ion o euka yo ic elonga ion ac o 2 (eEF2), a ibosomal anslocase whose phospho yla ion inhibi s p o ein syn hesis, is unde he p ecise and e e sible con ol o O 2 ension. Upon exposu e o hypoxia, phospho yla ion o eEF2 a Th 56 occu ed apidly (<15 min) and esul ed in modes ansla ional a es , a undamen al homeos a ic esponse o hypoxia ha spa es ATP and hus acili a es cell su i al. Acu e inhibi o y eEF2 phospho yla ion occu ed wi hou ATP deple- ion o AMP kinase ac i a ion. Fu he mo e, eEF2 phospho yl- a ion was mimicked by p olyl hyd oxylase (PHD) inhibi ion wi h dime hyloxalylglycine o by selec i e PHD2 siRNA silenc- ing bu was independen o hypoxia-inducible ac o ␣ s abili- za ion. Mo eo e , o e exp ession o PHD2 blocked hypoxic accumula ion o phospho yla ed eEF2. The e o e, ou indings sugges ha eEF2 phospho yla ion s a us (and, as a conse- quence, ansla ion a e) is con olled by PHD2 ac i i y. They un a el a no el pa hway o cell adap a ion o hypoxia ha could ha e pa hophysiologic ele ance in issue ischemia and cance . Cell adap a ion o educed oxygen a ailabili y o main ain ATP le els is a majo physiologic challenge because O 2 dep i- a ion, e en ansien , can p oduce i e e sible damage. Cellu- la esponses o hypoxia comp ise ansc ip ional and non- ansc ip ional mechanisms, he na u e o which is as ye only pa ially known (1–3). T ansc ip ional adap a ion o low O 2 ension depends mainly on hypoxia-inducible ansc ip ion ac o s (HIFs), 5 mas e egula o s o a b oad coho o genes whose exp ession dec eases he cellula O 2 demand and inc eases O 2 supply (3). Cellula le els o HIF- ␣ a e in e sely co ela ed wi h he ac i i y o p olyl hyd oxylases (PHDs), O 2 -sensingandFe 2⫹ -dependen enzymes ha hyd oxyla espe- ci ic p oline esidues in HIF- ␣ using O 2 as co-subs a e, hus allowing HIF- ␣ ubiqui ina ion and p o easomal deg ada ion (4, 5). In con as wi h he ela i ely well cha ac e ized PHD-HIF pa hway, he ansc ip ion-independen hypoxic adap i e mechanisms a e poo ly unde s ood. Ne e heless, hey a e essen ial o cell su i al du ing he i s minu es o hypoxia, be o e unc ional exp ession o he O 2 - egula ed genes can ake place (2, 6). Global inhibi ion o p o ein syn hesis (“ ansla- ional a es ”) has been classically ecognized as a undamen al adap a ion o hypoxia because mRNA ansla ion is an ene gy- cos ly p ocess ha consumes up o 70% o he ATP syn hesized by he cells. Hence, any small adjus men o he ansla ion a e can spa e su icien ATP as o immedia ely a end unc ions c i ical o cell su i al (7–10). Se e al kinases egula ed du ing cell ene gy s a a ion, in pa icula mTOR (mammalian a ge o apamycin), he endoplasmic e iculum kinase PERK, and AMP-ac i a ed p o ein kinase (AMPK), ha e been sugges ed o *This wo k was suppo ed by he Spanish Minis y o Science and Heal h, he Ma celino Bo in Founda ion, and he Andalusian Go e nmen . □ S This a icle con ains supplemen al “Me hods,” Figs. 1 and 2, Tables I and II, and addi ional e e ences. 1 Bo h au ho s con ibu ed equally o his wo k. 2 Fellow o he Spanish “Sa a Bo ell” P og am. 3 To whom co espondence may be add essed: Ins . de Biomedicina de Se - illa, Hospi al Uni e si a io Vi gen del Rocío, A . Manuel Siu o s/n, 41013 Se ille, Spain. E-mail: [email p o ec ed]. 4 To whom co espondence may be add essed: Ins . de Biomedicina de Se - illa, Hospi al Uni e si a io Vi gen del Rocío, A . Manuel Siu o s/n, 41013 Se ille, Spain. Tel.: 34-95-592-3049; Fax: 34-95-592-3101; E-mail: apascual- [email p o ec ed]. 5 The abb e ia ions used a e: HIF, hypoxia-inducible ansc ip ion ac o ; PHD, p olyl hyd oxylase; AMPK, AMP-ac i a ed p o ein kinase; eEF2, euka yo ic elonga ion ac o 2; eEF2K, eEF2 kinase; DMOG, dime hyloxalyl- glycine; DIGE, di e ence gel elec opho esis. THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 287, NO. 12, pp. 9651–9658, Ma ch 16, 2012 © 2012 by The Ame ican Socie y o Biochemis y and Molecula Biology, Inc. Published in he U.S.A. MARCH 16, 2012•VOLUME 287•NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 9651 con ibu e o HIF-independen inhibi ion o p o ein ansla- ion unde low O 2 condi ions (7, 9, 11–13). These phospho yl- a ion/dephospho yla ion cascades egula e p o ein syn hesis by p ecisely modula ing he ac i i y o ibosomal ini ia ion and elonga ion ac o s (10, 14). In his ega d, euka yo ic elonga ion ac o 2 (eEF2), a anslocase necessa y o he mo emen o he mRNA along he ibosome, is pa icula ly impo an as, in con- junc ion wi h aminoacyla ion, i uses mos o he ATP equi ed o p o ein syn hesis (15). eEF2 is inhibi ed by phospho yla ion a Th 56 by eEF2 kinase (eEF2K) (16, 17), which in u n is ac i- a ed by AMPK (18). Using a p o eomic app oach designed o de ec ea ly adap - i e changes du ing sho - e m (5–30 min) exposu e o hypoxia, we ha e iden i ied se e al p o eins acu ely egula ed by O 2 a ailabili y. Among hese p o eins, eEF2 was also modula ed by dime hyloxalylglycine (DMOG), an inhibi o o he O 2 -sensing PHDs. He ein, we show ha eEF2 ac i i y is apidly and e e s- ibly egula ed by O 2 ension in a PHD2-dependen manne . This phenomenon is un ela ed o HIF s abiliza ion, and i occu s p io o AMPK ac i a ion. These obse a ions, which un eil unexpec ed ac ions o he O 2 -sensing PHD2, could ha e p o ound implica ions o he pa hophysiology and pha macol- ogy o cell adap a ion o hypoxia. EXPERIMENTAL PROCEDURES Cell Cul u e and Hypoxic T ea men s PC12, Mlp9, HeLa, and HuH7 cells we e cul u ed in DMEM (BioWhi ake ) supplemen ed wi h 10% FBS (In i ogen), 100 uni s/ml penicillin, 100 uni s/ml s ep omycin, and 2 mML-glu- amine(BioWhi ake ).Ca diac en icula myocy eswe ep e- pa ed om 1–3-day-old animals and cul u ed as desc ibed (19). Hepa ocy eswe e isola ed om male Wis a a s bycollagenase (In i ogen) pe usion as desc ibed p e iously (41). U2OS Te -On and U2OS Te -On/pUHD-FLAG-PHD2 cells we e cul- u ed in DMEM supplemen ed wi h 10% e acycline- ee FBS (Clon ech), 100 uni s/ml penicillin, 100 uni s/ml s ep omycin, and5 ␮ g/mlblas icidinS(In i ogen). 200 ␮ g/mlhyg omycinB (Roche Applied Science) was used in U2OS FLAG-PHD2 cul- u es. Cells we e main ained unde a wa e -sa u a ed a mo- sphe e o 5% CO 2 and 95% ai . DMOG (BIOMOL In e na- ional) was used a a inal concen a ion o 1 mM, NaF (Sigma) was used a 25 mM, and 5-aminoimidazole-4-ca boxamide iboside (Sigma) was used a 1 mM. Okadaic acid (Calbiochem) and sanguina ine (Sigma) we e used a he concen a ions speci ied. Hypoxic condi ions (1% O 2 , 94% N 2 , and 5% CO 2 ) we e achie ed in a humidi ied a iable ae obic wo ks a ion (In i o 2 300, Ruskinn Technology L d.). In all expe imen s, cells we e pla ed a 30–50% con luence o p e en he de elopmen o anae obic condi ions a 1% O 2 . Be o e expe imen a ion, media we e p e-equilib a ed o e nigh o he expe imen al oxygen le el. P o ein Analysis Sample P epa a ion, Labeling, and Analysis by Two-dimen- sional Di e ence Gel Elec opho esis (DIGE)—S100 p o ein ex ac s om PC12 cells we e p epa ed and labeled as desc ibed (20). 150 ␮ g o labeled p o ein con aining h ee ex ac s o equal amoun s (no moxic, hypoxic, o DMOG- ea ed and he in e nal s anda d) we e used o wo-dimen- sional DIGE. Elec opho e ic condi ions, quan i a i e analysis o di e ences, and p epa a i e gels we e as desc ibed (20). Fou biological eplica es we e used in all o he expe imen s. eEF2 Pu i ica ion om Ra Li e —Ra li e s we e homoge- nized, and eEF2 was pu i ied as desc ibed (21) using an ÄKTA pu i ie (GE Heal hca e). Wes e n Blo ing, P o ein Syn hesis Measu emen s, and MALDI-MS/MS Analysis—These a e desc ibed unde supple- men al “Me hods.” RNA In e e ence T ans ec ions o siRNAs we e ca ied ou using Lipo- ec amine 2000 (In i ogen) a a inal concen a ion o 20 nM. P e iously alida ed sequences o siRNAs we e used (22, 23). Howe e , he e icacy o he ans ec ion in each expe imen was asce ained by immunoblo ing and/o eal- ime quan i a- i e PCR. Unpublished siRNAs and oligonucleo ide sequences a e p o ided unde supplemen al “Me hods.” ATP Measu emen s ATP le els in cell ex ac s we e de e mined wi h a CLS II ATPbioluminescenceassayki (RocheAppliedScience)usinga GloMax TM 96-mic opla e luminome e equipped wi h an au oinjec ion de ice (P omega). S a is ics Da a a e p esen ed as he mean ⫾S.E. and we e analyzed by one-way analysis o a iance, ollowed by Tukey’s es . p⬍0.05 was conside ed s a is ically signi ican . RESULTS Acu e O 2 -dependen Modula ion o Cell P o eome—We pe - o med p o eomic ( wo-dimensional DIGE) analyses o soluble subcellula ac ions (S100) (20) ob ained om homogena es o PC12 cells exposed o no moxia (21% O 2 ) o hypoxia (1% O 2 ) o 15–30 min. A wo-dimensional map o he S100 ac ion is shown in Fig. 1A, wi h spo s whose ela i e amoun s inc eased ( ed), dec eased (g een), o we e unchanged (yellow) unde hypoxia. Among he 2598 spo s de ec ed, 24 we e iden i ied in all o he expe imen al eplica es and changed hei ela i e amoun unde hypoxia e sus no moxia (⬎1.5- old; p⬍0.05) (supplemen al Table I). In wo-dimensional gels, eEF2 appea ed in a s ip o spo s ha could co espond o addi i e modi ica ions o he p o ein (Fig. 1A,boxed); hence, we ex ended ou analysis o addi ional neighbo ing spo s. These modi ica ions a e p obably due o di e en diph hamide con- en in eEF2. Among he se en spo s s udied (Fig. 1B,uppe panel), i e we e iden i ied as eEF2 by MALDI-MS(/MS) (as e - isks), and all we e ecognized by an an i-eEF2 an ibody (Fig. 1B, middle panel). Acco dingly, eEF2 pu i ied om a li e and sepa a ed by wo-dimensional elec opho esis showed a simila dis ibu ion (Fig. 1B,lowe panel). These esul s e ealed ha eEF2 is modi ied by sho - e m exposu e o hypoxia, hus sug- ges ing ha i could be a media o o he ea ly adap i e mech- anisms igge ed by low O 2 ension. PHD2 Ac i i y and T ansla ional A es unde Hypoxia 9652 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287•NUMBER 12•MARCH 16, 2012 I is known ha eEF2 ac i i y is inhibi ed by phospho yla ion a Th 56 h ough a speci ic eEF2 kinase (16, 17). The e o e, we es ed in se e al cell ypes whe he he changes in eEF2 obse ed in he wo-dimensional DIGE analyses co ela ed wi h modi ica ions in eEF2 Th 56 phospho yla ion s a us using speci ic phospho-Th 56 an ibodies. Hypoxia (1% O 2 ) p oduced a ma ked inc ease in eEF2 phospho yla ion ha was de ec able 15 min a e exposu e o cells o low O 2 ension. The o al amoun o eEF2 was unchanged by exposu e o hypoxia (Fig. 1, C–E, and supplemen al Fig. 1, Aand B). Inhibi o y eEF2 Phospho yla ion by Acu e Hypoxia Occu s be o e ATP Deple ion and AMPK Ac i a ion—As men ioned p e iously, se e al kinases modula ed by ene gy s a a ion a e also egula ed du ing sus ained hypoxic condi ions (7, 9, 12, 13, 18). AMPK, ac i a ed by phospho yla ion a Th 172 by LKB1 (li e kinase B1) when he AMP/ATP a io inc eases (24–27), can ac i a e eEF2K h ough phospho yla ion a Se 398 (18) and hus induce inhibi o y Th 56 phospho yla ion o eEF2 (16, 17). We pos ula ed ha he as O 2 -dependen modula ion o eEF2 epo ed he e should occu p io o and independen ly o AMPK ac i a ion. As expec ed, exposu e o hepa ocy es o 1% O 2 e ealed a clea inc ease in eEF2 Th 56 phospho yla ion de ec able as soon as 5 min a e O 2 dep i a ion, wi hou any inc ease in he le els o ac i e AMPK (phospho yla ed a Th 172 ). In con as , cells ea ed wi h 5-aminoimidazole-4-ca boxamide iboside, an AMPK ac i a o used as a con ol (28), showed s ong ac i- a ion o AMPK (Fig. 2A). Sho - e m exposu e o hypoxia also esul ed in an inc ease in eEF2 phospho-Th 56 and dec ease in AMPK phospho-Th 172 le els in o he cell ypes analyzed (Fig. 2B). In e es ingly, he amoun o ac i e AMPK du ing hypoxic exposu eslas ingup o4h emained e en lowe han ha unde no moxia (Fig. 2, Aand B). In ai ag eemen wi h hese da a, he cellula ATP concen a ion did no change (o e en inc eased) a e 15 o 60 min o hypoxia (1% O 2 ) (Fig. 2C). The ac i i y o eEF2K is inhibi ed by phospho yla ion a Se 366 ,a p ocess egula ed by p90 RSK /p70 S6K kinases (18). Sho - e m hypoxia did no p oduce any changes in he cellula le els o eEF2K phospho-Se 366 (Fig. 1C). Hypoxic ea men did no al e he o al le els o eEF2, eEF2K, o AMPK (Figs. 1Cand 2B). Hypoxia and PHD2 Inhibi ion Inc ease eEF2 Phospho yla ion in a HIF-independen Manne —To es whe he some o he iden i iedp o einsmodi iedbyhypoxiacouldbeunde hecon- ol o he O 2 -sensing PHDs, cells we e exposed o DMOG, a compe i i e inhibi o o 2-oxoglu a a e-dependen dioxyge- nases, including PHDs (4, 29). Con ol (un ea ed) and 15-min DMOG- ea ed cells we e p ocessed by wo-dimensional DIGE, and a ep esen a i e gel o he S100 ac ion is shown in Fig.3A, wi h p o eins whose ela i e amoun s inc eased (g een), dec eased ( ed), o we e unchanged (yellow) a e DMOG ea - men . Only spo 1, iden i ied by mass inge p in ing analysis as eEF2, changed bo h unde hypoxia (Fig. 1A) and a e DMOG ea men (Fig. 3A). Like hypoxia, DMOG ea men also elic- i ed, in se e al cell ypes, a simila phospho-Th 56 eEF2 accu- mula ion al hough wi h a slowe ime cou se (Fig. 3, C–E, and supplemen al Fig. 1B). DMOG- igge ed accumula ion o ei he HIF-1 ␣ o HIF-2 ␣ , used as con ols (Fig. 3C), was also delayed compa ed wi h hypoxia, hus possibly e lec ing DMOG cellula up ake and me aboliza ion o he ac i e N-ox- alylglycine inhibi o . As ans o med cell lines no mally show some esis ance o hypoxic ansla ional a es (30), we es ima ed i he inc ease in phospho yla ion o eEF2 by sho - e m hypoxia o DMOG induced a concomi an dec ease in p o ein ansla ion in p i- ma y cul u ed hepa ocy es (Fig. 3E). As expec ed, hypoxic and DMOG ea men s esul ed in a signi ican dec ease in he p o- FIGURE 1. Acu e inhibi o y eEF2 phospho yla ion (Th 56 ) is induced by hypoxia. A, ep esen a i e wo-dimensional gel. Ex ac s om no moxic (21% O 2 ; g een) o hypoxic (1% O 2 , 15 min; ed) PC12 cells mig a ed in he same gel. Numbe ed spo s co espond o he spo s lis ed in supplemen al Table I. The in e nal s anda d ex ac is no shown in he image. B, same spo s as hose boxed in Alabeled wi h CyDye om a di e en wo-dimensional gel (uppe panel). Fi e spo s (as e isks) we e iden i ied as eEF2 by pep ide mass inge p in ing analysis. Wes e n blo ing o he ma ked spo s was pe o med using an an i-eEF2 an ibody ( ␣ -eEF2;middle panel). eEF2 was pu i ied om a li e , sepa a ed by wo-dimensional elec opho esis, and de ec ed by SYPRO s aining (lowe panel). C, phospho yla ion o eEF2 a Th 56 in HuH7 cells subjec ed o no moxia (N) o hypoxia (H;1%O 2 ) o 15–240 min. The le els o eEF2K, eEF2K Se 366 , ␤ -ac in ( ␤ -ac ), HIF-1 ␣ , and HIF-2 ␣ we e analyzed in he same cell ex ac s. D, quan i ica ion o eEF2 phospho-Th 56 le els in C. .u., ela i e uni s. E, hepa ocy es subjec ed o no moxia o hypoxia (1% O 2 ) o 15 min. Two independen expe imen s a e shown. The o al le els o eEF2 and ␤ -ac in we e analyzed by Wes e n blo ing. PHD2 Ac i i y and T ansla ional A es unde Hypoxia MARCH 16, 2012•VOLUME 287•NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 9653 ein syn hesis a e as de e mined by [ 35 S]Me inco po a ion assay (Fig. 3F). Hypoxia and DMOG apidly inhibi PHDs; he e o e, he in ol emen o hese enzymes in he modula ion o p o ein syn hesis was s udied using cells incuba ed wi h siRNA speci ic o each PHD (22). We obse ed a selec i e inc ease in eEF2 phospho-Th 56 a e PHD2 silencing and no e ec a e inhibi- ion o ei he PHD1 o PHD3 (Fig. 4A). The e ec i eness o he siRNA ea men on PHD mRNAs was con i med by quan i a- i e RT-PCR (supplemen al Table II). PHD2 silencing was also demons a ed by selec i e s abiliza ion o HIF-1 ␣ (Fig. 4A). PHD down- egula ion s udied in se e al cell ypes always co - ela ed wi h hype phospho yla ion o eEF2 a Th 56 (supple- men al Fig. 2A). To u he p o e he di ec in ol emen o PHD2 in he eg- ula ion o eEF2 by acu e hypoxia, we used a cell line (U2OS) in which PHD2 exp ession is condi ionally induced by doxycy- cline. 6 O e exp ession o PHD2 is known o inhibi s abiliza- ion o HIF-1 and HIF-2 ␣ unde hypoxia (23, 31); hence, we 6 N. Masson and P. J. Ra cli e, unpublished da a. FIGURE 2. Inhibi o y eEF2 phospho yla ion unde acu e hypoxia is independen o AMPK ac i i y. A, phospho yla ion o eEF2 (Th 56 ) and AMPK (Th 172 ) analyzed by Wes e n blo ing in hepa ocy es subjec ed o hypoxia (H) o 5–240 min o ea ed wi h he AMPK ac i a o 5-aminoimidazole-4-ca boxamide iboside (AICAR) o 240 min as a con ol. B, phospho yla ion o eEF2 (Th 56 ) and AMPK (Th 172 ) analyzed by Wes e n blo ing in PC12 (le panels) and HuH7 ( igh panels) cells exposed o no moxia o hypoxia o 5 o 15 min. To al AMPK, eEF2K, and ␤ -ac in ( ␤ -ac ) we e used as loading con ols. C, ATP le els in HeLa (le ba s) and HuH7 ( igh ba s) cell lysa es exposed o no moxia o hypoxia o 15 o 60 min. Dashed lines ep esen no moxic le els. *, p⬍0.05 (n⫽5). FIGURE 3. Acu e inhibi o y eEF2 phospho yla ion (Th 56 ) induced by DMOG ea men . A, ep esen a i e wo-dimensional gel showing p o ein ex ac s om con ol ( ed) and DMOG- ea ed (g een) cells. The enci cled spo was di e en ially exp essed a e DMOG ea men and is simila o spo 1 in Fig. 1A. The boxed a ea ep esen s he same a ea o he gel shown in Fig. 1A.B, wo-dimensional gel egion showing spo s modi ied by hypoxia (uppe panel) and DMOG (lowe panel). C, phospho yla ion o eEF2 (Th 56 ) in HuH7 cells subjec ed o no moxia (N) o DMOG (1 mM) o 15–240 min. The le els o eEF2K, eEF2K Se 366 , ␤ -ac in ( ␤ -ac ), HIF-1 ␣ , and HIF-2 ␣ we e analyzed in he same cell ex ac s. D, quan i ica ion o eEF2 phospho-Th 56 le els in C. .u., ela i e uni s. E, hepa ocy es subjec ed o no moxia o DMOG (1 mM) o 15 min. Two independen expe imen s a e shown. The o al le els o eEF2 and ␤ -ac in we e analyzed by Wes e n blo ing. F, p o ein syn hesis ([ 35 S]Me inco po a ion) in ca diomyocy es exposed o no moxia, DMOG, o hypoxia (H) o 4 h. Da a a e ep esen ed ela i e o no moxia. *, p⬍0.05 (n⫽5). PHD2 Ac i i y and T ansla ional A es unde Hypoxia 9654 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287•NUMBER 12•MARCH 16, 2012 es ed whe he PHD2 o e exp ession also p e en s eEF2 phos- pho yla ion by acu e hypoxia. A e 4hindoxycycline, PHD2 exp ession was obse ed in U2OS cells (Fig. 4B) and was co e- la ed wi h a clea dec ease in he amoun o eEF2 phospho yl- a ed a e 30 min o hypoxic ea men (Fig. 4B; see quan i ica- ion in Fig. 4C,p⫽0.024). As a con ol, we es ed whe he HIF-1 ␣ s abiliza ion by hypoxia was also pa ially p e en ed by PHD2 exp ession (Fig. 4B; see quan i ica ion in Fig. 4C,p⫽ 0.044). Because inhibi ion o PHD2 by hypoxia, DMOG, o siRNA leads o HIF- ␣ accumula ion, we in es iga ed whe he HIF migh egula e eEF2 phospho yla ion a Th 56 . Al hough exp ession o ei he HIF-1 ␣ o HIF-2 ␣ unde hypoxia was clea ly abolished using speci ic siRNAs, eEF2 phospho yla ion was unal e ed (Fig. 4D,uppe panels). Hypoxia-dependen eEF2 phospho yla ion was obse ed e en hough HIF-1 ␣ and HIF-2 ␣ we e simul aneously silenced o p e en possible com- pensa o y e ec s o any o he iso o ms (Fig. 4D,lowe panels). The dispensabili y o HIF- ␣ o he hypoxic/DMOG-induced eEF2 phospho yla ion was u he con i med using a gene i- callymodi ied emb yonic s em cell line lacking HIF-1 ␣ (supple- men al Fig. 2B) (32). eEF2 Phospho yla ion by Hypoxia Is Re e sible and S ongly Con olled by Phospha ase Ac i i y—Accumula ion o phos- pho yla ed eEF2 unde hypoxia was apidly (⬍15 min) e e sed upon eoxygena ion, independen ly o he du a ion o he hypoxic ea men (Fig. 5A), hus suppo ing he no ion ha O 2 -dependen eEF2 phospho yla ion is indeed a igh ly egu- la ed p ocess. Fu he mo e, inhibi ion o Se /Th phospha ases wi h NaF p oduced a ma ked inc ease in he amoun o eEF2 phospho-Th 56 , which was no u he modi ied by subsequen exposu e o hypoxia. In con as , phospho yla ion o eEF2K a Se 366 , used as a con ol o phospha ase ac i i y, was unchanged. Hence, eEF2 phospho-Th 56 appea s o be highly suscep ible o he ac ion o phospha ases (Fig. 5B). To iden i y he phospha ase ac ing on eEF2, we used inhibi o s mo e spe- ci ic han NaF. I has been desc ibed ha eEF2 can be dephos- pho yla ed by PP2A and PP2C (33, 34). Hence, we es ed spe- ci ic inhibi o s o each phospha ase and analyzed whe e hey ep oduce he as changes in he le el o phospho yla ion o eEF2 obse ed unde hypoxia. A sho okadaic acid ea men (a PP2A inhibi o ) p oduced a s ong accumula ion o eEF2 (Fig. 5C,le panels), bu sanguina ine (a PP2C inhibi o ) did no a ec he phospho yla ion s a us o eEF2 ( igh panels). These esul s sugges ha PP2A con ols he phospho yla ion s a us o eEF2 unde no moxia. DISCUSSION In his wo k, we ha e desc ibed selec i e pos - ansc ip ional adap i e changes in he p o eome o mammalian cells elici ed by acu e hypoxia. We ocused on eEF2, a anslocase necessa y o p o ein syn hesis, al hough we iden i ied wo o he p o eins (a ginyl aminopep idase and p o ease inhibi o Se pinB6) also al e ed by lowe ing O 2 ension (supplemen al Table I). We ha e shown ha hypoxic inhibi o y eEF2 phospho yla ion is inde- penden o AMPK ac i i y and is mimicked by DMOG ea - FIGURE 4. Hypoxic eEF2 phospho yla ion is egula ed by PHD2 in HIF-independen manne . A, phospho yla ion o eEF2 (Th 56 ) and le els o HIF-1 ␣ examined by Wes e n blo ing 48 h a e siRNA ans ec ion. HeLa cells we e ans ec ed ei he wi h PHD siRNAs o a non-silencing sc ambled siRNA (Con ol). Con ol cells we e independen ly subjec ed o hypoxia (H) o 15 min. In all he expe imen s, sample loading was no malized wi h o al eEF2 and ␤ -ac in ( ␤ -ac ). B, hypoxic eEF2 phospho yla ion a Th 56 is p e en ed by acu e PHD2 o e exp ession. Shown is he phospho yla ion o eEF2 in con ol o PHD2-exp essing U2OS cells (FLAG-PHD2). Cells we e ea ed wi h doxycycline (Dox) o 4 h when indica ed (⫹) and subjec ed o no moxia (N) o hypoxia (1% O 2 ) o 30 min. The le els o HIF-1 ␣ and FLAG-PHD2 we e analyzed in he same cell ex ac s. The o al le els o eEF2 and ␤ -ac in we e also analyzed by Wes e n blo ing. C, quan i ica ion o he - old change p oduced in he le els o eEF2 phospho-Th 56 (black ba s) and HIF-1 ␣ (g ay ba s) by 30 min o hypoxia in he absence (⫺)o p esence (⫹) o doxycycline in FLAG-PHD2 cell cul u es. *, p⬍0.05 (n⫽3). D, hypoxic eEF2 phospho yla ion a Th 56 is independen o HIF- ␣ ac i i y. HeLa cells we e ans ec ed wi h a sc ambled siRNA (Con ol), HIF-1 ␣ (siHIF-1 ␣ ) o HIF-2 ␣ (siHIF-2 ␣ ) siRNA (uppe panels), o bo h (lowe panels) o 24 h and exposed o hypoxia o 0, 15, o 240 min. The le els o eEF2 phospho-Th 56 , eEF2, HIF-1 ␣ , HIF-2 ␣ , and ␤ -ac in we e analyzed by Wes e n blo ing. PHD2 Ac i i y and T ansla ional A es unde Hypoxia MARCH 16, 2012•VOLUME 287•NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 9655 men , hus sugges ing ha eEF2 ac i i y and p o ein syn hesis a e modula ed by PHDs. T ansla ional a es is a classically s udied cellula esponse o anoxia o long- e m hypoxia ( o a e iew, see Re . 14). I has also been epo ed ha a sho (2 h) exposu e o mode a e hypoxia (0.2–2% O 2 ) can lead o a modes bu ep oducible dec ease in ansla ion e iciency (35), an e ec ha is be e obse ed wi h longe (⬎24 h) ea men s (12, 13, 36, 37). We ha e shown he e ha acu e (5–15 min) hypoxia (1% O 2 )o DMOG ea men leads o changes in he eEF2 mig a ion pa - e n as a consequence o inc eased p o ein phospho yla ion a Th 56 . I is well es ablished ha phospho yla ion o eEF2 a Th 56 dec eases he a e o p o ein elonga ion (16, 17, 38), a mechanism ha can con ibu e o mRNA ansla ion inhibi ion by hypoxia and DMOG obse ed in p ima y cul u ed ca - diomyocy es (Fig. 3F). Simila esul s ha e been epo ed using umo /immo alized cell lines (12, 13), bu longe pe iods, mo e se e e hypoxia, o e en concomi an se um wi hd awal was equi ed o obse e a dec ease in p o ein syn hesis in hese cellula models ha a e no mally esis an o mRNA ansla- ion inhibi ion by hypoxia (30). The ac i a ion o he AMPK-eEF2K-eEF2 pa hway o a es p o ein ansla ion unde long- e m hypoxia o ischemia has been elegan ly demons a ed in p e ious epo s (12, 13). How- e e , he inhibi ion o eEF2 ac i i y obse ed unde acu e hypoxia desc ibed he e is independen o his pa hway. Indeed, we ha e shown ha sho - e m hypoxia e en dec eases AMPK ac i i y in all cells es ed (Fig. 2, Aand B). In ag eemen wi h hese da a, cellula ATP le els we e p ese ed o inc eased uponacu eexposu e ohypoxia(Fig.2C)(13,39).Theexis ence o complemen a y PHD2- and AMPK-dependen eEF2 modu- la o pa hways se in mo ion depending on O 2 a ailabili y is ad an ageous because i o e s a b oade adap i e epe oi e o ensu e ATP homeos asis. As DMOG inhibi s a b oad amily (⬎50 membe s) o 2-oxo- glu a a e-dependen dioxygenases (4, 29), we pe o med siRNA analyses o iden i y he PHDs in ol ed in eEF2 egula ion. We ound ha selec i e inhibi ion o PHD2 in se e al cell lines leads o sus ained eEF2 phospho yla ion a Th 56 , hus con- i ming ha p o einsyn hesis is modula ed by he ac i i y o he O 2 senso PHD2 (Fig. 4Aand supplemen al Fig. 2A). To achie e a s ong silencing o PHD2, cells we e ea ed wi h siRNAs o 72 h, inducing se e al HIF- egula ed pa hways ha could, indi- ec ly, lead o phospho yla ion o eEF2. Howe e , o e exp es- sion o PHD2 o 4 h p oduced a s ong inhibi ion o he hypox- ia-dependen phospho yla ion o eEF2 (Fig. 4, Band C), an e ec ha has been also shown o hypoxia-induced HIF s abi- liza ion (23, 31). Al oge he , hese da a sugges he con ol o PHD2 o e he elonga ion phase o p o ein ansla ion. PHD2- dependen egula ion o ansla ion is independen o HIF- ␣ ac i i y and akes place a ew minu es a e PHD inhibi ion (Fig. 4Cand supplemen al Fig. 2B). Sequence analyses e ealed he p esence in eEF2 o h ee mo i s consis ing o a p oline p e- ceded by wo neighbo ing leucines (LXLXP), esembling he hyd oxyla ion consensus mo i s (LXXLAP) exis ing in he O 2 -dependen deg ada ion domain o HIF- ␣ (4, 5). Howe e , addi ional wo k will be equi ed o de e mine whe he PHD2 egula es eEF2 by di ec hyd oxyla ion o hese esidues o , al e na i ely,modula es he ac i i y o o he p o eins, including eEF2K o any subuni o he cellula phospha ases. O 2 in lux o issues can be a luc ua ing condi ion (40); hus, he mechanisms egula ing hypoxic esponses mus be e e s- ible and easily adjus able o oxygen a ailabili y. In e es ingly, phospho yla ion o eEF2 is igh ly egula ed by he ac i i y o speci ic kinases and phospha ases (33, 34, 38). We ha e shown he e ha phospho yla ion o eEF2 a Th 56 by hypoxia is quickly e e sed upon e u ning o no moxia (Fig. 5A). A sho exposu e (10 min) o cells o a Se /Th phospha ase inhibi o leads o ma ked accumula ion o phospho yla ed eEF2, indica - ing a high basal ac i i y o phospha ases ac ing on eEF2 (Fig. 5B). Ou pha macological s udies e ealed ha acu e inhibi ion o PP2A ac i i y leads o a simila accumula ion o phospho yl- FIGURE 5. Phospho yla ion o eEF2 a Th 56 is dynamic. A, le els o eEF2 phospho yla ion (Th 56 ) in HuH7 cells subjec ed o hypoxia (H) o 15 (uppe panels) o 60 (lowe panels) min and eco e y a e e u ning o no moxia (N) o 15–240 min (Reox). B, phospho yla ion o eEF2 (Th 56 ) and eEF2K (Se 366 ) in HeLa cells ea ed wi h he Th /Se phospha ase inhibi o NaF (25 mM) o 10 min. Cells we e also exposed o no moxia o hypoxia o 15 min. The o al le els o eEF2K, eEF2, and ␤ -ac in ( ␤ -ac ) we e also analyzed by Wes e n blo ing. C, phospho yla ion o eEF2 a Th 56 is enhanced by a sho okadaic acid ea men (OA; 250 nM;le panels) and is no a ec ed by ea men wi h sanguina ine (SA; 0.5–2.5 ␮ M; igh panels). NaF (25 mM) ea men was pe o med as a con ol. PHD2 Ac i i y and T ansla ional A es unde Hypoxia 9656 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287•NUMBER 12•MARCH 16, 2012 a ed eEF2 (Fig. 5C), sugges ing a ole o his p o ein in he eg- ula ion o p o ein syn hesis by acu e hypoxia. The igh egula- ion and as e e sibili y o he eEF2 phospho yla ion/ dephospho yla ion s a us make his p o ein an ideal candida e o con ol he p o ein syn hesis a e in esponse o a iable condi ions o issue O 2 a ailabili y. In conclusion, we p opose a no el mechanism ha helps explain acu e ansla ional a es unde hypoxia, which is AMPK- and HIF-independen and elies on PHD2-media ed egula ion o eEF2 ac i i y (see scheme in Fig. 6). Howe e , a comple e unde s anding o he mechanism whe eby PHD2 eg- ula es eEF2 phospho yla ion mus awai u he in es iga ion. Besides i s ob ious in e es o he biology o hypoxia and hypoxia/ epe usion, he PHD2-eEF2 pa hway could p o ide new pe spec i es ele an o he pha macology o issue ische- mia and cance . Acknowledgmen s—We hank Pe e Ra cli e and No ma Masson o p o iding unpublished U2OS cell lines. We hank D s. M. L. Ma - ínez-Chan a and M. Va ela (CIC bioGUNE) o kindly p o iding he hepa ocy e p ima y cul u es. REFERENCES 1. Kaelin, W. G., J ., and Ra cli e, P. J. (2008) Oxygen sensing by me azoans: he cen al ole o he HIF hyd oxylase pa hway. Mol. Cell 30, 393–402 2. Lopez-Ba neo, J., Pa dal, R., and O ega-Sáenz, P. (2001) Cellula mecha- nism o oxygen sensing. Annu. Re . Physiol. 63, 259–287 3. Semenza, G. L. (2009) Regula ion o oxygen homeos asis by hypoxia-in- ducible ac o 1. Physiology 24, 97–106 4. Eps ein, A. C., Gleadle, J. M., McNeill, L. A., Hewi son, K. S., O’Rou ke, J., Mole, D. R., Mukhe ji, M., Me zen, E., Wilson, M. I., Dhanda, A., Tian, Y. M., Masson, N., Hamil on, D. L., Jaakkola, P., Ba s ead, R., Hodgkin, J., Maxwell, P. H., Pugh, C. W., Scho ield, C. J., and Ra cli e, P. J. (2001) C. elegans EGL-9 and mammalian homologs de ine a amily o dioxyge- nases ha egula e HIF by p olyl hyd oxyla ion. Cell 107, 43–54 5. Masson, N., Willam, C., Maxwell, P. H., Pugh, C. W., and Ra cli e, P. J. (2001) Independen unc ion o wo des uc ion domains in hypoxia-in- ducible ac o ␣ chains ac i a ed by p olyl hyd oxyla ion. EMBO J. 20, 5197–5206 6. Wei , E. K., López-Ba neo, J., Buckle , K. J., and A che , S. L. (2005) Acu e oxygen-sensing mechanisms. N. Engl. J. Med. 353, 2042–2055 7. A sham, A. M., Howell, J. J., and Simon, M. C. (2003) A no el hypoxia- inducible ac o -independen hypoxic esponse egula ing mammalian a ge o apamycin and i s a ge s. J. Biol. Chem. 278, 29655–29660 8. Hochachka, P. W., Buck, L. T., Doll, C. J., and Land, S. C. (1996) Uni ying heo y o hypoxia ole ance: molecula /me abolic de ense and escue mechanisms o su i ing oxygen lack. P oc. Na l. Acad. Sci. U.S.A. 93, 9493–9498 9. Koumenis, C., Naczki, C., Ko i zinsky, M., Ras ani, S., Diehl, A., Sonen- be g, N., Ko omilas, A., and Wou e s, B. G. (2002) Regula ion o p o ein syn hesis by hypoxia ia ac i a ion o he endoplasmic e iculum kinase PERK and phospho yla ion o he ansla ion ini ia ion ac o eIF2 ␣ .Mol. Cell. Biol. 22, 7405–7416 10. Wou e s, B. G., an den Beucken, T., Magagnin, M. G., Ko i zinsky, M., Fels, D., and Koumenis, C. (2005) Con ol o he hypoxic esponse h ough egula ion o mRNA ansla ion. Semin. Cell De . Biol. 16, 487–501 11. B owne, G. J., and P oud, C. G. (2004) A no el mTOR- egula ed phospho- yla ion si e in elonga ion ac o 2 kinase modula es he ac i i y o he kinase and i s binding o calmodulin. Mol. Cell. Biol. 24, 2986–2997 12. Connolly, E., B auns ein, S., Fo men i, S., and Schneide , R. J. (2006) Hy- poxia inhibi s p o ein syn hesis h ough a 4E-BP1 and elonga ion ac o 2 kinase pa hway con olled by mTOR and uncoupled in b eas cance cells. Mol. Cell. Biol. 26, 3955–3965 13. Liu, L., Cash, T. P., Jones, R. G., Kei h, B., Thompson, C. B., and Simon, M. C. (2006) Hypoxia-induced ene gy s ess egula es mRNA ansla ion and cell g ow h. Mol. Cell 21, 521–531 14. Ko i zinsky, M., and Wou e s, B. G. (2007) Hypoxia and egula ion o messenge RNA ansla ion. Me hods Enzymol. 435, 247–273 15. P oud, C. G. (2007) Signaling o ansla ion: how signal ansduc ion pa h- ways con ol he p o ein syn he ic machine y. Biochem. J. 403, 217–234 16. Nai n, A. C., and Pal ey, H. C. (1987) Iden i ica ion o he majo M 100,000 subs a e o calmodulin-dependen p o ein kinase III in mam- malian cells as elonga ion ac o 2. J. Biol. Chem. 262, 17299–17303 17. Ryazano , A. G., Shes ako a, E. A., and Na apo , P. G. (1988) Phospho yl- a ion o elonga ion ac o 2 by EF2 kinase a ec s a e o ansla ion. Na u e 334, 170–173 18. B owne, G. J., Finn, S. G., and P oud, C. G. (2004) S imula ion o he AMP-ac i a ed p o ein kinase leads o ac i a ion o euka yo ic elonga ion ac o 2 kinase and o i s phospho yla ion a a no el si e, se ine 398. J. Biol. Chem. 279, 12220–12231 19. Akao, M., Ohle , A., O’Rou ke, B., and Ma bán, E. (2001) Mi ochond ial ATP-sensi i e po assium channels inhibi apop osis induced by oxida i e FIGURE 6. Schema ic ep esen a ion o eEF2 cycle in ibosome (blue). Unde no moxia (le ), eEF2K/p o ein phospha ase (PP) balance is displaced o he unphospho yla ed o m o eEF2 (g een sphe es) in a PHD2-dependen p ocess. Acu e hypoxia (cen e ) inhibi s PHD2 ac i i y and leads o inc eased eEF2 phospho yla ion ( ed sphe es), which slows down he eEF2 cycle. Unde sus ained hypoxia o anoxic/ischemic condi ions ( igh ), AMPK is ac i a ed, which leads o s imula o y eEF2K phospho yla ion (Se 399 ), eEF2 phospho yla ion, and s ong inhibi ion o he p o ein syn hesis a e. PHD2 Ac i i y and T ansla ional A es unde Hypoxia MARCH 16, 2012•VOLUME 287•NUMBER 12 JOURNAL OF BIOLOGICAL CHEMISTRY 9657 s ess in ca diac cells. Ci c. Res. 88, 1267–1275 20. Pascual, A., Rome o-Ruiz, A., and Lopez-Ba neo, J. (2009) Di e en ial p o eomic analysis o ad enal gland du ing pos na al de elopmen . P o- eomics 9, 2946–2954 21. Ho man, S., B owne, G., K ause, U., Pa el, J., Ve ommen, D., Be and, L., La oinne, A., Hue, L., P oud, C., and Ride , M. (2002) Ac i a ion o AMP- ac i a ed p o ein kinase leads o he phospho yla ion o elonga ion ac o 2 and an inhibi ion o p o ein syn hesis. Cu . Biol. 12, 1419–1423 22. Be a, E., Beniz i, E., Ginou ès, A., Volma , V., Roux, D., and Pouysségu , J. (2003) HIF p olyl hyd oxylase 2 is he key oxygen senso se ing low s eady-s a e le els o HIF-1 ␣ in no moxia. EMBO J. 22, 4082–4090 23. Ginou ès, A., Ilc, K., Macías, N., Pouysségu , J., and Be a, E. (2008) PHD o e ac i a ion du ing ch onic hypoxia “desensi izes” HIF- ␣ and p o ec s cells om nec osis. P oc. Na l. Acad. Sci. U.S.A. 105, 4745–4750 24. Ha die, D. G., Sco , J. W., Pan, D. A., and Hudson, E. R. (2003) Manage- men o cellula ene gy by he AMP-ac i a ed p o ein kinase sys em. FEBS Le . 546, 113–120 25. Hawley, S. A., Da ison, M., Woods, A., Da ies, S. P., Be i, R. K., Ca ling, D., and Ha die, D. G. (1996) Cha ac e iza ion o he AMP-ac i a ed p o ein kinase kinase om a li e and iden i ica ion o h eonine 172 as he majo si e a which i phospho yla es AMP-ac i a ed p o ein kinase. J. Biol. Chem. 271, 27879–27887 26. Lizcano, J. M., Gö ansson, O., To h, R., Deak, M., Mo ice, N. A., Boudeau, J., Hawley, S. A., Udd, L., Mäkelä, T. P., Ha die, D. G., and Alessi, D. R. (2004) LKB1 is a mas e kinase ha ac i a es 13 kinases o he AMPK sub amily, including MARK/PAR-1. EMBO J. 23, 833–843 27. Shaw, R. J., Kosma ka, M., Ba deesy, N., Hu ley, R. L., Wi e s, L. A., DePinho, R. A., and Can ley, L. C. (2004) The umo supp esso LKB1 kinase di ec ly ac i a es AMP-ac i a ed kinase and egula es apop osis in esponse o ene gy s ess. P oc. Na l. Acad. Sci. U.S.A. 101, 3329–3335 28. Co on, J. M., Gillespie, J. G., Hawley, S. A., and Ha die, D. G. (1995) 5-Aminoimidazole-4-ca boxamide ibonucleoside. A speci ic me hod o ac i a ing AMP-ac i a ed p o ein kinase in in ac cells? Eu . J. Biochem. 229, 558–565 29. Jaakkola, P., Mole, D. R., Tian, Y. M., Wilson, M. I., Gielbe , J., Gaskell, S. J., on K iegsheim, A., Hebes ei , H. F., Mukhe ji, M., Scho ield, C. J., Maxwell, P. H., Pugh, C. W., and Ra cli e, P. J. (2001) Ta ge ing o HIF- ␣ o he on Hippel-Lindau ubiqui yla ion complex by O 2 - egula ed p olyl hyd oxyla ion. Science 292, 468–472 30. Höckel, M., and Vaupel, P. (2001) Tumo hypoxia: de ini ions and cu en clinical, biologic, and molecula aspec s. J. Na l. Cance Ins . 93, 266–276 31. Takeda, K., and Fong, G. H. (2007) P olyl hyd oxylase domain 2 p o ein supp esses hypoxia-induced endo helial cell p oli e a ion. Hype ension 49, 178–184 32. Ca melie , P., Do , Y., He be , J. M., Fukumu a, D., B usselmans, K., Dewe chin, M., Neeman, M., Bono, F., Ab amo i ch, R., Maxwell, P., Koch, C. J., Ra cli e, P., Moons, L., Jain, R. K., Collen, D., and Keshe , E. (1998) Role o HIF-1 ␣ in hypoxia-media ed apop osis, cell p oli e a ion, and umo angiogenesis. Na u e 394, 485–490 33. Redpa h, N. T., and P oud, C. G. (1990) Ac i i y o p o ein phospha ases agains ini ia ion ac o 2 and elonga ion ac o 2. Biochem. J. 272, 175–180 34. Redpa h, N. T., and P oud, C. G. (1991) Di e ing e ec s o he p o ein phospha ase inhibi o s okadaic acid and mic ocys in on ansla ion in e iculocy e lysa es. Biochim. Biophys. Ac a 1093, 36–41 35. Ko i zinsky, M., Rouschop, K. M., an den Beucken, T., Magagnin, M. G., Sa elkouls, K., Lambin, P., and Wou e s, B. G. (2007) Phospho yla ion o eIF2 ␣ is equi ed o mRNA ansla ion inhibi ion and su i al du ing mode a e hypoxia. Radio he . Oncol. 83, 353–361 36. Lang, K. J., Kappel, A., and Goodall, G. J. (2002) Hypoxia-inducible ac o 1 ␣ mRNA con ains an in e nal ibosome en y si e ha allows e icien ansla ion du ing no moxia and hypoxia. Mol. Biol. Cell 13, 1792–1801 37. Thomas, J. D., and Johannes, G. J. (2007) Iden i ica ion o mRNAs ha con inue o associa e wi h polysomes du ing hypoxia. RNA 13, 1116–1131 38. Redpa h, N. T., P ice, N. T., Se e ino , K. V., and P oud, C. G. (1993) Regula ion o elonga ion ac o 2 by mul isi e phospho yla ion. Eu . J. Biochem. 213, 689–699 39. Le eb e, V. H., Van S eenb ugge, M., Becke s, V., Robe oid, M., and Buc-Calde on, P. (1993) Adenine nucleo ides and inhibi ion o p o ein syn hesis in isola ed hepa ocy es incuba ed unde di e en pO 2 le els. A ch. Biochem. Biophys. 304, 322–331 40. B is ow, R. G., and Hill, R. P. (2008) Hypoxia and me abolism. Hypoxia, DNA epai , and gene ic ins abili y. Na . Re . Cance 8, 180–192 41. A ila, M. A., Ca e e o, M. V., Rod iguez, E. N., and Ma o, J. M. (1998) Regula ion by hypoxia o me hionine adenosyl ans e ase ac i i y and gene exp ession in a hepa ocy es. Gas oen e ology 114, 364–371 PHD2 Ac i i y and T ansla ional A es unde Hypoxia 9658 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 287•NUMBER 12•MARCH 16, 2012