scieee Open visual document viewer

Thioredoxin-linked processes in cyanobacteria are as numerous as in chloroplasts, but targets are different

Florencio Bellido, Francisco Javier; Lindahl, Anna Marika

Abstract

Light-dependent regulation of a growing number of chloroplast enzymatic activities has been found to occur through the reversible reduction of intra- or intermolecular disulphides by thioredoxins. In cyanobacteria, despite their similarity to chloroplasts, no proteins have hitherto been shown to interact with thioredoxins, and the role of the cyanobacterial ferredoxin/thioredoxin system has remained obscure. By using an immobilized cysteine 35-to-serine site-directed mutant of the Synechocystis sp. PCC 6803 thioredoxin TrxA as bait, we screened the Synechocystis cytosolic and peripheral membrane protein complements for proteins interacting with TrxA. The covalent bond between the isolated target proteins and mutated TrxA was confirmed by nonreducing/reducing two-dimensional SDS/PAGE. Thus, we have identified 18 cytosolic proteins and 8 membrane-associated proteins as candidate thioredoxin substrates. Twenty of these proteins have not previously been associated with thioredoxin-mediated regulation. Phosphoglucomutase, one of the previously uncharacterized thioredoxin-linked enzymes, has not earlier been considered a target for metabolic control through disulphide reduction. In this article, we show that phosphoglucomutase is inhibited under oxidizing conditions and activated by DTT and reduced wild-type TrxA in vitro. The results imply that thioredoxin-mediated redox regulation is as extensive in cyanobacteria as in chloroplasts but that the subjects of regulation are largely different.

Full text

Thio edoxin-linked p ocesses in cyanobac e ia a e as nume ous as in chlo oplas s, bu a ge s a e di e en Ma ika Lindahl* and F ancisco J. Flo encio Ins i u o de Bioquı´mica Vege al y Fo osı´n esis, Consejo Supe io de In es igaciones Cien ı´ icas, Uni e sidad de Se illa, Cen o de In es igaciones Cien ı´ icas Isla de la Ca uja, A enida Ame ico Vespucio s兾n, E-410 92 Se illa, Spain Edi ed by Bob B. Buchanan, Uni e si y o Cali o nia, Be keley, CA, and app o ed Oc obe 10, 2003 ( ecei ed o e iew July 15, 2003) Ligh -dependen egula ion o a g owing numbe o chlo oplas enzyma ic ac i i ies has been ound o occu h ough he e e sible educ ion o in a- o in e molecula disulphides by hio edoxins. In cyanobac e ia, despi e hei simila i y o chlo oplas s, no p o eins ha e hi he o been shown o in e ac wi h hio edoxins, and he ole o he cyanobac e ial e edoxin兾 hio edoxin sys em has e- mained obscu e. By using an immobilized cys eine 35- o-se ine si e-di ec ed mu an o he Synechocys is sp. PCC 6803 hio edoxin T xA as bai , we sc eened he Synechocys is cy osolic and pe iph- e al memb ane p o ein complemen s o p o eins in e ac ing wi h T xA. The co alen bond be ween he isola ed a ge p o eins and mu a ed T xA was con i med by non educing兾 educing wo- dimensional SDS兾PAGE. Thus, we ha e iden i ied 18 cy osolic p o eins and 8 memb ane-associa ed p o eins as candida e hio e- doxin subs a es. Twen y o hese p o eins ha e no p e iously been associa ed wi h hio edoxin-media ed egula ion. Phospho- glucomu ase, one o he p e iously uncha ac e ized hio edoxin- linked enzymes, has no ea lie been conside ed a a ge o me abolic con ol h ough disulphide educ ion. In his a icle, we show ha phosphoglucomu ase is inhibi ed unde oxidizing con- di ions and ac i a ed by DTT and educed wild- ype T xA in i o. The esul s imply ha hio edoxin-media ed edox egula ion is as ex ensi e in cyanobac e ia as in chlo oplas s bu ha he subjec s o egula ion a e la gely di e en . Di hiol兾disulphide exchange ca alyzed by hio edoxin (T x) o ms he molecula basis o ligh -dependen egula ion o many enzymes in he chlo oplas s o highe plan s and algae (1–3). Fe edoxin ecei es educing equi alen s om he pho- osyn he ic elec on anspo in he ligh and educes T x mand by means o e edoxin-T x educ ase. The T xs, in u n, con e disulphides o di hiols in hei espec i e a ge enzymes, he eby modula ing hei ac i i ies. The ea lies disco e ed a ge s o T x-media ed egula ion belong o he Cal in cycle o CO 2 assimila ion (1). Since he ini ial disco e ies, se e al mo e chlo oplas enzymes ha e been ecognized as subs a es o T x (3). A b eak h ough in he in es iga ion o chlo oplas edox egula ion came wi h he in oduc ion o a new me hod o isola e T x a ge p o eins (4–6). The me hod in ol es mu a ion o he bu ied edox-ac i e cys eine o T x, which a o s he o ma ion o s able mixed disulphides wi h i s a ge p o eins. This expe - imen al app oach con i med he in e ac ion be ween T x and i s known a ge s and mo e han doubled he numbe o po en ial T x- egula ed p o eins (6). Cyanobac e ia a e oxygenic pho osyn he ic p oka yo es ha p obably sha e a common ances o wi h he chlo oplas . The comple e sequence o he cyanobac e ium Synechocys is sp. PCC 6803 genome (Cyanobase, www.kazusa.o .jp兾cyano兾cyano. h ml) e eals ha his o ganism, he ea e e e ed o as Syn- echocys is, con ains e edoxin-T x educ ase and a leas ou di e en T xs. Ne e heless, a emp s o demons a e ligh - dependen edox egula ion o h ee cyanobac e ial enzymes o he Cal in cycle (phospho ibulokinase, uc ose-1,6-bisphos- pha ase, and glyce aldehyde-3-phospha e dehyd ogenase) we e unsuccess ul (7), al hough pu i ied phospho ibulokinase om Synechococcus sp. PCC 7942 could be ac i a ed by DTT in i o (8). The s a k con as be ween he g owing numbe o chlo o- plas T x a ge s and he lack o da a indica ing T x unc ion in cyanobac e ia migh lead o he conclusion ha ligh -induced T x-media ed egula ion e ol ed wi h he need o coo dina e chlo oplas ic and ex achlo oplas ic me abolism in pho osyn- he ic euka yo es. Su p isingly, Synechocys is T xA was ound essen ial o su i al unde pho oau o ophic as well as he e o- ophic g ow h condi ions (9). Among he di e se Synechocys is T xs, T xA is he one ha mos esembles he chlo oplas m- ype T x. Aiming a cla i ying he oles o T xA in Synechocys is, we ha e sc eened he Synechocys is cy osolic and pe iphe al memb ane p o ein complemen s o p o eins in e ac ing wi h T xA. To his end, we used a s a egy simila o ha desc ibed in e s. 5 and 6. Thus, we ha e iden i ied 18 cy osolic p o eins and 8 memb ane- associa ed p o eins as candida e subs a es o T xA. One o hese a ge s, phosphoglucomu ase (PGM), which ep esen s a me abolic b anch poin be ween s o age and u iliza ion o ca bohyd a es, was shown o be ac i a ed by T xA in i o. Ma e ials and Me hods Ma e ials. DEAE Sephacel ma ix, His-Bind esin, and Supe dex 75 we e pu chased om Sigma, No agen, and Ame sham Pha - macia Biosciences, espec i ely. DTT, T i on X-100, and NAD ⫹ we e om Sigma. O he chemicals we e o he highes g ade comme cially a ailable. Mu agenesis, Exp ession, and Pu i ica ion o T xA. The xA gene (ORF sl 0623, Cyanobase) was mu a ed, exchanging cys eine 35 o a se ine. The mu a ed T xA (T xA35) was exp essed wi h a C- e minal His- ag by using he pET22b ec o in Esche ichia coli BL21 DE3 and pu i ied o homogenei y by anion-exchange ch oma og aphy on DEAE Sephacel ma ix, Ni-a ini y ch o- ma og aphy wi h he His-Bind esin, and gel il a ion on Su- pe dex 75. The gel- il a ion s ep was p eceded by a 1-h incu- ba ion wi h 20 mM DTT on ice. Pu e T xA35 was elu ed in 25 mM T is䡠HCl (pH 7.9) and 150 mM KCl. P epa a ion o Cy osolic and Pe iphe al Memb ane P o ein F ac ions. Synechocys is cul u es we e g own pho oau o ophically a 30°C in BG11 medium (10) including 1 g兾l NaHCO 3 and bubbled wi h 1% ( ol兾 ol) CO 2 in ai unde con inuous illumina ion wi h whi e ligh a an in ensi y o 50 ␮ mol pho ons䡠m ⫺2 䡠s ⫺1 . Cells we e ha es ed and b oken wi h glass beads as desc ibed in e . 11 wi h bu e A (25 mM Hepes䡠NaOH, pH 7.0兾15 mM CaCl 2 兾5mM MgCl 2 兾15% ( ol/ ol) glyce ol兾1 mM PMSF). A e emo al o unb oken cells by cen i uga ion a 2,300 ⫻g o 5 min, o al memb anes we e pelle ed by cen i uga ion a 16,000 ⫻g o 20 min, yielding a supe na an con aining cy osolic p o eins. Mem- This pape was submi ed di ec ly (T ack II) o he PNAS o ice. Abb e ia ions: T x, hio edoxin; GS, glu amine syn he ase; GOGAT, glu ama e syn hase; Rubisco, ibulose 1,5-bisphospha e ca boxylase兾oxygenase; RbcL, la ge subuni o Rubisco; PGM, phosphoglucomu ase; P x, pe oxi edoxin. *To whom co espondence should be add essed. E-mail: [email p o ec ed]. © 2003 by The Na ional Academy o Sciences o he USA www.pnas.o g兾cgi兾doi兾10.1073兾pnas.2534397100 PNAS 兩 Decembe 23, 2003 兩 ol. 100 兩 no. 26 兩 16107–16112 PLANT BIOLOGY b anes we e washed h ee imes by esuspension in bu e A and cen i uga ion. Pe iphe al memb ane p o eins we e ex ac ed by 15-min incuba ion on ice wi h bu e A supplemen ed wi h 1 M NaCl and 0.02% ( ol兾 ol) T i on X-100 ollowed by cen i uga- ion a 16,000 ⫻g o 30 min. The supe na an was concen a ed, and NaCl and T i on X-100 we e elimina ed by cen i uga ion il a ion in Ul a ee-MC (Millipo e), 3-kDa cu -o , and suc- cessi e eplacemen wi h bu e A. P o ein concen a ions we e measu ed as desc ibed (12). Isola ion o T xA Ta ge P o eins. We added 1.2 mg o pu e T xA35 o 150 ␮ l o His-Bind esin in a 1.5-ml mic ocen i uge ube and incuba ed i a 4°C unde gen le agi a ion o 1 h. The esin was washed wice wi h 1 ml o bu e A a e emo al o he supe na an . Then 2.8 mg o cy osolic p o ein o 1.8 mg o pe iphe al memb ane p o ein was added ollowed by a 16-h incuba ion a 4°C unde gen le agi a ion. Unbound p o eins we e emo ed, and he esin was washed ou imes in 1 ml o bu e B (20 mM T is䡠HCl, pH 7.9兾0.5 M NaCl) wi h he addi ion o 60 mM imidazole. T xA35– a ge p o ein complexes we e eleased by addi ion o 120 ␮ l o bu e B including 1 M imidazole, and he elua es we e eco e ed by cen i uga ion o 10 sec a 2,000 ⫻g. Non educing兾Reducing Gel Elec opho esis. Elua es we e elec o- pho esed unde non educing condi ions in he i s dimension, main aining he T xA35– a ge p o ein complexes in ac , and educing condi ions in he second dimension, eleasing T xA35 om i s a ge s. Se en y mic oli e s o elua e was mixed wi h 15 ␮ l o See Blue p es ained p o ein s anda d (In i ogen) and 30 ␮ l o solubilizing bu e (0.25 M T is䡠HCl, pH 6.8兾12.5% ( ol/ ol) glyce ol兾10% (w / ol) SDS). A e sepa a ion by SDS兾 PAGE on a 16 ⫻18 cm, 1.5-mm hick gel, he lane was excised and incuba ed o 1ha 25°C in 2 ml o solubilizing bu e dilu ed 2.5 imes wi h he addi ion o DTT o 100 mM inal concen a- ion. The lane was ixed on op o an ac ylamide gel o iden ical size and, a e sepa a ion a 40 mA, he gel was s ained wi h Coomassie b illian blue R-250. Ac ylamide gels o 12% and 10% (w 兾 ol) we e used o cy osolic and pe iphe al memb ane p o eins, espec i ely. Appa en molecula masses we e de e - mined by using one-dimensional gels, educing condi ions, and nons ained p o ein ma ke s (Bio-Rad) ollowed by Coomassie s aining o maximal accu acy. P o ein Iden i ica ion. Coomassie-s ained p o eins we e excised and diges ed wi h ypsin. Pep ides we e subjec ed o ma ix- assis ed lase deso p ion ioniza ion– ime-o - ligh (MALDI- TOF) analysis pe o med on an Ul a lex appa a us (B uke , Bille ica, MA). P o eins we e iden i ied by using he MASCOT pep ide mass inge p in sea ch engine (Ma ix Science, London). Wes e n Blo . Fo glu amine syn he ase (GS), he an ibody (13) was used a 1:5,000 dilu ion. Fo NADH-glu ama e syn hase (GOGAT), an an ibody aised agains he la ge subuni , Gl B (F. Na a o and F.J.F., unpublished da a), was used a 1:2,000 dilu ion. Signals we e isualized wi h enhanced chemilumines- cence Wes e n blo ing de ec ion eagen s om Ame sham Pha macia Biosciences. Measu emen o PGM Ac i i y. The ac i i y was measu ed by coupling o educ ion o NAD ⫹ ca alyzed by glucose-6- phospha e dehyd ogenase (G6PD) om Leuconos oc mesen- e oides (Sigma). G6PD om L. mesen e oides is de oid o cys eines, and hiol eagen s do no a ec i s ac i i y. Each assay o 1 ml inal olume con ained 50 mM T is䡠HCl (pH 7.6), 4 mM MgCl 2 , 2 mM NAD ⫹ ,40 ␮ M ␣ -D-glucose-1,6-diphospha e, 1.4 uni s G6PD, and 23 ␮ go Synechocys is cy osolic p o eins. The eac ion was s a ed by addi ion o 4 mM ␣ -D-glucose-1- phospha e, and he abso bance a 340 nm was moni o ed. Incuba ions o cy osolic ex ac be o e measu emen we e pe - o med in 25 mM T is䡠HCl (pH 7.6) and 2 mM MgCl 2 a 25°C and a a p o ein concen a ion o 4 mg兾ml wi h addi ions as speci ied. Wild- ype T xA was p epa ed as in e . 14. Resul s and Discussion Isola ion o P o eins Bound o T xA35. Sc eening o he Synecho- cys is ex ac s o p o eins in e ac ing wi h he T xA Cys-35-Se mu an , T xA35, yielded cy osolic a ge p o eins (Fig. 1A)as well as pe iphe al memb ane a ge p o eins (Fig. 1B). The co alen bond be ween hese p o eins and T xA35 was con- i med by non educing兾 educing wo-dimensional SDS兾PAGE (Fig. 1C). The majo i y o he cap u ed Synechocys is p o eins mig a e unde non educing condi ions wi h an appa en molec- ula mass ha is ⬇15 kDa la ge han he appa en molecula mass unde educing condi ions (Fig. 1C). This di e ence in mass co esponds o one co alen ly bound T xA35 molecule, which is eleased a e educ ion. In some cases, he di e ence in mig a ion be ween he i s and second dimension is so g ea ha i implies mo e han one T xA35 molecule joined o a a ge and兾o an oligome ic s uc u e o he a ge p o ein media ed by in e molecula disulphides. T xA35 con ains only one cys eine, hus excluding he possibili y o T x ime s. A la ge amoun o T xA35 o ms homodime s du ing he binding assay and mi- g a es wi h coo dina es 33 kDa兾15 kDa, and some T xA35 emains a monome and mig a es a he diagonal (Fig. 1C). The diagonal is p ac ically de oid o Synechocys is p o eins. Iden i ica ion o Synechocys is T xA-Linked P o eins. Twen y-six o he esol ed p o eins we e posi i ely iden i ied wi h hei co esponding ORFs in he Synechocys is genome. The 18 cy osolic p o eins a e lis ed in Table 1, and he 8 pe iphe al memb ane p o eins a e lis ed in Table 2. One o he a ge s, he la ge subuni o ibulose 1,5-bisphospha e ca boxylase兾 oxygenase (Rubisco) (RbcL), was ound bo h in he cy osolic and he pe iphe al memb ane p o ein ac ions. The as majo i y o he a ge p o eins, 20 o 25, ha e no p e iously been associa ed wi h T x-media ed egula ion. Only wo o he Synechocys is a ge s, ansla ion elonga ion ac o s EF-G and EF-Tu (Table 1), ha e been epo ed ea lie o in e ac wi h chlo oplas T xs (6). I is s iking ha he e a e no o he common p o eins be ween he 35 chlo oplas T x a ge s iden i ied by Balme e al. (6) and he 25 Synechocys is T x a ge s epo ed in his s udy. The implica ions o ou indings in ela ion o he cu en knowledge abou he a ge enzymes and he me abolic p ocesses in which hey pa icipa e a e discussed below. Ca bon dioxide ixa ion. Cy osolic (Table 1) and memb ane- associa ed (Table 2) o ms o RbcL we e ound o be a ge s o T x. In bo h cases, RbcL mig a ed wi h much oo high appa en molecula mass unde non educing condi ions o be compa ible wi h an RbcL-T xA35 he e odime (Fig. 1C). Cys eine 247 o spinach RbcL, which co esponds o Synechocys is Cys-242, is known o media e dime o ma ion be ween neighbo ing RbcL subuni s (15). Nex o he ac i e si e o RbcL a e wo o he conse ed cys eines whose edox s a e a ec s enzyme ac i i y in Synechocys is (16). A bond be ween T xA35 and one o hese cys eines in addi ion o a disulphide be ween wo Cys-242 esidues would yield an RbcL 2 -T xA35 2 he e o e ame . Such a polypep ide wi h a complex b anched s uc u e could be mo e e a ded on SDS兾PAGE han would be expec ed om he sum o he molecula masses o he indi idual subuni s. The e is as ye no e idence o T x-media ed egula ion o RbcL in any o ganism, bu he small subuni o Rubisco in chlo oplas s has been iden i ied as a T x- a ge (5, 6). CcmM (Table 2) is a ca bon dioxide-concen a ing mechanism p o ein, which is equi ed o 16108 兩 www.pnas.o g兾cgi兾doi兾10.1073兾pnas.2534397100 Lindahl and Flo encio g ow h a ai le els o CO 2 and is p obably a cons i uen o he ca boxysome (17). We iden i ied CcmM as a p o ein mig a ing on SDS兾PAGE wi h an appa en molecula mass o 54 kDa (Fig. 1B), which is in ag eemen wi h Ogawa e al. (17) bu a a iance wi h Cyanobase, whe e i is epo ed as a 73-kDa p o ein. No s uc u al ca boxysomal p o eins ha e so a been epo ed o Fig. 1. (Aand B) Isola ion o Synechocys is T x a ge p o eins wi h he His- agged si e-di ec ed T x mu an T xA35 immobilized on a Ni-a ini y ma ix. To al p o ein be o e binding (TP), unbound p o eins (UB), and p o eins washed o wi h 60 mM imidazole (W) we e sepa a ed on 12% ac ylamide gels and s ained wi h Coomassie b illian blue. Six mic oli e s om each ac ion o cy osolic p o eins and 10 ␮ l om ac ions o pe iphe al memb ane p o eins we e applied on he gels, co esponding o 45 and 30 ␮ g o ini ial o al p o ein (TP), espec i ely. Se en y mic oli e s each o cy osolic p o eins (E1) and pe iphe al memb ane p o eins (E2) bound o T xA35 and elu ed wi h 1 M imidazole we e sepa a ed on 12% ac ylamide gels unde educing condi ions. The iden i ies o he mos abundan a ge s a e indica ed. (C) Sepa a ion o T x a ge p o eins by non educing兾 educing wo-dimensional SDS兾PAGE. Elu ed p o eins E1 and E2 we e sepa a ed on 12% and 10% ac ylamide gels, espec i ely. Be o e he second dimension, p o eins we e educed wi h 100 mM DTT. The molecula masses o he p es ained s anda d p o eins a e indica ed by numbe s. Table 1. Synechocys is cy osolic T xA-linked p o eins ORF Gene p oduc Appa en molecula mass, kDa Theo e ical molecula mass, kDa Numbe o cys eines conse ed* sll1499 Fe edoxin-GOGAT (GlsF) 200 169.4 16 (15) sll1502 NADH-GOGAT (Gl B) 186 169.0 10 (10) sl 0557 Valyl- RNA syn he ase (ValS) 105 102.7 5 (3) sl 1463 Elonga ion ac o EF-G 90 76.7 2 (2) sll0158 Glucan b anching enzyme (GlgB) 84 89.5 6 (0) sl 2076 60-kDa chape onin G oEL 63 57.5 1 (1) sll0726 Phosphoglucomu ase 58 61.1 1 (0) sl 0009 Rubisco la ge subuni (RbcL) 57 52.5 9 (8) sll1393 Glycogen syn hase (Glg2) 55 56.2 10 (4) sl 1176 ADPglucose py ophospho ylase 54 49.3 4 (3) sl 0585 A gininosuccina e syn he ase 53 44.5 2 (1) sll1099 Elonga ion ac o EF-Tu 51 43.7 1 (1) sll1212 GDP-mannose dehyd a ase 45 41.3 2 sll1994 Po phobilinogen syn hase 42 36.1 4 (0) sll0576 Suga -nucleo ide epime ase 35 34.9 1 (1) sl 1198 1-Cys pe oxi edoxin 29 23.5 1 sll1621 YLR109-homologue 25 21.2 2 (2) ss 3383 Phycobilisome linke (L c ) 13 7.8 1 *To al numbe o conse ed cys eines in homologues om any o ganism and hen, in pa en heses, he numbe o cys eines conse ed in plan chlo oplas homologues. Lindahl and Flo encio PNAS 兩 Decembe 23, 2003 兩 ol. 100 兩 no. 26 兩 16109 PLANT BIOLOGY in e ac wi h T xs, bu a chlo oplas ca bonic anhyd ase was ecen ly disco e ed as a a ge o T x (6). PGM. The in e con e sion o glucose-1-phospha e and glucose- 6-phospha e ca alyzed by PGM (Table 1) b idges glycolysis and glycogen me abolism. Al hough PGM is ecognized o play a key ole in suga me abolism, i s ac i i y has no been conside ed a a ge o me abolic con ol. PGM ac i i y in in ac chlo oplas s was in e es ingly ound o be up- egula ed in he ligh , bu p eincuba ion wi h 10 mM DTT in he da k had no e ec on his ac i i y (18). To es he possibili y o edox egula ion o PGM in Synechocys is, we examined he e ec s o DTT and T xA on PGM ac i i y. Incuba ion o cy osol wi h 5 mM DTT o 45 o 90 min be o e measu emen led o a subs an ial inc ease in PGM ac i i y (Fig. 2). In con as , con ol incuba ion wi hou addi ions esul ed in loss o ac i i y. Fu he loss was obse ed when p eincuba ion was pe o med in he p esence o a low concen- a ion o CuCl 2 , which is known o induce he o ma ion o disulphide bonds (5). Addi ion o DTT a low (0.2 mM) con- cen a ion main ained he ac i i y cons an h oughou he incuba ion and, inally, addi ion o 2.5 ␮ M pu i ied Synechocys is wild- ype T xA in he p esence o 0.2 mM DTT caused a signi ican inc ease in PGM ac i i y (Fig. 2). The ac ha 0.2 mM DTT alone p e en ed loss o ac i i y may be a ibu able o he p esence o endogenous T xs in he cy osolic ex ac . Syn- echocys is PGM con ains a single cys eine esidue, which implies ha he egula o y disulphide is in e molecula . Glycogen syn hesis. ADPglucose py ophospho ylase (Table 1) ca - alyzes he syn hesis o ADPglucose om glucose-1-phospha e and ATP. This e e sible eac ion cons i u es he i s s ep o glycogen syn hesis in bac e ia and s a ch syn hesis in he plas ids o plan s and algae (19). The highe plan enzyme is a he e o e - ame consis ing o wo la ge subuni s and wo small subuni s. Expe imen s in i o (20, 21) and in i o (22) ha e shown ha he N- e minal cys eine (Cys-12) o he small subuni s may o m an in e molecula disulphide bond be ween hem, he eby inac i- a ing he enzyme. The oxidized, inac i e enzyme is slowly educed and, hus, eac i a ed by DTT (20) o T x (21) in i o. This mechanism o educ i e ac i a ion has been p oposed o ope a e in all dico plan s, because Cys-12 is conse ed in hese plan s (20). In cyanobac e ia, ADPglucose py ophospho ylase is encoded by a single gene whose p oduc o ms homo e ame s (23). The cyanobac e ial enzyme closely esembles he small subuni o highe plan s. Glycogen syn hases in bac e ia and s a ch syn hases in plan plas ids ca alyze he ans e o ADPglucose o he non educing end o a g owing ␣ -1,4-linked glucan. The Synechocys is glycogen syn hase (Glg2) (Table 1) is simila o he soluble s a ch syn- hases om plan s and algae (24). Ou inding ha Glg2 is a po en ial a ge o T x is an indica ion o a pos ansla ional egula ion o his enzyme in pho osyn he ic o ganisms. The glycogen-b anching enzyme (Table 1) ca alyzes he inal s ep in glycogen syn hesis, ans e ing a segmen o a p eexis ing ␣ -1,4-linked glucan in ␣ -1,6 posi ion (19). Thus, he comple e pa hway o glycogen syn hesis seems o be linked o T x in Synechocys is. Suga -nucleo ide me abolism. Nei he he pu a i e suga -nucleo ide epime ase兾dehyd a ase no GDP-mannose dehyd a ase (Table 1) p e iously has been linked o T x. The la e ca alyzes he i s s ep o he pa hway ha leads o syn hesis o GDP-L- ucose, a componen o , e.g., bac e ial capsula polysaccha ides. Sul u me abolism. The i s s ep o educ i e sul a e assimila ion is he ac i a ion o sul a e ca alyzed by sul a e adenylyl ans- e ase. We ound his enzyme among he pe iphe al memb ane p o eins bound o T xA35 (Table 2). 3⬘-Phosphoadenylylsul a e (PAPS) educ ase, which was no ound among he T x a ge s in his s udy, educes ac i a ed sul a e o sul i e and is one o he bes known accep o s o educing equi alen s om T x in E. coli (25). Ins ead, we ound he nex enzyme o he pa hway, e e- doxin-dependen sul i e educ ase (Table 2), which is homolo- gous o he highe plan plas id e edoxin-sul i e educ ase (26). Cys eine syn hase, he las enzyme o he pa hway, ecen ly has been iden i ied as a a ge o chlo oplas T xs (6). Ni ogen me abolism. Ammonium assimila ion occu s h ough a cyclic pa hway in ol ing wo enzymes, GS and GOGAT. In his Table 2. Synechocys is pe iphe al memb ane T xA-linked p o eins ORF Gene p oduc Appa en molecula mass, kDa Theo e ical molecula mass, kDa Numbe o cys eines conse ed* sll1789 RNA polyme ase ␤ ⬘subuni 191 144.7 6 (5) sll1787 RNA polyme ase ␤ subuni 152 123.3 3 (3) sl 0335 Phycobilisome linke (L CM ) 92 100.2 3 sl 0963 Fe edoxin-sul i e educ ase 67 71.4 6 (5) sl 0009 Rubisco la ge subuni (RbcL) 56 52.5 9 (8) sll1031 Ca boxysomal p o ein (CcmM) 54 73.1 (56.9) † 2 sl 1165 Sul a e adenylyl ans e ase 43 43.7 4 (0) sll1804 30S ibosomal p o ein S3 32 27.1 1 (0) *See Table 1 o de ini ion o numbe s. †The numbe in pa en heses co esponds o he molecula mass o CcmM in e . 17. Fig. 2. Ac i i y o Synechocys is PGM. Con e sion o glucose-1-phospha e o glucose-6-phospha e in cy osolic ex ac s was measu ed be o e and a e incuba ion a 25°C o 45 and 90 min wi h he addi ions speci ied. Concen- a ions o CuCl2and T xA we e 25 and 2.5 ␮ M, espec i ely. The alues a e he means o h ee di e en expe imen s, and SDs a e p esen ed as e o ba s. The ini ial ac i i y be o e incuba ion was 0.38 ␮ mol䡠mg o p o ein⫺1䡠min⫺1. 16110 兩 www.pnas.o g兾cgi兾doi兾10.1073兾pnas.2534397100 Lindahl and Flo encio pa hway, GS has been conside ed he exclusi e a ge o egula ion a bo h ansc ip ional and pos ansla ional le els in cyanobac e ia (27). Su p isingly, bo h e edoxin-dependen GOGAT and he la ge subuni o NADH-dependen GOGAT (Gl B) we e cap u ed by T xA35 (Table 1), sugges ing ha glu ama e syn hesis may be edox- egula ed. Wes e n blo using an ibodies agains Gl B con i ms ha his p o ein is p esen in elua e E1, which con ains cy osolic p o eins bound o T xA35 (Fig. 3). In addi ion, we ound ha in i o mos o Gl B pa icipa es in high-molecula weigh complexes c osslinked by disulphides (Fig. 3). The appa en molecula mass o hese complexes canno be de e mined, because i exceeds ha o he la ges ma ke p o ein o 250 kDa, bu i is possible ha i co esponds o a dime o he 186-kDa Gl B subuni s. Howe e , i can be concluded ha T xA35 does no in e ac wi h he pa icula cys eine in ol ed in complex o ma ion, because ha would p oduce a Gl B-T xA35 he e odime o ⬇200 kDa mo- lecula mass (Fig. 3, lane 3). Chlamydomonas einha d ii GS (28) and plan GS (29) a e ac i a ed on educ ion by T x o DTT. In bo h e s. 5 and 6, GS is epo ed as a a ge o chlo oplas T xs. Synechocys is GS ( ype I) has ne e been ound o espond o hiol eagen s bu ins ead is known o be egula ed h ough he e e sible binding o wo small p o eins (27). In his a icle, we show ha Synechocys is GS is absen om elua e E1 despi e i s abundance in he cy osolic ex ac (Fig. 3). A gininosuccina e syn he ase (Table 1) is he a e-limi ing enzyme o bo h he u ea- and a ginine-ci ulline cycles. Like mos cyanobac e ia, Synechocys is s o es excess educed ni ogen in he o m o cyanophycin, a non ibosomally syn hesized poly- me o aspa a e and a ginine (30). Thus, he syn hesis o a ginine is analogous o he syn hesis o ac i a ed glucose o s o age o educed ca bon as glycogen. T ansc ip ion. RNA polyme ase subuni s ␤ and ␤ ⬘we e ound o eac wi h T xA35 (Table 2). In e es ingly, in i o ansc ip ion wi h pu i ied RNA polyme ase om he cyanobac e ium Anabaena sp. PCC 7120 equi es he p esence o hiol- educing agen s (31). Thus, i is possible ha polyme ase ac i i y in he da k is low because o oxida ion o egula o y hiols and ha ac i i y is eco e ed in he ligh a e educ ion by T x. Indeed, cyanobac e ial o e all gene exp ession dec eases d as ically a e a shi om ligh o da kness and e u ns o ini ial le els a e he onse o ligh (32). P o ein syn hesis and olding. A yeas homologue o alyl- RNA syn he ase (Table 1) al eady is known o unde go educ i e ac i a ion in a eac ion ha in ol es hiols (33). The ansla ion elonga ion ac o s EF-G and EF-Tu (Table 1) we e bo h isola ed as a ge s o chlo oplas T xs (6), and EF-Tu possesses p o ein disulphide isome ase ac i i y in E. coli (34). Howe e , i can be ques ioned whe he he 30S ibosomal p o ein S3 (Table 2) is a ue a ge , because i s single cys eine is conse ed in only Cyanopho a pa adoxa bu no o he species. The 60-kDa chap- e onin G oEL (Table 1) closely esembles he chlo oplas Rubisco-binding p o ein subuni s ␣ and ␤ , which we e epo ed ea lie as chlo oplas T x a ge s (6). Fu he mo e, he Hsp70 chape one is egula ed by a T x-like p o ein in plan s (35). Te apy ole biosyn hesis. I has been demons a ed ha syn hesis o p o opo phy in IX om 5-aminole ulinic acid is s imula ed by DTT in chlo oplas ex ac s (36). Th ee enzymes om he chlo oplas heme and chlo ophyll syn hesis pa hway ecen ly we e ound o in e ac wi h T xs (6). In his a icle, we epo ano he enzyme om he same pa hway, po phobilinogen syn- hase (Table 1). Oxida i e s ess esponse. Pe oxi edoxins (P xs) cons i u e an ubiq- ui ous amily o hiol-speci ic pe oxidases (37). A ligh - dependen , hiol-speci ic pe oxidase ac i i y was disco e ed in a ca alase dele ion mu an o Synechocys is (38), sugges ing he p esence o P x. Among ou T x a ge s we ound a 1-Cys P x and a homologue o he yeas pe oxidase YLR109 (Table 1). Yeas YLR109 has been shown o in e ac wi h an A abidopsis haliana T x h(4). T xs, o T x-like p o eins, a e known o educe chlo oplas 2-Cys P x (6, 39) and P x-Q (5). Ligh ha es ing. The phycobilisome co e linke polypep ide (L C ) (Table 1) links h ee allophycocyanin he e odime s, he eby o ming he co e o he phycobilisome (40). The Synechocys is L C con ains a single cys eine esidue, which is conse ed in se e al cyanobac e ial species, such as Synechococcus sp. PCC 6301, bu no in Anabaena sp. PCC 7120. The phycobilisome co e- memb ane linke polypep ide (L CM ) (Table 2) ancho s he en i e phycobilisome o he memb ane (40). One o he h ee abso- lu ely conse ed cys eines in he Synechocys is L CM polypep ide, Cys-190, is p obably in ol ed in ch omopho e binding (40). The p esence o edox-ac i e cys eines in bo h L C and L CM sugges s ha hey may bind each o he when joined unde oxidizing condi ions. This inding means ha he phycobilisome could be co alen ly associa ed o he memb ane unde ligh -limi ing condi ions by a disulphide ha is b oken by T x a e an inc ease in ligh in ensi y, ensu ing g ea e lexibili y in ligh ha es ing a highe ligh in ensi ies. In e es ingly, ea men o Synecho- coccus wi h N-e hylmaleimide dis up s ene gy ans e om phycobilisomes o pho osys em I wi hou pe u bing he in eg i y o ei he phycobilisomes o pho osys ems I o II (41). Conse a ion o Cys eines in Synechocys is T xA-Linked P o eins. Conside ing he possibili y o conse a ion o T x- ela ed eg- ula o y mechanisms h ough e olu ion o pho osyn he ic o gan- isms, we examined he p esence o conse ed cys eines in he Synechocys is amino acid sequences (see Table 3, which is published as suppo ing in o ma ion on he PNAS web si e). All T xA-linked Synechocys is p o eins ha e a leas one cys eine conse ed in homologous p o eins om some o he o ganism (Tables 1 and 2), and 17 o hese Synechocys is T xA a ge s ha e a leas one cys eine conse ed in chlo oplas homologues (Tables 1 and 2). Th ee p o eins wi h no cys eines conse ed be ween Synechocys is and A abidopsis chlo oplas coun e pa s [glucan b anching enzyme, PGM, and sul a e adenylyl ans- e ase (Tables 1 and 2)] display e y low o e all simila i ies o he chlo oplas p o eins. A close look a hese chlo oplas enzymes e eals ha hey a e much mo e simila o euka yo ic p o eins (da a no shown). I mus he e o e be concluded ha a possible T x-media ed egula ion has no been conse ed h ough chlo- oplas e olu ion in hese cases. Po phobilinogen syn hase in bac e ia, including cyanobac e ia, ungi, and animals, is a zinc- dependen enzyme ha uses h ee icinal cys eines o coo dina e he zinc ion (42). In plan chlo oplas s and some bac e ia, hese cys eines ha e been eplaced wi h aspa a es ha pa icipa e in coo dina ing a magnesium ion. The pu i ied zinc-con aining Fig. 3. Examina ion o NADH-GOGAT and GS by Wes e n blo analysis. Lanes 1 and 2 con ain 15 ␮ g o cy osolic Synechocys is p o eins; lanes 3 and 4 con ain 5 ␮ l o elua e E1 (cy osolic p o eins bound o T xA35). Lanes 1 and 3 do no include educing agen s; lanes 2 and 4 include 10 mM DTT. P o eins sepa a ed by SDS兾PAGE we e analyzed by using an ibodies agains GS and he la ge subuni o NADH-GOGAT (Gl B). Molecula mass ma ke s o 250, 64, and 50 kDa a e indica ed, as is he 186-kDa appa en molecula mass o he mono- me ic educed Gl B. Lindahl and Flo encio PNAS 兩 Decembe 23, 2003 兩 ol. 100 兩 no. 26 兩 16111 PLANT BIOLOGY enzyme depends on exogenous hiols o main aining he bound zinc and, consequen ly, o i s ac i i y (43), whe eas he plan enzyme does no (44). Thus, T x may unc ion as an an ioxidan , enabling zinc binding in i o and so con olling he ac i i y. This could be an example o a unc ion o T x los in one b anch h ough e olu ion. Synechocys is ADPglucose py ophospho y- lase sha es h ee cys eines (Cys-55, -325, and -330) wi h i s A abidopsis chlo oplas homologue. Mo e in e es ing, howe e , a e he wo N- e minal cys eines (Cys-2 and -3) ha a e no s ic ly conse ed bu appea only 2 aa ups eam o he chlo o- plas Cys-12, which has been implied in edox egula ion o his enzyme. Hence, he e is a possibili y ha cys eines could be unc ionally equi alen wi hou being conse ed in he sequence. Concluding Rema ks. The pu a i e Synechocys is T xA a ge p o- eins iden i ied in his s udy a e mainly enzymes pa icipa ing in anabolic p ocesses. A pic u e begins o eme ge in which he cyanobac e ial p ocesses o assimila ion and s o age o ca bon, sul u , and ni ogen in glycogen, p o eins, and cyanophycin, espec i ely, a e all possible a ge s o ligh -induced edox egula ion by means o T x. Based on he knowledge o T x ac ion in chlo oplas s (1–3), i is expec ed ha educed T x has a s imula o y e ec on hese anabolic p ocesses, hus signaling he a ailabili y o ligh ene gy. Howe e , he iden i ies and ela i e abundance o T x a ge s a e likely o change depending on g ow h condi ions. The new po en ial T xA a ge p o eins poin o di e ences in he T x-media ed egula ion be ween cyanobac e ia and chlo- oplas s. In con as o plan s, cyanobac e ia depend on he ca bon dioxide-concen a ing mechanism o a unc ioning CO 2 assimila ion and g ow h a physiological le els o CO 2 (45). The inding ha CcmM in e ac s wi h T xA sugges s ha he ca bon dioxide-concen a ing mechanism is subjec o T x-media ed edox egula ion in cyanobac e ia, whe eas he Cal in cycle has become a p ime T x a ge p ocess in chlo oplas s. Each Syn- echocys is enzyme o he comple e pa hway o glycogen syn hesis was ound o in e ac wi h T x and no only he a e-limi ing enzyme, ADPglucose py ophospho ylase, as seems o be he case in plas ids (21). Synechocys is GOGAT, a he han GS, we e ound o in e ac wi h T xA, implying a shi o he ocus o edox egula ion o ammonium assimila ion be ween cya- nobac e ia and chlo oplas s. This s udy pa es he way o u u e wo k on egula ion media ed by T x in cyanobac e ia, which has long been consid- e ed insigni ican . I also gi es an in e es ing pe spec i e o he molecula e olu ion o suscep ibili y owa d educ ion by T x. This wo k was inanced by G an BMC 2001-2635 om he Spanish Minis y o Science and Technology (MCYT). M.L. is a holde o a Ramon-y-Cajal con ac om he MCYT. 1. Buchanan, B. B. (1980) Annu. Re . Plan Physiol. 31, 341–374. 2. Jacquo , J.-P., Lancelin, J.-M. & Meye , Y. (1997) New Phy ol. 136, 543–570. 3. Buchanan, B. B., Schu¨ mann, P., Wolosiuk, R. A. & Jacquo , J.-P. (2002) Pho osyn h. Res. 73, 215–222. 4. Ve doucq, L., Vignols, F., Jacquo , J.-P., Cha ie , Y. & Meye , Y. (1999) J. Biol. Chem. 274, 19714–19722. 5. Mo ohashi, K., Kondoh, A., S umpp, M. T. & Hisabo i, T. (2001) P oc. Na l. Acad. Sci. USA 98, 11224–11229. 6. Balme , Y., Kolle , A., del Val, G., Manie i, W., Schu¨ mann, P. & Buchanan, B. B. (2003) P oc. Na l. Acad. Sci. USA 100, 370–375. 7. Tamoi, M., Mu akami, A., Takeda, T. & Shigeoka, S. (1998) Biosci. Bio echnol. Biochem. 62, 374–376. 8. Kobayashi, D., Tamoi, M., Iwaki, T., Shigeoka, S. & Wadano, A. (2003) Plan Cell Physiol. 44, 269–276. 9. Na a o, F. & Flo encio, F. J. (1996) Plan Physiol. 111, 1067–1075. 10. Rippka, R., De uelles, J., Wa e bu y, J. B., He man, M. & S anie , R. Y. (1979) J. Gen. Mic obiol. 111, 1–61. 11. Reyes, J. C. & Flo encio, F. J. (1995) Plan Mol. Biol. 27, 789–799. 12. Ma kwell, M. A., Haas, S. M., Biebe , L. L. & Tolbe , N. E. (1978) Anal. Biochem. 87, 206–210. 13. Ma que´s, S., Me´ ida, A., Candau, P. & Flo encio, F. J. (1992) Plan a 187, 247–253. 14. Na a o, F., Ma ı´n-Figue oa, E. & Flo encio, F. J. (2000) Plan Mol. Biol. 43, 23–32. 15. Ran y, B., Lo ime , G. & Gu e idge, S. (1991) Eu . J. Biochem. 200, 353–358. 16. Ma cus, Y., Al man-Gue a, H., Finkle , A. & Gu ewi z, M. (2003) J. Bac e iol. 185, 1509–1517. 17. Ogawa, T., Amichay, D. & Gu ewi z, M. (1994) Pho osyn h. Res. 39, 183–190. 18. Siche , R. C. (1989) Plan Physiol. 89, 557–563. 19. Ball, S. G. & Mo ell, M. K. (2003) Annu. Re . Plan Biol. 54, 207–233. 20. Fu, Y., Ballico a, M. A., Leykam, J. F. & P eiss, J. (1998) J. Biol. Chem. 273, 25045–25052. 21. Ballico a, M. A., F ueau , J. B., Fu, Y., Schu¨ mann, P. & P eiss, J. (2000) J. Biol. Chem. 275, 1315–1320. 22. Tiessen, A., Hend iks, J. H. M., S i , M., B anscheid, A., Gibon, Y., Fa e´, E. M. & Geigenbe ge , P. (2002) Plan Cell 14, 2191–2213. 23. Iglesias, A. A., Kake uda, G. & P eiss, J. (1991) Plan Physiol. 97, 1187–1195. 24. Cid, E., Ge emia, R. A., Guino a , J. J. & Fe e , J. C. (2002) FEBS Le . 528, 5–11. 25. Gleason, F. K. & Holmg en, A. (1988) FEMS Mic obiol. Re . 4, 271–297. 26. B uhl, A., Ha e kamp, T., Gisselmann, G. & Schwenn, J. D. (1996) Biochim. Biophys. Ac a 1295, 119–124. 27. Mu o-Pas o , M. I. & Flo encio, F. J. (2003) Plan Physiol. Biochem. 41, 595–603. 28. Flo encio, F. J., Gadal, P. & Buchanan, B. B. (1993) Plan Physiol. Biochem. 31, 649–655. 29. Choi, Y. A., Kim, S. G. & Kwon, Y. M. (1999) Plan Sci. 149, 175–182. 30. Ziegle , K., Diene , A., He pin, C., Rich e , R., Deu zmann, R. & Lockau, W. (1998) Eu . J. Biochem. 254, 154–159. 31. Schneide , G. J., Tume , N. E., Richaud, C., Bo bely, G. & Haselko n, R. (1987) J. Biol. Chem. 262, 14633–14639. 32. Singe , R. A. & Dooli le, W. F. (1975) Na u e 253, 650–651. 33. Black, S. (1986) Science 234, 1111–1114. 34. Richa me, G. (1998) Biochem. Biophys. Res. Commun. 252, 156–161. 35. Vignols, F., Mouaheb, N., Thomas, D. & Meye , Y. (2003) J. Biol. Chem. 278, 4516–4523. 36. Manoha a, M. S. & T ipa hy, B. C. (2000) Plan a 212, 52–59. 37. Wood, Z. A., Sch o¨de , E., Ha is, J. R. & Poole, L. B. (2003) T ends Biochem. Sci. 28, 32–40. 38. Tichy, M. & Ve maas, W. (1999) J. Bac e iol. 181, 1875–1882. 39. B oin, M., Cuine´, S., Eyme y, F. & Rey, P. (2002) Plan Cell 14, 1417–1432. 40. Sidle , W. A. (1994) in The Molecula Biology o Cyanobac e ia, ed. B yan , D. A. (Kluwe , Do d ech , The Ne he lands), pp. 139–216. 41. Glaze , A. N., Gind , Y. M., Chan, C. F. & Saue , K. (1994) Pho osyn h. Res. 40, 167–173. 42. Shoolingin-Jo dan, P. M., Spence , P., Sa wa , M., E skine, P. E., Cheung, K.-M., Coope , J. B. & No on, E. B. (2002) Biochem. Soc. T ans. 30, 584–590. 43. Spence , P. & Jo dan, P. M. (1993) Biochem. J. 290, 279–287. 44. Liedgens, W., Lu¨ z, C. & Schneide , H. A. W. (1983) Eu . J. Biochem. 135, 75–79. 45. Badge , M. R. & P ice, G. D. (2003) J. Exp. Bo . 54, 609–622. 16112 兩 www.pnas.o g兾cgi兾doi兾10.1073兾pnas.2534397100 Lindahl and Flo encio